EPA-460/3-74-011
July 1974
                                     A  STUDY
                            OF THE DIESEL
                         AS A LIGHT-DUTY
                             POWER  PLANT
             U.S. ENVIRONMENTAL PROTECTION AGENCY
                Office of Air and Waste Management
             Office of Mobile Source Air Pollution Control
                Emission Control Technology Division
                    Ann Arbor, Michigan  48105

-------
                                              EPA-460/3-74-011
       A  STUDY  OF  THE  DIESEL
AS  A LIGHT-DUTY  POWER  PLANT
                            by

         M. L. Monaghan, C. C. J. French, and R. G. Freese

             Ricardo and Company Engineers (1927) Ltd.
                        Bridge Works
                Shoreham-by-Sea, Sussex BN 45FG
                    Contract No. 68-03-0375

                     EPA Project Officers:

          T. C. Austin, J. J. McFadden, and K. H. Hellman



                         Prepared for

           U.S. ENVIRONMENTAL PROTECTION AGENCY
                Office of Air and Waste Management
            Office of Mobile Source Air Pollution Control
               Emission Control Technology Division
                   Ann Arbor, Michigan 48105

                          July 1974

-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers.  Copies are
available free of charge to Federal employees, current contractors and
grantees,  and nonprofit organizations - as supplies permit - from the Air
Pollution Technical Information Center,  Environmental Protection Agency,
Research Triangle Park, North Carolina 27711; or, for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia  22151.
This report was furnished to the Environmental Protection Agency by
Ricardo and Company Engineers (1927) Ltd. , in fulfillment of Contract No.
68-03-0375. The contents of this report are reproduced herein as received from
Ricardo and Company Engineers (1927) Ltd.  The opinions, findings, and
conclusions expressed are those of the author and not necessarily those
of the Environmental Protection Agency.  Mention of company or product
names is not to be considered as an endorsement by the Environmental
Protection Agency.
                       Publication No. EPA-460/3-74-011
                                     11

-------
                     1-1

                    CONTENTS
SECTION 1
            EXECUTIVE SUMMARY


SECTION 2
            INDEX AND LIST OF FIGURES
SECTION 3

            GENERAL SUMMARY AND INTRODUCTION


SECTION 4
            LITERATURE SURVEY, ASSESSMENT OF
            PROBLEM AREAS AND TRADE OFFS


SECTION 5
            ENGINE CONFIGURATION STUDY


SECTION 6
            POWER PLANT RATING


SECTION 7
            PROGRAMME PLANS
SECTION 8
            APPENDIX 1 - LIST OF KEYWORDS


SECTION 9
            APPENDIX 2 - LIST OF REFERENCES


SECTION 10
            APPENDIX 3 - GLOSSARY OF TERMS

-------
                             1-2
         THE DIESEL ENGINE AS A LIGHT DUTY POWER PLANT

                              SECTION 1
                       EXECUTIVE SUMMARY
Introduction

In America today the gasoline engine is virtually the only engine
used for light duty purposes due to its low initial cost and
refined performance.  However, current and proposed legislation
and the prospect of increased fuel costs may render the low
emissions gasoline engine less attractive than some of the other
available power plants.

In other parts of the world the diesel engine, which is the most
efficient practical engine produced today and which has good
emissions characteristics, has long been used as a light duty power
plant in applications where fuel economy and durability are of great
importance.  The lightweight, high speed diesel  has been developed
specifically for European and Japanese conditions while low fuel costs
have until now prevented any major effort being  expended on a light
duty engine for America.

The purpose of this study was to see if the diesel engine could be used
as a viable power plant for an American passenger car.  It was felt
necessary to carry out the study because European  effort has tended
towards the lighter weight European vehicles while  American effort
has been towards meeting legislative limits with the gasoline engine
while retaining traditional performance standards.

Proposed legislation indicates that a light duty vehicle for the near
future  should be able to meet emissions targets  of  : -

                 HC               0.41 g/mile
                 CO               3.4 g/mile
                 NOX               1.5 g/mile

when tested by the CVS-CH procedure.

-------
                           1-3
A light duty vehicle for longer term application should be able to meet
emissions targets of :-

                 HC               0.41 g/mile
                 CO               3.4  g/mile
                 NOX              0.4  g/mile

For the purposes of this study the vehicle considered was a 4/5 seat
sedan with a loaded weight of about 1600 kg (3500 Ib), capable of
0-9(7 km/h  (0-60 mph) in 13.5 s and 32-113" km/h (20-70 mph)
in 15 s,  i.e. a compact size sedan but with "standard1 performance
capabilities.

The first phase, a literature survey, involved a study of all existing
light duty and relevant heavy duty diesel literature from published
sources in Europe, America, Japan and also from Ricardo in-house
technical reports.  Visits to diesel manufacturers and users  were also
made. At the end of this phase, broad conclusions on the feasibility and
likely problem areas of an American light duty diesel could be made.


The next phase required that brief design studies be made to cover
all the potentially viable diesel power plants.  This involved the
calculation of performance and outline design of any type of diesel
which could power the target vehicle  in the above  emissions climate.


For the third phase a rating methodology was derived  which allowed
a numerical comparison of all the potentially viable power plants to
be made.
The fourth and final phase of the study involved a consideration of the
results of the previous sections in order to make recommendations
for further action and  effort to achieve the most desirable light duty
vehicle.   Although these recommendations were to be aimed mainly
at light duty use, other areas uncovered in the course of the study
were not specifically excluded.

-------
                         1-4

Conclusions

On completion;  of the study the following broad conclusions were drawn  : -

     1.     .A diesel pow.eredpassenger car could be built using
           present technology.

           It would have equal acceleration and general performance
           to typical gasoline powered passenger cars.

           The noise level would be perfectly acceptable and in fact
           would be less than 1 dBA greater than a gasoline powered
           vehicle in an S.A.E. drive-by test.

           The engine would be hardly any larger than the equivalent
           V-8 gasoline engine and would weigh only 68 kg (150 Ib)
           more.

           There would be very little if any visible smoke under
           normal driving conditions.

           The diesel powered vehicle would meet the primary
           emission targets without the use of catalysts or other
           special equipment.

           The diesel powered vehicle will deliver up to 50% greater
           fuel economy than the equivalent gasoline powered vehicle,
           depending on the driving cycle.

     2.     The study indicates that a diesel powered vehicle could not
           meet  the secondary emissions target of  0.4 g/mile NOx
           if the target vehicle and performance standards are adhered
           to and if current technology is assumed.

           Virtually no work has been carried out to determine the
           emissions capabilities of the light duty diesel for these
           secondary targets..  More basic research is required to
           determine the ultimate emissions potential of the light
           duty diesel.

           It would seem that a diesel powered vehicle could meet
           a 1.0 g/mile NOX  standard with the use of modulated exhaust
           gas recirculation but lower standards than 1.0 g/mile would
           require a reduction in vehicle performance and/or weight.

-------
                           1-5

Recommendations

As-a: result of the study the following recommendations were made :-

     1...    A V-8 97 kW(!30 bhp) naturally aspirated indirect
           injection diesel engine should be constructed and
           demonstrated in a  4/5 seat sedan.

     2-,    A 6 cylinder indirect injection diesel engine should be
           built and work carried out with the engine to investigate
           the application of turbo-chargers and "Comprex" pressure
           exchangers to automotive engines.

     3,.    Advanced fuel injection  systems could help to bring about
           ftirther reductions in pollutants and make it possible to
           use the direct injection diesel: as a possible automotive
           power plant, while novel forms of construction may allow
           some reduction in  the cost of fuel injection equipment.
           Thus work should be pursued on advanced and novel forms
           of fuel injection equipment.

     4*.    A. thoro ugh investigation should be made of the potential
           of the low compression  ratio , ignition assisted, indirect
           injection diesel engine in view of its possible fuel economy
           advantages.

     5-    A fundamental investigation should be carried  out into the
           formation of unburnt hydrocarbons in diesel engines.  The
           development of a catalyst for control of hydrocarbons is also
           desirable.

     6..    Work should be initiated to develop control systems for
           load and speed modulation of exhaust gas recirculation.

     7.    Work on the development of improved starting  aids such as
           'instant1 glow plugs and programmed starting  sequences
           should be carried out.

     8.    The clear superiority of the diesel for such specialised
           applications as taxi cabs and light delivery vehicles indicates
           that a programme  to demonstrate and encourage the conversiot
           of these vehicles to diesel power should be instituted.

     9%    An investigation into particulate formation in the diesel engine
           should be carried out together with the development of smoke
           and particulate traps.

-------
Literature Study

Current light duty diesel experience is mainly concerned with European
type vehicles and thus the literature study was carried out to indicate
the feasibility of the diesel engine for American light duty use and the
likely problem areas and possible trade-offs  if  it were adopted.

The survey covered the period 1919 to the present day although the
great bulk of the literature studied was published in the period 1950
to the present.  A key-word indexing system  was  devised so that
pertinent points could be extracted easily from  the literature and so
that data relevant to a particular point could be assembled quickly
and thoroughly.  Sources of information in this  section of the study
were proceedings of learned societies , technical journals , Ricardo
published information,  Ricardo unpublished information and discussions
with manufacturers and users of light duty diesel engines.

The overall conclusion of the literature study was that the diesel engine
is a potentially viable power plant for the target vehicle and that it could
meet the primary emissions targets without the use of a catalyst.

Specific conclusions on various performance  aspects were as follows :-

Smoke - The target vehicle should emit very  little visible smoke and
under normal driving conditions there should be virtually none.

Odour -  Simple timing  controls should minimise the slight odour of the
high speed diesel under light load conditions while the use of an indirect
injection combustion chamber and the proposed restricted power rating
should minimise the full load odour.

Gaseous Emissions - A naturally aspirated indirect injection engine
capable of powering the prototype vehicle should be able to meet the
primary emissions targets of 0.41 g/mile HC,  3«4 g/mile CO  and
1.5 g/mile NOX  without the use of catalysts, although some small amount
of exhaust gas recirculation may be required to ensure sufficient margin
for production compliance.  The secondary target of 0.4 g/mile NOX could
not be achieved with current technology if the target vehicle size and power
to weight ratios are assumed.  The  use of conventional direct injection
systems would make it  difficult to meet both NOX and CO standards while
the injection timing retard required would cause high HC levels.

-------
                           1 - 7
Particulates   - The diesel, like all engines employing heterogeneous
combustion,  emits more exhaust particulates than the gasoline engine,
but neither the effect of particulates on health nor the importance of
the nature of particulates is known at present. Filter  systems could
probably remove diesel particulates if necessary.

Noise - Attention to engine and vehicle design would make the diesel
passenger car perfectly acceptable. Drive-by noise  levels would be
only slightly higher (1  or 2 dBA) than the gasoline powered vehicles.
Idling noise can be annoying with current diesel engined vehicles but
it is unlikely to be a significant cause for complaint with the target
vehicle.

Volume  - The volume of the diesel engine will be greater than that
of the equivalent highly rated gasoline  engine, but there should be
no installation problems in the target vehicle.

Weight  - A  97  kW (130 bhp) diesel engine may be 136 kg  (300 Ib)
heavier than  a highly rated gasoline engine of the same power output
but only about 68 kg (150 Ib) heavier than a more normal and lower
rated gasoline engine.

Fuel Economy  -  The diesel engine is  undoubtedly the most efficient
current prime mover for light duty use and in city conditions the
fuel consumption advantage may be as  much as 50% (twice the miles
per gallon).

Fuel  - The increasing cost of fuel makes the high economy diesel
more attractive and there should be no difficulty in increasing the
quantity of automotive diesel fuel.

First Cost -  The  diesel engine will cost between 1.5 and 2 times as
much as the  gasoline engine in America (i.e. % 300 against  % 200
production cost  )  and about half of the cost increase  is due to the
fuel injection equipment.

Maintenance   - Longer periods between overhauls and the similar
nature of the maintenance requirements to those of the gasoline
engine will mean that overall maintenance costs will  be less for
the diesel powered vehicle.

Starting  -  Starting is inferior to the gasoline engine but the maximum
delay with heater plugs will be 30 seconds.

-------
                              1-8

Hot Driveability  - The hot driveability of the diesel engine will be
very similar to that of the gasoline engine and no major transmission
changes will be required.

Cold Driveability - The cold driveability of the diesel engine will be
as good as its hot driveability and very much superior to the gasoline
engine.

Torque Rise  - The torque rise of the diesel engine will be similar
to that of the gasoline engine.

Durability  - The diesel engine will have greater durability than its
gasoline counterpart.

Coolant. Heat Loss - The diesel has reduced heat loss at low load and
idle so that ' slow speed traffic overheating ' will not be a problem
although winter morning de-icing may be more difficult. The
increased heat losses at full load may dictate the use of a larger
radiator, although this may only be required for trailer-towing
vehicles.

Vibration and Torque Recoil -  The unthrottled, high compression ratio
diesel undoubtedly vibrates  more than the gasoline engine especially at
idle.  This tends to give an impression of harshness with current
vehicles but this impression disappears  once the vehicle is in motion.
The problem  can be minimised by careful attention to engine and
transmission mountings.

Manufacture   - The similarities between diesel and gasoline engines
allow them to be made on the same production lines if necessary although
the diesel  engine does require  closer control of tolerances.

Ancillaries -  Ancillaries are similar and often identical to those of
the gasoline engine.

Production techniques for the high volume production of conventional
fuel injection equipment have already been developed but there may be
scope for cost reductions with greater production numbers.

Lubrication  - The diesel engine requires oils with greater dispersant
and anti-corrosive qualities than does the gasoline engine.

-------
                          1-9
Engine Configuration Study

The literature study and Ricardo in-house knowledge indicated that
a diesel engine could provide a viable power plant for a passenger
car and thus the study required that all potentially viable diesel
variants be designed in sufficient detail to allow their likely performance
to be assessed realistically. The vehicle to be powered by any of the
candidate  power plants was a 4/5 seat sedan weighing less than 1600 kg
(3500 Ib) , and capable of meeting the EPA standard car performance
specifications, i.e. 0- 97 km/h (0-60 mph) in less than 13.5 s. ,
32-113 km/h (20-70 mph) in less than 15 s and capable of overtaking
a 80 km/h (50 mph) truck in less than 15 s.  The emission targets were

     Primary (or short term)

            HC                    0.41 g/mile
            CO                    3.4  g/mile
            NOX                    1.5  g/mile

     Secondary (or long term)

            HC                    0.41 g/mile
            CO                    3.4  g/mile
            NOX                    Oo4  g/mile

and were to be obtained when the vehicle was tested according to the
CVS-CH procedure. Computer calculations indicated that a 3-speed
automatic gear box would require an installed power of approximately
97 kW (130 bhp) for engines with normal torque characteristics, and
thus all the diesel power plants were designed with this in mind.

While no specific  gasoline configuration exercise was required, two
gasoline engines were  studied to provide a basis for comparison
with the diesel engine types.

The diesel engine types considered covered all of the major  combustion
systems,  all likely engine configurations  and operating cycles (e.g.
4 stroke,  rotary, turbo-compound) and different possible installations
(e<»g. air-cooled, water-cooled), but the only ones schemed in detail
for this study were those engine configurations and types which were
considered at all  suitable for passenger car application.

The following paragraphs summarise the results of the configuration
study while Table 1 shows some of  the more important engine
parameters for the diesel engines.

-------

o mm
Bore
in
c. , rnm
Stroke
in
kW
Power ,
hp
Swept litre
Volume in3
Weight ^
Box m3
Volume ft3
Specific power- kW/1
o
swept volume hp/in
Power/unit kW/mm
piston area hp/in^
Specific power- kW/m3
box volume hp/ft3
Specific weight kg/1
- swept volume lb/in3
Specific weight kg/kW
- power Ib/hp
V8
Gaso-
line
97
3.82
76
3.00
96
128
4.5
275
250
550
_
21.3
0.47
.
-
55.6
2.00
2.60
4.30
IL6
Gaso-
line
88
3 .46
82
3.22
96
128
2.99
183
186
410
_
32.11
0.70
-
-
60.2
2.19
1.87
3.12
V8
NA
IDI
88
3.46
98
3.86
96
128
4.78
292
320
700
0.32
11.2
20.2
0.44
0.0020
1.70
302
11.4
66.9
2.41
3.31
5.47
V6
TC
IDI
90
3.54
100
3.94
96
128
3.84
234
309
680
0.33
11.6
24.4
0.55
0.0025
2.15
291
11.05
80.9
2.91
3.22
5.31
In Line
6 TC
IDI
90
3.54
100
3.94
96
128
3.84
234
327
720
0.36
12.7
24.4
0.55
0.0025
2.15
265
10.1
85.6
3.08
3.41
5.62
V6
TC
DI
93
3.66
94
3.70
96
128
3.84
234
300
660
0.31
11.0
24.4
0.55
0.0023
2.03
306
11.6
78.5
2.83
3.12
5.16
In Line
6 TC
DI
93
3.66
94
3.70
96
128
3.84
234
310
680
0.34
12.0
24.4
0.55
0.0023
2.03
280
10.7
80.7
2.91
3.22
5.31
V6
2 Stroke
Loop IDI
99
3.89
114
4.50
96
128
5.25
320
340
760
0.36
12.7
18.3
0.40
0.0021
1.82
265
10.1
65.7
2.37
3.59
5.93
In Line 6
2 Stroke
Uniflow DI
83
3.28
114
4.50
96
128
3.74
228
365
800
0.47
16.8
26.0
0.57
0.0029
2.52
201
8.0
98.4
3.55
3.78
6.25
4 Cyl.
Compound
DI
93
3.66
93
3.66
96
128
2.52
153
305
670
0.32
11.3
38.0
0.83
0.0035
3.04
299
11.3
120.6
4.34
3.17
5.23
2 Rotor
2 Stage
Rotary
-
-
96
128
^
227
500
0.26
9.2
-
-
368
13.9
_
2.36
3.91
                                                                            I
                                                                            *-»
                                                                            o
     SUMMARY TABLE OF MAJOR CHARACTERISTICS OF


POWERPLANTS CONSIDERED IN ENGINE CONFIGURATION STUDY
CO

£

-------
                         1-11
      1.    V-8 Gasoline Engine

            8 cylinders 97 mm x 76 mm (3.82" x 3.00")
                                 4.5 litre (275 CID)
            96 k'w (128 bhp) at 66.7 rev/s
            290  N.m  (210 Ib.ft) at 41.6 rev/s
            250  kg (550 Ib) estimated weight

For the primary emissions target (1.5 g/mile NOX) the engine would
require sophisticated close tolerance carburettors,  modulated exhaust
gas recirculation, air injection into the exhaust manifold and an oxidation
catalyst in the exhaust system.

For the secondary emissions target (0.4 g/mile NOX>  a reducing
catalyst and a further catalytic device 'getter box1  or  some similar
system would be needed.

In primary emissions build, fuel consumption would be approximately
18 1/100 km (13 mpg)  over the CVS-CH cycle while the need to run
closer to stoichiometric in the secondary emissions build would reduce
this to 19 1/100 km (12.5 mpg).

Drive-by noise level would be 74 dBA under US Federal Test conditions.

      2.    In-Line Gasoline Engine

            6 cylinders 88 mm x 82 mm (3.46" x 3.i'.3")
                                 3 litre (183 CID)
            96 kW (128 bhp) at 83.3 rev/s
            236  N.m  (171 Ib.ft) at 50 rev/s
            186  kg (410 Ib) estimated weight

For the primary emissions targets the engine would be equipped with
gasoline injection, modulated exhaust gas recirculation, air injection
into the exhaust  manifold and an oxidation catalyst.

For the secondary emissions targets the addition of a reducing
catalyst and protective 'getter1 box will be required.

Fuel consumption over the CVS-CH" cycle should be 15.5 1/100 km (15 mpg)

Drive-by noise level will be 77 dBA.

-------
                           1 - 12
       3.   Naturally Aspirated 4-stroke
            V-8 Indirect Injection piesel Engine

            8 cylinders 88 mm x 98 mm (3.46" x 3.86")
                                 4.78 litre (292 CID)
            96 kW (128 bhp) at 66.7 rev/s
            290 N.m (210 Ib.ft)  at 33.3 rev/s
            320 kg (700 Ib) estimated weight)

For the primary emissions targets the engine should only require
retardation of the injection timing to reduce NOX levels to 1.2 - 1.5
g/mile although some slight exhaust gas recirculation might be
necessary to ensure production compliance.

It is not thought that this type of engine could meet more stringent
NOX limits than 1.0 g/mile with conventional injection equipment
and current combustion systems.

Fuel consumption with this engine should be 11.5-10.5 1/100 km
(20-22 mpg) on the CVS-CH cycle.

Drive-by noise should be 76 dBA.

       4.   Boosted 4-stroke 6 cylinder
            Indirect Injection Diesel Engine

            6 cylinders - 90 mm x 100 mm (3.54" x 3.94")
                                   3.84 litre (234 CID)
            96 kW (128 bhp) at 60 rev/s
            290 N.m (210 Ib.ft)  at 33,3. rev/s
            327 kg (72CHb) (in-line-6) estimated weight
            309 kg (680-lb) (V-6) estimated weight

This engine would be lightly boosted by a turbocharger to attain the
above performances.  In-line and V configurations were schemed.

For the primary emissions targets, only injection retard should be
required although some slight exhaust gas recirculation might be
needed to ensure production compliance.

As for the naturally aspirated engine, no secondary emissions
configuration was schemed.

The target vehicle, equipped with this  engine,, should achieve 12-11
1/100 r.,n (19-21 mpg) during the CVS-CH cycle, the poorer part load
consumption of the turbocharged engine causing this penalty when
compared with the naturally aspirated engine.

-------
                              1 -r 13

Drive-by noise level should be 75 dBA.

       5.    Comprex on 4-stroke 6-cylinder
            Indirect Injection Diesel Engine

            6 cylinders - 90 mm x  100 mm (3.5-i" x 3.94")
                                  3.84 litre (234 CID)
            96 kW (128 bhp) at 60 rev/s
            290 N.m  (210 Ib.ft) at 33.3 rev/s
            327 kg (720 Ib)estimated weight

This engine, only schemed in in-line form due to manifolding problems
with the pressure  exchangers, would  employ the 'Comprex' to provide
a slight improvement in torque over more of the speed range than with
a turbocharger and to give a more driveable engine.  It is considered
that suitable sound insulation couli be fitted to the 'Comprex'.

The emissions of the engine will be the same as  for the turbo-charged
version and fuel consumption and drive-by noise should be similar at
12 - 11 1/100 km (19 - 21 mpg) and 75 dBA.

       6.    Boosted 4-stroke 6-cylinder Direct
            Injection Diesel Engine	

            6 cylinders - 93 mm x 94  mm (3.66" x 3.70")
                                  3.84 litre (234 CID)
            96 kW (128 bhp)  at 60 rev/s
            290 N.m (210 Ib.ft) at 33.3 rev/s
            310 kg (680 Ib) (In-line 6) estimated weight
            300 kg (660 Ib) (V-6) estimated weight

It was considered  that only a boosted direct injection engine would
have adequate power for a reasonable  weight and volume and would
allow sufficient injection timing retard for emissions control.
The engine schemed out would only be  capable of meeting  NOX limits
of 2.5 g/mile and would be retarded by 10° crank and have 10%
exhaust gas recirculation even to achieve this.   HC levels at these
retarded timings would be very high and an efficient oxidising catalyst
has been assumed to allow the 0.41 g/mile target to be met.  (Such a
catalyst does not exist at present.)

Predicted fuel consumption is 11 - 10 1/100 km (21 - 23 mpg) on the
CVS-CH cycle and predicted drive-by  noise level is  82 dBA.

-------
                             1-14
      8.    Loop Scavenged 2-stroke 6-cylinder
            Indirect Injection Diesel Engine
            6 cylinders - 99 mm x 114 mm (3.89" x 4.50")
                                  5.27 litre (321 CID)
            96 kV  (128 bhp) at 45 rev/s
            400  N.m (290 Ib.ft) at 33.3 rev/s
            340  kg (760 Ib) estimated weight

For this engine a 90  V-6 configuration  was chosen.

It was considered  that the primary emissions NOX level of 1.5 g/mile
could be met, but  there is little doubt that the HC target of 0.41 g/mile
woqld be unattainable with present combustion systems and fuel
injection equipment while the levels are likely to be so high that
no current catalyst is known to be capable of ensuring compliance.

Predicted fuel consumption is 13 - 11.5 1/100 km (18 - 20 mpg) arid
the predicted drive-by noise level is 75 dBA.

      9.    Uniflow 2-stroke 6-cylinder
            Direct Injection Diesel Engine
            6 cylinders -  83  mm x 114 mm (3.28"  x 4. 50" )
                                    3.74 litre (228 CID)
            96 kW  (128 bhp) at 45 rev/s
            414  N.m (300 Ib.ft) at 33.3 rev/s
            365  kg (800 Ib) (In-line 6)  estimated weight
            354  kg (780 Ib) (V-6) estimated weight

Two configurations of this engine were schemed out,  an in-line 6
cylinder and a 90  V-6.

With this form of combustion system it  is unlikely that the NOX targets
can be met and there is some doubt as to whether the HC levels will
be below the target without the  use of a  catalyst. These predictions
assume current technology.

This engine should allow the target vehicle to achieve 12.5 - 11.5 1/100 km
(19-21 mpg) over the CVS-CH cycle,  and drive-by  noise level  should be
75 dBA under U.S. Federal Test conditions.

-------
                           1-15
      10.    Compound 4-stroke 4 cylinder Direct
            Injection Diesel Engine	

            4 cylinders - 93 mm x  93 mm (3. 66" x 3. 66")
                                   2.52 litre (154 CID)
            Turbine and compressor directly geared to crankshaft
            96 kW (128 bhp)  at 50 rev/s
            318  N.m (230 Ib.ft)  at 50 rev/s
            305  kg  (670 Ib) estimated weight

For light duty application  the directly geared turbo-compound
configuration was chosen  since it gives  good part load fuel
consumption and is the most suitable construction for  mass-
produced passenger car application.  The torque curve is such
that a new transmission system is required since the torque is
at maximum at full speed full load and falls at all conditions
below this.  The high rating will prevent the primary NOx target
of 1.5 g/mile being achieved while  the need for  low compression
ratios and the very wide range of fuel quantities to be  burnt in
the combustion system will almost  certainly give a very high
hydrocarbon level.   If the engine were produced and if a suitable
transmission could be used,  fuel consumption might be 11.2 -
10.2 1/100 km (21 - 23 mpg) on the CVS-CH cycle.
Drive-by noise level could be as  low as 75 dBA.

      11.    2-staqe 2 bank Rotary Diesel Engine

            2-stage, 2 rotors - 5.35 litre  (326 CID)  L.P.  rotor
            96 kW (128 bhp)  at 66.7 rev/s  (output shaft)
            290  N.m (210 lb,ft)  at 25 rev/s
            227  kg  (500 Ib)  estimated weight

This engine would consist of two banks of a two-stage rotary engine.
The first L.P. stage would be 5.35 litre (326 CID) per lobe  (2:3
geometry) while the second stage would be  1.343 litre (82 CID) per
lobe.  The combustion system would be  of swirl chamber type.
The output shaft  would be  linked to the rotors to provide a 2:3 speed
reduction.

This engine would probably be able  to meet the NOX limit of 1.5 g/mile.
However, the difficulties  of achieving good combustion with this type
of engine imply that both HC  and  CO levels will be extremely high and
the target levels certainly would  not be  attainable without some major
improvement in  catalyst technology.

-------
                             1-16
Fuel consumption would be little better than with the gasoline engine over
the CVS-CH cycle and is predicted to be 16.4 - 14.7  1/100 km  (14-16 mpg).
There is some evidence that the engine would be quiet and a drive-by
noise level of 75 dBA is predicted.

Power Plant Rating

One of the major aims of the study was that a methodology should
be derived which would allow a quantitative  assessment of the
relative merits  of various power plants for light duty vehicle use.
Although the study was concerned only with a comparison of gasoline
and diesel configurations, the methodology was developed so that it
could be applied to any liquid hydrocarbon power plant and thus should
be of value in other studies.

The advantages of such methodology are that its application will allow
a direct quantitative rating of various power plants and that  it should
also be possible to identify those factors and aspects which render a
particular power plant suitable for a particular duty.   The second
advantage allows an  assessment of changes in a particular area as well
as high-lighting areas worthy of effort to make  a particular  configuration
more suitable for use in a given environment.

The fitness of a  power plant for light duty use can be assessed under
the following broad headings :-

     a)    Emissions
     b)    Package (Size, weight,  etc.)
     c)    Costs
     d)    Nature (Driveability)
     e)    Others (Convenience  and minor safety aspects)

These headings were too broad for a detailed measurement of different
power plants and a more detailed  list of performance aspects was drawn
up such that the various aspects would cover all facets of the operation
of light duty vehicle power plants  fuelled by  liquid hydrocarbons.
The individual performance aspects  were as follows  :-

            1.    Smoke
            2.    Particulates
            3.    Odour
            4.    NOX
            5.    HC
            6.    CO
            7.    S02
            8.    HC reactivity
            9.    Evaporative emissions

-------
                          1 - 17
            10.   Misc. emissions
            11.   Drive-by noise
            12.   Package volume
            13.   Package weight
            14.   Fuel economy
            15.   Fuel
            16.   Vehicle first cost
            17.   Maintenance cost
            18.   Startability
            19.   Hot driveability
            20.   Cold driveability
            21.   Torque rise
            22.   Durability
            23.   Coolant heat loss
            24.   Fire risk
            25.   Idle noise
            26.   Vibration and torque recoil
The performance aspects were generally those studied in the literature
survey although certain others were added so that the rating methodology
would be complete.

The final rating involved the assignment of weighting  factors to each
of the performance factors, followed by the rating of each candidate
engine under the same  heading.

The weighting factors were assigned by a group of eighteen Ricardo
engineers, each with experience of the emissions field and of the
American automotive situation, who were each instructed to act  in
isolation in assessing the factors.  The weighting factors, which are
listed in Table 2, are the arithmetic means of the values  assigned
by the members of the  group.

The rating  of the individual engines involved the devising  of a scale
which would cover both quantitative and qualitative assessments.

As expected,  some difficulty was experienced in  relating purely
subjective  impressions to a linear quantitative scale, but after some
consideration the following system was adopted as giving several easily
relatable,  subjective key points to the numerical scale,  the numbers
without definition being an interpolation of the surrounding merit
definitions.

-------
         1 - 18
 Merit Rating Scale




 0      Totally unacceptable




 1




 2      Bad




 3




 4      Poor




 5      Acceptable




 6




 7      Good




 8




 9      Best practical




10      Perfect

-------
  1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
              1 - 19
            TABLE 2
     Final Weightings Used in Study
Aspect                              Weighting
Smoke                              4.48
Particulates                         2.14
Odour                              4.48
NOX                                3.92
HC                                 3.99
CO                                 3.61
SO2                  .              3.48
HC reactivity                       1.83
Evaporative Emissions               1.60
Miscellaneous Emissions             0.98
Noise (Drive-by)                    6.32
Package volume                     2.61
Package weight                      2.59
Fuel economy                      12.20
Fuel cost                            5.40
Vehicle first cost                    4. 65
Maintenance cost                    4.35
Startability                          4.85
Hot driveability                     4.48
Cold driveability                    3.52
Torque rise                         1.98
Durability                          4.80
Heat loss                            2.18
Fire risk                            3.55
Idling noise                         3.83
Vibration  and torque recoil           2.18

-------
                            1-20
The rating system evolved allowed an immediate quantitative
assessment of the overall merit of the power plant which was
accomplished by multiplying each aspect  "rating1  by its
appropriate 'weighting1 and summing all  products.  With a
total weighting of 100 and a merit scale of 0 - 10 as above,
the maximum possible score is 1000.

A committee was used to  assign ratings to the various power
plants. The committee consisted of five experienced members
of the Ricardo staff and great care was taken to  ensure that the
committee had no bias to  either diesel or gasoline power plants.
The power plants considered were those described in the "engine
configuration" section of the report with the addition of the two
gasoline engines described briefly in the same section.

Each of the power plants was to be considered for the primary
and secondary emission levels of the study, i.e.

            Primary Targets

                  HC                 0=41 g/mile
                  CO                  3.4  g/mile
                  NOX                  1.5  g/mile

            Secondary Targets

                  HC                  0.41 g/mile
                  CO                  3.4  g/mile
                  NOX                 0.4 g/mile

but since it was considered that none of the diesel power plants
could possibly meet the secondary targets (in particular the 0.4
g/mile NOX figure) the rating system was only applied to an
environment embracing the primary targets <,

The ratings for each performance aspect are given in Table 3.
The designation  (O) indicates that the particular engine is completely
unable to meet the particular target, which might be considered to
be a disqualification for light duty use.

-------
          1-21
      TABLE 3
Final Ratings for Each Aspect






Aspect

1 Smoke
2 Particulates

3 Odour
4 NOX
5 HC
6 CO
7 S02
8 HC reactivity
9 Evap. emissions

10 Misc. emissions
11 Noise (drive by)
12 Package volume
13 Package weight
14 Fuel economy
15 Fuel cost
16 Vehicle first cost
17 Maintenance cost
18 Startability
19 Hot driveability
20 Cold driveability
21 Torque rise
22 Durability
23 Heat loss
24 Fire risk
25 Idling noise
26 Torque recoil


o>
c
•iH
"o
w
CO
0
oo

9
7

7
5
5
5
7
5
5

5
7
7
6
5
5
6
5
6
7
5
7
6
5
5
8
8


_c
I-H
o
w
CO
t
1 — 1
CJ

8
7

7
5
5
5
7
5
5

5
6.5
8
7
5.5
5
7
6
6.5
7
6
7
5 = 5
5
5
8
8





5
I-H
oo

6
2

4
5
5
6
4
7
7

5
6
6
5
7.5
5
' 5
6
5
7
8
7
7
5
6
5
6


0

1— 1
Q
o
1— 1
o

4,5
2

4
5
5
5
4
7
7

5
6
5
5
7
5
5
6
5
6
8
7
6
5
6
5
5

mprex
0
O

Q
0
O
vO

5
2

4
5
5
5
4
7
7

5
5
4
5
7
5
5
6
5
i
8
7
6
5
6
5
5


^

3
i — i
o

:4.5
2

3
(0)
(0)
5
4
7
7

5
4
5
5
8
5
5
5
5
6
8
7
6
5
6
3
5

X
CD
!H
a
0
u

3
i-H
O

5
2

3
(0)
(0)
5
4
7
7

5
3
4
5
8
5
5
5
5
7
8
7
6
5
6
3
5

O
(H
Crt
iv
3
CM .
i
2 |
2

2
5
(0)
(0)
4 I
7
7
c
5
6
8
7
5
5
4
2
2 i
7 i
8
7 s
2 |
4 !
6
5
7

-------
                           1-22
 The product of the weighting and rating for each performance aspect
 was summed up for each power plant and the results of this operation
 are shown in Table 4  : -

                         TABLE 4
                  Power Plant                     Final Score
                                             (Rounded to nearest
                                                 whole number)
1
2
3
4
5
6
7
8
9
10
11
V-8 gasoline
6 cylinder gasoline
V-8 IDI
6 cylinder IDI T/C
6 cylinder IDI Comprex
6 cylinder DI T/C
6 cylinder DI Comprex
2-stroke loop scavenge
2-stroke uniflow
Compound
2 stage rotary
608
620
587
556
554
(500)
(497)
(516)
(515)
(465)
(434)
In order to establish the validity of the  "committee method1  of
rating two re-runs were done for the gasoline engines and the
IDI diesels .    The final  runs for these power plants were found
to be within - 2-J% of the above figures and the relative order of
scores was not found to change.  Because of this the final scores
from the first complete run shown in Table  4 were taken as
representative of the relative merit of the various power plants.

The rating methodology has the disadvantage that a rating of 0 in
any aspect can be  hidden by good ratings for other aspects,  and
the bracketed figures in Table 4 are for those power plants which
scored a 0 (i.e. totally unacceptable) in one or more performance
aspects.

The results indicate that the gasoline power plants are superior to
the diesel power plants and that only the indirect injection 4-stroke
diesels are viable for the duty considered.  All the other diesel
power plants are unacceptable due to the inability to score better
than zero on  one or more performance aspects  (generally emissions)

-------
                         1-23

With the weightings adopted, the gasoline engine's superiority for
passenger car use in America is demonstrated but the closeness
of the final scores indicates that only quite minor changes in the
weightings would bring the scores equal.  It is apparent that in an
emissions and fuel conscious environment there  are many applications
which would change the individual weighting systems sufficiently to
make the automotive diesel an attractive alternative to the gasoline
engine.

Without a breakthrough in the reduction of exhaust emissions of
diesel engines, no diesel power plant can meet the secondary
emissions target however.

-------

                                             N&:*%&

           	GASOLINE  ENGINE

           K:;:!:;:;:;:;::::^ DIESEL  ENGINE
OUTLINES  OF  A  96 kW V-8 NATURALLY  ASPIRATED  I.D.I.

    DIESEL  ENGINE  AND  A  CURRENT 51 (3OO CIDJ

         AMERICAN  V8  GASOLINE   ENGINE
                                                               IV* (/I
                                                               •s
                                                               22

-------
                         2  -  1
                 SECTION 2
         INDEX AND LIST OF FIGURES             Page

      Contents                                   1-1

1.    Executive Summary                        1-2
         Conclusions                             1-4
         Recommendations                        1-5

2.  .  Index and List  of Figures
         Index                                   2-1
         List of figures                          2-2

3.    General  Summary & Introduction           	
         Summary                               3-1
         Introduction                             3-4

4.    Literature Survey
         Summary                               4-1
         Introduction                             4-2
         Scope of survey                        4-3
         Performance aspects                    4-3
         Conclusions                             4-50
         Areas requiring future work            4-56

5.    Engine Configuration  Study
         Summary       .   .                     5-1
         Configuration study                      5-2
         C.VS-CH consumption $  emissions
             estimation                           5-49

6.    Power Plant Rating
         Summary                               6-1
         Performance aspects                    6-2
         Weighting factors                        6-3
         Rating scale                            6-5
         Results of rating                        6-7
         Final rating                             6-33
         Results                                 6-34
         Conclusions                             6-35

7.    Programme Plans                          7-1

8.    Appendix 1  - Keyword System             8-1

9.    Appendix 2  - List  of References            9-1

10.   Appendix 3  - Glossary of Terms           10-1

-------
                          2-2
                    LIST OF FIGURES
SECTION 1

Fig.  1-1     Outlines of a 96 kW V-8 Naturally Aspirated I.D.I.
            Diesel Engine and a Current 51  (300 CID)
            American V-8 Gasoline Engine.

SECTION 4

Fig.  4-1     Bar Chart of Institution and Conference Papers Surveyed

     4-2     Bar Chart of Journals Surveyed

    4-3     Study of European Diesel Vehicle Operators Contacted to
            Obtain User Experience Data

    4-4     2 litre  D.I.  Conversion
            Performance Comparison - D.I. and Comet V Builds

    4-5     Comparison  of  Full Load Characteristics  for Three
            Combustion Chambers on 2-litre  4-cylinder  Engine
    4-6     NO Emission and Performance Characteristics of
            D.I. and Swirl Chambers

    4-7     Effect of Exhaust Gas Recycle on Performance and
            Smoke of a Six Cylinder Swirl Chamber Engine over its
            Load Range at 28 rev/s (1700 rev/min)

    4-8     The Response of Naturally Aspirated Engine
            NOX Emissions to Load

    4-9     The Response of Turbocharged Engine
            NOX Emissions to Load

    4-10    Light Duty Diesel Vehicle Emission Levels - 1975 FTP (CVS-CH)

    4-11    Data on Current Small 4-Cylinder  High Speed Diesel
            Engines for Automotive Applications

    4-12    Comparison of Diesel and Gasoline Engine
            Weight/Swept Volume

    4-13    Diesel versus Gasoline Weight Analysis
            for 4 Cylinder Engines

    4-14    Power/Litre of Some Current Gasoline and Diesel Engines

-------
                           4-51
 Odour at full load can be minimised by combustion chamber development.
 The proposed reductions in smoke levels should alleviate this problem.

 The identification of several odorous components has been achieved
 but quantitative assessment has yet to be perfected.  The A.D. Little
 Odormeter may advance technology in this area by a significant step.
Gaseous Emissions
In general,  turbocharging increases NOx further by increasing the
charge temperature but it allows further retard for the same smoke
limit.

Exhaust gas recirculation is effective in reducing NOx levels (particularly
over the CVS-CH cycle) but durability has yet to be proved and it does
tend to increase smoke emissions.

Although water injection has the benefit  of reducing NOx without
significantly affecting engine performance, the logistics of the installation
and the problems of engine  durability make this measure unattractive.

Timing retard is undoubtedly the most effective single parameter for
the reduction of NOx and the fact that the smoke limited performance
of the DDI engine tends to deteriorate less with retard than the DI
gives it a major advantage in this field.

The limited  data available indicates that emission levels from 2-stroke
engines should be of the same order as from 4-stroke engines of
similar performance.

Heavy duty experience leads to the conclusion that the use of a
conventional direct injection chamber will increase both NOx and CO
levels while HC  levels might rise rapidly with retarded timings.  It
seems almost certain that a high speed (67 rev/s) (4000 rpm)
conventional, naturally aspirated direct injection engine would not
achieve the  primary emission levels due to its low smoke limited
performance at retarded timings.

For a naturally aspirated 4-stroke indirect injection engine it can be
predicted that 3.4 g/mile CO can be achieved; 0.41 g/mile HC could
be attained on prototype vehicles although this figure may not be
held in production ; and 1.5 g/mile NOx could just be obtained from
a prototype  current generation engine although some exhaust gas
recirculation may be necessary to allow a margin for production
compliance.

-------
                           4-52
Although 0.4 g/mile NOx has been achieved with a highly modified
prototype engine in a European type vehicle, it is extremely unlikely
that this figure would be achieved with a heavier vehicle and with a
higher power to weight ratio.

Any diesel powered vehicle would have less difficulty in achieving
the target objectives if both weight and power to weight ratio were
reduced.  A lighter, lower powered vehicle would also have improved
fuel economy.
Particulates
The early suggested Californian requirement would present a major
problem for all engines with a heterogeneous combustion system and
this is undoubtedly a problem area for the diesel engine.  However,
the true effect of particulates on health is unknown at the moment
and the problem could be solved by the addition of filter systems
although this move would carry a high cost penalty.

It is considered that work should be initiated into the distribution of
particulates from different types of engines and their true health
hazard determined before any legislation is finalised.  An understanding
of their formation within the engine might also be a useful tool for their
control and thus a fundamental investigation using experimental and
analytical techniques should also be started.
Noise
The drive-by noise levels of diesel powered vehicles are slightly higher
than those of gasoline powered vehicles but there is no reason why
light duty vehicles should not meet proposed noise legislation.

The idle noise is annoying to the by-stander as well as the driver with
present European vehicles.

There is no reason why diesel powered light duty vehicles should be
unacceptable to  either the driver or by-stander if sufficient attention
is paid to details of the engine  and vehicle construction.
Volume
For the same power output the diesel engine is likely to be greater in
volume than a highly rated gasoline engine, but the increase in volume
is unlikely to pose any major problems.

-------
                           2-3
Fig. 4-15   Vehicle Fuel Consumption versus Inertia Weight
           during LA4 1975 (CVS-CH) Test Cycle

    4-16   Tolerances Controlling Piston to Cylinder Head Clearance

    4-17   Effect of Inlet Valve Timing on Low Speed Torque

    4-18   Torque Comparison on a 300.HP D.I. Diesel Engine
    4-19   Heat Rejected to Coolant for Similar Diesel and
           Gasoline Engines over the Load and Speed Range

SECTION 5

    5-1    Estimated Torque Curve for 0 97 x 76 mm V-8
           Gasoline Engine in Low Emissions Build
    5-2    Estimated Fuel  Consumption Curves for  a097x76mm V-8
           Gasoline Engine in Low Emissions (1.5 g/mile NOX) build
    5-3    Estimated Torque Curve for a 088 x 82 mm 6 Cylinder
           "European Type" Gasoline Engine in Low Emissions Build

    5-4    Estimated Load Range Consumption Curves for a  6 Cylinder
           088 x 82 mm "European Type" Gasoline  Engine in 1.5 g/mile
           NOX build

    5-5    Estimated Performance Curve for Naturally Aspirated
           088 x 98 mm V-8 Comet V Engine

    5-6    Load Range Fuel Consumption Curves for Naturally Aspirated
           088 x 98 mm V-8 Comet V Engine

    5-7    N/A - V-8 - I.D.I. Diesel Engine Installation Drawing

    5-8    N/A - V-8 - I.D.I.'Diesel Engine
           Comparison of  Crossflow/Unisided Cylinder Heads

    5-9    N/A - V-8 - I.D.I. Diesel Engine Cross-Sectional Arrangement

    5-10   Estimated Performance Curve for Boosted Six Cylinder
           090 x 100 mm I.D.I.  Engine in Low Emissions Build
           (1.5 g/mile NOX in a 3500 Ib Passenger Car
    5-11   Estimated Load  Range Fuel Consumption  Curves for a
           Boosted  Six Cylinder 090 x 100 mm I.D.I. Engine in
           Low Emissions  Build (1.5 g/mile NOx) in a 3500 Ib
           Passenger Car

-------
                         2-4
Fig. 5-12   96 kW 6 Cylinder Turbocharged I.D.I. Diesel Engine
           Installation Drawing showing Comparison between
           In Line 6 / V-6.

    5-13   6 Cylinder Turbocharged I.D.I. Diesel Engine
           Cross and Longitudinal Arrangement Drawing

    5-14   Preliminary Layout of V-6 60 Bank Angle Turbocharged
           Diesel Engine with 6 Throw Crankshaft

    5-15   Preliminary Layout of V-6 Turbocharged Diesel Engine
           with 3 Throw Crankshaft

    5-16   V-6 Turbocharged  I.D.I. Diesel Engine
           Cross and Longitudinal Arrangement Drawings

    5-17   Estimated Performance Curves for Boosted 093 x 94 mm
           Six Cylinder D.I. in Minimum Emissions Build

    5-18   Estimated Load Range  Fuel Consumption Curves for a
           Boosted 093x94mm  Six Cylinder D.I. Engine
           in Minimum Emissions Build

    5-19   96 kW Six Cylinder Turbocharged D.I. Diesel Engine
           Installation Drawing showing Comparison between In-Line6/V-6

    5-20   6 Cylinder Turbocharged D.I. Diesel Engine
           Cross-Sectional Arrangement Drawing

    5-21   Estimated Performance Curve for 2 Stroke Loop Scavenge
           I.D.I. Diesel Engine

    5-22   96 kW Loop Scavenge - I.D.I. 2 Stroke Diesel Engine
           Installation Drawing

    5-23   V-6 - I.D.I. - 2 Stroke Loop Scavenge Diesel Engine
           Cross-Sectional Arrangement Drawing

    5-24   Estimated Performance Curve for 2 Stroke Through
           Scavenge D.I. Diesel  Engine

    5-25   96 kW 2 Stroke Through Scavenge Diesel Engine
           Installation Drawing

    5-26   6 Cylinder  2 Stroke   Through Scavenged  D*!. Diesel Engine
           Cross and Longitudinal Arrangement Drawing

-------
                             2-5

 Fig. 5-27   V-6  2 Stroke Through Scavenged D.I. Diesel Engine
            Cross and Longitudinal Arrangement Drawing

     5-28   Estimated Performance Curve for Four Stroke
            093 x 93 Compound D.I.  Engine

     5-29   96 kW Compound In-Line  4 Cylinder D.I. Diesel Engine
            Installation Drawing

     5-30   Compound In-Line 4 Cylinder D.I.  Diesel Engine
            Cross and Longitudinal Arrangement Drawing

     5-31   Estimated Performance Curve for Two Stage
            Rotary Diesel Engine

     5-32   96 kW 2 Stage/2 Bank I.D.I. Rotary Diesel Engine
            Installation Drawing
     5-33   2 Stage/2 Bank I.D.I. Rotary Diesel Engine
            Preliminary Cross Sectional Arrangement Drawing

     5-34   2 Stage/2 Bank I.D.I. Rotary Diesel Engine
            Preliminary Longitudinal  Arrangement Drawing
     5-35   Naturally Aspirated 96 kW  V-8 I.D.I. Engine
            Predicted HC ppm Levels in Low Emissions Build
     5-36   Naturally Aspirated 96 kW  V-8 I.D.I. Engine
            Predicted NOX ppm Levels in Low Emission Build
     5-37   Naturally Aspirated 96 kW  V-8 I.D.I. Engine
            Predicted HC ppm Levels using Current  Fuel
            Injection Equipment

     5-38   Naturally Aspirated 96 kW  V-8 I.D.I. Engine
            Predicted NOX ppm  Levels in  Optimum Performance Build

     5-39   Typical HC Emissions from a Lightly Boosted Conventional
            D.I.  Engine in Optimum Performance Build

     5-40   Typical NOX Emissions from a Lightly Boosted Conventional
            D.I.  Engine in Optimum Performance Build
     5-41   Typical HC Emissions from a Lightly Boosted Comet V
            Engine in Optimum Performance Build

     5-42   Typical NOX Emissions from a Lightly Boosted Comet V
            Engine in Optimum Performance Build

SECTION 6

Fig.  6-1     Light Duty Vehicle Power Plant Survey Results

-------
                            3-1
                         SECTION 3

          GENERAL SUMMARY AND INTRODUCTION
This report summarises the results of a study which was conducted
on behalf of the Office of Mobile Source Air Pollution Control of the
Environmental Protection Agency and which was aimed at identifying
the problem areas and estimating the effect of the wide use of the
diesel engine  as a light duty power plant in the United States.

The diesel engine is widely used in Europe for light duty vehicles and
for taxi cab services, roles in which its well known advantages in
respect of fuel economy and emissions are of great importance, and
in which the off-setting problems of weight, bulk, noise, cost and
possibly odour are accepted. The prospect of the high costs, greater
weight, increased bulk, and lower specific power of an emissions
controlled gasoline engine may  reduce the impact of some of these
problems while the undoubted fuel economy advantage of the diesel
engine may be accentuated in an emissions controlled United States
environment.

Since there is little United States experience  of light duty diesel
engines, the first stage of the study entailed the collection of related
data and a preliminary account  of the feasibility of the diesel as a
light duty power plant.  This first stage involved a study of published
and unpublished literature from technical journals and learned societies
throughout the world.  Visits were also made to existing diesel users
and manufacturers throughout Europe so that  operating and service
factors could be estimated realistically.  A careful assessment of the
literature and visits indicated that a diesel engined passenger  car should
be able to meet the proposed emission targets of 0.41 g/mile HC,
1.5 g/mile NOx and 3.4 g/mile CO although the secondary target of
0.4 g/mile NOX did not appear likely.  It appeared that the main problems
of the diesel engine might be odour, particulates, noise, bulk, cost and
starting.   In the areas of reliability,  driveability and economy there seemed
to be considerable advantages to the diesel engine.


The next stage of the study involved the preliminary design of potentially
viable diesel power plants.  These power plants had to be capable of propellint
a 1600 kg (3500 Ib) car with an acceleration potential equal to that of a
'standard' American car.  (Essentially they  were 97 kW (130  bhp) engines).

-------
                          3-2

The concepts studied included reciprocating and rotary diesel type engines
with and without turbocharging, compounding and low heat loss components.
Two gasoline power plants were also outlined for comparison purposes.
The configuration study covered the following contenders for the power
plant :-

         1.        8-cylinder 'American' gasoline
         2.        6-cylinder 'European1 gasoline
         3.        8-cylinder IDI diesel 4-stroke
         4.        6-cylinder IDI diesel 4-stroke turbocharged
         5.        6-cylinder IDI diesel 4-stroke with 'Comprex1
         6»        6-cylinder DI  diesel 4-stroke turbocharged
         7.        6-cylinder DI diesel  4-stroke with 'Comprex1
         8.        6-cylinder IDI diesel  2-stroke loop scavenged
         9.        6-cylinder DI diesel  2-stroke uniflow
        10.        4-cylinder compound diesel 4-stroke
        11.        2-stage, 2-bank rotary diesel


Drawings were prepared for all the diesel power plants and performance
curves and emissions levels were calculated.

The third stage of the study  was an assessment of the relative merits
and viability of the power plants for both the primary and secondary
targets.

In order to compare the various power plants a quantitative rating
method was devised.   This  involved an estimate of the relative
importance  of 26 performance aspects and assigning to them
numerical 'weightings'.  Assessment of the merit with which each power
plant  met an aspect allowed a numerical  'rating' to be assigned.
Summation of the weighted ratings allowed the power plants to be
compared numerically.

Use of this rating system on the power plants indicated that the gasoline
engines were slightly superior to the diesel engines for the primary
emissions targets and that only the 4-stroke IDI diesels were really
viable for the duty considered.

None  of the  diesel engines could be considered as being practical
power plants for the secondary emission targets.

The fourth and final stage of the study was an assessment of the
implications of the above results so that proposals for further
work  and research could be made if it seemed desirable.

-------
                              3-3
It was concluded that :-

     a)       In view of the large extrapolations required
              from current European practice an 8-cylinder
              IDI diesel 4-stroke engine should be built and
              demonstrated.

     b)       A 6-cylinder IDI diesel should be built and
              work on turbochargers and 'Comprex1
              pressure exchangers should be carried out.

     c)       Work should be pursued on advanced and
              potentially cheaper fuel injection  systems.

     d)       The low compression ratio, ignition assisted
              IDI engine should be investigated  thoroughly.

     e)       A fundamental investigation into the formation
              of hydrocarbons  in diesel engines should be
              carried out together with an attempt to produce
              catalysts with a lower light-off temperature.

     f)        A control system for the  modulation of exhaust
              gas recirculation should be developed.

     g)       A quick warm up starting aid should be developed.

     h)       Diesel engines should be  demonstrated in taxi cab
              and light delivery truck service to encourage a
              rapid switch to diesel power for these duties.

     i)        A study into particulate formation in diesel exhaust
              together with the development of soot and particulate
              filters is desirable.

-------
                             3-4
 Introduction

 There is little doubt that the low emissions gasoline engine offers
 the prospect of higher cost, lower specific power and high specific
 bulk and weight while the increasing cost of fuel renders higher
 efficient alternative power plants  more attractive.


 The diesel engine is demonstrably the most efficient practical
 prime-mover produced today and  it is known to have favourable
 emissions characteristics.  The most efficient diesel engines
 are not used in light duty vehicles because of their bulk and
 limited speed range while those diesel engines currently used
 in light duty vehicles are smaller than would be required for an
 American passenger car and have not yet been developed to meet
 very stringent emissions targets*

 In Europe and Japan small high speed diesel engines are used widely
 for light duty vehicles and taxi cab service while in some countries
 there is also a minority of diesel  powered passenger cars.  Thus
 there is already a great mass of experience of light duty diesel
 vehicles in most roles.

 The present American environment, however, cannot be related
 directly to European and  Japanese experience. The diesel engined
 passenger cars used in Europe and Japan have been of smaller size
 than typical American cars and have also had lower power to weight
 ratios.  Current and proposed American emission  standards are
 different from those obtaining in the rest of the  world so that there
has not,  as  yet, been any attempt  to run existing light duty diesel
engines in service in low emissions build.

Consideration of the diesel engine for an American passenger car
immediately reveals that it will be considerably  larger and will
spend most of its life running under different  conditions than those
currently found elsewhere. It was these considerations which led to
the objectives of this study, i.e. an examination of the feasibility
of using the diesel engine as a power plant for an American passenger
car, an examination of potential problem areas and a study of the
trade-offs possible within these areas. The determination of the
possibilities of finding solutions to any outstanding technical  problems,
and finally to put forward recommendations for the future increase in
light duty diesel engines (if any be thought desirable) and to  indicate
the best routes to achieve this increase.

-------
                           4-1
                        SECTION 4

          LITERATURE SURVEY, ASSESSMENT OF
             PROBLEM AREAS AND TRADE OFFS
This section contains  a digest of the articles  considered in a
survey  of light,  duty  (and appropriate heavy duty) diesel
literature and the conclusions obtained from visits to manufact-
urers and  users of light duty diesel engines in  Europe.  The
published literature covered the period from  1919 to the present
although most of the information came from the period from
1950 to the present.  Where possible and pertinent the  published
information was augmented  by  technical data from Ricardo in-
house reports»

Study of the literature, which was conducted under the  headings
of 'Performance Aspects',  allowed problem areas and potential
trade-offs  to be identified as well  as indicating some of the
possible gains  from the  use of light duty  diesel  engines in  America.

The major  conclusions were that there were no major  problem
areas in the use of light duty diesel engines in America.   Smoke,
odour, participates,  noise,  volume,  weight,  cost,  starting  and
refinement would be  inferior to existing gasoline engines but  there
would be advantages  in respect of  emissions, fuel economy,
maintenance, driveability,  durability and overall economics.

Emissions predictions indicated that  the diesel engined American
passenger car  might just achieve  the primary emissions targets
of HC  - Oo41 g/miie.  CO » 3o4 g/mile, NOx -  1.5  g/mile with-
out catalysts, but that the  secondary emissions targets of 0.4
g/mile NOx would  be practically impossible to achieve.

-------
                          4-2
Introduction
Although diesel engined light duty vehicles and taxi  cars are
widely used in Europe  and Japan, all the  current service
experience undoubtedly relates  to engines  which are smaller and
which run  under different load  conditions  from those which  would
be applicable  to  present  and proposed American light  duty con-
ditions.  Current service experience moreover is with a gener-
ation of engines  which were not  emissions biased although much
work has  been carried out recently to determine the emissions
characteristics of these  engines.

It was these conditions which indicated that a study of published
literature, Ricardo in-house reports and  data,  and discussion with
current light duty diesel manufacturers and users would yield
much useful information  on the likely feasibility of the diesel
engine for American light duty use  and which would also indicate
the problem areas and potential  trade-offs if diesel engines were
adopted for American use.

Although defined in  the configuration study, the goals  of the com-
plete study had to be kept in mind throughout this  literature study.

The vehicle and  application considered was a  1600  kg  (3500 Ib)
4/5 seat sedan capable of achieving 0-97 km/h  (0-60 mph)
in 13.5 s, 32 -  113  km/h (20  - 70 mph)  in 15 s,  and the  D.O.T.
high speed pass  manoeuvre in  15 s so that an installed power of
about 96 kW  (128 bhp) was required.

The emissions  targets  were that the vehicle should  achieve  the
following figures measured by  the CVS-CH test procedure.

           a)   Primary Targets

               HC         0.41 g/mile
               CO         3.4  g/mile
               NOx        1.5  g/mile

           b)   Secondary Targets

               as above but

               NOx        0.4  g/mile

-------
                           4-3
Scope of the Survey
Since its first application, much has been written about the diesel
engine and a  considerable proportion of the  published  work has
dealt with light duty  engines.   For the survey the period 1919 to
the present day was  selected as  appropriate  although the majority
of the information  was uncovered for  the period  1950  to the pre-
sent.  For this period all published and unpublished reports and
translations were checked and  Figs. 4-1  and 4-2 show the  scope
of the survey.  For convenience  in analysis the light duty diesel
engine was considered under various performance aspect para-
meters,  as  shown  in Table I below:-
                          TABLE  I

                    Performance  Aspects

               1.                 Smoke
               2.                 Odour
               3.                 Gaseous Emissions
               4.                 Particulates
               5.                 Noise
               6.                 Volume
               7.                 Weight
               8.                 Fuel  Economy
               9.                 Fuel
               10.                First Cost
               11.                Maintenance
               12.                Starting
               13.                Hot Driveability
               14.                Cold Driveability
               15.                Torque  Rise
               16.                Durability
               17.                Coolant Heat Losses
               18.                Fire  Risk
               19.                Vibration & Torque Recoil
               20.                Idle Noise
               21.                Reliability
               22.                Economics
               23.                Manufacture
               24.                Performance
               25.                Design
               26.                Ancillaries
               27.                Lubrication

-------
                           4-4
In order to assist in the identification of  major conclusions and
problem areas, a keyword system  was used to classify the items
of information uncovered in the literature search  and this  keyword
system, a three level  one,  is  described  in  Appendix 1  (section 8).

The literature  search covered some 700 items  and although it  is
not possible to refer to them all in detail,  a complete  list may be
found in Appendix 2  (section 9).

In addition to the literature study,  several  European companies
involved with the  day to  day operation of  diesel powered light  duty
vehicles were  contacted to obtain a unbiased opinion of the true
operational  characteristics of diesel passenger cars  in  current
European  conditions.  The companies contacted are listed in Fig.
4-3.
Exhaust Smoke
Any  suggestion  to make wider  use of the  diesel engine  may come
under public attack  because of the diesel's anti-social reputation in
this  respect.   It is important to realise however  that black smoke
is not  an inherent characteristic of a diesel engine but only of one
that  is either ill-maintained or is  overloaded.

Apart from Thermal or Mechanical Loading limits,  which are  a
function  of engine design, it  is the onset  of excessive black smoke
in the  exhaust which limits diesel  engine  output.   Smoke limitations
are therefore output limitations.
Black Smoke
Exhaust smoke may be of one of two forms, either black or  blue/
white.  Black smoke occurs when there is insufficient oxygen for
complete  combustion of the fuel and since in the  diesel  engine com-
bustion is heterogeneous, it can occur even  though there is an
overall excess of oxygen in the charge.

-------
                           4-5
When a fuel drop is heated in a combustion chamber, complete
oxidation of the carbon and hydrogen occurs provided there is a
plentiful supply of oxygen and provided further that there is time
for combustion to be completed before the reactions  are quenched
by expansion of the gases.  If there is a shortage of oxygen however,
the hydrogen preferentially takes it quota, leaving too little oxygen
for complete combustion and under this condition carbon particles
are formed which appear as black smoke in the exhaust (19).

The exact chemical reactions leading to the formation of smoke are
however unknown although they may involve €2 or €3 fragments of
hydrocarbon decomposition as a basic building brick leading to a
final polybenzenoid structure for the final smoke particle.  It may
even be that there are a number of alternative paths  leading to such
formation.

Soot production  takes place during the early part of the combustion
process of all diesel engines but under conditions in  which the
exhaust smoke emission is low,  the soot is consumed during  the
later parts (9).

The formation of soot particles is a characteristic of diffusion flames
such as are typical of combustion in a diesel engine and photographs
of such combustion exhibit intense white or yellow radiation of
incandescent particles under all engine conditions. There is  a
dramatic difference in this respect from premixed combustion in a
gasoline engine,  which  exhibits  a much lower degree of luminosity.

At high fuel inputs,  with a consequent smoky exhaust, the radiation
persists until cooling takes place during expansion which leads to
quenching of the combustion products and the emission of black
exhaust smoke (38).
Blue/White Smoke
The smoke emission from diesel engines does not consist uniquely of
black smoke but at times may include white or blue/white smoke
resulting from misfire,  the colour of the smoke depending on the
size of the droplets.

This smoke is produced by fuel droplets in  abundant oxygen which do
not reach ignition temperature and which in a partially vaporised form

-------
                            4-6
pass out of the exhaust as a cloud.  The problem can occur during
the winter when the engine is first started after a long soak at low
temperature and on some engines the smoke can persist for a
considerable time after start-up.  The most general solution is an
increase in compression ratio but at very low temperatures even
this may not be a complete solution (19).  Other solutions are the
use of a higher Cetane Number fuel or by decreasing the mid boiling
point  of the fuel (38) .

Blue smoke can also result from excessive lubricating oil consumption
(573) and while this may be the result of wear due to high mileage,
problems may arise early in the engine life due to unsatisfactory
running in of the piston rings.
Combustion Chamber Type and Black Smoke
Diesel engine combustion chambers may be classified into one of three
types. The first of these, and the one most extensively used for truck
and larger engines, employs direct injection, where the fuel is
injected directly into the space above the piston.  Prechamber engines
employ injection into a flask shaped chamber into which part of the
air charge has been compressed, with a restriction between the pre-
chamber and the main chamber.  Swirl  chamber engines are a special
form of prechamber engine in which the entrance to the prechamber  is
so arranged as to give an intense swirling motion to the air charge in
the prechamber; a larger portion of the air charge is  compressed into
the swirl chamber than would be the case with a prechamber,  50% as
against 25% or so, and the restriction between the swirl and main
chamber is less pronounced than would  be the case in a prechamber
engine.

The three types of combustion chamber  have different smoke character-
istics.  Direct injection engines have a  relatively slow onset of smoke
as load is increased and  furthermore have a lower  effective air
utilisation.  As a result, with low smoke limits, the direct injection
engine must be more severely derated than a  swirl chamber engine
(Fig.  4-4).  Prechamber engines on the other hand, while they have
similar smoke levels to swirl chamber  engines at full load, (Fig.4-5)
(20,695)  can give plateau' smoke as the load is  reduced,  i.e.  the
smoke does not fall as  low as it does on a swirl chamber  or  with
direct injection (16,  18, 20).

-------
                           4-7
Fuel/air mixing is a vital factor affecting smoke production.  On direct
injection engines, both increased rate of injection and advance in
injection timing reduce the amount of exhaust smoke (110) .  Air swirl
rate is also important and it is essential to optimise swirl and rate of
injection (Fig. 4-6).

Indirect injection engines are less sensitive to injection rate but have
a reverse characteristic in so far as injection timing is concerned,
retardation of timing giving reduced smoke output. The insensitivity
of injection rate is presumably due  to the mixing being controlled by
high air velocities rather than by jet mixing.
Fuel Type and Black Smoke
While prechamber engines tend to be less exacting in their fuel require-
ments than direct injection engines, increasing fuel volatility in all
engines helps to reduce black (and indeed white) smoke.  While, as
has already been mentioned, increasing the Cetane Number helps
reduce white smoke, it should not be increased too far or increased
black smoke results (294).

Improvements in exhaust smoke can be obtained by fumigating, all (543)
or part (568) of the fuel with the induction air. Ricardo results- siagjgest
however that this can lead to blue smoke in the exhaust at part load
unless the fumigated fuel be cut off under these condition®.

One minor but interesting smoke problem was reported! during: tfee
Vehicle User Survey by the British Petroleum Company, who mentioned
that vehicles with fuel pump maximum fuel stop settings set to- eomply
with smoke requirements on continental European fuel with a relative
density of 0.825 - 0.835, had a smoke problem in the United Kingdom
where the average relative density of fuel is 0.84.  This could well
cause difficulties in other locations.

Dual fuel engines with gaseous main fuel's, ignited by injecting a small
charge of diesel fuel,  can also give a cleaner exhaust (2l) (61) r but
emission tests on such dual fuel engines show high levels of hydrocarbons
in the exhaust.

-------
                           4-8
Anti-Smoke Fuel Additives and Hang-on Devices
There are numerous references to tests with alkaline (usually barium)
based fuel additives.  All agree that large reductions in smoke levels
are obtained but most long term  tests show that combustion chamber
deposits are increased by the use of these additives (7, 9, 76, 109,
292, 293, 294, 80), since the resulting fuels have a higher ash content
than untreated fuel  (19) .  There is also concern as to possible health
hazards with metallic additives in the fuel.

It is clear that anti-smoke additives do affect the combustion process
and do not merely act as soot dispersants since there is a reduction
of mass soot emission as well as a reduction in invisible smoke  (9).

Tests with exhaust catalysts have resulted in slight changes in exhaust
conditions, there being  a small improvement in the smoke levels at
the  full load/speed  condition in one test engine  (107) .  Very little
work with such devices  has however been reported in the literature.
Smoke in Turbocharged Engines
With the levels of turbocharging likely to be applied to a passenger car
engine, the turbocharger will be matched with the engine  at the peak
torque condition.  Hence with any acceptable torque back-up curve
the exhaust will  be clean under any steady state condition at this and
higher speeds.  At lower speeds however the torque will be limited
by the onset of exhaust smoke and in addition, with turbocharging'
there could be a transient smoke problem over most of the speed range
unless a boost sensor be fitted to the  maximum fuel stop.

In this area,  the Comprex may have decided advantages over
conventional turbochargers but more  experience is necessary before it
is possible to guarantee that smoke would be eliminated under starting
and transient conditions.
Smoke Measurement
Smoke measurement is not the simple problem which it appears to be
at first sight as  may be seen from the large number of attempts that
have been made  to design and develop smoke meters.  Only a few have

-------
                            4-9
survived as practical instruments and these may be classified as
follows: -

a.     Sampling light-obscuration meters (Volvo, Hartridge,
                                          DDR.RD M.4)

b.     Sampling filter soiling meters (Bosch, Von Brand)

c.     Full flow obscuration meters (UTAC, USPHS, Beckman,
                                      Celesco)

d.     Sampling, ligh-dispersant meters (Bosch, Tyndall)
As has been shown, smoke emission from diesels does not consist
uniquely of black carbon particles but at times includes white smoke.
Light obscuration meters indicate such white smoke as increasing
the apparent density whereas filter soiling meters,  where the soiling
is measured by light reflection, indicate reduced density.  Thus the
direct correlation betweeen these meters, which should in theory be
possible with black carbon particles, breaks down.
Smoke Legislation
Current United States smoke legislation for truck diesel engines
limits the average visible smoke from the vehicle to 20% opacity
during acceleration and 15% during the lug mode of the Federal smoke
cycle with a 50% limit to the peak cycle.  While these levels may be
acceptable for a limited number of trucks, many of which spend the
greater part of their  time on highways and not in cities, Ricardo think
that such levels from diesel engined light duty vehicles would be
unacceptable and would suggest a level  of approximately 5-7% opacity
at the rated speed and 8-10% at peak torque speed.

These proposed levels are in fact in line with those laid down in Britain
for this size of engine, and it is Ricardo's belief that current European
diesel passenger cars could comply with these limits.

Typical smoke legislation for countries other than the United States is
as follows: -

In Britain control of diesel engine smoke in road vehicles is covered
by British Standard AU 141a:1971. This requires a 100 hour type test

-------
                            4-10
for each basic engine type for certification.  The AU 141a smoke
limit is in the form of a curve which defines the smoke limit at
different levels  according to the engine output;  the higher the
output the lower the smoke limit.  The limits apply to speeds at
full load from 100% down to 45% of the maximum  speed, or  16.7 rev/s
 (1000 rev/min) whichever is  the higher,  smoke density being mea-
sured by a light obscuration method.

The European Economic Community (EEC) have produced a smoke
test code for diesel engined road vehicles (EEC test code 72/306/EEC).
This includes steady state testing over the same road/speed range as
the British Standard and with limits similar to but slightly higher than
those of BS AU 141a. In addition a free acceleration test procedure
is included with a smoke limit set slightly above the limit of the steady
state tests, so that the transient response can be checked and also
road-side tests  carried out on vehicles. In this test code, smoke
measurement is by opacity smoke meter.

Czechoslovakia
The test is a filter soiling type using meter type NC 112, allowing 40%
discoloration of the filter in new vehicles and 50% discoloration in other
vehicles.

Finland
A full load acceleration test, in gear, is specified using a filter soiling
smoke meter.

E. Germany
Steady state and free acceleration tests are under investigation using
RD M-4 smokemeters.

Japan
A steady state, full  load test at 40, 60 and 100% rated speed with smoke
measurement by filter soiling smoke meter and a limit of 50% discolor-
ation at each test condition.

Jugoslavia
Exhaust density must not exceed No.  2 Ringelmann.

Mexi co
Smoke emission not to exceed No. 2 Ringelmann for more than 10 seconds.

-------
                               4-11
South Africa
A free acceleration test with Hartridge meter is specified, with a
smoke limit of 70 HSU (Hartridge Smoke Units).

Spain
Free acceleration limit of 65 HSU operates from 1970 in Madrid
only.

Sweden
The test involves a loaded vehicle in gear, limiting the maximum
speed to 45 - 50 km/h.  Full load acceleration is tested using either
Bosch or Hartridge meters.  Limits for passenger vehicles for more
than 30 persons are 30 HSU or 2.5 Bosch. The limit for other vehicles
is 45 HSU or 3.5 Bosch.

Switzerland
Full  load steady state using Bosch meter and free acceleration test
using Bosch integrating smokemeter.  For steady state test, limits
range from 6 Bosch for up to 3 litre engines  to 4.5 Bosch above 8 litres.
This applies up to 600 m altitude, 0.5  Bosch  being allowed additionally
for each 400 m above this.
Odour
Exhaust odour, like smoke, is a social problem of the diesel engine
but is much more difficult to quantify.   Nevertheless, a considerable
amount of work has been carried out in this area, mainly using
subjective testing with 'sniff panels' but also including chromatography
in association with subjective testing.

Early work indicated that exhaust odour  was related to aldehyde concen-
tration but that it was dependent on the sulphur4 content of the fuel (6).
Later work by A.D. Little has suggested however that only two groups
of components are important in so far as diesel odour is concerned.
One group is composed of aromatics such as alkyl benzenes,  indenes,
indans, tetralins and naphthalenes.  The other group of components
is composed of partial oxidation products of components found in the
aromatic fraction of the diesel fuel.  This work also suggests that the
sulphur content of the fuel is not important (14).

Peak odour intensities  have been found to occur at no load and full load
while engine speed is not an important parameter (652). Comparisons

-------
                              4-12
between two stroke and four stroke diesel engines gave much higher
odour levels from two stroke engines. Later model two stroke engines
with modified low sac volume fuel injectors had however similar
odour characteristics to equivalent four stroke engines (653) .

The high odour levels at no load,  which can be accentuated at high
speeds by misfire, can be minimised by  advancing the injection timing
at light loads to prevent misfire conditions. Steps taken to reduce
smoke at full load by limiting the engine  load will similarly reduce
full load odour.

Major changes in fuel constitution, e.g.  the use of a fuel with very
low aromatic content,  have been reported to give changes in odour
characteristics.  The use of such a fuel also gives a lower intensity
of odour as compared with a normal  No. 1 or No. 2 diesel fuel (655) .

It has been stated that the levels of exhaust odour found in a Mercedes
diesel car are not low enough to prevent a significant amount of
objectionability (654), but so  far as  Ricardo  are aware,
exhaust odour from light duty engines has not been a serious cause
of complaint in Europe even in taxi service where the density of such
vehicles can be quite high in congested urban areas.  While it is very
difficult to estimate the likely effect of a large increase in the
density of such diesel vehicles, it would seem reasonable to assume
that densities in American cities up to those found in Europe should
not cause offence.

Experimental work has been carried out  with catalytic exhaust mufflers
(653).  These have been  shown to reduce odour under idle and acceleration
conditions but have no effect on other loads. Such mufflers are believed
to have been used on some bus fleets.

As diesel fuel does not readily evaporate, fuel odours can be detected in
diesel engined vehicles as a result of bad housekeeping.  It is essential
to make sure that there are no leaks from the fuel system and there
is no spillage when refilling the fuel tank. With these precautions such
fuel odours should not be detectable.
Gaseous Emissions
The three major gaseous pollutants are NOx, CO and HC.  NOx is largely
formed as NO but also contains small amounts of NO2-  Only the NO is

-------
                            4-13
measured if the exhaust is metred by means of a non-dispersive
infra-red analyser „  HC will contain a very wide spectrum of
hydrocarbons of all types, some of which will be of similar form
to those found in the raw fuel while others will have partially
oxidised or otherwise changed in the engine.  They arise from
incomplete combustion of the fuel.

The oxides of nitrogen are formed from the constituents in the air
although it has been suggested that dissolved  nitrogen in the fuel
can play an important part.  The oxides of nitrogen are formed
under high temperature and pressure conditions during the
combustion part of the cycle.  The rate of formation  is highly
temperature dependent and anything that can be done to reduce the
peak cycle temperatures will reduce NOx levels. Unfortunately
it is difficult to do this without loss of efficiency since this is also
a function of peak  cycle temperature.

The carbon monoxide levels produced by the diesel are very low, at
least by gasoline engine standards, since the flame conditions do not
favour this product of partial combustion under normal operating
conditions.  It is only when excessive fuel is  injected and the
controlled pattern of mixing and combustion breaks down that carbon
monoxide emissions rise to significant levels.  This  can occur,  although
normally at fuels air ratios above the limits set by legislative smoke
requirements °  Thus  within these smoke limits, control of CO is not
a problem and for example CO levels from  a  well developed pre-
chamber engine are less than 5% of those emitted by  non-emission
controlled gasoline engines.

As has been explained, the hydrocarbons range from pure fuel through
to complex aldehydes, oxygenated hydrocarbons etc.,  and at very high
fuel:air ratios can include 3.4 benzpyrene, recently suspected as
being carcinogenic.  HC emissions can show apparently random
variations across  the engine operating range, mainly due to the
extreme complexity of the reactions involved in chemical-kinetic terms.
Sources are possibly the same locally over-rich flame zones responsible
for soot formation and in some  cases weak  mixture zones in which the
flame is unstable.  Late admission of fuel is known to increase the
levels significantly since the falling chamber temperature effectively
freezes all reactions before combustion is completed.

Whereas NOx formation is the result of efficient, high temperature
combustion  (and is therefore difficult to reduce without seriously
affecting engine efficiency) ,  both CO and HC  are the results of marginal

-------
                           4-14
combustion conditions either throughout or within local pockets
(sometimes on a molecular scale) of the chamber.

The diesel engine emits much larger quantities of one other gaseous
pollutant than does the gasoline engine. This is SC>2 which arises
from the sulphur in the fuel. Current diesel fuels contain up to
0.5% of sulphur while gasoline contains only about 0.05%.  It is not
possible  to reduce the SC>2 levels by engine modifications, and while
it would be technically possible to remove the sulphur during the
refining process, the cost is likely to be high.
Effect of Combustion Chamber Type on Emissions
It has been recognised for many years that, when optimised for
performance, the direct injection engine emits more HC, CO and
NOx than does the indirect injection engine.  As a number of authors
have pointed out however, and as may be seen from Figure 4-6,  the
direct injection engine has a much steeper response of NOx reduction
with retard  of ignition timing, and at  very  retarded timing there is
little to chose between the two types in so far as  NOx levels are
concerned.  These degrees of retard of timing introduce excessive
hydrocarbon levels however, and with naturally aspirated direct
injection engines,  the high smoke levels would involve  excessive
derating to maintain an acceptable exhaust smoke level»

Pischinger I 21)  selects the direct injection engine  as a potential low
emissions power plant because of the steep NOx/injection timing
response and claims to prevent the onset of smoke by induction port
and combustion chamber design modifications.  As mentioned above
however,  in Ricardo's view the high hydrocarbon levels likely to
result from the required degree of retardation will present a problem.

EPA tests on vehicles with prechamber (Mercedes  220D) and swirl
chamber (Peugeot 504, Opel 2100D, Nissan Datsun 220c) engines  have
indicated no significant differences in the emission characteristics
between these engine types although any engine may have high HC
levels due to injection irregularities.

-------
                           4-15
Effect of Engine Load on Emissions
As the load on most engines is reduced at constant speed, the specific
hydrocarbon emissions increase.  Similarly on all indirect injection
engines and on many direct injection engines, the specific NOx levels
increase with  reduction of load. As a result, for a given power output,
a larger, lower rated engine will have higher emission levels than
will a smaller higher rated one.  With a given vehicle weight therefore
and driving  over a fixed driving cycle, provided that all vehicles can
drive the cycle, the one with the lowest power/weight ratio, i.e. the
smallest engine, will have the lowest HC  and NOx emissions.  In
general terms however there will be little to choose between their
CO levels.
Turbocharging
In general, turbocharging without aftercooling gives somewhat higher
NOx levels than those obtained with naturally aspirated engines.  It will
however clean up the smoke which otherwise would arise from retarded
injection timing.  If the compression ratio is reduced to limit maximum
cylinder pressures, high HC levels may result at low load where the
turbocharger is giving no effective boost.

With turbocharging  and aftercooling, as has been shown by Walder (16),
a reduction of NOx  of approximately 2Q% is possible together with  a
smaller improvement in CO levels.  The HC levels are unaffected.
Gaseous Emissions from Two Cycle Engines
As the majority of engines used in truck service are four cycle, the
greater part of the data on emissions from diesel engines is from
four cycle engines.  The very successful Detroit Diesel 71 and 53
series are however through scavenged two cycle engines and available
data indicates that these engines have similar levels of NOx and HC
emissions to four cycle engines. It may be inferred therefore that
both through scavenged two cycle and four cycle engines have similar
emissions levels.

So far as Ricardo know, there is no data available for the exhaust
emission levels from loop scavenged two cycle engines. With exhaust

-------
                           4-16
ports, oil control and hence HC levels may give problems and, with
less effective scavenging, smoke and hence CO levels may tend to be
higher than with through scavenging,  unless severe derating be
employed. On the other hand,  the internal exhaust gas recirculation
inherent in poor scavenging should give lower NOx levels although
there will be of course no cooling of the recycled gas.
NOx Reduction
The most effective way of reducing the quantity of NOx in the exhaust
is to retard the injection timing.  On a direct injection engine however
retarding the timing leads to an increase in smoke levels and hence
for  a constant smoke limit it is necessary to reduce the power output
of the engine. Some improvement is possible by retarding the timing
and increasing the rate of injection.  This can lead to reduced levels
of oxides of nitrogen with less derating of the engine but normally
there is still some loss of efficiency.

On indirect injection engines, at least over the range of injection
timings of interest, smoke levels reduce as the timing is retarded,
and hence there  is no need to derate the engine.  There is still however
a loss of efficiency due to the retardation of timing.

Apart from  injection timing retard,  there are two other effective
methods of NOx  reduction.  One is by recirculation exhaust,
preferably after cooling it.  and the other is by water injection either
with the induction air or as  an emulsion with the fuel.  Both methods
reduce the temperatures within the charge and EGR also reduces the
oxygen available to react with the nitrogen in the air.
Exhaust Gas Recirculation
The effectiveness of exhaust gas recirculation in reducing NOx has
already been proven beyond doubt by many authorities with both light
duty (647) and heavy duty(6, 16,  12) cycles but durability, particularly
that of the lubricating oil, has yet to be established. From data
available  to Ricardo., there are indications that  at high exhaust re-
circulation levels the wear rate of engine components may be higher
than without such recirculation.  There is insufficient data however
to predict whether or not a problem will arise from such wear although

-------
                           4-17
oil thickening is likely to be a problem with a consequent need for
frequent oil changes.

By reducing the oxygen available for combustion,  exhaust gas re-
circulation results in a need for derating to avoid excessive smoke.
Fig. 4 -7 demonstrates the smoke response of an  EDI engine over the
load range with varying amounts of EGR.  It will be noted that increasing
recycle rates significantly reduce the smoke limited performance of
the engine while the fuel consumption is unaffected.  Assuming therefore
that the engine is rated near to its smoke limit without EGR, no
appreciable improvement in full load emission levels can be achieved
using recycled exhaust unless the engine is derated. As load is
reduced, increasing amounts of recycle gas may be introduced.

The scope for such modulated exhaust gas recirculation will vary from
engine type to engine type.  With indirect injection,  the NOx emissions
actually rise with reducing  load for some way down the load range
and hence modulated EGR can be very effective. With direct injection
however, where the NOx levels  fall quite rapidly with load, the effect
of modulation is less pronounced (Fig.4-8).

Turbocharged engines, with both direct and indirect injection, have
rapidly falling NOx concentrations (Fig. 4-9) with reduction in load but
normally have a considerable excess of air for combustion at full power
conditions.  Hence there is little need to restrict  exhaust gas recirculation
at full power.  There may however be smoke problems at the maximum
torque speed as derating here will make the torque curve from the
engine less attractive.

It  must be remarked however that when driving a  light duty vehicle with
a high power/weight ratio over the CVS-CH test procedure, the engine
will not run at full power and modulation of exhaust gas recirculation
to reduce NOx at these conditions may not therefore be necessary.
Water Injection
Water injection has been found to be a powerful agent for reducing NOx
(6,16), but there are major problems in practice in applying this
method.  The major advantage is that significant reductions in NOx can
be obtained with no loss in engine performance.

Problem  areas are:-

-------
                          4-18
      A water tank approaching the size of the fuel tank is
      required, the actual size depending on the proportions
      of water/fuel employed.

      Antifreeze is necessary in winter and this may lead to
      HC problems.  When Ricardo experimented with alcohol
      as an antifreeze the exhaust hydrocarbons were excessive.

      To guarantee low emission levels it is necessary to
      incorporate a mechanism to shut the engine down in case
      of failure of the water injection system.

      Cylinder liner corrosion and wear and/or lubricating oil
      contamination with water may be a problem.

Some experimental work is being carried out with injection of water/
fuel oil  emulsions. This may overcome some of the problems but
it will probably be very difficult to obtain acceptable  stability of the
emulsion.  Either partial or complete breaking of the emulsion would
be disastrous  from a fuel handling and engine operating point of view.
Reduction of NOx by Reduced Engine Compression Ratio
It has been suggested that one way to reduce NOx levels would be to run
the diesel engine at a low compression ratio.  In Ricardo's experience
however, the lowest overall levels of HC and NOx are obtained at high
compression ratios,due to the onset  of misfire at light loads and
starting when using a low compression ratio.  This results in high HC
emission levels. The use of more effective starting and ignition aids
has been proposed to overcome this  but to the best of our knowledge
no-one has  yet successfully demonstrated such a device.
Emissions under Light Duty Cycle (CVS-CH) Conditions
While, as has been pointed out earlier, the power/weight ratio of
current light duty diesel engined vehicles is much lower than is necessary
to satisfy the performance requirement for this survey, it is of greatest
interest to look at results from vehicles operating under the CVS-CH
cycle test procedure in  order to predict emission levels for a diesel
engined passenger car.

-------
                           4-19
The literature contains a large number of references relating to the
effect of engine variables on diesel exhaust emissions when
operating over the 13 mode steady state test procedure.  There are
unfortunately however only a very small number relating to light
duty CVS-CH cycle testing.  Furthermore, in Ricardo's experience
there is no constant relationship between 13  mode and CVS-CH test
results so that it is not possible to read directly across from one
to the other. Despite this, there is in general terms a relationship
between the two types of testing in that parameters which tend to
reduce emissions in one test are very likely to do so in the other.

Although Ricardo have some data arising from tests which they have
carried out for their clients, very little additional CVS-CH data
has been found apart from EPA tests on standard European diesel
cars and one Mercedes 220D with modified injection equipment.
All these vehicles have indirect injection engines. The results of these
tests are summarised in Figure 4-10.

One other report on light duty emissions tests on diesel vehicles
covers comprehensive tests by Daimler Benz on an experimental
vehicle.  After considerable development and using high rates of
exhaust gas recirculation the secondary target levels were just
achieved. Significant reductions in HC and CO levels were obtained
by the introduction of a catalyst into the exhaust system and HC
levels were reduced by modifying the fuel injection equipment  (647).

All of the light duty vehicles tested are somewhat lighter than  the
specified 3500 Ib car for this study.  In addition, power levels are
much lower and in fact the vehicles are only normally just able to
drive the cycle.

To a  first order of accuracy, the emissions in g/mile when driven
over a fixed cycle and with the same power/weight ratio for each
vehicle, will vary with the weight of the vehicle.  Furthermore,
since specific NOx and HC levels on indirect injection and many
direct injection engines rise as load is reduced, an increase in
power/weight ratio will increase the emissions from the vehicle
when driven over a fixed cycle.

The best emission levels which have been obtained from current
generation diesel engined light duty vehicles are:-
                     HC                    0.2 g/mile
                     CO                    1.5 g/mile
                     NOx                   1.5 g/mile

-------
                           4-20
HC levels of 0.21 g/mile have been achieved by EPA on a standard
Nissan Datsun 220c and Ricardo tests on other vehicles have
indicated that substantial reductions in HC levels can be achieved
by minor modifications to fuel injection equipment without altering
engine performance characteristics.  Thus it is felt that the HC
emission levels of 0.3 - 0.4 g/mile achieved in the Mercedes and
Opel vehicles could be reduced to circa 0.2 g/mile.

As a result of applying a correction factor for vehicle weight and
increased engine size, predicted emission levels from a 130 bhp
3500 Ib inertia weight vehicle over CVS-CH cycle are:-
                     HC     '       0.4 g/mile
                     CO            3.0 g/mile
                     NOx           2.0 g/mile

Timing retard will reduce NOx  levels to the order of 1.2 g/mile
without seriously affecting CO levels. Combustion noise levels in
this retard mode should be lower and vehicle driveability will be
unaffected.  The fuel consumption will probably  increase by about
10%, water jacket heat losses will remain constant but blue smoke
and misfire and hence HC levels, particularly when cold, may be
a problem.  Some form of temperature sensor controlling injection
advance at light load may be necessary to control this.

Some further reduction in HC and CO levels may be possible by
optimisation of the gear ratios  within the transmission.  Some
preliminary tests by Ricardo on a diesel powered vehicle in which
gear ratios were altered showed substantial benefits could be
obtained by changing from a four speed to a three speed gear box.
The tests were in fact run with  the standard four speed box but in the
second run first gear was not used, but all gear change points were
selected as if the gearbox were of standard three speed form.  While
the NOx emissions were unaffected, the HC and CO levels were
reduced by approximately 50%.  While as large a change as this might
affect the  ability of the vehicle  to give the target performance,
transmission ratio modifications must be regarded as a possible
method of reducing  HC and CO  levels from diesel vehicles.

The secondary project objectives, i.e. to obtain 0.4 g/mile NOx, are
very much more difficult to achieve.  Daimler Benz have already
demonstrated a passenger car  (647) which just  attains this level in a
highly modified build,  but with  no room for production variations or
deterioration throughout the life of the vehicle.  The level of 0.4 g/mile
was only achieved with 'increased1 EGR flow rates (exact value  not known)

-------
                           4-21
and although the value of EGR in terms of effectiveness in reducing
NOx is not questioned, engine durability with high recycle rates
has yet to be proved.,  The practicality of achieving this level of
NOx with a production solution in a European type car is very much
doubted and the chances of achieving 0.4 g/mile NOx with even a
prototype version of an American type high-powered vehicle are
thought to be very remote.

In summary, a 97 kW (130 bhp), conventional  naturally aspirated
swirl chamber diesel engine in a practical 3500 Ib vehicle should be
able to achieve the project target emission levels of CO  3.4 g/mile,
HC 0.41  g/mile and NOx 1.5 g/mile  with timing retard and relatively
minor modifications to the fuel injection equipment. Some EGR may
be necessary to ensure a sufficient margin for production compliance.
A less powerful engine in the same vehicle would result  in the
additional benefits of lower HC and NOx but an inability to meet the
performance requirements.
Particulates
With its complex heterogeneous combustion, local rich areas leading
to particulate emissions must exist in diesel engines.  It has been
postulated by Wheeler in an unpublished Ricardo note that it is in these
rich areas immediately surrounding the fuel droplets that pyrolysis
occurs forming graphite sub particles of great reactivity which
subsequently oxidise rapidly upon meeting freely available oxygen or
coagulate into clumps like strings of beads.  If the latter happens
slightly quicker than the former then soot escapes from the chamber
still actively coagulating.  Even the highest temperature black  soot
is still a hydrocarbon containing about 3% hydrogen by weight (CsHs).
The most important property of this soot is its high absorptive
capability for all active molecules. Thus as it cools and its attraction
increases it absorbs first high boiling hydrocarbons, then  lower
boiling, then aldehydes and other oxygenates, then SO2  and finally
when down to ambient temperatures, fill-in of the remaining 'holes' is
completed by water molecules.

As mentioned in the smoke  section, the droplet aerosols of blue and
white smoke  are also products of very incomplete combustion  (in
this  case mainly due to marginal temperature conditions within the
cylinder).

-------
                           4-22
Thus there are two completely separate particulate problems in the
diesel engine, blue/white aerosol emissions at low load which can be
relatively easily removed by engine modifications, and the gas
carbon (soot) formed by pyrolysis  in rich areas which increases
as stoichiometry is approached and is regulated by legislation
limiting its visual appearance.

Californian suggested legislative levels were set at 0.1 g/mile.
This level was originally intended for enforcement in 1975 (383).
Typical diesel powered, European type passenger cars  appear  to
emit particulates in the order of several times this level (0.5 - 3
g/mile) and it is difficult to envisage practical methods of reducing
particulate levels at source. A few authorities, notably Berliet,
(649, 650) have attempted to develop soot filters.  Their early system
used Kaowool as a filter element, this being impregnated with an
oxidation catalyst. More recently, it has been shown that the catalyst
is not necessary for burning off the carbon particles, although it is
of some secondary use in that it can oxidise the CO resulting from the
carbon burn off to CC>2 (as well as  oxidising the in cylinder  generated
CO and HC).

The pure filter  (i.e. non catalytic)  system works by reason of the
fact that the residence time of the soot particles is increased
sufficiently to allow natural burn off to occur.  With burn off occurring
at 450 - 500  C  ( i.e. at a high load factor), the major problem behind
the development of a practical system is the incorporation of a
sufficiently large storage system in the exhaust to trap and store (hold)
soot particles emitted at part loads without causing excessive exhaust
back pressure increases in typical  duty.  It is in this area that much
research would be necessary into vehicle operating modes and engine
smoke characteristics before a system could be developed for
production.

To sum up therefore, at present the only controls on the particulate
emissions from diesel engines are  in the form of smoke legislation
and most modern engines have been developed to comply with the most
severe of these (U.K. smoke regulation based on recommendation
BS AU 141a) . Californian suggested particulate controls of 0.1  g/mile
over the CVS-CH cycle are much more severe than this, current diesel
engines emitting 0.5-3 g/mile, and in order to comply with 0.1 g/mile
levels it would be necessary to develop a soot filter system.  Much
development  would be necessary to ensure a sufficient storage volume
for particulates emitted at light load  and at below burn off temperature.

-------
                           4-23
Noise
To the passer-by the noise of the engine is probably the first obvious
indication that a particular vehicle is fitted with a diesel  engine,
and for this reason it is an obvious cause of consumer resistance.
The larger the vehicle the greater is the likelihood that the engine
will be of direct injection form and the more obtrusive the noise to
be.  However, the smaller high speed diesel engines fitted to
passenger vehicles and delivery vans will probably be of  indirect
injection form and consequently less noisy (34).  Although little
direct comparison is available, the indications are that for a similar
size, the IDl engine may be up to 5 dBA less noisy than the DI under
standard drive-by test conditions (39)  and a comparison  with a
gasoline powered vehicle shows an increase of 2.5 dBA for the diesel
after direct conversion from gasoline to a Ricardo Comet type of
diesel engine  (142)  (Ref. 112 shows a predicted 2 dBA increase,
Ref.  479 shows a predicted 4 dBA increase).

Noise at idle conditions is a prime consideration as, although the
absolute levels may be low, the characteristic of the noise is subject-
ively annoying and of much greater impact than for gasoline engines ,
especially for passenger vehicles (18). Comparisons have shown the
diesel to be considerably more noisy than the gasoline engine at idle
(2) but these tests were carried out on a four  cylinder engine where
low frequency vibration of the diesel due to torque reaction under
high compression ratio conditions, since the engine is not throttled
when idling, gave subjectively a harsh  idle characteristic.  This
problem would be largely eliminated in a 6 or  8 cylinder  engine and
the idle noise would be less objectionable but would still be very
pronounced as compared with a gasoline engine.

It has been shown that the majority of noise in the diesel  engine is
combustion generated and it follows therefore  that modifications to
the combustion and/or fuel injection system which gave a smoother
pressure diagram, would result in a quieter engine (2, 3, 631).
These measures include the use of pilot injection (2, 33, 89, 200) ,
injection timing  changes (2,  201, 221,  324), nozzle changes (454)
and injection rate changes (3, 200). For instance,  refs.  2, 33, 89,
and 200 quote reductions of 2 - 5 dBA using pilot injection, and refs.
201 and 324 quote reductions of l£ to 4 dBA with up to 6  injection
timing retardation.  Although pilot injection is known  to reduce the
noise by modifying the pressure diagram its application, in practice,
has not been achieved over the whole load  and  speed range of an
engine due to the variable fuelling characteristics of current fuel
injection equipment  and the difficulty of obtaining an accurate balance

-------
                           4-24
between cylinders.  The effect of injection equipment in reducing
the characteristic diesel clatter (164, 2)  is important at idle
conditions, and one manufacturer (Peugeot) has introduced a
device which effectively reduces the injection rate at. idling.

Attention to the compression ratio is important in keeping combustion
noise to a low level. Refs. 153, 201, 253 and 531 report significant
reductions in noise with increases in ratio. (201 reports 1 dBA
reduction with  CR increased from 18 to 20 in a Ricardo Comet
combustion chamber).

Fuel quality is  also important in keeping diesel knock to a minimum.
Low Cetane number increases the noise level due to poorer ignition
quality  and therefore longer delay between injection and combustion,
resulting in a more rapid pressure rise (2, 131, 243).  However,
no advantage is to be gained above about 55 Cetane.

Noise of the engine is caused by vibration of the outer surface
resulting from  forces within the engine being transmitted by the
engine structure  to the outer surfaces.   It follows therefore that
reduction of noise can be achieved by modification to these forces
and to the transmission paths, and by structural changes, bringing
about a reduction in the vibration of the engine outer surfaces.  Much
research has been,  and is being, carried out on noise reduction by
attention to engine construction and many papers have been published
including those by Priede (83, 301 etc.), Jenkins  (4), List (48) and
Scott (2). In addition researchers have investigated the effects of
changes to piston pin offset (3, 378, 380), valve mechanisms (23, 377)
and timing gear drives  (2), but the main conclusions are that more
noise reduction can be obtained by attention to the engine structure
than by other means.

By isolation and damping of the unstressed cover, a noise reduction of
5-8 dBA has been achieved (53, 111), others record 3-5 dBA
reduction (23,  84). By using  a complete 'structured1 engine (174, 4)
up to 10 dBA reduction in noise wag obtained.  However, this engine
was  of experimental form, having a stiff frame containing the crankshaft
and cylinders,  with the areas  in between covered with non-vibrating
material, and it is not a practical' manufacturing proposition at the
present time.  Future engines, however, may show a trend towards
'unconventional1  construction techniques if noise limiting legislation
becomes more  severe.

-------
                            4-25
In order to reduce the noise of existing engines, in addition to
isolation and damping of the unstressed parts of the engine as above,
resort may be made to shielding and enclosureso  Much research
has already been carried out on these techniques and it has been
shown.that 5-8 dBA  reduction may be obtained by shields to the
engine (23, 267, 174, 84, 272,  371, 373), and by  total enclosure
up to 20 dBA reduction has been obtained (23,  83,  84, 375).  These
experiments were mostly carried out on the test bed where
conditions are ideal but, when attempting to shield or enclose an
engine in a vehicle, a more difficult situation'arises. It has been
shown that 5-6 dBA  reduction can be obtained under these conditions
however  (275), although_such a system of shielding or enclosing
increases the problems associated with maintenance.

Priede (83) has shown that engine noise can be calculated using an
empirical prediction  formula, i.e. the noise (dBA) of a diesel engine
at 1m = 30 log-jQ N +  50 logjQB - 48.5, where N is the engine speed in
rev/s and B is the cylinder bore diameter in mm.  This  applies to a
4-stroke normally aspirated engine.  Variants of the formula can be
used to predict noise  values for gasoline engines,  2-stroke engines
and boosted engines „   These for mulae show that the noise of the
engine is a function of the rotational speed and cylinder  bore diameter,
and ref „  83 .concludes that for a lower noise 'engine a configuration
of. small  bore, more  cylinders, lower rotational speed and turbo-
charging would be desirable.  Experience has shown that drive-by and
test bed noise can be  related and that the noise of a vehicle at 7.5m
is some 15 -_17»5 dBA less than an engine at 1m „  This relationship
is dependent to a certain extent on the installation  in the individual
vehicle due to variabilities in shielding by bodywork etc.

To keep vehicle noise to a minimum, careful attention should be paid
to the design of exhaust and intake silencers.  It should  be noted that
the intake on a diesel is unthrottled all the time (367)„ Optimisation
of silencing systems  can more rapidly be carried out by use of a
computer aided silencer design program,

It is  likely .that radiator fan noise could be a problem in both gasoline
and diesel powered vehicles (175, 373), and the use of thermostatic
control devices would aid the solution. With air cooled engines fan
noise can be excessive but by careful design this can to  a large extent
be overcome (43, 257).  Air cooling may have other advantages, for
instance  the ducting system could.be used as a noise shield if careful
attention was paid to design.

-------
                           4-26
Interior noise can be a considerable problem in diesel engined
passenger cars unless care is taken.  All the measures mentioned
above would reduce the amount of noise transmitted to the interior
and particular attention should be paid to engine mountings to
prevent vibration - particularly at idling conditions. Mos"t of the
published data on interior noise is concerned with commercial
vehicles but a small amount of passenger car information is
available  (142), and shows  an increase of 2 - 3 dB over the gasoline
version of the vehicle at 48 km/h (30 mph)  and 3^- 4^ dB at 80 km/h
(50 mph). During acceleration the comparisons were ps follows:
48 km/h (30 mph) 3 dBA louder, 80 km/h (50 mph) 2^ dBA louder
and 113 km/h  (70  mph) 1^ dBA louder. 'The latter figure suggests
that at speed the diesel car is less noticeably noisier than the
gasoline version and with more efficient sound insulation would be
no noisier o  Provided the small diesel vehicle is constructed in such
a way that the engine noise is not too obtrusive, there is little reason,
on grounds of noise level,  why such a vehicle should not be acceptable.
Volume
The close physical resemblance between the diesel and gasoline engine
results in very similar specific box volume when the comparison is
made on  the basis of swept volume. Due to the slightly increased block
height (to ensure adequate piston compression height)  and the greater
need for  water cores between cylinders, the diesel is however  slightly
bulkier than its gasoline counterpart.  The cylinder head of the diesel
engine may also be slightly deeper due to its increased complexity.

When  however a comparison is made on the basis of engine power
output, the diesel suffers a large penalty.  This, as emphasised in the
next section, is due to the substantially higher specific output of the
gasoline  engine.  As a result, the gasoline engine may have to  be
compared with a naturally aspirated diesel engine of some 50%  higher
swept volume.

Increasing the specific output of the diesel by boosting would reduce
this disadvantage although it  is likely that there -wduldstill be a  height
penalty against the diesel (depending on engine configuration).  The
improvement would be achieved at the cost of the increased technological
requirements  of the engine (higher cylinder pressure demanding  more
sophisticated gasket materials and stud patterns as well  as higher
thermal loadings leading to the necessity of oil cooled pistons,  oil

-------
                            4-27
 coolers etc.).  It is felt that reduced box volume alone is insufficient
 to warrant this degree of complexity although the attendant reduction
 in engine weight may be.

 Forthcoming U.S. Vehicle Safety regulations are likely to have an
 effect on engine length with a need to allow sufficient crush space
 and it may be in this  area that the diesel is at its greatest disadvantage
 with regard to physical size, although the use of the vee configurations
 in order to minimise length is equally applicable to the diesel as it is
 to the gasoline engine.

 In conclusion, the diesel  in its naturally aspirated form has a significantly
 larger box volume than a gasoline engine of the same power output.
 Boosting the diesel would reduce this penalty but it is debatable whether
 the increased engine  complexity is warranted by reduction in box volume
 alone.  Due to the normal excess hood space in cars of this  class,  this
 is not considered a major problem area although the introduction of
 severe vehicle safety regulations may demand extra hood length for the
 diesel powered passenger car, thus incurring a  first cost and fuel economy
 penalty.

 The increase in  volume over the gasoline engine will be of the order of
 50% which, assuming similar bore/stroke ratios, will give an increase
 of linear dimensions of 12%.
Weight
Experience with European powerplants and engine conversions reveals
that for reasons explained in the last section,  there is only a small
difference (12 - 18%) in the specific weight of the two when the
comparison is made on the basis of swept volume (648 and Figs. 4-11
and 4-12)=  Important reasons for the increased weight of diesel engined
vehicles are the weight of most diesel engine accessories (starter
motor,  generator,  fuel injection equipment, battery etc.) which are
substantially greater than their gasoline engine counterparts -
approximately 30% of the total engine weight penalty in Fig. 4-13 (648)
being due to the extremely large, heavy duty battery fitted to the diesel
powered vehicle»

For a given application, the dominant factor governing the relative weights
of diesel and gasoline engines however is that of specific power output.

-------
                          4-28
Because of its heterogeneous combustion mode, the high speed
indirect injection engine can use at most 90% of the air available
within the cylinder without exceeding socially acceptable levels
of exhaust smoke, and engines with other combustion systems
would use substantially less. The gasoline engine is capable of
consuming and efficiently  using 100% of its  air charge; this,
combined with its lower friction losses (Curve No. 5, ref 648)
and higher useful operating speed accounts for the much higher
specific power output of the gasoline engine in terms of bhp/litre.

Fig. 4-14 (Fig. 4 from 648) demonstrates the difference in specific
power output between the two engine types.  At best the high speed
diesel engine delivers  24 kW/litre (0.5 bhp/in3)  whilst its
European gasoline counterpart delivers a minimum of  30 kW/litre
(0.65 bhp/in3).  Many American cars do in fact have a somewhat
lower rated gasoline engine, of similar specific output to the diesel
engine, but current American compact cars are powered more
closely following 'European philosophy1  although they are normally
somewhat heavier.

A greater use of aluminium components  would go some way to
reducing  the weight penalty at the expense of cost but it would of course
be open to the gasoline engine to move in the same direction.
Alternatively, boosting of the diesel engine can be employed although
any disadvantages of this must then be accepted.

For the 1600 kg  (3500 Ib)  vehicle under consideration, it is estimated
that an engine power output of 96 kW (128 bhp) is required to enable
performance targets to be  achieved.  Current American gasoline
engines of this output would have a swept volume of about 4.5 litres
(275 in ) and would weigh  about 250 kg  (550 Ib).  It would however
be possible to produce a more compact gasoline engine of about 3 litres
(180 in3) which  would produce 96 kW  (130 bhp) at   80 - 100 rev/s
and would weigh about 186  kg . (410 Ib).
Fuel Economy
The fuel consumption of U.S. cars has been increasing over the years
due to increases in vehicle weight and engine size as a result of
increases in vehicle size and the fitting of automatic transmissions,
air conditioning and emission and safety devices  (461, 464). The
standard size Ford sedan weight increased for example from 1610 kg

-------
                           4-29
 (3550 Ib) in 1965 to 1940 kg (4275 Ib) in 1973, and it was necessary
 to increase the engine size from  5 litre to 5.75 litre (300 to 350 in3)
 Over this period the fuel economy deteriorated by 20% (463) .

 The demand for gasoline in the U.S. is predicted to increase by 50 -
 70% between 1973 and 1985. Three possible ways to moderate this
 growth in fuel demand and still protect air quality criteria are:-

           1.  Wider use of mass  transportation and car pools
           2.  Use of smaller cars
           3.  Use of diesel cars
Of various methods considered to conserve transportation energy, the
greatest saving can be achieved by conversion to small cars and
economy vehicles (467).

The calculated peak thermal efficiency for ideal engines shows the
gasoline to be 45% and the diesel 55%; the diesel advantage increases
however under low load conditions (20) .  The fuel economy gain by the
diesel diminishes as the engine  speed is increased due to the higher
friction and pumping losses (18).

While there have been a limited  number of models of light duty diesel
engined vehicles in  service, there are a considerable  number of
references to the comparative fuel economies of diesel and gasoline
engined vehicles . Typical data are given in the following  extracts from
the literature :-

    The potential saving in fuel  for the diesel as compared to the
    gasoline vehicle is  5% for highway driving, 30% for mixed
    duty and 50% for taxi cab service (479) .

    Early London taxi company fuel consumption data suggested diesel
    taxi 11.1 1/100 km  (21.2 mpg) against 21.4 l/100km (ll mpg)
    for gasoline, the diesel figure being taken over  8.8 x 10  km
     (5.5 x 10  miles) but gasoline figures are likely to have im-
     proved since that time (15).

    Fuel consumption for Paris  vehicles , 11.3 1/100 km (20.8 mpg)
    for diesel and 15.7 1/100 km (15 mpg) for gasoline  (15).

    In London taxi service  the gain in mpg is  almost 2 : 1 for the
    diesel over gasoline.  In general use the  diesel  engined car

-------
    is 30 - 50% better than gasoline (112).
    By fitting diesel engines in place of gasoline engines in
    U.S. Post Office ^, 1 and 5 ton vehicles, fuel economy
    increases ranging from 18/23%, 10/28% and 150/180%
    respectively were obtained, in controlled tests at
    Aberdeen Proving Ground. From vehicles in actual service,
    the £ and 1 ton vehicle economy was 50% greater than the
    gasoline (31 ) .

    From customer and service records,  770 kg (1700 lb)
    vans fitted with Perkins 4-99 engines used for  local
    delivery, municipal work and distance haulage showed
    fuel economy between 71.5 and 129. 8% better than gasoline
    engined vehicles (363).

    EPA tests carried out on diesel engined cars - Mercedes
    220D,  Opel Rekord 2100D and Peugeot 504. Diesels give
    70% better fuel economy than  gasoline, by the same
    test procedures (l9l).

Ricardo measured  road fuel economies of three different model diesel
engined cars were as follows : -
Litre/100 km
    7.0
    5.9
    6.0
mpg  (country route)
33.8
39.8
39.4
litre/100 km
    8.9
    7.2
    8.2
mpg (town)
26.3
32.8
28.8
Reference 490 gives constant speed consumption.
                  Peuqeot 504
                    Mercedes 220D
                    Opel Rekord
litre/lOOkm
60
80
100
120
140
km/h
km/h
km/h
km/h'
km/h
5
6
8
10
12
.0
.1
.0
.3
.2
mpg
46.6
38.7
29.2
22.8
19.2
litre/ 1 00km
4.
5.
7.
8.
11
9
8
2
7
.4
mpg
47
40
32
27
20
.9
.5
.5
.1
.6
litre/ 100km
5.
6.
7.
9.
11
3
3
8
9'
.6
mpg
44.6
37.5
30.0
23.8
20.2
Austin A40 - measured fuel economy 45% better with diesel than with
gasoline engine (198).

-------
                           4-31
Comparing direct injection and indirect injection diesels of the same
cylinder size, the DI has no significant fuel advantage over the EDI
for automobile use. The DI gives about 10% better fuel economy on
the test bed but no difference on the road. This is due it is believed
to the ability to hold on longer in a high gear with the high torque
of the indirect injection engine.

To sum up therefore, the features of good fuel economy combined
with relatively low emissions are the major arguments for the case
of diesel powered passenger cars.  In practical terms the fuel
economy advantage of the diesel is at its  greatest at the light load
end of the engine operating range.  European experience has shown
significant economic gains in the use of diesel powered vehicles
particularly for light load, local delivery and in city use. Rising
fuel prices as well as increasing service costs are steadily
increasing this advantage.

Actual fuel economy gains of the diesel vehicle over its gasoline
powered cousin are difficult to quantify because of the dominating
effect of vehicle operating mode.  European in-city use has revealed
fuel economy gains of 50% to 100% in terms of miles per gallon.  In
general use this improvement reduced to 30 -  50%.  No figures are
available from  vehicles with a significant highway driving mode but it
is felt that such vehicles would return fuel economy advantages of
less than 25%0  EPA data comparing diesel with gasoline powered
vehicles on a simulated .inertia weight basis showed that economies
of approximately 70% (again mpg) were realised (Fig.  4-15).

In the prototype vehicle, the high power/weight ratios must cause
the engine to be driven at light specific load factors and will also
allow higher gear ratios to be used, resulting  in the use of low
average engine rotational speeds.  Both these factors should enhance
the fuel economy advantage of the diesel  (as opposed to comparing
two equal but low powered diesel and gasoline  vehicles where  the high
load factor and rotational speeds would tend to minimise the advantage
of the diesel)»

Although current European diesel powered vehicles comply with the
project emission targets while in production build, the high powered
heavier prototype vehicle will need engine modifications to ensure
compliance with the primary project targets.  Data obtained from
the European vehicle suggests that if similar technology engines are
used, large amounts of timing retard will be needed to better  2 g/mile
NOx and a small but significant fuel economy penalty will be incurred.

-------
                           4-32
Assuming approximately 10  injection retard is used over the lower
half of the useful torque band, the resultant fuel consumption
penalty at steady state conditions would be in the order of 8 - 10%.
In real life driving conditions the penalty in fuel economy would be
less,  probably 5 - 8%.
Fuel
It must be remembered that some 16% or so of the increase in miles
per gallon between the diesel and gasoline engined vehicles arises
from the higher density and corresponding higher calorific value of
diesel  fuels as compared with gasoline.  Whether this is a real gain
which can be fairly attributed to the diesel engine can only be
decided from a study of the energy economies of the production of
distillate fuels from crudes.

The world consumption of mid-distillate  fuels is increasing rapidly
in both the under-developed and in the highly industrialised countries.
In those countries in the early stages of  industrial development,
the directing of the greater part of their resources into construction
channels is reflected in measures taken  to encourage the building of
all types of vehicles which use diesel engines and the discouragement,
by discriminating taxes or other means, of the  manufacture of motor
cars for private sale and pleasure use.  In highly industrialised
countries  which have a moderately or very cold winter, space heating
of all new and many old buildings is increasing, with  a consequent
increase in demand for light distillates for the heating installations (277)

The use of residual fuels is becoming less popular due to their high
sulphur content and there is a trend towards the use of lower viscosity
fuels,  i°e. the mid-distillates (192).

In the U.S. , crude oil consumption is increasing at a rate of 100 million
barrels/year, and the estimated increased demand for mid-distillate
fuels is some 44 million barrels/year.  By changes in refinery
techniques it would be possible to double the current production of
mid-distillates at the expense of gasoline and residual fuel production.
By means  of these changes in refinery techniques and based on the
projected 1978 U.S. consumption, it is estimated that the production of
automotive diesel fuel could be increased from  130 to 1610 million
barrels/year, i.e. a 12-fold increase (194).

-------
                           4-33
Widening the cut of mid-distillates would give a fuel having a lower
Cetane number and flash point and an increased aromatic content.
The reduction of Cetane number would adversely affect noise and
starting (173).  The increased aromatic content would result in
some increased engine wear (640).

For the high speed diesel engine the most important fuel qualities
are:-

     1.   High ignition quality to give short ignition delay
          and good cold starting.

     2.   A boiling range which will give the highest possible
          air utilisation and ease of starting combined with a
          minimum of deposits.

     3.   Flash point low enough to enable use of the
          volatile fractions.

     4.   Vapour pressure low enough to avoid vapour
          lock,,

     5.   Sulphur content generally below 1%.
                                                  (555)

(Ricardo Note:- The level of this last item  will probably be driven
down by future legislation) .

Amyl Nitrate is an ignition accelerator for diesel fuels - its use would
effectively increase the Cetane rating of the fuel.

Barium based additives are very effective in reducing diesel exhaust
smoke although this effectiveness  varies depending on the engine
design. (109)°  However, these anti-smoke additives are not attractive
because of pollution problems concerning particulates (173) .

A series of tests carried out at South West Research Institute on a rail-
road engine operating on propane as fuel was suspended before satis-
factory operation had been achieved (537).

On transit vehicles it is claimed that the use of propane results in
reduced fuel costs compared with diesel or gasoline vehicles.  It is
also claimed that with propane as a fuel, engine wear and maintenance
is less and there are no problems with smoke andodbur (541).

-------
                          4-34
(Ricardo Note: - This last item refers to spark ignition propane
engines - the use of propane in diesel engines does not appear
attractive) .
Fuel at Low Temperatures
When fuel is cooled between cloud point and pour point, some
paraffinic constituents preciptate as wax crystals and can accumulate
in fuel systems causing fuel starvation.  Flow improving additives
alter the crystal growth to give better cold flow (643).

Flow improvers give crystals which are smaller,  less cohesive and
do not form a gel structure.  These improvers do not alter cloud point
or the quantity of wax which separates. (642)

The flow of automotive diesel fuel at low temperature can be improved
by fuel dilution with kerosene. (314)
First Cost
The manufacturing cost differential between gasoline and diesel
engines is difficult to establish, in part because manufacturers who
make both types will seldom provide the necessary breakdown of costs.
The selling price of diesel vehicles may be no guide to the manufacturing
cost, being more an indication of what the market will stand or a means
of controlling the extent of the market. The relative production
quantity affects the cost of the engine and also that of the fuel injection
equipment which forms a  major item in the increased cost. The data
available indicates that in Europe the light duty diesel engine costs 50%
more to produce than the  gasoline engine. About half of this increase
is the cost of the fuel injection equipment. The remainder is attributed
to the greater complexity of the engine design in the combustion
chamber region,  superior materials, larger starter motor etc. (479).

Three reasons account for the present difference in cost between the
diesel and gasoline engines: greater weight and complexity, closer
manufacturing tolerances on some parts, and more limited scale of
production (277).

-------
                            4-35
First cost is inevitably influenced by the number of engines built and
the degree to which common tooling can be used between diesel  and
gasoline versions of a basically common engine.  The diesel is
likely always to be more costly than the gasoline engine for the
following reasons:-

      1.   Unless there is a major breakthrough to some simpler
          form of injection equipment, the cost of this item will
          be higher than for gasoline engine ignition system and
          carburettor.

      2.   Because of the higher cylinder pressures  in the diesel,
          some of the  basic engine parts need to be stronger and
          due to the combustion chamber form the cylinder head
          will be more complex.

      3.   Because of the heavier starting load, the starter motor
          and battery will need to be larger.  (15)

For the diesel engined vehicle  the difference in cost will be of the
order of 8 -  15% depending on the scale of production and on whether
or not a basically common design with common tooling is used for
the engine ( 15).

Ricardo believe that comparing British engines of equal capacity, i.e.
1.6 litres (98 CID)  in four  cylinders developing 37-54kW(50-72
bhp) respectively, the difference in cost between the diesel and
gasoline engine is approximately % 120.  Of this difference over half
is accounted  for by the increased cost of the fuel injection equipment
and starting aid compared with the cost of the ignition  system and
carburettor for the gasoline engine.

Increasing the power output of the diesel engine to equal that of  the
gasoline engine,  either by increasing the engine capacity or by turbo-
charging, will increase the cost of the diesel engine by % 120.

This high first cost is  one of the few economic disadvantages of  the
diesel engine.  Basically it is this initial penalty that the improved
fuel economy of the diesel must overcome before the economic
advantages of this engine type  can be realised, although the longevity,
reliability and fuel economy advantages  of the diesel should reduce
the depreciation rate of the vehicle compared  with a gasoline powered
car unless the limit is set by body deterioration.

-------
                           4-36
Investigation of available data therefore reveals the major contributory
causes to the current first cost penalty of the diesel to be:-

      1.   Lower specific output
      2.   Complex fuel injection equipment
      3.   Relative production levels
      4.   Greater complexity

From the limited data available to Ricardo, it is estimated that the
true production costs of European  automotive diesel engines are
approximately  2-2^ times that of a comparable (non-emission
controlled) gasoline engine. These figures are representative of true
current production costs, therefore production volume could have a
large influence.  Absolute cost figures are not known with authority
but it is felt that the true production costs of current high volume
European diesel engines are probably in the order of % 3.25/kW
(# 2.4/hp).  The cost of fuel injection equipment for the diesel engine
is obviously far higher  than for ignition/carburation systems, and
although the general technology and in particular production tolerances
on injection equipment are far more severe than in any other section
of the engine, the basic design for most components (for example
pumping elements, control valves and injector nozzles) lend them-
selves to high volume production.  Thus increased production  levels
could reduce not only bare engine  production costs, but also those of
ancillary equipment and in particular fuel  injection equipment.
Further, the cost of fuel injection  equipment does not increase pro  rata
with engine power output, whether this is achieved by larger or a
greater number of cylinders and the relative percentage cost penalty
of a 97 kW(l30  bhp)  diesel engine  compared with a gasoline engine  of
the same output is likely to be smaller than from the two comparable
engines of lower output.

Other factors which  contribute towards the higher cost of the diesel
include more complex and accurate machining requirements around the
combustion chamber and the need  for sophisticated  materials  in pre-
chamberso

In these high compression ratio engines,  the total combustion volume
is very small and tolerances controlling piston/head bumping  height
must be minimised in order to remain within the band  of correct
chamber proportions for optimum performance.  This  means that all
components which have some control over this must be manufactured
to more exacting requirements than their gasoline equivalents (see

-------
                           4-37
Fig. 4-16), and a slight additional cost penalty is thus incurred.

Although the diesel has been shown to be considerably more expensive
than the gasoline in terms of first cost, it must not be forgotten that
in a typical passenger car the cost of the gasoline engine accounts
for generally less than 10% of the total retail price of the vehicle.  This
proportion would increase with a diesel  engine to perhaps 15 - 20%,
still only a modest proportion and one which might easily be justified
in a fuel economy conscious climate.
Maintenance
User experience indicates that major maintenance intervals for light
duty diesels may be three times as long as for gasoline engines (15).
The use of pintle type nozzles in IDI engines gives good reliability
and long service,  290,000 km (180,000 miles) is being quoted with
routine cleaning (100).  Other sources quote 160,000 - 240,000 km
(100,000 - 150,000 miles) without removing injectors, even for
cleaning (112) and vehicle operation of 320,000 km (200,000 miles)
without attention was experienced by one taxi fleet operator.

Routine maintenance costs are difficult to define.  Some operators find
the cost less for diesel than for gasoline engines due to the lack of
spark plugs, HT components, etc. (479), but others suggest that
routine servicing is slightly more costly due to the higher frequency
at which filter elements should be changed. Ref. (15)  has a table
showing service costs of % 98 for  gasoline and % 112 for diesel. In
general, the opinion of operators  suggests that major engine maintenance
is much less frequent for diesel than gasoline engines but is probably
slightly more expensive per service.  The net result however is reduced
maintenance cost and less time off the road due to breakdowns and
servicing. Extra cost of diesel vehicle recovered in first 40,000 km
(25,000 miles) was quoted by one operator (479).

From the literature survey it is apparent that the routine maintenance
costs of the diesel are very similar to those of the gasoline engine
and that if anything the shorter diesel lubricant life may make-the
diesel's servicing marginally more expensive.

Service life of pintle nozzles of 160,000 km plus (100,000 miles plus)
with the rest of the fuel injection equipment having similar maintenance

-------
                           4-38
requirements, will virtually eliminate major overhauls in a
passenger car with an average life of 160,000 km (100,000 miles).
Starting
While the diesel engine is totally immune to the problems of damp
ignition systems which affect gasoline engines, there are difficulties
in obtaining a quick start under cold ambient conditions.

The limiting temperatures for a start without aids depend on the
combustion chamber type as follows :-

      Pre-chambers -  It  is necessary to use a starting aid (heater
      plug) to obtain a quick and smoke-free start from below
      10 - 15°C  (50 - 59°F) with this chamber.

      Swirl chambers - Some form of starting aid needed for the
      first start of the day when ambient temperature is below
      15  C  (59 F) , subsequent starts are satisfactory if Pintaux
      nozzles  are used.

      DI - Easy cold starting without aid down to -10  to -15 C
      (14  - 5 F) is normal.

      'M1 System  - The cold starting of the 'M' system is inferior
      to that of the DI because of the heat losses arising from the
      high air swirl employed and a starting aid is required (34) .

The cranking speed of  the engine is important for effective starting
and a minimum  of  2 rev/s(l20 rpm) should be maintained.  It is
important that the starter should not throw out of engagement when
the engine first  fires  (277).

The cranking speed can be increased under low temperature conditions
by changing to low viscosity lubricating oil. By changing from 10 to
5 SAE lubricating oil an increase of 1/3 in cranking speed can  be
obtained.

A sufficiently high compression ratio is vital for good starting and
indeed the ratio is  normally chosen from consideration of good starting
and low noise and  is set  at a higher value than would be desirable from

-------
                           4-39
considerations of best fuel economy.

A number of aids to assist starting are available in addition to the
provision of excess fuel which is universally employed.

    1.   Heaters installed in water jacket and/or oil sump.
         These are usually household current operated and
         provided the engine is isolated from the atmosphere
         so  that  a general area of warmth is generated around
         the engine, they are effective.

    2.   Heater  plugs installed in the combustion chamber.
         The term  heater is a misnomer as they really ignite
         the fuel rather than heat the chamber.  There are
        three types of heater plug: heavy gauge, shielded,  and
         thin wire. Tests showed that the ignition delay period
         was least with the heavy gauge and longest with the
         thin wire  (54).

         For satisfactory starting the element should reach at
         least 1000°C (1832°F).  (423)

         With swirl chamber engines the gain in starting with a
         heater plug can be as much as 30  C (54 F) and the
         lowest limit for starting with heater plugs is around
         -25°C (-13°F).  (423)

         A wait of  15 - 30 seconds is necessary for the heater
         plug to  warm  up. (112)

    3.   Heater  installed in the inlet manifold.
         Manifold heating  can be achieved by an electric heater
         or by combustion of fuel within the manifold. The
         probable gain in starting with an electric heater for
         a DI is  of the order of 10°C( 18°F) . (423)

         Combustion heaters such as the CAV Thermostart or
         Kygas burn fuel in the manifold and give gains in the
         order of 10 - 15°C (18 - 25°F)  on both DI and EDI.
         (423)

         With combustion heaters it is important to control the
         fuel quantity or excessive flows will starve engine

-------
                           4-40
         combustion of oxygen. (423)

         In IDI engines manifold heaters are less effective than
         heater plugs in the cylinder but may be easier and
         cheaper to install and can operate with smaller delay
         time. (277)

    4.   Aids based on fuels having low ignition temperatures.
         Starting fluids are usually based on ether with addition
         of lubricating oil and are very effective and can start
         DI or IDI to -40°C. (423)

    5.   Devices applied to injection equipment.
         Simple retard of injection timing  for starting and the use
         of Pintaux nozzles is employed with swirl chamber
                                               o   /  o
         engines which can improve starting by 10 C (18 F).
         (423)

    6.   Other Methods.
         Adding a small quantity of oil or gas oil to  the inlet mani-
         fold directly behind the inlet valve assists starting, by
         improving the seal of the piston rings and increasing
         compression  ratio. (423)

         Closing the inlet valve earlier gives an increase in
         effective compression ratio but will give a  penalty in
         engine performance at full speed.

         While Cetane  number is an important fuel quality for
         starting, at lower temperatures where a starting aid of
         some form is necessary, the Cetane number is less
         important.  (564)

         White/blue smoke on starting is reduced by using higher
         Cetane fuel (38, 19)

         White/blue smoke on starting is reduced by using more
         volatile fuel.  (38)

As mentioned earlier, some diesel fuels give trouble in vehicles in
cold weather due to clogging of fuel filters by wax crystals.  Improved
design of the fuel system can help as can the use of flow improvers
in the fuel or kerosene dilution of the fuel.

-------
                           4-41
Starting of Passenger Cars in the USA
For the first cold start of the day, no insurmountable problems are
foreseen for the diesel powered passenger car provided the correct
starting procedure is followed. If heater plugs are allowed to
achieve working temperatures of 10OO C (1832  F)(entailing a delay
of 2O - 30 seconds with current devices) a satisfactory start in
                                    0,0       o (
ambient conditions down to -20 to -25 C (-4  to -13 F) can be
achieved;  the additional use of simple ether-based aids extends this
range down to -40  C.  Care must be taken when using aspirated
special fuels with  a very high ignition quality (e.g. ether)  as
excessive amounts can cause damage due to extremely high cylinder
pressures and it is preferred that if this form of aid is to be used
regularly it should be  administered automatically.

White/blue smoke can  be a problem  for some engines immediately
after a cold start, particularly at light load when fuel atomisation
and mixing is often poor.  Any fuel introduced into the chamber late
in the combustion process is thus a  potential cause of blue  smoke,
and it is  mainly in the field of development of fuel injection equipment
and reduction of production tolerances on injection timing settings
that  improvements can be made-  On the combustion chamber side,
any feature which decreases warm up time will minimise the time
period during which blue smoke is likely to occur, and both Daimler-
Benz with their pre-chamber and Ricardo with the Comet Mk.V
incorporate insulated chamber members in order to reduce this
problem  of warm up time. Other means of cleaning up blue smoke
include increasing the Cetane number of the fuel or increasing the
compression ratio of the engine (already raised for starting consider-
ations far above the optimum performance requirement), or
increasing the exhaust back pressure, thus raising the mean load
factor and in-cylinder temperatures.

One subjectively annoying aspect of  starting small high speed diesel
engines is the delay while waiting for the heater plugs to warm up.
Prototype plugs with virtually instant warm up times have been tested
but these have always  encountered durability problems and it is
unlikely that an instant warm up plug could be developed to a production
stage within the next two or three years. A fast warm up plug (say
1O seconds) would be extremely desirable, or alternatively current
production plugs could be made more attractive if  a programmed
start were adopted in which the heater plugs were  activated manually
or automatically via some remote sensor/switch before the driver
gets into the car so that the subjective delay time is minimised.

-------
                          4-42
Even in warm ambient conditions,  e.g.  15  C  (59 F),it is desirable
to use some degree of pre-heat of the heater plugs in order to avoid
excessive cranking times (with fuel being injected and passing straight
through the engine) .  Thus some form of programmer which
determines the  minimum required pre-heat time to ensure an
immediate start would be advantageous. In order to predict required
pre-heat times both ambient and cylinder head temperatures would
need monitoring.

Thus the starting problem of the light duty diesel falls into the
following categories :-

1.    Subjective annoyance during delay while heater plugs warm
      up.  This  can be a problem even in mild ambient conditions.
      A possible cure is a simple programmed start device or the
      development of a fast warm up heater plug.  'Instant1 warm
      up heater plugs would eliminate this problem altogether but
      experience has shown durability and allied problems.

2.    Excessive cranking with a warm engine due to the driver over-
      riding the heater plugs leading to unburnt fuel being passed
      through the engine.  This problem could be eliminated by
      incorporating an automatic programming system  which senses
      cylinder head and possibly ambient temperatures and
      automatically selects optimum preheat time before the starter
      can be engaged.  The ultimate solution is a completely
      automated start.

3.    White/blue smoke - can be minimised by careful development
      of fuel injection equipment for low speed operation.  If this is
      not successful, other techniques such as raising the  exhaust
      back pressure thus increasing engine working load are known
      to be effective, but in this case it is accompanied by a fuel
      economy penalty.

4.    Starting during extreme winter conditions - in this respect the
      starting of the diesel is inferior to that of the gasoline engine
      in that a specific starting sequence must be followed to ensure
      a start - as in (2) automatic programming  devices may help
      to reduce the subjective annoyance but the delay is still there.

-------
                           4-43
Hot Driveability
With the same torque curve for the diesel as the gasoline engine,
driveability will be similar to that of the gasoline engine.  The diesel
engine normally has an all speed governor, the throttle control being
an engine speed control.  If desirable however it would be possible
to fit a load control together with a governor giving idling and over-
speed control.

Current flywheels are small enough to give easy and rapid gear
changes, particularly on six cylinder engines.  Transmission system
requirements are similar to those of the gasoline engine and no
special provisions need be made.
                          • *      : t
Diesel engined vehicles in production, even with four cylinder engines,
can be driven down to about 25 km/h(l5 mph) in top gear. Engine flexibility is
very good and hot driveability should not therefore present any problems.
(Fig. 16, Ref. Nos. 489, 648).

Driveability is unaffected by emission control measures, such as retard
of injection timing. (112)
Cold Driveability
Once the diesel engine is running it will deliver power unfailingly
without the danger of stalling or  hesitation which is a common fault with
gasoline cars if the accelerator is  depressed suddenly when the engine
is cold. Driveability is also unaffected by emission control measures
such as retard of injection timing. (112)
Torque Rise
In the gasoline engine the torque curve shape reflects primarily the
breathing capacity of the engine modified somewhat over the speed range
by the change of mechanical and indicated thermal efficiencies. The
diesel engine torque curve is influenced by the same factors but can
be radically altered by a change in injection equipment specification.
Additionally, changes in combustion efficiencies will be greater in the
diesel. At low speeds the maximum torque must be reduced,  in part
because of the higher proportionate heat losses and in part because of

-------
                           4-44
    mis-matching of the air and fuel movements resulting in an early
onset of exhaust smoke.  At high speeds the torque will hold up
within the limits of the breathing of the engine r

The ideal torque curve shape depends on the engine application. By
changing valve timings, injection timings or combustion chamber
characteristics the torque curve shape can be varied. (287)
(Fig. 4-17)

A torque curve back up of rather more than 10% would seem reason-
able for automotive diesel engines  whether normally aspirated or
turbocharged. The speed range over which the torque back up will be
achieved will be less with the turbocharged than with the normally
aspirated engine, in general 65% and 55% of maximum speed,
respectively.  (39l).

With Comprex supercharging the rated output can be increased by 40%
and at half speed, the torque can be increased to 70% above the
normally aspirated values.  By suitable matching, the peak torque can
be set at whatever speed is required between 50 and 75% of full speed.
(l76)(Fig.  4-18).

A naturally aspirated swirl chamber engine producing 96 kW (128 bhp)
from 4.8 litre (292 CID) at 66.7 rev/s (4000 rpm)  6 Bar (88  lb/in2
bmep) would have,  say,  10 - 15%torque back up down to a speed of
33.3 rev/s (2000 rpm).

The torque to be transmitted will depend on the engine type chosen. If
a high speed gasoline engine is replaced by a low speed diesel engine,
the increased torques may require a somewhat heavier transmission
but if the two engines run at similar speeds the same transmission
may be employed.
Durability
It is difficult to put definitive values on the durability of the high speed
diesel engine.  Most user experience is concerned with commercial
vehicle operation; trucks and buses, etc., but operators of passenger
car vehicles which have been surveyed indicate that their experience
may be similar to that of the larger vehicle operators.

    durability of the diesel engine is tied to wear rates,  since excessive

-------
                           4-45
wear is the normal reason for component failure.  This in turn may
be attributed to the duty for which the particular vehicle is subjected
and to the sulphur content of the fuel.  Research has shown that
diesel wear on start up is much less than for the gasoline engine but,
due to higher loadings, steady state wear may be higher (456) .
Overall durability, however, should be greater than for the gasoline
engine if only because of the higher quality materials and more
rugged construction required for the diesel engine  (15). Wear rates
of O.lmm/1000 hours  at top ring level and 0.0016  mm/1000 km for
liner wear are quoted  (100) if the liner temperature is kept  above
100  C (212 F).  The effect of fuel sulphur content was shown to
produce three times the wear rate when the sulphur content is
increased from .06% (desulphurised) to 1.4% (564).  The sulphur
content was also shown to have an effect on carbon deposits which in
turn affect durability.  Carbon deposits were also shown to be
affected by fuel Cetane number; reducing the number from 55 to 35
doubled the amount of  carbon (564).

Experience with taxis suggests 320,000 km (200,000 miles) on one vehicle
without attention.  Comparison of the durability of  fuel injection
equipment with spark ignition equipment shows the diesel equipment
to require less maintenance and to have a greater life (15).   The life
of injection equipment is variable but is quoted as 288,000 km
(180,000 miles) (100), and 176,000 km (110,000 miles)  (639).  In
this  period routine cleaning of the nozzles may be required but some
users do not consider  this necessary,  particularly for the self
cleaning pintle type of nozzle used in indirect injection engines.

The factor which determines the time at which the diesel engine should
undergo a major overhaul is cylinder bore wear.  When run on a low
sulphur fuel (i.e.  less than 0.4%), current European engines achieve
approximately 160,000 km (100,000 miles)  between overhauls and
during this time it may be necessary to inspect and clean the fuel
injectors once or perhaps twice.

Thus the durability of the complete engine package  is known to be at
least as good as that of the gasoline equivalent.   It is likely in fact
that  the diesel engine will have a durability that considerably exceeds
that  of the bodywork of the vehicle.

-------
                            4-46
Coolant Heat Losses
Heat transfer from the combustion gases is dependent on the temperature,
pressure and velocity of the gases, and in the diesel engine, on radiation
from carbon particles in the flame.  At high loads,  radiation plays an
important part, and while the mean gas temperatures of the diesel are
somewhat below those of the gasoline engine, the pressure and gas
velocities will be greater and as a result the heat losses  from an
indirect injection dieseL at high load are greater than from a gasoline
engine of the same power output.

At part load, radiation plays a much smaller part and the mean gas
temperatures in the diesel fall rapidly and as a result,  although the
pressures and gas velocities are not reduced appreciably, the diesel
engine has a lower heat rejection at low loads than does the gasoline
engine.

The differences are clearly brought out in Figure 4-19 from which it
can be seen that the heat loss to the coolant at full load can be 25%
higher for the diesel engine.  In view of this higher  heat loss, it is
normally necessary to fit a larger radiator than for the equivalent
gasoline engine where the size is often set by the need to prevent
boiling of the coolant under idling conditions with a high ambient
temperature. A radiator size of some 15% larger is necessary. (150).

The lower heat loss to the coolant from the diesel at light loads could
be a minor disadvantage in winter conditions in that it may marginally
affect cab warm up time with a cold engine.  Under  light load
conditions it may be desirable to fit shutters to blank off the radiator
to ensure a  sufficiently high coolant temperature to  give effective
defrosting of the windshield and to provide cab comfort.  (31)

It is of course possible to use air cooling for the engine.  This over-
comes the problem of leakage and coolant freezing.  (257)

There are disadvantages with air cooling in that:-

1.   With higher temperatures and the difficulty of getting effective
     cooling of hot spots, the air cooled engine is generally rated
     at a lower power.

2.   The hotter cylinder walls result in a lower volumetric efficiency.

3.   With no water jackets to absorb sound, the air cooled engine is
     noisier.

-------
                            4-47
4.    It is not so easy to provide vehicle heating.

The first of these is the most difficult, and the problems are accentuated
by the very high local heat flows which occur in portions of the cylinder
heads of indirect injection engines. There are in fact no air cooled
diesel engines employed in light duty service today.
Vibration and Torque Recoil
On approaching idling conditions from a higher speed on a diesel engine,
the exciting torque and the transmission of the engine mounting both
increase so that a rapid increase in the magnitude of the vibratory
forces applied to the vehicle frame is to be expected.  With the
gasoline engine the exciting force is substantially smaller and is
further reduced by throttling of the air intake  (367) .

After installing diesel instead of gasoline engines in £, 1 and 5 tonne
vehicles, driver reaction was that diesel vehicles were noisier and
suffered more from vibration (31).

If one makes the engine mounting soft enough to accommodate the
displacement involved at idling speeds, it is almost impossible to
avoid the engine passing through the natural frequency of the mounting
on starting up or stopping (15).

For these reasons very low natural frequencies of the engine on its
rubber mounting are chosen; i.e. below the lowest operational speeds
of the engine (367).

The use of automatic transmissions in vehicles has a beneficial effect
on the problem of torque recoil vibration by virtue of the fluid coupling
involved (15).

In general, very close attention to detail in the development of the
engine/body package is necessary to ensure an acceptably 'smooth'
vehicle, especially at low engine speeds and at idle, but Ricardo
believe that it is possible although it is never  likely to be as good as
with a throttled  gasoline engine.

-------
                           4-48
Manufacture
In overall design the automotive diesel engine is conventional and
could be made on the same production line as a gasoline engine.
Many gasoline engine parts could in fact be incorporated.  The -special
feature which the diesel engine demands above all others is the
reduction to a minimum of the dead space outside  the combustion
chamber.  Clearance between the piston and cylinder head must
therefore be kept to a minimum and it is chiefly for this reason
that a simple shape, i.e. a flat cylinder head with vertical valves,
is maintained.  The necessity for close tolerances on a large number
of parts in order to hold the piston to head clearance down to a
reasonable value will add to the cost of the engine as compared with
the gasoline engine. Other features of the diesel, apart from the
fuel injection pump drive, which differ from gasoline engine practice
result from the higher cylinder pressures  in the diesel cycle. Thus
there is a main bearing between each cylinder, the piston and
connecting rod are extremely sturdy and the cylinder head is held
on by a thicket of studs (626).

The compression ratio of a typical 2^ litre, 4 cylinder Comet engine
can vary considerably due to production dimensional tolerances. By
far the greatest contribution  to the variation arises  from the
dimensions controlling the piston head clearances (122).

With regard to ancillaries, all these are very similar to those
currently used on American gasoline engines (although some, such
as the starter motor and alternator,may need to be slightly larger)
apart from the fuel injection  equipment and a small  air pump (vacuum
or positive pressure).  The latter is needed  for the  power braking
system as the diesel, which does not throttle its intake air,  does not
have a convenient,  usable low grade pressure system as does the
gasoline engine.

Fuel injection equipment, although manufactured to very close tolerances,
is designed specifically for high volume production and is currently
manufactured in very large quantities on current technology equipment.
Lubrication
The internal combustion engine imposes a severe duty on its lubricant.
The lubricant must perform several independent functions including
the control of wear and corrosion yserve as a piston seal and coolant as

-------
                           4-49
well as flush away carbonaceous, partly burnt fuel residues from
critical areas.  It must also function satisfactorily over a wide
temperature range with both temperature extremes introducing
their own individual problems.

Diesel engine lubricants are formulated to control corrosive wear of
piston rings, liners and bearings and to prevent the fouling of various
engine parts by the build up of deposits (570).

All modern lubricants contain additives to improve their performance.
The following properties are normally bestowed on the lubricant by
the addition of additives: -

1.    Detergent/dispersant qualities so that insolubles such as
      soot or oxidation products remain in suspension in the oil
      and do not deposit on engine surfaces.   This is much more
      important in a diesel engine than in a gasoline engine.

2.    Anti-oxidant qualities to discourage lacquer formation in
      piston ring grooves  and consequent ring sticking  and to
      reduce the rate of viscosity increase caused by oxidation products.

3.    Anti-corrosive qualities to prevent the corrosion of bearing
      liner materials. Also adequate alkalinity to neutralise any
      sulphuric acid products that condense on the liner and hence
      to suppress corrosive wear of liner and rings arising from
      the higher sulphur content of diesel fuels.

4.    Extreme pressure qualities to reduce valve train scuffing,
      particularly  tappets and cams.

5.    Viscosity index improvements to give the oil multi-grade
      characteristics.

6.    Pour point depressants to prevent oil thickening at low
      temperatures.

7.    Anti-foaming and anti-rust properties.

It appears that prechamber engines are more severe from the additive
depletion aspect in their demands on heavy duty lubricating oil than
are their direct injection  counterparts.  (570)

-------
                              4-50
Oil change periods are normally dictated by additive depletion although
other factors can influence the change period,  e.g. contamination of
lubricant with fuel, coolant or abrasives (570).

A reasonable oil consumption for an automotive diesel engine is 0.5%
of its gross fuel consumption.

Analysis of oil samples after short engine runs both with and without
exhaust gas recirculation showed that the oil used during the test with
exhaust gas recirculation had a significantly higher insolubles content
than the oil from the non-EGR test, suggesting  that some engine wear
increase may occur as a result of EGR and that more frequent oil
changes will be required with EGR.
CONCLUSIONS

Smoke
Black Smoke - This should not be an aesthetic problem if the engine
complies with the Federal Smoke Regulations for heavy duty vehicles.
(This should be obtainable by attention to local mixing and the overall
air:fuel ratio at rated conditions.  Ratings can be controlled to take
account of altitude effects).

The high power;: weight ratio of the American car should mean that
visible smoke conditions will only be obtained for extremely short
periods during hard accelerations.

Turbocharged engines may have a low speed transient problem but
the 'Comprex' pressure exchanger might be a solution to this.
Blue/White Smoke - This can be unpleasant from the sidewalk
particularly as it is formed under idle conditions,  but the problem
can be minimised by careful attention to combustion chamber design
and fuel injection characteristics.


Odour
The small high speed diesel can have an odour problem, particularly
at light load conditions if misfire is approached.  The problem  can be
minimised by the addition of a light load advance mechanism.

-------
                           4-51
 Odour at full load can be minimised by combustion chamber development.
 The proposed reductions in smoke levels should alleviate this problem.

 The identification of several odorous components has been achieved
 but quantitative assessment has yet to be perfected.  The A.D. Little
 Odormeter may advance technology in this area by a significant step.
Gaseous Emissions
In general, turbocharging increases NOx further by increasing the
charge temperature but it allows further retard for the same smoke
limit.

Exhaust gas recirculation is effective in reducing NOx levels (particularly
over the CVS-CH cycle) but durability has yet to be proved and it does
tend to increase smoke emissions.

Although water injection has the benefit  of reducing NOx without
significantly affecting engine performance, the logistics of the installation
and the problems of engine durability make this measure unattractive.

Timing retard is undoubtedly the most effective single parameter for
the reduction of NOx and the fact that the smoke limited performance
of the IDI engine tends to deteriorate less with retard than the DI
gives it a major advantage in this field.

The limited data available indicates that emission levels from 2-stroke
engines should be of the same order as from 4-stroke engines of
similar performance.

Heavy duty experience leads to the conclusion that the use of a
conventional  direct injection chamber will increase both NOx and CO
levels while HC  levels might rise rapidly with retarded timings.  It
seems almost certain that a high speed (67 rev/s) (4000 rpm)
conventional, naturally aspirated direct injection engine would not
achieve the primary emission levels due to its low smoke limited
performance at retarded timings.

For a naturally aspirated 4-stroke indirect injection engine it can be
predicted that 3.4 g/mile CO can be achieved; 0.41 g/mile HC could
be attained on prototype vehicles although this figure may not be
held in production ; and 1.5 g/mile NOx could just be obtained from
a prototype current generation engine although some exhaust gas
recirculation may be necessary to allow a margin for production
compliance.

-------
                           4-52
Although 0.4 g/mile NOx has been achieved with a highly modified
prototype engine in a European type vehicle, it is extremely unlikely
that this figure would be achieved with a heavier vehicle and with a
higher power to weight ratio.

Any diesel powered vehicle would have less difficulty in achieving
the target objectives if both weight and power to weight ratio were
reduced.  A lighter, lower powered vehicle would also have improved
fuel economy.
Particulates
The early suggested Californian requirement would present a major
problem for all engines with a heterogeneous combustion system and
this is undoubtedly a problem area for the diesel engine.  However,
the true effect of particulates on health is unknown at the moment
and the problem could be solved by the addition of filter systems
although this move would carry a high cost penalty.

It is considered that work should be initiated into the distribution of
particulates from different types of engines and their true health
hazard determined before any legislation is finalised.  An understanding
of their formation within the engine might also be a useful tool for their
control and thus a fundamental investigation using experimental and
analytical techniques should also be started.
Noise
The drive-by noise levels of diesel powered vehicles are slightly higher
than those of gasoline powered vehicles but there is no reason why
light duty vehicles should not meet proposed noise legislation.

The idle noise is annoying to the by-stander as well as the driver with
present European vehicles.

There is no reason why diesel powered light duty vehicles should be
unacceptable to either the driver or by-stander if sufficient attention
is paid to details of the engine  and vehicle construction.
Volume
For the same power output the diesel engine is likely to be greater in
volume than a highly rated gasoline engine, but the increase in volume
is unlikely to pose any major problems.

-------
                              5-35

The main variables are listed below :-

           Engine type - 4 stroke
                        2 stroke (valve in head, loop
                        scavenged-opposed piston)
           Engine compression ratio
           Boost pressure
           Cylinder pressure limits
           Air fuel ratio
           Valve timing
           Turbine efficiency
           Compressor efficiency
           Use of charge cooler  - charge cooler efficiency
           Engine /turbo components drive arrangement
           Thermal loading limitations
           Bore/stroke ratio
Although there is a fairly extensive literature study of the subject,
the little practical experience that exists is concerned with large
engines for which this configuration is most suited.  The performance
predictions are also concerned with large engines and are mostly
confined to maximum output conditions without consideration for
part load operation.

In order to arrive at fairly broad estimates for the performance of
a small engine, the available information has been used and modified
where possible to take account of the scaling down of the components.
Engine Configuration

The simplest form of compound engine is that in which the output is
taken from the diesel engine with the turbo-compressor unit geared
to the crankshaft. The gearing has to cope only with the difference
in power between the compressor and the turbine, and there is no
requirement for power equilibrium between these components
such as exist in normal turbocharging.

-------
                               5-36
With a gas generator system all the power output is taken from the
turbine through reduction gearing.  For the relatively low power
requirements of the engine under study the small turbine would
have a very high rotational speed.  Although  this arrangement has
the theoretical ad vantage of a high-low speed torque compared with
the compound engine, it is not considered to  be feasible for the
present application in view of these high rotational speeds and
the unknown performance of small turbines at high expansion
ratios.
Engine Type
From the available data a 2-stroke engine is seen to offer the best
thermodynamic performance and power/weight ratio with an opposed
piston type being superior to the valve in head form.   It also permits
greater control over the division of power between the diesel engine and
the turbine compared with a 4-stroke engine.  However, although the
2-stroke engine is basically simpler it is considered that the 4-stroke
might be preferred since it has  much lower thermal stresses and
piston heat flow and has greater volume  of experience behind its
development.
Estimated Performance
The values tabulated below have been derived from data predicted
for large engines with the following assumptions :-

           Boost ratio 4:1- Inlet pressure  4.12 Bar ( 60 lb/in2 abs)
           Charge cooler effectiveness 80%
           Charge cooling water temperature 85 C
           Compressor efficiency 74%
           Turbine efficiency 82%
           Cylinder pressure limit 207 Bar (3000 lb/in )

The final column shows the engine swept volume for 96 kW (128 bhp)
at an engine speed of 50 rev/s.

-------
                            4-53
Weight
When compared with a highly rated gasoline engine of 97 kW (130 bhp)
a diesel engine of the same power output may be as much as 136 kg
(300 Ib)  heavier, but the weight penalty may be as little as 68 kg
(150 Ib)  if a more normal low rated gasoline engine is chosen for
comparison.
Fuel Economy
The diesel engine is the most efficient current and practical power
plant for light duty use.

Its fuel economy advantage over  the gasoline engine will increase
as the load is reduced and may be as much as 50% (twice the miles
per gallon) in city use.
Fuel
As fuel costs increase the high economy diesel engine becomes more
and more attractive.

Automotive diesel fuel production could be increased by up to twelve
times the current levels by 1980.

If it became necessary to introduce wider cut fuels noise could be
increased and cold starting worsened,  emission levels could also
increase.
First Cost
The first cost of the diesel engine will be significantly higher than that
of the gasoline engine for the same power output and emission levels.

Such evidence  as is available indicates that the diesel engine in
America will cost between 1.5 and 2 times  as much as the gasoline
engine.  For a 97 kW (130 bhp) engine this means some #  300 against
£ 200 production cost and about half of this  extra cost is due to the
extra cost of fuel injection equipment.
Maintenance
Minor maintenance requirements of the diesel engine (oil change,
filter elements, etc.) are equivalent to the gasoline engine although
the cost of these items may be slightly higher.

-------
                            4-54
The periods between major overhauls for the diesel are greater so that
over the life of the vehicle overall maintenance costs will be less than
for a gasoline engine-
Starting
Starting is inferior to the gasoline engine particularly in mild ambient
conditions although a programmed start would reduce the annoyance
of the 30 second (maximum) delay for heater plugs.

The light duty diesel engine will start down to -25 C (-1.3 F) with heater
plugs, but without external aids, and down to -40 C  (-40 F) with
simple external aids. Below this temperature more complex external
aids are required.
Hot Driveability
For engines of equivalent power the low speed torque will be greater
with the diesel than with the gasoline engine.

Although the operational speed range of the diesel is less than that
of the gasoline engine its low speed smoothness is frequently better.
Cold Driveability
The cold driveability of the diesel engine is as good as its hot drive-
ability and is considerably superior to that of the carburetted gasoline
engine.
Torque Rise
The gasoline and diesel engines will have similar torque characteristics.
If the diesel engine runs at an appreciably slower speed than the
gasoline however, the high torques may require a slightly heavier
transmission.
Durability
The durability of the diesel engine will be greater than that of its
gasoline counterpart.  The diesel may in fact be too durable  and out-
live the vehicle, but it is difficult to reduce engine  life without
seriously impairing durability and reliability.

-------
                            4-55
The durability of the diesel fuel injection system is greater than that
of the gasoline engine's ignition and carbitration system.
Coolant Heat Loss
The reduced heat loss at idle and low load will reduce the problems
of * traffic- jam -boil-over1 but may make winter morning de-icing
and warm-up slightly more difficult.

The increased heat loss to coolant  at full load may dictate the use of
a larger radiator, but this will possibly only be required for vehicles
towing trailers.
Vibration and Torque Recoil
The torque recoil from the unthrottled, high compression ratio
diesel is undoubtedly greater than from the gasoline engine, particularly
at idle.  This tends to give the impression of harshness which disappears
once the vehicle is in motion.

The problem can be minimised by attention to the engine mountings.
Manufacture
The diesel engine is sufficiently similar to the gasoline engine to allow
it to be made on the same production line if necessary, although
greater control of tolerances is required.

Ancillaries  are all similar to those of the gasoline engine and in many
cases are identical.

The production techniques for the high volume production of fuel
injection equipment have already been developed in Europe although
there may be some scope for cost reductions with even greater
production quantities.
Lubrication
The diesel engine demands lubricants with higher dispersant and anti-
corrosive properties  than the gasoline engine.  With present day oils
this implies that oil changes might be slightly more frequent.

-------
                            4-56
Areas likely to Benefit from further Work
1. Emissions
A fundamental attraction of the diesel engine is that its emission
characteristics can avoid the use of catalysts.  A careful study of
the origins of the controlling parameters in the formation of unburnt
hydrocarbons in light duty diesels would yield much valuable
information and perhaps extend the use of non-catalyst controlled
power plants.
2. Particulates
Current ignorance on the spectrum of particulates from a given
engine, the relative dangers of particulates of various sizes and
compositions, and the effects of blanket legislation would indicate
that work should be  in:' ated into these unknown areas before any
legislation is passed.
3. Starting
Further work on 'instant1 glow plugs and programmed starts may
increase the acceptability of the diesel engine.
4. Fuel Injection Equipment
About 50% of the cost differential between diesel and gasoline engines
is due to fuel injection equipment while both emissions and noise
might benefit from more versatile injection systems.  Because of
this research into cheaper and more versatile fuel injection equipment
should be considered.

-------
                                       EPA PROJECT RFP WA 73-R176
                         BAR CHART OF INSTITUTION AND CONFERENCE  PAPERS  SURVEYED
    PUBLICATION









SAE TRANSACTIONS




INDIVIDUAL SAE PAPERS




I.MECH.E. PROCEEDINGS




I.MECH.E. A.D. PROC.




J.S.M.E. BULLETIN




J.S.A.E. BULLETIN




CIMAC CONFERENCE PAPERS




FISITA CONFERENCE PAPERS




MIRA JNCLASS. REPORTS




MIRA V.A. REPORTS




MIRA E.A. REPORTS




ASME PAPERS




ASME TRANSACTIONS




ASME JOURNAL
                                                                   YEAR
50















51










^_




52















53















5*















55









!





56















57















58















59















60












,
f


61















62















63















6*4















65















66















67















68















69















70















71















72















73

















-------
                                       trM
                                                   r\rr- WH
                                     BAR CHART OF JOURNALS SURVEYED
       JOURNAL





MTZ


ATZ


DIESEL 6 G.T. PROG.


DIESEL 6 G.T. PROG. W.W. ED.


DIESEL & G.T. PROG. U.S. ED.


SAE JOURNAL


C.M.E. (I.MECH.E. JOURNAL)


I.MECH.E. A.D. JOURNAL


AUT. DESIGN ENGINEERING


SI A JOURNAL


AUTOMOBILE ENGINEER


GAS & OIL POWER


AUTOMOTIVE INDUSTRIES


MECHANICAL ENGINEERING

   (ASME JOURNAL)


G.M. ENGINEERING JOURNAL


COMBUSTION ENGINE PROGRESS
YEAR
50

















51

















52

















53

















5*4

















55

















56

















57















E.

58

















53

















60

















61

















62

















63

















6/4

















65

















66

















67

















68

















69

















70








,








71

















72

















73

















                                           i
                                           
-------
                     4-59
                                       FIGURE 4-3
A STUDY OF THE DIESEL AS A LIGHT DUTY POWER PLANT
     LIST OF EUROPEAN DIESEL VEHICLE OPERATORS
     CONTACTED TO OBTAIN USER EXPERIENCE DATA



     OPEL, RUSSELSHEIM, ¥. GERMANY.

     PEUGEOT,  PARIS, FRANCE.

     DEUTZ,  PORZ, W. GERMANY.

     B.L.M.C.,  LONGBRIDGE, ENGLAND.

     B.P., LONDON, ENGLAND.

     C.A.V.,  LONDON,  ENGLAND.

     LONDON GENERAL CAB CO., LONDON, ENGLAND,

     VERBAND FUER DAS PERSONENWERKEHRSGEWERKE,
          HAMBURG, W. GERMANY.

     SLOTA TAXIS, PARIS,  FRANCE.

     AUSTIN HIRE & TAXI SERVICE LTD., WORTHING,
          ENGLAND.

-------
                               4-60
                                                      FIG. No. 4-4
                                                      Drg. No 0 2-GtOCtB
                                                      Date   £8-7-74
         2  LITRE  D.I. CONVERSION
PERFORMANCE COMPARISON - DJ. AND COMET  V  BUILDS
  EXHAUST SMOKE  MEASUREMENTS AT OPTIMUM CONDITIONS.
FIXED STATIC PUMP TIMING
OPT. SWIRL 55 rev/s  FOR D.I.
                                    JbARO; D.!. 7£4 *r»m Hg.
                                           COMET 77O mm

-------
COMPARISON
                      t,     <             FIG. No. 4-5

                      4-61                DRG.No. D26203

                                         DATE*-  JULY 74

       OF  FULL  LOAD  CHARACTERISTICS FOR  THREE
COMBUSTION  CHAMBERS  ON 2-LITRE  4-CYLINDER  ENGINE
                                   i

       (FROM PAPER  BY  DR. EISELE - DAIMLER  BENZ)
D.I. - DIRECT INJECTION

P.C - DAIMLER BENZ  PRECHAMBER

S.C. s SWIRL  CHAMBER   COMET  V
                                     BMEP
I- 5
8*
o
CD

 I
D.l.y



PC.



S.C,
                  EXHAUST  SMOKE
                                                     HO i
                                                     IOO-
                                                     9O -
                                                     BO-
                                                      70J
                                                 -7-5





                                                 - 7




                                                 -6-5
                                                             o
                                                             .0
                                                        h 6  I
                                                  55
                                                         - 5
                  FUEL  CONSUMPTION
      I OOP
          i

         2O
                 rcv/min
            2OOO        3OOO

                     	1	
4OOO
          3O      4O      5O     6O

             ENGINE  SPEED-r«v/s
                                                    •=0-5-
                                                     O-4-
                                                    0-3-J
                                                         -35O
                                                         -300
                                                             O
                                                             UJ
                                                         -25O
     U
     ui

•200ft
   7O

-------
                              4-62
                                                    FIG. No.  .4-6

                                                    DRG No. D as71 3 A

                                                    DATE:-  £8-7-74
           NO EMISSION  AND PERFORMANCE
     CHARACTERISTICS OF D.I. AND SWIRL  CHAMBERS.
AVERAGE DATA FROM NATURALLY ASPIRATED ENGINES AT

2.1:1 AIR : FUEL. RATIO & PISTON SPEED CIRCA  G-Sm/SEC.
E
q
d
O
   asoo
  aooo
   I5OO
1000
   50O
      oL
   350
O
U.
   £50
   aoo
                              4 SPRAY D.I.

                              SWIRL CHAMBER
                             NO
              SPECIFIC CONSUMPTION
                'EARLY
                                      LATE
                                                              0)
                                                              I

                                                              
-------
                                                              FIG. No.   4-7
                                  I^b£.2!i.f£5£9Sl^^  Drs NO.  D a&c
&  SMOKE OF A, SIX  CYLINDER  SWIRL CHAMBER          Date    £8-7-74
   EFFECT  OF  EXHAUS
             F3 C~ .* i. j- #•' i -~ *«*-,  ,*> A. ft   /* f»i ,»*i ;  «" f*«,  "«• /-^,  r" .^> £*,.»  ; j .« -• (5 P« '
             K&L.."< w-ufctJ  L**X£»  L-OC-u.fc.lJ   i 
-------
                             4-64
      THE  RESPONSE OF NATURALLY ASPIRATED
         ENGINE  NOX EMISSIONS TO  LOAD.
              FIG No   4-6
              Drg. No.  OZ6O&9
              Date  38-7-74
GCYL.
                    L COMET V ENGINE
	G CYL. 5-40 L Dl  ENGINE.
1   FULL SPEED,
j   RETARDED £°
TIMING

-------
                          4-65
      THE RESPONSE  OF TURBOCHARGED
                                                   FIG. No 4-9
                                                   Drg. No. DZ&O70
                                                   Date   £8-7-74
ENGINE NOx  EMISSIONS TO LOAD.
         GCYL. G-81  COMET  V ENGINE
	G CYL. 5-41 L  D.I. ENGINE
}
                                      FULL SPEED, TIMING
                                      RETARDED 6°.

-------
VEHICLE
MERCEDES 220D
* MERCEDES 220D
OPEL REKORD
PEUGEOT 504
DATSUN 220C
DATSUN 330C
INERTIA
WT
3500
3500
3000
3000
3500
4500
SWEPT
VOL. IN3
134
134
127
129
132
198
COMB.
CH.
PRECH
PRECH
COMET
COMET
COMET
COMET
TEST
LAB
EPA
EPA
EPA
EPA
EPA
EPA
G'BOX
4 AUT
4 AUT
3 AUT
4 MAN
4 MAN
4 MAN
NO. OF
TESTS
5
5
4
5
2
3
HC
g/mile
0.34
0.28
0.40
3.11
0.21
1.70
CO
g/mile
1.42
1.08
1.16
3.42
1.69
3.81
NOX
g/mile
1.43
1.48
1.34
1.07
1.72
1.71
MPG
(US gall)
23.6
24.6
23.8
25.2
24.0
21.4
1/100 km
10.0
9.5
9.9
9.3
9.3
11.0
                                                                                                 E
                                                                                                 o
                                                                                                 x
                                                                                                 H

                                                                                                 d
                                                                                                 c
                                                                                                 H
                                                                                                 w
                                                                                                 f
                                                                                                 O

                                                                                                 £

                                                                                                 M
                                                                                                 2
                                                                                                 h— I
                                                                                                 W
                                                                                                 CO
                                                                                                 HH

                                                                                                 §



                                                                                                 s
                                                                                                 w
* MODIFIED F.I.E. - ALL OTHER ENGINES IN PRODUCTION BUILD (NOTE DATSUN 330C

  WAS A NON-STANDARD MARRIAGE WITH A 1973 FORD F250 PICK-UP)
-J
en



3
T3



f)
                                                                                                 O

                                                                                                 2C
     2
     S'



     0)


     i
     h-k
    To

-------
Engine Make
and Model
Peugeot XLD
Prototype ®
BLMC 1.5
Perkins 4.108
Prototype
Peugeot XDP4.88
Isuzu C190
Opel 2100
Prototype
Peugeot XDP4.90
Nissan SD 22
Citroen CRD90
Mercedes 220D
Land Rover 2\
Ford York
Mercedes 240D
BLMC 2.5
Mitsubishi 4DR50
Bore x
Stroke
mm
75 x 71
78 x 71
76 x 76.
73 x 89
79 x 89
80 x 89
88 x 80
86 x 84
88 x 85
86 x 90
90 x 83
83 x 100
90 x 85.5
87 x 92.4
90 x 89
94 x 86
91 x 92.4
89 x 101.6
92 x 100
Swept
Volume
1
1.25
1.36
1.39
1.49
1.76
1.80
1.95
1.95
2.07
2.09
2.11
2.16
2.17
2.20
2.29
2.36
2.40
2.52
2.66
Power
Output
kW
(DIN)
30
33.5
27
29.5
36.5
38
45
46 gross
45
49 gross
48.5 .
48.5
46
45
50 gross
46
48
48
56.5
Rated
Speed
rev/s
83.3
83.3
66.7
66.7
66.7
66.7
75.0
73.3
73.3
70.0
75.0
66.7
75.0
70.0
66.7
60.0
70.0
58.3
58.3
Weight
kg
152K
216
185
208
174
170
205
199
208
185
186
206
184
243
222

268
261
kg/1
122*
155
125
118
97.5
87.4
105
96.5
99.5
87.5
85.5
95
83.9
106.5
94.5

106
98.2
kg/kW
5.10*
7.80
6.29
5.70
4.58
3.80
4.44+
4.45
4. 65"*"
3.82
3.84
4.49
4.11
5.39+
4.80

5.62
4.60
kW/1
23.7
24.8
19.8
19.7
20.7
20.4
23.0
21. 4+
21.6
21. 6+
23.0
22.4
21.0
20.4
19. 7+
19.6
20.1
18.9
21.3
® Included for reference, circa 1956
Weight includes integral gear box



Assuming DIN power 10% less than
                                                                                                      d
                                                                                                        g
                                                                                                      M



                                                                                                      O
                                                         o
                                                                                                      w
                                                                                                      to
                                                                                                      d
                                                                                                      H
                                                                                                      O

                                                                                                      2


                                                                                                      §
                                                                                                      
-------
4-68
  COMPARISON OT DICTSEL AND GASOLINE  ENGINE
            WEIGHT/SWCPT VOLUME

GASOLINE CUBVg TAKEN FROM I. MgCM-g-PAPEB C34S/73
                             FIG No.  4-12
                             Drg No 0246J2A
                             Date  0.1.74

-------
             4-69
DIESEL V. GASOLINE WEIGHT ANALYSIS
      FOR 4 CYLINDER ENGINES
                                           Figure 4-13
Component
Combustion &
Ignition
Equipment

Cylinder Head
(Cast Iron)
Diesel Gasoline
Item
4 Cyl. DPA +
Delivery Valves
4 Injectors
Complete
4 Heater Plugs


Statistical
Weight
kg
7.13
1.85
.34

9.32
25
Item
Distributor
Lucas
Plugs + HT Lead
Coil Lucas LA12
Carburettor

Statistical
Cylinder block difference due to increased height etc.
Pistons & Con.
Rods
Crankshaft
Flywheel

Starter Motor
Vacuum Pump .
Battery
Sound absorption
material
Vehicle


4 Units

Insufficient
statistical
data - estimated
only
1 Lucas M45G

1 off Pierburg
1 Lucas BT11A
72 amph
Estimated

8.1

16.5

10.5

1
29.5
2

4 Units

Insufficient
statistical
data - estimated
only
1 Lucas 2M100

None
1 Lucas BT7A
43 amph


Weight
kg
1.13
.79
1.47
3.84
19

5.9

12.3

7.9

-
19.5


Diesel
Penalty
kg



5.5
6
2
2.2

4.2

2.6
17
1
10
2
13
ESTI MATED TOTAL PENALTY 35.5
Mercedes 220D
Peugeot 504D
Peugeot 204D XLD
1375
1279
955
Mercedes 220
Peugeot 504
Peugeot 204
(1130 cc)
1335
1230
935
40
49
20

-------
                        4-70
POWER/LITRE OF  SOME  CURRENT
   GASOLINE' & DIESEL  ENGINES.
FIG. No.   4-14
Drg.No. DZ6O7I
D.te   £8-7-74
    O  GASOLINE  ENGINES
    *  DIESEL  ENGINES
            ENGINE CAPACITY - Litr«

-------
4-71
FUEL ECONOMY - m.p.g. (U.S.) FUEL CONSUMPTION - L/IOO Km. ._
FIG. No. 4-15
DRG. No. D26O9
VEHICLE FUEL CONSUMPTION Vs INERTIA 	 _ _
VEJGHT DURING LA4 1975 (CVS-CH) TEST CYCLE.
I

ao
15
10
5
£5
10
15
10
5
1C
+ DIESL P

JRES FOR 1957- &7 GASOLINE VEHICLES.
JRES FOR 1973 GASOLINE VEHICLES.
OWERED VEHICLES- EX EPA DATA.




500 1000
VEHICLE INERTIA W






xvs^
X



.^'"
«
: t
isoo aooo
EIGHT -kg.

• t^m

^^^^rr-

3
74
i
>oo aooo 3000 4000 sooo
VEHICLE INERTIA WEIGHT- Ib.

"\


-------
                     •*- I £.
                                               FIG. No. 4-16
                                               DRG. No. S5802
                                               DAU:-   31-7-74
 0.25
BLOCK
       GASOLINE
0.10
THROW
1

/



                                      DIESEL
                                             0.050
                                            CON ROD
                                            CENTRES
  0.15
CON ROD
CENTRES
           TOLERANCES CONTROLLING  PISTON_TO
             CYLINDER  HEAD  CLEARANCE  MM.' "
    FOR ENGINES  OF b5 m/m  TO IIOm/m  STROKE
                                             0.100
                                             BLOCK

-------
              4-73
                                      FIG. No A-H
                                      Drg. No. D2GO7£
                                      Date   Z8 - 7 - 74
EFFECT OF INLET VALVE TIMING ON
LOW SPEED  TORQUE.
   NLET CLOSING
                               - rev/s
          ENGINE SF

-------
                              4-74
                     TORQUE - COMPARISON!


             ON A. BOO HP  Dl-DIESEL   ENJQNE
                                                      FIG. No  4-18

                                                      DRG. No. 025 7 22

                                                      DATE.-
    IOO
 o
8-
01
3
a
(*
o
M^XIDYNE TC.y* \

(MTZ 1912)      \
                                                    NATURAL ASPIRATED

-------
                           4-75
    HEAT  REJECTED TO  COOLANT FOR  SIMILAR
DIESEL & GASOLINE ENGINES OVER THE LOAD AND
                SPEED  RANGE
                     FIG. NO 4-19
                     Drg. No D 26O73
                     Date  £8-7-74
              DIESEL.
	GASOLINE.
         SWEPT VOLUME, BOTH ENGINES = I • 49 L
                                 ES OF CONSTANT*
                                 HEAT REJECTED
                                 BRAKE OUTPUT
                             ,.
                           , i , i I . . . ,X4 . . . - , .....   -t
                           '• • - • I ! ' ' t  • I : • * • < i  i t • ; -i
                           • • ' • I • t • I  < . • ^ • • ,  » ..... • 4
                     IG-7rev/s J
             ENGINE BRAKE OUTPUT

-------
                            5 - 1

                         SECTION 5
              ENGINE CONFIGURATION STUDY
This section contains details of the diesel engine configurations which
were schemed as potentially viable light duty diesel power plants.
Information on two comparable gasoline power plants is also included.

The diesel power plants were all designed to propel the target vehicle
for the study, a 4-5  seat sedan with a loaded weight of about 1600 kg
(3500 Ib) capable  of 0-96 km/h (0-60 mph) in 13.5 s  and 32-112 km/h
(20-70 mph) in  15 s.  Computer calculations indicated that a bare
engine power of about 97 kW (130 bhp) was required if a conventional
3-speed automatic transmission,  was assumed.

The engines included in the configuration study were as follows :

      U    V-8 Gasoline 4^stroke

      2.    IL-6 Gasoline 4-stroke
      3,    V-8 Indirect Injection Diesel 4-stroke

      4.    6 Cylinder Indirect Injection Turbocharged Diesel 4-stroke

      5.    6 Cylinder Indirect.Injection 'Comprexed1 Diesel 4-stroke

      6.    6 Cylinder Direct Injection Turbocharged Diesel 4-stroke

      7.    6 Cylinder Direct Injection 'Comprexed1 Diesel 4-stroke

      8.    6 Cylinder Indirect Injection Loop-scavenged Diesel 2-stroke

      9.    6 Cylinder Direct Injection Uniflow Diesel 2-stroke
     10.    4 Cylinder Direct Injection Compound Diesel 4-stroke

     11.    2-stage 2-bank Rotary Diesel  ' 4-stroke'

Drawings and performance curves were prepared for all the diesel power
plants which were schemed  for the primary emissions environment,
i.e.   HC - 0.41 g/mile,  CO -  3.4 g/mile,  NOX -  1.5 g/mile.
No diesel configurations were prepared for  the secondary emissions
target (HC - 0,41 g/mile,  CO -  3.4 g/mile,  NOx - 0 .4 g/mile)
since it was considered that no current technology diesel engine could
meet this NOy target.

-------
                          5-2
Introduction

The literature survey and Ricardo in-house knowledge both indicate
that a diesel engine would provide a viable power plant for a
passenger car.  Given this general conclusion the study required
that the most promising diesel variants be schemed in sufficient
detail to allow a reliable assessment of their potential.
Requirements of Vehicle

For the purposes of this study two target emission levels were
envisaged.  These were :-

      Primary  (or short term)

                   HC                 Oo41g/mile
                   CO                 3,4  g/mile
                   NOx                1.5  g/mile

      Secondary  (or long term)

                   HC                 Oo41 g/mile
                   CO                 3<,4  g/mile
                                       °-4
The vehicle for these environments was to be a passenger car
(typically a 4/5 seat sedan) weighing less than 1600 kg (3500 Ib)
test weight and capable of meeting the EPA standard car performance
specifications, i.e.  0-97 km/h  (0-60 mph) in less than  13.5 s,
32 - 112 km/h.  (20-70 mph) in less than 15 s ,  capable of over-
taking a 80  km/h   (50 mph) truck in less than 15 s.

To assist in the definition of the power plant a computer program was
written to allow the  maximum power output of the engine , the shape
of its torque curve and the transmission system necessary to achieve
these acceleration capabilities to be calculated.

Use of this program  showed that with a three-speed automatic gearbox
and the final drive ratios selected to give 136 km/h (85 mph) top speed
a bare engine power  of 97 kW (130 bhp) was required at the rated speed
to satisfy the acceleration target.   This  implies that the vehicle has

-------
                          5-3
 acceleration "in hand1 at its designed top speed and this is indeed true
 of the diesel power plants which will have a governor cut-off at their rated
 speeds in order to protect the engine.  Various gear ratios and amounts
 of torque back up  were tried and the minimum power plant requirements
 were finally selected as  :-

            Rated Speed                 96 kW         128 bhp
            .75 Rated Speed            76 kW         102 bhp
            .5  Rated Speed            53 kW           71 bhp
            .25 Rated Speed             17 kW           22.7 bhp

 i.e. 96 kW ( 128 bhp) engine with 10% torque back up at 50% speed and
 a 4:1 speed range.
Feasibility of achieving Target Emissions

While each and every type and layout of diesel engine will have problem
areas when used as a passenger car engine, the only area in which
performance targets have been set is that of exhaust emissions.

As has been explained in the report on Task 1, the great majority of
diesels emission data available refers to tests carried out on engine
test  beds under the 13 mode test procedure and in the experience of
Ricardo and others it is not possible to correlate this data directly
with results which would be obtained with CVS-CH testing.   It would
in fact be possible to estimate  the results of cycle testing from ' maps'
of data collected under steady state conditions but such maps are not
available for many of the engines of interest.

It is estimated however that a  128 bhp indirect injection engine
fitted in a 1600 kg (3500 Ib) car would require an NOX reduction
of some 30-40% from current levels in order to achieve the
1.2 to 1.4 g/mile necessary to guarantee 1 = 5 g/milejin a production
car.  Such reductions should be possible, with some penalty in
fuel economy, by retarding the injection timing but it might be
necessary in some cases to employ modulated exhaust gas
recirculation.

Ricardo do not believe however that it is possible with current
knowledge to  achieve the secondary target levels of 0=4 g/mile NOX
with a diesel  engine fitted  to the specified vehicle.  No attempt has
therefore been made in this report to rate any of the diesel  engine
designs at any tighter emissions target than 1.5 g/mile NOX.

-------
                          5-4

GASOLINE ENGINE

While no engine configuration studies have been carried out on
gasoline engines, it was necessary to have estimates of performance
and package size for comparison with the various diesel engines.

Two gasoline engines were studied, the first being a V-8 following
current American practice both in design and performance, and the
second a 6 cylinder engine more typical of European practice.

V-8 Gasoline Engine
Specific Performance and Emissions
Using typical specific performance levels for this type of engine, i.e.
21.0 -22,0 kW/1    (0.46 - 0.48  bhp/CID) , the exercise target
power output demands a swept volume  of 4.5 1  (275 CID).  This
capacity in conjunction with a rated speed of 6607 rev/s (typical of
most standard American engines)  produces 96 kW (128 bhp)  at a
brake mean effective pressure  of 6.26 Bar (92 Ib/in )  and a peak
torque of 285 N. m (210 Ib.ft) at 41. 6 rev/s gives 25% torque back
up resulting in satisfactory driving characteristics with a simple
three or four speed transmission.   The torque curve for this engine
can be seen in Figure 5-1.

The precise engine specification considered is as follows :

           0 97  mm (3.82")  x  76  mm  (3oOO") x 90°,  V-8
           4.5 1  (275 CID)
           96 kW (128 bhp) at 66.7 rev/s  (6026 Bar, 92 Ib/in2 bmep)
           285 N.m (210 Ib.ft)  at  41.6 rev/s (7.83 Bar,  115 Ib/in2 bmep)
           Estimated weight 250 kg (550  lb)»

The above performance  levels can  be achieved with the engine in low
emissions build  but external hang  on emissions control devices are
necessary, their complexity depending on the severity of the emissions
target. The primary project targets of 1.5 g/mile NOX,  0°4 g/mile HC
and 3.4 g/mile CO in a  1600 kg  (3500 Ib) inertia weight car  can be
achieved by using close tolerance  sophisticated carburettors in
conjunction with modulat.ad exhaust gas recycle, air injection into the
exhaust and an oxidation catalyst.  All these devices can be considered
current technology and in this case the trade-off for low emissions
is well known (increased power plant  weight, first cost,  maintenance

-------
                           5-5

 cost, a demand for lead free fuel,  and a depreciation in vehicle fuel
 economy).  The load range fuel consumption curves shown in Figure 5—2
 are estimated for this engine in 1.5 g/mile NOX build.

 The secondary project emissions objectives of 0»4 g/mile NOX while
 maintaining HC and  CO levels demand the additional  use of a
 reducing catalyst and possible  further catalytic devices (getter box)
 to protect the reducing catalyst from oxygen spikes.  Durability of
 such systems has yet to be proven although recent developments
 appear promising (Gould 3 System).  Due to the need to run with
 low oxygen concentration in the exhaust (less than 0.5%)  it is
 necessary to run with  a mixture strength approaching stoichiometric
 conditions with a resultant fuel economy penalty.  Depending on the
 exact baseline emission  levels of the engine, it may  be possible to
 reduce EGR levels compared with the 1.5  g/mile NOX build with
 resultant increases in thermal efficiency, particularly at the light load
 end of the operating  range. The combined effect of these two changes
 is likely to be a small loss in vehicle fuel economy compared with
 current vehicles.

 From EPA data, a fuel consumption level of 18  1/100  km (13 mpg) has
 been assumed for the gasoline powered vehicle when  in 1.5 g/mile NOX
 build.  Reducing NOx emissions to 0.4 g/mile will probably increase this
 to a level in the region of 19 1/100 km (12.5 mpg)»

 From data on current American V-8 engines, engine  weight will be
 250 kg (550 Ib).

 Using established noise prediction formulae, bare engine noise levels
 on a test bed would be  95 dBA at rated load and speed»  This correlates
 with a drive-by noise level of 74 dBA at 15.23 m  under standard
 American test conditions,,
6 Cylinder Gasoline Engine
The first and most obvious difference between American and European
engines of this power range is that of specific performance.  Whereas
the American V-8 gasoline engine runs to around 67 rev/s and develops
21 kW/1  (0.47 bhp/CID), the European engine runs to at least 83 rev/s
and develops a minimum of 32 kW/1 (0.7 bhp/CID)  normally nearer
36 kW/1    (0.8 bhp/CID).    The major reason for these vastly differing

-------
                           5-6

philosophies is one of economics;  mainly because of taxation
European fuel prices have always been high in terms of real spending
power with a resultant continuing trend towards small capacity, high
economy,  engines and cars.  On the other hand, the price of fuel in
America has always been low and fuel  consumption has never been a
major consideration; engine size has  therefore increased over the
years to improve driveability and to some extent as a sales feature.

A total power requirement of 96 kW  (128 bhp) and a torque back up  of
20% to enable the prototype vehicle to achieve  its target performance
could be achieved  easily with a 3 litre, six cylinder engine running  to
83.3 rev/s.  A typical oversquare cylinder configuration of 88 mm  bore
x 82 mm stroke (3.46"  x 3.22") has been selected.  The use of petrol
injection will ensure good distribution  between cylinders and should
allow the engine to run at generally leaner mixture strengths than if
carburettors were used.  Estimated performance in the form  of torque
curves and load range fuel consumption loops can be seen in Figures
5-3 and 5-4.  The  load range consumption curves have been based on
levels which are currently being achieved by current production engines.
Note that fuel consumption levels at light load conditions (on a bmep
basis) are similar to those of the V-8  engine.  At high load factors
most current V-8  engines still suffer a throttling effect (often the
intake valves or ports are restrictive in order to further improve
low speed torque)  with a resultant fuel consumption penalty whereas
the more high tuned six cylinder engine with  its need for improved
breathing characteristics should return low fuel consumption at these
conditions.

These performance figures and the project primary emissions targets
could be achieved  using the same external hang-on control devices  as
the V-8 gasoline engine, i.e.  modulated exhaust gas  recycle, air
injection and an oxidation catalyst.

Again, as in the case of the V-8 engine, the more severe NOX levels
of Oo4 g/mile should be achieved by the addition of a reducing catalyst
system to the above package with the same order of magnitude increase
in vehicle fuel consumption levels compared  with the 1.5 g/mile NOx
build.

Overall fuel consumption levels during CVS-CH testing have been predicted
at 15.5 1/100 km (15 mpg). This results in a total fuel consumption saving
of over 10% compared with the  V-8 gasoline engine.

-------
                         5-7

From European data, the predicted engine weight is 186 kg ( 410 lb),
a reduction of 64 kg ( 140 lb) compared with the V-8 engine.

Estimated drive-by noise levels of this engine in the prototype vehicle
are 77 dBA at 15 m (50 ft).  The reason for theSdBA increase in
drive-by levels compared with the V-8 is the higher rotational speed
of the 3 litre engine.

The completed specification for this engine is  :-

           0 88 mm (3.46") x 82 mm  (3.22")  x 6 (in-line)
           2.991     (183 CID)
           96 kW (128 bhp) at 83.3 rev/s
           232 N.m (171 Ib.ft) at 50 rev/s
           Estimated  weight 186 kg (410 lb)

-------
                         5-8
NATURALLY ASPIRATED IDI ENGINE

Specific Performance and Emissions
Of all the diesel candidates, this configuration is the closest to
current technology and as such its emissions and performance
predictions are those in which Ricardo have the greatest degree
of confidence.

It would be possible to use either a swirl chamber or a prechamber
of the form used by Daimler-Benz. The swirl chamber has a
higher air utilisation at high speeds however and its smoke level
drops more markedly as load is reduced, avoiding the 'plateau
smoke1 at low load which can be a problem with prechambers  It
was decided therefore to use a  swirl chamber for the design study.

The effects of vehicle inertia and engine swept volume have already
been calculated with the conclusions that this candidate in a 1600 kg
(3500 Ib) passenger car would  give approximately 2 g/mile NOX
during CVS-CH testing, with the engine in optimum performance
build.  Retarding the injection timing by 6 - 8 (crank) should lower
NOX levels to 1.2 - 1.5 g/mile, and it is at this retarded timing
condition that engine performance characteristics have been
calculated.  No other devices should be necessary to achieve the
1.5 g/mile NOX target although a modest degree  of exhaust gas
recycle might be necessary if large production compliance type
safety margins were demanded. The CO target of 3.4 g/mile should
be achieved without any modification to the engine but some fuel
pump/injector development might be necessary before the HC target
could be attained.  The modifications required to reduce HC levels
are unlikely to affect engine performance adversely.

Thus, when considering the primary emissions targets,  the only low
emissions device likely to affect engine performance is the timing
retard necessary for low NOX levels.  This will  adversely affect both
the specific  power output and fuel consumption of the engine. In
order to maintain the necessarily low smoke levels for passenger car
operation, the specific power output of the engine has been taken as
20 kW/1    (0.44 bhp/CID) - normally 22.0 - 22.5 kW/1    (0.48 -
0.49 bhp/CID) for this type of engine in optimum performance build.
A fuel consumption penalty of approximately 5% compared with
optimum performance has been assumed over the load and speed
range.

-------
                          5-9
 In the inters! of fuel economy, engine speed is normally limited
 by friction and breathing considerations to a minimum piston
 speed of 13.0 m/s (2600 ft/min).  Thus with a rated speed of
 66.7 rev/s  (in order to accept the same drive line as the
 gasoline engine) and a known swept volume, engine bore and
 stroke are automatically defined as follows :-

           Bore                       0   88 mm (3.46")
           Stroke                         98 mm (3.86")

 Figures 5-5, 5-6 illustrate predicted test bed performance levels.

 As stated earlier, fuel economy and power  outputs are all based
 on current practice,  allowing for  some derate and fuel economy
 penalty due to the retarded injection timing.

 From the test bed fuel consumption levels a fuel economy level of 11.5-
 10.5  1/100 km (20-22 mpg) has been predicted for the 1600 kg
 (3500 Ib) inertia weight prototype while driving CVS^CH tests;
 current lower powered European diesel passenger cars of the same
 inertia weight average 10-9 1/100 km  (24-26 mpg).

 With current technology  engines, empirical formulae have been devised
 which enable overall noise levels  of engines to be predicted at the
 design stage with  a remarkable degree of accuracy (normally - 1 dBA).
 The predicted levels of this engine on a test bed is 97 dBA which
 should result in Californian drive-by noise  levels of 76 dBA when it
 is installed in a typical American vehicle.

 DESIGN NOTES
V-8 Specification                        Figures :- 5-7,  5-8 and 5-9

           Bore                         88 mm  - 3.46"
           Stroke                       98 mm  - 3.86"
           Displacement                 4.781  - 292 in
           CR                           20:1
           bhp                          96 kW (128 bhp) at 66.7 rev/s
           bmep                        7.5 Bar (109 Ib/in  ) at 33.4 rev/s
           power/unit piston area        0.00196 kW/mm
           Piston speed                  13 m/s - 2560 ft/min

-------
                          5-10

Crankcase Controlling Features

Crankshaft

Crank shaft pins and journals based on maximum firing pressure of
86 Bar (1250  lb/in2).
                                                       2
            Crankpins                420 Bar  (6100 lb/in )
            Journals                 283 Bar  (4100 lb/in2)  aS    y

Connecting Rod and Piston

            Small end eye             690 Bar  (10,000 lb/in2)
            Piston Boss               530 Bar  ( 7,600 lb/in2)  aS °n y

Straight cut rod -  170 mm  (6.7") CRS( —  = 3.48 )

Low piston height of 55 mm (2.16")» 63% bore, was chosen as
no piston cooling oil was required.

Cylinder

Press fit dry liners for engine compactness and low cost.

Cylinder centres at 1.31 x  cylinder bore controlled by bearing layout.

Camshaft

Central camshaft, operating valves through push rods.

Single camshaft chosen for minimum engine width.

Water circuit

Water pump on front face of block, discharging  coolant through both
banks of crankcase and cylinder  heads through transfer pipes.

Cylinder Head

Figure 5-8 shows two possible port arrangements, cross flow
and uni-sided.

Due to the complexity of inboard push rod,Comet chamber and a
regular bolting pattern, a uni-sided head with up-swept inlet port
was chosen.

This gives more freedom for the cylinder head casting and reduces
the overall length of the engine.

-------
                          5-11
 Fuel Pump
 Centrally mounted pump, gear driven from camshaft.  Inboard
 injectors give short  fuel pipe lengths.

 Auxiliary Drives

 Internal gear train chosen for long life and versatility.  'V belt
 drives to alternator, vacuum pump and water pump. Gear driven
 hydraulic pump.

 Overall Engine Package

            Length                    758 mm - 29.84"
            Height                    604 mm - 23.75"
            Width                     692 mm - 27.25"
            Box Volume               0.32 m3- 11.18 ft3
            Estimated weight          320 kg - 700 Ib
Boosted - Indirect Injection

Whilst the naturally aspirated engine should make an attractive
power plant, the use of boosting will give an improvement in
specific power output and hence a reduction in bulk and weight.

Boosting with both turbocharging and with the use of Comprex have
been considered.  Turbocharging does of course introduce other
difficulties.  A waste gate will be essential with the turbocharger
matched to the engine at peak torque speeds.  A boost responsive
maximum fuel stop, or the equivalent, is necessary to prevent
puffs of black smoke when accelerating  and the time lag on
acceleration due to turbocharger inertia may require a change in
driving patterns.

The use of the Comprex should go a long way to eliminate this lag
and to give a lift to the torque curve at low engine speeds.  It is
also quite easy to introduce modulated exhaust gas  recirculation
without the need of additional valves and controls.  On the other hand,
the device is somewhat bulky,  is unproven in service, and at this
time it is difficult to estimate its first cost. There can also be
difficulties with smoke under starting conditions.

The indirect injection engine has a more severe thermal loading than
does the direct injection engine, and this is accentuated by boosting.
Oil cooled pistons will be essential for the boosted  engine.

-------
                          5-12

The required power output is conveniently given by a six cylinder engine
of 90 mm (3.54 inch) bore and 100 mm (3.94 inch) stroke, and
comparative arrangement drawings on both in-line and Vee versions
are given in Figure 5-12.  Once again, the Vee engine shows a sub-
stantial gain in length but the difference in bulk and weight is less
significant.

It is predicted that a lightly boosted six cylinder IDI will have very
similar emissions characteristics to the naturally aspirated engine
although NOX levels at the same combustion timings are likely to be
marginally higher because of the increase in cylinder  temperatures
and pressures encountered in the boosted engine.  These increased
NOX levels can be compensated for by means of an aftercooler to
reduce intake charge temperatures, although this approach may be
impractical in passenger car applications because of first cost,
weight and bulk penalties,  or by further retarding the  injection'
timing.  The extra amount of retard required to arrive at a compar-
able NOx level to the naturally aspirated V-8 is difficult to quantify
but Ricardo have assumed 2  which would result in a fuel consump-
tion penalty of 2 - 3%.  Therefore the total fuel consumption penalty
of a  1.5 g/mile NOX boosted IDI engine is likely to be up to 8%
compared with the same engine in optimum performance build:
a penalty of 6 -  8% has been assumed when constructing the load
range consumption curves.

In the interest of improved vehicle economy, the maximum rotational
speed of this engine has been limited to 60 rev/s.

Overall Configuration :-

            Bore                     0 90 mm (3.54")
            Stroke                   100  mm (3.94")
            Rated Speed               60 rev/s

Figures 5-10 and 5-11 show detailed predicted performance levels
of this engine. It is assumed that identical performances could be
obtained from both in-line  and Vee configurations.

The boosted engines will  suffer a small but definite fuel economy
penalty at light load conditions (only just becoming apparent at the
extreme lower end of the load range curves) because  of the power
required to  drive the boosting device, and this in conjunction with
a need for a greater timing retard will result in a  small vehicle
fuel economy penalty compared with the naturally aspirated engine.

-------
                          5-13

A level of 12-11  1/100 km (19-21 mpg) during the Federal test procedure
has been predicted.  Bare engine predicted noise levels are 96 dBA at 1 m
which should result in Californian drive-by noise levels of 75 dBA when
installed in a typical American passenger car.
DESIGN NOTES

In-Line 6 Specification

           Bore
           Stroke
           Total Cylinder Displacement
           CR
           bhp
           bmep
           power/unit piston area
           Piston speed

Crankcase Controlling Features

Crankshaft
Figures 5-12, 5-13

90 mm - 3.54"
100 mm-3.94"
3.84  1 - 234 CID
 19: !
96 kW (128 bhp) at 60 rev/s
9.3 Bar (1351b/in2 at 33.4 rev/s
0.00247 kW/mm2 (2.15 hp/in2)
12.0 m/sec - 2350 ft/min
Crank shaft pin and journal loadings based on maximum firing pressure,
Pmax of H3 Bar  (1650 lb/in2)
          Crankpins
          Journals
450 Bar (6550 lb/in  )  _
345 Bar (5000 lb/in2)  GaS °nly
Connecting Rod and Piston

           Small end eye
           Piston bosses
640 Bar (9300 lb/in2)  „
565 Bar (8250 lb/in2)  GaS °nly
Conventional straight cut connecting rod with centres of 175 mm (6.9")
                                      3.5 )
Piston compression height controlled by the requirement for oil cooling
with fixed crankcase located spray nozzles.

-------
                          5-14

Cylinders
Dry sleeve cylinder liners adopted for overall engine compactness
and low cost.

Cylinder  centres of 1.23 x cylinder bore controlled by coolant
jacket requirement between adjacent barrels.

Tappets removed through side of crankcase.

Oil Galleries
Individual lubricating oil and piston cooling oil galleries. (Piston
cooling oil unfiltered, hence smaller filter requirements).

Coolant Pump Position

Coolant pump located on the front face of the  block discharging
directly into the block jacket.

Cylinder  Head Features
Uni-sided port lay out adopted for reasons of casting, simplicity
and manifold compactness (turbocharger and manifold on same
side of engine).

Pushrods, combustion chamber and  injector positioned together
on opposite sides  of ports.

Fuel Pump

High mounted distributor type fuel pump,  gear driven off the main
camshaft timing train.

Mounted on same  side of engine as the combustion chamber, hence
compact fuel.pipe runs.

Auxiliary Drives

Two sets  of twin ' V ' belt drives.  Alternator and water pump together
with individual drive to the vacuum  pump.

-------
                          5-15
Overall Engine Package

          Length
          Height
          Width
          Box Volume
          Estimated Weight
           976 mm (38.41")
           678 mm (26.7" )
           543 mm (21.4" )
           0.36 m3 (12.7 ft3)
           327 kg  (720 Ib)
                  (incl SAE.4
                  bell housing)
V-6 Specification
          Bore
          Stroke
          Total Cylinder Displacement
          CR
          bhp
          bmep
          power/unit piston area
          Piston speed
           Figures 5-12,  5-14, 5-15, 5-16

           90 mm (3.54")
           100mm (3.94")
           3.84  1  (234 CID)
            19:1
           96 kW (128 bhp)  at 60 rev/s
           9.3 Bar (135 lb/in2) at 33.4 rev/s
           0.00247 kW/mm2 (2.15 hp/in2)
           12.0 m/sec   2350 ft/min
Crankcase Controlling Features

Crankshaft and Cylinder Bank Displacement

       Table of Cylinder Bank Displacement and Crankshaft
       	Configurations for a V-6 Engine	

Cylinder Bank Displacement   Firing    Primary and
and Crank Configuration       Intervals  Secondary Balance
120  simple 3 throw crank    Even
120  - 6 throw crank
pins displaced 60
Uneven
Primary balance shaft
required.  Two secondary
balance shafts required
for complete balance.

Primaries in balance.
Two secondary shafts
required for complete
balance.
90   simple 3 throw crank    Uneven
           Primaries in balance.
           Two secondary shafts
           required for complete
           balance.

-------
                          5-16
90-6 throw crank          Even        One primary balance
pins displaced 30                         shaft required.  Two
                                         secondary shafts required
                                         for complete balance.
  o
60-6 throw crank          Even        Primaries in balance.
pins displaced 60                         Two secondary balance
                                         shafts required for
                                         complete balance.
   o
120  bank displacement rejected due to excessive overall engine width.
  o
90   bank angle considered with a simple 3 throw crankshaft as
opposed to the 6 throw arrangement.  Although uneven firing
intervals result,the primary forces are in balance.  Crankcase
length of 425 mm (16.75") can be realised with angled split
connecting rods.
  o                                                             o
60  bank angle considered with a 6 throw crankshaft, pins offset 60 .
This arrangement has both even firing intervals and primary balance
as shown in the above tables.  Crankcase length of 507 mm  (19.96")
can be achieved with angled split connecting rod.
  o
90  bank angle and simple 3 throw crank selected on its merits of
minimum length and inherent primary balance.

Crankshaft pin and journal loadings based on maximum firing pressure
       of 113 Bar (1650 lb/in2).
 max
           Crankpins                   455Bar (6600 lb/in )
           Journals                     345 Bar (5050 lb/in2)     °n y

Connecting Rod and Piston
                                                         2
           Small end eye                640 Bar (9 300 lb/in )
           Piston bosses                565 Bar (8250 lb/in2)    S    y

Angle-split connecting rod with centres of 175 mm (6.9")


                     (  7-    =   3.5)

Piston compression height controlled by requirement for oil cooling
by fixed crankcase located spray nozzles.

-------
                          5-17

Cylinders
Dry sleeve cylinder liners adopted from cost considerations.

Cylinder centres of 1.45 x cylinder bore controlled by the
crankshaft requirements.

Camshaft
Single camshaft located within the centre of the Vee with pushrod
operated valves adopted for reasons of minimum cost and overall
height profile.

Tappets removed through side of crankcase.

Oil Galleries
Individual piston cooling galleries for each bank located with the
crankcase side walls.  Separate lubricating oil gallery, located
adjacent to the camshaft, thus minimising filter requirement.
(Piston cooling oil unfiltered).

Coolant Pump Location
Single water pump mounted on frcnt face of block,  transfers across
to the opposite bank via external pipe at rear.

Cylinder Head Features

Cross-flow porting arrangement adopted for simplicity of turbo-
charger and manifold installation*  Inlets on the inside of the Vee,
exhausts on the outside.

Pushrods, combustion chamber and injector located together on
inside of Vee.  (Due to wide.cylinder centres, reasonably uncongested
head of layout should result).

Fuel Pump
Distributor type fuel pump mounted within the Vee directly above the
camshaft, and gear driven off the main timing gear train.

Fuel pump adjacent to> injectors, hence compact fuel pipe runs.

Auxiliary Drives
Two sets of twin ' V ' belt drives.  Alternator and water pump together,
with individual drive to vacuum pump.

-------
                          5-18
Overall Engine Package
           Length                      731 mm  (28.8")
           Height                      693 mm  (27.3")
           Width                       650 rr.m  (25-6")
           Box Volume                  0. 33 m3 (11. 6 ft3)
           Estimated Weight             309 kg (680 Ib)
In-Line 6 Cylinder with Comprex Blower
                                       Figure 5-12

Identical concept and layout to the T/C version but for the
substitution of the exhaust driven turbocharger for a  twin  'V
belt driven blower.

Overall engine package box volume will be reduced  marginally.
Naturally Aspirated - Direct Injection

Due to its high thermal efficiency and long life, the direct injection
diesel engine is  used almost universally in the heavy duty,
commercial vehicle.  While the operating speed range of such an
engine is adequate for  heavy vehicles, there are however difficulties
in obtaining the wide speed range which is desirable for an engine
for a  light duty vehicle.

The main reason for this difficulty is to be found in the fuel/air
mixing requirements which are necessary for good combustion.
In the engine bore size under consideration, it is necessary to
employ considerable air swirl in the cylinder.  This swirl is
normally generated by the appropriate design of the inlet air port but
it is found that if the fuel and air flows are matched for the high
engine speeds then the air swirl is insufficient at low speeds and the
engine power has to be reduced at low engine speeds to avoid smoke.
Conversely, if the swirl is matched  at low speeds,  air swirl is
excessive at high speeds and smoke  will result at the high engine
speeds.

-------
                            5-19

 It may also be necessary to employ very high injection pressures at
 high engine speeds to avoid excessively long  injection periods with
 the nozzle hole sizes which are required to give good combustion at
 low speeds.

 Another difficulty with high speed direct injection engines is the
 high noise level which arises from the long combustion delay period
 as a result of which much of the fuel is in the cylinder before
 combustion is initiated.

 For  these reasons  direct injection engines, despite their higher
 thermal efficiencies, are not employed in any high  speed application,
 i.e. with speeds much in excess of 50  rev/s.  Such a restriction in
 speed involves a considerable increase in the engine swept volume
 and  hence in the bulk and weight of the engine for a given power output.

 As a matter of interest, several years ago, Ricardo converted a  small
 high speed indirect injection engine    (21 swept volume) to direct
 injection and carried out a limited amount of test bed and vehicle
 running.

 (Emission measurements were only carried out over the 7-mode
 cycle but the direct injection engine had twice the NOX and CO levels
 of the  comparable indirect injection engine and much higher HC levels. )

 The following point emerged from the result :-

 On the test bed, the direct injection engine gave about 10% better
 fuel  economy but with 8% less smoke limited power  output.  The
 improved fuel economy was not borne out by road tests with city
 driving, however, when  no difference in fuel  consumption could be
 measured.  This was believed to be  due to the higher torque of the
 indirect injection engine enabling it to be driven with high gears
 engaged for a longer period of operating time.  A similar finding
 had been reported by one of Ricardo's clients with a pair of somewhat
 larger engines.

The direct injection engine was excessively noisy.

As would be expected, the direct injection engine was more
sensitive to injection  timing and required more advance over the
speed range.  Provided that one is not aiming for very low emission
levels  it is indeed not necessary to provide any speed advance for
an indirect injection engine.

-------
                            5-20
The major problem with direct injection is however set by the
emissions targets.  It is unlikely that an NOX level of 1.5 g/mile
could be achieved on production engines, even with a considerable
amount of exhaust gas recirculation, and the retardation of exhaust
timing which will also be necessary will give excessive hydrocarbon
levels and the fuel consumption will be little if any different from an
indirect injection engine of the same emission levels.

While it might be possible to use an exhaust catalyst to reduce the
hydrocarbon levels, there are likely to be difficulties in developing
a system with adequate catalyst activity especially under light load
running when the hydrocarbon emissions will take the form of blue/
white smoke due to misfire.

Due to these problems and the severe derating which would be necessary
under retard timing  conditions to avoid excessive smoke, no proposals
have been put forward for a naturally aspirated direct injection engine.
Boosted - Direct Injection
The use of boost will of course give a higher specific output and hence
will reduce engine bulk and weight.

It will also enable the injection timing to be retarded without a heavy
reduction in power since the exhaust smoke levels will be much
reduced.

The turbocharged engine will have similar problems of turbocharger
lag and transient smoke as the turbocharged indirect injection engine,
but the latter may be treated again by  means of a boost controlled
maximum fuel stop. The Comprex again offers possibilities.

While it seems unlikely that a boosted  direct injection engine could
meet  the emission  targets, it is clearly desirable to include such
an engine in the overall comparison of types.

The specific power output of the boosted DI has been assumed to be
exactly the same as that of the boosted IDI.  This has allowed the
same engine capacity to be used but in the case of the DI, virtually
'square1  (0 93 mm x 94 mm) cylinder dimensions have been selected.
This choice  was  not affected by combustion chamber considerations
but was selected simply to show the other  extreme to the under-square
IDI configuration.

-------
                          5-21
The resultant weight reduction by adoption of a square configuration
was estimated to be in the order of 18. kg   (40 lb) for in-line variants
and 9 kg   (20 lb) for the vee engines.  A noise penalty of approximately
1 dBA will be incurred with this stroke:bore ratio, when compared with
a DI of the same proportions (0 90 mm x 100 mm)  as the IDI engine.
Overall predicted noise level of the bare engine is 104 dBA with a
resultant predicted drive-by noise level of 82 dBA.

Estimated performance levels of this candidate are shown in Figures
5-17 and 5-18.  Identical performances should be obtained from both
in-line and vee configurations.

The load range consumption curves in Figure 5-18 have been constructed
assuming the injection timing  has been retarded 10  in an attempt to
achieve low NOX levels with a resultant fuel economy penalty of 10%.
From heavy duty data it is estimated  that this timing retard in conjunction
with 10% EGR should result in NOx levels in the order of 2.5  g/mile.
This amount of EGR should not affect fuel economy levels.

HC levels at these retarded timings are likely to be very high and it
is assumed that an efficient oxidising catalyst would be necessary in
order to achieve 0.1 g/mile.  Such a system is unlikely to affect
fuel economy levels.

From  the predicted fuel consumption  curves it is estimated that vehicle
fuel economies of 11-101/100  km (21-23 mpg)  are likely to be returned
during the Federal test cycle.
DESIGN NOTES


In-Line 6 Cylinder Specification         Figures :-  5-19

           Bore                        93 mm (3.66")
           Stroke                       94 mm (3.70")
           Displacement                3.841    (234 CID)
           CR                          17.5:1
           bhp                         96 kW (128 bhp) at 60 rev/s
           bmep                        9.3 Bar (135 lb/in2)  at 33.3  rev/s
           power/unit piston area       0.00233 kW/mm2 (2.03 hp/in2)
           piston speed                 11.3 m/s 2230 ft/min

-------
                           5-22

Crankcase Controlling Features

Crankshaft
Pins and journals based on a maximum firing pressure of 124 Bar
(1800 lb/in2).
                                                          2
           Crankpins                   450 Bar (6500 lb/in )
           Journals                    290 Bar (4200 lb/in2)  3S °n Y

Connecting Rod and Piston

           Small end eye               450 Bar (6500 lb/in2) Gas Qnl
           Piston bosses                580 Bar (8400 " "
Angled split connecting rod with centres of 169mm (6. 65")


                     (-   =  3.6)
Piston height controlled by piston oil cooling requirements.

Cylinder

Press fit dry liners for engine compactness and low cost.

Cylinder centres controlled by liner and water jacket clearances
and chosen at 1.25 x cylinder bore.

Camshaft
Single block mounted camshaft, with pushrod operated valves,
adopted for minimum engine cost and overall height profile.

Oil Galleries
Individual lubricating and piston oil cooling galleries.

Piston cooling oil is unfiltered.

Water Circuit

Water pump discharging directly into front face of cylinder block.

Cylinder Head Features

Uni-sided port arrangement chosen for simplicity in the cylinder head
casting,  and ease of directing manifold pipes to and from  the turbocharger.

-------
                           5-23
 Fuel Pump
 High mounted distributor pump gear driven from camshaft giving short
 fuel pipe lengths.

 Auxiliary Drives
 One set of twin 'V belt drives to alternator and water pump.  One set
 of 'V belts to vacuum pump.

 Hydraulic pump mounted on front gear cover and gear driven.

 Overall Engine Package

           Length                       939 mm - 36.95"
           Height                       667 mm - 26.25"
           Width                       543 mm - 21.375"
           Box Volume                  0.34  m3 - 12 ft3
           Estimated weight             310 kg - 680 Ib.
COMPREX BLOWER SYSTEM

The comprex blower was mounted on the side of the block and twin belt
driven.  The manifolding is simplified and gives a slimmer engine and
a reduction in box volume.

The auxiliary drives were repositioned, and the position for the vacuum
pump was changed to block mounted and driven from the camshaft.

V-6 Specification                      Figures :- 5-19 and 5-20

          Bore                       93  mm (3.66")
          Stroke                      94  mm (3.70")
          Displacement               3.841     (234 CID)
          CR                          17.5:1
          bhp                        96  kW (128 bhp) at 60 rev/s
          bmep                       9.3 Bar (135 ib/in2) at 33.3 rev/s
          Power/unit piston area       0.00233 kW/mm  (2.03  hp/in2)
          Piston speed                11.3 m/s   (2230 ft/min)

-------
                            5-24
Crankshaft Controlling Features
Crankshaft
Pins and journals on a maximum firing pressure of 124 Bar
(1800 lb/in2).
                                                         2
           Crankpins                   460 Bar (6700 Ib/in0) r
                                              ,2* *jas onlv
           Journals                     325 Bar (4700 lb/in )         y
Balancing  requirements as stated for V-6 Comet engine
Connecting Rod and Piston
                                                           2
           Small end eye                700 Bar (10,200 lb/in )
           Piston bosses                580 Bar ( 8,400 lb/in )        y
Angle cut  split connecting rod with centres of 174 mm (6.86")

                    
-------
                          5-25
Fuel Pump

Centrally mounted punap, gear driven from camshaft. Inboard injectors
give small fuel pipe lengths.

Auxiliary Drives
Internal gear drive used for long life and versatility.  ' V' belt drives to
alternator, vacuum pump, and water pump.  Gear driven hydraulic pump.
Overall Engine Package

           Length
           Height
           Width
           Box Volume
           Estimated weight
.708 mm - 27.87"
654 mm - 25.75"
674 mm - 26.50"
0.31 m3T 11 ftS-
300 kg -   660 Ib.
Comprex blowers were not deemed a viable proposition due to
complexity of manifold pipes to and from the blower.

-------
                            5-26
Loop Scavenged, Two Cycle, Indirect Injection

While it is difficult to obtain a reasonable performance and fuel
economy with a two cycle indirect injection engine, the lower NOX
emissions with indirect injection make it desirable to assess the
possibilities.

An engine of this type was manufactured for a short period by the
Turner Manufacturing Company in the 1950's (689) and the published
performance data from this engine has been taken as the basis for
prediction.

The resulting engine has six cylinder and a bore of 99 mm (3.89")i
a stroke of 114 mm (4.50") and runs to  45 rev/s.  This gives a high
value of  4450mm rev/s (10500  in. rpm) for the bore and speed
parameters which control breathing  on a two cycle engine and therefore
gives a penalty in fuel economy.  While it would have been possible
to reduce the parameter by employing eight cylinders of 85 mm (3.3611) ,
the very wide scavenge port belts necessary with loop scavenging give
such wide cylinder spacing that the engine  would have been excessively
large and heavy although of Vee  form.

It  is difficult to predict the emission levels of this power unit.  Due
to the poor fuel economy and the need therefore to burn more fuel
the levels could be higher than for an equivalent four cycle engine.
It  has been assumed however that the NOX primary target can be  met
but there is no guarantee of this.  Hydrocarbon levels will almost
certainly give problems.

Estimated vehicle fuel consumption levels during CVS-CH are 13-11.5
1/100 km (18-20 mpg).

Predicted noise levels are impossible to quantify with any degree of
confidence but it is estimated that combustion noise levels from this
engine under Californian drive-by conditions will be around  75 dBA.

The arrangement drawing of the  engine is shown in Figure 5-22.
The effect of the wide cylinder spacing on engine length is clearly
shown and in Figure 5-21 the performance curves show the rather
poor fuel economy to be expected.

-------
                           5-27

DESIGN NOTES

V-6 Specification                          Figures :-   5-22 and 5-23

           Bore                        99 mm  (3.89")
           Stroke                       114mm (4.50")
           Engine Displacement         5.261    (321 CID)
           CR                           19:1
           bhp                         96 kW (128 bhp)
           bmep                       4.7 Bar (68 Ib/in ) at 33.3 rev/s
          power/unit piston area       .00207 kW/mm2 (1.82 hp/in2)
           Piston speed                 10.2 m/s  (2000 ft/min)

Crankcase Controlling Features

Crankshaft

Pin and journal loadings based on maximum firing pressure of 110 Bar
(1600 Ib/in ).
                                                         2
           Crankpin                    470 Bar (6800 Ib/in  ) Gas only
           Journals                    307 Bar (4450 Ib/in2  )

For crankshaft configuration and pin offset  Table 1 (next page) has been
assembled.

From the results a  90  'V engine with a 30  crankpin offset was chosen.
A 120°  'V engine is too wide for serious consideration, and a 60° 'V
is too tall.

Scheme Fig.37 shows the crankshaft layout with flying webs between
adjacent pins.
Connecting Rod and Pistons

           Small end eye                745 Bar (10,800 Ib/in2)
           Piston bosses                515 Bar (  7,500 Ib/in2)

Angle split connecting rod with centres of 228 mm (8.96").

                       <±-  -   4)

This high  —  ratio is caused by piston skirt and the porting of inlet
and exhaust gases. Another effect on this ratio is the angle split
connecting rod.

-------
                          5-28
                              Firing
Residual Primary
   Couple
120
90
75
* 90
75
90
90
60
0
0
30
15
20
15
60-60-60
90-30-90
75-45-75
60-60-60
60-60-60
70-50-70
75-45-75
0
0
0.53 vert.
0.53 vert.
0.77 vert.
0.35 vert.
0.27 vert.
*  This configuration gives even firing and primary couple of low
magnitude that should be removable through correct positioning
of flexible mountings.

-------
                          5-29

Cylinder

Thick walled, press fit dry liners for compactness.

Cylinder centres fixed at 1.7 x bore due to porting through sides of
adjacent barrels.  (In-line version of this engine was rejected for
the reason of excessive engine length).

Oil galleries
Individual lubricating oil and piston cooling oil galleries.
Piston cooling oil is unfiltered.

Water Circuit
Water pump on front face of block, discharging coolant through
both banks of crankcase and to cylinder heads through transfer
pipes.

Cylinder Head Features

Due to large cylinder centres no inlet or exhaust ports;  simple
individual head was adopted.

Pre-combustion chamber is therefore centrally mounted.

Manifolding

The Rootes type blower is centrally mounted inside the ' V' and
gear driven from the crankshaft through idler gears.

An inlet manifold casting  is attached to both banks of the ' V'
and connected directly to the blower.

Exhaust manifolding is a simple conventional cast iron pipe.

Fuel Pump
The distributor fuel pump is driven through the Rootes type blower,
and the high position allows small fuel pipe lengths,

Auxiliary Drives

Two sets of ' V' belt drives.  Alternator and water pump together with
an individual drive to the  vacuum pump.  Hydraulic pump is gear driven.

-------
                          5-30

  Overall Engine Package

           Length                      803 mm (31. 60")
           Height                      692 mm (27.25")
           Width                       648 mm (25.5")
           Box Volume                 0.359 m3 (12.7 ft  )
           Estimated weight             340kg  (7601b).
Uniflow Scavenged Two Cycle Direct Injection
The considerable success of the General Motors series of truck
engines of this type suggests that it is a suitable candidate for study.

For good fuel economy, it is desirable to keep the product of
speed and bore size to a modest value and a similar value to
that used on the GM71 series engines was adopted.  A value of
10.2 m/s. (2000 ft/min) was chosen for the piston speed.

While there are difficulties in running fuel injection equipment at
engine speed on two cycle engines, the running speed of 45 rev/s
on this size of engine should not give rise to difficulties but a pump
injector system could be used as an alternative.

A direct injection combustion system was chosen due to the known
difficulties of using indirect injection on a two cycle engine and still
obtaining an acceptable fuel consumption.  Ricardo do not believe
however that it will be possible to meet the primary NOX target
and the hydrocarbon levels are also in doubt.

As with the loop scavenge engine, the low rated speed of this engine
along with modest bmep levels results in this engine being heavy and
because of the higher torque levels necessary to obtain the target
power at this speed, a heavier transmission may be needed.

Predicted noise level of this engine under drive-by conditions at a
distance of 15.24 m (50ft) is likely to be 75 dBA.

A fuel consumption of about 12.5 - 11.5 1/100 km (19-21 mpg) under
CVS-CH test conditions is predicted assuming optimum performance
conditions.   There is insufficient data to enable Ricardo to predict
any performance parameters for this engine in low emissions build.
The comparative arrangement drawings for In-line and V-6 cylinder
versions of an engine of 83 mm (3.28") bore and 114 mm (4.5")  stroke
are given  in Figure 5-25 with comparative performance data in Figure
5-24.

-------
                            5-31

 DESIGN NOTES

 In Line 6 Specification                  Figures :- 5-25 and 5-26

           Bore                        83 mm (3.28")
           Stroke                      114mm (4.50")
           Cylinder displacement       3.74  1   (228 CID)
           CR                          17:1
           bhp                         96 kW (128 bhp)  at 45 rev/s
           bmep                       6.9 Bar (100 lb/in2) at 33.3 rev/s
           power/unit piston area       0.002 9 kW/mm2 (2.52 hp/in2)
           Piston speed                 10.2 m/s  (2000 ft/min)

 Crankcase Controlling Features

 Crankshaft

 Crankshaft pin and journal loadings based on maximum firing pressure
 Pmax of 138 Bar (200° lb/in2).

           Crankpin                    451 Bar (6550 lb/in2)
           Journal                      342Bar (4970 lb/in2)

 Connecting Rod and Piston

           Small end eye                7 60 Bar (11,000 lb/in2)
           Piston bosses                570Bar (  8,300 lb/in2)

Angle split connecting  rod and cap with centres 254 mm (10")


                        (±     = 4.48 )

 Piston compression height controlled by requirement for oil
cooling  from fixed crankcase located spray nozzles.

Cylinders

Thick dry sleeve cylinder liners adopted for minimum cost.

Cylinder centre of 1.3 x cylinder bore controlled by water jacket
and air  belt between adjacent cylinders.

Camshaft

Single crankcase located shaft with pushrod operated exhaust valves
adopted for minimum engine cost and overall height profile.

Tappets removed through top face of block i.e.  head must firstly
be removed from the block.

-------
                          5-32

 Oil Galleries

 Individual lubricating oil and piston cooling oil galleries (piston oil
 unfiltered, hence smaller filter requirement).

 Coolant Pump Location

 Located on front face of block,  discharging directly into the crankcase
 water jacket.

 Blower and Fuel Pump

 Rootes type blower mounted on  the front gear casing and gear driven
 from the camshaft.  Blower air is directed through the crankcase into
 the middle ducting around the cylinder.
 The fuel pump is driven through the blower.

 The exhaust manifold is a simple  cast iron 'Rake1 casting.

 Cylinder Head Features

 The cylinder head is simplified by only containing the exhaust porting
 and injector bosses.  Due to cylinder centres  of 1.3 x bore a one piece
 head was chosen.

 Auxiliary Drives

 Two sets on twin 'V  belts. Alternator and water pump together, with
 individual drive to the vacuum pump.  Hydraulic pump is directly
 mounted on gear casing and gear  driven.

 Overall Engine Package

           Length                       946 mm (37.25")
           Height                       874 mm (34.38")
           Width                        575 mm (22.63")
           Box Volume                  0.47  m3 (16.77 ft  )
           Estimated Weight             365kg   (800 Ib)

V-6 Specification                       Figures :- 5-25 and 5-27

           Bore                         83 mm  (3.28")
           Stroke                       114mm (4.50")
           Engine Displacement         3.74 1   (228 CID)
           CR                          17:1
          bhp                         96 kW (128 bhp) at 45 rev/s
          bmep                        6.9 Bar (100 lb/in2) at 33.3 rev/s
           power/unit piston area       ,0029 kW/mrp2 (2.52 hp/in2)
           Piston speed                 10.2  m/sec.   (2000 ft/min.)

-------
                         5-33

 Crankshaft Controlling Features

 Crankshaft
 Pin and journal loadings based on a maximum firing pressure of
 138 Bar (2000 lb/in2).

           Crankpin                     448 Bar  (6500 lb/in2)
           Journals                     290 Bar  (4200 lb/in2)

 For crankshaft configuration see table prepared for the loop
 scavenged 2-stroke V-6 engine.

 For this engine application with the possibility of  a shorter engine
 casing with cylinder centres of 1.6 x bore, a 90°  'V bank angle
 with uneven firing but with the advantage of no vertical primary
 couples .WHS selected.

 (The cylinder centre ratio is lowered when compared to the loop
 scavenged engine, by the removal of exhaust gas  ducting from the
 crankcase to cylinder head).

 Connecting Rod and Piston
                                                            o
           Small end eye               760 Bar  (11,000 lb/in )
           Piston bosses               570 Bar  ( 8300  lb/in2 )     °n y

 Angle split rod, with centres of 254 mm  (10")    ( — = 4.48).
 This high  •—-  ratio is caused by porting of the inlet gases, piston
 skirt, angled split rod and small diameter bore.

 Piston height controlled by oil cooling requirements.

 Cylinders

 Thick walled, press fit dry liners for compactness.

 Cylinder centres fixed at 1.6 x bore, controlled by bearing require-
 ments and porting of  inlet gases,  through cylinder block  around barrels.

Camshaft
Single shaft centrally crankcase mounted with push rod operated
exhaust valves.

Oil Galleries
Individual lubricating oil and piston cooling oil galleries.
Piston cooling is unfiltered.

-------
                               5-34


Water circuit
Water pump mounted on front face  of block,  discharging coolant
directly through both banks of crankcase and to cylinder heads
through transfer pipes.

Cylinder Head Features
Due to large cylinder centres and with only the exhaust porting in
the casting, a simple individual head was chosen.

Auxiliary Drives
Two sets of 'V belt drives.  Alternator and water pump together
with an individual drive to the vacuum pump.  Hydraulic pump is
gear driven.

Internal gear drive was chosen for long life and versatility.

Overall Engine Package

           Length                       704 mm (27.7")
           Height                       685 mm (27.0")
           Width                        768 mm (30.211)
           Box Volume                 0. 387 m3 ( 13. 66 ft )
           Estimated Weight             354   kg (780 Ib)
UNCONVENTIONAL ENGINES

Compound. Direct Injection Engine

Compounding of diesel engines for the attainment of improved power/
weight ratios and thermal efficiency implies the use of air compressors
and exhaust driven turbines with different drive arrangements between
the three components. In the ultimate form the diesel engine is used
solely to produce high temperature, high pressure gas for use in the
gas turbine which forms the power unit - this  arrangement is generally
termed a gas generator.

The possible operational variables which can affect the overall perform-
ance and efficiency in varying degrees are very numerous and the
prediction of performance to include the effect of these variables is
very complex.

-------
                              5-35

The main variables are listed below :-

           Engine type - 4 stroke
                        2 stroke (valve in head, loop
                        scavenged-opposed piston)
           Engine compression ratio
           Boost pressure
           Cylinder pressure limits
           Air fuel ratio
           Valve tinning
           Turbine efficiency
           Compressor efficiency
           Use of charge cooler  - charge cooler efficiency
           Engine /turbo components drive arrangement
           Thermal loading limitations
           Bore/stroke ratio
Although there is a fairly extensive literature study of the subject,
the little practical experience that exists is concerned with large
engines for which this configuration is most suited.  The performance
predictions are also concerned with large engines and are mostly
confined to maximum output conditions without consideration for
part load operation.

In order to arrive at fairly broad estimates for the performance of
a small engine, the available information has been used and modified
where possible to take account of the scaling down of the components.
Engine Configuration

The simplest form of compound engine is that in which the output is
taken from the diesel engine with the turbo-compressor unit geared
to the crankshaft. The gearing has to cope only with the difference
in power between the compressor and the turbine, and there is no
requirement for power equilibrium between these components
such as exist in normal turbocharging.

-------
                               5-36
With a gas generator system all the power output is taken from the
turbine through reduction gearing.  For the relatively low power
requirements of the engine under study the small turbine would
have a very high rotational speed.  Although this arrangement has
the theoretical advantage of a high-low speed torque compared with
the compound engine, it is not considered to be  feasible for the
present application in view of these high x otational speeds and
the unknown performance of small turbines at high expansion
ratios.
Engine Type
From the available data a 2-stroke engine is seen to offer the best
thermodynamic performance and power/weight ratio with an opposed
piston type being superior to the valve in head form.   It also permits
greater control over the division of power between  the diesel engine and
the turbine compared with a 4-stroke engine.  However, although the
2-stroke engine is basically simpler it is considered that the 4-stroke
might be preferred since it has much lower thermal stresses and
piston heat flow and has  greater volume  of experience behind its
development.
Estimated Performance
The values tabulated below have been derived from data predicted
for large engines with the following assumptions :-

           Boost ratio 4:1- Inlet pressure  4.12 Bar ( 60 lb/in2 abs)
           Charge cooler effectiveness 80%
           Charge cooling water temperature 85 C
           Compressor efficiency 74%
           Turbine efficiency 82%
           Cylinder pressure limit 207 Bar (3000 lb/in )

The final column shows the engine swept volume for 96 kW (128 bhp)
at an engine speed of 50 rev/s.

-------
                             5-37
                     bmep     2    Thermal        Displacement
 Charge cooled      Bar / Ib/in     Efficiency *       (litres)
2-stroke gas
generator           19 3    280       .29            1.0

2-stroke
compound           22.4    325       .35            0.85

4-stroke gas
generator           24,2   350       .34            1.57

4-stroke
compound           27.6    400       .39            1.38

Uncooled

2-stroke gas
generator           17 -2    250       .32            1.1

2-stroke           .  ,       .  .-.
compound           19.0    275       .3             1.0

4-stroke gas
generator           18 3    265       .36            2.08

4-stroke
compound           20.7   300       .41            1.63

    * Based on fuel energy.

It is pointed out that these figures must be regarded as approximate
and are intended to show the order of magnitude only and the relative
position of the different operating  conditions.

Specific output can be increased.significantly with increasing manifold
pressure but will be accompanied  by higher thermal and mechanical
stresses.  It is for this reason that a pressure ratio of 4:1 has been
chosen in the above table.

From this table it  is seen that an engine capacity of 1.0 - 2.0 litres
would be required to produce 96 kW (128 bhp) at 50 rev/s depending
on the operating conditions.  There are a number of factors to be taken
into account however in the  application of this data to a small engine.

-------
                         5-38

Firstly, the smallest cylinder size dealt with in the preparation of these
results was 2.8 litres  (170 in^), involving a scaling down of about 1/5.
This will obviously have some direct adverse effect on the engine
performance but the most serious problem would be associated with
the compressor and  turbine.   The dimensions of these would be very
small to match the low gas flows and would have high rotational speeds.
It is difficult to estimate the efficiencies of these components at high
pressure ratios but they would be lower than those used in the predicted
performance and would therefore lead to some increase in engine size.
This problem could be  alleviated by the  use of two stage operation but
this would add to the complexity and bulk of the power unit.  For the
sake of simplicity it would be preferable to dispense with charge cooling
in automotive applications - this also results in an increase in engine
size but has the benefit of increased thermal efficiency.   The  maximum
cylinder  pressure of 20.7 Bar (3000 lb/in2) used in the predicted
performance is very high.   If this is reduced to a more reasonable
level of 13.8 Bar (2000 lb/in ) by a reduction in pressure ratio, the
performance will be penalised by possibly 20%.

Alternatively, it could be lowered by retarding the injection timing which
would depress the engine performance  , but part of this loss would be
recovered in the turbine due to the increase in exhaust temperature.
This method would also be beneficial from the point of view of NOX exhaust
emissions.  It is probable that from the above considerations the engine
size would be increased by 25 - 30% above those quoted in the table.

Choice of Engine Configuration

Assuming that a 4-stroke engine would  be preferred on account of lower
thermal and mechanical stresses and superior thermal efficiency, and
that it would be non-cooled to reduce complexity and bulk, the choice
remains  between a gas generator and a  compound arrangement.  The
engine swept volumes  would be about 2.6  and 2.0 litres respectively,
both of which fall in  the range of large production naturally aspirated
diesel engines.  With a bore/stroke ratio  of 1:1 the corresponding
bore sizes are  94 mm  and 86 mm and piston speeds at 50 rev/s of 9.4
and 8.6 m/s.  Although the compound engine is smaller and requires
a lighter gear connection to the  engine crankshaft, it has poor low speed
performance and would therefore require a more complex transmission.
It is the question of part load behaviour that poses the greatest difficulty
in prediction owing to  the interaction between the three main components,
and the lack of  a suitable computer program.

-------
                          5-39

 Emissions

 The behaviour of the compound engine from the point of view of exhaust
 emissions is also difficult to predict.  Exhaust smoke should not be
 a problem as it should be possible to operate at an air :fuel ratio giving
 low smoke values  e.g.  25:1, although low speed, low load boost
 characteristics will tend to cause some smoke.

 The high combustion temperatures will produce high NOX levels but
 these could be  reduced with charge cooling.  On the other hand, this
 would have an adverse effect on HC as would the low compression
 ratios essential with high intake pressures, particularly at light load.
 These features can only  be established by actual engine measurements.


 Conclusions
The general conclusions are that a 4-stroke diesel engine of about 2£ litres
would be required to produce 96 kW (128 bhp) at 50 rev/s. Although there
is a vast amount of experience with diesel engines of this size with natural
aspiration, extensive development would be required to produce a robust
enough engine configuration and to establish the performance of high speed
small turbo-units.  It is considered that the complexity and durability  of
the high reduction gearing would present a  major disadvantage and expense.

Starting and light load hydrocarbons will undoubtedly be a problem and
ignition assistance may be necessary at idling and light load as well as
when starting.

Figure 5-29 gives the arrangement drawings for  a four cylinder  engine
of 93 mm (3.66 inch) bore and stroke, running at 50 rev/s.  The good
fuel economy but poor torque curve shape are clear from the performance
data in Figure 5-28.

It should perhaps be stated that in a limited study of this kind, it is
not possible to optimise the design for all variants of compound
engine as it would be possible to design  for lower boost ratios.
This would improve the torque curve shape, help starting and idling,
and possibly reduce  NOX levels but at the expense of a  bulkier and
heavier engine with a worsening  of fuel economy.

-------
                            5-40
DESIGN NOTES

In-Line 4 Cylinder Specification

           Bore
           Stroke
           Cylinder Displacement
           CR
           bhp
           power /unit piston area
           Piston speed

Turbine Specification

           Wheel tip diameter
           Wheel tip length
           Exducer diameter
           Speed

Compressor Specification

           Wheel tip diameter
           Pressure ratio
           Speed

Crankshaft Controlling Features

Crankshaft
Figures :- 5-29 and 5-30

93 mm (3.66")
93 mm (3.66")
2 = 54 litre (154 CID)
13:1
96 kW (128 bhp) at 50 rev/s
0.00346 kW/mm2 (3.05 hp/in2)
9.3 m/sec (1820 ft/min)
63 mm (2.48")
6.5 mm (0.256")
48 mm (1.89")
2220 rev/s at engine rated spee<
68 mm (2.68")
4:1
2220 rev/s at engine rated speec
Crankshaft pin and journal loadings based on maximum firing pressure
pmax °f 138 bar (2000 Ib/in2)
           Crankpins
           Journals

Connecting Rod and Piston

           Small end eye
           Piston bosses
460 Bar (6550 lb/in2)  _
342 Bar (4960 lb/in2)  GaS °nly
700 Bar (10,000 lb/in )
584 Bar ( 8,300 lb/in2)
                                                                         •
Angle split connecting rod and cap with centres 175 mm (6.9")

                                  =     3.75 )
                            r
Piston compression height controlled by requirements for oil cooling
from fixed crankcase located spray nozzle.

-------
                          5-41
Cylinders
Dry sleeve cylinder liners for overall engine compactness and
minimum cost.

Cylinder centres of 1.23 x cylinder bore controlled by coolant
jacket requirement between adjacent cylinders.

Camshaft
Single crankcase located shaft with pushrod operated valves adopted
for minimum engine cost and overall height profile.

Tappets removed through side of crankcase.

Oil Galleries
Individual lubricating oil and piston cooling oil galleries  (piston
cooling oil unfiltered, hence smaller filter requirement).

Coolant Pump: Location                                  .
Located on front face of the block discharging directly into the block
water jacket.

Fuel Pump
High mounted distributor type fuel  pump gear driven off.the main timing
gear train  located at the front of the engine.

Cylinder Head Features
Uni-sided port layout adopted for casting simplicity and manifold
compactness,  turbine on same side.

Pushrod and injector positioned together on opposite side to the ports,
and same side as fuel pump^ hence compact fuel pipe runs to injectors.

TVurbine Drive
Gear train located at flywheel end of engine.

Crankshaft located driving gear T driven through the flywheel via
'cush drive'clutch mechanism.

-------
                           5 - 42
Gear

Crankshaft driving
Compound Set - Engine side
               Turbine side
1st Idler
2nd Idler
Turbine driven
Speed (rev/s)

      50
     300
     300
    1200
    1200
    2220
Pitch Line Velocity
47 m/s
47 m/s
189 m/s
189 m/s
189 m/s
(154 ft/s)
(154 ft/s)
(620 ft/s)
(620 ft/s)
(620 ft/s)
189 m/s (620 ft/s)
Note : -     Two idlers necessary to mount the turbine outside the
            flywheel profile.

Turbine Location

Separate gear train casing mounted between engine block and flywheel
housing.  Turbine housing bolted onto rear face of casing and compressor
housing to the front, thus resulting in a compact exhaust and inlet
manifold arrangement.

Auxiliary Drives

Two sets of twin  'V belt drives.  Alternator and water pump together
with individual drive to vacuum pump.
Overall Engine Package
                Length
                Height
                Width
                Box Volume
       802 mm (31.58")

    685 mm (26.96")
    584mm (23")
    0.32  m  (11.3 ft3)
      (incl. SAE 4
      flywheel housing)
                Estimated weight    305 kg (670 Ib)
Rotary (Wankel) Engine
With the considerable interest in the rotary gasoline engine over the
past fifteen years or so, a somewhat smaller interest has been
displayed in the rotary diesel engine.  One company $ Rolls Royce,
have gone quite a long way but have now abandoned their efforts
as they do not believe that a commercially attractive engine can be
produced.  Due to the difficult combustion chamber shape at diesel
compression ratios, the only engines that have run reasonably
successfully have used two stage compression.

-------
                            5-43
      I

 The engine tested by Rolls Royce had a larger 'diameter' first stage
 than second but the design schemed for this report has equal
 'diameters' with longer first stage rotors to give a lower package
 height.

 There must, in view of the experience of all who have worked in
 this field, be grave doubts on the ultimate success of developing
 a rotary diesel engine.  Fuel consumption, hydrocarbon and carbon
 monoxide emissions, smoke and durability must all have a question mark
 over them unless there be a breakthrough  in further developments.
Geometric Limitations - Operating Cycle

The 2.3 * or 4 stroke geometry is undoubtedly the most attractive
form for diesel operation.  The only alternative would be the 1.2
or 2 stroke form, which, whilst giving scope for much higher
compression ratios, has insufficient space within the rotor to
accommodate an  adequate bearing.

Compression Ratio
In order to achieve acceptable cold starting and light load operation
the diesel engine must operate with a high compression ratio;  in the
range 18-22:1 depending upon the combustion system used in the type
of application under consideration.

The 2.3* Wankel mechanism with no combustion chamber in  either
trochoid or rotor can reach a ratio of about 25:1 with a K ratio
.generating radius   of around 14-15. However, as the K ratio
v eccentric radius >
increases and eccentric radius is reduced, the diameters of the
phasing gears are reduced in order to maintain a 1.5:1 ratio and this
limits the diameters of the  main and rotor bearings. Consequently a
K ratio greater than 10 would not be acceptable for a diesel  engine
from the point of view of bearing loads. This limits  the theoretical
maximum compression ratio to 21:1 without any combustion chamber.
(* 2.3 refers to number of trochoid lobes and number of rotor faces),

-------
                            5-44

Combustion Volume
Whether direct or indirect combustion systems are used, piston engine
experience has shown that the volume outside of the actual chamber
must be kept to a minimum.  In the case of small piston engines,  about 25%
of the clearance volume is in unavoidable clearances.  To maintain this
situation in the rotary engine would mean accepting a compression ratio
of about 5:1.  Clearly a compromise must be accepted in which a reasonabl<
compression ratio can be achieved with a minimum of dead space in the
combustion system. Since the compression ratio is going to be well below
that normally used in high speed diesel engines it will be necessary to
utilise two-stage compression and expansion.

As a starting point one might select a first stage compression ratio
of 2:1 and second stage of, say, 10.5:1.  This will mean that at
IDC 50% of the total clearance volume will be in the dead volumes.
Again to achieve piston engine conditions a 4:1 first stage and 5.25:1
second stage would be required.

Combustion Chamber Type
The choice here is between direct injection (Dl) or indirect injection  (IDI).
It  is known that in the DI chambers, air swirl is essential to give good
combustion when wall impingement occurs.   Attempts to use DI chambers in
rotary diesel engines have failed because it is quite impossible to generate
adequate air  motion in the Wankel rotary engine (unpublished Ricardo
experience).

It  is therefore concluded that the combustion system must be of IDI type.
Although the  rotary engine may have special requirements  as regards
chamber geometry,  at this stage, we can only predict combustion
performance from well tried systems such as the Ricardo Comet  on
piston engines.  IDI systems such as the Comet depend for their mixing
gas motion on having the .bulk of the clearance volume divided into two
by a throat or orifice through which the air is compressed  and combustion
gases are expanded.  The proportions in the prechamber and main volume
over the piston are fairly critical to performance and should ideally be a
50/50 distribution for the Comet.  For prechamber systems such as
Caterpillar,  the prechamber volume can be only 25% of the total.

Clearly there must be some dpearture from this ideal because with the
selected compression ratios there is already 40% of the clearance
volume over  the rotor before any chamber volume is incorporated
in the rotor similar to those normally present in the piston crown of
conventional  engines.  At the very least a channel must be provided
in the rotor face to allow gas transfer from trailing to leading
volumes at TDC.

-------
                         5-45

 Piston engine experience shows some fall off in air utilisation at
 a volume proportion of 40-60, i.e. 40% in the prechamber, at 20%
 in the prechamber the loss of air utilisation is about 33%  and the
 loss of brake performance about 40%.  Presumably the precombustion
 chamber will suffer loss in this way although no data is available.
This is the area for maximum growth and where maximum development
should be applied to the diesel rotary engine.  We will assume that
some improvement is possible and say at this stage that due to loss of air
utilisation in the Wankel engine the indicated performance will be down
by 20% on the piston engine.

We must consider the relative motoring losses of the rotary and piston
engines.  Compared on the basis of Wankel rotational speed = 1.5 x
piston engine rotational speed, there would appear to be an increase of
some .75 bar mep in the case of the Wankel engine.  This  difference
probably contains increases in heat, mechanical and pumping losses
resulting from the 2 stage arrangement and which are difficult to
separate.

Taking a typical small automotive IDI performance and making allowance
for the above two differences, the performance is predicted as in Figure
5-31, and to achieve the required power the engine will have two high
and two low pressure stages. The swept volume of the low pressure
stages will be 5.35 litre (326 in3) per  lobe.   The engine will give two
working strokes per revolution of the eccentric shafts and with an output
shaft reduced in the ratio to 3:2 the working cycle frequency will be
as for a 6 cylinder  in-line engine.   Top eccentric speed will be
100 rev/s and output shaft 66.7  rev/s.

The fuel pump repetition rate of 100 rev/s is too high for one element.
Therefore each injector will be fed by two lines of a modified 4 cylinder
pump siamesed together within the pump.

The layout in Figures 5-32,  5-33 and 5-34 is suggested having a length
of about 752mm(29.6"),width of  6.79mm (26.7") and height about 508 mm
(20")  without auxiliaries.   There are ample take-off points for driving
auxiliaries,  the alternator could be contained within the engine envelope
being on the other end to the fuel pump.

-------
                              5-46
 Low Heat Loss ( ' Adiabatic ' ) Engine
 Preliminary calculations were carried out to assess the likely
 improvement  in fuel economy which would result from large
 reductions in  the heat transferred from the  charge during the
 working cycle.

 Cycle calculations were carried out using a simplified engine
 performance simulation program in which heat is fed into the
 cycle by means of an assumed heat release  diagram shape.  It
 was assumed  that the heat release diagram  shape was unaffected
 by the reduction in the heat loss which was simulated by increasing
 the assumed temperatures of the combustion chamber walls.

 While the results are not of a high order of accuracy, they are
 believed to give a reasonable picture  as to likely trends . Reducing
 the heat transfer during the power stroke by 70% gives a reduction
 in fuel consumption of only 4. 5% with  a resulting increase in maximum
 cycle temperature from 1460°C to 1510° C  (2658°F to 2788°F)  and
 an increase of exhaust release  temperature from 600 C to 705°C
        to 1300°F).
These calculations assume that the volumetric efficiency is unchanged
which is likely to be an optimistic assumption and any reduction in
this will of course increase the fuel consumption due to a lowering
of mechanical efficiency.  The increase in exhaust temperature may
lead to problems with exhaust valves but on the other hand, this
increase could lead to an increase in power recovery in a compound
engine.

NOX emissions are likely to be increased, perhaps dramatically, due
to the strong temperature dependence of NOx formation at these levels
but hydrocarbon levels should be reduced.

Provided that some practical method of reducing heat losses can be
developed, the 'Adiabatic1 principle could be applied to most of the
engines under consideration;  on the other hand,  for the reasons given in
the preceding paragraphs ,  the   overall result may not be too attractive
and no further analytical studies: have been made at this time.

-------
                               5-47
 Low Ratio, Ignition Assisted Diesel
 There has been a considerable interest in recent years in the possibility
 of developing a low compression ratio diesel engine.  The interest has
 been initiated by, amongst others, the following considerations :-

 1.      For reasons of acceptable cold starting and the prevention of
        high speed light load misfire and blue smoke after a cold start,
        the compression ratio of current high speed diesel engines is
        appreciably higher than the compression ratio for optimum
        efficiency.  A reduction in ratio would therefore give an
        improved fuel consumption.

 2.      A reduction in peak cycle pressures could give a reduction in
        engine scantlings and hence of weight  and cost.

 3.      Reduction in peak cycle temperatures  should give reduction in
             emissions.
While there is sone doubt as to the validity of the second and third of
these arguments , it would undoubtedly be of interest to run such an
engine to examine its performance and potential.

The major difficulty of such a test arises from the difficulty of
guaranteeing reliable and consistent ignition and it is useful to
consider the alternatives ;-

1.      Heater plugs as presently used for a starting aid would have
        an unacceptably high power  consumption even if they did
        give consistent ignition.   For cold starting ,  additional aids
        would also be necessary.

2.      Inlet air manifold heaters would  have a similar power loss if
        electrically heated and even if fuel heated, would give an
        unacceptable increase in fuel consumption, especially at
        light load when the need for an ignition aid will be at its
        greatest.

3.      Spark ignition must be the most attractive aid due to its low
        power consumption.  The use of a high energy surface discharge
        plug may be necessary however and a repetitive spark system
        of the kind used in the TCCS system may be essential.  Even
        with this , there is no guarantee that consistent and regular  firing
        will result and high hydrocarbon levels due to misfire may prove
        insuperable.

-------
                            5 - 48
4.      Exhaust gas recycle to heat the cylinder charge under light load
        conditions could possibly be developed as an effective ignition aid
        but Ricardo are already concerned about the effect of high EGR
        flow rates on lubricating oil and engine durability when it is applied
        at high loads.  Increasing the range during which this is used may
        aggravate this potential problem.

5.      The use of exhaust back pressure to eliminate blue smoke
        problems  (by increasing cylinder temperatures) is  known to
        be  effective in conventional diesel engines and  would
        undoubtedly be effective in improving ignition conditions in
        a low ratio engine.  However, as in the case of inlet manifold
        heaters it works by increasing the amount of fuel being burnt
        and like intake charge heating is in direct opposition to a high
        economy prime mover.

6.      Charge air heating by means of the hot exhaust gas could possibly
        be  employed once the system has warmed up but will not provide
        a solution to high hydrocarbon levels due to misfire on start up.

7.      Variable compression ratio pistons would limit cylinder pressures,
        but can only be applied to quiescent direct injection  combustion
        systems, would not give improved economy at  light  load, are
        expensive and it is difficult to arrange piston cooling for high
        ratings.
In view of these difficulties it was thought that a low ratio, ignition assisted
diesel engine was not a proved and practical proposition at this time but
it is suggested that for experimental purposes it would be worth
attempting to find a solution of the  ignition problem in order that  the
future potential might  be investigated.

-------

mm
Bore
in
Stroke
in
kW
Power ,
hp
Swept litre
Volume in^
Weight j^
Box m3
Volume ft^
Specific power- kW/1
swept volume hp/in
r\
Power/unit kW/mm
piston area hp/in^
Specific power- kW/m^
box volume hp/ft^
Specific weight kg/1
- swept volume Ib/in1-^
Specific weight kg/kW
- power Ib/hp
V8
Gaso-
line
97
3.82
76
3.00
96
128
4.5
275
250
550
:
21.3
0.47
-
-
• 55.6
2.00
2.60
4.30
IL6
Gaso-
line
88
3.46
82
3.22
96
128
2.99
183
186
410
:
32.11
0.70
-
-
60.2
2.19
1.87
3.12
V8
NA
IDI
88
3.46
98
3.86
96
128
4.78
292
320
700
0.32
11.2
20.2
0.44
0.0020
1.70
302
11.4
66.9
2.41
3.31
5.47
V6
TC
IDI
90
3.54
100
3.94
96
128
3.84
234
309
680
0.33
11.6
24.4
0.55
0.0025
2.15
291
11.05
80.9
2.91
3.22
5.31
In Line
6 TC
IDI
90
3.54
100
3.94
96
128
3.84
234
327
720
0.36
12.7
24.4
0.55
0.0025
2.15
265
10.1
85.6
3.08
3.41
5.62
V6
TC
DI
93
3.66
94
3.70
96
128
3.84
234
300
660
0.31
11.0
24.4
0.55
0.0023
2.03
306
11.6
78.5
2.83
3.12
5.16
In Line
6 TC
DI
93
3.66
94
3.70
96
128
3.84
234
310
680
0.34
12.0
24.4
0.55
0.0023
2.03
280
10.7
80.7
2.91
3.22
5.31
V6
2 Stroke
Loop IDI
99
3.89
114
4.50
96
128
5.25
320
340
760
0.36
12.7
18.3
0.40
0.0021
1.82
265
10.1
65.7
2.37
3.59
5.93
In Line 6
2 Stroke
Uniflow DI
83
3.28
114
4.50
96
128
3.74
228
365
800
0.47
16.8
26.0
0.57
0.0029
2.52
201
3.0
98.4
3.55
3.78
6.25
4 Cyl.
Compound
DI
93
3.66
93
3.66
96
128
2.52
153
305
670
0.32
11.3
38.0
0.83
0.0035
3.04
299
11.3
120.6
4.34
3.17
5.23
2 Rotor
2 Stage
Rotary
-
-
96
128
:
227
500
0.26
9.2
-
:
368
13.9
:
2.36
3.91
    SUMMARY TABLE OF MAJOR CHARACTERISTICS OF
POWERPLANTSXONSIDERED IN ENGINE CONFIGURATION STUDY

-------
                               5-49
 CVS-CH Fuel Consumption and Emissions Estimation
 For rating purposes it was necessary to estimate the expected emission
 levels and fuel consumptions when the target vehicle, fitted with the various
 engine combinations,  is driven over the LA4 cycle.

 It is clear that the final accuracy cannot be better than that of the input data
 and there are considerable difficulties in giving a definite forecast of the
 results to be expected from some of the candidates.  The emission levels
 to be expected from the compound engine and the rotary  engine are clear
 examples of this difficulty as is the fuel consumption of the rotary engine.

 It was decided therefore to make detailed calculations for the case where
 the highest degree of accuracy for the input data could be expected and to
 estimate the likely variations from  these calculated values for the other
 engines.

 The detailed calculations were carried out using a cycle  synthesis computer
 program which has been developed by Ricardoo  The program  is flexible in
 terms of the driving cycle which it will accept but for the purposes of these
 calculations, the LA4 cycle was of course employed. The cycle is fed in
 in terms of vehicle speed against time and the program calculates the engine
 performance which is  necessary to  drive the cycle.

 Engine performance maps  showing exhaust constituents,  brake specific fuel
 consumptions and volumetric efficiency, all against brake mean-effective
 pressure and engine speed have also been entered in and the program then
 uses this data to calculate the gross fuel consumption and the mass of the
 individual exhaust pollutants in 1 second steps for  the whole cycle.

 As  CVS-CH test data is  only available,  albeit for lower power/weight ratios
 and for somewhat lighter vehicles,  for the naturally aspirated indirect
 injection engines, it was decided to use the computer program for this
 engine.  The predicted engine performance data and emission maps given
 in Figs. 5-35 and 5-36 were used for this purpose  and the following estimates
 resulted :-

       Fuel  consumption            11.3 1/100 km  (20.7 miles/gallon)

             NOX                   1.17g/mile
             HC                    0.46g/mile

A second set of calculations were run using typical ppm levels achieved
from engines of this cylinder size when in optimum performance build.
 (Figures 5-37 and 5-38).

-------
                              5-50

In this case for the 97 kW engine the following emission levels were
predicted :-

             NOX                  1.9 g/mile
             HC                   0.7 g/mile

Vehicle fuel economy levels were not  calculated with the engine in this
build but a CVS-CH fuel consumption  approaching 10.5 1/100 km
(22 mpg) would be expected.

CO levels were not calculated in either case as experience has indicated
that provided the engine is operated at acceptable smoke levels, the
CO levels will be below the target value of 3. 4 g/mile.

The arguments employed in the prediction of the values  to be expected
for the other engines are given in the  Power Plant Rating section of
this report.

For comparison purposes predicted emission maps were also prepared
for the boosted DI and IDI engines.  These are shown in Figures 5-39,
5-40, 5-41 and 5-42.   Due to limitations in the availability of air flow
data for turbocharged engines, however, no computer calculations were
carried out for the turbocharged engines.

-------
                               5-51
         ESTIMATED TORQUE CURVE FOR
                                FIG. No  5-1
                                Drg Nc  D26O84
                                Date   £8-7-74
097x76 mm V8  GASOLINE ENGINE  IN  LOW  EMISSIONS BUILD
                                      BARO 7£0 mm  Hg.
A.I.T. 20°c
BARE ENGINE
              ;>Tbmep-bar
                     ^TT/TfiUlLp  FOR. LOW EM 155
                     - 1 /L' 4-1-14-U.i-l I I I i-L-li-l I I I I i |-rr
                          5g/mile NOx-CLOSE
                     OLERANCE SOPWISTICAJED
                           RBURETTOR MODULATED
                             AJR INJECTION^:
                        OXIDATION CATALYST
                  0-4 a/mi le NOx - AS ABOVE
                     REDUCIN6 CATALYST
                  FUEL CONS.-a/KW-h
                   ENGINE  SPEED  rev/s.

-------
                                  5-52
       ESTIMATED FUEL CONSUMPTION  CURVES FOR A


     097x76mmV8 GASOLINE  ENGINE IN


     LOW EMISSIONS Q-Sg/mile NOx) BUILD.
         BARE ENGINE
                     A.I.T. £C°C.
-•5
abo:-t
                             11
                                  JJi

                                 1 I <

                                 HI

                                               4
                                               4
                              EL CONSUMPTION LEVEL6 IN*
                   :: : 0-4q/mile NOx. BUILD ARE  LIKELY TO BE 4
                   • i • • | ! • • •¥• •  i  -  i  • • i • • . i t"i ' 11 (tt : r11 • i Trnrt li

                   --r UP TO 5% HIGHER THAN SHOWN HERE
FIG. No 5-2.
Drg No D 2 40 85
Dat. Zfl-7-74
mm Hg.




:::
.;:
.i




--
--
--'•
- j-
-i
1




--
-


N

--






' TOBI
rir-trt
if rt t~
HE
4
.
- -




...
_.
—





































\ -
IT
i
*
h*
..

-














































_
--
c
j:
--











































--










.-

-



































-





-







"i


i*
.
:.:



-

















• -











--


-

--
--
4
1










.
"
•;





































-







C
J









" _^



--









































l"

14-
ft"
U
n
j


•-

--

•'•
1-
4

-•

-.


-
'-"
jt
F



•

c


- •










-L"


J

.
_ N


i.
4






I
C
1
a,
c
D
^
1 1

t





•
...







J~


r
0
p
H-
5
1
4

--



-


--


..-

i-
t-
r
i-
['
t
ii

-
-

•














-





-;-


•




Js

1


.•
...

_


::


--



1 •
1
t








f
- "I-
* 1
4--
I
t
1 4
H
4-
I
"|*
i
4-

-
m
(

-1
-
10




L



.
><








_„







"4

















y





5















to






J '






--

(



::





J






I



:-






0





C








-


c


:





c















c






c





i(



-

£
1


                        bmeo - bar

-------
                                    5-53
           ESTIMATED TORQUE  CURVE  FOR A

       0 88 x 8£mm G  CYLINDER "EUROPEAN TYPE"

       GASOLINE ENGINE  IN  LOW  EMISSIONS BUILD
                                                         FIG. No 5-3

                                                         Drg No D26O86

                                                         Date   £8-7-74
         BARE  ENGINE
                          A.I.T.  20°C.
                      BARO 7GOmm
"805=^0^
~7&
 60-
 50'
        i:
               n
          j ioo6T
             t >(*
             fcr
                 - -» n
                 -1-n
                     t  t
                                                          "IU '
                                BUILD FOR LOW  EMISSIONS:! f
                                                            4-
                                          NOx.- PETROL INJECTION
     : j; MODULATED EGR., AIR INJECTION.,,
     1 '  '/^vi OATirMci r A.T&V\/eV i ' t ;• • i 4^—1
   OXIDATION CATALYST!
;;::T::-U::;:1;;;:1—UuOiuI
^  IT;"~ri ~rr:T~r~^"irrrrn': :!'!'!! i
                         0-4gj/miiVN'Ox -AS ABOVE-f RE
                         LAJALYST" SYSTEM"^
                          ,4
                          i.l-
                             li
                                          I T | f
                                          t i-h
                                                  !1
                   SPEC. FUEL CONS. (WOT) -fl/KW -h
                 tr
             t; :20 : i!
             :;T!i '• ':
                 i.t M
                             i •)
                             rev/min.
aooo
         30OO
                                DUCINGt
                                                 -t i H-
                                                  T I •*-* '
                                                   IT-
                                                      ;li:-:
    4000
                                                         ::t
-»t-
 ;1
                                                          t;f
                                                               i-..
                                                              JJ:::
                                                              P
                                             r1 I
                                                                  •4oo :>
                                                                  300;$
                                                                  •aoo
                                                                  te
\-^ H
 j:  io-!|.
 •' T. 1.11 H :
      40
           50
6O
    il
                      ENGINE  SPEED  rev/S.

-------
	.	5-54	

                                                        FIG. No 5-4


   ESTIMATED LOAD  RANGE CONSUMPTION CURVES FOR  Dr9 No D£4O87

   ^ &|X CYUNDER 0 e8x 62mm "EUROPEAN TYPE"     Dat*   2fl""7-74


 |  GASOLINE ENGINE IN  l-Sg/mile NOx BUILD.



   PETROL  INJECTION. EGR, AIR INJECTION + OXIDATION  CATALYST.
                                                                  500
                               sooo rev/min)
   3rev/s  (
                   	h 44-- -4-4 ---J4---4-; -44.	4---
                   Hii. 14 t. *-.-l. l J. jj i 4. u+-, i_|-f- 4-f-H-H'tt
                   5O rev/s  (300O rev/m n)j
                   33'3 rev/s (abob rev/min]
£•7 rev/s  (looo rev
                                                                      r
                                                                      i
                                                                      *

                                                                      01
                                                                      i
                                                                      j
                                                                      U)

                                                                      u.

                                                                      o
                                                                      UJ
                                                                      a.
                   JIN:.!, l}J!i mi'
                   t r5.}mln
-------
                                  5-55
         ESTIMATED PERFORMANCE CURVE  FOR
                    FIG. No.   5-5


                    Drg. No. DE6O74


                    Date   28-7-74
   NATURALLV ASPIRATED  0 88 x 98mm V8  COMET V  ENGINE




  BARE ENGINE.  LOW EMISSIONS BUILD (l-5g/mile NOx IN A 3500 Ib


        PASSENGER CAR)  AIT aO°C   BARO 7GO mm Hg.
BO-= -fcQ-*
                     -STEADY STATE
SMOKE - °/o OPACITVrl
                 BUILD FOR LOW EMISSIONS c
                 INJECTION TIMING RETARDED 6-
                             i.; :n ::;. i!::: i:.:; i
                tHC'&
                -+-H-H ! l.M
                      ECTION EQUIPEMENT OPTIMISED FO
           FULL LOAD FUEL CO
                  ENGINE SPEED - rev/5.

-------
  LOAD  RANGE FUEL CONSUMPTION CURVES FOR
FIG. No.  5-G
Drg. No OSLGfO75
Oate   £8 -7-74
NATURALLY ASPIRATED #88x98 mm V8  COMET  V ENGINE.

  BARE. ENGINE.  LOW  EMISSIONS BUILD (l-Sg/mile NOx IN A
  350O Ib  PASSENGER  CAR) AIT 2O5c.   SARO 7^O mm  Hg.
                      I NOTE: INJECTION TIMING RETARDED
                     Hill!.: J:: :; U:_,.L;I;_LLLI...-; ii LJ-
                       G-8° FROM OPTIMUM PERFORM
                       SETTING
             li.Sprev/a;
              33-3 rev/
                       bmep- bar

-------
HYDRAULIC
 PUMP
                                           6ELL HOUSING
                                STARTER MOTOR
                                                      ftORE:-
                                                      STUOKE:-
                                                      6.H.P:-
                                                      »ME.P :-

                                                      r». MAX :-
          (88mm)
384."      (96mm)
I2B(3> 40OO rpm
»9lb/in»   (
-------
                                                                                                                                                             i
                                                                                                                                                            U1
                                                                                                                                                            00
                                                                             POETS
DEC. N? 30S4/1
             E.RA.  DIESEL  IMPACT'  PBO.JECT
          PHASE a ENGINE CONFIGUgATION  STUDY
           N.A.  VS I.O.I DIESEL ENGINE
COMPABISON OF CeOSSFLOW/UNISIDED CYLINDEB HEADS
                                                                                                                                                                 u>
                                                                                                                                                                  i

-------
                                                                                crur-CBi mtc
                                                                                     y
                                                                                      I£AI
                                                                                      L ._^	I
                                                                                                               t
                                                                                                               Ul
DRG.N? 3O84/2
                                E.PA. DIESEL IMPACT' PROJECT
                             PHASE HI  ENGINE CONFIGURATION STUDY

                         N.A.-V8-ID.I. DIESEL ENGINE CROSS-SECTIONAL

                                      ARRANGEMENT.

-------
     ESTIMATED PERFORMANCE CURVE FOR BOOSTED
                FIG. No.   5-IO
                Drg. No. D SL&O7G*
                Date   28-7-74
SIX CYLINDER 0 90x 100mm  I.D.I. ENGINE IN  LOW  EMISSIONS
   BUILD  Q-Sq/milfc NOx  IN A 2>5OO Ib PASSENGER CAR)
    BARE  ENGINE    AIT  2O°C
   ioo
BARO 7(iOmm Hg.
                ENGINE SPEED- rev/S.

-------
	5-61   /(.	

                                                   FIG. No 5"H


 ESTIMATED LOAD RANGE  FUEL CONSUMPTION CURVES  Dr9 No OefcO77


 FOR A BOOSTED SIX CYLINDER 0 9O x IOO mm  I PI

 ENGINE IN LOW EMISSIONS  BUILD  (l-5q/mile NQx

 IN  A 350O Ib  PASSENGER CAR.)
                             Date
                    £8-7-74
   BARE ENGINE
  rSOO
AIT  ZO'C.
BARO 7&O mm  Hg.
                           : INJECTION  TIMING RETARDED
                                                               cr
                                                               I

                                                               U)
                                                               Z
                                                               J
                                                               UJ
                                                               D
                                                               U.
                                                                •
                                                               o
                                                               111
                                                               Q.
                                                               
-------
        731
                                   GK5 Ak/IP
                                   ALTERNATOR
                           -SAE *bCs^^
                             |&aoibC3Q9nq')
                                   &ELL HOUSlMO
 -VACUUM PUWP
ORG.No. 3O84/15
                       STARTER MOTOR
                                                                             BLOWER ON IN LINE 4, CYLINDER
                                              TUR8OCHARGED IN LIME G> CYL.INDER
                                                                                             &o AMP ALTERNATOR
                              E.PA. DIESEL 'IMPACT'  PROJECT
                           PHASE  HI  ENGINE CONFIGURATION STUDY
                      6 CYLINDER TURBOCHARGED I.D.I. DIESEL ENGINE
               NSTALLATION DRAWING  SHOWING  COMPARISON BETWEEN  IN LINE 6/ve
                                                                                                                            Tl
                                                                                                                            P
                                                                                                                            u>
                                                                                                                            i
                                                                                                                            io

-------
                                                                                            •Dependent on cylinder liner
                                                                                                and water jacket
DEG. N9 3O84/I4
                E.RA.  DIESEL IMPACT' pecuecT
           PHASS a EMGINF.  CO^FIGUgATION  STUPV
   
-------
? 30S4/II
                                      E.=A. ggSEL IMPACT PSQj;CT
                                   PHASE Z EMS'NJc CCSJ'iGUEATlCN 5TLOY
                                     LAYOUT  c^ \/6_ g;O°  5A_NK,  ANGLE!
                                                                           n
                                                      71-9
                                                     -JJSSL
                                                                                                   H
                                                                                                     -

                                                                                                    "P
                                                                                  i?.i
                                                      iUL
1
';!; '
'•I i
 H
T'J
                                                                                                                    i

-------

p
i
1


1

(3.54") E
i 73
i? R91


r
c:

ore Dia.

i
(&*


l-MD










'


                                                                   Crankshaft Arrangment with
                                                                   Straight-C~ut Connecting Rod
                                                                    Block Lenath (18.15"!  461
                                                                       »tOJD-OT CCTXrlTKi TOO
N? 3O84/I2
                ERA.  DIHSEL 'IMPACT'  P2OJECT
             PHASE in ENGINE  COMFIGU^TION STUDY
PBELIMINA2Y  LAYOUT OP V<5  TIJESOCHAEGSD  IMESEL  ENGINE
              WITH  3 THEOW

-------
                                                                                    Dependent on Crankshaft Design
OEG. N9 3064/13
            ERA.  DIESEL 'IMPACT' PBOJECT
        PHASE ffl ENGINE  CONFIGURATION STUDY
       TUBBOCHABGED LDI.  DIESEL  ENGINE
CBOSS  & LONGITUDINAL  ABgANGEVIENT DBAWINGS.

-------
5-67

ESTIMATED PERFORMANCE CURVES FOR
FIG. No .5-17
Drg. No Oa6O78
r- • as -7-74
BOOSTED 0 95x94mm SIX CYLINDER D.I. IN MINIMUM EMISSIONS BUILD.
(=2: E-5q/mile NOx IN A 35OOIb PASSENGER CAR)
ENGINE BARE AIT 2O°C. BARO 7GO mm
130-


J20-
,
iio-
....

100-


90-
807
HI :
70-


60-
50-


40*
_.-• ..
30-
T '

- 1




T :":':"
IUC




-90
	 ^


. on
•80






. ^t\-
'70
. ._.!_,.
•GO
• — — •*-
•50
....
i :.• ;


•

i . ..


: : ^
-»— r — *—






: r- •

J_^_





/ '













1

._ 	 1

^t
-^
1

H "
D
Q.

K- rr|
2'

; j CL ' '
•40






•30^
• • j

• • r
U20



• • • t





UJ , t -
o.i
OL j

!


' •! 1


! -r
— i-— *— .-t
t 1 * •
I i :
, * *




: : : 1


^
j •


----- —






.... _^-. -





uiL .
X .
\
; >

: . . ,
. t . •
__

1 . •

i I '



! ! . -t
h • /
: /
/
/
4-
:FU
;-H«.
' '



ooo
:i. : :

' '• 'Z.(

:;::):.::
' 	
i ...

	 r" - 	 .-.--..


j^~ i
r . 1

1 ' ' I '.. ,'
......






. <

1
. . 1 - ^^^
) - bar .

.....


I


W. s<^: :
!»-. . -.T*"*^
;:::.: . . : : . ; . . .
:• -f::.:







.....

A
/ IN
••••;•••__ / ; o-
! . . / '




:::: :: j
k.. /
v 	 /
:>v ! /
:>KJ/

i : : : / : 1















lit j r T T
/ OXID
/.:}::..


bTEADY
— . , -. + . ^f .
^N»^^^
: : :^>*^

. . . . i . .

i


*







, » . f .
LL LOAD FUEL i
,..,...

i . .
I



...;.. .

D::: -;'3



* .
... i
:
. 1 -

. . . j . . . .

^^*S-<__ '

.... 1 ....
- • • • ! '/


/
/
f: • •
/ i • •

r . . i : : '


.j
f
' '^r
A
^/

|X^X'N^: :
. . : : :^s











• t . •

	 '• •


i


	


.... ! ....






.... j ....
...(•-..



BUILD FOR LOW EMISSIONS
JECTION
. . . .
-10* 10'
ISING Cft


\
STATE S
I
T
....(....
. | .




I 1


: • : : I : . :



! '
j

r
:oNs.-g
_.: • • 1 • i -
• • • • t 	

i rev/min..
£000
•:-:r.5C
TIMING R
/o EGR_
.TALYST

.........

vIOKE - '


















t


/kW-h :
'.'.'.'. ! .-X1
*— —"-( •

t .


... i ...
i

lOO

, ' " "' 	
o;;: : 40 ::j;' so ;:;: . G>(
ETARDEC
	


FOR HC
	 _i_ ...
. . . n . . :
. . . 1 . , .

/0 opAcr


: : : :
, . . , . .
f -



	 —










. . ; . . . . ! ;
	 : • -



. . T~. , .

. . . '


•






• • i .

-~nt
• c
•\
1 ' n





Hg.
' ' : '
'140-



130-


!??:




MO"


i
•::: 100-
..,,!.,..
....
D :T:

.1UJ.
90-

S, ::



1
t~
O
. : . . i. ,0.
fY ;':;0






1



1

t




i


,
\
* ; JC
*v


D



' 4OOO : : : ;


:.: ::: J-: ::
D 7

O. .

Hi*
t ' i

• i *••

o
— 0)

1




•fc -


J-5-B

*


•4-
j .




Tit)


. 1
• 9 :
T71^
• '
r • r4-
4 • •- -t-
; r:.T
	 : t r
-:^:ir
0 .


•


. ,
	 •-

i-
t * '
I


t : . ;
1 | r-i
- . • t

!'! ' '-
•30C
• -; -t r

1 1 r
. j : 1
.' i i :.
! 9 /\t



_L> ».i : ' '
.... f ....
• • S

O ;

u : : .
i ! t ..
a; r;
u ::i
E:fi

j — „
i t-r-
i . 1
-. — i — ~-i—
.. .. ..
1 ' f-
' '"t"
I- f
J-l
•* 01
-<- '. :
Z
'-3
J
"tul
[U -
):J:
:ui
I of)
* . 1 1-
• i
ENGINE SPEED- rev/s.









-------
i ESTIMATED LOAD RANGE
FOR A BOOSTED 093x 94
FIG. No 5.19
FUEL CONSUMPTION CURVES Drg. No DZ6O79
mm SIX CYLINDER D.I. ENGINE
D«e 28 - 7-74
MINIMUM EMISSIONS BUILD. (^ a-5q/mile NOx IN A 35OO Ib
PASSENGER CAR)
'Tr

H-- •
.>7-
f
-0 r
— : j : • - : : -
. . i .
	
. * ......

..._. ;;.;
.•!••••


AIT £0°C
1




• • • i

. , . .



ev/s


. . .





... |
: • :: j








....)....


... !
....


. . . , ...
~
••^•MM
NOTE: INJECTION
B-IO°t.
MEEDED
• • • • I • • •

: : : 50 rev/s

- - --(•-•- —
::::b3-3
^*^--H

.... i ....

. . .
....
rev/
•~«—




'ft:-
«••*•
\ — i. ...
. . . . i . , . ,
6-7 revy/s
. : . . : ; .





! i .50 : : : 60 . 70
'i 1 'i ' ^
a • : ;
'. .'. A
» : ; : ; ;. •
l::;i:-i









: : 80:


M


i • ' '
• • C
0%
(SHC
AFFE
L 	 i
....
i ...


....




i

BARO /GO mm Hg,

, .

... *




' ' ' • 4

(







... , . . . .


Uv^vw*
TIMING F
EGR. Ox. CAi
5ULD NOT SE
ECT 'PERFORI









' i !•

b/ir
90
i



bmep-

I*-
: 10
:•::::::
.......
.... i ....



....





* 1 1 1 II





— — •




O IIO
...

. ' ' ' ' '
bar






: '. : i

''H

:;;;









iW*v
I
IETA
•ALy
RIOL
v1AN(
•. 	 -4
RDEI
ST-
JSLY
-e);
.,..)...,

	 t •
^***^
~ • 	
. i ...



**r*, i




;'fc
B' : :
; )

— , t ; ,
• • 1 •
• • t »
w*^*+*
i; i :
1 M '
• ; 1 t-
• 1 V f
• * t •

• • i !
• ,- 1 j




;!',
! . i l
, 1 . i
, I ' .




.... 1





D. . .



4 1 i .
iili












— -r—






. . 1 ,


, . J .

, , I I
1 . . •!
• I • i
. . 1 .
. . i I
t • • •
; ' • t
: ; LI
m;
[ ; ;u j
• t • i- •
1 : !r
Ut'J.
!|t.|

• : • I
! ' ' t
) : ISO :.
• • • '4! : '
. ; i . . j .




	 	 1 1



. i . t
1 , . .
• i • 1
M ! i
t • * 1-
t'-i
•i H •
i-H
in

. . t .
, . ,.,
40
•it ; •
• ' i •.
... 1
t ! ; s .
Hit
i . • t
• i • *







I



. i . .
i t * *



•

•G> -

. i 1 .

j
. ; . •
: i
Q.
•;JC
I'll
i ' • -t
t', * r
! i t Z
1 : 1 •
;4-
i » » »-

ii .+
o; i

4 ; ! -J-
'. ! . j. . .
:!ij
; • :••*




. . i ..
. , , -


* .



~Ar\f
***UL
I- 4--
|i
• • *"£-
• • r-f-
rrit
• i 1 -t—
. 1 4 f.
-• — •••-*— r-
t f" ft-
lirji
1 4-L-1-.
r t •*•*-
• ( -4-f-
,;ttt
Hit
::|.|.
ST?
• * • i f-
. (.;J-
. » i >.
.. . 4-4-
-3/V
j-T f t-
4; t; •-
:;|-t
':•!!

)
•
£.
1
^
0»
1
ui
§
ui
3.
U.
0
111
0.
(0

-------
            152
                     4-S4-
                                                                 337
                                                                            ,—• CO*» PUOW »Of»T
                                                                           /   CVLIMOCK HCAO
HYOQkUUC
PUMP
           (47S)
                                                                                                                                                  Ul
                                                                                                                                                   I
                                                                                                                                                 in
DRC.Mo
                E.PA.OIESEL IMPACT  PROJECT
           • PHASE HI ENGINE CONFIGURATION  STUDY
96 KN  6 CYLINDER TURBOCHARGED D.I. DIESEL ENGINE
INSTALLATION DRAWING SHOWING COMPARISON BETWEEN IN-LINE 6/V6

-------
                                                                                        in-line.Crankshaft —
                                                                                                                          u.
                                                                                                                    I

                                                                                                                 '	1.25D    I
                                                                                                             Dependent on CylinderLiner
                                                                                                                    I and Watei Passage Core
                                                                                           Crankshaft
                                                                                                   L
                                                                            -4
                                                                                                                     1.46D
                                                                                                          Dependent on Bearing Requirements
DEG. N9  3084/4
          E.P.A.  DIESEL  IMPACT PROJECT
      PHASE m ENGINE CONFIGURATION STUDY

6  CYLINQER TUeBOCHARGED D.l. DIESEL ENGINE

 CROSS-SECTIONAL A&I2ANGEMENT DBAWING
                                                                                                                                                       171
                                                                                                                                                       I
                                                                                                                                               Ui
                                                                                                                                                I

-------
                               5-71
     • ESTIMATED PERFORMANCE CURVE  TOR
    STROKE  LOOP SCAVENGE IPI DIESEL  ENGINE.

          OPTIMUM  PERFORMANCE CONDITIONS.
          NOx > 2Lg/mile   HC > O-4g/mUe..
FIG. No  5-2.1
Drg. No D£6O8O
Date   aS-7-74
fEHI.100
                     rwjtrMiFntt
               ENGINE SPEED- rev/S

-------
                                                                        394
                                                                       0275")
                                                                                                                                                  vn
                                                                                                                                                  i
                                            PLATK CLUTCH MAY M
                                       MCtuiMO Oft &AE*3 MIL
                                       HOU«M« WITH UAMtt WA.CLUTCM
                                                                                                  »TROKC
«SO"   (114 mm)
I2» 9> 2TOO rpm
      ! (4-7»AR)^000 R.P.M.
                                                                                                  CAPACITY:-   321 C.I.D. O-2ta Litre*)
                                                                                                  •OX VOL :-
                                                                                                  UT WI*MT:- •
                                                     E.PA DIESEL'IMPACT  PROJECT
                                                 PHASE IE ENGINE CONFIGURATION STUDY
                                96kWLOOP   SCAVENGE-I.D.I. 2  STROKE  DIESEL  ENGINE
                                                       INSTALLATION  DRAWING
DOG No 30B4/7A
                                      H
                                      5
                                      u»
                                      rO
                                      tO

-------
                                                                                                           1.7D
                                                                                                    Dependent on Exhaust and
                                                                                                    ntake Passage at Cylinders
                                                                                                                      J
DUG. N?  3084/6
               E.PA. DIESEL IMPACT PROJECT
           PHASE m ENGINE  CONFIGUBATION STUDY
V6-ID.I- ZSTBOKE  LOOP SCAVENGE  DIESEL  ENGINE
        CBOSS-SECTIONAL  AEEANGEMENT  DRAWING
                                                                                                                                                u>
                                                                                                                                                to

-------
                            5-74
  ESTIMATED PERFORMANCE CURVE FOR


STROKE THROUGH SCAVENGE D.I. DIESEL ENGINE



     OPTIMUM PERFORMANCE  CONDITIONS.

     NOx > Zg/mile,1 HC >0-4g/mile.
FIG NO. * 5-24


Drg No DZfrOSl


Date   Z8-7-74
                                       t)4OOO
                                      4 < 4 •(• I- m •»•

-------
4 O^hMP .
                                       V6 - 2 STROKE  THROUGH  SCAVENGE 0 I  DIESEL ENGINE
                                                                aat,        28?
                                                            .. (" »3 .^, fna.'
                                                           IN LIME 6>CYUNpm-2»T>OKe-THBOLiaM

                                                              SCAVINOC D.I MKML
                                                                                                                                           OR &«e " i aei
                                                                                                                                           WTI LARGER Ol*
                                                                                                                                           CLUTCH.
 3084/83
                                                           E.PA. DIESEL'IMPACT'  PROJECT
                                                       PHASE HI ENGINE CONFIGURATION STUDY
                                        3» L.trts)
                        VEE
                       13-613 Pt-3
                       (•387m3)
                        7BO Ib
                        (354K3)
           IN LJNE
60X VOL.-   l«-77 ft 3
           (•47mS)
EST. WEIGHT:-  800 Ib
            ( 365 K9)
                                                s

-------
.DEG.N? 3O64/I6
               E.PA. OESEL MPACT  PROJECT
            PHASE ffl  ENGINE CONHGUCATION STUDY

6 CYL. - 2 STBOKE - THROUGH SCAVENGED- D.I. DIESEL  ENGINE

      CEOSS & LONGITUDINAL  AE2AMGEMENT DRAWING
                                                                                                                          u


























































                                                                                                                                  m
                                                                                                                                  i

-------
                                                                                                                                    VI
                                                                                                                                    I
DBG. N9  3O84/2I
            E.P.A. DIESEL 'IMPACT  PBOJECT
         PHASE m ENGINE CONFIGURATION STUDY

V&   2 STBOfcE  THBOUGH-SCAVENGED D.I. DIESEL ENGINE
  CROSS &  LONGITUDINAL  ABBANGEMENT DRAWING

-------
                                5.78
        ESTIMATED  PERFORMANCE CURVE FOR
    FOUR STROKE 093x93 COMPOUND Dl. ENGINE.
FIG No.  5-28
Drg No DE6O82
Date   ^6-7-74
          OPTIMUM  PERFORMANCE CONDITIONS
          NOx > 2g/mite    HC > 0-4g/mile.
mr
I30r
                 t- tttnitti
                 SPEC. FUEL

-------
                                                                                                       602
                                                                                                      (SI-SB )
                                 feOAMP
                                 ALTERNATOR
                                                                                                                                                       Ul
                                                                                                                                                       I
                                                                                                                                                       -a
                                                          3-fefc'OIA  (93mm)
                                                          3-fefe*    (93mm)
                                                          128 (5) 30OO rpm.
                                                          222 Ib/in2 (l53&AR)^!)3OOOrpm
                                                          2OOOIb/in2 (3T5API)
                                                          153 CID   (2'5 Litres)
                                                          ll-3cuft-  (0-32m»)
                                                          feTO Ib    (305 l<9 )
                                                          WHEEL DIA'.- 2-H8*(S>3mm)
                                                          TIP wiOTH:- O-25fe" (fe Smm)
                                                          EXOUCER WA:-  I 89*(46mm)
                                                          SPEED :-
                                                                  RATED &PEED
                                            COMPRESSOR:- WMCELDIA:-  2-«>a*
                                                          PRKSSURE  RATIO.'-
                                                                                CAPACITY :-
                                                                                BOX. VOL :-
                                                                                EST. WEIGHT :-
ORG. No 3084/18
                             96 kN
           EPA. DIESEL 'IMPACT' PROJECT
      PHASE  m ENGINE CONFIGURATION STUDY
COMPOUND IN-LINE 4 CYLINDER D.I  DIESEL  ENGINE
               INSTALLATION DRAWING
                                                                                                                                               CP

-------
                                                                           1.2S x Bnro  _
                                                                    Dependent on Cylinder Liner
                                                                         _and Water Jacket
                                                                                                                             .Nodal Turbine.
                                                                                                                                Gear Train
DEG. NP 3O84/n
                                                                                                      "Cush Drive
                E.P.A. DIESEL IMPACT  PBOJECT
            PHASE m ENGINE CONFIGURATION STUDY

COMPOUND  IN-LINE 4 CYLINDER D.I.  DIESEL  ENGINE
 CEO5S & LONGITUDINAL  AggANIGEMENT  DRAWINGS.
                                                                                                                                                  i
                                                                                                                                                  00
                                                                                                                                                  o

-------
                           5-81
  ESTIMATED  PERFORMANCE CURVE FOR
TWO STAGE ROTARY DIESEL ENGINE.

        OPTIMUM PERFORMANCE CONDITIONS
         NOx  > 2g/m»le   HC > O-4g/mile
FIG. NO  5.31
Drg. Nc  DZ6O83   '
Date   Z8-7-74
         SPEC. FUEL CONS.-Q/kW-h
           F.NfilME SPEED - TCV/S.

-------
                  LOW PRESSURE ROTOR
DRG  No. 3084/10
                E.PA  DIESEL  IMPACT PROJECT
            PHASE  El  ENGINE CONFIGURATION STUDY
SGKKI  2 STAGE/  2  BANK  I.D.I  ROTARY DIESEL ENGINE
                                                             BMP  -
                                                             6 M E =>
                                                             p
                                                             CAPACITY
                                                             BOx VOL  -    9-ibcof
                                                             EST WElGrlT •  5>OO B (
                                                                                                            IZ6 d&4OOO r pm (.OUTPUT SHAST)
                                                                                                            51lb/in2 (5-56 S*!?} .i 400O r p.
                                                                                                            feSibl'"* (4-3ABAR1) .J) '500'?^
                                                                                                            ifcOOii>/inz ( i >o BAR)
                                                                                                                  o. |
                                                                                                                                                   Tl
                                                                                                                                                   O
                                                                                                                                                            I
                                                                                                                                                            00
                                                                                                                                                   I
                                                                                                                                                  U)

-------
                                                                                                                                                                                   Ul
                                                                                                                                                                                   I
                                                                                                                                                                                   00
                                                                                                                                                                                       O
                                                                                                                                                                                        i
                                                                                                                                                                                       u>
                                                                                                                                                                                       o>
PRO. N? 3084/19
      E.PA. DIESEL IMPACT PgQjECT
   PHASE a ENGINE CONFIttUgATION STUDY

 j STAGE/ta^NK 1.01. BOTAgY OieSEL BNGINE

PgEUMINAgY CBOSS SECTIONAL AggANGtMEMT DgG.

-------

                                                      I  ,
                                                      •H
DEG. N9 3O84/2O
          E.PA. OES6L IMPACT PgQJECT
      PHASE Bl ENGINE CONPK3UBATION STUDY

 2 STAGE II BANK IQI. BOTAIZY DIESEL ENGINE
PgELIMINAEY LONGITUDINAL  ABBANGEMENT D&G.

-------
5-85
NATURALLY ASPIRATED 9& kW V8 IDI ENGINE
PREDICTED HC ppm LEVELS IN LOW EMISSIONS
NOTE ;
ITHC LEVEL
.-n_;..n.,"Ui.i
^Rl,JN s 0-
' " ' * '
.....








. i *--i —

....

....


*
. . . .-



r'H
» .+- A I
*-f--t- •
.i , 4 -,
• u:
ti:
* - -
- rt •





....


1 - • •








....

-*—• ! 	




....








: : 1 1
. . ; .
~ " t '
" !
t ' '
. . i
i »'! i
• - -t




• • *
. . 1
• : i

.... j ....
t • ; i










aoc
i .


T 1 !







1 1 1 1
• ' t f
' t- J *
1 • • i
! J-
.1 i


! '•• i
i h

1ft
1 ii

' i ; t
' 'it
i t- t I

"t
1'

i : '
. . . . 1 . . i . ; ,





THE ppm LEVELS SHOWN HERE (
GUARANTEED FROM THE PROTOT
S ACHIEVED DURING CVS SIMULATIC
4Gc

— 	




1 ,v^
;*


' ^
'Jr






. ; ,
'!'
ii'
; j f
t ( ;
1 . i
Mi '
tpr
i : .

'• , , '

... i
> • • f

/mil

9. '.:.

_„_»-: .,-L

,*r^ ....



^ i


£ '. '


' ' j^

jr
^

1






x













... l
I ...
t • •

; t i
* ' ' '
•-H-H-
,.. t .
• t • •
|!H

•-I

....
i- ' . •

* -^j.
ooo . :
... , ......

HC
!::;
1 ' • '
;! !:

(44,
(...
M - -1





•fT;



^••— •



















"PF


. •
: : * *

-*. - t .
• 4 t •
f * • *
t- • • ;




"i^
.,41

: • •

TT-*-N

















m C



i 1 ! -

-t ] * •



. . , ^

ST
^ritr-
...."'.. .
"ULL TO
, . . L/ -
—4

"*>ls.
^^^. -
: ; IL... .x
::-:T ^




.

'

t
t





: : ; i


i • • (

_^._
....
1 1 • •




T , , i
. i . .
; ; ; i
p.i

:: 100
^ .

• -r-- T



rev/min '.•'•
: : : : aoob ;
. , . . i .

.11 ! !VNI


OO
...^

"x

k • • '

V
: : : : 'x

j : •







1 4 . i
Tilt

H ; ' t
1 ' ' •
• it.
i i i .
y






/
: 3OOO
i; so ;;



* t ' '

| • . 4
I'M
M • '
i ; I *

i . . i
: : 2.
•^
* . . i
....
FIG. No. 5-35
Drg. No. D 2 6092
Date 2.8-7-74
BUILD
1OULD NOT BE
YPE VEHICLE.
)N .

:-;:.;:::.
RVE

~*


.......

... j ....

	




----


\ ' ; ; '


.' ' ' :
s^. H


; : i\



,..,


; ! ; :
•M1

" Y
00


:>T
i . j .
* • • i
. , . i
; : ; i
t •+ * •
rh:
no-




::;: 100-

* ! i " ! '' " ' "




Kaob ;;
^~



^
::::
..,.
; • • :
::-j:
M i i
i • » >
' • ! t
"it

' i i '••
• ' t '
i . ! i
. . i i

. . ,



c
l^jv.

t" * ' '

l.; ;;
•Hi

'•"
! : • I.


3 .;
4OOO ::;
....,.,...,..,.

90-


:::•
P°


70-
fco-
-lit'
50-
•J • ;;
1 i -• •
40-
t 1 1 •
~t !•*-*-
i ' • •
30-
2.0-
i • • •
h ! :
JO'
! !•< •
! t l '
; i 1 ;
H ! i"
• t ' •
i , i •


-^-1
;";
• } -i-4 4-
-i- 1 i-4-
IJ-tt
TTH"
. i , i-
liii:
•

i : ' ! T
..,,..

^
~

• • i .
' ';
i i . ,
i
....
. 4 4-
i 4 J--
i i j I
-ti-4-
4 *-1~
4|
-4'flT

13 !
t •* i
•i-i n
-* 1 4 .
; . -J -i-.
•at
* 1 * '
. i
. M
• i±i
. *- * *
,11.
O-

(-•* 4 •
iii!
Q ;t
\£
TT
_Of"j]
M{TF
•; f pp
•4
T*~T
:Ht
S
11
1 1 t
ENGINE SPEED -rev/S.
V

-------
    NATURALLY ASPIRATED 96 kW V8  IDI ENGINE
                                                 Drg. N«. D 14094

                                                 DM*   lft-7-74
  PREDICTED  NOx ppm LEVELS IN LOW  EMISSIONS  BUILD
NOTE; CVS-CM COMPUTER PROGRAM PREDICTS  1-17 Q/milV i'H !
rrrtrrnrii-iTtr;!!]::;:!:: •. |- • ;• i :•• "t ••:!:: > -i- • MJ ;;:. i;: i;.. ^< inUlM:
     t-J-
     I I
   ... H
   fi H
       Ox PROM THIS ENGINE BUILD IN
      -M-t-H
L3SOO Ib
Iff!
PASSENGER CAR.
                         U-L-.:.
                                < i •
                                A 1590 Kg
                                     •t i i : i * ' *
                                     r '  IZO
                          FULL TORQUE CURVED  IIO.r
                     ENGINE  SPEED rcv/S
                                                          Htf
                                                                !:tt

-------
                                   5-87
       NATURALLY  ASPIRATED 9C> KW V8 IDI ENGINE



    PREDICTED  HC ppm  LEVELS  USING  CURRENT



             FUEL  INJECTION  EQUIPMENT.
'^ NOTE: CVS-CH COMPUTER  SIMULATION  PROGRAM
    TTLfrr-: tr; ::i::-. •. • • • -1 • •! • f • • i r: • • • i;: • :r : i • i    • <'' • < t • >:: if;; j

    t-zr PREDICTS O-Gag/mile HC WITH THESE LEVELSftr
    .1 rU.| i 1.4 . | .- . , - ) , . . . | . . . . I ?/. . I . . , . | , . i . | . . . . , . , . | . , , . I . , . M • • - • 4 ,. r

    ::,r IN A 1590 kg (35OOIb) PASSENGER CAR';!!-{ij.ij- 'i |

    i.:::|:!::i!..:;;I:;fr!T:-:|::::':!i!l::[:i:;::|:::;h:;; !i!i :!;: nil
                                   FULL TORQUE CURVE '
                      ENGINE SPEED-rfcV/S.
*

t
.. 1
::
Li
] ^

VE
1!
• ;





; i
li

" ;
H
.!
i
T *-
-
, [
. . .
I ' '
-t
t
•4
-f
.'
T *



• T


J-r
0
r*i
M




»

i

















j

}










r


f
r
c

i i .
t *"
1 !-i

; i !





i ; ,
t!'


;
-][
t !
T '" I '
t * *

H"
1
-. .
<\l
c
ID
i .
il!
•lt1
H
n1
lit
4 H

l'.i" i
f- r *--
. . L
. . i
t - •
• t •

•1 -t
1 1
i ,

t- t 1
j; ;
_L:I

IG. Nc
>rg. N<
>ate

t ' t *
£O-

-t Ml
: i n
110 -
. I ...

100"

.

.11.
fak A
30 "
MM
oX 1
SO"
ffi!
ill;
....
Hi;
; i i
£O "
teu
i i"' j
50-
•t-l->- •
40-
nn
; 1
J3d-
r


t ' t- *
io-
1 i !-.

- -j-i •
ill

t . , .
1 : *
' t-U

>. i
> D£
28

T]Tj~
•^A r-
t
7 -I
i.


' ! I '

. . i ..
61'
T

^


• i 4
' ' * 1
t i
"* ^T '
i ; . :
i }-i
' 4 -*•

li:
_ 5 i
"{:

11..
• I.I I
• t-tj
". 1 llj
' f-
' i *

. [ 4- i

t1
1 i t
rr-t [-
flit'

5-37
!&0£
-7-

4-Ul.
M
it tt
t-4 -t r-
4^
'Jtff
. 4- - -4-.
~i-r-j-
i 4 i- f-

' ' T '
j j
! (...
.. t ..
i T"
jiit
T~
|I
iii"
i-J-
-'Tt
rt "* 1
W "T *
1 -tt-
atr
A\ j~i
E it
m
-h-
-1 4 -
, t T
1-4 4--
'Ifl""
I-U-L

lit
*- i— f--*--
T T~rl~
M-~vt-
1 • i i I
•;4
-!(-

)5
74






































-------
       NATURALLY ASPIRATED 9G 
-------
                              5-89
  TYPICAL HC EMISSIONS  FROM  A

LIGHTLY  BOOSTED CONVENTIONAL  Dl JNl

      OPTIMUM  PERFORMANCE BUILD.
FIG NO.  5-33

Drg. No. D£GO88

Date   28-7-74
               Xo ENGINE RATED SPEED
                                                                    i
                                                                   Q
                                                                   Q)
                                                                   E
                                                                   12

-------
                                  5-90
     TYPICAL NOx  EMISSIONS  FROM A
LIGHTLV BOOSTED CONVENTIONAL D.I. ENGINE IN
       OPTIMUM  PERFORMANCE BUILD.
FIG. Nr  5-4O
Drg No D2
-------
                              5-91
 TYPICAL  HC  EMISSIONS FROM A
LIGHTLY BOOSTED COMET  V  ENGINE IN
  OPTIMUM PERFORMANCE  BUILD.
FIG. No.   5-41
Drg. No DE609I
Date   aS-7-74
                 ENGINE RATED

-------
TYPICAL  NOx EMISSIONS FROM A LIGHTLY
        BOOSTED COMET V ENGINE
 IN OPTIMUM PERFORMANCE  BUILD
FIG r i 5-42.
Drg No D2.6O9O
Date   £8-7-74
           7o ENGINE RATED SPEED

-------
                                    6-1

                                 SECTION 6

                            POWER PLANT RATING
The quantitative comparison of different power plants was one of the major
aims  of the  study.  This section describes the rating methodology which was
developed, how it was applied to the eleven potentially viable power plants
and the results of that application.

The rating methodology involves the identification of performance aspects or
requirements for a  power plant.  Each aspect  is  given a weighting indicating
its  relative importance.   Each power plant is then given a  'rating1  indicating
how well it met the requirements of each performance aspect.  Multiplication
of each  'rating1  by the appropriate 'weighting1 arid summation of the products
then gives a numerical  'overall rating1.  Comparison of the  'overall ratings'
allows the power plants to be compared on  a numerical basis*

The individual  ratings for each performance aspect for each power plant are
listed  and the  reasons for these ratings are discussed.

The 'overall ratings'  indicated that the gasoline  engines  were some  10%
superior to the diesel engines for  an average light duty application  in the
primary emissions environment (HC - 0.41 g/mile,  CO  -  3.4 g/mile,  NOX -
1.5 g/mile.)  The results,  however, indicated that greater  weighting on fuel
consumption  and certain  other economic and durability aspects would make  the
diesel at least as attractive  as  the gasoline engine.  Only  the four-stroke
IDI diesel power plants were viable, all the  other diesel variants being incap-
able of meeting the emissions targets.

Since it was considered that none of the diesel power plants could meet the
secondary emissions targets (HC - 0.41 g/mile, CO -  3.4 g/mile, NOX -
0.4 g/mile), no attempt was made to  carry out  the  rating  exercise for this
secondary environment.

-------
                                6-2

Introduction

One of the major aims  of the study was that a methodology should be  derived
which would  allow  a quantitative assessment of the relative merits of various
power  plants for light duty vehicle use.  Although the study was concerned
only with a comparison of gasoline and diesel configurations  the methodology
was developed  so that it could be applied to any  liquid hydrocarbon power
plant and thus  should be of  value in other similar studies.

The  advantages of such  a methodology are that its application will  allow a
direct  quantitative  rating of various power plants and  that  it  should also be
possible  to identify those factors and aspects  which render a  particular
power  plant suitable for a particular duty.   The second  advantage allows  an
assessment of  changes in a particular  area  as well as  highlighting areas
worthy of effort  to  make a particular  configuration more suitable for  use in
a given environment.

Approach of  Rating  System

The  fitness of any  power plant for a given duty is  a combination of the  excell-
ence with which  it meets various performance aspects or requirements and  the
relative  importance of those individual aspects.  The  application of any rating
system must thus involve five stages:

     l)   Identification of performance aspects
     2)   Estimation  of relative importance of those aspects
     3)   Estimation  of how well a particular power plant meets  a performance
         aspect

     4)   Assessment of  overall  merit of that power plant

     5)   Comparison with overall merit of other power plants

Performance Aspects

The  fitness of a  power plant for light  duty use can be assessed under  the
following  broad headings:

     a)   Emissions

     b) •  Package • (Size,  weight, etc.)
     c)   Costs

     d)   Nature  (Driveability)

     e)   Others   (Convenience and minor safety aspects)

-------
                                6-3

Unfortunately these headings are much too broad for a detailed  assessment
of different power plants and a more  detailed list of performance aspects
was  drawn  up such that the various aspects would  cover all facets of light
duty vehicle power plants when fuelled by liquid  hydrocarbons.  The  indivi-
dual performance aspects are  as follows:

                    1.  Smoke
                    2.  Particulates
                    3.  Odour
                    4.  NOX
                    5.  HC
                    6.  CO
                    7.  S02
                    8.  HC reactivity
                    9.  Evaporative emissions
                   10.  Misc. emissions
                   11.  Drive-by Noise
                   12.  Package volume
                   13.  Package weight
                   14.  Fuel economy
                   15.  Fuel
                   16.  Vehicle first  cost
                   17.  Maintenance cost
                   18.  Startability
                   19.  Hot  driveability
                   20.  Cold driveability
                   21.  Torque rise
                   22.  Durability
                   23 .  Coolant heat loss
                   24.  Fire risk
                   25.  Idle noise
                   26.  Vibration and  torque  recoil

It will be seen that although the performance  aspects are generally those
studied in the literature  survey certain others have been added  so that the
rating methodology would  be complete.

Relative  Importance of Performance Aspects

If the  summated importance  of all the performance aspects is 100 then each
individual aspect can  be  assigned a numerical importance or''weighting'.
For this study a committee of six  Ricardo  personnel was selected and they
were  asked to assign a 'weighting1 to each of the performance aspects.  Al-
though the committee members were selected so that each member had
experience  and knowledge of both the  emissions field and the  American auto-

-------
                                6-4

motive situation it was felt that the corporate results  may  have carried a
bias and it was thus decided to try to eliminate this bias.   A further group
of eighteen persons was then selected (again with experience of emissions
and American automotive conditions)  and this  group of people was  asked to
assign weightings  in isolation.   A comparison of this  survey and committee
results  (Fig.  6-1) shows that although there was general agreement between
the two,  discrepancies were  sufficient to indicate that there may well have
been slight committee bias and  thus the survey  figures were adopted for
this  study.  These weightings  are shown in Table 1 below.

                                 TABLE 1

                      Final Weightings Used In Study

                      Aspect                             Weighting
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Smoke
Particulates
Odour
NOX
HC
CO
so2
HC reactivity
Evaporative Emissions
Miscellaneous Emissions
Noise (Drive-by)
Package volume
Package weight
Fuel economy
Fuel cost
Vehicle first cost
Maintenance cost
Startability
Hot driveability
Cold driveability
Torque rise
Durability
Heat loss
Fire risk
Idling noise
Vibration and torque recoil
4.48
2.14
4.48
3.92
3.99
3.61
3.48
1.83
1.60
0.98
6.32
2.61
2.59
12.20
5.40
4.65
4.35
4.85
4.48
3.52
1.98
4.80
2.18
3.55
3.83
2.18

-------
                                 6-5

How Well a Performance Aspect or Requirement is  Met

It is necessary that a rating  scale be devised so that a quantitative assess-
ment of how well a particular power plant  meets a given performance aspect
can be made.   The above list of performance  aspects shows that although
some  aspects could be  quickly  assessed in a numerical  fashion, many others
are essentially  qualitative  and any rating scale should be able to cover  all
aspects.

As expected,  some difficulty  was experienced  in  relating a purely subjective
impression to a linear  quantitative scale,  but  after some consideration the
following system was adopted as giving the numerical scale several easily
relatable,  subjective  key points, the  numbers  without definition being an inter-
polation of  the  surrounding  merit definitions.

     Merit rating  scale

     0  Totally unacceptable
     1
     2  Bad
     3
     4  Poor
     5  Acceptable
     6
     7  Good
     8
     9  Best practical
     10 Perfect

Assessment of  Overall  Merit of the Power  Plant

The rating system  evolved allows an  immediate quantitative assessment  of the
overall merit of the power  plant and this is accomplished by multiplying each
aspect  'rating1  by  its appropriate 'weighting1  and summing all the products.
With a total weighting of 100  and a merit scale of 0-10  as  above the maximum
possible is 1000.

The relative merit  of various power plants can be assessed immediately by
comparing their total scores,  the  power plant with the highest  score being the
best.  An idea  of  the absolute  merit of the power plants can also be obtained
if the  score is  divided  by  100 and the quotient related to the above rating
scale,   e.g.  a score of 1000/100 =  10 is a 'perfect1 power plant.   A  score
of 500/100  = 5  is  an 'acceptable'  power plant.

-------
                                6-6
Use of the Rating System in the Study
In order to apply the  rating system  to the power plants  considered in this
study a committee: was uscu to assess tucr various ratings.  *ne  comrriii.cG
consisted of  five experienced members  of the Ricardo  staff and great care
was taken  to ensure that the committee had no  bias  to either diesel or
gasoline  power plants.  The power plants considered were those described
in the 'engine  configuration1  section  of the report with the addition of the
two gasoline engines described briefly in the same section.
     i.e.    l)   V-8  'American1 gasoline
             2)   I.L.-6 'European'  gasoline
             3)   V-8 I.D.I, diesel
             4)   6 cyl. I.D.I,  diesel turbocharged
             5)   6 cyl. I.D.I,  diesel with domprex
             6)   6 cyl. D.I. diesel turbocharged
             7)   6 cyl. D.I. diesel with Comprex
             8)   6 cyl. 2-stroke loop scavenged I.D.I, diesel
             9)   6 cyl. 2-stroke uniflow D.I. diesel
            10)   4 cyl. 4-stroke compound D.I. diesel
            11)   2-stage 2 bank rotary diesel
Each of the above power plants was  to  be considered for the primary and
secondary  emission levels  of the study.
     i.e.    Primary Targets
             HC    0.41 g/mile
             CO    3.4  g/mile
             NOx  1.5  g/mile
             Secondary Targets
             HC    0.41 g/mile
             CO    3.4  g/mile
             NOX  0.4  g/mile
The above  emission levels  being  measured according to  the  CVS-CH  test
procedure.

-------
                                6-7

In fact  as  it was  considered that none of  the diesel power plants could pos-
sibly meet the secondary targets (in particular the 0.4 g/mile NOX figure)
the rating  system was only applied  to  an  environment embracing the primary
targets.

RESULTS OF  RATING  ASSESSMENT
The following 26 sub-sections give the 'ratings'  given  to each  of the eleven
power plants  on each performance aspect.

Each  sub-section is preceeded by a summary table giving the numerical ratings
and the sub-section proper  then  follows giving notes  on the derivation of the
various scores. Since a verbatim account of the committee deliberations
which led to  each score would be tedious to  the  reader the first aspect
'smoke1 is described  in some detail  to illustrate the process while following
aspects are  covered briefly unless major contentious points arose.

The final  ratings are  also shown in Table 2 at the end of the 26 sub-sections.
1.  SMOKE
Engine                         Score

V-8  gasoline                     9
IL6 gasoline                     8
NA V-8  I.D.I.                   6
TC 6 I.D.I.                    4.5
Comprex 6 I.D.I.                5
TC 6 D.I.                      4.5
Comprex 6 D.I.                 5
Loop scav. I.D.I.                4
Uniflow D.I.                     5
Compound D.I.                   3
2 stage rotary                   2

For  commercial vehicles  the absolute smoke level at which the engine is
limited varies considerably from country to  country and even between manu-
facturers ,  some simply complying with legislative requirements  whereas other
companies  aim  for  significantly  lower levels.  For passenger car use,  maxi-
mum smoke levels  must be selected on  an aesthetic basis in order to avoid
public criticism.  From  European experience,  Ricardo would recommend that

-------
                                6-8
maximum steady state smoke levels  of  5-8% opacity  should be aimed for.
The combination of these low smoke  levels,  with the high power/weight ratio
of the vehicle  should  result in the exhaust being at least  acceptable from
most  of  the candidate diesel power plants.

V-8 gasoline engine

Although the use of full load  mixture enrichment devices  can caused visible
black smoke with hard accelerations, the gasoline  engine per se  is  com-
pletely free of visible smoke problems  and can be classed  in the category
of "best  practical".   Blue  smoke  can be formed in a worn  gasoline  engine
but this  is also true of any other conventional internal  combustion engine.

Score -  9
6 cylinder gasoline engine

Again the smaller more highly  rated gasoline engine will only suffer a
smoke problem under acceleration conditions  but  its smaller size infers
that  mixture enrichment will occur  over a larger part of its load range
and therefore for a longer proportion  of a given acceleration.   Although
marginally inferior to  its larger brother the small  180 CID  gasoline engine
must still be regarded  as extremely good when considering  visible smoke
emissions.

Score -  8
NA V-8  I.D.I.
Black smoke is the major  criterion which controls the output of any naturally
aspirated diesel engine.   The smoke level  selected for  the  performance esti-
mates will result in a very small  amount of smoke being visible from the
kerbside, although at levels  and for time periods which are better than just
acceptable to the general public.   As engine load is reduced,  black smoke
disappears rapidly from  I.D.I,  engines so that this power plant would be
better than  acceptable from the point of view of black smoke.

White/blue smoke due to incomplete  combustion of fuel  should not be a pro-
blem from any of the indirect injection chamber engines.

Score -  6

-------
                                 6-9
TC  6 cyl.  I.D.I.

The  smoke  characteristics from  this lightly boosted engine will be very
similar to those of the naturally aspirated V-8  although it will  probably
suffer a transient  smoke  problem.  This occurs when the turbocharger
inertia  prevents the  turbocharger output  from keeping pace with the engine's
demands, the resultant over-rich fuel/air ratio leading  to black smoke.
For  an  engine of this rating  such a problem  would normally  be confined  to
accelerations from very low speeds where turbine and compressor effici-
encies are  low.

The  magnitude of  this problem in terms  of smoke  would be slight but would
probably be termed as  annoying  to the casual observer.   Experience has
shown it to be almost  insuperable  if vehicle  driveability is to remain un-
altered.  Another  annoying aspect  of this problem is  that it  brings the
diesel's smoke right to the city centre and suburbs  (the  slight smoke problem
of the naturally aspirated diesel  occurs at  a  higher speed/load and will not
normally be apparent during in-city driving).

Score - 4.5
Comprex boosted  6 cyl.  I.D.I.

The reputedly instantaneous flow response of the Comprex  pressure exchanger
to load and speed  should improve the smoke emissions from a diesel engine
considerably.  However, much  time has been put into development of this
device without a production solution  being achieved  and the full range per-
formance and durability of Comprex has yet to  be proven,  At its present
state  of  development there is some confidence that  this power plant  will be
better than its turbocharged  equivalent.

Score  -  5
TC  D.I,
Black smoke levels from  the turbocharged D.I. should be similar to those
from the turbocharged I.D.I, assuming a practical  method of achieving the
required  NOX levels is  developed (current technology indicates  that the D.I.
engine will have great difficulty  in achieving 1.5 NOy during CVS-CH) with-
out resorting to excessive amounts of timing retard.   Even with small
amounts of retard a blue smoke problem may  exist.

Score -  4.5

-------
                               6-10

Comprex boosted D.I.

Again the theoretical advantage of instant transient response  of the pressure
exchanger  make this candidate more  attractive than the  turbocharqed D.I.
and this power plant should be as good as  the  I.D.I, version.

Score - 5
Loop scavenge 2 stroke I.D.I.

Black smoke response  from the 2 stroke I.D.I,  should not be any worse than
from a four stroke  I.D.I,  but  blue smoke may be a problem if large quanti-
ties of lubricating oil are burnt - due to the proximity of the inlet and exhaust
ports and their position within  the cylinder there is every likelihood  of this
occurring.   It is this latter aspect which renders it likely that  this engine will
be less than acceptable.

Score - 4
Uniflow 2 stroke D.I.

The blue smoke problem associated with the loop scavenge two stroke should
not occur on this engine and providing adequate scavenge  can be provided black
smoke should not be a problem even with some timing retard (note it  is
extremely unlikely that  this engine with a D.I. chamber could achieve  the pri-
mary NOx target and some device -  e.g.  exhaust gas recirculation. would be
required in addition to timing retard).  It was considered that this engine
would be capable of achieving an acceptable smoke rating.

Score - 5
Compound  D.I.

With its  small radial compressor driven  direct from the crankshaft, parasitic
losses at low engine speeds in addition to low boost pressures would force the
manufacturer to  aim for maximum  air utilisation at low  speeds giving high
smoke levels.  The extremely high boost ratio would require the adoption  of
low compression ratios which would introduce light  load  misfire  problems  and
white/blue smoke during and immediately after a cold start would  almost cer-
tainly be found.  This  engine would be somewhat worse  than poor from the
point  of  view of  smoke.

Score -  3

-------
                                 6-11
2 stage rotary  engine

The  successful  development of an automotive  rotary diesel engine would need
years  of  research into  optimum combustion systems and  even then it is un-
likely  that competitive  air utilisation levels could  be achieved.  The clear-
ance volume between the rotor and trochoid is far removed  from the optimum
as it contains  large areas which are too shallow and inaccessible to  support
or allow  the propagation of combustion.  This could lead to  smoke problems
at both ends of the load range,  blue  smoke at light load,  and black smoke at
high load with the distinct possibility of a "plateau" smoke problem bridging
the two.  There is  little doubt  that this engine will  be  the worst of all those
assessed  from  this aspect and  should be rated as  'bad'.

Score  - 2
2.    PARTICULATES

Engine                          Score

V-8  gasoline                     7
IL6 gasoline                      7
NA V-8  I.D.I.                   2
TC  6 I.D.I.                      2
Comprex 6 I.D.I.                2
TC  6 D.I.                        2
Comprex 6 D.I.                  2
Loop scav. I.D.I.                2
Uniflow  D.I.                     2
Compound D.I.                   2
2 stage  rotary                   2

Currently the  data is insufficient to come to a final decision concerning  this
topic especially  as  the  true reactivity of particulates has  yet to  be deter-
mined.  It was argued  that  if the gasoline engine running on lead  free fuel
were regarded as good, or  worthy of  seven merit  points, the Diesel  engine
with particulate  levels  of 10-20 times that of  the gasoline must warrant  a
rating of bad or  two  points.   Ricardo  do not feel qualified to predict  relative
particulate levels between the different Diesel variants and  therefore  award
the  same  score  to all.   If  particulate levels are to be limited by  legislation,
the  Diesel would  need hang-on  soot or particulate filters and it is  in  this
area that  further work  could be usefully carried out.   The true health hazards
of various sizes of oarticles should also be determined before leqislation is oassed.

-------
                                6-12

3.    ODOUR

Engine                          Score

V-8 gasoline                      7
IL6 gasoline                      7
NA V-8  I.D.I.                    4
TC  6 I.D.I.                      4
Comprex 6 I.D.I.                 4
TC  6 D.I.                        3
Comprex 6 D.I.                  3
Loop scav. I.D.I.                 3
Uniflow D.I.                      3
Compound D.I.                    3
2 stage rotary                    2

It was generally agreed that the gasoline power plants are good  from the point
of view of odour even though the exhaust fumes are noticeable under  cold con-
ditions.   All the diesel plants,  however, are noticeably odorous  and although
the modern automotive I.D.I, engines are  good as diesel  engines go they are
poor by  comparison with the gasoline engines.  Light  load odour is perhaps
the most objectionable while the full load odour at low air/fuel ratios  is  also
noticeable.

The  direct injection four-stroke engines will  be worse at  both  light load and
full  load than the indirect injection  engines due to the more  limited capa-
bilities of their combustion  system.

The  two-stroke  engines will have similar characteristics to the direct  injection
four-strokes.

The  compound engine  may have  a problem  with odour  at low loads but is un-
likely to be much worse overall than the direct injection  engines.

The  rotary engine will be very  odorous  due to poor scavenging of  end  spaces
and  the difficulty of arranging for complete combustion over a wide load  and
speed range.

-------
                                6-13

4.   NOX

Engine                          Score

V-8 gasoline                     5
IL6 gasoline                      5
NA V-8  I.D.I.                    5
TC  6 I.D.I.                      5
Comprex 6 I.D.I.                 5
TC  6 D.I.                       (0)
Comprex 6 D.I.                 (0)
Loop scav. I.D.I.                 5
Uniflow D.I.                    (0)
Compound D.I.                  (0)
2 stage rotary                    5

On  performance aspects involving legislative requirements,  the manufacturer
will be forced by mainly  economic  considerations to produce  a powerplant
which just satisfies the mandate (unless the engine already betters these
requirements in optimum performance build).   Thus for all candidate power
plants  where some  reduction in pollutants is necessary to achieve the project
targets,  a classification of acceptable was awarded even though the potential
of the engine might have  been substantially better  than regulatory require-
ments  (since further  reduction  would involve other penalties).

Gasoline engines have already demonstrated  their ability to achieve 1.5  g/mile
NOX during  CVS-CH and it is predicted that a 130  bhp I.D.I, diesel should be
capable of attaining thi s level although with  a  significant  degree of timing
retard.  Heavy duty experience indicates  that  the D.I. engine will not be cap-
able of attaining this  level  and  all variants with this chamber have been
awarded a zero score.  When considering the  secondary NOX target of 0.4
g/mile,  it is known that several gasoline engines have achieved this level at
zero miles but catalyst durability is  currently low.  Daimler-Benz have demon-
strated a 65 bhp Diesel  powered vehicle with NOX levels  of 0.4 g/mile but
with the heavier vehicle and higher power/weight ratio of the vehicle for the
study it  was felt that this target is not achievable with current technology and
production methods.

-------
                                6-14
5.   HC

Engine                          Score

V-8 gasoline                      5
IL6 gasoline                      5
NA V-8  I.D.I.                    5
TC  6 I.D.I.                      5
Comprex 6 I.D.I.                 5
TC  6 D.I.                       (0)
Comprex 6D.I.                 (0)
Loop scav. I.D.I.                (0)
Uniflow D.I.                     (0)
Compound D.I.                   (0)
2 stage rotary                   (0)

As  demonstrated earlier, it  is  estimated that with minor development,  the
I.D.I, engine should be  capable of complying with the  0.41 g/mile  require-
ment over CVS-CH.   The D.I.  in retarded mode  (in order to  minimise NOX
but still not achieving  the target of 1.5 g/mile)  is likely to suffer  extremely
high HC's and in Ricardo's opinion, none of the D.I. candidates would be
capable of complying with the HC requirement of this project.  The poor
combustion chamber  shape of the 2 stage  rotary engine is hardly conducive
to efficient combustion and unburnt  HC's are likely to be very high.  Unfort-
unately the rapid  drop in exhaust temperature  as load  is  reduced ,on the diesel
engine is likely to restrict catalyst activity to the higher loads only and their
effectiveness during CVS-CH on a high powered  diesel vehicle is open to doubt.
Investigations into low temperature  light-up catalyst  and catalyst matrix design
(for carbon storage and  to ensure sufficiently  free access of the HC's to  the
catalyst)  could  be an area worth further investigation.
6.   CO

Engine                          Score

V-8  gasoline                      5
IL6 gasoline                      5
NA V-8  I.D.I.                    6
TC 6 I.D.I.                      5
Comprex 6  I.D.I.                 5
TC 6 D.I.                        5
Comprex 6  D.I.                  5
Loop scav.  I.D.I.                 5
Uniflow D.I.                      5
Compound D.I.                    5
2 stage rotary                   (0)

-------
                                 6-15
There should be no problem in achieving  the  CO targets for any of these
engines  apart  from the two stage rotary  engine where it is felt  that the
inefficient combustion  will again result in higher  emissions.

The naturally aspirated V-8 I.D.I,  should achieve the targets without any
modifications and therefore merits  6 points whereas  the remaining candi-
dates score 5.
7.   S02

Engine                         Score

V-8  gasoline                    7
1L6 gasoline                     7
NA V-8  I.D.I.                   4
TC  6 I.D.I.                     4
Comprex 6 I.D.I.               4
TC  6 D.I.                       4
Comprex 6 D.I.                 4
Loop scav. I.D.I.               4
Uniflow D.I.                    4
Compound D.I.                  4
2 stage rotary                   3

This pollutant is totally dependent  on the sulphur content  of the fuel and the
amount of fuel  being burnt.   In  assessing or rating  the Diesel  forSC>2  it was
assumed  that the sulphur content of US  light distillate (DFl)  is  similar to
that found in Europe (i.e. 0.2-0.5%).  In theory the sulphur  can be completely
extracted from  the fuel but economic considerations have ruled this out to  date<
The  reactions of SC>2 in  the exhaust to  form  sulphates/sulphides should not be
overlooked;   the  addition of oxidation catalysts would probably  increase the
extent of these.  The sulphur content of gasoline is  very low (<0.1%) and  thus
the gasoline engines were rated as good.  Diesel  engines must rate a  merit
number of only 4  apart from the 2 stage rotary engine, where the extremely
high fuel consumption will drop  the  rating by one  further number.

-------
                               6-16

8.   HC REACTIVITY

Engine                        Score

V-8  gasoline                    5
IL6 gasoline                    5
IN A V-8  I.D.I.                  7
TC  6 I.D.I.                    7
Comprex 6  I.D.I.               7
TC  6 D.I.                       7
Comprex 6  D.I.                7
Loop scav.  I.D.I.               7
Uniflow D.I.                    7
Compound D.I.                  7
2 stage rotary                  7

This subject is  an  important aspect of HC emissions  and as mentioned earlier,
HC's emitted by the Diesel under high load conditions  may be  conveniently
wrapped in  the  soot skeleton to form a coagulum which effectively  bars their
escape to the atmosphere or animal tissues.  At light  load conditions, blue
smoke can  be found and  here the HC's are  freely available.   However it  has
been demonstrated that HC reactivity from the  gasoline engine  is 10  times
higher than  from the Diesel.  Thus if all gasoline variants warrant an accept-
able rating  then the diesels should  be at  least two points better.
9.   EVAPORATIVE EMISSIONS

Engine                        Score

V-8  gasoline                    5
IL6 gasoline                     5
NA V-8 I.D.I.              '    7
TC  6 I.D.I.                     7
Comprex  6 I.D.I.               7
TC  6 D.I.                      7
Comprex  6 D.I.                 7
Loop scav. I.D.I.               7
Uniflow D.I.                    7
Compound D.I.                  7
2 stage rotary                  7

-------
                                6-17

With modern control systems  using  carbon filter/storage canisters,gasoline is
an acceptable and practical fuel.  Using the  same  systems, mid-distillates
such as  used in nigh  speed Diesel engines would suffer marginal evaporative
losses and  must merit a good rating.  This  assumes fuels similar to current
No.  1,  wider cut fuels  may use more  of the light  end  fraction which could
close the gap between diesel fuel and gasoline.   One disadvantage of having
a fuel  of low volatility is that if a fuel leak  occurs the fuel does  not evapor-
ate and its  odour can be detected for some considerable time.
10.   MISCELLANEOUS EMISSIONS

Engine                         Score

V-8  gasoline                     5
IL6 gasoline                     5
NA V-8  I.D.I.                   5
TC 6 I.D.I.                     5
Comprex 6 I.D.I.                5
TC 6 D.I.                       5
Comprex 6 D.I.                 5
Loop scav. I.D.I.                5
Uniflow D.I.                     5
Compound D.I.                   5
2 stage rotary                   5

At the time of the study none of the power  plants was  regarded as producing
major quantities of any pollutants other than those already considered so that
all were  rated as equal and acceptable.

-------
                               6-18
11.  DRIVE-BY NOISE

Engine                        Score

V-8  gasoline                    7
IL6 gasoline                     6^
NA V-8 I.D.I.                  6
TC 6 I.D.I.                     6
Comprex  6 I.D.I.               5
TC 6 D.I.                      4
Comprex  6 D.I.                 3
Loop scav. I.D.I.               6
Uniflow D.I.                    6
Compound D.I.                  6
2 stage rotary                  6

As mentioned earlier,  the Diesel powered vehicle is  only  slightly noisier  than
its gasoline counterpart when driven at speed.   (Based on experience with
naturally  aspirated indirect  injection engines).   The V-8 gasoline  powered
vehicle,  being very quiet under drive-by  conditions (75-80 dB(A)  at  15m
(50  ft))  must be given a drive-by rating of good whereas  the 6 cylinder
European  engine working  with lower  drive ratios  will be marginally noisier.
Within the Diesel range,  the indirect chamber engines will be inherently
quieter than  their D.I. equivalents.   With all engines operating in  a  retarded
timing mode, absolute  noise levels of the I.D.I, engines will be  only margin-
ally  higher than the gasoline power plants.  Predicted noise  levels of the
candidate power plants are  listed below.   These predictions are for the bare
test  bed  engine  at  a  distance of 1 m (3.2 ft).   To correlate these with
American drive-by noise test conditions one must subtract approximately  20-
22 dB(A).

Bare engine test bed noise levels  (l  m)

Gasoline V-8  (97  mm bore, 66.7 rev/s, 93 dB(A))
Gasoline 6 (88 mm bore, 82.5 rev/s,, 96  dB(A))
NA V-8 I.D.I.              '         97
TC 6 I.D.I.                          96
Comprex  6 I.D.I.                    96
TC 6 D.I.                          102
Comprex  6 D.I.                     102
Loop scav. I.D.I.                    93
Uniflow D.I.                         96
Compound D.I.                      100
2 stage rotary                       93

-------
                                 6-19

The formulae used in predicting the noise levels are empirical only and assume
current technology is used in the crankcase design - significant reductions
could be obtained by substantial modifications to the basic engine design.  These
formulae also only predict combustion noise which is normally dominant in
these engines and take no account of other possible sources which could raise
the overall noise levels.  When considering the boosted DI and IDI four stroke
engines allowance has to be made for the high sensitivity of the Comprex to
exhaust back pressure and general design restrictions throughout the flow
system (because of this the silencing of both intake and exhaust could  not be
made as effective as the silencing of the turbocharged engines and some
additional noise penalty will be incurred).  Of the last four candidates, so
little information is known of their noise characteristics, particularly in this
power  range, that the degree of confidence in the predicted noise levels must
be low.
12. PACKAGE VOLUME

Scores

Engine                       Score
V-8 gasoline                   7
IL6 gasoline                    8
NA V-8 IDI                    6
TC 6 IDI                       5
Comprex 6 IDI                 4
TC 6 DI                        5
Comprex 6 DI                  4
Loop Scav. IDI                 4
Uniflow DI                     3
Compound DI                   5
2 stage rotary                 8

Box volumes as calculated during the design configuration phase are listed as
follows :-

                             Box Vol.ft3
V-8 IDI                        11.2
IL 6 IDI TC                     12.7
V-6 IDI TC                     11.6
IL 6 IDI Comprex               12.1              (.34)
V-6 IDI Comprex
IL 6 DI TC                      12.0          '    (.34)

-------
                               6-20


V-6 DITC                     11.0              (.31)
IL 6 DI Comprex               12.2              (.34)
V-6 DI Comprex                 -                 -
V-6 2 str. Loop Scav-           12.7              (.36)
IL 6 2 str. Uniflow             16.8              (.46)
4 cyl. Compound               11.3              (.32)
2 stage rotary                  9.2              (.26)

All are given merit ratings roughly in accordance with their calculated box
volume apart from the Comprex boosted engine where the disadvantage of
taking the exhaust from near the front of the engine is not illustrated.

Since package shape is of at least equal importance to package volume, this
was taken into account in applying ratings.  Engine length, for example, was
believed  to be important in view of safety regulations.
13. PACKAGE WEIGHT

Scores

Engine                       Score
V-8 gasoline                  6
IL 6 gasoline                  7
NA V-8 IDI                    5
TC 6 IDI                      5
Comprex 6 IDI                5
TC 6 DI                       5
Comprex 6 DI                 5
Loop Scav. IDI                4
Uniflow DI                    4
Compound DI                  5
2 stage rotary                 7

The weight penalty of the small high speed diesel engine is large when compared
with a European type gasoline engine but this penalty is reduced considerably
if a larger American type V-8 is used as basis for comparison. It is Ricardo's
opinion that only the European gasoline engine  can warrant a rating of good (7 points)
all other power plants apart from the rotary diesel considered in this exercise
are considerably heavier and must therefore warrant lower scores.

-------
                                6-21
Engine                 Est. weight Ib
V-8 gasoline                  550
IL-6 gasoline                  410
NA V-8 IDI                    700
TC 6 IDI                      720
Comprex 6 IDI                 680
TC 6 DI                       680
Comprex 6 DI                 660
Loop Scav. IDI                 760
Uniflow DI                    800
Compound DI                  670
2 stage rotary                 500
14. FUEL ECONOMY

Scores

Engine                      Score
V-8 gasoline                  5
IL 6 gasoline                  5\
NA V-8 IDI                    71
TC 6 IDI                      7
Comprex 6 IDI                 7
TC 6 DI                       8
Comprex 6 DI                 8
Loop Scav. IDI                 6£
Uniflow DI                    7
Compound DI                  8
2 stage rotary                 5

For this section it was felt that the rating system must be redefined on a quantitat-
ive basis and the following scale was devised.

Fuel economy mpg   (1/100 km)     Rating (normal subjective scale)
                                       0 (totally unacceptable)
        6              (39.4)            1
                                       2 (bad)
       10              (23.6)            3
                                       4 (poor)
       15              (15.7)            5 (acceptable)
                                       6
       20              (11.8)            7 (good)
                                                       (contd.)

-------
                                 6-22
Fuel economy mpg     (1/100 km)    Rating  (normal subjective scale) (contd.)
                                        8
        25             (9.45)            9 (best practical)
                                       10 (perfect)
Predicted fuel economy figures were based on estimated and measured fuel
consumption levels from engines of each type.  For the rotary engine, little
data are available but from basic combustion considerations and for the
higher friction levels of the rotary engine, specific fuel consumption curves
have been prepared from which comparative vehicle fuel economy levels
could be estimated.

Estimated vehicle fuel  consumption and'ecbnorriy levels during CVS-CH are :-

                      1/100 km           mpg
V-8 gasoline          18.2               13
IL6 gasoline           15.7               15
NA V-8 IDI           11.8-10.7         20,22
TC 6 IDI              12.4 - 11.2         19-21
Comprex 6 IDI        12.4-11.2         19-21
TC 6 DI              (11.2-10.3)      (21-23)
Comprex 6 DI        (11.2 -10.3)      (21-23)
Loop Scav. IDI       (13.1-11.8)      (18-20)
Uniflow DI           (12.4-11.2)      (19-21)
Compound DI         (11.2 -10.3)      (21-23)
2 stage rotary        (16.8-14.7)      (14-16)

Fuel economy levels in brackets indicate that although the injection timings of
the engines have been retarded as far as is practical,it is estimated that at
least one primary emissions target (normally NOx - see emissions section)
has not been  achieved.  The poorer  fuel economy of the boosted IDI as compared
with the naturally aspirated V-8 is due to its need for further timing retard to
comply with 1.5 g/mile NOx.
15. FUEL COST
Scores
Engine
V-8 gasoline
IL 6 gasoline
NA V-8 IDI
TC 6 IDI

-------
                                 6-23

Comprex 6 IDI                5
TC 6 DI                       5
Comprex 6  DI                5
Loop Scav. IDI                5
Uniflow DI                    5
Compound DI                 5
2 stage rotary                5

The fuel requirements of all the diesel candidates are similar to current light
distillate (DFl).  Due to a lack of real data on the economics of fuel refineries
and because the switch towards greater use of diesel fuels does not require
technological innovation, it is assumed that the fuel situation for both gasoline
and diesel candidates must be regarded as acceptable and a score of 5 has thus
been awarded.

It is estimated that the adoption of lead free gasoline would incur a price
penalty of 10% however, and if  sulphur free diesel fuel were stipulated,  a severe
cost penalty might arise,  but due to the uncertain future situation with regard to
fuel prices no attempt has been made to compare diesel and gasoline engine fuels.
16. VEHICLE FIRST COST

Scores

Engine                      Score
V-8 gasoline                  6
IL 6 gasoline                  7
NA V-8 IDI                   5
TC 6 IDI                      5
Comprex 6 IDI                5
TC 6 DI                       5
Comprex 6 DI                 5
Loop Scav. IDI                4.5
Uniflow DI                    4.5
Compound DI                  4
2 stage rotary                4

If the existing gasoline powered American car is regarded as slightly better than
acceptable in terms of first cost then the European type engine should give
sufficient advantage to warrant an extra point.  The first cost of all diesel
engines is undoubtedly a penalty and all the evidence  available indicates that a
cost penalty of some 10% will be incurred with a diesel powered vehicle even

-------
                                 6-24

after allowing for the cost of catalysts for the gasoline engine.  Because of this
the ' conventional' diesel powered vehicles would only be just acceptable in
terms of first cost.  (The potential cost savings with the  smaller boosted engines
are offset by the cost of the supercharging equipment).  The higher development
time and cost for the two stroke engines  will impose a surcharge on the vehicle
first cost so that these power plants must be rather less than acceptable.  The very
high development, manufacturing and material costs of both the compound and
rotary engines will pull them down to a 'poor' category  for this aspect.
17. MAINTENANCE

Scores
Engine                      Score
V-8 gasoline                  5
IL 6 gasoline                  6
NA V-8 IDI                    6
TC 6 IDI                      6
Comprex 6 IDI                6
TC 6 DI                       5
Comprex 6 DI                 5
Loop Scav. IDI                5
Uniflow DI                    6
Compound DI                  4
2 stage  rotary                 2

Data extracted from the literature survey and the experience of taxi operators
in Europe all suggest that minor maintenance costs for diesel vehicles  are
similar to those of gasoline equivalents but that major engine services for the
diesel are not required within the normal operating life of the vehicle.  The
average operating load factor for American passenger car engines is some-
what lower than  European vehicles (solely because of their higher power:weight
ratio) and the maintenance periods quoted for European use should be increased
when  considering the American situation.

Maintenance costs for  gasoline engines are regarded as satisfactory and
European experience indicates that the four stroke IDI engines should be better
than this. The direct injection engine uses plain hole nozzles which require
cleaning at more frequent intervals than the variable orifice pintle nozzle
used in  the indirect injection combustion systems.  Otherwise maintenance
requirements  are the  same as for the IDI engine.

Of the two stroke engines, the direct injection uniflow engine has been  given a

-------
                                 6-25

higher reliability rating mainly because of the excellent reputation of GM's
two stroke engines, to which this engine is closely akin.

The compound engine which is the candidate with the highest thermal loading -
0.00353 kW/mm2 (3.04 hp/in2\ and mechanical stresses - 137 bar (2000 lb/in2)
cylinder pressure - is likely to be a much more sophisticated design particularly
around  the combustion chamber area,and major cylinder head maintenance
periods are likely to be far shorter than for the other variants.

li; is felt that the 2  stage rotary engine would also need frequent servicing even if
only because the high cylinder pressures would cause high rates of seal wear and
the use of only one injector per bank with very high injection frequencies could
increase servicing demands in this area.
18. STARTABILITY

Sicores

Engine                      Score
V-8 gasoline                  6
IL 6 gasoline                  6.5
NA V-8 IDI                   5
TC 6 1D1                      5
Comprex 6 IDI                5
TC 6 DI                       5
Comprex 6 DI                 5
Loop Scav. IDI                4
Uniflow DI                    5
Compound DI                  2
2 stage rotary                2

With the ability to start instantaneously under most environmental conditions
experienced in America,  the gasoline engine should merit a very high rating
but the hot starting of low emissions vehicles is known to be poor and is not
likely to improve substantially beyond today's levels.  Thus the V-8 engine was
awarded 6 points but the 6-cylinder  engine with its fuel injection system should
start more easily and was thus awarded 6.5.

The starting characteristics of the naturally aspirated diesel could only be
classified as acceptable even with a programmed  start system.  The develop-
ment of an instantaneous warm up heater plug  which gives a clean start in all
conditions would improve the start up of the die^el to a significant extent.

-------
                                 6-26


With the low boost pressure ratios used on the 6 cylinder IDI engines it should
be possible to use the same compression ratio as in a naturally aspirated
engine without encountering excessively high cylinder pressures at full load
conditions.  Starting characteristics should therefore be unaltered.

Although large DI engines have extremely good starting  characteristics, tests
run by Ricardo on a high speed DI with a cylinder capacity of 0.5 1 (30 in^)
have indicated startability to be on a par with the IDI. This is due to higher
heat losses to the cylinder walls (a) because  of the higher swirl rates needed
for high speed operation,and (b) because of the higher surface areatvolume
ratio of the smaller cylinder. The compression ratio of the DI was similar
to that of the swirl chamber engine.

It is likely that the loop scavenge IDI will have a low  effective compression ratio
at cranking speeds with resultant inferior starting. It is difficult to quantify
this but a rating of 4 has been awarded if only to point out a likely problem area.
The uniflow two stroke should start as well  as the four stroke engines.

In order for the compound engine to attain its extremely high outputs without
exceeding its designed maximum cylinder pressures, its compression ratio will
need to be limited to 13:1 and severe starting/misfire/blue-white smoke
problems are foreseen.  Methods might be  developed for reducing the misfire
and light load smoke problems after starting but with current technology it is
extremely unlikely that a clean, reliable start could be  achieved in ambient
American winter  conditions.  A score of 2 points has therefore been awarded.

With the rotary engine, cranking heat losses between stages will be extremely
high as will be blowby losses at low speeds and it is difficult to envisage its
starting characteristics to be anything better  than bad.
19. HOT DRIVEABILITY

Scores

Engine
V-8 gasoline
IL 6 gasoline
NA V-8 IDI
TC 6 IDI
Comprex 6 IDI
TC 6 DI
Comprex 6 DI

-------
                                 6-27

Loop Scav. IDI                7
Uniflow DI                    7
Compound DI                 7
2 stage rotary                7

Hot driveability of all candidates apart from the two turbocharged four stroke
engines will in no way be inferior to current generation passenger car engines
(this assumes that all-speed governing used on most European diesel cars
is discarded in favour of simple idling and overspeed governing), and a score
of 7  points has been  awarded to all but these types.  The turbocharged engines
will  suffer a slight lag during acceleration because of tubocharger inertia and
the resultant feeling that the driver has not complete control over the vehicle
can be disturbing. It is claimed that the Comprex does not suffer this delay
problem.
20. COLD DRIVEABILITY

Scores

Engine                      Score
V-8 gasoline                  5
IL 6 gasoline                  6
NA V-8 IDI                   8
TC 6 IDI                      8
Comprex 6 IDI                8
TC 6 DI                       8
Comprex 6 DI                 8
Loop Scav. IDI                8
Uniflow DI                    8
Compound DI                  8
2 stage rotary                8.

With flat spots due to carburation problems at cold temperatures the V-8
gasoline engine can barely be regarded as acceptable.  The 6 cylinder gasoline
engine, however, will be better than this due to the use of fuel injection.  The
diesel engine pulls away immediately without hesitation once a start has been
achieved, the only difference between hot and cold driveability being due to the
higher friction levels of the cold engine.  Eight points were awarded therefore
to all the diesel candidates.

-------
                                6-28
21. TORQUE RISE

Scores

Engine                     Score
V-8 gasoline                 7
IL 6 gasoline                 7
NA V-8 IDI                   7
TC 6 IDI                     7
Comprex 6 IDI                7
TC 6 DI                      7
Comprex 6 DI                7
Loop Scav. IDI                7
Uniflow DI                   7
Compound DI                 2
2 stage rotary                7

All candidates achieve the torque characteristics demanded by the performance
specifications,  assuming a three speed automatic gearbox, with the one
exception of the compound four stroke engine.  The extremely poor torque
curve of this engine would need a very sophisticated transmission system to
achieve any degree of acceptance with the general public and only two points
were awarded to this engine.
22. DURABILITY

Scores

Engine
V-8 gasoline
IL 6 gasoline
NA V-8 IDI
TC 6 IDI
Comprex 6 IDI
TC 6 DI
Comprex 6 DI
Loop Scav. IDI
Uniflow DI
Compound DI
2 stage rotary
Durability of 1.5 g/mile NOx American gasoline engines is assumed to be satis-

-------
                                  6-29

factory on the assumption that catalyst life can be increased.  The European type
engine with its higher rating and operational speed range is likely to suffer
slightly higher wear rates but a satisfactory life of 100,000 miles can easily
be achieved.

The life of the diesel powered passenger car engine in Europe is significantly
greater than that of the gasoline engine.  The same arguments should apply in
American use although absolute life in terms of mileage will be greater in
America because of higher power: weight ratios (and therefore lov/er relative
load factors).  It is therefore Ricard's opinion that the durability of the naturally
aspirated IDI should be classified as good (7 points). The boosted engines, both
DI and IDI, will have marginally lower useful lives because of higher mechanical
and thermal loadings in and around the combustion chamber.   However, their
longevity will still be comparable to the American V-8 gasoline engine.

Due to their simplicity the life of the two stroke engines should be equivalent to
or better than the four stroke engines. The excellent reputation of GM 's
uniflow engines cannot be overlooked and for this reason the uniflow DI has been
rated at 8 points.

With considerable development the compound engine could achieve 160,000 km
(100,000 miles) in passenger car service (acceptable - 5 points) but the
durability of the rotary engine is open to speculation and  in Ricardo's opinion
it could not be developed to a satisfactory degree using current technology.
23. COOLANT HEAT LOSSES

Scores

Engine
V-8 gasoline
IL 6 gasoline
NA V-8 IDI
TC 6 IDI
Comprex 6 IDI
TC 6 DI
Comprex 6 DI
Loop Scav. IDI
Uniflow DI
Compound DI
2 stage rotary

-------
                                 6-30
Although the heat loss to coolant of the IDI diesel is some 25% higher than that
of the gasoline engine, cooling under full load conditions is not the critical
case in passenger applications and although some increase in radiator size
may be necessary for diesel cars, it certainly would not be a major increase
and it may be possible to use the same radiator (Peugeot use the same
radiator in the gasoline and diesel cars but the diesel engine is of much
lower output). It is therefore assumed that no significant front end changes
are needed for the application of a diesel engine to passenger cars and the
same acceptable rating  was  awarded to all candidates except the rotary
diesel where the extremely poor fuel utilisation will significantly increase heat
losses - 4 points were awarded in this case.
24. FIRE RISK

Scores

Engine                      Score
V-8 gasoline                  5
IL 6 gasoline                  5
NA V-8 IDI                   6
TC 6 IDI                      6
Comprex 6 IDI                6
TC 6 DI                       6
Comprex 6 DI                 6
Loop Scav. IDI                6
Uniflow DI                    6
Compound DI                 6
2 stage rotary                6

Apart from the fact that 0.2% of accidents involve fire, no quantitative data
relating to fire risk was uncovered during the literature  survey.

Causes of fires in automotive accidents are inevitably very complex but the
major  source of ignition is a high energy electrical discharge or a fire in the
wiring loom due to a dead short.  Thus the flash point of  the fuel has very little
influence on fire risk but its volatility does in that a highly volatile fuel  such
as gasoline can create a large  area of mixture with sufficient strength to support
combustion,  in this manner it can 'transport' itself effectively to the  ignition
source.  Under normal climatic conditions the volatility  of diesel  fuel is too
low to  allow the formation of a combustible mixture.

The lack of H.T. ignition systems in diesel vehicles must also lower their fire

-------
                                 6-31

risk (whether an accident occurs or not) .

For these reasons the diesel vehicle has been given one point more on the merit
rating scale than the gasoline vehicle.  It is assumed the fire risk of the gasoline
vehicle must be regarded as acceptable (i.e. 5 points).
25. IDLE NOISE

Scores
Engine
V-8 gasoline
IL 6 gasoline
NA V-8 IDI                   5
TC 6 IDI                      5
Comprex 6 IDI                5
TC 6 DI                       3
Comprex 6 DI                 3
Loop Scav. IDI                5
Uniflow DI                    4
Compound DI                  2
2 stage rotary                5

The idling noise of the gasoline engine is low in terms of both objectionability
and overall noise level and must be approaching the best practical level that
a power plant can achieve..  Although idling noise levels of the IDI diesel
engines are low, the characteristics  of the noise are subjectively annoying
and have an immediate impact on the casual observer in the street.  The overall
noise level from the IDI is sufficiently low to  make this just acceptable.  With
its greater instantaneous  pressure rise the DI engine sounds harsher than the
IDI and noise  levels are normally somewhat higher; Ricardo feel it would be
regarded as approaching bad or even unacceptable by the general public.  The
two stroke engines  were given the same score as the four stroke engines with
the same chamber apart from the uniflow engine which was given the benefit
of one extra point,  again mainly because of the reputation of GM's engines.

The compound engine with its  very low compression ratio will  have very long
delay periods at idling conditions with attendant harsh, loud combustion noises.
Mechanical gear noise in  the complex drive may also be a problem because of
torque recoil.

In the case of the two stage rotary engine it is difficult to predict idle noise but

-------
                                6-32
the shape should limit noise transmission and it was therefore awarded 5 points,
i.e. the same as the reciprocating IDI engines.
26. VIBRATION AND TORQUE RECOIL

Scores

Engine                      Score
V-8 gasoline                 8
IL 6 gasoline                 8
NA V-8 IDI                   6
TC 6 IDI                     5
Comprex 6 IDI               5
TC 6 DI                      5
Comprex 6 DI                5
Loop Scav. IDI               5
Uniflow DI                   5
Compound DI                 4
2 stage rotary                7

The aspect of vibration and torque recoil is undoubtedly a major contributor to
the feel or refinement of a vehicle.

Both the gasoline power plants are so smooth that the user is totally unaware
of any reciprocating motion and they must be regarded as the best practical
expression of power plants.

The unthrottled diesel engines with higher compression ratios are all inherently
rougher than the gasoline engine,  and the V-8 would be only good while the 6
cylinder engines would all tend to drop down to an ' acceptable' rating due to the
lower number of cylinders.

The compound engine is essentially a 4 cylinder engine from this aspect and
must be considered poor.

Of all the diesel power plants it was felt that the rotary engine would score best
on this aspect and although not quite as  unobtrusive as the gasoline engines, it
would certainly be good due to its inherent good "balance.

-------
       6-33
          TABLE 2



Final Ratings for Each Aspect
; ;


.
« ' ,,
C ! O
g -a j £•
•— t W ! 1— I
O ^0 ' f"*\
w en »_ ; H
03 Q
01 j; H
00 -^
Aspect ! ' °
^* ^O
1 Smoke 9 8
2 Particulates 7 7
3 Odour 7 7
4 NOx 5 5
l~~!
OO ! ^
1
o
vO
6 4.5
2
4
2
4
5 5
5 HC 555
5
6 CO 5565
1 7 SO2 77
8 HC reactivity 5 5
9 Evap. emissions 5 5
10 Misc. emissions 5 5
11 Noise (drive-by) 7 6.5
12 Package volume 7 8
4 4
7 7
7 7
5
6
6
5
6
5
13 Package Weight 6 7 j 5 5
14 Fuel economy 5 5.5 7.5 : 7
15 Fuel cost 5 5
16 Vehicle first cost 6 7
17 Maintenance cost 5 6
18 Startability 6 6.5
19 Hot drive ability 7 7
20 Cold driveability 5 6
21 Torque rise 7 7
22 Durability 6 5.5
23 Heat loss 5 5
24 Fire risk 5 5

25 Idling noise 8 8
26 Torque recoil 8 8
5
5
6
5
5
6
5 5
7 6
8 8
7
7
5
6

5
6
7
6
5
6

5
5

x !
QJ i
(H
£
0
i
._,
/->,
Q
i
">,
0
vO
.
, 5
2
4
5
5
5
4
7
7
5
5
4
5
7
5
5
6
5
7
8
7
6
5
6

5
5
! 0)


U
H
I— I
Q

Ix
O
SO
4.5
X
Q)
a
o
U
c
0)
CO
o
w
a
r\
,, ) T*S



o
d
'e c
Q 2 3 3
«
^
CJ
5
2 2
3 3
(0) (0)
(0) (0)
5 I 5
4 : 4
7
7
7 7
5 5
4 |3
5 ! 4
5 [5
8 8
5
5
! 5
5
6
8
7
6
5
6

3
j 5
5
5
5
5
7
8
7
6
5
6

3
5
.
SH
W
CM
4
2
3
u i &
fc ! E
1/3 i o
CM O
5 3
2 2
3 t 3
5 (O)j(O)
(0)
5
(o)j(o)
5 5
4 | 4
7
7
5
6
4
4
6.5
7
7
5
6
3
4
7
5 5
4.5 4.5
5
4
7
8
7
7
5
6

5
5
6
5
7
8
7
8
5
6

4
5
4
7
7
5
6
5
5
8
5
4
4
2
7
8
2
5
5
6

2
4


i
i
t'
cc
C


cv
CC
-*-'
CM
2
2
2
5
(0)
(0)
4
7
7
5
6
8
7
5
5
4
2
2
7 !
8.
7 i
2 :
4 !
6

5
7 '

-------
                             6-34
RESULTS

The product of the weighting and the rating for each performance aspect was
summed for each power plant and the results are shown in Table 3 below:-

                               TABLE 3
               Power Plant              Final Score
                                       (Rounded to nearest whole number)
1
2
3
4
5
6
7
8
9
10
11
V-8 gasoline
6 cyl. gasoline
V-8 EDI
6 cyl IDI T/C
6 cyl EDI Comprex
6 cyl. DIT/C
6 cyl. DI Comprex
2 str. loop scavenge
2 str. uniflow
Compound
2 stage rotary
608
620
587
556
554
(500)
(497)
(516)
(515)
(465)
(434)
In order to establish the validity of the 'committee method' of rating two re-
runs were done for the gasoline engine and the IDI diesels,when the final
runs for these power plants were found to be within - 2^% of the above figures
and the relative order of scores was not found to change. Because of this the
final scores from the first complete  run shown in Table 3 were taken as
representative of the relative merit of the various power plants.

The rating methodology  has the disadvantage that a rating of 0 in any aspect
can be hidden by good ratings for other aspects and the bracketed  figures
in Table 3 are for those power plants which scored a 0 (i.e. totally unaccept-
able) in one or more performance aspects.

The results are revealing  in that they immediately indicate that the gasoline
power plants are superior to the diesel power plants and that only the indirect
injection 4-stroke diesels are viable for the duty considered.  All the other
diesel power plants are unacceptable due to the inability to score better than
zero on one or more performance aspects (generally emissions).  With the
weighting adopted, the gasoline engine's  superiority for passenger car use
in America is demonstrated but the closeness of the final scores indicates that
only quite minor changes in the weightings would bring the scores equal.  It
is apparent that in an emissions and  fuel  conscious environment there are

-------
                              6-35
many applications which would change the individual weighting system
sufficiently to make the automotive diesel an attractive alternative to the
gasoline engine.

The smaller 6 cylinder gasoline engine has 12 more points than  the V-8
entirely on the strength of its better fuel economy and lower first cost.

The potentially viable IDI diesel power plants scored some 50 points or so
below the gasoline engines.  This is well outside the 15 point scatter found
in the re-runs but is sufficiently close to permit their consideration for
many applications.

It can be seen that all  the other diesel power plants were unacceptable,
mainly due to their poor HC and NOx characteristics.

If the total score for each contender is divided by  100, this averaged score
can be related to the original merit table.  It can immediately be seen that
no plant can be called  'good' although both gasoline engines score better
than 6,  i.e. better than 'acceptable' although not  quite 'good' .

The IDI diesel plants on this basis are still better  than 'acceptable1 although
an average score of 6  is not quite achieved.

Certain aspects beyond the scope of this study, such as depletion of world
stocks of precious metals due to the use of catalysts or the implications
of particulate legislation,  cannot be covered by a rating method such as
this but, where  possible, these other aspects should be considered before
any major  decisions on automotive power plants are made.
CONCLUSIONS

The rating methodology evolved has allowed different power plants to be
compared on a numerical basis.

Its application to the eleven power plants in this study has indicated that
gasoline power plants are some 10%'superior'  to the diesel power plants
for an average light duty application.

The viable diesel power  plants, four-stroke IDI's, are sufficiently close to
the gasoline engines to allow changes in the importance of some performance
aspects,  notably fuel economy, to bring them up to the same effective rating.

-------
                             6-36
None of the other diesel power plants were viable for the primary emissions
environment.

No diesel power plant would be viable for the secondary emissions environ-
ment without considerable advances in technology.

-------
                                        6-37
                                                                     FIG.6-1
                     LIGHT  DUTY  VEHICLE   POWER  PLANT
                               SURVEY   RESULTS
 1.                 SMOKE
 2.           PARTICULATES
 3.                 ODOUR
 k.                   N02
 5-       .             HC
 6.                    CO
 7-                   S02
 8.          HC  REACTIVITY
 9.  EVAPORATIVE EMISSIONS
10.        MISC. EMISSIONS
] 1 .       NOISE  (DRIVE BY)
12.         PACKAGE  VOLUME
13.         PACKAGE  WEIGHT
1/4.           FUEL ECONOMY
15.              FUEL COST
16.     VEHICLE FIRST COST
17.       MAINTENANCE COST
 18.           STARTABILITY
 19.       HOT DRIVEABILITY
20.       COLD DRIVEABILITY
2V.             TORQUE RISE
 22.             DURABILITY
 23.               HEAT LOSS
 2k.               FIRE RISK
 25.            IDLING NOISE
 26.           TORQUE RECOIL
                                            10               20
                                              °/o  WEIGHTING

-------
                                   7-1

                                SECTION 7
                            PROGRAMME PLANS

While this study has confirmed  that the diesel  engine is  a viable alternative
power plant  for  a passenger car and indeed  would have advantages in a
number of areas, notably in respect of fuel  consumption, it has brought out
other areas  where the diesel engine is less  attractive than the  gasoline
engine.   The report has also shown up areas where  the  technical data is
not as complete as  would have  been desirable  for  the purposes  of the  study.

It  would be  advantageous therefore  to  carry  out a  number of experimental
programmes  aimed  partly at filling these gaps in knowledge and partly at
overcoming  the areas of deficiency in  performance.

1.  Construction of V-8 naturally aspirated, indirect injection diesel  engine

While it  is believed that accurate estimates of fuel consumption and exhaust
emission levels have been made, these have involved predictions  with  vehicle
power:weight ratios much higher than  any for  which actual test data are avail-
able.

For a convincing demonstration that these performances  are possible  it is
desirable to  fit a suitable diesel engine  into a 1600  kg  (3500 Ib)  passenger
car and  obtain a full set of vehicle performance data.  This would also  pro-
vide a convenient demonstration vehicle.

While the time scale for the design and  construction of a new engine  is  likely
to be in the  region  of 30-36 months,  it  would  be possible to produce an
engine in a  shorter time by carrying out a conversion of an existing  gasoline
engine.   This would in  no way  affect the accuracy of the vehicle performance,
fuel consumption or emissions  measurements but could lead to  a necessity
for compromise in  so far as some of  the other performance aspects  are con-
concerned.

2.  Pressure Charging  Investigations

More data are necessary to refine the predictions  for pressure charged  engines.
With a turbocharger for example, it may be difficult to  drive a cycle  such as
the LA4  with a  large number of transients and still  have freedom  from exhaust
smoke, with resulting excessive exhaust  CO levels.

Theoretical simulation is very difficult in this  area  and  there would be much
to be gained by carrying out comparative test  bed and vehicle studies  using
a  turbocharger  and  as an alternative,  the Comprex.  It would of course be
necessary to use a  smaller  engine  and once again while it would be possible
to design and develop a new engine, it will  probably be preferable either  to

-------
                                    7-2
use an existing six cylinder  diesel engine such  as  the  Ford "York" or alter-
natively to carry out  a conversion of  an existing gasoline engine.

3.  Improved Fuel Injection  System

There are two problems in the area of  the fuel injection equipment.  The
first results from the high degree of  accuracy necessary for acceptable
performance and  low  emissions.   This high accuracy leads  to  high cost  and
hence to the cost of the fuel injection equipment accounting for some 50% of
the difference  in  cost  between  a  diesel  engine and  a gasoline engine.

The second arises from the  need,  for reasons  of low emission levels,  to
have an injection timing which varies with load and  speed.   This leads to
additional expense and to a considerable complication with most mechanical
injection systems.  Indeed it is this  requirement that  has  spurred  work with
piezo,  accumulator,  electro-magnetic and other  injection systems.   If it
could be satisfactorily carried out it.  might  prove advantageous to  vary the
pumping  rate with either  load or  speed, to  use pilot injection, or even  to
vary the pumping rate during each delivery stroke.

There is much to be said therefore for  pursuing work  in the areas of novel
and lower cost fuel injection equipment.

4.  Low  Compression  Diesel Engines

The interest in the low compression diesel engine  lies in the expectation
that it will give a lighter, more efficient engine.  If exhaust hydrocarbon
and odour levels  are  not  to  be unacceptable however,  it is  essential  to  have
a reliable starting and low load running,  ignition aid.   This might  take  the
form of a high energy spark, a heater  plug, an inlet air heater, exhaust
recirculation or indeed be of some other  form.

It  may or may not be possible to produce such a device but in view  of  the
possible  advantages it is  desirable that  work should  be carried out to investi-
gate the  feasibility of  the aid and, if  successful, to  investigate the actual
advantages of a low compression  diesel engine.

5.  Exhaust Hydrocarbon  Formation

While the chemical reactions leading  to  the  formation of NOX are,  at least
partially, understood  and it  should be possible to meet the  primary NOX
emissions  target  of 1.5   g/mile,  further reductions in NOX are made  diffi-
cult by resulting  rises  in hydrocarbon levels.   Indeed  it may  prove necessary
to use a catalyst to hold  hydrocarbon levels  down  low  enough  with  this primary
emissions  target.  With direct injection engines, it is  the  rising hydrocarbon
levels which make it  impractical  to attain 1.5  g/mile  NOX-

-------
                                    7-3
It is recommended therefore that a fundamental study should be carried out
into the mechanism  of exhaust  hydrocarbon formation.   Are the hydrocarbons
associated for example with particulate material  arising from combustion?
Do they result from flame quenching  close to the combustion chamber walls?
Do they arise from  slowburning and subsequent quenching  of part of the fuel
or alternatively  do the last portions of the fuel to be injected have excess-
ively large drops  due to  changes in the hydraulics of the  fuel injection system.

It is possible that some other mechanism may be involved but  if the mech-
anism is understood it could lead to a control of  hydrocarbon formation and
hence to the  development of lower emission diesel engines.

As hydrocarbon  levels are always  likely  to be a  problem  at the retarded
injection timings which are essential  for  very low NOX emissions,  it is
desirable to  carry out work aimed at the development of  a catalyst with a
low "light off temperature which would be effective at the low exhaust
temperatures which  occur at part load in a diesel engine.

6.   Modulation Systems  for the  Control of Exhaust Gas Recirculation

To  avoid excessive derating of naturally aspirated engines, it is necessary to
modulate exhaust gas recirculation  quantities  with load and it may  also  prove
desirable to modulate with speed.   It  may prove to be necessary to reduce
the recirculation at  peak torque conditions with a pressure charged engine.

Control  devices  for  such modulation in diesel engines are not,  so far as
Ricardo  know, available  at this time and  those being developed for gasoline
engines  may  suffer problems due to the particulate  material in diesel
exhaust.  It is suggested therefore that work be  initiated  to develop a
practical control system.

7.   Improved Starting  Aids

The delay required to bring heater  plugs  up to operating temperature is a
considerable  subjective problem.  The use of a programmed starting sequence
can help and  indeed such devices are  in  use, if not actual articles of com-
merce.   "Instant", or more realistically, fast heating plugs would be a con-
siderable advantage  however, and a general programme aimed  at quick  start-
ing under cold ambient conditions is desirable.

8.   Demonstration of Diesel Engines for  Specialised Applications

This study demonstrates  the clear  superiority of  the diesel  engine  for taxi
cab and light delivery  vehicles.   Demonstrations in these applications in
America in the  past have  failed due to the inadequate power/weight ratio of
the demonstration vehicle  or due to the use of unsuitable  vehicle/engine com-
binations .

-------
                                    7-4
It  is recommended therefore that following the construction or  conversion of
a suitable engine and an investigation of its performance in  a suitable  vehicle,
a programme should be instituted with a number of these vehicles  to demon-
strate  the advantages and to encourage the conversion of current and future
taxi  cabs and light delivery trucks to diesel  power.

9.  Particulate  Levels

The relatively high particulate  levels which occur in the exhaust from a diesel
engine even under clean exhaust conditions, could pose a long term problem.
It  would be desirable therefore to  carry out  a fundamental programme aimed
at finding the source of these  particulates  and hopefully therefore  at a method
of controlling their formation.

It  would also  be desirable to carry out  a programme aimed at the development
of a trap which  would  eliminate, or  at any rate  reduce,  smoke and particulates
from an exhaust which contains excessive levels.

Since neither  the 'spectra1 of  particle sizes from  a given power plant nor
the relative harm which can be caused by  particles of  a given  size and
nature is known, it is  apparent that  this lack of knowledge should  be remedied
before any legislation is finalised.  It is thus  suggested that work  be intensi-
fied  in  the field of particulate  emissions to ensure  that legislation  is relevant
to the  needs of the population.

-------
                          8-1

                      SECTION 8

             APPENDIX 1 - LIST OF KEYWORDS

 1.   SMOKE
 2.   ODOUR
 3.   GASEOUS EMISSIONS
 4.   PARTICULATES
 5.   NOISE
 6.   VOLUME
 7.   WEIGHT
 8.   FUEL ECONOMY
 9.   FUEL
10.   FIRST  COST
11.   MAINTENANCE
12.   STARTING
13.   HOT DRIVEABILITY
14.   COLD DRIVEABILITY
15.   TORQUE RISE
16.   DURABILITY
17.   COOLANT HEAT LOSSES
18.   FIRE RISK
19.   VIBRATION AND TORQUE RECOIL
20.   IDLE NOISE
21.   RELIABILITY
22.   ECONOMICS
23.   MANUFACTURE
24.   PERFORMANCE
25.   DESIGN
26.   ANCILLIARIES
27.   LUBRICATION

-------
                                    8-2
1.   SMOKE
     Measurement techniques




     Legislation




     Formation




     F.I.E. -  timing  -  injection rate - pilot injection - nozzle




     Combustion chamber  -  D.I. - I.D.I. - compression ratio




     E.G.R.




     Boosting - turbocharging  -  comprex - supercharging - ramming




     Diesel vs gasoline




     Levels vehicle




     Levels Test bed
     User experience




     Public opinion




     Endurance




     Light load - blue - white - aids




     Fuel - quality - dual - additives  -  alternate




     Exhaust system - catalyst - soot filter




     Ambient conditions




     D.I. vs I.D.I.
     Mathematical models
     Water injection

-------
                                    8-3
2.   ODOUR
      Measurement techniques - subjective - meter




      Legislation




      Formation




      F.I.E. -  timing  -  injection rate - nozzle




      Combustion chamber  -  D.I.  - I.D.I.




      E.G.R.
     Diesel vs gasoline




     Levels - vehicle




     Levels - test bed -  subjective - meter




     User experience




     Public opinion




     Endurance




     Fuel - quality - dual additives - alternate




     Light load aids




     Exhaust system - catalyst - soot filter




     Ambient conditions




     Mathematical models

-------
                                    8-4
3.   GASEOUS EMISSIONS




     Measurement techniques - heavy  duty  -  light duty - miscellaneous




     Legislation  - heavy duty - light duty




     Formation




     F.I.E. -  timing  - injection rate - nozzle - pilot  injection




     Combustion  chamber  -  D.I. - I.D.I. -  compression ratio




     Fuel - quality - dual - additives  - alternate




     Water  injection




     E.G.R.




     Boosting - comprex




     Diesel vs gasoline




     Levels -  vehicle




     Levels -  test bed




     Endurance




     Light load aids




     D.I. vs  I.D.I.
     Ambient conditions




     Catalyst




     Cost




     Mathematical models




     Trends




     Exhaust reactor

-------
                                    8-5
4.    PARTICULATES




      Measurement techniques




      Legislation




      Formation




      Combustion chamber - D.I. - I.D.I.




      Boosting




      Diesel  vs gasoline




      Levels  -  vehicle




      Levels  -  test bed




      Endurance




      Fuel -  additives




      D.I, vs I.D.I.




      Exhaust system - catalyst - soot filter




      Ambient  conditions




      Mathematical models




      Toxicity

-------
                                    8-6
5.   NOISE
     Measurement techniques




     Legislation




     D.I,  vs I.D.I.




     Exhaust
     Intake




     F.I.E. -  timing  - injection rate - pilot  injection - nozzle




     Combustion chamber  - D.I. - I.D.I. -  compression ratio




     Fuel - quality - dual - additives  - alternate




     Boosting




     Shields




     Enclosures
     Structure - isolation  - materials




     Diesel vs gasoline




     Levels - vehicle - interior - drive by - stationary




     Levels - test bed




     User experience




     Public opinion




     Air  cooled  vs water cooled
     Engine configuration




     Combustion




     Mechanical




     Endurance

-------
                                    8-7
6.   VOLUME




     Configuration




     Diesel ys gasoline




     Vehicle  installation..




     Air  cooled  vs water cooled




     Specific

-------
                                   8-8
7.   WEIGHT
     Materials




     Configuration




     Diesel vs gasoline - engine - vehicle




     D.I, vs  I.D.I.




     Bare engine




     Air cooled vs water cooled




     Specific

-------
                                   8-9
8.   FUEL ECONOMY




     D.I,  vs I.D.I.




     Fuel -  dual - additives - alternate  -  quality




     Boosting




     C onfig ur ation




     Diesel  vs gasoline




     Vehicle




     Test bed
     User experience




     Economics




     Endurance




     Light load aids




     Thermal efficiency

-------
                                   8-10
9.   FUEL




     Cost - tax. - prime




     Refinery  technique




     Quality




     Alternate




     Additives
     Diesel vs gasoline




     Trends




     Production - trends

-------
10.  FIRST COST
     F.I.E.




     Boosting




     Legislation - smoke -  emissions - odor - noise




     Diesel vs gasoline




     D.I, vs  I.D.I.




     Vehicle
     Engine - components




     User  experience




     Economics




     Air cooled vs water cooled




     Specific cost




     Tolerances

-------
                                   8-12
11.  MAINTENANCE




     Engine




     F.I.E.
     Ancilliaries




     Diesel vs gasoline




     Methods of  improving




     I.D.I, vs D.I.




     Vehicle
     User  experience




     Economics




     Air cooled vs water cooled




     Fuel - quality -  additives - alternate




     Lubrication

-------
                                   B-13







12.  STARTING




     Temperature




     F.I.E. - excess fuel - nozzles  - timing




     D.I,  vs  I.D.I.




     Fuel - quality  - alternate - additives - filter clogging




     Boosting




     Aids  - heater  plugs  -  manifold heaters - ether




     Diesel vs  gasoline




     User experience




     Smoke




     Compression ratio




     Sequence




     Battery /starter




     Cranking speed




     Valve timing




     Lubricating oil

-------
                                   8-14
13.  HOT DRIVEABIUTY




     Diesel vs gasoline




     D.I, vs I.D.I.




     Vehicle
     User  experience




     Light  load aids




     F.I.E.




     Response time

-------
                                     8--15
14.-  COED; EMVEABILITY




     Dieaeli vs gasoling-




     D.I. vs r.D.,L..




     Vehicle
     User experience




     Light Icrad aids




     Odor




     Fuel —  alternate. -  quality

-------
                                   8-16
15.  TORQUE RISE




     Boosting  - comprex




     F.I.E.




     Diesel vs gasoline




     I.D.I, vs D.I.




     Valve timing

-------
                                   8-17
16.  DURABILITY
     F.I.E.
     Emissions devices




     Boosting




     Diesel vs gasoline




     I.D.I, vs D.I.




     Vehicle
     Test  bed




     User experience




     Economics




     Air vs water cooling




     Combustion chamber




     Lubrication




     Exhaust  system




     Wear

-------
                                   8-18
17.  COOLANT HEAT LOSSES




     Boosting




     Diesel vs gasoline




     D.I. vs  I.D.I.
     Vehicle heaters
     Test bed
     User  experience




     Specific heat loss




     Air vs  water cooling




     Warm-up  time




     Thermal  loading




     Temperature control




     Oil cooler




     Piston cooling

-------
                                   S-19







18.  FIRE RISK




     Legislation - fuel tanks - general




     Diesel vs gasoline




     Vehicle




     User experience




     Statistics




     Exhaust  system - catalyst




     Fuel

-------
                                   8-20
19.  VIBRATION AND  TORQUE RECOIL




     Engine mountings




     Diesel vs gasoline




     Vehicle




     User  experience




     Configuration




     Engine installation




     Damping

-------
                                     8-21
20.  IDLE NOISE
     Vehicle - absolute - subjective




     F.I.E.  - pilot injection  - injection rate - nozzle  -  timing




     Diesel  vs gasoline




     D.I, vs I.D.I.




     User experience




     Public opinion




     Air vs  water cooling




     Configuration




     Fuel  -  quality




     Combustion




     Mathematical models




     Shields

-------
                                    8-22
21.  RELIABILITY
     F.I.E.
     Booster




     Diesel vs gasoline




     D.I, vs  I.D.I.




     Vehicle
     Test bed




     User experience




     Economics




     Air  vs water cooling




     Lubrication

-------
                                   8-23







22.  ECONOMICS




     F.I.E.
     Boosting




     Diesel vs gasoline




     D.I, vs  I.D.I.




     Vehicle
     User experience




     Specific  costs




     Legislation - smoke -  emissions - odor - noise

-------
                                   8-24
23.  MANUFACTURE




     F.I.E. - economics - technology




     Combustion system - economics - methods - tolerance




     Diesel vs gasoline - economics  -  technology




     D.I, vs  I.D.I.




     Air vs water  cooling




     Configuration




     Tolerance

-------
                                     «-25







24-   PEaFOBMAiNCE




      Vehicle - legislation - boostiiKj -  diesel vs  gasoline




              - D.I. vs  I.D.I- - 'durability  - torque rise




              - road test - F.I.E.




      Test bed - fuel additives  - F.I.E.  - combustion chamber




               - combustion - fuel quality - E.G.R.  - boosting




               - diesel vs gasoline - D.I.  vs  I.D.I.  -  specific




               - durability -  friction - volumetric efficiency




               - thermal efficiency - compression ratio - cylinder




               - pressure -  fuel alternate - dual  fuel - swirl




               - valve timing -  ambient conditions




               - component. temperatures




      Mathematical models

-------
                                    8-26
25.  DESIGN
     Configuration




     Legislation




     Combustion system




     Boosting




     Diesel vs gasoline




     Mechanical  - bearings + liners




     Tolerances




     Air cooled vs  water cooled

-------
                                    8-2:7
26.  AKCILMARIES




     Diesel vs  gasoline - cost




     Emission' Gorttroilers




     Volume
     Weight




     Servopamp

-------
                                    8-28
27.  LUBRICATION





     Water  injection




     E.G.R.





     Diesel vs gasoline




     D.I, vs  I.D.I.





     User experience




     Durability




     Oil change period




     Oil consumption




     Fuel - quality




     F.I.E.





     Cooling

-------
                                    9-1
                                  SECTION 9
                       APPENDIX 2 - LIST OF REFERENCES
As mentioned in section 4 the scope of the literature review was wide rang-




ing, covering internal Ricardo reports as well as freely available published




articles.









While many of the internal Ricardo reports  are the  result of Ricardo spon-




sored test programmes, others have  arisen from research  or  development




projects which have  been funded by individual  clients  of Ricardo.  These




reports are  regarded by Ricardo as confidential between themselves  and  their




client and because of this  the  report  titles  in  several of the references have




been modified to ensure this confidential nature.

-------
                                   9-2
1.   Diesel emissions  as predictors of observed Diesel  Odor.
     H.E.  Dietzmann et  al - SAE 720757

2.   Noise  of small indirect injection  Diesel engines.
     W.M.  Scott - SAE 730242

3.   Automotive Diesel engine  noise and its control.
     M.F. Russell - SAE 730243

4.   Design aspects of Low Noise Diesel engines.
     S.H.  Jenkins et al  - SAE 730246

5.   Diesel engine exhaust smoke - its  measurement, regulation and control.
     M. Vulliamy & J. Spiers  -  SAE 670090

6.   Diesel emissions  as related to engine variables  and fuel characteristics
     Marshall & Fleming - SAE 710836

7.   Design factors that  effect Diesel  emissions.
     R-C.  Bascom et al  - SAE 710484

8.   Pre-combustion chamber Diesel  engine emissions -  a  progress report.
     R.E. Bosecker & D.F. Webster -  SAE  710672

9.   Diesel exhaust -  a  European Viewpoint.
     B.V.  Millington & C.C.J. French - SAE 660549

10.  A  stratified  charge  multifuel military engine - a progress  report.
     E. Mitchell et al  -  SAE 720051

11.  Exhaust characteristics  of the automotive Diesel.
     R.C.  Schmidt et al  - SAE 660550

12.  Exhaust emission  control  in medium swirl  rate direct injection Diesel
     engines.
     Parker & Walker - SAE  720755

13.  Some effects of fuel injection system  parameters on Diesel exhaust
     emissions.
     R.J. Hames et al - SAE 710671

14.  The characterisation of  odor components in Diesel  exhaust  gas.
     R.S. Spindt et al  -  SAE 710605

15.  The development, of  the  small automotive Diesel in western Europe and
     its likely role in  the U.S.A.
     J.H.  Pitchford -  SAE 215B

-------
                                    9-3
16.  Reductions of emissions from Diesel engines.
     C.J. Walder - SAE 730214

17.  Measurement and evaluation  of Diesel smoke.
     R.C. Bascom et  al  -  SAE 730212

18.  Some problems encountered in the design and development of high speed
     Diesel engines.
     C.J. Walder - SAE 978A

19.  Diesel fuel properties and exhaust gas - distant relations?
     G.  McConnell  &  H.E. Howells  -  SAE 670091

20.  Daimler Benz passenger  car  Diesel engines  - highlights of  30 years of
     Development.
     E. Eisele - SAE  680089

21.  Combustion system parameters and their  effect  upon Diesel engine  exhaust
     emissions.
     Pischinger & Cartellieri - SAE 720756

22.  Towards higher speeds and outputs from  the small Diesel engine.
     D.  Broome - SAE  730149

23.  Diesel engine and highway truck  noise  reduction.
     R.M. Law - SAE 730240

24.  Cooperative  evaluation of techniques  for  measuring nitric oxide and carbon
     monoxide.
     J.M. Perez  et al - SAE  720104

25.  Turbocharged Diesel engine performance  at altitude.
     J.W. Dennis - SAE 710822

26.  Developing a new stratified charge combustion system with fuel injection
     for reducing emissions in small  farm and industrial  engines.
     M. Miyake - SAE  720196

27.  Combustion Characteristics of rotary engines.
     K.  Yamamoto  et al - SAE 720357

28.  A  new Diesel combustion chamber - the  variable  throat chamber.
     B. Brisson et al - SAE 730167

29.  A  preliminary model  for  the formation of nitric oxide in D.I. Diesel
     engines and  its application in parametric  studies.
     S.M. Shahed et al -  SAE 730083

-------
                                   9-4
30.  School fleet Dieselisation Program =
     R. Davis  - SAE 650711

31.  Delivering the mail with Diesels  - the  post office department  looks  at
     Diesel engines„
     G.C. Nield - SAE  650710

32.  An  example of development in automotive small  high  speed Diesel
     engines.
     S. Tanaka et  al - SAE  978C

33.  Pilot injection in Diesel engines.
     D.P.  17093 - 1973

34.  Review of Combustion Chamber Characteristics.
     D.P.  11353A  -  1968

35.  The automotive  Diesel engine in perspective.
     Donald -  1st Symposium  on low pollution power  systems development and
     AAPS coordination  meeting.  Ann  Arbor, Oct. 1973.
36.  Development of an automotive particulate sampling device  compatible
     with the CVS  system.
     Musser & Bernstein - Esso publication  -  1973

37.  Smoke Measurement  -  Instruments & comparison of  methods.
     Dodd & Spiers - I.Mech.E. proc. 1968-69, 183  pt 3E p!57.

38.  Influence  of fuel properties on Diesel exhaust emissions.
     Burt and Troth - I.Mech.E.  proc. 1968-69,  183  pt 3E p!71.

39.  Comparison of 2 litre D.I. and Comet  V automotive  engines.
     D.P.  12897 - 1970

40.  The Turbocharged Diesel as  a Road Transport Power  Unit.
     Holmer & Haggh - I.Mech.E.  1970

41.  The Breathing and Combustion Requirements of the Small High Speed
     Diesel Engine - D. Broome - I = Mech.E.  1966

42.  Where  Phillips stands on the Stirling engine.
     Automotive Engineering - July 1973 - p37

43.  The state of Development of the Air Cooled Diesel engine.
     Dr.  Ing.  M.  Weidenmuller  -  Ric. Trans. 330

44.  The Rotary engines of Yanmar outboard motor.
     Yamaoka  & Tado - SAE 710581

-------
                                   9-5
45.   Experiments with a single  cylinder engine to aid development of highly
      supercharged high speed Diesel engines.
      MIRA Translation 14/72

46.   Performance with economy - The  RamAire System.
      Percival & Ahrens - SAE 670109

47.   Low Compression Ratio IDI Diesels with Glowplug added ignition - a
      summary of relevant  experiments.
      D.P. 17555-  1974

48.   Dealing with the design problems  of  present day Diesel engines.
      Prof. Dr.  H. List - Ric. Trans. 321

49.   Results of development of the Daimler Benz  OM 360 Diesel engine.
      Dipl. Ing. Klinder & Ing. Kern -  Ric. Trans. 318

50.   Research on diesel  engine  exhaust pollution  control.
      Prof. R. Pischinger - ATZ - 1972

51.   Influence of exhaust gas  composition  by means of exhaust gas recir-
      culation  in  a pressure charged  swirl  chamber diesel engine.
      Dipl. Ing. Manfred Fortnagel - MTZ  - 1972

52.   Some problems of current  diesel  engine development and the  outlook
      for other prime movers.
      Urlaub - ATZ  - 1972

53.   Quieting  the diesel  with structural changes.
      Automobile Engineering  -  1972

54.   Some investigations on cold starting  phenomena  in  diesel engines.
      Austen & Lyn - Proc. I.Mech.E. No.  5 -  1959-60

55.   Contribution to  the  problem of starting and  operating diesel vehicles
      at low temperatures.
      Prof. M. Brunner & Dr.  H. Ruf - Proc. I.Mech.E.  No. 5 - 1959-60

56.   Improving  temperature control in  diesel engined vehicles.
      Bisiker & Benford - Proc.  I.Mech.E.   (AD) No.  8 - 1961-62

57.   Some experiences with a differentially supercharged diesel engine.
     Dawson et al - Proc. I.Mech.E. Vol.  178,  Part 2A No.  6  -  1963-64

58 .   Vehicle particulate emissions.
      Campbell & Dartnell - I.Mech.E.  Conference - Air Pollution Control
      in Transport Engines - 1971

-------
                                   9-6
59.  Experiments in the  control of diesel engine emissions.
     Torpey et al -  I.Mech.E. Conference - Air Pollution  Control in
     Transport Engines - 1971

60.  Metabolism and dietics of "Moteur  diesel compense B"  and its  result
     on air pollution,
     Brille  & Baguelin -  I.Mech.E. Conference - Air  Pollution Control in
     m        , TT*  •       * f\ *7 -1
     iranapoi'L c/uyines  — IV il

61.  Controlling  exhaust emissions from a diesel engine by  LPG dual fuelling,
     Lyon et  al - I.Mech.E. Conference - Air Pollution Control in
     Transport Engines - 1971

62.  Stratification  and air pollution,,
     Witzky - I.Mech.E.,  Conference  - Air Pollution Control  in Transport
     Engines  - 1971

63.  Diesel engine  exhaust  emissions and effect  of additives.
     Nesr et al  -  I.Mech.E.  Conference - Air  Pollution Control in
     Transport Engines - 1971

64.  The  mechanisms of soot release from  combustion of  hydrocarbon fuels
     with particular reference to  the diesel engine.
     Broome  and Khan -  I.Mech.E.  Conference - Air Pollution Control in
     Transport Engines - 1971

65.  Recent automotive air  pollution  control legislation in  the  U.S. and  a
     new  approach  to achieve  control:   Alternative engine systems.
     Brogan - I.Mech.E.  Conference  -  Air  Pollution Control in Transport
     Engines  - 1971

66.  Prediction of  soot and  nitric oxide  concentrations in diesel engine
     exhaust.
     Khan et al  -  I.Mech.E.  Conference - Air  Pollution Control in
     Transport Engines - 1971

67.  Factors  affecting emissions of smoke and gaseous  pollutants  from
     direct injection engines.
     Khan & Wang  - I.Mech.E. Conference -  Air Pollution Control in
     Transport Engines - 1971

68.  Exhaust  emission control system for the rotary engine.
     Muroki - I.Mech.E. Conference -  Air Pollution Control in Transport
     Engines  - 1971

69.  Symposium  on  Multi-fuel engines.
     Gas  & Oil Power 1959

-------
                                    9-7
70.   Compact long  life diesel engine.
      Timoney - I.Mech.E. Conference - Transport Engines of Exceptionally
      High Specific Output - 1968

71.   The differential compound engine:   Interim  test results and assessment
      of  future development.
      Wallace et al - I.Mech.E. Conference - Transport Engines of Except-
      ionally High Specific Output  -  1968

72.   Curtis-Wright's  rotating combustion engine  - compact,  lightweight
      power.
      Jones - I.Mech.E.  Conference - Transport  Engines  of  Exceptionally
      High Specific Output - 1968

73.   Small high  speed diesel  engines.
      Westwell &  Blum  -  I.Mech.E. Conference -  Mechanical  design of diesel
      engines - 1967

74.   Effects of multiple  introduction of  fuel on performance of a compression
      ignition engine.
      C.P. Guptor et al - SAE 929A

75.   The nature  and cause of diesel emissions.
      B.W. Millington - I.Mech.E. Conference -  Motor vehicle air pollution
      control  - 1968

76.   Some  experiments on the mode of action of a diesel smoke suppressant
      additive.
      B.E. Knight &  C.H.T.  Wang  - I.Mech.E. Conference -  Motor vehicle
      air  pollution control - 1968

77.   Natural gas  engine for buses meets 1975 emission limits.
      Automotive  Engineer - June  1972

78.   Mid-Range Diesels  Mean Savings for Delivery Fleet.
      Diesel  and  Gas Turbine - July 1973

79.   Hybrid Diesel  -  Electric bus may  help cities  to lick noise-air-pollution
      problems.
      Automotive  Engineering - August 1970 p42.

80.   Influence of fuel properties and influence of anti-smoke additives  on
      Diesel  exhaust smoke.
      McConnell et al  - SAE Journal,  March '68  p36.

81.   Novel  supercharger retains smaller engine  economy while providing
      larger  engine performance. SAE  Jni.  March '68,  iro 62-65.

-------
                                  9-8
82.  Alternative non polluting power sources.
     SAE  Journal, Dec. '68 p40.

83.  Noise in engineering  and transportation and its effect on the Community.
     T. Priede - SAE  710061

84.  The  use of specially designed covers and  shields to reduce  Diesel engine
     noise.
     G. Thien - SAE 730244

85.  Diesel exhaust odor - its evaluation and relation to exhaust gas com-
     position.
     F.G. Rounds  & H.W.  Pearsall - SAE 863 - 1956

86.  A  sampling  smokemeter for automotive Diesel engine testing.
     E.G. Searle  - SAE 436C -  1961

87.  CRC  Investigation of Diesel smoke measurement.
     J.B. Durant  - SAE 801A -  1964

88.  The  measurement and Control  of Diesel exhaust smoke emission.
     J.D. Savage  - SAE 440B -  1962

89.  The  effect of the Vigom Process on the combustion in Diesel engines.
     SAE  929B

90.  1971 European Diesel  engine performance survey 10-500 bhp.
     D.P. 13718

91.  Nissan Emissions Results
     Ex E.P.A.  -  1973

92.  1DI Diesel  Engine - Performance development and piston cooling tests.
     D.P. 13679  - 1971

93.  6 cylinder  IDI Diesel  Engine,  Progress Report No.  1,  Jan-March 1971.
     D.P. 13677  - 1971

94.  For Passenger Cars Diesels Must Be Better.
     C.J. Walder  - SAE Jnl. Nov.  '65,  Vol. 73, No.  11, p78.

95.  A  European  contribution to  lower  vehicle exhaust  emissions.
     (Eindhoven Conf.  1971) - D.P. 13538

96.  Work carried out on  IDI engine, August  '69  - November  '70.
     D.P. 13277

-------
                                   9-9
97.   EDI engine - effect of valve timing on performance.
      D.P.  12921 - 1970
                                                                    3
98.   Note  on the effect of exhaust  manifold arrangements on a  138  in
      Comet  V engine.
      D.P.  12812 - 1970

99.   NOX  emissions  from single cyl.  engine in D.I. and Comet V builds.
      D.P.  12429 - 1970

100.  Design  Developments in European Automotive Diesel engines.
      Ricardo & Pitchford -  SAE  Journal  (trans) Vol. 41-3 p405  -  1937

101.  Exhaust Emissions Test  Program  on automotive IDI  diesel engine and
      dies el car.
      D.P.  14442 - 1971

102.  The Diesel is Friendly to its Environment.
      Garthe  - Deutz Publicity -  1971

103.  Reducing  Hydrocarbons and Odor in Diesel Exhaust  by  Fuel Injector
      Design.
      Ford  et al - SAE  700734

104.  Diesel  Fuel Specification and Smoke Suppressant Additive Evaluations.
      J.G.  Brandes - SAE 700522

105.  4 cylinder Comet Mk  V Engine - Progress Report no.  18.
      D.P.  4680 - 1958

106.  6 cylinder 5.5 litre, Comet V,  Progress Note No.  3.   CARB Emissions
      Tests.
      D.P.  16524 - 1973

107.  Performance of a Catalyst  Box on a Direct Injection Engine.
      D.P.  15867 - 1972

108.  Effect of  Inlet Air Humidity and Temperature on Diesel Exhaust Emissions,
      S.R.  Krause et  al - SAE 730213

109.  Surveying Tests of Diesel  Smoke Suppression with fuel additives.
      Takeshi Saito  -  SAE 730170

110.  Factors Affecting  Smoke and Gaseous Emissions from Direct Injection
      Engines and a Method  of Calculation.
      I.M.  Khan et al - SAE  730169

-------
                                  9-10
111. Diesel Engine Noise Reduction  by Combustion  and Structural  Modifi-
     cations .
     D.D. Tiede & D.F. Kabele  - SAE 730245

112. The Light Duty Diesel Engine.
     D.P.  16494 - 1973

113. 6 cylinder 5.5 litre.  Comet V.  Progress  Note No-  4,  CARR Emissions
     Tests.
     D.P.  16660 - 1973

114. 6 cylinder 5.5 litre,  Comet V,  Progress  Note No.  5,  GARB Emissions
     Tests.
     D.P.  16736 - 1973

115. The Diesel
     R. Wakefield - Road & Track - Sept.  1973

116. 6 cylinder 5.5 litre,  Comet V,  Progress  Note No.  6,  CARB Emissions
     Measurement.
     D.P.  16878 - 1973

117. Noise Test on IDI engined vehicle.
     D.P.  16947 - 1973

118. Engine and vehicle emissions reduction program.
     Interim Report No. 1.
     D.P.  17095 - 1973

119. NC>2 in Diesels  - Ricardo Research.
     D.P.  17138 - 1973

120. 6 cylinder IDI Diesel.   Progress Note No.  4.
     D.P.  14208 - 1971

121. 6 cylinder IDI Diesel.   Progress Note No.  3.
     D.P.  14068 - 1971

122. Investigation into  the Compression  Ratio limits of a Typical  Comet III
     type production  engine.
     D.P.  4494 - 1957

123. Note on  the Effect of Nozzle Heat Shields on  a typical Small Comet  V
     engine.
     D.P.  4417 - 1957

124. Diesel car - Ricardo Report - Jan.  1956

-------
                                   9-11
125. Diesel car fuel  consumption with and without  fluid drive.
     Ricardo GO.3712 - 1956

126. Ricardo Memo - Diesel  car - Nov.  1956

127. Comet V  automotive  diesel.  Progress Report No. 5.
     D.P.  9715 -  1966

128. Note on the Mechanical  behaviour  of the IDI diesel  engine during test
     running.
     D.P.  3926 -  1956

129. Comet Mk V Diesel Engine - 6 cylinder.
     D.P.  4566 -  1958

130. Progress  Report on diesel engine - 4 cylinder.   Ricardo Comet  Mk V
     Combustion System.
     D.P.  3963 -  1956

131. Influence  of Cetane Number on Compression Ignition Engine Performance.
     D.P.  13740 -  1971

132. Diesel Car  - Fuel & Oil Consumptions - Internal Ricardo  memo 1.10.54

133. Diesel Car.
     Ricardo GO.2930 - 1954

134. Diesel Car.
     Ricardo GO.3279 - 1954

135. 4 cylinder,  Comet Mk V.  Report on an investigation into Factors
     Affecting  Auxiliary Hole Blocking of the Pintaux nozzle.
     D.P.  4673 -  1958

136. Diesel fuel rating at  high speed - Report on preliminary tests on  the
     El2 H.S.  unit to observe ignition delay.
     D.P.  4547 -  1958

137. 4 cylinder,  Comet Mk V Progress  Report No.  2.
     D.P.  4443 -  1957

138. 4 cylinder,  Comet Mk V Engines - Progress Report No. 12.
     D.P.  4550 -  1958

139. 4 cylinder,  Comet Mk V Engines - Progress Report No. 13.
     D.P.  4573 -  1958

-------
                                  9-12
140. 4 cylinder,  Comet  Mk V Engines  -  Progress Report No.  15.
     D.P.  4617 - 1958

141. Induction  Ram.
     D.  Broome  - Automobile Engineer - April,  May,  June 1969.

142. Report on the Conversion of a Ford Zephyr car to diesel.
     D.F.  12093 - 1969
143. V8 IDI Diesel engine.  Progress  Note No. 10.
     D.P.  5761 - 1960

144. Further supercharging tests on a 6 cylinder  Comet V engine.
     D.P.  5379 - 1960

145. 4  cylinder automotive Comet Mk  V Engine - Progress Note No. 1.
     D.P.  5084 - 1959

146. 4  cylinder automotive Comet Mk  V Engine - Progress Note No. 2.
     D.P.  5137 - 1959

147. 4  cylinder automotive Comet Mk  V Engine - Progress Note No. 3,
     D.P.  5294 - 1959

148. 4  cylinder automotive Comet Mk  V Engine - Progress Note No. 4,
     D.P.  5353 - 1960

149. 4  cylinder automotive Comet Mk  V Engine - Progress Note No. 6,
     D.P.  5574 - 1960
150. 4  cylinder automotive Comet Mk  V Engine - Progress Note No. 9.
     D.P.  5678 - 1960

151. Diesel Car  - Conversion to Comet V.
     D.P.  3405 - 1955

152. Diesel engine - Modifications  to  Combustion Chamber.
     D.P.  3394 - 1955

153. Report on work  carried  out on diesel car.
     D.P.  3672 - 1955

154. The Ricardo  Comet V Combustion System.
     D.P.  5834 - 1960

155. Notes on  the Nozzle  life of the diesel engined vehicle.
     D.P.  5545 - 1960

-------
                                   9-13
156. Diesel engined vehicle - (Effect of mods to combustion chamber).
     D.P.  4881 - 1959

157. Further Notes Regarding Nozzle Life on the diesel engined vehicle.
     D.P.  5754 - 1960

158. V8  IDI diesel engine - Progress Note  No. 2.
     D.P.  5020 - 1959

159. V8  diesel engine - Progress  Note  No. 4.
     D.P.  5158 - 1959

160. 4-J litre Comet Mk V  - Progress Report No.  1.
     D.P.  5169 - 1959

161. V8  IDI diesel engine - Progress Note  No. 6.
     D.P.  5213 - 1959

162. 4-J litre Comet Mk V  six cylinder engine - Progress  Report No. 2.
     D.P.  5260 - 1959

163. V8  IDI diesel engine - Progress Note  No. 7.
     D.P.  5323 - 1959

164. Quiet Idling Investigations - Assessment  of Diesel Car Fitted  with
     Quiet Idling Device.
     D.P.  11615 - 1969

165. A Technical Appraisal  of the Clarke Rotary Piston Machine as a Basis
     of an Internal Combustion Engine. - D.P.  16199  - 1973

166. A new class of Rotary Piston Machine suitable for Compressors, pumps
     and internal combustion engines.
     Clarke et al - SAE  Proceedings Vol. 186 - 62/72.

167. 6 cylinder D.I. Engine Performance Calibration.
     D.P.  15811 - 1972

168. Results  from a short  series of CARB Emission Tests  run  on a Comet
     V Diesel  Engine.
     D.P.  16815 - 1973

169. A Study of  the Exhaust Emissions  Characteristics of a Turbocharged
     Direct Injection Engine.
     D.P.  16272 - 1973

-------
                                   9-14
170. A Study of the Exhaust  Emission Characteristics of a Naturally
     Aspirated Direct Injection Engine.
     D.P.  16470 -  1973

171. A Summary of a Research Project - Diesel Emissions.   S.N.  16374.

172. Unclassified data sheet listing  Comet  engines -  1973.

173. Meeting between  a client and Ricardo.
     D.P.  17358 -  1973

174. A review  of low  noise Diesel engine design at ISVR.
     Grover &  Lalor - Jnl. Sound &  Vib. No. 3, 1973 p403.

175. A review  of vehicle  noise studies - with a reference to some recent
     research on petrol engine noise.
     J.A.  Raff & R.D.H.  Perry - Jnl. Sound & Vib. No.  3  1973 p433.

176. Comprex Supercharging - a progress  report - Presentation to E.P.A.
     by A . Mayer, Brown Boveri.

177. Comprex Supercharger  for Passenger  Car Diesel engines.
     Prof.  E.  Eisele - D.Benz report to  E.P.A. - Sept. '73

178. Notes on a visit  to Brown Boveri, re.  Comprex, Dec.  1972.
     E.P.A. note by C.F. Bachle

179. Characterization  of particulates and  other non-regulated emissions from
     Mobile sources and  effects of exhaust  emissions control  devices on
     these emissions.  Report for E.P.A, APTD 1567, 29 Oct. '73.  Pages
     160-163 -  Mercedes 220D vehicle.

180. A Report on the  Emission Performance of the Ford Stratified Charge
     Engine Using the 1975 Test  Procedure.
     E.P.A. Report  72-2   -   August 1971

181. An evaluation  of  three Honda compound Vortex controlled combustion
     (CVCC) Powered vehicles.
     E.P.A. Report  73-11   - Dec.   1972

182. Exhaust  Emissions Analysis of  Two Wankel  Powered Cars Furbished by
     the  U.S.A.T.A.C.
     E.P.A. Report  73-10  -  Sept.  1972

183. Evaluation  of the Texaco Stratified Charge  (TCCS) M151 Army Vehicle.
     E.P.A. Report  73-27   -  June  1973

-------
                                  9-15
184. Vehicle Test Report No. 3 - Peugeot diesel.
     E.P.A.  Report  - 16th  July  '73

185. Vehicle Test Report No. 2 - Opel  diesel.
     E.P.A.  Report  - 16th  July  '73

186. Vehicle Test Report No. 8 -  '72 EPA Ford.
     E.P.A.  Report  - 7th Sept.  '73

187. Vehicle Test Report No. 7 - EPA  Ford.
     E.P.A.  Report  - 7th Sept.  '73

188. Final Report: Exhaust  Emissions from  a  Mercedes-Benz Diesel Sedan.
     E.P.A.  Report  73-6 ~  July  1972

189. Emissions  from a pick-up truck retrofitted with  a  Nissan Diesel Engine.
     July 1973  - From E.P.A. 29/10/73

190. EPA  diesel study status -First symposium on low pollution power
     systems development.
     Oct.  19th  1973  - Presented by  J.J.  McFadden -  from  EPA 29/10/73

191. Exhaust Emissions  from three diesel-powered passenger  cars.
     March 1973 - from EPA 29/10/73

192. M.W. Kellog Study, Ann Arbor, Michigan 48105.
     Letter from J.  DeKany  to E. Stork, MSAPC - from EPA 21/9/73

193. Letter referring to  Meeting with Mobil on Impact of Diesel on the oil
     industry.
     Letter dated  21.8.73 from T. Austin of EPA in reply to  AFAE - from
     EPA  21/9/73

194. Task 17 final report.   Potential increased production of  Automotive
     diesel fuels.  Submitted to EPA office of Air Programs Division of
     Control Systems Contract No. CPA 70-68  - from EPA  21/9/73.
     Submitted  by M.W.  Kellog Company,  Texas, Dec.  1st  1972

195. Notes on  a visit of representatives to Bridge Works on 26th Sept.
     1962 -  by C.J. Walder  - D.P.  6817

196. Notes on  the performance of wide cut gas  oil in a light duty IDI
     diesel vehicle.
     D.P. 6820 - 3rd Oct.  1962

197. 6 cylinder 2/4 litre - Comet Progress Report No. 1.  - Oct.  10th 1962
     D.P. 6839

-------
                                   9-16
198. Road fuel consumption for the petrol and  diesel versions of the Austin
     A60 and Morris Oxford  saloon cars.
     D.P.  7529 - 27th Nov.  1963

199. Notes on the 6-cylinder 6 litre engine.
     D.P.  6806 - 26th Sept.  1962

200. Ricardo High Speed  Engine Pump Injector  Tests.  Interim Report.
     D.P.  15200 - 1972

201. Report on a program of noise measurements  carried  out on  an IDI
     Diesel Engine.
     D.P.  15247 - 1972

202. List of  Ricardo Comet V Diesel Engines currently in production.
     D.P.  14378 - 1971

203. Diesel Engine Exhaust Emissions.  Report No. 5.
     D.P.  14168 - 1971

204. Client/Ricardo Technical Meeting.
     D.P.  14278 - 1971

205. 3 cylinder Comet V.
     D.P.  4381 - 1957

206. 4 cylinder Comet Mk V Engines.  Report No.  17.
     GO File 4661 -1958

207. 6 cylinder 4^ litre  engine fitted with Ricardo Comet V  cylinder head.
     D.P.  5652 - 1960

208. An  analysis of some engine test results on an IDI diesel engine.
     GO File 4725 -  1958

209. 4 cylinder Comet Mk V Engine.
     GO File 4776 -  1958

210. 4 cylinder Comet Mk V Diesel Engines. Report No.  18.
     GO File 4817 -  1958

211. Supercharging tests  on a Comet V Engine.
     GO File 5233 -  1959

212. Report on Diesel Engine 6 cylinder.  - Ricardo Comet V head.
     GO File 458 - D.P. 4458 - 1958

-------
                                   9-17
213. Volumetric Efficiency of Comet Engines.
     D.P. 8066 - 1964

214. Emission tests on  a 2.25 litre Diesel Engined vehicle.
     D.P. 16800 - 1973

215. Road test on a diesel car. Unclassified document.

216. Noise tests on a diesel vehicle.   4 cylinder  2  litre Comet Mk  V engine.
     GO  File 4933 - 1959

217. 4  cylinder Comet  Mk V  diesel engine.  Report No.  1.
     GO  File 4409 - 1957

218. Diesel  engine exhaust emissions.   Report No.  6.
     D.P. 14443 - 1971

219. Estimated  friction  losses  for 6 cylinder Comet Mk V and  D.I.  engines.
     R  &  Co. Drg. No. D18662 -  1966

220. Hybrid heat  engines/electric  systems.   A summary of  a feasibility
     study carried out.
     D.P. 14670 - 1972

221. Interim  report of noise measurements  of a BLMC  2.521 engine.
     D.P. 14767 - 1972

222. Ford Zephyr car fitted with a 4 cylinder IDI  diesel  engine.
     1975 CVS  Emission Tests.
     D.P. 14913 - 1972

223. Some notes on the selection of cylinder size  and arrangement for small
     high speed automotive  engines  using  the  Ricardo Comet  Mk V combus-
     tion  system.
     D.P. 15108 - 1972

224. Timed inlet and exhaust  systems for a 6 cylinder  prechamber diesel
     engine.
     D.P. 15130 - 1972

225. The  prospects of the diesel engine to satisfy  the  U.S. pollution regul-
     ations .
     D.P. 15448 - 1972 and R & Co. Memo 1973

226. 6  cylinder engined diesel vehicle.
     D.P. 15537 - 1972

-------
                                    9-18
227. 4 cylinder 2 litre Comet V.   Performance  calibration in modified build.
     D.P.  16210 - 1973

228. Future Power Units  for  Automobiles and Commercial Vehicles.
     D.P.  16256 - 1973

229. Opel  Rekord Diesel  2100D.
     Described briefly in ATZ  Vol.  75,  No.  1 - 1973 p27.

230. Preliminary  study of a low emissions  diesel engine  for  use in passenger
     cars.
     D.P.  16424 - 1973

231. Summary of Ricardo results showing effect of  fuel quality on diesel
     engine emissions.
     D.P.  16484 - 1973

232. The dynamic injection timing requirements  of  small  high speed diesel
     engines using Ricardo Comet combustion chambers.
     D.P.  4143 - 1957

233. The unloading mechanism  in distributor  pumps.   Diesel engine with 3 mm
     fuel  pipes.
     D.P.  4170 - 1957

234. Further thoughts on the mechanism of unloading in  distributor pumps.
     D.P.  4145 - 1957

235. Comet V  fuel injection pump tests.
     D.P.  3910 - 1956

236. Comet V  fuel injection pump tests.
     D.P.  3990 - 1956

237. Comet V  fuel injection pump tests.
     D.P.  3995 - 1956

238. Comet V fuel injection pump tests.
     D.P.  3550 - 1955

239- Test to establish the acceptable range of injection timings  on the  Comet
     III diesel engine when fitted with a 6  mm. fuel pump.
     D.P.  4255 - 1957

240. Report on the work carried out on  a high  ratio Comet  Mk  V engine bet-
     ween April - November 1958.
     D.P.  4862 - 1958

-------
                                    9-19
241. Comet  Mk III diesel engine - Comparison  of exhaust  smoke  rating with
     high and  low compression ratio.
     D.P. 4899 -  1959

242. Comparison  tests on 2 /4  and 2 litre engines using fuels  with and
     without the addition of ^%  by volume isopropyl nitrate.
     D.P. 4221 -  1957

243. Ricardo letter to Client -  28th  Jan.  1966

244. IDI Diesel engine.
     GO File 3971 - 1956

245. Modifications carried out on a  diesel car.
     GO File 4508 - 1958

246. Report  on Comet V Diesel Engine 6 cylinders.
     GO File 4718 - 1958

247. 4  cylinder Comet Mk V  Engine. Report No. 22.
     GO File 5309 - 1959

248. Note on high speed running of an IDI diesel engine.
     D.P. 5464 -  1960

249. Notes  on condition of nozzles in a diesel car.
     GO File 5469 - 1960

250. Comments on indicator diagrams and peak  pressure  measurements
     obtained on  an  IDI diesel engine.
     GO File 5612

251. Diesel  Estate Car.
     GO File 5297 - 1959   .

252. 3  cylinder Comet V  engine cold starting tests.
     GO File 4586 - 1958

253. Diesel  Estate Car.
     GO File 5494 - 1960

254. 4  cyl.  IDI diesel engine.
     D.P. 5785 -  1960

255. The  influence of fuel composition on emissions of  carbon  monoxide
     and  oxides of nitrogen.
     Carr et al - Univ. of California -  SAE 700470

-------
                                  9-20
256. Investigation of Combustion Phenomena  in a  Swirl  Chamber Compression
     Ignition Engine using Schlieren Techniques.,
     A<,B.  Allen  and A. Khin  - Univ.  of Toronto  -  SAE 700500

257o The FL413 - A New Series of Deutz  Air-Cooled V  type Diesel Engines.
     Hans-Ulrich Houe - Deutz - SAE 700028

258. Diesel emissions  as  related to engine variables  and fuel  characteristics.
     Marshall and Fleming -  U.S. Bureau of Mines - SAE 710836

259 o Automotive  Particulate Emissions  and  their Control.
     Habibi - Du Pont  de  Nemours  - SAE  710638

260o Emission Characteristics of  Natural  Gas as  an Automotive  Fuel.
     Fleming and Allsup  - Bureau of  Mines  - SAE 710833

261. The variable - Displacement Engine:   An Advanced concept powerplarit.
     Welsh and Riley - Thermo Mechanical Systems - SAE 710830

262. The effect of exhaust system geometry  on  exhaust  dilution and  Odor
     intensity.
     Colucci and Barnes  - G.M.  Corp. -  SAE 710219

263. Cooperative evaluation of techniques  for measuring hydrocarbons in
     diesel exhaust.
     (A  CRC report) Wagner  and Johnson  -  SAE  710218

264. Effectiveness of exhaust  gas recirculation with extended use.
     Musser et al - SAE  710013

265o Influence of Operating  Cycle on  Noise of Diesel Engines.
     Anderton &  Baker - ISVR - SAE  730241

266. Reducing noise from  heavy diesel trucks by engine compartment
     shielding.
     Ronnhult -• Saab Scania - SAE  730682

267. Diesel engine noise reduction  hardware for  vehicle noise control.
     Jenkins & Kuehner - Cummins  Engine Co. - SAE 730681

268. Photochemical Reactivity  of Diesel Exhaust.
     Bureau  of Mines ~ RI 7514 - 1971

269 o Diesel emissions  reinventoried.
     Bureau  cf Mines - RI 7530 •- 1971

270. Merits  of the Ricardo combustion systems.
     Letter to  Editor - Gas and Oil  Power - 1953

-------
                                    9-21
271. The small high speed diesel  engine.
     B.V. Millington - Proc.  I.Mech.E. Vol.  18 pt  IV - 1965

272. Optimisation of diesel engine parameters at high specific output.
     D. Broome - UDK 629.  1-445:  621.436.003.1 -  1973  (Conference -
     Belgrade  Yugoslavia)

273. Reducing  exhaust emissions from diesels.
     Automobile Engineer - 1971

274. Looking in on diesel combustion.
     W.M.  Scott - SAE  690002

275. Recent developments in diesel engine research  at the Ricardo labora-
     tories .
     W.M.  Scott - Entropie No. 48 - 1972

276. Some problems encountered  in the  design and development  of high speed
     diesel engines - C.J. Walder - SAE  978A - repeat of  ref.  18

277. What problems still restrain the small automotive diesel engine.
     Pitch ford et al  - FISITA  Conference  1964

278. Some more light on diesel combustion.
     Alcock & Scott - Proc. I.Mech.E.  1962-63  No. 5

279. Development of a High Speed Four  Cylinder Diesel Engine under con-
     sideration of the existing machine  tools equipment  for the production
     of the gasoline  engines.
     H. Weitzel  - I.Mech.E.  Conf.  Publication 19 - 1973

280. High Speed Diesel  Engines.
     H.R. Ricardo - Institution  of Automobile Engineers - 1926-30

281. Diesel engines.
     H.R. Ricardo - Royal Society of Arts - 1931

282. High Speed diesel engines  applied to Motor Vehicles.
     R.L. Stafford - Newark Engineering Society - 1932

283. High Supercharging of a  C.I. Engine.
     C.W.R. Smith - Engineer - 1938

284. Looking at diesel combustion.
     Hempson & Scott - New  Scientist - Vol.  6  pp 1134-37.

-------
                                   9-22
285. Diesel engines for road transport.
     Sir H.R. Ricardo  -  Engineering - 1948

286. Ricardo Combustion Systems for High Speed Compression Ignition
     Engines.   Unclassified publication - 1947.

287. Papers  on  small high speed diesel engines.
     Proc. I.Mech.E.  1954-55  - No.  8 - pp 209-246

288. Road Test Reports -  Peugeot 204, Mercedes  220D, 240D & 230/4,
     Opel Rekord,  and Peugeot 504.  Diesel & Gasoline Comparisons
     (French) L1 Action - Feb. 1974

289. Some problems arising from a wider use of the small  diesel engine.
     J.H.  Pitchford - The Oil Engine and Gas  Turbine - 1953

290. Future of the  high speed  reciprocating internal  combustion engine.
     J.H.  Pitchford -  CME -  1960

291. The high speed diesel engine.
     H.R.  Ricardo  - Diesel engine  users association  -  1930

292. Diesel smoke suppression by fuel additive treatment.
     C.O. Miller - SAE 670093

293. Smoke reduction in diese] engines.
     A.W. Carey-  SAE  670224

294. Diesel engine exhaust smoke.   The influence of  Fuel Properties  and
     the effects of using  Barium-containing fuel  additive.
     D.W. Golothan -  SAE 670092

295. The differential compound engine.
     F.J. Wallace - SAE  670110

296. Factors influencing  diesel emissions.
     Marshall and  Hurn  - SAE 680528

297. Relation of lean combustion  limits in diesel engines to exhaust odor
     intensity.
     Barnes  - SAE 680445

298. Effect of design revisions on two stroke cycle diesel engine exhaust.
     D.F.  Merrion  - SAE  680422

299. A  flame ionisation technique for  measuring  total hydrocarbons in diesel
     exhaust.
     Johnson et al - SAE 680419

-------
                                    9-23
300. Some Notes on Diesel  Cold Starting Aids.
     D.P. 17750 -  1974

301. Relation between noise and basic structural vibration of  diesel engines.
     T.  Priede  et al - SAE  690450

302. Does turbocharging  increase diesel engine noise? -  observations on the
     generation, emission and reduction of diesel  engine  noise.
     R.  Wolfgang Hempel. - SAE 680406

303. The noise  problem of air cooled diesel engines - measures towards
     its  reduction with general observations and specific  results.
     O.  Cordier& G.  Reyl - SAE 680405

304. Unaided starting of  diesel engines.
     T.W. Biddulp & W.T. Lyn -  SAE 680103

305. Development of Mitsubishi  DC2 Series V-Type Diesel Engine.
     Kenj  Okamura & Kiji  Yamanda - SAE  690745

306. Measurement  of Automotive Passby Noise.
     Ralph K.  Hillquist & Richard A. Battis  -  SAE 725275

307. Motor Vehicle Noise identification and  analysis of situations contri-
     buting to  annoyance.
     William J. Galloway & Glenn Jones - SAE 725276

308. Future trends in energy conversion systems.
     H -W. Welsh - SAE  660603

309. A method  for estimating and graphically  comparing  the  amounts of
     air pollution emissions attributable to automobiles,  buses,  commuter
     trains and rail transit.
     J.W. Scheel - SAE  720166

310. Factors affecting diesel smoke in highway operation.
     W.A. Howe - SAE Golden anniversary diesel  engine  meeting - 1955

311. Dimensions of diesel fuel performance:  Design, depressants  and
     response.
     B.L.  Michel and L.D.  Fergesen - SAE  660371

312. Conversion of high  speed,air  cooled diesel engines from precombustion
     chamber process to direct injection.
     H.  Ldnnenkohl  - SAE 660010

-------
                                   9-24
313. Future development  of free piston gasifier turbine combinations for
     vehicle traction.
     F.J. Wallace et al - SAE  660132

314. The role of flow improvers in solving autodiesel winter  fuel problems.
     K.A. Beyreis et al  - SAE 660372

315= Piston Cooling.
     C.C.J. French  -  SAE 720024

316. Noise  source definition -  interior vehicle  noise.
     R.J. Vargovick - SAE 720274

317. The effect  of fuel and vehicle variables on polynuclear  aromatic hydro-
     carbon and phenol emissions.
     G.P.  Gross - SAE 720210

318o Reduction of diesel  smoke in California.
     M.L. Brubacher - SAE 660548

319. Effect  of air swirl on smoke and gaseous  emissions from D.I. diesel
     engines.
     I.M.  Khan et al - SAE 720102

320. Comet V automotive diesel engine - Progress Report No.  4.
     D.P.  9458  - 1966

321. Analysis of reported tests on a 6 cylinder turbocharged engine.
     D.P.  9451  - 1966

322. Comet V automotive diesel engine.
     D.P.  9419  - 1966

323. Ricardo Petter high speed 80 x 95 mm D.I.
     D.P.  9361  - 1966

324. 1.5 litre diesel car.
     D.P.  9319  - 1966

325. Report of some tests  carried out on a 1.5 litre diesel  car.
     D.P.  9254  - 1966

326. Comet V automotive diesel engine.
     D.P.  9241  - 1966

327. 4 cylinder IDI diesel  engine of about  3 litres.
     D.P.  9231  - 1966

-------
                                   9-25
328.  4  cylinder ID I dies el engine of about 3 litres.
      D.P. 9056- 1966

329.  Comet  V automotive diesel  engine.
      D.P. 9048 - 1966

330.  6  cylinder indirect combustion diesel engine  5^ litre.
      D.P. 10794 -  1968

331.  1.5  litre Comet VA (4 cyl).
      D.P. 10700 -  1968

332.  PHS 80 x 95 mm engine.   A comparison of toroidal D.I. and Comet
      Mk  V combustion  systems.
      D.P. 10698 -  1968

333.  Comet  V automotive diesel  engine.
      D.P. 10591 -  1968

334.  Comet  V automotive diesel  engine.
      D.P. 10485 -  1967

335.  Comet  Vb engine,  4 cyl,  2^ litres.
      D.P. 10191 -  1967

336.  Comet  Vb engine,  4 cyl,  2£ litres.  D.P. 10042  -  1967

337.  Comet  V automotive diesel  engine.
      D.P. 10018 -  1967

338.  2  litre  Comet  V - Heat losses to water and  lubricating oil.
      D.P. 9942 - 1967
                 3
339.  4  cylinder 2/4 litre D.I. engine.
      D.P. 9931 - 1967

340.  Comet  Vb engine,  4 cyl,  2^ litres.
      D.P. 9924 - 1967

341.  Toothed belt timing drives endurance tests on a 2 litre diesel engine.
      D.P. 9841 - 1967

342.  Comet  Mk Vb  engine,  4 cyl,  2£  litres.
      D.P. 9840 - 1967

343.  Comet  Mk Vb  engine,  4 cyl,  2\  litres.
      D.P. 9814 - 1967

-------
                                   9-26
344. Some notes on the theory of induction ram.
     D.P.  9745 - 1967

345. The application of induction ram techniques to four stroke diesel engines,
     D.P.  9740 - 1967

                 3;
346. 4 cylinder 2/4  litre D.I. engine..
     D.P.  9690 - 1966

347. The position in regard to  small  engine D.I. development to  Nov. 1966.
     D.P.  9634 - 1966

348. 4 cylinder IDI diesel  engine of just  less than 3 litres.
     D.P.  9480 - 1966

349. 4 cyl. diesel Comet V conversion.
     D.P.  13320 - 1970

350. 6 cyl. IDI diesel.
     D.P.  13811 - 1971

351. Exhaust emissions  of  an IDI diesel engine.
     D.P.  13766 - 1971

352. 1000  hr endurance test of a DI  diesel engine.
     D.P.  11449 - 1968

353. Notes on  work with ram  pipes on an IDI diesel engine.
     D.P.  11507

354. 1.5 litre  Comet  VA.
     D.P.  11020 - 1968

355. 2.5 litre  Comet  Mk Vb engine.
     D.P.  11105 - 1968

356. Possibilities for  induction ram on small high speed diesel engines.
     D.P.  11166 - 1968

357. Notes on  exhaust manifold layout for four stroke naturally  aspirated
     engines and  its  effect on engine performance.
     D.P.  11167 - 1968

358. Exhaust emission investigation on  a  typical prechamber  diesel engine.
     D.P.  11187/11188 -  1968

-------
                                    9-27
359. Comet  V automotive diesel  engine.
     D.P. 10848 - 1968

360. 4  cyl.  23/4 litre Comet V.
     D.P. 10975 - 1968

361. Diesel  engine design study.
     D.P. 10990 - 1968

362. Turbocharged loop scavenge engine.
     S.H. Henshall - Proc. I.Mech.E. Vol. 177 No.  6 - 1963

363. The  future of lightweight high-speed diesel engines in the automotive
     and  agricultural fields.
     J.G. Dawson & N.M.F. Vulliamy - Proc. I.Mech.E.  Vol. 177 No.  38 -
     1963

364. Nitrogen oxides and other toxic gases in diesel  engine  exhausts.
     H. Middledich - Proc. I.Mech.E. Vol. 179 Part  1  1964-65

365. The  diesel engine in association with the gas turbine.
     E. Chatterton - Proc.  I.Mech.E.  Vol.  174 No.  10  - 1960

366. The  high speed  heavy duty diesel  engine,  its  development, design and
     application.
     E. Schmidt - Proc. I.Mech.E. Vol. 174  - 1960

367. Noise  and  vibration problems in  commercial  vehicles.
     T. Priede  -  J.  Sound Vib.  (1967)  5(l),  129-154.

368. A  Summary  of Turbocharging  Investigations in 1966.
     D.P. 17543  - 1974

369. Automotive Piston engine noise and its reduction - a literature survey.
     W.W. Soroka &  C-S.F. Chien  - SAE 690452

370. Engineering know-how in engine design. Part  20. SAE SP-369.1972
     Acoustics  and noise.   Vehicle noise  measurement.  Noise legislation
     on product design.  Abatement of structural  noise.  Practical noise
     control.

371. The  diesel engine as a source of  commercial vehicle noise.
     P.E. Waters et al -  Proc.  I.Mech.E.  -  1970

372. Statistical  investigation into diesel engine noise.
     Dr.  Ing. W.  Hempel - CIMAC Working Group Paper - given to I.Mar.E.
                (r V=-<- V  IT,,'   -?o%

-------
                                   9-28
374. The  Performance of BP Vanellus S3-20W in an Austin Taxicab.
     B.P. Technical Bulletin - 1972

375. Origins of reciprocating  engine noise.
     D.  Anderton et al  - ASME 70-WA/DGP-3.

376. Some studies into  origins of automotive diesel  engine noise and its
     control.
     T. Priede -  FISITA  Congress - 1966

377. Identification of mechanical  sources of noise in a diesel  engine:   sound
     emitted from the valve mechanism.
     B.J. Fielding & J.  Skorecki - Proc. I.Mech.E.  Vol.  181 Part I No.  19
     1966-67

378. Identification of mechanical  sources of noise in a diesel  engine:  sound
     originating from piston slap*
     B.J. Fielding & J.  Skorecki - Proc. I.Mech.E.  Vol.  184 Part I No.  46
     1969-70

379. Medium  speed diesel  engine noise.
     R. Bertodo & J.H.  Worsfold - Proc. I.Mech.E.  Vol.  183 Part I No.  6 -
     1968-69

380. On  piston slap as  a source  of engine noise.
     D.  Ross  & E.R. Ungar -  ASME 65-OGP-10

381. Urban  noise  legislation.
     C.  Caccavari - SAE 720902

382. Air  swirl on a  road vehicle diesel engine.
     D.  Fitzgeorge & J.L*  Allison - Proc.  I.Mech.E. (A.D.) No.  4 -
     1962-63

383. Particulate sampling.
     D.P. 12553 - 1970

384. Single  cylinder  engine. Investigation into the merits  of the D.I.  and
     Ricardo Comet Va  Combustion systems in regard to  exhaust emissions.
     D.P. 12080 - 1969

385. Tests on a structures research car.
     D.Po 12162 - 1969

386. Air  utilisation in diesel engines.
     D.P. 12205 - 1969

-------
                                    9-29
387.  Slant 4 Comet Mk Vb.  Note on light spring diagrams of cylinder and
      inlet manifold pressures obtained at  4500 rpm.
      D.P. 12288 -  1969

388.  Exhaust emission investigation on anIDI diesel saloon car.
      D.P. 11603 -  1969

389.  Exhaust Control  of Diesel Engine (EGR & Water Injection)
      M. Tashiro, Isuzu Motors - E.P.A.  Report  Feb.  1974

390.  Small diesel engines for the 1970's.
      D.P. 11832- 1969

391.  Torque back up in current automotive diesel engines.
      D.P. 11284 -  1968

392.  Motoring  compression  pressures.
      D.P. 7569 - 1963

393.  Note on sound insulation carried out by Ricardo on a compact diesel
      sedan.
      D.P. 7599 - 1964

394.  4 cylinder Comet V.
      D.P. 7723 - 1964

395.  Injector Maintenance.
      D.P. 7735 - 1964

396.  Summary of incoming reports concerning an automotive  IDI diesel
      engine.
      D.P. 7776 - 1964

397.  4 cyl. Comet III diesel engine - Progress Report No. 1.
      D.P. 8416 - 1965

398.  Report on the calibration  tests of a 4 cylinder IDI diesel engine.
      D.P. 8465 - 1965

399.  Report on the recalibration  tests of a 4 cylinder  IDI  diesel engine with
      combustion chamber to latest specification.
      D.P. 8639 - 1965

400.  4 cyl. Comet Va (diesel  engine) - Progress Report No.  2.
      D.P. 8722 - 1965

-------
                                   9-30
401. Comet V automotive diesel engine.
     D.P.  8748  - 1965

402. Diesel car road tests.
     D.P.  8770  - 1965

403. Survey of European Diesel engines.
     D.P.  8811  - 1965

404. Diesel car road tests.
     D.P.  8934  - 1965

405. Note on further induction ram tests on  an IDI diesel engine.
     D.P.  12083 - 1969

406. 4 cylinder  Comet  Mk V.
     D.P.  7248  - 1963

407. 4 cyl. Comet Vb 2.8 litre  (normally aspirated  high  speed version).
     D.P.  7158  - 1963

408. 4 cyl. Comet Va 2.8 litre  (normally aspirated  high  speed version)
     D.P.  7159  - 1963

409. 4 cylinder  Comet  Mk V.
     D.P.  7976  - 1964

410. 4 cylinder  Comet  Mk V.
     D.P.  7554  - 1963

411. 4 cyl. 2^ litre diesel  engine.
     D.P.  7981  - 1964

412. Noise tests of two diesel sedans.
     D.P.  8047  - 1964

413. 2 litre Comet V.
     D.P.  8220  - 1964

414. California  exhaust air pollution test on  a Diesel car.
     D.P.  8264  - 1964

415. 2 litre Comet V.
     D.P.  8298  - 1965

416. Comet Mk  V diesel  engine.
     D.P.  8310  - 1965

-------
                                   9-31
417.  2 litre Comet V.   .
      D.P. 8365 -  1965

418.  3 litre" Comet V.
      D.P. 8386 -  1965

419.  Report on calibration  tests of the 4 cylinder  in line diesel engine after
      conversion to  Ricardo Comet V combustion system
      D.P. 8839 -  1965

420.  Report on the  road  testing of a diesel car.
      D.P. 8840 -  1965

421.  1 litre Comet V engine (4 cyl) low speed.
      D.P. 8570 -  1965

422.  4 cyl. diesel engine.
      D.P. 8910 -  1965

423.  Notes on cold starting of diesel  engines.
      D.P. 13905 -  1971

424.  Comet V automotive diesel engine.
      D.P. 10164 -  1967

425.  Application of the flash temperature concept  to cam and tappet wear
      problems.
      A. Dyson & H. Naylor -  Proc. I.Mech.E.  1960-61 No. 8

426.  Measurement of piston ring and  cylinder liner wear by using radio-
      active tracers.
      R.F.  Pywell and S.T.  Walker - Proc.  I.Mech.E.  (AD) 1960-61  No. 8

427.  Fuel  injection  system  calculations.
      B.E.  Knight - Proc. I.Mech.E. (AD)  No. 1  1960-61

428.  Calculations of the  effect of  heat release on  the  shape of the cylinder
      pressure diagram and cycle efficiency.
      W.T. Lyn - Proc. I.Mech.E. (AD) No.  1  1960-61

429.  Air  cooled automobile engines.
      J.  Mackerle -  Proc.  I.Mech-.E.  (AD) No.  2 1961-62

430.  Relation  between fuel  injection  and heat  release in a direct injection
      engine and the  nature of the combustion process.
      A.E.W.  Austen & W.T. Lyn - Proc. I.Mech.E. (AD) No. 1 1960-61

-------
                                  9-32
431. Effect of engine structure on noise of diesel engines.
     T. Priede, A.E.  Austen &  E.G. Grover - Proc. I.Mech.E. Vol.  179
     Part 2A No.  4 1964-5

432. Automobiles  and  Petroleum:  Past, present & future.
     Prof. JoJo Broeze - Proc. I.Mech.E. (AD) No. 7  1953-54

433. Development and application  of automotive fuels - diesel and gasoline.
     A.E. Felt & G.C. Wilson (1954)

434. Some notes on the design, development and production of high  speed
     compression ignition  engines.
     S. Markland & N. Tattersall - Proc. I.Mech.E. 1947-8

435. Symposium on superchargers and supercharging.
     Proc. I.Mech.E.  No. 6  - 1956-7

436. Performance of vehicles under trans antarctic  conditions.
     D.L. Pratt - Proc. I.Mech.E.  (AD)  No.  6  - 1958-59

437. Commercial  Vehicle performance and fuel economy.
     G.L. Smith - SAE SP 355 (700194)

438. What can the turbocharger  do for  the engine?
     W. Lang  - SAE 660473

439. The  influence of induction and  exhaust  system design on power pro-
     ducing characteristics of diesel engines.
     H.G. Holler - SAE 700535 (SP 359)

440. The  turbocharger - a vital part of the engine intake and exhaust systems
     W.E. Woolenweber - SAE 700534 (SP 359)

441. Recent developments in variable compression ratio engines.
     J.C. Basiletti and E.F.  Blackburne - SAE 660344 (SP  280)

442. Diesel combustion at high  MEP with low compression  ratio.
     W.Po Mansfield & W.S. May  - SAE 660343 (SP 280)

443. Design and development of a very high output multifuel  engine.
     R.T. Paluska et al -  SAE 670520  (SP 290)

444. Daimler-Benz high output engine - a study in compact design.
     O. Herschmann - SAE  670519  (SP 290)

445. Comparative analysis of Stirling and other combustion systems.
     S.R. Davis & N.A. Henein - SAE 730620

-------
                                    9-33
446. Light diesels  give promise in mail service.
     G. Nield U.S. P.O.  Dept. - SAE  Jnl., July 1966 Vol.  74  No. 7

447. New diesel combustion chamber gives smoother  combustion and cleaner
     exhaust.
     W. Henny  & R. Herrmann (Hispano-Suiza) SAE Jnl. July 1966 Vol.  74
     No. 7

448. High speed diesel rates improvement.
     SAE Jnl.  Dec. 1960  - J.H. Pitchford

449. European small diesel comes of age in light commercial vehicles.
     J.G. Dawson  - SAE  Jnl. -  1961

450. List of manufacturers in current production with engines having Comet
     Mk III & Mk V chambers.
     D.P. 7271

451~ 4  cyl. Comet V  engine  -  Progress Note No. 2
     D.P. 7317

452. 4  cylinder Comet Mk V.  Report on tests carried out with two differ-
     ent   injection-purnps during the period March to June 1963.
     D.P. 7290

453. 4  cyl. Comet Vb 2.8 litre normally aspirated high  speed version engine
     No. 2 - Progress Note  No. 5.  Noise analysis  tests.
     D.P. 7340

454. Relation between form of cylinder pressure diagram and noise in  diesel
     engines *
     T.  Priede  - Proc. I.Mech.E.  (AD) No.  1 1960-61

455. An investigation  into factors which affect piston  ring wear under starting
     conditions.
     J.  Cree & J. Thiey - Proc. I.Mech.E. (AD)  No. 8 -  1960-61

456. A  comparison of the wear of  small gasoline and diesel laboratory
     engines.
     W.C. Pike & P. Newman - Proc. I.Mech.E.  (AD) No. 8 1960-61

457. MWM Diesel features a  new pre-combustion  chamber-excepts from  paper
     by H.L. Hockel - SAE Jnl.  Sept.  1957

458. Diesel for vehicles  under 5000 Ib -
     based on discussion  and paper by J.S. Bright -  F.  Perkins (Canada)  Ltd.
     SAE Jnl.  Feb. 1959

-------
                                  9-34
459. How to  step up  fuel economy -
     based on paper  by L.D. Conta & P. Durbetaki - SAE Jnl. March 1959

460. Total  energy situation in the United States.
     D.R.  Oliver - SAE 730514 (SP 383)

461. Impact of automotive emissions regulations  on gasoline  demand.
     D.H.  Clewell  and W.J. Koehl  - SAE 730575  (SP 383)

462. Current and future trends  in United States gasoline supply.
     E.J. Cahill -  SAE  730516  (SP  383)

463. Fuel consumption trends in today's  vehicles.
     H.C.  MacDonald - SAE 730517 (SP  383)

464. Energy and the automobile - General factors affecting vehicle fuel  con-
     sumption.
     G.J.  Huebner Jr.  &  D.J.  Gasser - SAE 730518 (SP 383)

465. Alternative powerplants.
     J-J. Brogan - SAE 730519 (SP 383)

466. Synthetic fuels for transportation  and national energy needs.
     D.P.  Gregory & R.B.  Rosenberg - SAE 730520 (SP  383)

467. Mass  transit impact  on energy consumption.
     R. Husted - SAE 730521 (SP 383)

468. Criteria for evaluating vehicle in use inspection/maintenance impact
     on emissions  and  energy  conservation.
     J.C.  Elston - SAE 730522  (SP 383)

469. Low voltage ignition will start cold  diesel.
     L.P. Atwell et al  - SAE Jnl. May 1960

470. Diesels gaining: Some problems ahead.  E.R.  Klinge -  SAE Jnl., -
     May 1960

471. The I.H.C.  D-301  Diesel.
     W.E.  Peterson - SAE Jnl.  - Jan.  '61

472. Mobile powerplant  trends  forecast.
     H.W.  Welsh - SAE Jnl. Aug. 1967

473. Some notes  on stratified charge combustion systems.
     D.P.  16744 -  1973

-------
                                    9-35
474. A status  report on the development  of the 1973 California diesel
     emissions standards.
     R.C. Bascom & G.C.  Hass - SAE 700671

475. Ricardo electronic smoke  sensor.
     D.P. 16547 - 1973

476. Diesel combustion research.  A  review  of 20 years combustion
     chamber  development.
     D.P. 11378 - 1968

477. Fuel consumption  of diesel cars.
     D.P. 17334 - 1973

478. The  development of a  variable compression ratio  diesel engine which
     can  greatly reduce the diesels  emissions of NOX.
     J. Witzky -  Unclassified Document - Ex. E.P.A.  1970

479. The  light  weight diesel engine.
     D.P. 17311 - 1973

480. Diesel and gasoline fuel consumption comparisons.
     D24923 Ricardo

481. Ethyl Ether best for starting  cold diesels.
     F. Nelson & C.J., Ulzheimer - SAE  Jnl.  March 1950  p42

482. Diesel compared with gasoline  engine.
     F.B.  Lautzenhiser - SAE Jnl. June 1950 p73

483. Making diesel engines deliver greater power.
     C.R. Maxwell - SAE Jn.l.  July  1950 p48

484. How engine was developed for Worlds Fastest Diesel  Car.
     J.C. Miller & C.R. Boll - SAE Jnl.  Oct. 1950 p53

485. Diesel economy stems from higher BTU  usage. •
     M.C.  Horine -  SAE Jnl.  Oct. 1950  p92

486. Thermal gains seen as Diesel edge over  gasoline engine.
     F.G. Shoemaker - SAE Jnl. Nov. 1950 p°57

487. Power Booster Fuels for Diesel engines.
     E.J.  Mclaughlin et  al - SAE Jnl. June 1952, p42.

488. Diesels can  be started at - 65 F.
     D.E. Kiltv et al - SAE Jr',.

-------
                                  9-36
489. Un Diesel dans  le monde du silence.
     B. Carat -  L1 auto-Journal 15 Dec.  '73 p26

490. Peugeot 504 diesel.
     Quattroruote (Italian Magazine) December '73

491. Ricardo curves D24426/A - 30, 24494  - 1973

492. Nitrogen oxides - diesel villian - gases attack oil and engine parts.
     SAE  Jnl. Heavy Duty  Issue - Sept. 1963  - Article based on material
     drawn from paper No. 714B

493. 2 Neue Motoren fur Mercedes-Benz.
     Mittelklassewagen - MTZ Sept. '73

494. Results obtained on  a  single cylinder  Comet  V.
     D.P. 6853  - 1962

495. Notes  on preliminary  road  tests using wide cut fuel.
     D.P. 6881  - 1962

496. Comet Vb diesel engine - Progress Note  No. 5.
     D.P. 6905  - 1962

497. Tests to determine  firing requirements to avoid misfire at all  con-
     ditions in an IDI diesel engine.
     D.P. 7038  - 1963

498. Comet V -  tests with a distributor type pump.
     D.P. 7057  - 1963

499. Comet V diesel engine - Progress Note No. 1.
     D.P. 7086  - 1963

500. The  operation of compression  ignition  engines  on wide boiling range
     fuels.
     Donel R. Olson,  Nelson T.  Meckel,  & R.D. Quillian,  Jr.
     SAE Trans. Vol. 70,  1962,  p551.
501. Some steps toward  calculating diesel engine behaviour.
     A.E.W. Austen & V.T. Lyn - CAV Ltd. presented to  I.Mech.E. Nov.
     1960

502. GMR 4-4 Hyprex free piston turbine engine.
     AoF. Underwood -  Head Mech. Dev'ment.  G.M. Corp. SAE Summer
     Mtg. June  5th  '56

503. Hybrid engines.
     P.H. Schweitzer & L.J. Grunder  - Richfield Oil Corp.
     SAE Trans. Vol. 71,  1963,  p  541-562.

-------
                                   9-37
504. Charge stratification by fuel  injection  into swirling .air.
     A.W. Hussman et al (work done under contract to U..S.  Army Ord-
     nance)  c.1964

505. 1.5  litre Comet V engine (4  cyl) - Progress Report No. 21 - Nov. 1962.
     Effect of changes  in water and lubricating  oil temperatures on engine
     performance and volumetric efficiency.
     D.P. 6919 - 29th Nov.  1962

506. 1.5  litre diesel engine.
     D.P. 6836 - 6th Dec.  1962

507. 1.5  litre Comet V engine (4  cyl) Progress Report No. 20. Covering  work
     done -  Sept.-Dec. 1962.
     D.P. 6990 - 16th Jan.  1963

508. The  effect of thermal flow in Comet chamber walls.
     D.P. 7036 - 12th Feb.  1963

509. Cold starting in the small high  speed  diesel engine.
     D.P. 17771 - 1974

510. Comet V diesel engine - Progress  Report No.  4.
     D.P. 7937 - 1963

511. Current small 4  cylinder engines for automotive applications..
     Ricardo  data sheet - 1974

512. Untitled Ricardo report  - D.P. 17357 - 1973

513. Comet V diesel engine - Progress  Report D.P. 6419 - 1961

514. IDI diesel engine  -  Progress  Report -  D.P. 6421  -  1961

515. Comet V diesel engine - Progress  Note No. 5 - D.P.  6480 - 1962

516. Comet V diesel engine - Progress  Note No. 7 - D.P.  6559 - 1962

517. Comet V diesel engine - Engine No. 2 -  Progress Note No. 2 - 1962
     D.P. 6509

518. Comet V diesel engine - Progress  Note No. 3 - D.P.  6600 - 1962

519. 4 cylinder Comet Vb -  weight of component parts - D.P. 6458  -  1962

520. 6 cylinder IDI  diesel engine - weight of major components.

-------
                                   9-38
521. 180 cu.in gasoline engine - weight of engine parts.
     Unclassified document 1973

522. 75 cu.in gasoline engine - weight of component parts - 1968

523. 120 cu.in Comet V diesel  engine - weight of component parts.
     D.P.  6183 -  1961

524. 80 cu.in gasoline engine - weight of component parts.
     DoP.  9952 -  1967

525. 100 cu.in gasoline engine - weight of component parts.
     Unclassified document

526. 150 cu.in gasoline engine - weight of component parts.
     Unclassified document

527. 6  cylinder   D.I. weight of component  parts.
     D.P.  6645 -  1962

528. 4 cylinder Comet V diesel engine - Progress Note No.  8.
     D.P.  6607 -  1962

529. 4 cylinder Comet V diesel engine - Progress Note No.  6.
     D.P.  6525 -  1962

530o Work  carried out  on No. 2  engine (IDI  diesel) prior  to despatch.
     D.P.  6739 -  1962

531. IDI diesel engine no.  2 -  Progress Note  No. 3.
     D.P.  6773 -  1962

532. 4 cylinder Comet V - Progress Note No. 1.
     D.P.  6057 -  1961

533. 1.5 litre  Comet V.   Progress Report No.l.
     D.P.  6064 -  1961

534. 1.5 litre  Comet V -  review of temperature tests on the various
     piston forms.
     D.P.  6122 -  1961

535. Light  engine oils for  improved subzero  operation.
     V.G.  Raviolo -  SAE Trans. AP.  1950 No. 2  pl6l

536. Matching  fuels  to diesel combustion systems.
     Dr. C.G.A.  Rosen - 1962 Horning Lecture

-------
                                   9-39
537. Dual fuel combustion of propane in a railroad diesel engine.
     J.M. Clark & H.M. Bunch - S.W. Research Institute c.1960

538. Can you afford diesel  engines for  highway hauling?
     R.W.  Richardson  & E.G. Caputo:   Simmonds Precisions  Products  Inc.

539. Swirl and combustion in divided combustion chamber type diesel
     engines -
     F.  Nagao & H. Kakimoto - SAE Trans.  Vol 70 - 1962

540. Gas temperatures during compression in motored and fired diesel
     engines.
     K.C. Tsao, P.S.  Myers and  O.A. Uyehara - SAE Int.  Congress  -
     Jan. 1961

541. Advantages  of propane  as  a transit vehicle fuel.
     R.  Lee  - SAE Trans. No.  2 Vol.  6 April 1952

542. I.H. high-speed  lightweight diesels.
     A.  Dewsberry  et  al - SAE Trans.  Vol. 68 -  1960

543. Fumigation kills  smoke  -  improves diesel  performance.
     M. Alperstein  et  al - SAE Trans.  Vol. 66  1958

544. Subzero winterization of diesel engine power  equipment.
     P.W.  Espenschade et al -  presented  to  SAE Nat. Diesel engine mtg,
     Chicago Oct. 30th 1951  -  Vol. 6 No. 4 Oct.  1952

545. Compression temperatures in diesel engines under starting conditions.
     Pennsylvania State University - W.E. Meyer et  al

546. Report  concerning turbocharged IDI  diesel engine - 1961.

547. Report  concerning turbocharged IDI diesel  engine -  1962.

548. Report  concerning turbocharged IDI diesel  engine -  1962.

549. Report  on temperature  measurement tests  in  the hot plugs for an IDI
     diesel  engine - 1967.

550. Weights of component parts of various  engines.
     120 cu.in diesel  engine (Comet V) - D.P. 6183 - llth Sept.  1961

551. Effect of valve size and timing on the torque of small  Comet  engines.
     D.P.  17499 & 17544

-------
                                   9-40
552. Diesel car road tests.
     D.P.  8805 - 4th  Oct. 1965

553. 4 cylinder Comet Mk V - Progress Report No.  2 - June-Sept.  1961.
     D.P.  6216

554. Constant power diesel engine - Gas & Oil Power -  April  1964

555. Fuel requirements of the small high speed diesel engine.
     G.  Barrett et al  - Gas & Oil Power - Feb. 1956

556. Investigation into nozzle blockage in high speed diesel engines.
     G.  McConnell  - Gas & Oil  Power - April  1960

557. The Volvo dual powerplant for military vehicles.
     S. Kronogard - Turbine & Automatic Trans.  Div, AB  Volvo,
     SAE 660017

558. The challenge of  pollution.
     P.  Myers - Jnl of Automotive Engineering - April  1970

559. Towards quieter diesels.
     M. Russell  -  CAV Ltd.  Jnl of Automotive Engineering - Dec.  1970

560. Exhaust emission legislation.
     A.  Aitken - Ford Motor Co.  Ltd.  - Jnl.  of Automotive Engineering  -
     July 1971

561. Car Maintenance - a reappraisal by Marcus Jacobson - Jnl. of
     Automotive  Engineering - July 1971

562. Recent developments  in high speed oil engines.
     Prof. S. Davies - March 1938

563. Oxides  of nitrogen in diesel engine exhaust gas:  their formation and
     control.
     G.  McConnell  - Proc.  I.Mech.E.  1963-64

564. Research  on the compression ignition engine and its fuels.
     P. Vaile - Abstract  of I.Mech.E.  paper, read before North Western
     Branch  in Manchester - 24th April 1948

565. Combustion in diesel engines.
     H.R.  Ricardo  - lecture delivered 21st Feb. 1950 -  London

566. Producer  gas  for road transport.
     J. Hurley & A. Fitton - 30th July 1947

-------
                                   9-41
567. An experimental investigation into the  effects of fuel addition to intake
     air on  the performance of a compression ignition engine.
     W. Lyn - paper  accepted  by Institution Council  for publication  12th March
     1953

568. The effect of auxiliary fuels on the smoke limited power output of diesel
     engines.
     L. Derry et al - paper received  at Institution 14th Jan.  1953

569. Comparative studies of methane and propane  as  engine fuels.
     N. Moore - paper accepted  by  Council for   publication  23rd Dec. 1955

570. Diesel  engine  lubricants.   Their selection and utilization with parti-
     cular reference  to oil alkalinity.
     A. Dyson et al - paper received by Institution on 27th July  1956

571. Horizontal diesel engine Comet V.           .
     Progress  Report No.  7 -  D.P.  6159 -  llth August  1961

572. Comet  V  4  cyl.  2.8 1 normally  aspirated, high speed version.
     Progress  Note No.  1.
     28th August  1961 - D.P.  6170

573. Diesel  engine  smoke  and  pollutants.   Diesel  engineers  and users  asso-
     ciation  -  J.  Spiers & M.  Vulliamy - Publication 342  -  18th Feb. 1971

574. CRC Correlation of diesel smokemeter measurements.
     F. Hills,  T. Wagner, D.  Lawrence.  SAE 690493- May 23rd  1969

575. Steady-state correlation of diesel smokemeters.
     SAE Task Force  Report -  A. Carey - SAE 690492 -  May 1969

576. The measurement of diesel engine smoke.
     A. Dodd & Z. Holubecki. - MIRA Report 1965/10.

577. Curves for  4 cyl,  IDI diesel  engines.
     Dec.  1968 - Unclassified  document.

578. Combustion tests on an IDI diesel engine  - Jan. 1965
     Unclassified document.
579. Curves for  EDI diesel  engine  dated 17.3.64.
     Effect of heater  plug  position in Mk VI chamber. Unclassified' document.

580. Test report dated 1.4.63.
     Cam form effect on  performance of an IDI diesel engine @ 4500 rpm.
     Unclassified document.

-------
                                   9-42
581. Test report dated 20.6.69.
     Ram pipe tests on an IDI diesel engine d> 3000  rpm.
     Unclassified document.
582. Motoring  loss tests  on an IDI diesel  engine.
     c.  1965  - Unclassified document.
583. Test report dated Feb. 1964.
     Noise tests on an IDI diesel  engine at 4500 rpm.  Unclassified  document.

584. Tests on an IDI diesel engine Mk VI  chamber.
     Piston  combustion chamber tests at various F.I.E.  timings - 27.3.64
     Unclassified document.
585. High speed C.I.  engine combustion systems  - D.P.  327E

586. Performance data, N.A.  4 stroke,  blower scavenged 2 stroke with
     Ricardo combustion  systems - D.P.  199A & B.

587. D.I. emissions.                             «.
     GO.373, 6th March   1948

588. Combustion systems in high speed diesels.
     D.P. 380, A, B, C.

589. Relative  advantages  of aluminium versus cast iron crankcases for
     high speed diesel engines.
     D.P. 651 - 1st  April 1949

590. Note on  turbocharging of high speed Comet units.
     D.P. 771 - 8th  Sept. 1949

591. E12 Comet III/IV.
     GO.1044

592. Torque reaction tests.
     GO.2136

593. Comet III, Camshaft tests.
     GO.2470 - 28.3.55

594. Cold starting  diesel  engines.
     D.P. 2711 - 23rd June 1953

595. Comet comparisons  of Mk II, III,  IV.
     D.P. 2779 - 28th Sept.  1953

596. Diesel car.
     D.P. 2904 - 4th March 1954

-------
                                   9-43
597. Tests on 4 cylinder diesel engine.
     D.P. 2925 - 30th March 1954

598. Inlet manifold starting  device.-
     D.P. 3057 - 25.11.54

599. Notes on  tests to assess relative performance of Comet III;,  V, VI and
     inverted V chambers..
     GO.3100 - 11.11.55

600. 21  Comet V  Progress Report No. 3.
     D.P. 3144 - Nov.  1956

601. Diesel engine.
     D.P.. 3238 -29th Sept. 1954

602. High speed high  power road transport engine^.
     D.P. 3267 - 29th Oct.. 1954

603. Diesel engine.
     D.P. 3324 - 17th Dec.  1954

604. Diesel engine.
     D.P.. 3359 - 29th Jan.. 1955

605. Diesel engined taxi.
     D.P.. 3372 - 4th, Feb..  1955

606. Diesel engine.
     D.P.. 3394 - 25th Feb.  1955

607. Diesel tractor engine.
     D.P. 3395 - 25th Feb.  1955

608. 90  BHP diesel/petrol comparison.
     D.P.. 3400 - 3.355

609. Diesel car.
     D..P. 3405 - 7th March 1955

610. Noise reduction.
     D.P. 3413 - 16th March 1955

611.- Gold starting tests on  2 /4 litre Comet III. chamber engine.
     D.P. 3427 - 30.3.55

612. Small Comet V  engines.
     D.P. 3460 - 12.5.55

-------
                                  9-44
613. High speed loop  scavenge diesel engine.
     D.P.  3472 - 23.5.55

614. Comparative weights - diesel/petrol.
     D.P.  3474 - 1.6.55

615. Volumetric efficiency  of  diesel engine.
     D.P.  3493 - 25.6.55

616. 2J/4 litre  Comet V.
     D.P.  3550 - 21.9.55

617. Diesel tractor  engine.
     D.P.  3700 dated 25.1.56

618. Automotive diesel engine.
     D.P.  3710 dated 1.2.56

619. Diesel engine.
     D.P.  3714 dated 10.2.56

620. Comet V  engine.
     D.P.  3724

621. Comet V  engine  2 1.
     D.P.  3730 - 27.2.56

622. Diesels for road transport.
     D.P.  4799 - 4.11.58

623. High speed multi-cylinder Comet V engines of up to 90  mm bore.

     D.P.  4847 - 12th Dec. 1958

624. Report of work carried out on a 2 /4 litre high ratio Comet V  engine
     between April-Nov.  1958.  K.A. Atkins - D.P. 4862 - 30th Dec-  1958

625. Tests with hot  plug throats of various shapes E12/4.
     D.P.  5175 - 27th August 1959

626. The small  diesel engine for  cars.
     D.P.  5284 - 11.11.59

627. IDI diesel  engine with Bosch heater plugs.
     D.P.  5296 - 23.11.59

-------
                                    9-45
'628.  Small high speed diesel.
      D.P. 5609  -  4.8.60

 629.  PHS progress note no.  1.
      D.P. 5640  -  10.8.60

 630.  High speed diesel engine proposals.
      D.P. 5703A - 24.11.60

 631.  The effect of diagram  shape on combustion noise in diesel engines
      D.P. 5742  -  24.10.60

 632.  Motoring friction.
      D.P. 5844  -  4.1.61
 633.  1^1  Comet V engine - Progress Note No. 12'
      D.P. 5948  - 8.3.61

 634.  Comet  V diesel  engine.
      D.P. 6009  - 20.4.61

 635.  Automotive diesel, nozzle tests.
      D.P. 6019  - 28.4.61
636.  1^1  Comet V engine - Progress Note No. 13.  D.P. 6026 - 2.5.61

637.  4 cylinder Comet  Mk V  engine progress note no.  5.
      D.P. 6480 - 23rd Feb.  1960

638.  The 2 stage rotary engine - a new concept in rotary power.
      F. Feller  - I.Mech.E. Proc. 1970-71  - Vol.  185 -  13/71

639.  Maintenance frequency in certain  bus operating undertakings as it
      affects injection equipment  and combustion zone components.
      D.P. 3870 - 1956

640.  Classification of  diesel  fuels.
      R.P. Linderan,  D.K.  Lawrence and T.O. Wagner  - SAE 680467  -  May
      1968

641.  Avoidance  of diesel fuel filter plugging  in winter months.
      Esso - Petroleum  Technical Service No.  73/29 -  16.11.73

642.  The cold filter  plugging point:  A practical operability task for middle
      distillate fuels.  Paramins  (Esso)  RBM 5SU 73 April 5 - 1973

-------
                                   9-4b
643. Minimising low  temperature sensitivity of diesel vehicle fuel systems.
     (Esso) Paramins RBM 54SU-X  73.  - 19.11.73

644. How to  run diesel engines in cold  weather.
     W.  Sloan  et al - SAE Jnl. - June  1951

645. Fleets,  but not  private cars, are likely customers for LPG.
     SAE Jnl.  Dec. 1951

646. 4000 rpm is  practical aim  for automotive diesels.
     N.M. Reiners & R.C. Schmidt - SAE Jnl. Dec.  1951.

647. Daimler Benz Test  results - EPA report -  1973

648. The small high  speed diesel engine - a power unit for future passenger
     cars.
     D.P.  17478 - 1974

649. Berliet Pat. Brit. Pat.  1,128,661

650. Berliet Pat. Brit. Pat.  1,213,206

651. Institut  Gornogo Dela Pat.  1,270,782

652. Diesel Exhaust Odor  - its evaluation and relation to  exhaust gas com-
     position.
     F.G.  Rounds  & H.W.  Pearsall.
     SAE Trans.  Volume 65, 1957,  Pages 608/627.

653. Smoke and Odour Control for Diesel Powered Trucks and Buses.
     R.C.  Stahman,  G.  Kettridge, K.J. Springer.
     SAE 680443.

654. Investigation of  Diesel Powered Vehicle Odor  and Smoke.
     Part V, SWRI  Project No. 11-2340-005.   K.J. Springer.

655. Note on a visit  to Messrs.  Opel,  Russelsheim.  by C.A. Beard  on
     November 27th,  1973  - D.P. 17387.

656. Peugeot letter to Ricardo  dated 20th December 1973 + attachment.

657. London General  Cab. Co. Ltd. letter to Ricardo dated 15th January  1974
     + attachment

658. Note on discussion with Herr U= Alen  of Verband Fuer  das
     Personenwerkehrsgewerke,  Hamburg, on May 31st,  1974 - D.P. 18175

-------
                                    9-47


659. Note  on visit to  Slota Taxis, Paris on 9th May,  1974 - D.P.  18100.

660. CAV  letter to  Ricardo  dated  30th November,  1973.


661. Note  on visit to  Deutz, Porz, on 28th November  1973 - D.P.  17394

662. Note  on visit to  Austin Hire  &  Taxi Service Ltd., Worthing on
     19th November 1973 -  D.P.  17320.

663. Note  on a visit by C.A.  Beard to BLMC on 10th January  1974  -
     D.P.  17510.

664. Note  on discussion  with B.P. in London on 21st January 1974 -
     D.P.  17551.

665. A  new  light  aircraft engine  - the  radial 2-stroke diesel.
     J.L.  Dooley - Ex. EPA data  -  1970.

666. Extract from letter  sent by  W.A.  Bareham to  Mr. S. Oshika

667. Mass emissions  and fuel economy from  LDV  diesel engines.
     1975  FTP - Ex EPA  data.

668. Some engine ideas of the Past  and Future.
     C.F.  Bachle

669. Manufacturability and costs  of  proposed  low emissions  automotive
     engine  systems.
     National Academy of Sciences report.
                                                         3
670. Ricardo  Comet   Mk V  Emission levels from a 416 in  truck engine under
     turbocharged conditions.
     D.P.  15176 -  11  May 1972.

671. McCulloch is developing lightweight  aircraft diesel -  Automotive
     Engineering - Sept.   '73, Vol.  79. No. 9, p53.

672. Discussion of diesel operation  at. ]ow temperature at  Shoreham on
     18th January 1974.
     D.P.  17540 - Jan.  '74

673. Some observations on the nature of  blue  smoke in engine exhaust.
     W.T.  Lyn M.I.M.E.  -  CAV  Ltd. - Unclassified document.

674. Exhaust  Particulates  -  D.P.  16471 - April 1973.

-------
                                   9-48
675. Fuel economy and emissions control - EPA report - Nov.  '72.

676. A report on  automobile fuel economy.
     EPA  Report - Oct.  1973.

677. The  Low Emission Car for 1975 - Enter the Diesel.
     K.J. Springer & H.A. Ashby - SAE 739133.

678. Diesel for  passenger cars.
     F. Bachle -  EPA report - Feb.  1974.

679. Nitrogen oxides and variables  in pre-combustion chamber type diesel
     engines - E.W.  Landen -  SAE 714B  - 1963.

680. Feasibility  of meeting the  1975-76  exhaust emission standards in actual
     use.
     National Academy of Sciences  - June 1973

681. Characterization and control of emissions from heavy duty diesel and
     gasoline fuelled engines - prepared for EPA by Bartlesville Energy
     Research Center,  Bureau  of  Mines - Dec.  1972

682. An  evaluation of alternative  power sources for  low emission automobiles
     National Academy of Sciences  - April 1973

683. Manufacturability and costs of proposed low emission automotive engine
     systems -  National  Academy of Sciences  - Jan. 1973

684. Diesel exhaust odor analysis by sensory techniques.
     D.A. Kendall, P.L. Levins & G. Leonardos  -  SAE  740215 - 1974

685. Chemical analysis of diesel  exhaust odor  species.
     P.L.  Levins  et al - SAE 740216 -  1974

686. Public opinion of diesel odor.
     C.J. Hare et al - SAE 740214  - 1974

687. Progress in  diesel odor research.
     C.W. Savery, R.A.  Matula & T. Asmus - SAE 740213 -  1974

688= Letter from K.J. Springer (Southwest  Research Inst) to R.C. Stahman
     (EPA) - Dec.  13,  1973.

689. Notes on Turner 2 stroke  IDI engine  - taken from Commercial Motor &
     Engineering - 1953

-------
                                    9-49
690. Higher  BMEP's in automotive diesels.
     D.  Broome -  reprint from Automotive Design Engineering - March 1966

691. A Designer's Viewpoint.
     H.  Barnes-Moss  - I.Mech.E. paper -  Conference Pub. 19 - No. C343/73

692. Survey  of  diesel  engine emissions - 3rd report.
     Project Group CEC-CF  12, June 20th,  1973.  Unclassified document.

693. Unaided starting  of  diesel engines.
     T.W.  Biddulph &  W.T. Lyn - Proc.  I.Mech.E. 1966-67, Vol. 181,  Part  2A

694. Influence of atmospheric pressure and temperature  upon  the performance
     of the naturally aspirated four  stroke  C.I.  engine.
     C.B.  Dicksee - I.Mech.E. - 1959

695. Diesel engine combustion  processes with a  view to  their use in pass-
     enger cars.
     Prof.  Dr.  Ing. E. Eisele - The Internal Combustion Engines Conference,
     Bucharest  - 1970, Paper No. 1C.

696. Diesel engines operating in  underground mines.
     Dr. Ing. G. Reyl -  Unclassified document

697. Unclassified document  received from EPA -  Chapter XVI - Diesel Engine.

698. Letter from K.J. Springer (Southwest  Research Inst)  to  R.C.  Stahman
     (EPA) - Feb.  15, 1974.

699. Durability  of advanced  emission controls for heavy  duty  diesel and
     gasoline fueled engines  -  prepared for EPA  by Dr.  R.D.  Fleming  and
     T.R. French,  Bartlesville  Energy Research Center,  Bureau of  Mines,
     Sept.  1973.

-------
                                   10-1

                               SECTION 10


                    APPENDIX 3 - GLOSSARY OF TERMS

Some of the terms  mentioned in the text  of this report may not  be fami-
liar to  the  reader and the following  glossary has been compiled  to  clarify
matters.
CID       - engine swept volume in cubic  inches.

CVS-CH  - 1975 cold/hot start  Federal  test procedure using  CVS  (constant
            volume  sampling) equipment.   For diesel powered vehicles HC's
            are measured by means of  a  heated flame ionisation detector,
            the gas sample  being  extracted via a heated  line to minimise
            condensation of  the  heavier  hydrocarbons  (mass  emissions at
            the end of the test are determined by integrating the continu-
            ous trace obtained  from the flame ionisation detector).   This
            is  the only  difference in the 1975 test procedure between gaso-
            line and diesel powered light  duty vehicles.

Comprex - belt driven  engine boosting  device working on a pressure ex-
            changer principle.   Currently being developed by Brown-Boveri,
            Switzerland.

DI        - direct injection  (open chamber).

IDI       - indirect injection (pre-cup or swirl chamber).

EGR       - exhaust gas recirculation^

F.IoE.    - fuel injection equipment.

getter box - catalytic device fitted upstream of a reducing catalyst  to pro-
             tect it from oxygen "spikes".

mpg       - fuel economy in miles per  U.S.  gallon.

NA       - normally  aspirated»

T/C       - turbocharged

DP	)
          - internal identification  for  classification of Ricardo reports.
SN	)

-------
                                 10-2
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. , 2._
EPA -460/3-74-011
4. TITLE AND SU&TITLE
A STUDY OF THE DIESEL AS A LIGHT DUTY
POWERPLANT
7. AUTHOH(S)
C.C.J. FRENCH, M.L. MONAGHAN, R.G. FREESEo
9. PERFORMING ORGANIZATION NAME AND ADDRESS
RICARDO & CO. ENGINEERS (192?) LTD.,
BRIDGE WORKS, SHOREHAM-BY-SEA, SUSSEX.
BN4.5FG. ENGLAND.
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency, Office of Air & Water
Programs, Office of Mobile Source Air Pollution Control,
Emission Control Technology Division, Ann Arbor,
MiVhirpn 48105-
3. RECIPIENT'S ACCESSIOWNO.
5. REPORT DATE
JULY 1974
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NC
D.P. 18410
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-03-0375
13. TYPE OF REPORT AND PERIOD COVEREC
FINAL RKPOKT
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
is. ABSTRACT Tnjs project was carried out to determine whether the diesel engine was a
possible power plant for light duty vehicles in America. The light duty vehicle consid-
ered was a 4/5 seat compact sedan with good acceleration capabilities and was consid-
ered for a primary emissions  environment of HC - 0.41 g/mile, CO - 3.4 g/mile, NOX-
1.5 g/mile.  A secondary environment of HC-0.41 g/mile, CO-3.4 g/mile,  NOX-0.4
g/mile was also considered.
  A literature survey was carried out covering existing light duty diesel work and exper
 ence throughout the world.  This indicated that the diesel engine was a viable power
 >lant for light duty use.
  An engine configuration study was then carried out when 9 diesel engines developing
 96kw were designed.  The diesel types covered were indirect and direct injection two-
cycle and four-cycle engines;  rotary and compound versions were also included. Two
gasoline engines were also outlined for comparison purposes.
  A method of rating the various power plants was devised and this was applied to the 9
diesels and the 2 gasoline engines. It was concluded that although the gasoline engine
was slightly superior to the diesel engine for passenger car use many applications with
an emphasis on fuel consumption and durability would give an equal or better rating to
the diesel.
  Only the indirect injection four-cycle diesels were capable of meeting the primary
 emissions target with current technology but they did not require any cataly.st.s to do thi:
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
EXHAUST EMISSIONS
DIESEL ENGINES
POWERPLANT RATING METHODOLOGY
ENGINE DESIGN
LITERATURE REVIEW
19. DISTRIBUTION STATEMENT
RELEASE UNLIMITED
b.lDENTIFIERS/OPEN ENDED TERMS
LIGHT DUTY VEHICLES
LIGHT DUTY ENGINES
DIESEL/GASOLINE
COMPARISON
EMISSION CONTROLS
FUEL ECONOMY
19. SECURITY CLASS ( ilns Re/,,'ril
UNCLASSIFIED
20. SECURITY CLASS (This page!
UNCLASSIFIED
c. COSATI Held/Group
-
21. NO. OF PAGE'S
22. PRICE
EPA Form 2220-1 <9-73)

-------