THE STUDY OF LOW EMISS ION VEHICLE
POWER PLANTS USING GASEOUS WORKING FLUIDS
FINAL REPORT
REPORT NO. SR-20
AUGUST 1972
THERMO MECHANICAL SYSTEMS COMPANY
7252 REMMET AVENUE
CANOGA PARK, CALIFORNIA 91303
COPY NO.
-------
STUDY OF LOW EMISSION VEHICLE
POWER PLANTS USING GASEOUS WORKING FLUIDS
FINAL REPORT
REPORT NO. SR-20
)
AUGUST 1972
By
H. W. Welsh
J. L. Harp, Jr.
R. A. Yano
T. P. Oatway
C. T. Riley
L. Nawroczynski
Prepared For The
Environment Protection Agency
Office of Air Programs
Advanced Automotive Power Systems Division
Robert B. Schulz, Project Officer
Contract No. EHSH 71-003
Thermo Mechanical Systems Co.
7252 Remmet Avenue
Canoga Park, California 91303
-------
ABSTRACT
This report presents a preliminary technical study of low
emission vehicle powerplants which use gas as the powerplant working
fluid.
The scope of the program concerns evaluation of the thermo-
dynamics and preliminary design of several cycles which include but are
not necessarily limited to:
l.
2.
The External Combustion Piston Engine
The Closed and Open Brayton Cycle Engines
The Ackeret-Keller Cycle Powerplant
3.
4.
5.
The Stirling Cycle Engines
Evaluation of Rankine Cycle and Other Closed
Cycle Working Fluid Hazards
This report includes consideration of those factors which affect
the general suitability of the powerplant to the automotive type vehicle,
and a comparison of the overall relative metits of the several power
systems.
i
-------
1.0
2.0
3.0
4.0
5.0
6.0
5.6
5.7
TABLE OF CONTENTS
SECTION
PAGE NO.
INTRODUCTION
SUMMARY
1,
2
RECOMMENDATIONS
EXTERNAL COMBUSTION PISTON ENGINE (ECPE)
7
8
4.1
4.2
8
10
Introduction
Principle of the ECPE
Control Operation
46
52
4.3
4.4
Heat Transfer Analysis
Automotive Engine Performance
62
78
4.5
4.6
AutOmotive Design and Installations
General Characteristics
88
91
4.7
4.8
Conclusions
GAS TURBINE (BRAYTON)
5.1 Introduction
92
92
5.2
5.3
93
119
General
Closed Cycle Gas Turbine
Open Cycle Gas Turbine
137
179
5.4
5.5
Single-Shaft and Free-Turbine Analysis
Comprex
193
198
Conclusions
STIRLING ENGINE
Introduction
201
201
6.1
6.2
6.3
6.4
6.5
6.6
6.7
principle of the Stirling Engine
Drive Mechanisms
202
207
General Performance
Automotive Engine Performance
215
228
Automotive Engine Design
Conclusions
245
257,
ii
-------
7.0
8.0
9.0
10.0
RANKINE AND OTHER CLOSED CYCLE WORKING FLUID HAZARDS
7.1
7.2
Introduction
Hazards Associated with Closed Cycle Working Fluids
Vehicle Accidents
7.3
7.4
Fluid Flammability Hazard
Fluid Toxicity Hazard
7.5
7.6
Hazards of Candidate Closed Cycle Working Fluids
Conclusions
7.7
7.8
Attachments
POWERPLANT COMPARISONS
8.1
8.2
Introduction
8.3
Comparisons
Conclusions
REFERENCES
APPENDICES
I.
OAP Vehicle Design Goals and Road Load Characteristics
for the Six-passenger Automobile
iii
260
260
260
260
261
263
265
266
269
318
318
318
327
329
335
-------
1.0
. INTRODUCTION
Both the gas turbine and the Rankine cycle engines are included
in the EPA Advanced Automotive Power Systems Program because of their
potential for achieving low exhaust emissions that meet Federal 1976
standards.
The prime reason that these engines merit gerious consideration
is that they use a continuous combustion system in which the combustion
process can be controlled to produce relatively low values of exhaust
emissions.
These emissions inelude carbon monoxide, unburned hydrocarbons,
and oxides of nitrogen.
The achievement of the Federal 1976 oxide of
nitrogen standards for the internal combustion gas turbine has not been
demonstrated, but encouraging results have been reported.
The Rankine
cycle engine, which utilizes an external combustion system, has demon-
strated in combustion rig tests that this engine can achieve Federal 1976
emission standards.
The gas turbine or open Brayton cycle engine has the potential
of achieving low emissions.
The Rankine cycle engine is an external
combustion engine but there are other engine thermodynamic cycles based
on gaseous working fluids that utilize potentially low emission external
combustion which also merit investigation.
This program includes a
study of both the open Brayton cycle and the following closed cycle
heat engines:
1.
2.
Closed Brayton cycle
Stirline engine
3.
4.
External Combustion Piston Engine
Ackeret-Keller cycle.
Factors considered during the evaluation of the above cycles included
efficiency (at full and part load), power plant weight and volume, oper-
ating characteristics, cost hazards, etc.
An additional objective of the present program was to evaluate the
potential hazards associated with organic Rankine cycle and candidate
gaseous cycle working fluids.
The combustive, explosive, and toxic
characteristics of these fluids were compared to similar hazards now
common to existing motor vehicle power plants.
1
-------
2.0
SUMMARY
This final report presents the results of an analysis and design
study of closed and open cycle gaseous working fluid powerplants and an
evaluation of their suitability for use as low emission vehicle power-
plants.
The primary areas of investigation and the conclusions drawn
therein are as follows:
1.
An in-depth study of all Brayton cycle concepts which hold
potential as automotive powerplants was performed. The first
step in the Brayton cycle investigation was to determine
the relative merits of open versus closed cycles.
For the
closed cycle, working fluids (helium, argon, freon, etc.)
otherthan air were evaluated and, in addition, the complex
Ackeret-Keller cycle was also investigated with varying
degrees of intercooling and reheat.
In general, the closed
cycle gas turbines did not appear as feasible as the open
cycles due to their lower efficiencies, increased complexity,
and added cost.
A thorough investigation into state-of-the-art turbine technology
was then undertaken to determine the turbomachinery components
and associated characteristics which were best suited for an
automotive application.
It was found that any automotive
gas turbine powerplant will be limited to certain maximum
temperatures and pressures as determined by the availability
and cost of refractory alloys.
The number and sizes of compressor,
turbine, and heat exchanger components must be kept to a minimum
to reduce the very important initial cost.
The best powerplant
package was determined to be a single stage radial compressor
and single stage radial turbine with a maximum compressor pressure
ratio of about 7:1 (depending upon heat exchanger effectiveness)
and maximum turbine inlet temperature of 1800oF. The issue
of single-shaft versus free-turbine wag found to allow consider-
able trade-offs in such areas as weight, size, cost, and oper-
. "
ational characteristics.
For this reason, either form is
feasible, although the single-shaft seems to have the advantage
with respect to initial cost and "amount of high temperature
Eaterial required.
2
-------
Having determined the best combination of turbomachinery for
the automotive powerplant, detailed cycle analyses were per-
formed.
It became immediately clear that the simple cycle with
a single stage radial compressor and maximum turbine inlet
o
temperature of 1800 F had unacceptable part load fuel consump-
tion.
This led to the analyses of gas turbine cycles which
use various types of heat recovery systems.
Both regenerative
and recuperative heat exchangers were examined, with the pJate-
fin type recuperator appearing to be the superior type of unit
with respect to size, cost, and complexity, although the reliabil-
ity and expected life of this type of heat exchanger have not
been proven.
The tubular recuperator, though reliable, was
shown to be somewhat bulky and expensive, while the c~ramic
or metallic regenerator, though comparable in size and perfor-
mance to the plate-fin recuperator, was determined to be more
complex because of possible seal and bearing problems.
Further cycle analysis with various plate-fin recuperators
revealed that the optimum recuperator size depends on the
available powerplant space and the associated heat exchanger
manufacturing cost.
An exhaust gas bypass arrangement was evaluated in combination
with the various types of heat exchangers.
Bypassing the heat
exchanger at higher engine power levels gives a more compact
less costly unit, since the heat exchanger is sized for low
power operation only.
such an arrangement can provide the
power of a medium pressure ratio non-recuperative cycle for
acceleration and high speed operation, while retaining the
efficiency of a low pressure ratio recuperative cycle at low
road load power levels.
The study showed that the bypass
concept offers advantages with respect to size, weight, and
cost while giving somewhat poorer fuel consumption than could
be obtained with a larger heat exchanger designed for full
power operation without the bypass feature.
In addition, a preliminary investigation was also directed to-
wards determining the potential of the "Cornprex" as the gas
3
-------
1___--
generator unit for a small free-turbine powerplant.
A liter-
ature search and preliminary cycle calculation shows that this
concept holds potential as a successful automotive powerplant.
2.
The External Combustion Piston Engine (ECPE) is a new and un-
usual powerplant concept which is intended to improve the
emissions of conventional reciprocating piston engines in the
power class which is needed for automobiles.
The concept is
centered on the use of an external combustor which transfers
the energy of combustion to a closed cycle reciprocating
engine through an external heat exchanger. Open cycle oper-
ation was investigated originally and shown to require excessive
displacements to produce the required horsepower (160 BHP) .
Various closed cycle working fluids (helium, argon, freon, etc.)
other than air were also investigated; however, the performance,
availability, and cost factors were such as to make air the
most attractive.
Optimization studies were conducted to
determine the best combination of cycle parameters with close
attention being paid to such factors as material cost, material
availability, stress limits and design sophistication.
prelim-
inary cycle analysis showed that a 267 CID, 4 stroke engine
o
with maximum closed cycle temperature and pressure of 1300 F
and 1500 psia with a pressure ratio of 10 would produce the
required ~60 BHP at a design speed of 3600 rpm. A conventional
six cylinder 250 CID Chevrolet engine was used for prototype
layouts, although a somewhat heavier expander will be required
for the development program to withstand the maximum 1500 psia
cylinder pressure.
This base engine was modified with a slight
increase in cylinder bore and a new head modified for the
four valves which are required to establish the hot and cold
air flow paths.
The pressure ratio and maximum cycle temper-
ature were such as to make regeneration of the closed cycle
air unattractive. Maximum combustion products temperature
was limited to 25400F to reduce both the burner NO emissions
x
and the amount of critical metals in the combustion heat
exchanger.
4
-------
Control system operation, heat exchanger designs, and engine
characteristics, as a function of speed and power, were deter-
mined and used to evaluate engine performance under all power
demands.
Acceptable efficiency and fuel economy was obtained
for the ECPE powered vehicle throughout the normal operating
range.
The ECPE powerplant engine with all auxiliaries and
accesso~ies was found to fit in the engine compartment of a
medium size vehicle (1970 Chevrolet Caprice).
The ECPE powerplant was evaluated with respect to cost,
hazards, and various other economic and social considerations
and shown to be a potential solution to the air pollution
problem.
3. A very thorough literature survey of past and present effort
on the Stirling engine was the first step in determining the
feasibility of this engine as an automotive powerplant. The
information obtained was analyzed and coalesced into a form
from which the most feasible design was constructed.
Various
working fluids were investigated with hydrogen proving to be
the most efficient and helium the least hazardous.
Weight and size considerations indicate that the Stirling
engine may not be a very practical automotive powerplant
even with its capability for high efficiency and low emissions.
Major design problems must be solved to produce a suitable
automobile engine.
4.
A study was conducted to evaluate the potential hazards of a
number of candidate closed cycle working fluids for automotive
powerplants.
The study considered such characteristics as
toxicity, cOmbustibility, and susceptibility to explosion, and
compared the proposed closed cycle working fluids with fuels
which are now in common automotive use.
In general, it was determined that any closed cycle system re-
quiring a toxic or flammable working fluid should not be
considered unless the total quantity of fluid is sufficiently
small so as to minimize the hazard.
5
-------
5.
The final effort in this study was a comparison of the various
powerplant concepts investigated, in addition to a comparison
with other existing more conventional powerplants (Otto, diesel,
Wankel, etc.).
The various powerplants were compared with
respect to weight, cost, complexity, full and part power
efficiency, driveability and, of prime importance, emissions.
The comparisons have indicated that:
(1) the gas turbine,
Rankine, ECPE, and other concepts being considered will require
further development work to prove them acceptable automotive
powerplants; (2) the Stirling engines appear to be less
attractive than other contending powerplants; and (3) based
on demonstrated emission levels and suitability for automotive
application, various versions of the internal combustion
engine have significant leads over other proposed powerplants.
6
-------
3.0
RECOMMENDATIONS
Based on the results of the present program, the following recom-
mendations are presented:
1.
Further effort should be directed toward developing the medium
pressure ratio (6 or 7:1) plate-fin type recuperative gas
2.
turbine powerplant.
Additional effort is recommended to determine the feasibility
of using a "Comprex" type gas generator unit for an automotive
free-turbine powerplant.
3.
Due to similarities in size, cost, complexity, etc., the ECPE
and Rankine cycle engines may be considered as equally feasible
low emission automotive powerplants.
Sufficient design and
development work should be carried forward on the ECPE to bring
it to a stage where a more authoritative comparison can be made
4.
with the Rankine and other powerplant concepts.
The cost and complexity of the present stirling engine make it
impractical for an automotive powerplant.
The use of hydrogen as the
working fluid, higher engine speeds, and a barrel engine config-
uration can improve the performance of this engine.
Re-evalu-
ation of the Stirling engine for the automotive powerplant
application should be made after recent development details
5.
are made available from Phillips.
Additional effort is recommended on promising variations of the
otto and diesel engines such as the stratified charge concept,
low compression ratio diesels, and variable displacement
6.
engines.
Any system requiring a toxic or flammable working fluid should
not be considered unless the total quantity of fluid is
sufficiently small so as to minimize any hazard.
7
-------
4.0
EXTERNAL COMBUSTION PISTON ENGINE (ECPE)
4.1
INTRODUCTION
The External Combustion piston Engine is an unusual powerplant
concept which is intended to improve the emissions of conventional recipro-
eating piston engines in the power class which is needed for automobiles.
While there has been much progress in reducing emissions from conventional
Otto and diesel cycle engines, and more progress can be anticipated in the
future, progress to date has been achieved at the expense of additional
complexity and it is anticipated that further reductions, which will be
essential to meet future goals, will be achieved at the expense of further
increases in complexity, fuel consumption, maintenance, etc. At the
present time, the intermittent combustion process of the internal combustion
engine produces higher than desire able levels of emissions (CO, CH , NO).
x
While it would be unwise to predict that the emissions of the internal
combustion engine cannot reach the low levels required by current and
future regulations, available data indicates that some form of power system
which involves steady flow, external combustion would be more favorable.
cycles.
Such external combust~rs are most effectively used with closed
The total emissions of an engine of a given power, however, is
the product of the total emissions of the combustor and the overall effic-
iency of the engine.
Thus, it is important that the energy converter
has a high efficiency over a broad range.
The Stirling engine is one form
of closed gas cycle which has demonstrated a relatively flat, high. efficiency
level (on the order of .30-40%) and should, therefore, exhibit very favorable
overall emission characteristics.
While many millions of dollars and many years have been spent on
the Stirling engine, the promising performance characteristics have not
resulted in a production type engine.
The Stirling engine uses inter-
mittent flow heat exchangers and hydrogen or helium working fluid.
It is
necessary to achieve a fine balance between heat transfer coefficients,
fluid friction, and void volume of the heat exchanger to achieve the
desired performance of the engine.
Possibly the Stirling type of engine would be much more favorably
received if it were possible to retain the high efficiency reciprocating
8
-------
compression and expansion processes and replace the intermittent flow
heat exchangers with steady flow heat exchangers.
This would remove the
requirement for special working fluids, and ordinary ambient air could
then be used.
Engine (ECPE).
This is the basis for the External Combustion Piston
It was the objective of the ECPE cycle to eliminate as.many of
the undesirable factors of the Stirling engine as possible while retaining
its favorable features.
High on the list of changes considered was the
development of a concept which would be capable of attaining Stirling engine
efficiency, but which would use ambient air as the working fluid.
This
means that the heat exchanger design had to be changed radically to
eliminate the engine's dependence on heat exchangers which must be so
compact that a hydrogen working fluid is required to maintain acceptable
pressure losses.
To achieve this objective, it appeared most feasible
to design a cycle which used steady flow, or essentially steady flow,
through the heat exchangers so that the required heat can be transferred
with minimum pressure loss.
with a reciprocating system, steady flow can
be approximated by using multiple cylinders so that the fluctuations are
minimized.
In addition, in small sizes, such as engines in the 160 HP class,
the reciprocating compressor (and expander) is usually more efficient than
the corresponding turbomachinery units.
It is not unusual for small recip-
rocating compressors to attain over 90% efficiency, but it is unusual for
a small turbine or compressor to achieve over 80%, and more often, the value
is on the order ot 70-75%.
A turbocharger compressor for a 160 HP diesel
engine, for example, attains an efficiency of about 72%.
The reciprocating
compressor and expander thus has a distinct advantage from the overall cycle
efficiency standpoint for the ECPE cycle.
In its basic form, the cycle
is similar to the classical diesel cycle, and the total efficiency is of
the order of 20-25% at maximum cycle temperature of about 13000F using air
as the working fluid.
Both regenerative and non-regenerative versions have
been studied; however, as will be shown, the results indicate that regener-
ation is of marginal or no value for the range of pressure ratios and temper-
atures under consideration.
Several working fluids have been investigated
and while a monotomic gas such a Argon appears to be favorable from the
9
-------
efficiency standpoint, the overall availability and performance of the
cycle using air was found to be more favorable in the practical sense.
The literature shows that there have been attempts to design
reciprocating open Brayton cycle engines (or modifications thereof).
They have always failed, however, because such concepts have always
used one cylinder (or set of cylinders) for compression and an equal
number of cylinders for expansion.
In all cases, extreme design sophis-
tication is essential to cool the expansion pistons and cylinders due to
the exposure. to the hot gas.
No successful engine has been developed,
although there are many patents and many designs which have been tried.
However, in the case of the ECPE design, both compression and expansion
take place in the same cylinder as is the case for the conventional
internal combustion engine.
The cylinder wall is heated during the
expansion stroke and cooled during the compression stroke.
To control
the process events, four valves are used per cylinder and the cycle
operates on the 4-stroke principle.
Modifications to the valve events can
be made to convert the ECPE classical diesel cycle to a reciprocating
Brayton cycle.
The intake valve could be maintained in the open position
during part of the compression stroke so that the compression ratio would
equal the expansion ratio thereby increasing the cycle efficiency.
The
work per cycle would decrease, of course, but the power can be increased
by increasing the average gas pressure.
In addition, a hybrid version of the basic cycle can ',use the classi-
cal diesel cycle at high power and convert to the Brayton cycle for higher
efficiency at lower power requirements, or the intake valve timing can be
modified to provide any cycle between the two.
This can be used as an
alternate control method compared to that conventioaally used in closed
gas cycles (control of base cycle pressure) .
Since broad power and torque
range is important for applications such as automobile powerplants, the
intake valve method of controlling the cycle output may have some advant-
ages which merit investigation in a later program.
4.2
PRINCIPLE OF THE ECPE
The Extexnal Combustion piston Engine is unique in that the com-
pression and expansion processes take place in the same cylinder while the
10
-------
heat addition and removal processes are conducted external to the cylinder.
Thus, the engine retains the well established reciprocating engine character-
istic of using the cooler gas of compression to cool the cylinder structure
which was heated by the expansion process.
The ECPE cycle can be designed
for either open or closed cycle, and with or without regeneration.
The concept of the ECPE is illustrated in Figure 4.1 for the closed
cycle version.
In both the open and closed cycles, the engine machinery
is similar, and Figure 4.1 can be used for descriptive purposes.
Cold
gas is inducted into the cylinder through the intake valve as the piston
moves downward.
This valve is closed at or near bottom center and the gas
is compressed to some value which is determined by the overall cycle thermo-
dynamics and is discharged from the cylinder through valve no. 2.
The gas
then flows through the regenerator (if there is one) and through the com-
bustor heat exchanger or burner, adding heat to the working fluid.
expansion valve (no. 3) then opens and the piston moves downward.
The
This
valve remains open for part of the piston downstroke, and the point at
which it closes is determined by the overall cycle thermodynamics.
Once
this valve is closed, the gas in the cylinder expands to the end of the
stroke.
The exhaust valve (no. 4) then opens and the cylinder gas "blows
down" to the exhaust pressure.
This valve remains open as the working
fluid remaining in the cylinder is discharged while the piston moves to
top center.
If there is a regenerator, the exhaust gas flows through it
before being discharged to the atmosphere (open cycle) or the waste heat
exchanger.
Figure 4.2 shows the basic P-V diagram for the ECPE cycle.
The working fluid is drawn into the cylinder from 0' to 1, compressed from
1 to 2, forced out of the cylinder from 2 to 2', heated at almost constant
pressure and returned to the cylinder from 3 to 4, expanded within the cylin-
der from 4 to 5, blown down to base pressure from 5 to 6, and forced out
of the cylinder from 6 to o.
The difference in pressure levels between 0
and 0', and 2' and 3 is due to valve losses and heat exchanger pressure
drops.
At this point it was necessary to determine whether the open or
the closed ECPE cycle was the most feasible as an automotive powerplant,
keeping in mind the fact that the closed cycle has the potential of using
various gaseous working fluids other than air.
various monatomic and
11
-------
EXhau8t"t~ases
Combustion Air
P P-i P-T.
r---
I
. i
I
I
-_J.
Q&lw~ent
air
'/Iorking
Air Heater
~
-c-J
Control Valve
Atmospheri "
Inlet
Air stora'Je
I[Q:.r2 ]h'
1
2
Open Loop - - - -Combustion Air
Closed Loop Working Fluid
o
3
4
",
I
I
J
j
FIGURE 4.1
Diagram of External Combustion piston Engine (ECPE).
l~
-------
r.--~
I ~
p
2
4
6
1
v
FIGURE 4.2.
Basic p/V diagram for the External Combustion
Piston Engine (ECPE).
13
1012 a
-------
<-
diatomic gases were reviewed as possible ECPE working fluid candidates,
and the following are chosen as having the greatest potential:
1. Air (open and closed cycle)
2. Argon (closed cycle)
3. Helium ( closed cycle)
4. Hydrogen (closed cycle)
5. Freon (closed cycle)
Freon was used for comparative purposes inasmuch as it is not capable of
withstanding the peak cycle temperature used in the analysis.
For this
preliminary investigation, assumptions of the closed cycle parameters
incl uded: peak cycle temperature of °1540oF, peak pressure of 1000 psia,
pressure ratio of 10, and minimum cycle temperature and pressure of 1400F
and 100 psia respectively. For the open cycle the peak pressure was 147
psi a and cycle inlet conditions of 850F and 14.7 psia were considered (all
other parameters comparable).
At this point in the analysis cycle parameters
(T ,P ,etc.) were chosen which would give relatively high efficiencies
max max
in a reasonably sized powerplant while taking into consideration material
limitations of reciprocating engine components.
It will be shown later
that heat exchanger limitations will change the cycle parameters as pre-
sented here; however, such changes do not affect the general results of
the working fluid analysis given here.
Figures 4.3 to 4.6 present ideal P-V diagrams for the closed air,
argon, helium and hydrogen ECPE cycles, plotted using logarithmic scales.
The cycle calculations were performed assuming isentropic expansion and
compression with no valve, heat exchanger of pumping losses.
In short,
the work output and efficiencies were determined on an ideal basis to be
used for comparison purposes only, not to be used as engine performance
capabili ties.
For computing working fluid mass flow rate and engine size
(CID) , an indicated horsepower requirement of 250 HP was assumed. The
absolute ~~lue of this horsepower requirement is unimportant as the results
here are to be used for relative comparisons only.
The results of the cycle calculations for the various gases are
tabulated in Table 4-I.
The open air cycle requires an engine with a
displacement of .1471 cubic inches which eliminates it ~s a possible
contender for an automotive type powerplant.
Of course, decreasing the
14
-------
6.0
o
.ro!
+J
~ 4.0
Q)
H
::J
en
en
Q)
H
Po.
T 5" 13800R
2.0
15.0
!llB.
10.0
A.O
1.0
0.1
Volume (Fraction of initial)
FIGURE 4.3
External Combustion Piston Engine
Closed cycle, ideal pressure-volume
diagram for air.
15
1029
-------
15.0
10.0
8.0
6.0
°
oM
....
~
GI
~
III
III
GI
...
a..
4.0
2.0
1.8.1
FiGURE 4.4
ARGON (k .. 1.67
T2-15120R T4D 20000R
0.2
0.4
0.8
0.6
Volume (Fraction of initial)
External Combustion Piston Engine
Closed cycle, ideal pres8ure-volume diagram
far argon.
16
TS =-95g0R
Tl-6000R
1.0
1030
-------
o
.~
+J
~
Qj
~
::3
1/1
1/1
Qj
~
~
15.0
10.0
8.0
6.0
4.0
2.0
1.8.1
FIGURE 4.5
HELIUM (k m 1.66
2=14980R
'I4'" 20000R
0.2
0.4
Volume (Fraction of initial]
External Combustion piston Engine
Closed cycle, ideal pressure-volume diaqram
for helium.
17
o
T Szs971 R
Tl=6000R
1.0
1031
-------
15.0
HYDROGEN
T2D 11580R
10.0
o
..-4
+I
~ 4.0
cu
1-1
~
II)
II)
cu
~
~
T )13130R
2.0
1.0 .
0.1
0.2
0.4
0.6
Tl=6000R
1.0
0.8
Volume (Fraction of initial)
FIGURE 4.6
External Combustion piston Engine
Closed cycle, ideal pressure-volume diagram
for hydrogen.
13
1032
-------
TABLE 4-1
EXTERNAL COMBUSTION PISTON ENGINE WORKING FLUID CYCLE CHARACTERISTICS
Tl T2 T4 T5 Ideal CID
CYCLE PI P2 P4 P5 Cycle for 250 IHP Mass Flow Ideal Cycle
VI V2 V5 Work at Rate Efficiency
V4 BTU/lb 3600 RPM Ib/min.
in3
AIR 6000R 11300R 20000R 13800R
(Closed Cycle) 100 psia 1000 psia 1000 psia 231 psia 90 253 119 .396
2.22 ft3jlb .420 ft3/1b .744 ft3/1b 2.22 ft3jlb
ARGON 6000R 15120R 20000R 9590R
100 psia 1000 psia 1000 psia 159.5 psia 34 484 313 .560
(Closed eycle) 1. 613 ft3 lIb .406 ft3/1b .537 ft3/1b 1.613 ft3 /lb
HELIUM 6000R 14980R 20000R 971 oR
100 psia 1000 psia 1000 psia 162 psia 350 472 31 .554
(Closed Cycle) 16.10 ft3/1b 14.025 ft3JJtb 5.373 ft3/11 16.10 ft3 /lb
HYDROGEN 6000R 11580R 20000R 13130R
(Closed Cycle) 100 psia 1000 psia 1000 psia 218 psia 1220 267 9 .407
31.96 ft3 lIb 6.17 ft3/1b 10.64 ft3/1 31. 96 ft3 /lb
FREON 6000R 7820R 20000R 16980R
(Closed cycle) 100 psia 1000 psia 1000 psia 240 psia .35 154 305 .202
0.525 ft3/1b 0.0685 ft3/1 0.175 ft3/ b 0.525 ft3/1p
5450R 10410R 20000R 14200R
AIR
14.7 psia 147 psia 147 psia 38.2 psia 95 1471 112 .377
(Open Cycle) 13.73 ft3/1b 2.62 ft3/1b 5 . 04 f t 3 / Ib 13.73 ft3/1b
-------
pressure ratio would increase the work output and reduce the required
displacement; however, decreasing it enough to give a reasonable size
powerplant would produce an unfavorable cycle efficiency. Hydrogen
requires a displacement of 267 cubic inches and provides an ideal thermal
efficiency of 4~%, values which are essentially the same as calculated
for the closed air cycle, Inasmuch as there is a fire hazard with hydro-
gen, this gas it not considered a likely working fluid for the ECPE cycle.
The two monatomic gases, argon and helium, provide a significantly improved
ideal efficiency of about 56%. However, the engine size increases to 472
cubic inches with helium and to 484 cubic inches with argon. Of course,
just as for the open air cycle, the displacement can be reduced by reducing
the pressure ratio; however, this will lower the efficiency to the point
where there is little advantage over the closed air cycle, even with suitable
regeneration.
Furthermore, since there is a very limited supply of helium,
it does not appear to be a strong candidate for the ECPE.
In addition,
the use of Argon will require a large resupply source for the working fluid
(and the associated distribution problems) thereby eliminating the advantage
of using readily available ambient air. Therefore, a closed cycle using
air as the working fluid was chosen for further ECPE analysis and design
studi.es.
Parametric studies to determine the best combination of inlet
pressure, pressure ratio, maximum system pressure, and maximum system
temperature on engine size, efficiency, and power output were then
initiated. The maximum system temperature was limited to 13000F by heat
exchanger metal temperature limitations and the need to avoid the use of
high alloy stainless steels (Hastelloy, etc.). The maximum system pressure
was limited to 1500 psia, a level intermediate between conventional Otto
and diesel cycle maximum cylinder pressures. The mini~um closed cycle
temperature was ~650F and the waste heat exchanger is designed to control
o
this temperature under all operating conditions on an 85 F day. In
addition, it was decided to design the engine for a maximum rotational
speed of only 3600 rpm to reduce friction losses and excessive engine
stresses. Finally, the peak brake horsepower requirement at the design
speed was calculated to be ~60 BHP (see Appendix A) .
Figures 4.7 and 4.8 show the results of these parametric studies
for the closed ECPE air cycle.
since the maximum pressure (1500 psia)
20
-------
500
Q
H 400
u
~
c:
Q)
~
U
III
~ 300
0. 1=100 psia
1/1
-Pi
Q P1=150 ps'a = 12
1
P2/P1=10
200
30
De ign 1=100 psia
P2/ = 12
1
1=150 psia
c1P = 10
25
Q)
~
III
~
IJ:I
~
20
1600 1700 1800 1900 2000
MAXIMUM CYCLE TEMPERATURE - oR
FIGURE 4.7
Effect of Maximum Cycle Temperature on Engine
Displacement and Brake Cycle Efficiency.
21
1165
-------
30
* No In et Pressure Affects for Range
of 10 to 150 psi.
dP
25
..
Q) Regen.
~
H
SQ
~
20
500
P1 - pSia
100
o 400
H
()
,
+I 125
~
Q)
rii
u
lIS 300 150
r-f
p..
I/)
.,.j
0
200
6
8
10
12
PRESSURE RATIO, P2/Pl
FIGURE 4.R
Effect of Pressure Ratio and Inlet Pressure on
Brake Cycle Efficiency.~nd Engine Displacement.
(T 4 = 1760oR.)
22
14
1166
-------
I----~-'
and temperature (1300oF) limits were defined, it was merely necessary to
determine the best combination of base pressure level and pressure ratio
which gives the best combination of engine size and efficiency.
It can
be seen that a base pressure of 150 psia with a pressure ratio of 10 gives
a medium size engine (267 CID) with a relatively high brake cycle efficiency
(25.5%).
It should be pointed out that this efficiency does not take into
account the combustion loop heat losses and that it applies only at the
design point.
As expected, Figure 4.7 shows that the highest efficiency
and smallest powerplant is obtained at the highest temperature, and there-
fore there is no question here but to use the maximum allowable, or 1300oF.
Figure 4.8 shows that to use a higher pressure ratio than 10 has the
effect of increasing the efficiency slightly and increasing the required
displacement (for 160 BHP) quite significantly.
Therefore, the more
detailed total efficiency calculations will now be presented for the ECPE
closed air cycle using base pressure level of 150 psia and a pressure ratio
of 10 as the design point parameters.
4.2.1
Cycle Analysis
Figures 4.9 and 4.10 are the P-V and T-S diagrams for the closed
air cycle ECPE concept at the design point.
Point a represents the air
as it is just about to be inducted into the cylinder through the intake
valve (no. 1).
As the piston moves from top center (TC) to bottom center
(BC) on the intake stroke, the air is drawn into the cylinder (0-0'-1)
with a resulting pressure drop due to losses across this valve.
This
pressure drop was estimated to be about 15% from valve pressure drop and
volumetric efficiency information, and will be discussed in detail in the
next section.
This pressure drop is handled by merely letting the pressure
at a be .15% higher than the desired induction pressure at 1, (in this case
P = 176 psia, P , l = 150 psia) .
00,
1 lb. of air was chosen as the amount inducted during the intake stroke.
Therefore, for induction conditions of 150 psia and 6250R we have:
R T~
- -
P~
To simplify the cycle calculations,
v =
1
(53.3) (625)
(150) (144)
=
1,548 ft3/1bm
and
23
-------
ECPE p- ' " T ' CYCLE DIAGRArviS
I . .
II . ...
160 HP DESIGN POINT CHARACTERISTICS
. . .. .
1600 21 2000
2
- 3~., 4
:>
1600
cu.
'U; 1200 0::
c.. 0
a. 1000 t-.. 1200
..
IJJ IJJ
0:: 0::
tv => =>
~ (f) 800 t-
(f) «
IJJ 0::
0:: IJJ
a.. a. 800
0:: 600 :2:
LL.I I.1J
0 t- 0
z
- 0::
-I «
>- 400
(.)
5 400
200 6
Of 1
0 0
0 0.5 1.0 1.5 2.0 AIR SPECIFIC ENTROPY
CYLINDER VOLUME, V - cu. ft. sru/ Ibm oR
FIGURE 4.9 FIGURE 4.10
1241
-------
ft-lbf
R = air gas constant (Ibm OR)
v = air specific volume (ft3/lbm)
V = cylinder volume (ft3)
The work done by the gas on the piston during the induction stroke is:
V = 1. 548 ft3
1
Where:
PI vI
WO'-l = J
- (144) (150) (1.548)
- 778
= 43.0 BTU/lbm
Where:
W = work (BTU/lbm)
J = mechanical conversion factor
(778 ft-lbf/BTU)
The air is then compressed to the maximum system pressure (1500
psia) from state 1 to 2, and this compression process is assumed to occur
at about 98% isentropic efficiency (T[ s). This value was estimated from
information available on reciprocating compressor design technology.
Therefore, for a 10 to 1 pressure ratio we calculate*:
T 2s - T 1 (: ~) k ~l
= Ha60R
k = ratio of specific heats for air = 1.386
Wl-2 = u2s - ul
Tis
=
(u2 - ul)
101.90 BTU/lbm
u = air specific energy (BTU/lbm)
where
and
u2 = Wl-2 + ul
T2 = .l198oR
v = RT2 = .295 ft3/lhm
2 -
P2
V = .295 ft3
2
When the pressure of the gas reaches 1500 psia (or the volume
reduced to .295 ft3), the compression valve (no. 2) opens, allowing the
high pressure air to flow through the combustion heat exchanger.
The
characteristics of this valve will be discussed in the next section, and the
reason that no regeneration is employed will be brought out at the end of
*Values of u, h, Cp, and k used in the cycle calculations are taken from
Keenan and Kayes Air Tables.
25
-------
this analysis. The piston continues doing work to TC pushing the gas out
of the cylinder (from 2 to 2'). This work is equal to:
W2-2' P2 V2 (144) (155) (.295) = 82.04 BTU/lbm
= =
J 778
When the air has been pushed out of the cylinder and through
the combustion heat exchanger, it will be forced back into the cylinder
through the expansion valve (no. 3) as the piston moves from TC to BC.
The temperature of the gas will be at the maximum cycle temperature
(1760oR); however, the pressure will not be at the maximum pressure
(1500 psia) since the air has experienced pressure drops through valve
nos. 2 and 3, and through the heat exchanger.
The pressure drop across
the heat exchanger (2"-3) was found to be about 2% (see ECPE heat ex-
changer section), and the pressure drops across valve no. 2 (21-2") and
valve no. 3 (3-3') were calculated to be 3% and 4% respectively. The
pressure drops across these valves will also be discussed in more detail
in the next section.
The pressure at position 3' and 4 is, therefore,
1365 psia, and the amount of heat added through the combustion is:
Qin = hl760 - h1l98 = ~48.21 BTU/lbm
h = air specific enthalpy (BTU/lbm)
Valve no. 3 will remain open until the 1 lb of air at 1365 psia
and l7600R has been inducted into the cylinder or until the cylinder
volume is:
v - RT4 - (53.3) (1760)
4 - P4 - (~365) (144)
3
v4 = .477 ft
3
= .477 ft /lbm
Before the valve closes, the. gas will have done work on the piston
equal to:
P4V4
W 3 1 -4 = J
= (144) (1365) (.477)
778
= 120.57 BTU/lbm
After the valve closes the gas expands, again with 98% isentropic efficiency
( II e)' to position 5. The thermodynamic conditions at this point are:
3
v5 = vI =1.548 ft /lbm
26
-------
"
........
3
Vs = VI = 1.S48 ft
TSs ~ T4 (~: ) k-l
= 11S4oR
The work done by the gas on the piston is:
W4-S = (u4 - uSs) x
T( e = u4 - Us = l1S.S BTU/lbm
and
Us = u4 - W4-S
T = 116SoR
S
Where:
~ is the isentropic expansion
e efficiency = .98 (chosen as an
average value for the given
expansion pressure ratio, S)
The pressure at position S is:
P - RTS - (S3. 3) (116S)
S - V S - (1. S48) (l44)
= 278 psia
At this point in the cycle the piston is near BC and valve no. 4,
the exhaust valve, begins to open.
The high pressure gas in the cylinder
immediately "blows down" to the discharge pressure (state S to 6). This
----.
discharge pressure is lS% higher than the initial induction'pressure (+.5S%
due to waste. heat exchanger pressure loss), or:
p = P6 = l76 psia
discharge
The complete blowdown process is best described by referring to the
T-S diagram of Figure 4.11.
The air in the cylinder, before valve no. 4
opens, is at state i, in the sketch.
The "first particle" of air to leave
the cylinder expands from state i to state 2 and accelerates through the
exhaust port (this is shown as an isentropic process in the sketch) .
After
entering the discharge line, the air dissipates its kinetic energy at the
constant discharge pressure (Pd) along path 2-3-4. Note that during the
short time required for the particle to execute this process, the flow
is essentially a steady throttling process so that the initial and final
enthalpies are equal (hi = h4) .
A subsequent particle will expand
to state I, using its energy as change of
,/
with:n the CY1~der from state i
vol~e work in forcing the previous
/
27
-------
h
T
FIGURE 4.11
FIOUM 4.12
I
I
4
ir-------
1
--
2
S
T-S Diagram of Slowdown Process
r----'"
I ,
) 0 I ,
, i I
I I
I
L.------,
Initial State
t
r----'"
I Df-'-:"_.:
I b a I
I t'--.-.-.-..i
',------U' .
Pd
o
Final State
Sketch of Initial and Final Slowdown Conditions
28
1075
V-'",; " ';C='. ~
-------
particles from the chamber.
It may leave the cylinder at state 1, for
example, and accelerate through the discharge port to state 2.
Its
velocity at this point will be lower than that of previous particles
since its energy potential for accelerations is only (hI - h2)' This
will dissipate its kinetic energy in the discharge line to a final state
of enthalpy h3 - hI' a lower value than previous particles, due to its
lower kinetic energy. All the air which has not left the cylinder when
the process is complete (when the cylinder and discharge line pressures
are equal) will have followed the process i-2 and have a final enthalpy
h2' having given up their energy as change of volume work, and not having
been accelerated. Thus, all air particles having left the cylinder will
have a mean energy level lying between h4 and h2' whereas all air particles
remaining in the cylinder will have the same final state, h2' Figure 4.12
shows the state-of-the-air before and after the b1owdown process.
The
temperature of the fluid remaining in the cylinder after blowdown (Tb) is:
Tb " Ti (:: ) k~l
10290R
Where:
Pi = Ps = 278 psia
T. = T = 116SoR
1. 5
Pa = Pb = P6 = 176 psia
V = 1.548 ft3
b
The mass of the fluid remaining in the cylinder after blowdown is:
~=
Pb Vb
R Tb
- (144) (176) (1. 548)
- (029) (53.3)
= .715 Ibm
Now, the energy equation for the blowdown process is:
Irreversible Work = Initial Energy - Final Energy
PaVaMa =.M ui = (Maua + ~~)
Where:
M = Ma + ~ = 1 Ibm
~ = 82.29 BTU/lbm
u. =
1.
107.21 BTU/lbm
or
M
a
h
a
= ~ ~ = M \\
.285 (h ) + .7l5 (82.29) = (1) (107.21)
. a
h = 169.6 BTU/lbm
a
T = 10960R
a
29
-------
Therefore, the temperature of the air pushed out of the cylinder is 10960R
and the temperature of the air remaining in the cylinder is 10290R.
. Having completed blowdown, the piston begins its upstroke in the
cylinder, pushing the remaining air at state b out of the cylinder (6-0).
The work done by the piston on the gas is:
P6v6 - (144) (176) (1.548)
W6-0 = -y-- - 778
The final mixed state of the gases after the blowdown and exhaust
= 50.7 BTU/lbm
process is:
Ma ua + ~ ub = M6 u6
(2.85) (94.50) + (.715) (82.29) = u6
u6 = 85.76 BTU/lbm
T6 = 10480R
T6 is the mixed final exhaust temperature which would be available
for regeneration. The temperature at state 2, however, is higher than this
temperature, therefore making regeneration impossible.
It should be pointed
out here that regeneration may have been feasible if a lower pressure .ratio
or a higher maximum cycle temperature had been chosen. The maximum cycle
temperature was limited to l300oF, however, due to heat exchanger consider-
ations.
Due to this somewhat low maximum temperature, regeneration would
not be feasible unless the pressure ratio was dropped below 8 (see Figure
4.8) and even then the cycle efficiency is several points lower than that
which can be obtained with the present design parameters.
Thus, the air flow through the waste heat exchanger transferring
to the coolant air heat equal to:
Q6-0 = h6 - ho = l03.6 BTU/lbm
The air is now at its original state, point 0, and is again ready
to repeat the cycle process.
The net ideal cycle work is:
W) = (+43.0) + (-lO1.9) + (-82.04) + (+120.57) +.(+llS.52) + (-50.7)
net ideal
= 44.5 BTU/Ibmx l Ibm = 44.5 BTUs
therefore, for 1 Ibm of air, the cycle gives 44.5 BTUs of. mechanical work
(ideal) .
The average frictional loss for internal combustion reciprocating
engines (at 3600 rpm) is about 15% of full load power (this will be expanded.
in the performance section).
Therefore,
30
-------
W) - 44.5 x 85% = 37.8 BTU/lbm
brake
and W)brake
11.brake = 37.8 25.5%
= =
Q in 148.21
The required air flow rate and engine displacement to produce 160 BHP
at full load (3600 rpm) are:
g = (160 BHP) (42.44 BTU/min) = 180 Ibm/min.
(37.8 BTU/lbm) (1 BHP)
(180 Ibm/min) (1. 548 ft3/1bm) (1728 in3)
CID = -
(1800 cycles/min.) (ft3)
In the above closed ~ycle analysis and in the following open cycle
= 267 in3
analysis, the only heat losses included are those from the waste heat
exchanger and the exhaust gases from the recuperator.
The heat losses from
these areas will "account for most of the heat lost from the engine since
these areas act at the only significant heat sinks from both the open and
closed loops.
Direct heat losses to both the coolant and the surrounding
air have been neqlcct.ed.
However, due to the low average operating temper-
ature of the ECPE and to the insulation which will be required on the hot
air path, these direct heat losses should be kept to a minimum and should not
significantly affect the cycle performance calculations.
Having analyzed the closed loop air system, it is now necessary to
analyze the open loop combustion system to determine the overall powerplant
performance characteristics.
The characteristics of the control system and
the heat exchanger designs presented in Sections 4.3 and 4.4 will be used
here and their derivation saved for those sections.
It will be shown in
the performance section (4.5) that it is possible to retain high levels
of efficiency over a broad range of vehicle operating conditions with the
chosen power control system.
The basic components of the ECPE cycle are illustrated in Figure 4.13.
The basic function of the waste heat exchanger is to remove waste heat from
the closed cycle; however, it also assists in warming the ambient air before
it passes to the recuperator for additional heating.
In effect, this pre-
heating of open cycle air in the waste heat exchanger reduces the heating
requirements of the burner.
Both the recuperator and the combustion heat
exchanger are designed for an effectiveness (~) of .75 at 75% of design power.*
*Design power is defined to be ~60 BHP at full throttle (150 psi base
pressure and full speed, 3600 rpm) .
31
-------
T ::0 6250R
cl .
Engine
Block
t
~
E
=11980R =
T a17600R
dJ
Td
=104S0R
~mo
r-
I
I
-....--
FIGURE 4.13
Diagram of ECPE Basic Cycle and Temperature History
at 75' ~8ign Power.
IT
I Ol~
= 30000R
p:
o
en
~
\0
r-I
II
Combustion
8°
r-
I
~-
I
. ~-
I
I
,
t..._L_L_..J
Heat
Exchanger
--
E= 0.75
Waste
Heat Exchanger
I
I
Open Cycle
Ambient Air
32
--,
I
I
I
I
I
I
+10'
Recuperator
E = 0.75
= 13970R
-~
Q)
r-I+J
U If)
-.. (; ~
.c:
cl1
~
~....
o
\0
0\
en
II
'"
o
E-<
tr .=6450R
1°2
I
. °
T =545 R
I 01 (85°F)
. I
1179
-------
For the ECPE performance analysis it was assumed that the heat transfer
effectiveness, combustion efficiency, and the burner outlet temperature
(30000R) will be constant for the entire engine operating range.
In reali ty ,
however, heat exchanger effectiveness values will increase as the flow rate
through them is reduced, and for all power conditions corresponding to less
than 75%, the performance analysis is somewhat conservative.
Accordingly,
heat exchanger effectiveness will be somewhat less than design for higher
power conditions, but automatic transmission shift characteristics (the full
throttle shift point occurs approximately at 75% of design speed) essentially
limit the operation of the engine to less than 75% power except for the very
high speed regime (over 100 mph) .
completely in Section 4.4.
This effect will be developed more
The above heat transfer characteristics were used to calculate the
engine efficiency and fuel consumption at full power.
Referring to Figure
4.13, the subscripts "c" and "0" refer to the closed cycle and open cycle
respectively. Initially, the open cycle (ambient) air enters the waste
heat exchanger (WHX) at a temperature (TOl) of 5450R. This air is heated
as it passes through the WHX to approximately 6450R, and is directed
through an air cleaner to the recuperator.
Before the heating effect of the recuperator on the air can be
determined, the temperature of the hot combustion gases into it (or similarly,
the temperature of the gases out of the CHX) must be determined.
Using the
75% effectiveness value for the CHX as determined in Section 4.4, we have,
(CHX = .75 =
i'
o -
4
T
o -
4
T
05
T
c2
or
T = 3000 = .75 (3000-1~98)
05
= 16480R
with
Now the temperature of the air having been heated in the recuperator
75% effectiveness is given by,
':C. T
= .75 = 03- 02
T T
05 - 02
(
recup
33
-------
or
T = 645 + .75 (1648-645)
03
= 13870R
and similarly the temperature of the hot combustion products leaving the
. recuperator is:
T = 1648-.75 (1648-645)
06 = 8960R
o
Now since the combustion temperature is equal to 3000 R (constant),
the temperature rise across the combustor is 30000_13970 = 16030R. The
ideal fuel-air ratio required to provide this temperature rise (assuming
constant pressure combustion and a fuel lower heating value of 18,700
BTU/lb, from Ref. 4). is .0265.
ratio becomes .0268.
Assuming 99% combustion efficiency, this
It has been shown previously that 148.21 BTU/lbm of heat is trans-
ferred in the CHX per cycle, and that at full speed, full power, the flow
rate is 180 1b/min.
Therefore, the heat transfer rate is:
QCHX = 148.2l BTU/lbm x 180 Ibm/min.
= 26,677 BTU/min
Knowing this and the temperature of the combustion gases into and
out of the CHX, the mass flow rate of the combustion products (g ) can
cp
be determined at full power as:
o
m
cp
=
Qtransf
C (T -T )
P 04 05
- 26,677
- (.2975) (3000-1648)
C = specific heat of the
p combustion products
= 66.3l lbm/min
and knowing the fuel air ratio = .0268, the mass flow rate of the air
o 0
(ma) before the burner is 64.58 Ibm/min. and the flow rate of fuel (mf)
is 1.73 lb/min.
Therefore, the fuel heat input (Qf) at full power is:
o
Qf = mf (LHV)
= (1. 73) (18,700)
= 32,360 BTU/min
and the overall efficiency of the combustion loop (~CL) is:
34
-------
'1)/ -- Qtransf
II x 100
CL Qf
26,677 x 100
32,360
=
= 82.4%
Finally, the total system efficiency is the product of the cycle
efficiency and the combustion loop efficiency or:
II total = (.255 x .824) x 100
=21%
and the brake
specific
o
mf x 60
BHP
fuel consumption at design is:
BSFC =
(1. 73) (60)
160
= .649 lb/BHP-hr
The pressure and temperature history as a function of crank angle
for the reference cycle is plotted in Figure 4.14.
The analysis to deter-
mine the crank angle at which valves no. 2 and no. 3 must open and close
will be presented in the following section.
4.2.2
Valve Analysis
A detailed valve analysis is necessary so that the correct valve
pressure drops can be incorporated into the ECPE cycle analysis.
Since
there are four valves in the ECPE cycle, a simple estimating procedure
for the valve pressure drops could result in large errors.
A preliminary
timing sequence for the valve events has been estimated and is shown in
Figure 4.15.
To understand the mechanics behind this diagram, one must
first understand the role played by each of the valves in determining cycle
performance.
The following discussion will examine the important character-
istics of each of the valves.
ward.
The operation of the intake valve (no. ~) is relatively straightfor-
On the intake stroke the piston moves from top center (TC) to bottom
center (~C) drawing a new charge of air into the cylinder.
To obtain the
maximum amount of work from the cycle, it is necessary to take in as much
air as possible during this stroke.
Therefore, the valve is opened slightly
35
-------
W
0\
1600
1400
1200
1000
lIS
..-j
!II
a.
..
IV
~
;j
II)
!II
IV
~
~
~
..-j
IC(
SOO
600
400
o
200
J
o
o
T.C.
2
/2
/
/
---
FIGt.:RE 4.14
1
180
B.C.
360
T.C.
Crank A:.gle, degree
54J
B.C.
20:;0
a::
o
1600 ~
Q)
H
;j
+J
lIS
H
Q)
1200 a.
=
Q)
E-t
~
" .....
0<
" ....~O
'"
Working Fluid (Air) Te~perature-Pressure Histor] for the ECPE Cycle.
00
o
720
T.C.
12'A
-------
0.3
o.~
I-t
QJ
~
Q)
s
~
.,..j
c:.
w)
-..JI.H
.,..j
H
0.1
TC
L = ":::.1 ve
D = ':::.1 ve
o
'.
FIG','?'':: 4.15
i.f~ {in.)
.':'~-:-.eter (in.)
I,ta;-:e
\' lye
:: . 1
8J
BC
160
TC
ornpressicn
alve
Jo. /
24J
'< '.""
-'... -'
400
C=an~5taft Rotation,
Ex~a"1sio:-
Valve
:.;". 3
"'-t=.J
::.eg:!."'ee3
BC
T'C
560
Prelimi~ary Valve Tirninq Se~~e~ce Ciagrarn for Cne C=~~lete EC?E Cycle.
Ex:-,aust
Valve
No.4
640
I .G -
-------
before TC (100) and closed slightly after BC (300), giving a total open
angle of about 2200. (The detailed analysis of the valve overlap angles
will not be attempted here.
combustion engine values.)
The angles used are merely typical internal
Next, as the piston travels from BC to TC,
the fresh charge of air is compressed to a pressure and temperature as
determined by the cycle thermodynamics.
Referring to the reference cycle
calculations of the previous section, the pressure, temperature, and
specific volume of the air at the maximum compression point are 1500 psia,
o 3
1198 R, and .295 ft Ilbm.
To hold these conditions at state 2, valve no. 2,
the compression valve must open when the piston is .20 (= V2/Vl) of its
stroke from TC.
To determine the crank . angle at which valve no. 2 opens requires
knowing the piston position as a function of crank angle.
This relation-
ship and its derivation are shown in Figure 4.16 where the piston was
assumed to move with simple harmonic motion. The crank angle corresponding
to X = .20 is about 580 BTC. Valve no. 2 will remain open until just past
TC (about 50) at which time valve no. 3 begins to allow the high temper-
ature air back into the cylinder.
It has been assumed that about 10% on
each side of the lift curve is a zero flow region. Therefore, the total
opening of valve no. 2 including the zero flow region is about 750.
Now valve no. 3, the expansion valve, will have started to open
slightly before TC (5°) and continues to open as the piston moves downward,
allowing the high temperature air to fill the cylinder.
When the original
mass of air has again filled the cylinder, valve no. 3 will close, allowing
the air to expand for the remaining portion of the downstroke.
For the
reference cycle, the properties of the air at the required closing position
are T4 = 17600R, P4 = ~365 psia, and V4 = .477 ft3llbm. From Figure 4.16
this gives a crank angle at X = .31 (= V4/v5) of 700 ATC. Therefore,
including the 10% zero flow regime on each side of the valve event, the
total open angle for valve no. 3 is about 900.
Now that valve no. 3 has closed, the gases expand in the cylinder
until BC at which time valve no. 4, the exhaust valve, opens.
Actually,
this valve opens slightly before B€ so that the blowdown process is
completed before the piston starts its upward stroke.
Valve no. 4 then
remains open during the exhaust stroke to allow the spent air to be pushed
38
-------
~
+I
> s:::
QI
~ ~
+I U
-rl f1I
U .-I
o 0.
.-I III
QI -rl
:> CI
~*
xt
T.C.
V
1.0
B.C.
V
H
200
.8
.6
.4
.2
o 0
80
120
. 160
.40
Crank Angle
degrees
o V t}
x .. ( 1-c~se)- t
V a sin e -(0.262 ns)
X a distance from TC (In.)
v .. piston velocity (fps)
s - stroke (in.)
n .. rpm
FIGURE 4.16
Diagram Showing Piston position and Velocity as a
Function of Crank Angle.
3"
1078
-------
out of the cylinder and through the waste heat exchanger. It closes
o
slightly after TC (20 ) to allow complete removal of the spent gases from
the cylinder, giving a total open angle of about 2200. Valve no. I, the
induction valve, is again opened about 100 BTC to allow the new charge
of air to purge the spent air from the cylinder and to provide the fresh
charge for the next cycle.
Now that the operation of the four valves has been discussed, it
remains to determine the appropriate sizes and the resulting pressure drops.
Taylor (Ref. 1) uses a term called the Mach Index to determine the volu-
metric efficiency through valves of various lifts and diameters.
Mach Index (Z) is defined as:
This
2
Z = (biD) x (Sic. a)
~
Where:
b = piston bore, inch
D = valve diameter, inch
S = average piston speed, fpm
C. = average flow coefficient
~
a = speed of sound = 2940~. fpm
s
T = static temperature, oR
s
The average flow coefficient used in computing Z was obtained by
integrating the flow coefficients at various LID ratios over the schedule of
valve lift versus crank angle.
Taylor shows that lifting the valve more
than one quarter of its diameter gives diminishing returns in flow capability.
Therefore, a maximum LID ratio of .25 was used for all valves (as shown in
Figure 4.15) and the average flow coefficient for this lift for typical
internal combustion engine valve design was found to be .33 (see Taylor
Volume I, page 176).
Taylor also shows that for the best volumetric efficiency (~v)' and
therefore, the best inlet valve design, the Mach Index should remain below
about 0.5 or 0.6.
Figure 4.17 shows that as Z exceeds 0.6, the volumetric
efficiency falls rapidly.
Therefore, Z = 0.6 was chosen as the Mach Index
for the inlet valve design of the ECPE engine.
The resulting inlet valve
diameter and pressure drop for a proposed 4.0" bore, 3.530" stroke engine
having
full load at 3600 rpm is:
2
Z = (biD) x (siC, a)
~
D = b (S/C.a zrl/2
~
40
-------
0.9
0.8
:>
~
..
:>.
u
~
Q) 0.7
.r!
u
.r!
~
~
W
U
.r!
~.
.jJ
Q) 0.6
!3
M
0
>
0.5
------ """,,.
"
"
"
"
'
''
'
~
'
"
''
"-
.,
.....
0.4
0.4
1.2
1.4
0.0
1.0
0.6
Mach Index,
B a (!?\ 2 -L
OJ C.a
1.
FIGURE 4~ 1?
Volumetric Efficiency Vs Inlet-Valve Mach Index.
41
1080
-------
where:
b = 4.0 inches
S = (3600r~v) (3.53 in) (2
mJ.n stroke
Z = 0.6
C. = .33
J.
a = 2940 x i/62S = 73,500 ft/min
strokes) ( f~) = 2100 ft/min
rev 12 J.n
gives
D = 1.63 inch and from Figure 4.17,
7lv =
.82 at Z = .60.
If all the volumetric losses are assumed to occur across the inlet valve,
then the inlet valve pressure loss is 18%.
The assumption that the entire
volumetric loss is due to pressure drop across the inlet valve is a very
good one for the ECPE engine since very little heating will occur between
the cylinder walls and the fresh air since there is not a large temperature
difference between them.
The volumetric efficiency that Taylor gives takes
into account the higher air heating that occurs in the spark ignition engine.
Therefore, the ECPE volumetric efficiency and the pressure drop are probably
less than 18%, probably closer to about 12%. Thus, due to lack of better
data, and to be conservative, a value of 15% was chosen.
For the compression valve (no. 2), the Mach Index can again be cal-
culated, however, the volumetric efficiency curve of Figure 4.17 cannot be
used to determine the resulting pressure drop.
The pressure drop for this
valve is determined by finding the pressure ratio corresponding to the
calculated Z value from the Mach number tables.
A valve diameter of 1.25
inches was chosen as a reasonable size and the resulting Mach Index calcu-
lation was:
2
Z = (bID) x (SIC. a)
J.
Where:
b = 4.0 inch
D = 1.25 inch
C. = .33
l.
a = 2940 x -V 1200 = 102,000 fpm
The value of S, the average piston speed, is the average speed from
the time at which valve no. 2 opens to TC.
This average speed can be deter-
mined from Figure 4.16 where the piston velocity is plotted as a function
o
of crank angle. Valve no. 2 opens at about 58 BTC for the reference cycle
at which time the piston velocity is about .80 V .
max
42
-------
Therefore, V
P
= S = .4 V
max
S = (.4)(.262)(3600)(3.530)
S = 1340 ft/min.
Consequently,
4.0
Z = 1. 25
Z = .41
2
1340
(.33) (102,000)
From the Mach number tables of Ref. 59, the pressure ratio is 0.0875
or a 12.5% drop across the valve.
This analysis assumed that the entrance
to the valve seat was shart-cornered and that there was no pressure recovery
past the valve.
Reference 3 gives the correction factors (K ,C ) to
c c
account for actual valve design as:
jp = K C ~ P
. actual c c
where:
K
c
C
c
therefore:
/J.P2
--
P.
~n
= contraction loss coefficient = .43
= entrance correction factor
= .50 (Ref. 3, page 87)
(.43) (.50) (12.5) = 3%
For the expansion valve (no. 3), the Mach Index is again used in
conjunction with the Mach number tables. Several sized valves were studied
in an attempt to keep the pressure losses through this valve to a minimum.
The final design calculation used a 1.63 inch valve, the same size as the
intake valve, and gave the fOllowing Mach Index calculations:
2
Z3 = (b/D) x (s/cia)
where:
b = 4.0 inch
D = 1.63 inch
C. = .33
~
a = 2940 xl/1760 = 123,480 fpm
The value of S was determined'in the same manner as for valve no. 2.
Valve no. 3 closes at about 700 ATC for the reference cycle giving:
V = S = 2000 ft/min.
p
Therefore,
2 2000
Z = (4.0/1.63) x (.33) (123,480)
= .29
43
-------
From the Mach number tables of Ref. 2, the resulting pressure ratio
is 0.944 or a 5.6% drop across the valve.
The correction factor (K ) to
e
pressure drop is given in Ref. 3 as:
for the sudden expansion
tJ.p = K 6p
actual e
K = .74
e
therefore:
fj P3
p-:-- - (.74) x (5.6%)
~n = 4.2\
account
The pressure drop across the exhaust valve (no. 4) was not rigorously
calculated due to the fact that the blowdown process includes this pressure
drop, and therefore, it does not directly affect the cycle characteristics.
The only limitation on valve no. 4 is that it be large enough to allow
essentially complete blowdown to discharge pressure before the piston starts
its exhaust stroke.
With the proper valve opening, it was felt that a 1.25
inch diameter valve would be sufficient for complete blowdown.
The proposed valve arrangement is shown in Figure 4.18 for a 4.0 inch
bore cylinder having 4 in-head valves.
The design is symmetrical with two
1.63 inch n inlet valves, and two 1.25 inch D outlet valves.
(These are
standard production Chevrolet valves.)
The valve design criteria given
by Taylor (Ref. ~) was used to determine the minimum spacing between valves
and between the valves and the cylinder wall.
In addition, if the calculated
clearances between the valves is not sufficient then additional area can be
obtained by merely doming the cylinder heads.
4.2.3
Clearance Volume Analysis
In estimating the clearance volume required for the ECPE head design,
it must first be determined whether the piston need be recessed to prevent
interference between the piston and valves.
The sizes and operating charac-
teristics of the four valves has been reviewed in the previous section and
it was deter.mined that lifting any of the valves more than one quarter of its
of its diameter gives diminishing returns in flow capability.
Therefore,
the maximum lift for valves no. 1 and no. 3 (D = 1.63 inch) is .407 inch, and
the maximum lift for valves no. 2 and no. 4 (D = 1.25 inch) is .312 inch.
The valve and piston displacements are plotted as a function of crank angle
44
-------
, -
FIGURE 4.18
For all Valves
Min.Clearance, Cl-0.06 x (Cyl.Bore)
Proposed ECPE Engine Valve Sizes and
Arrangement in Cylinder Head.
45
For Valve-Cyl.Wall
Min. Clearance
C2=0.03 x (Cyl.Bore
1081
-------
in Figure 4.19.
The opening and closing angles for the valves are as deter-
mined in the previous section and both the valves and piston are assumed
to move with simple harmonic motion.
The ECPE powerplant is being developed around a conventional 250
CID L-6 OHV Chevrolet engine.
The standard bore and stroke is 3.88 inches
x 3.53 inches; however, to provide the required power output (160 BHP), the
ECPE displacement needs to be increased to 267 CID.
To accomplish this
increase in displacement, the bore has been increased to 4.00 inches (stroke
remaining constant).
This increased displacement and bore should be within
the tolerab}e bore limitations of the stock engine.
F~gure 4.19 reveals that there will be no interference between valves
no. 1 and no. 4 and the piston for a design having .050 inch mechanical
clearance between piston and head.
However, to keep the minimum .050 inch
mechanical clearance between the valves and pistons (Taylor Ref. 1) requires
recessing the piston about .025 inch beneath these valves.
Valve numbers
2 and 4, however, will interact with the piston and it appears that the
minimum recess required for these valves is about .2118 inch plus the .050
inch mechanical clearance, or .2618 inch.
Therefore, allowing .050 inch
mechanical clearance on the valve diameters, the total clearance volume is:
1.60 in3/cyl. x 6 cyl. = 9.6 in3
% clearance volume = 9.6/276.6 = 3.4%.
Therefore, the total clearance volume in the proposed engine is only
about 3%.
The effects of this clearance volume on the previously calculated
cycle performance data would be quite small and can be considered to be
taken into account in the somewhat conservative value of volumetric effi~
ciency (85%) used in that analysis.
4.3
CONTROL OPERATION
Three power control methods were initially proposed for the ECPE
powerplant.
These were:
(1) control of inlet pressure and pressure ratio,
(2) control of maximum cycle temperature, and (3) control of inlet density.
It was shown early in the study that the cycle efficiency as a function of
power level tor each of the three methods was comparable, and since the
first two methods require a complex variable valve timing system, the third
46
-------
:-.C.
'" ~
.. .'.....
B.C.
. ~ :
,I
"
() J.
.,'
Piston
'!J
'"
~I
QJ
'f1
~ ~:
..-j
"
;.,
U
Q)
:>
o
~
0.4
0.2
/-
I '
/ \
I ',valve I
I \
I '
I \
I \
/ \
I \
", '-
Q)
U
s::
<1J
+J
I/)
'--1
U
0240
160
80
o
80
160
240
Crank Angle, degrees
a)
Valves #1 and #4
0.8
B.C.
r:
.,-1
Piston
,-
,. 0.6
<1J
(J
~~
Q)
'0
r:
.,..j
M 0.4
'.,
()
cu
~-
0
.()
t.,. 2/
Q) 0.2 Valve
u
c:: .
<1J /
+J
(f)
.,..j
Q
0
240 160 80 0 80 160 240
Crank Angle, degrees
b} Valves #2 and 13
FIGUPE 4.19
Valve and Piston Motion as a Function of Crank Angle.
1244
47
-------
method was chosen as the power control system for the ECPE engine.
This
power control system is illustrated in Figure 4.20.
I
The power controller
is envisioned as being similar to the conventional gas pedal in that a
mechanical actuation is used to regulate:
1.
The base pressure in the closed cycle loop
through power valve A, and
2.
The fuel flow and air flow to the combustor.
Additional control auxiliaries to the ECPE powerplant include an
ignition device to start the burner (some type of battery powered glow plug) ,
a starter motor, a burner air and fuel control, an air storage tank to hold
the closed loop air, an air compressor to provide system make-up air, and
a conventional automatic transmission.
The air storage tank is a cylindrical steel bottle designed to hold
1.46 ft3 of air at a maximum pressure of 375 psia. The maximum tank pressure
was defined to be equal to the maximum cycle pressure at the proposed cycle
idle condition, and the tank was designed to provide almost twice the mass
of cycle air required at the full power condition.
The air storage bottle
is pressurized during deceleration by closed loop bleed air passing through
the power valve (A), which is actuated by a deceleration signal from the
throttle pedal. Upon completion of the deceleration to idle, the pressure
in the high pressure bleed line artd the tank should have equalized at 375
psia minus the small drop due to the mass of air which has been lost to the
atmosphere by means of blow-by past the rings and valves.
The mass of air
which remains in the closed loop at idle power is assumed to be lost during
extended engine shutdown periods, while the air in the storage tank is
retained for the next powerplant start-up.
The mass of air lost each time an engine shutdown occurs can be
estimated from the component average volumes, temperatures, and pressures.
Bu summation of the engine, heat exchangers, and manifold fluid volumes,
it was found that the total ECPE fluid volume was 3030 in3. The "low
pressure" side of the engine, including the waste heat exchanger, low
pressure intake and exhaust manifolds, and one-half of the engine dis-
placement, comprises 2170 in3 of the total volume. The approximate mass
of air contained in this volume at idle can be calculated from volume-
averaged pressures and temperatures (37.5 psi a and aOOoR) as:
48
-------
,
,
I
I
,
,
\
\
\
\
,
I
I
Intake
, Pressure
, Sensor
I "'-,
, \
\
\
\
,
\
,
,
,
,
"
Power
Valve
(A)
Tank
Pressure
Sensor
Air
Comp .
One-way
Make-up
Valve
(B)
1,.001> ----
C1.0sed co~o1.----
?~s\).~
----
Engine
Block
I\)
~ I\)
I/J C
I/J
I\)
tt
----
Power
ontroller
'I
\
\open Loop
Air and Fuel
\Control
\
Combustor
Combustion
Heat
Exchanger
Waste
Heat Exchanger
FIGURE 4.20
Schematic of ECPE Power Control System.
.10
1167
-------
P V
(37.5) (l44) (2l70)
M10w siae = R T low= (53.3) (800) (1728) = .l6 lbm.
A similar procedure applied to the "high pressure" side, including
the combustion heat exchanger, high pressure intake and exhaust manifolds,
and one-half of the engine displacement, results in a mass (at idle) of:
(350) (l44) (860)
~igh side = (53.3) (1400) (1728) = .34 Ibm
where:
T = l4000R
P. = 350 psia
therefore, the total mass of air lost during an extended engine shutdown
period is:,
= .50 Ibm - (airmass at ambient conditions)
= .50 - ..l28
= .372 Ibm
The storage tank has been pressurized to contain:
M = P V = (375) (144) (l.46) = 2.72 lbm
tank R T (53.3) (545)
The shutdown loss is therefore seen to represent almost 12% of the total
system mass.
A loss of this amount of air plus an additional :amount of
blow-by past the rings and valves during normal operation must be replaced
by the air compressor during the next start-up.
An air compressor installed
for this purpose will be actuated automatically at idle power conditions
when the tank pressure falls below 735 psia.
Once actuated, the compressor
will supply make-up air to the system through a one-way valve (B) located
on the low pressure intake side of the cycle.
The compressor will provide
approximately 3:1 pressure ratio air from the atmosphere at such a rate
that operation at idle power for 10 second upon starting will resupply
the system with the mass loss and pressurize the tank to 375 psia.
The
compressor requires about 2.0 HP at a maximum volumetric flow rate of
.l9.3 ft3/rnin. to supply this required make-up air.
A description of the operation of the system from "ignition on" to
"ignition off" will 'be useful in describing the function of the individual
components.
It begins with the transmission in neutral, the ignition key
turned to start, and the accelerator pedal depressed.
Current from the
battery
heats the igniter,(glow) plug and engages the starter motor,
which "turns over" the engine.
A preliminary investigation has shown that
50
-------
the power required for the starter should be only slightly larger than
that required for an internal combustion engine. starter.
This is due
primarily to the slightly higher than ambient base pressure, which results
in a higher work of compression while turning over the engine.
The mechanical motion of the power controller pedal then opens the
power control valve (A) to increase the working pressure level.
The
position of the pedal is also sensed by the burner control, which precisel~
regulates the air and fuel quantities through the open burner cycle.
closed cycle air should be brought up to idle conditions within a few
The
revolutions of the engine, and with the glow plug operating, the engine
should sustain idle power.
The air compressor will immediately begin to
supply system make-up air if needed at the defined idle inlet conditions.
When a dashboard light (for instance, a red light indicates that the air
compressor is operating) goes off, the driver can put the car in gear
and begin to accelerate, assured that he has a fully pressurized storage
tank for maximum acceleration requirements.
As the pedal is depressed, the air from the storage tank flows
through the power valve (A) to bring the cycle up to the desired power
level (full pedal will be 150 psia inlet pressure). It should be men-
tioned here that the time lag to go from, say, idle to full power, has not
been fully determined.
The time lag due to increasing the combustion heat
exchanger internal pressure from that at idle power (300 psi a) to that at
full power (1500 psia) has been roughly determined to be under 1 second
(at the idle speed of 600 rpm). However, it is felt that the heat transfer
lag is of secondary inportance.
Since burner operation (including transients)
is being studied by other GAP contractors, it will be necessary in any follow-
on study to combine the ECPE transients with the burner transients to deter-
mine the overall engine performance with respect to such considerations.
At the full power condition, almost 50% of the total system mass remains in
the air storage tank, which would'be sufficient to supply cycle air lost
due to blow-by over extended driving periods or to supply air for a sudden
burst of power above the 100% design power level.
Deceleration is accomplished by release of the pedal, which allows
high pressure air to flow back into the air storage tank through the
power valve (A).
Engine power and vehicle speed are reduced as the closed
51
I,::;
-------
cycle air and the open cycle fuel and air flow are simultaneously reduced
to the desired level.
In addition, during deceleration, the engine acts
as a compressor, storing energy in the high pressure storage tank and
increasing engine braking ability.
Finally, turning off the ignition
switch will extinguish the igniter spark and stop the burner fuel flow,
thereby stopping the engine.
4.4
HEAT TRANSFER ANALYSIS
The three heat exchangers employed in the ECPE cycle (disregarding
the coolant radiator, for the moment) are the combustion heat exchanger
(CHX) , the waste heat exchanger (WHX) , and the recuperator.
These three
components exert a significant influence on engine performance and oper-
ating characteristics, cycle efficiency, engine cost, etc.
Because of
the importance of these heat exchanger elements, considerable effort has
been devoted to establishing the interrelationship between cycle parameters
and heat exchanger designs in order to reduce the required sizes (reduce
weights and costs), improve the shapes (for better packaging in the engine
compartment), and reduce the fan power requirements. The interrelationship
between the open and closed cycle loops and between the various heat
exchanger elements is illustrated in Figure 4.21. The closed loop air flow
path has been reviewed in a previous section and the open loop flow path
is as follows:
Ambient air (B50F) is pulled through the WHX by a fan (not shown in
rigure 4.21) which sits directly behind it and which is belt driven
from the engine.
This fan may be the same fan or a separate fan
from that which is needed for the coolant radiator.
erature after the WHX is approximately l8SoF.
The air temp-
A portion of this heated air is then ducted to the recuperator
inlet, where another fan will provide the required pressure rise
to force the air through the entire open loop combustion system
(including air cleaner, recuperator, burner, CHX, and exhaust
ducting). This fan will be an integral part of the fuel control
system and it is expected that it will be electrically driven to
provide a controlled amount of air-flow for proper air-fuel ratios
and clean emissions. .
52
-------
ECPE COMBUSTION LOOP SCHEIVIATIC
U1
tv
l;> CYLINDER ENGIN~...BLQC.K
00000
COMBUSTIO~
HEAT EXCHANGER
WASTE
HEAT
EXCHANGER
......
tv
.r:.
U1
FIGURE
4.21
-------
Having passed through the air cleaner, the heated air passes
directly through the counterflow, plate-fin recuperator. The
air is heated to approximately 10000F and ducted to the combustor
where it is mixed with fuel and burned to its maximum temperature
o
of 2540 F.
The high-temperature combustion gases then pass
straight through the cross-flow combustion heat exchanger where
heat is transferred to the closed cycle working fluid, these
gases leaving the CHX at about 1200oF. Then the combustion gases
are turned into the recuperator where heat is transferred to the
incoming open cycle air, and finally the gases leave the recuper-
at or and enter the exhaust pipe at about 400oF.
It should be evident from Figure 4.21 and the above flow descrip-
tion that the ECPE heat exchanger designs play an important part in the
feasibility of the entire ECPE concept. The efficiency of the powerplant
is very dependent on the ability of the heat exchangers to transfer the
heat efficiently from the open loop combustion gases to the closed loop
working fluid (air).
In addition, the heat exchanger designs must be such
that they fit comfortably in the available under-the-hood space of the
prototype vehicle.
Therefore, high efficiency and compact size were the
two prime requisites for the ECPE heat exchanger designs.
To facilitate high heat exchanger efficien~y in: a compact package,
it was decided to use 75% of maximum power for heat exchanger design pur-
poses.
This 75% of maximum power was arbitrarily selected as a compromise
between the high speed cruise (40 to 50% power) and maximum speed/acceler-
ation vehicle operating conditions, and results in less drastic heat
exchanger off-design operation.
By designing for 75% of full power, the
heat exchanger material requirements were made less severe, and overall
sizes and costs were reduced.
This was directly due to' the fact that lower
mass flows, pressures and temperatures are required at 75% power than 100%.
~ull power will still be available, however, and this will be possible by
pushing the heat exchanger maximum pressures and temperatures slightly
above design for these short high power periods (less than 2% of time spent
near full power (Ref. .15».
The decision to use 75% of maximum power as
the design point was also motivated by the fact that with an automatic
transmission the engine rpm will rarely exceed 75% of design speed due to
54
-------
design shifting characteristics (except at high road speeds in high gear) .
Additional assumptions and characteristics used in the heat exchanger and
combustion system analyses are:
1.
Waste heat exchanger and recuperator fan air
is 1850p and 14.7 p~ia on an 850p, day.
Normal combustion products temperature is 2540op.
2.
3.
The open and closed loop temperatures remain approximately
constant for all normal operating conditions.
This will
be true except for slight heat exchanger effectiveness
(e) changes as a function of speed and ambient temper-
ature variations.
4.
The burner pressure drop is 3% of inlet.
(The low-emission
5.
external burner is being developed under other OAP contracts.)
The air cleaner and exhaust pipe pressure drop was calculated
to be about 4%.
The combustion loop thermal efficiency (defined as heat transferred
in CHX divided by heat released from fuel) was calculated to be about 82%
at design point.
The detailed open loop calculations for this efficiency
. have been presented in section 4.2.
will now be presented.
The individual heat exchanger designs
4.4.1
Combustion Heat Exchanger (CHX)
The representative CHX and pertinent design data are presented in
Pigure 4.22. It is a bare tube, four pass, cross-counterflow design based
o
on a maximum combustion products temperature of 3000 R and exchanger
design effectiveness of 75%.
The overall dimensions are 7 x 15.5 x 18
inches (H x W x D) with an estimated total weight of about 35 Ibs, based
upon the use of type 347 stainless steel tubes with .0125 inch wall
thickness.
1.
Among the more important of the CHX design characteristics are:
1/8 inch OD tubes with vertical in-line spacing to increase
the air-side heat transfer coefficient (h), thereby
o
lowering the metal temperature (1400 F limit). Harrison
Radiator states that 1/8 inch diameter tubes are realistic,
however, they tend to be expensive.
Future optimization
studies should investigate increasing tube diameter and
55
-------
,
"
Cb
.o~ ~l.i
~ ~~
(J>Q l.i~ oJ'o
'Ooo~~ '.I}
<\>.1 .r",
I-~
.It},,
;/
.,/
./
/"
;/
./
./
;/
/'
./
./
./
./
;/
,/
,/
/'
./
./
,,/
7"
Combustion
Products Out
( 16S00R)
Flow of ~
Combustion
.
Products
,.
.
Deaton Characteristics:
0.125"Tube Dia.
. .
Tube OUtside Diameter = .125 inch
Tube Arrangement ~.Yertical, In-line (see sketch)
Tube Material - .18-S stainless steel6 .0125 inch thickness
Maximum Tube Wall Temperature = 1400 F
Header Diameters, Inlet and Outlet = 2 inch
Estimated Total Weight (frames, shells, headers, etc.) = 35 Ibs.
: Estimated Cost About $160.00
~eratinq Characteristics:
Combustion ~roducts Fiowrate = 50 lb/min.
Combustion Products P.ressure Drop = .6%
Ideal Air Fan Horsepower = 0.3 HP
Closed Loop Air" P'lowrate = 135 lb/min.
~losed Loop Air Pressure Drop = 2.2%
" Effectiveness «() = 75%
FIGURE 4.22
Representative ECPE Combustion Heater Design.
56
1168
-------
using internal turbulators or ring dimpled tubes to
increase the internal h coefficient.
2.
Type 347 stainless steel, which is representative of a
stabilized 18-8 grade, is used in the CHX primarily because
of its known favorable high temperature, corrosion-resistant
brazing characteristics. It has adequate short-time tensile
o . h
strength and ductility at 1200 F, W1t stress rupture pro-
perties high enough for uprating up to possibly 1400oF. It
has adequate corrosion resistance and excellent resistance
o
to oxidation at temperatures up to 1700 F. It is also possible
that with the great strides being made in the use of ceramics
for high temperature applications, much of the CHX casing,
tube supports and manifolds can be made from this material,
thereby greatly reducing the overall cost.
It is unlikely,
however, that the tubes can be made of ceramics due to the
large internal pressures (1500 psia) .
The CHX design is such as to allow easy manifolding between the
burner and the recuperator, and to and from the no. 2 and no. 3 valves
of each of the six cylinders.
4.4.2
Recuperator
4.23.
The recuperator and pertinent design data are presented in Figure
It is a counterflow, strip-fin, plate-fin heat exchanger designed for
an effectiveness of 75%.
The overall dimensions are 9 x 9 x 18 inches
(H x W x D) with an estimated total weight of about 30 lbs based on the use
of a low-alloy steel with .004 inch fin thickness.
AiResearch Manufacturing
Co. has stated that the counterflow design as presented here can be manu-
factured as easily as a crossflow design and with a significant increase in
e,ffectiveness.
Future optimization studies should investigate reducing the
plate thickness (perhaps from .01 inch to .006 inch) and also reducing the
plate spacing.
The recuperator manifolding allows for the exhaust gas to be
ducted toward the ground for easy exhaust pipe pickup and back towards the
rear of the car (the exhaust piping will further lower the exhaust temper-
ature to a safe level).
57
-------
ECPE RECUPERATOR DESIGN
COUNTERFLOW STRIP-FIN, PLA TE-FIN HEAT EXCHANGER
SURFACE 1/4(a) - 11.1 FOR AIR SIDE
SURFACE 3/32 - 12.22 FOR COMBUSTION PRODUCTS SIDE
S \\ ---'
'\.. . I
CO~BUSTION-
r PRODUCTS IN
(1648oR)
DESIGN CHARACTERISTICS:
Matrix material = low alloy steel
Fin thickness = 0.004 in.
Plate thickness = 0.01 in.
Estimated total weight = 30 Ibs.
(frames, shells, ducting etc.)
I
I
I I
I I
I I
I I
II
I,
Ii
U1
00
AIR IN
(G450R)
OPERATING CHARACTERISTICS:
Air flowrate = 48.4 lb/min.
Air side press.drop = 6%
Gas flowrate = 50.0 lb./min.
COMBUSTION PRODUCTS OUT
( 89GoR)
Gas side press. drop = 5%
Ideal fan hp = 7 hp
Effectiveness (E) = 0.75
I-'
N
.c:.
en
FIGURE 4.23
~ AIR OUT
~ - (l474oR)
..
I~TERNAL
ARRANGEMENT
1r- : 0 1"
~
~ ~.25"
-------
4.4.3
Waste Heat Exchanger (WHX)
The waste heat exchanger and pertinent design data are presented in
Figure 4.24.
It is a two-pass, cross-counterflow with plain, plate-fin
surface on both sizes (19.86 fins/inch).
The overall dimensions are 22 x
18 x 4 inches (H x W x D) with an estimated total weight of about 25 Ibs
based on the use of aluminum with .006 inch fin thickness and .01 inch plate
thickness on both sides.
This design was found to be much more effective
and also much less expensive than a finned-tube type arrangement (as used
in automotive radiators) of the same size.
The two pass arrangement on
the working fluid side permits easy manifolding to and from the no. 1 and
no. 4 valves of each of the cylinders.
In addition, the manifolding arrange-
ment to and from the WHX has been designed to allow as much cooling as
possible from the headers and piping.
With finned manifolds it is felt
that at least 10% of the waste heat can be removed in this manner.
As
mentioned previously, the WHX serves the double purpose of cooling the
working fluid and also preheating the ambient air into the recuperator.
One additional point to note is the fact that the ideal fan horsepower
required does not take into account the fact that, under cruising condi-
tions, the WHX will take advantage of the dynamic head due to vehicle
motion.
For instance, for cruising (no acceleration) at 75% of maximum
power, the vehicle will be travelling approximately 95 mph and will supply
almost 80% of the required ideal HP, or the ideal HP required will only
be about 2 HP.
In addition, it has been determined that with the proper
fan arrangement (including a slipping clutch or possibly even an electri-
cally driven fan) and ducting, fan efficiencies on the order of 65-75%
should be obtainable.
4.4.4
Off-Design Heat Exchanger performance
As mentioned previously, the heat exchangers were designed at 75%
of full power and the performance characteristics «( ,LJp) determined at"
that level were used for the entire power range.
This was done to reduce
the number of laborious heat exchanger calculations and, as will be shown,
represents a very conservative approach for engine performance over the
normal operating re9ime.
At power levels below the design point, which
represents most driving conditions (about 95%) the effectiveness will increase
59
-------
ECPE WASTE HEAT EXCHANGER (WHX)
CROSS-COUNTERFLOW, PLAIN, PLATE-FIN HEAT EXCHANGER
19.86 FOR BOTH SIDES
DESIGN CHARACTERISTICS:
AIR IN
( ~45°~J
/
Matrix material = aluminum
Fin thickness = 0.006 in.
Plate thickness = 0.01 in.
Estimated total Weight = 25 lb.
Estimated cost = $35.00 (OEM)
OPF.RATING CHARACTERISTICS:
COoling air flowrate = 524 lb/min
(]I
o
Cooling air sideAP = 1.3%
Working Fluid flowrate=135 lb/mit?2"
Working fluid AP = 0.55%
Ideal fan HP = 9 hp
Effectiveness = 0.82
-I-'
tV
~
-..J
FIGURE 4.24
,-
WORKING
---- -
FLUID OU1
(625°R)
WORKING
FLUID IN
CI005°R)
-------
and the pressure drops decrease, thereby increasing cycle efficiency and
reducing fan horsepower requirements.
In fact, at idle conditions, the
flow rate in both the open and closed loops is only about 5% of that at
design and the resulting fan horsepower requirement is almost negligible.
Also of importance, however, is the performance at 100% which requires
about a 33% increase in both open and closed system flow rates from the
design point (75% power).
that:
The 100% power level calculations have indicated
1.
The CHX will suffer very little in effectiveness (about 2
points) .
This is due to the fact that the working fluid flow-
ing through it is very highly turbulent and an increase in
flow shows an almost corresponding increase in hot side h
coefficient.
The combustion products side is in a transition
flow region and shows a similar increase in the cold side h
coefficient (although not as large).
These h increases tend
to keep effectiveness constant as flow rate increases.
Of
course, the pressure drops on both sides will increase (about
as the 1.8 power of flow rate), however, the design pressure
drops were low enough so that this does not significantly
effect engine performance.
2.
The recuperator will suffer more of a drop in effectiveness
(about 4 points) than the CHX due to the fact that the flow
through both hot and cold sides is in the laminar flow regime
(characteristic of most plate-fin heat exchangers) and an
increase in flow does not change the h coefficient much.
There-
fore, an increase in flow rate decreases the NTU value and
correspondingly decreases f.
This decrease in f will require
a higher fuel-to-air ratio in the burner to hold the same maximum
combustion temperature, which tends to reduce cycle efficiency
(increase BSFC) .
In addition, the pressure drop will go up
almost linearly with the flow rate, thereby increasing the
recuperator fan power requirements.
Thus, at high cycle flow
rates, the cycle efficiency and the power output are somewhat
lower than predicted based on the present recuperator design.
Full power will be available, however, by increasing either the
maximum cycle pressure or temperature above design conditions,
since full power will only be required for short intervals.
6l
-------
3.
The wax will suffer in a manner similar to the recuperator
since it is also a plate-fin type design, although the flow
through it on both sides is more in the transition flow regime
than the laminar. It will experience a slight drop in effective-
ness, however, it is felt that with proper manifolding de~ign
(finned manifolds) that the WHX will be able to provide the
required cooling (closed cycle minimum temperature of 6250~
even at maximum flow. The pressure drops will go up as the
mass flow rates on both sides go up and this will result in
higher losses, especially on the cooling air side.
However,
with a properly designed fan and shroud arrangement, it is
felt that these losses can 'be kept to an acceptable level.
Ambient temperature effects were also investigated as they
directly affect the WHX and indirectly affect the powerplant
performance. An increase in ambient temperature of 200 (85
o
to 105 F) showed up
o
of about 15 :F (from
as an increase in minimum cycle temperature
6250R to 640°R). This increase in minimum
cycle temperature had only a slight effect on cycle efficiency
and power output calculations and it is felt that any further
definitive studies should be saved for actual engine dynamo-
.meter testing.
In summary, it is felt that the heat exchanger designs presented here
should compliment the ECPE powerplant package during the initial prototype
development stage. We are fully aware that they may not be the optimum
, '
designs; however, it is felt that further studies must be guided by actual
hardware testing as the entire field of heat exchanger preliminary design
is quite uncertain until proven in the actual operating environment.
Such
factors as cycle efficiency, power output, fan norsepower, fan noise, etc.
should be considered in such a heat exchanger development stage to optimize
the powerplant for the combination of on and off-design operating conditions.
4.5
AUTOMOTIVE ENGINE PERFORMANCE
The cycle analysis presented in Section 4.2 indicated that a 267 CID
powerplant would be capable of producing the required 160 BHP at the design
speed of 3600 rpm.
The maximum closed cycle temperature and pressure were
62
-------
defined to be 17600R and 1500 psia at the design point.
The total power-
plant efficiency at the design point was found to be 21% with a resultant
BSFC of about .65 lb/Bhp-hr.
It now remains to determine the powerplant
performance characteristics for the entire range of vehicle operation.
Evaluation of the engine power output at part speed and at part load
required an estimation of the valve pressure losses and the frictional horse-
power requirements at these part load conditions.
Pressure losses for valves
no. 2 and no. 3 as a function of engine speed are shown in Figure 4.25 and
were determined by the same method as shown in Section 4.2 for the design
point pressure loss calculations.
Since the volumetric losses are assumed
to occur across the inlet valve (no. 1), and since these losses do Dot
necessarily decrease with decreased eng~ne speed, it was decided to retain
the design point pressure drop of l5% for valve no. 1 at all speeds.
In
addition, the pressure drop across the exhaust valve (no. 4) is included
in the blowdown process and therefore, does not directly affect cycle
calculations (as long as the valve is large enough to allow proper blowdown) .
Frictional horsepower data from a larger Ford engine was scaled down to be
compatible with the requirements for the ECPE powerplant and is presented
in Figure 4.26.
As can be seen, the frictional horsepower increases with
rpm, reaching a maximum of 28.2 BHP or 15% of the design indicated horse-
power at 3600 rpm.
The brake horsepower, brake efficiency, torque, and brake mean
effective pressure of the powerplant were developed from cycle calculations
(with the above losses) as a function of engine speed at power levels of
~oo, 80, 50, 33, and 20%.
Power level, as defined for the ECPE, refers
to the percentage of maximum closed cycle base pressure (150 psia) , which
is not necessarily the percentage of maximum power as a result of various
cycle losses (valve and friction) .
The brake horsepower as a function of
engine speed, presented in Figure 4.27, shows the effect of the increased
frictional losses at high speeds, especially at the low power levels where
the brake horsepower may even decrease as engine speed is increased.
This
is due to the fact that the frictional horsepower requirements increase
at a greater rate than the indicated horsepower.
63
-------
.05
o
o
*Losses fo valves no. I
and 4 ass med constant
for all seeds.
.04
~
.......
~
r-4
~
.01
1000
2000
3000
4000
ENGINE SPEED - rpm
rIGURE 4.25
Valve'Pressure I:Irops va.. Engine Speed.
64
1180
-------
40
~
~ 30
----
).f
QI
)
8.
QI
II)
).f
0 20
:I:
r-4
lIS
c:
0
....
~
u
....
).f
~ 10
FIGURE 4.26
o
o
3000
4000
1000
2000
ENGINE SPEED - rpm
Frictional Horsepower Requirement vs. Engine Speed.
for the ECPE.
65
.p.r31
-------
200
NOTE:
% Powe~
of c10s
pressur
inlet
\ Power Level
40
33
160
~
IX!
120
H
QI
~
&
QI
(I)
H
g
QI 80
~
H 50
IX!
20
1000
2000
3000
4000
ENGINE SPEED - rpm
FIGUPE 4. Z7
Brake Horsepower va. Engine Speed for ECPE Powerp1ant.
66
1182
-------
The engine torque and BMEP as a function of engine speed at full
throttle are shown in Figure 4.28 and illustrate the characteristic of the
ECPE to develop maximum torque for acceleration at very low engine speeds.
This fact makes it possible for the ECPE 267 CID engine to match the perfor-
mance of larger, more powerful spark ignition engines in acceleration runs,
and yet demonstrate the economy that a smaller displacement engine can
provide both in initial cost and fuel savings.
The published maximum torque
@ rpm ratings for several standard 1970 domestic spark ignition engines are
shown to illustrate the favorable torque characteristics of the ECPE power-
plant.
It can be seen that the ECPE provides more torque than comparable
size six-cylinder engines, and provides almost the same maximum torque as
the somewhat larger V-8 engines, although at much lower engine speeds
(where it is most needed) .
The ECPE total thermal efficiency and brake specific fuel consump-
tion, presented as a function of engine speed for the various power levels,
is shown in Figure 4.29 and 4.30.
The reduction in efficiency at higher
speeds is caused primarily by the increased frictional horsepower require-
ments at these speeds.
As mentioned in Section 4.4, the effectiveness of
the various heat exchangers will increase at lower than design flow rates
(75% of maximum) and decrease at higher than design flow rates.
The effect-
iveness (and pressure drop) values used in the analysis here were assumed
constant and equal to the design point values for all engine speeds.
This
is somewhat of a simplification.
However, due to the unknown degree of
error present in other areas of the analysis caused by the rather vague infor-
mation available on such things as valve losses, isentropic compression and
expansion efficiencies, blowdown losses, etc., it was felt that using the
design point values would be sufficiently accurate at this point in the
study.
In addition, since about 95% of all automotive vehicle operation is
below 50% power, the efficiency and horsepower curves presented here are
actually somewhat lower than realizable for the area of operation where we
are most concerned.
As shown in Figure 4.30, the ECPE, unlike most auto-
motive powerplants (both conventional and unconventional), operates most
efficiently at low speeds.
This is due to the fact that the ECPE cycle
losses do not increase at low speeds (such as the Otto cycle with its pumping
losses); in fact, the losses decrease due to the lower valve and friction
losses.
The increased BSFC at lower power levels for the same engine speed
67
-------
320
318 CID
Dodge V-8
<> 302 ID
. Ford V-8
+J
Ij.f
. 280
~
~
fiI
~
240 250 CID In-line 0
(267 CID,6";cy1
6-cy1. evro1et
200
180
160
."j
en
C1I
I
g 140
NOTE: Power Le e1 = 100\
120'0
1000
2000
3000
ENGINE SPEED - rpm
FIGURE 4.28
ECPE Full Power Torque and BMEP vs. Engine Speed
(Compared to Conventional Spark Ignition Engines
Published Max~ Torque @ RPM).
60
4000
1183
-------
~
40
~ 30
>. % Power Level
u
c:
Q)
..-4
U
..-4
~ 100
~ ~
~ I 20
.....
"' 50
~
Q)
.c:
8
..... 33
"'
4-1
g
10
20
00
1000
2000
3000
4000
ENGINE SPEED - rpm
FIGURE 4.'29
ECPE Total Thermal Efficiency V8. Engine Speed.
69
1184
-------
1.6
% Power Level
1.4
80
100
1.2
.
~
.c
I
a.
.c
~ 33
~ 1.0
~
CI1
II:!
0.8 50
0.6
0.4
o
1000
2000
3000
4000
ENGINE SPEED - rpm
FIGURE 4.30
Brake Specific Fuel Consumption vs. Engine Speed
for the ECPE Powerplant.
70
1185
-------
is due to the fact that the friction losses remain approximately the same
as power level decreases (at constant speed), therefore, taking a greater
pe~centage of the indicated power at the lower power levels.
The individual
BSFC curves level out at idle speed (600 rpm) due to the fact that the
engine is considered to be putting out about 5 BHP at idle to handle the
accessory load (see Appendix I).
Increased horsepower can be obtained from the ECPE powerplant (keep-
ing maximum temperature the same) by several methoQs:
1.
OVerloading the present 267 CID engine by increasing
cycle base pressure (up to 180 psia) ;
2.
Increasing the engine displacement;
3.
Increasing the maximum speed (3600 rpm).
The first of these, overloading, may be admissible only for short bursts of
extra power, since the maximum cylinder pressure will reach 1800 psia (at a
compression pressure ratio of 10), and this pressure is considered too high
for sustained operation.
In addition, the air storage tank must be enlarged
to provide additional cycle air (an additional 32 BHP is available with a
tank volume increase of 1.2 ft3) for this overload condition. The second
means of power gain, increased engine displacement, requires a simple scaling
of all powerp1ant components, which, of course, adds to both cost and weight.
Similarly, the third method, increasing maximum engine speed, would require
enlarging many of the cycle components to handle the increased flow rate and
would require a greater degree of design sophistication for the valves, cams,
bearings, etc.
After having evaluated the engine performance as a function of engine
speed, it now becomes pertinent to select the engine drive train components
and evaluate the overall vehicle performance.
A conventional 3-speed auto-
matic transmission with gear ratios of 2.52, l.52, and 1.00:1 and the
corresponding rear axle gear ratio of 2.73:1 were chosen as being compatible
wi th the ECPE powerplant.
The shift characteristics of the transmission
both as a function of power level and engine speed were combined with the
previously shown curves of engine performance to give the vehicle performance
as a function of vehicle speed.
Brake horsepower of the ECPE powerplant as a function of vehicle speed
is shown in Figure 4.31.
Typical shift patterns are shown which demonstrate
71
-------
160
140
120
G: 100
IX!
I / Veh' 1e Power
,..
Q)
) Req irement
8.
Q) 80
I/)
,..
~
Q)
~
lIS
,.. 60
IX!
33
20
00
20
40
60
80
100
120
VEHICLE SPEED - mph
FIGURE 4.~1
ECPE Brake Horsepower vs. Vehicle Speed.
72
1186
-------
.
the tendency of transmission shift points to be a function of both throttle
position and engine speed.
Power available for acceleration is represented
by the difference between full throttle BHP and the vehicle power requirement,
which has been superimposed on the figure.
(The derivation of the vehicle
power curve is presented in Appendix I.)
The vehicle power curve shows that at
a near zero vehicle speed the power requirement is about 5 HP.
The 20% power
level will conveniently handle this power at about 600 rpm, and therefore,
this power level and engine speed have been defined as the engine idle
condition.
As shown in Appendix I, and again here, the maximum vehicle
speed is approximately 107 mph since at that speed the vehicle power require-
ment (including all auxiliaries and accessory requirements) is equal to the
maximum engine output.
The total efficiency of the ECPE powerplant as a function of vehicle
speed is presented in Figure 4.32 where the non-accelerating vehicle operation
curve has been determined from the vehicle power curve of Figure 4.31. Total
efficiency for non-accelerating vehicle operation is shown to range from about
18 to 22%.
This represents a significant advantage over gas turbine and
conventional spark ignition engines, which are generally known to display
poor efficiency at low vehicle speeds.
Several sources have been identified,
including the AiResearch Manufacturing Co. 's mid-term report to EPA, "Auto-
mobile Gas Turbine Optimum Configuration Selection Study", and they. indicate
that load range requirements of 0-15 BHP, corresponding to vehicle speeds
of approximately 0-30 mph, comprise over 60% of most average vehicle
driving time.
This fact points out the importance of good efficiency at
low load for good fuel economy and low emissions.
Next, the brake specific fuel consumption of the ECPE was evaluated
as a function of vehicle speed and is shown in Figure 4.33.
The sarne general
conclusions made concerning thermal efficiency apply to these curves, and
BSFC at non-accelerating vehicle operation is shown to range from .63 to
.76 lb/Bhp-hr.
As in Figure 4.30, these BSFC values are based on 5 BHP
output at idle or zero vehicle speed to run the accessories.
Finally, the steady-speed fuel economy (in high gear) was calculated
for the ECPE powered vehicle and the results are presented in Figure 4.34.
The calculated fuel economy, which was developed from a generally conser-
vative performance analysis, peaks at about 24 mpg at 20 mph, and there-
73
-------
#
.30
%
ower level
~
--...... 100
~ -
r:i .20
H 50
u
H
~
~
~
~ 33
NOTE: ngine opera ion 18 generally
~ .10 hed line.
ove the da
fi 20
o
o
20
40
60
80
100
120
VEHICLE SPEED - mph
FIGURE 4,)2
Total Thermal Efficiency va. Vehicle Speed for ECPE
Powerplant.
74
1187
-------
1.6
1.4
1.2
.
1-1
.c:
I
a.
.c:
~
~ 1.0
u
c..
tf.I
~
0.8
FIGURE 4.33
Powe
Level
0.6
20
0.4
o
40
60
80
100
120
20
VEHICLE SPEED - mph
Brake Specific Fuel Consumption vs. Vehicle Speed~
75
llAR
-------
28
NOTE: In ludes Co stant
24 4 P Access ry Load
~
I 20
~
0
~
(J
~
H 16
~
12
FIGURE 4. J4
8
o
40
60
80
100
20
VEHICLE SPEED - mph
Steady Speed aoad Load Fuel Economy of ECPE
Powerplant.
76
120
1189
-------
after decreases to 12.5 mpg at 80 mph.
Fuel economy of this level is
considered quite good, and will be shown comparable or superior to other
types of automobile powerplants in the powerplant comparison section~
It
is also noteworthy that the ECPE, like other external combustion cycles,
requires less highly refined fuels such as kerosene, which promises to
be an additional benefit in vehicle operational cost.
It should be noted here that the vehicle operation curves in the
previous figures take into account all parasitic, auxiliary, and accessory
losses in addition to the rolling and aerodynamic power requirements (see
Appendix I) .
In short, what has been done is that all power losses other
than engine friction losses have been lumped together in the vehicle power
curve, so that the BHP, as calculated by the cycle analysis of Section 4.2,
represents power available to satisfy all power draws (auxiliaries,
accessories, wind resistance, etc.).
This appeared to be the best method
of presenting the results since the ECPE eng~ne can now be evaluated in
combination with any vehicle by retaining all the engine performance
curves and merely using the vehicle power curve which represents the parti-
cular vehicle.
For instance, additional auxiliaries and accessories can be
added to the ECPE engine and the only effect on the results presented here
will be to shift the position of one curve, the vehicle power curve.
In
addition, the vehicle power curve used here may be slightly lower than that
required at the high speed end since average values of transmission, fan,
accessory, etc. losses were used.
Conversely, the curve may be slightly
higher than that required at the low speed end.
This will tend to improve
performance characteristics for the low speed region and degrade them for
the high speed region; however, as mentioned previously, the low speed
region is of highest concern since the vehicle will spend most of its oper-
ating life in this region.
Finally, the ECPE vehicle performance with the three speed automatic
transmission and associated rear end was checked against the OAF vehicle
design goals. .as presented in Appendix I and found to satisfy all requirements.
This was done by actually determining vehicle acceleration as a function of
vehicle speed for full power operation.
The proper vehicle shift points for
full pedal operation were adhered to and the 4300 lb vehicle was found to
77
-------
have the necessary acceleration to meet all OAP goals. This was done
without taking into account any torque multiplication which would result
from the standard torque converter.
4.6
AUTOMOTIVE DESIGN AND INSTALLATION
Early in the study it was decided that a six-cylinder in-line
engine arrangement would be best suited for the ECPE powerplant due to the
fact that the heat exchangers and associated manifolding could be installed
alongside the block in the available under-the-hood space of a medium
sized vehicle.
A search of the technical literature revealed that the
250 CID IL-6 OHV Chevrolet engine would provide a convenient base powerplant
for the ECPE concept.
The required 267 CID could easily be obtained by
increasing the bore from the stock 3.87 inches to 4.00 inches and leaving
the stroke (3.53 inches) the same.
In addition, preliminary calculations
have shown that there is arnple bore area in the head to accomodate the
four valves (see valve analysis Section 4.2) and therefore engine design
changes would be basically limited to:
the cylinder head, valves, valve
train, manifolding, and fan and accessory pulleys. As will be show, these
changes do not produce any significant changes in engine volume.
Generally, the effort was made to keep changes to the basic stock
engine to the very minimum to reduce prototype costs while not infringing
on the optimum cycle performance.
of the proposed ECPE powerplant.
Figure 4.35 shows the front view section
Due to the number of carn operated valves
(24) and also because of the necessary porting arrangement, the double
overhead earn design was chosen.
The stock camshaft has been left on this
figure due to the fact that it may be required to operate various power-
plant auxiliaries (fuel pump, oil pump, etc.).
The'proposed arrangement
of the cylinder head, roughly indicating valve and port locations, is
indicated in Figure 4.36.
As shown, the valves are 80 arranged that the
same function (i.e., exhaust, intake, high pressure release, and high
pressure return) parts of two adjacent cylinders may be combined into a
single manifolding port. The two sets of high pressure valve ports are
both located on the left side of the engine, in order to make them acces-
sible to the combustion heat exchanger rnani~olding, while the intake and
exhaust valve ports are located on the right.
This valve arrangement and
78
-------
...,
""
...
...
...
...
1'3.27
7.63
1.30
~
1-
6,12
5.15
>.61
-.
10.00
3.88 0tA
3.5301.
11.27
- --, "-.-.-
9,40
7,50
FIGURE 4.35
THERMO MECHANICAL SYSTEMS INC.
CANOGA PARK CALIF.
SCAli' FULL APPIiOIIEO BY.
DATE' 6-16-71
OAA:NNBY~~.
REvIS££):
CHEVROLET 250 CU.IN.- 6 CYL.IN LI NE
STUDY- ECPE ENGINE SECTION, FRONT VIEW
DRAWl""'; HI,),
LO-3IR
- --'-.-..
-------
VI=W ON TH:::
TJ='
'.Z8
,"00
, .
\ I
28.29
~.~
~~~
1-
FIGURE: 4. 6
THERMO MECHANICAL SYSTEMS INC.
CANOGA PARK. CAUF.
, FUll .....,""',
om, 6-17-71
DRAwN 8T~
"EVISED:
CHE\iROU:.T 250 CU IN. - 6 CYL IN LINE
STUOY-- ECPE ENGINE,CILINDER HEAD
LO-32R
- 8C\ -
1115.
-------
the tolerance between valves, and valves and cylinder wall, are shown in
more detail in the valve analysis section (4.2).
The head has been designed
to utilize the standard mounting bolts and tentatively to use the sarne
water cooling passages between the head and block.
The rocker arm and earn housing design, shown in Figure 4.37, indicates
the general approach to bearing support for the carns and valve train.
Cam
lobes have been indicated but are only of a preliminary conceptual nature.
The finalized details of optimum cam shape, shaft and rocker stress, and
bearing loads are oritical but were not calculated at this point in the ECPE
development program.
A very preliminary analysis into such area as valve
seals and piston ring designs was done and it is felt that the technology
is available, when the time carnes for prototype design, to justify the
preliminary configurations presented here.
The importance of acceptable
mechanical designs to the feasibility of the ECPE powerplant is realized,
however, and detailed designs will be of first concern in any follow-on
program.
Next, the belt drive system for the carns and various auxiliaries
and accessories was determined and is shown in Figure 4.38.
This belt drive
system has a separate belt for the two carn pulleys and for each of the fan
and accessory pulleys which are to the right and left sides of the engine.
The carn drive will use a cog-toothed belt similar to that employed on the
Ford Pinto four-cylinder engine and the accessories and fans will be driven
by conventional V-belts and pulleys.
As indicated in the figure, the carn
drive incorporates a dri~ing cog-toothed pul~ey shaft driven by the original
carn drive gear, an idler pulley mounted on the front of the new cylinder
head, and driven cog-toothed pulleys at the front of each of the camshafts.
The only change to the basic engine block involves replacement of the carn
drive gear access plate with a mounting plate for the carn cog-toothed
pulley drive.
All of the drive system are typical of current automotive
production techniques.
After having completed the basic engine design phase, attention was
di~ected towards designing the heat exchapgers to fit comfortably around
the engine in the available under-the-hood space of a medium size automobile.
Actual under-the-hood dimensions were used, taking into account such
restrictions as wheel wells, steering column, master cylinder, etc.
The
81
-------
-----.-
ROCKER ARMS
CYL.Nl6
.;.--
ROCKER ARMS SHAfT ~-- ( !i'f'i' --.----- ------.--
CAMSHAFl l f i(;'-
I L...:-:"
CAMSHAFT Ii.'
ROCKER ARMS SHAFT (.-
TOP VIEW
CAMSHAFT (
".40
,B
1F" - ~
f... I If II '.- I
:! I. I " --\.
!j I I ! I T I L 1 I 3" ' r
~ i -. < c ~ ~~ .. I ~
I I I
'-- - -- -- -_u -_.. --- ,~- '
",.n~-- I I ' ! --- --- ._~- .-- -- .! I I I .-...
r ' I I -
, i I
I i i I i I I i i I i
.. .' .,. .. ,.
! I i I r ; I I I I '
~CIitER AFlatS SHAFT \.
SECTION A-A
--1B
{. CRANKSHAFT
--.--
SECTION B-B
FIc;u!~. 4, 7
THERMO MECHANICAL SYSTEMS INC.
c. ".RK IF
uu.
- -,
--=
STUOY-ECPE EN;II'£.ROCKER ARMS BOX
0tEVR0LET ~OCU'H..-6 CYLIN uN£
to-JBR
- a~
-------
. ,
,
I i
! !
i :
I
I
II
-. --
~~~THED SELf
:l.MSHJ,F~ P'Q
2 PLACES
~
L6El1 CAI GEAR
~~fT~o;62p.n
R STEERING ~~p
/ ! ') . ..
r\+ .
l~-' ..
'-
.-.--
~
~
I
,...
V-BELT 00
CRANKSHA~~LE PllLEY 8,50 P'Q
,...
'....
C t
R.W
-------
heat exchanger designs are presented in section 4.4 and the recuperator-
burner-CHX combination is shown in Figure 4.39.
As will be shown in Figure
4.40, the heat exchanger arrangement was such as to make maximum use of
the available space.
Further heat exchanger optimization studies will be
a part of future ECPE development work, however, it is felt that such
studies will not change the general arrangement as shown in Figure 4.39.
Finally, the complete three-dimensional representation of the ECPE
powerplant installed in a m~dium size automobile is shown in Figure 4.40.
This representation includes all the required accessories, among which
are:
the complete air conditioning system, the make-up air compressor,
the alternator, and the power steering pump.
It can be noted that the
system make-up air compressor has been mounted on a common shaft with the
alternator and that the reserve air cylinder has been located at the back,
right hand side of the engine compartment.
In addition, the power controller
and associated valves, transfer lines, linkage, etc. referred to in Section
4.3 should present no design or packaging problems and can be installed
where. most convenient on the right or left side of the engine.
To
facilitate manifolding, the combustion heat exchanger and the combustor
are mounted along the left side of the engine opposite the cylinder head,
with the recuperator located directly beneath.
The close packaging arrange-
ment plus a certain amount of insulation should keep convective heat losses
from the air heater, recuperator, and manifold to a minimum.
The waste
heat exchanger and radiator are indicated to be located side by side in
front of the engine with a separate fan for each; however, it is felt
that perhaps one fan is sufficient to provide the required air flow for
both.
In fact, analysis of the engine cooling system has indicated that
the ECPE radiator should be much smaller than that required for the stock
250 CID engine due to the fact that the ECPE average operating temperature
is well below that of the typical internal combustion engine.
Future
development work will include a detailed cycle heat loss analysis and will
investigate the possibility of removing the radiat0r altogether through
.proper design of the head to air cool and properly isolate the high temper-
ature no. 3 valve and valve seat.
A preliminary weight estimation for the ECPE powerplant and a com-
parison with that of the conventional 250 CID powerplant is given in Table 4-11.
84
-------
; [
q
I
,i "
-----
c<>~,;'''~
:' .:~ . :il
I
+
,.,+
"""".
(/('--"-"-----'/
,:t !-
I --r-r--f
I
1_-
-'--,-
+
I
,--'-
AtAFILTER
-------
---- ---
- .------
-"--
"-
,--.-----
r- Ct..OSm L.OOP
!11"Pf1£SSl.f>£
. CO~J6US n~N
- .. Irlli.",c~~,
I~;;}
~~f~t~A
,~ - "oiio-;m;~ -j
~b'~~N£ '\, -
-«[~~)
'~J~t n'
- on - - --_m_-_m_-----_----------- onon'J-:=----
RfCUPEAATOA
oPEN LOOP CUCTS
TO COMBUSTOR
85
I
. -I I A .:
- R INAT M T -. f
O£VRa.ET ~OQ.LIN.-eCYLINI..IHf LO-40R I
- - - --.J.Z!L.._-
-------
EXTERNAL COMBUST
: I ~ PISTC I ~ E~ ~
~ :
.
-
L
:..
FIGURE 4.40
-------
TABLE 4-II
PRELIMINARY POWERPLANT WEIGHT ESTIMATION
ECPE 267 CID
6-cyl engine
Engine (including manifolds)
Radiator
600
25
Starter
20
20
Generator
Combustor (plus ducting)
combustion Heat Exchanger (plus ducting)
20
35
Recuperator (plus ducting)
Waste Heat Exchanger (plus ducting)
30
40
Air Compressor (plus pulleys, belts)
Air Storage Tank (plus lines and controls)
20
25
To tal
835 Ibs
Transmission
170
170
Rear Axle and Drive Train
Battery
Fuel and Tank (200 mile range)
40
145
Exhaust
40
50
Miscellaneous
Total
615 Ibs
Total Propulsion System Weight
Vehicle without Propulsion System Weight
1450
2700
Vehicle Curb Weight
4150
87
Conventional
S.L 250 CID
6-cyl engine
400
35
20
20
475 Ibs
170
170
40
145
40
50
615 Ibs
1090
2700
3790
-------
The ECPE pr0pulsion system is estimated to weigh a total of about 1450 lbs,
or about 350 Ibs more than the conventional six-cylinder propulsion system
from which it was developed.
Since a somewhat heavier expander than that
of the 250 CID engine will be required to withstand the full 1500 psi
pressure, the ECPE base engine weight (including manifolds) has been
increased over that of the standard engine. It must be remembered, however,
that the predicted ECPE performance (horsepower, torque) is superior to
most conventional six-cylinder engines and therefore, when compared to a
V-8 of similar performance, the ECPE total powerplant weight should be more
comparable.
Unlike the gas turbine powerplant, the total amount of critical
materials used in the ECPE (nickel, chromium, and cobalt) is not considered
to be a major problem.
In the ECPE essentially all of the critical metal
usage is in the open combustion loop components, including the combustor,
combustion heat exchanger, recuperator (or pre-heater), and their associated
hot manifolding and ducting.
The estimated total weight of these metals
is for nickel, about 4.5 ~bs, for chromium, about 8 Ibs, and for cobalt,
about .02 Ibs.
These estimations are, unfortunately, very rough at the
present time and may vary considerably depending on further optimization
studies and development work on the heat exchangers and burner.
The
critical metal cOntent of the burner, for instance, was determined using
available information from the OAP contractors engaged in burner. develop-
ment work (Solar, Ref. 60).
Since burner development work is still pro-
gressing, it becomes a very difficult task to approximate the content of
critical materials that will be required.
The primary purpose for estimating
the critical material content of the ECPE powerplant was to show that
critical Eaterial availability is not a major problem.
4.7
GENEAAL CHARACTERISTICS
The previous sections have reviewed the principle and design of the
ECPE concept and have shown that it holds good promise as an automotive
powerplant.
The performance and fuel economy characteristics were shown
to be quite good in Section 4.5, and the design of the various system
components in Section 4~6 was shown to be well within the reach of present
technology.
This section will attempt to identify some of the major social
88
. .
-------
and economic considerations applicable to the ECPE which will necessarily
playa major role in its acceptance or non-acceptance as an automotive
vehicular powerplant.
These major social and economic considerations are:
l.
2.
cost
hazards
3.
4.
emissions
development time, costs, and risks.
4.7.1
Cost
The first of these considerations, cost, can only have meaning when
evaluated in comparison to the present internal combustion engine.
For the .
ECPE powerplant, the primary cost difference as compared to the I.C. engine
is in the heat exchangers and the external burner.
Preliminary cost
estimates made for these components (see Heat Exchanger Analysis, Section
4.4) and also for the additional engine equipment (valves, cams, belts,
fans, air compressor, storage tank, etc.) place the initial cost at about
1.5 to 2.0 times the cost of a conventional I.C. engine of comparable
performance.
This value is estimated to represent the cost as compared
to an I.C. engine without emission controls.
With controls, to provide
comparable emissions, the ECPE cost may be favorable to the I.C. engine.
These cost estimates account for the additional equipment required by the
ECPE and not required by the 1. C. engine and, in turn, also account for
the equipment needed for the I.C. engine and not needed by the ECPE.
In addition to initial cost, the maintenance and operating cost of
the ECPE should be comparable to the I.C. engine, especially if emission
control equipment is taken into account.
As mentioned, the troublesome
electrical system of the I.C. engine is essentially eliminated with the
ECPE (.1 glow plug required), therefore eliminating the frequent, expensive
tune-ups characteristic of the I.C. engine.
Also, because of the clean
(combustion-free) air circulating in the engine, carbon deposits should
be eliminated and overall engine life extended.
Finally, fuel costs
should be less due to the fact that the external burner can operate with
less highly refined fuels th~1 the I.C. engine.
89
-------
4.7.2
Hazards
The hazards of the ECPE engine have been investigated and the gener-
al conclusion is that the ECPE poses no additional hazards over those of
the conventional I.e. engine.
It is felt that with proper design and the
necessary safety controls any danger of a fuel or lubricant fire or explo-
sion can be eliminated to the same degree as presently realizable in the
I.e. engine. The proper design will include: (1) high tolerance, high
performance piston rings with the very minimum of lubricant leakage,
(2) the possible use of a lubricant having a higher maximum operating
temperature than conventional automotive lubricants, (3) an oil trap (or
scrubber) within the air storage tank to continuously clean the closed
cycle working fluid (air), and (4) valve seals with a high degree of seal-
ing capability (possibly a labyrinth type seal).
It is felt that the
danger of a lubricant explosion will be further reduced due to the fact
that the crankcase is not pressurized (in fact, it will be continuously
vented to the burner) and also because after each shutdown the closed
cycle air not stored and filtered in the air storage tank will escape
from the system ta~ing entrained lubricant with it.
In addition, the
danger of a fuel explosion is of about the same magnitude as the danger
present in conventional I.e. engines, which is quite small.
Additional hazards, such as excessive exhaust temperatures, high
rotative speeds, etc., appear to be no more or no less a danger than that
present by conventional I.e. powerplants.
4.7.3
Emissions
Recent developments have shown that the type of external burner
used in the ECPE powerplant is capable of meeting the ~976 Federal emission
standards.
The burner development studies are being handled by other OAP
contractors and their most recent test results are tabulated below:
EMISSION STANDARDS (GM!MILE)
1971 Federal ~976 Federal ~eo
Standards Standards Ext. Burner
(test results)
AEROJET
Ext. Burner
(proj ected
.30
CO 47 4.7 .25
UHe 4.6 .14 .2
NO .4 .27
x
.01
.10
90
-------
4.7.4
Development
The development of the ECPE into a road-tested, low-emission vehicle
powerplant appears to be one of minimum lead time, few risks, and relatively
low cost.
Its close resemblance to the present I.C. engine should require
very few retooling changes for the automotive industry.
The production
of the heat Jexchangers should be relatively straightforward, the primary
effort here being to keep material and manufacturing costs to a minimum.
This, however, should be a simple task for the automotive industry, which
has developed the techniques to produce I.C. engines at little more than
the cost of the material used to make them.
4.8
CONCLUSIONS
The preceding ECPE concept study has shown that the ECPE powerplant
could provide a solution to the air pollution problem.
Preliminary design
work has been completed and the next phase will involve further detailed
desiqn and analysis, hardware development, and powerplant installation in a
test vehicle.
Hardware development includes head design and heat exchanger
fabrication (or acquisition), while powerplant installation primarily
includes the acquisition and coordination of the burner, control valves,
air storage tank, auxiliary air compressor, and the various drive belts,
manifolds and supporting equipment.
r
91
-------
5.0
BRAYTON CYCLE ENGINES
5.1
INTRODUCTION
The outstanding success of the gas turbine engine in aircraft appli-
cations has generated significant interest in utilization of small turbo-
machines for automobile powerplants.
In aircraft applications very large
gains were established quite readily in terms of reliability and maintain-
ability characteristics over reciprocating internal combustion engines.
In
addition, the gas turbine engine has demonstrated a potential for low weight
and size, for quiet, smooth operation, good torque at low speeds (for free-
turbine type), and finally, low exhaust emissions.
But despite intensified
efforts in the past decade to demonstrate the advantages of turbine power
for passenger cars, disappointingly slow progress has been made toward the
introduction of a mass produced turbine powered vehicle.
Failure of previous efforts to capitalize on the good features of gas
turbines have been largely economic.
During this period the greatest challenge
has been to show sufficiently superior economics and operational characteris-
tics over the reciprocating engine to persuade the automobile industry to
replace the massive investment in tooling and capability to produce the
conventional reciprocating engine developed over the past 50 years.
However,
with the adoption of stringent federal emission standards for years 1975
and 1976, the position of all new automotive powerplant concepts has been
strengthened and renewed effort has been generated to overcome remaining
difficulties with the automotive gas turbine powerplant.
The most commonly noted detriment to extended application of gas
turbine power is deficiency in fuel economy when it is necessary to operate
a gas turbine at low power settings for extended periods of time (as in
the case of the automobile).
Other areas of concern include acceleration
lag from idle to full power, an increased requirement for air contamination
protection due to its high specific air consumption, and sensitivity to
ambient temperature conditions.
Additionally, gas turbine production requires
large quantities of costly materials of limited availability, close manu-
facturing tolerances, and necessitates an extensive training program to
provide competent mechanics for servicing of precision turbomachinery
components.
92
-------
The objective of this study was to examine the various gas turbine
concepts under consideration for passenger car application and define the
characteristics of these systems so that an overall comparison could be made
with conventional and Rankine cycle systems.
Past and present work in auto-
motive gas turbine development has been reviewed to determine progress to
date, including, among others, the fifty Chrysler produced turbine cars of
the mid-sixties, the Williams Research turbine powered Hornet, and gas
turbine programs sponsored by the Office of Air Programs to study feasibility,
optimize designs, and further the development of gas turbine systems.
Such
efforts in the field render valuable assistance in evaluation of the vehicular
gas turbine powerplant and have provided the direction needed to channel
efforts in the areas requiring improvement.
Both open and closed Brayton
cycle systems were considered.
Various gaseous working fluids were compared
to determine their relative merit in a closed Brayton system.
5.2
GENERAL
5.2.1
COmbustors and Gas Turbine Emissions
Intensified interest in the development of low cost, small gas turbines
for automobile and industrial usage have resulted in significant investments
in component improvement research.
In addition, the impact of emission
standards has required that a great deal of attention be given to gas turbine
combustors.
Attention focused upon combustor development includes both
independent research and research funded by government agencies (such as
the EPA funding of low NO burner development). It is not the intent of
. x
this study to contribute to combustor or other component technology, but
rather to correlate known characteristics in synthesizing a realistic
evaluation of gas turbine systems.
A variety of gas turbine combustor designs are possible, including
the more conventional can type (single or multi) and annular combustors.
A cross section of a typical annular combustor is illustrated in Figure 5.1,
a design which uses several discrete fuel nozzles to inject fuel into the
primary zone as sheets of droplets.
Fuel-rich combustion results and the
primary mixture flows downstream where secondary air is introduced and mixed
with the primary combustion products to reduce the overall mixture strength
93
-------
Compressor
Diffuser
.
Fuel
Ho;z:zle
FIGURF. 5.1
Primary Air
Secondary
Air
.,
Representative Gas Turbine Combustor.
94
Turoine
Nozzle
115'
-------
(fuel-air ratio) to a value of the order of 0.02 to 0.025 at the design
turbine temperature level.
Providing the mixture is more or less uniform,
there is very little free hydrocarbon or carbon monoxide left in the com-
bustion products and, if the design is proper, there are very few particu-
lates.
However, the oxides of nitrogen with this type of combustor can be
. a problem because there is a variable mixture field around each fuel
nozzle, part of which may burn in a high temperature region which for the
maximum flame temperature is slightly leaner than stoichiometric, thereby
forming excessive NO .
x
combustors is on the order of .2 to .3, so there is inadequate time for the
NO to dissociate and the combustion process remains essentially frozen
x
as the gas flows through the turbine.
The through-flow Mach number of these types of
Thus, while the present gas turbine
can have low emissions of hydrocarbons, carbon monoxide, and particulates,
it does have relatively high emissions of the oxides of nitrogen.
Adding to the emissions problem in open cycle gas turbine engines
is the high specific air consumption (about 3 to 4 times that of the con-
ventional otto cycle engine) and the generally unfavorable part load effi-
ciency characteristics of gas turbine engines.
Even with high specific
air consumption, however, the open Brayton cycle can provide low emissions
and, in overall terms, can provide the basis of a very favorable low
emission engine.
Williams Research, for example, has recently emission
tested (in a federal test facility) its 80 horsepower free-turbine engine
in a Volkswagen squareback (53), and the results indicate its ability to
meet 1975 emission goals:
Proposed 1975 Williams proposed 1976
Standard g/mi Gas Turbine g/mi standards g/mi
Hydrocarbons 0.46 0.34 0.14
~~~ Monoxide 4.7 4.5 4.7
Oxides of Nitrogen 3.0 2.15 0.4
These results, although promising, require extensive additional improvement.
Three important characteristics of the open cycle gas turbine
combustor which affect gas turbine performance are combustion efficiency,
total pressure drop, and average outlet gas temperature.
It has been
convenient in preliminary cycle analyses done for this and other gas turbine
studies to make the simplifying assumptions of constant efficiency and
95
-------
pressure drop over the entire spectrum of combustor operation.
Throughout
this study these values have been fixed at 99% and 3% respectively.
Combustion efficiency should always be close to 100% if fuel and
air are well mixed in proper proportions, ignited, and given time to
burn.
In typical aircraft gas turbines the combustor size is critical
and combustion may.be somewhat less than complete, resulting in off-design
efficiency degradation.
The automotive gas turbine combustor, however,
will be required to burn cleanly and evenly over the entire load range in
order to meet stringent emissions standards.
Such an advanced burner would
have a burner efficiency near ~OO% over the entire operating range.
combustion system pressure drop is made up of two components (as-
suming the diffusion of high velocity compressor air is accounted for in
compressor efficiency); the friction loss through the combustion chamber,
and the momentum loss associated with acceleration of the gases to a high
exit velocity.
These losses are related to reference velocity, defined
as the theoretical velocity for flow of combustor inlet air through the
maximum burner cross-section flow area.
Mass flow varies with load, and
static pressure varies in a similar manner, so volumetric air flowrate
or velocity is nearly constant with load.
This characteristic permits the
use of a constant combustor pressure drop for all operating conditions.
Fuel types also influence combustor performance primarily because
of the effect of differences in viscosity and evaporation rate on atomi-
zation and volatility (22).
Performance calculations have assumed air-
craft-type fuels, (JP4, JPS), with lower heating values of approximately
18,400 BTU/lb, although a variety of fuel types are possible.
The materials chosen for manufacture of the combustion chamber are
subjected to the highest temperatures prevailing in the gas turbine.
major requirement of the alloys is good oxidation resistance, ease of
A
fabrication in the desired shape, and satisfactory vibration resistance.
Refractory alloys such as Hastelloy X have been commonly used for high
temperature cycles, but designs utilizing ferritic alloys containing reduced
quantities of cobalt and nickel should be investigated in the interest of
limiting the use of these critical materials to the absolute minimum.
Again,
cost is the overwhelming cr~teria upon which the practicality of the gas
turbine as an automotive powerplant will be judged.
96
-------
5.2.2
Compressors
Advancements in the state of technology of small compressor machinery
have resulted largely from attempts to develop compressors for small indust-
rial and aircraft engines, including new high bypass ratio turbofan engines
in small thrust sizes which necessarily have very small gas generator com-
ponents.
Efforts in these fields have led to significant improvements in
the maximum stage loading capability (pressure ratio) and efficiency of small
compressors, particularly centrifugal compressor machinery, which has long
been the choice for small turbine designs.
Table 5-r lists some important
characteristics of radial and axial compressor types in regard to vehicular
gas turbine applications.
Aside from these general characteristics, some significant improve-
ments have resulted from advanced centrifugal compressor research.
Earlier
passenger car turbine efforts, including free-turbine regenerative designs
announced by General Motors, Chrysler, Ford, and Williams Research, have
utilized a single stage centrifugal compressor of approximately 4:1 pressure
ratio.
Recent developments, however, have demonstrated the ability to
generate pressure ratios on the order of 6 to 7 per stage with compressor
efficiencies of the order of .80 and no significant decrease in the surge
I
free range.
Utilization of high stage pressure ratios has several beneficial
effects on the cycle, most important of which is better optimization of cycle
conditions at lower load conditions for improved part load BSFC.
This is
illustrated in Figure 5.2, where a comparison of the SFC characteristics
of comparable regenerative engines with design compression pressure ratios
of 4 and 6 is presented.
At the design point there is little to choose
between the fuel consumption of these two cycles.
However, at part load
as the compressor pressure ratio falls and turbine inlet temperature is
reduced, the operating line of the higher pressure ratio cycle intersects
lines of constant T4 at nearer the ~inimum BSFC for that turbine temperature.
The equilibrium operating line of a perfectly optimized cycle is defined
by the locus of the minimum BSFC points at any operating temperature.
Figure 5.2 shows that a regenerative cycle «(=.75) with a design compressor
pressure ratio of about 6 tends to follow this locus of minimum BSFC for
reduced turbine inlet temperatures.
Because of the tendency to much higher
97
-------
(1)
(2)
(3)
(4)
( 5)
(6)
(7)
(8)
TABLE 5-1
COMPARISON OF GENERAL CHARACTERISTICS OF
RADIAL AND AXIAL FLOW COMPRESSOR TYPES
RADIAL COMPRESSORS I
Have high structural integrity,
allowing high tip speeds and
higher pressure ratio per
stage than axials.
Are less costly due to the
ability to investment cast the
parts and also require little
or no finish machining.
Are relatively insensitive
to losses associated with small
~irflow sizes and are competitive
with axials in efficiency for
the same cost.
Are inherently more vibration
resistant than axials.
Adapt easily to flowpath requ-
irements for ducting tO'a heat
exchanger or burner which are
typically located outside of
the axial flowpath.
Are shorter in axial length
for the same pressure ratio
but usually exibit larger
frontal area.
Are generally heavier and have
higher polar moment of inertia
and therefore are slower to
accelerate to high speeds
from idle.
Are less sensitive to erosion
caused by air contamination.
98
AXIAL COMPRESSORS:
(1)
Are limited aerodynamically and
structurally to lower tip speeds
and stage pressure ratios so
more stages are'required for a
given pressure ratio.
(2)
Are costly due to the requirement
for close attention to manufacturing
details such as leading and trailing
edge sharpness and additional stages
required to achieve a given pressure
ratio.
(3)
Are more sensitive to small flowpath
heights and clearance effects.
(4)
Require sophisticated vibration
damping methods.
(5)
Require tortured ducting flowpa ths
to burners and heat exchangers
outside the axial flowpath.
(6)
Provide a low frontal area profile
due to smaller O.D. but a longer
axial length (due to larger number
of stages).
(7)
Are usually lighter and have low
polar moment of inertia if acce-
leration of the gasifier is an
important consideration.
(8)
Are especially sensitive to sand
and dust erosion.
-------
0.7
H
.r::
I
p,
;x:
........ R
~ 0.6
I 1600
u
r...
U)
I:Q
0.5
0.8
0.4
0.3
2
4
2000
6
8
10
12
Compressor Pressure Ratio
FIGURE 5.2
Effect of Design Compressor Pressure Ratio
on Part-Load Specific Fuel Consumption.
99
14
'I ,c, ,r'
-------
fuel consumption at lower cycle operating temperatures and the large amount
of time spent in part load conditions in normal passenger car operation, it
is absolutely necessary to optimize the design compressor pressure ratio
for any cycle, whether simple or recuperative, in this manner.
Optimization of the cycle to higher compression pressure ratios
has benefits not only in reducing specific fuel consumption but also in
increasing specific air consumption, thereby reducing size, weight, and
cost of the entire engine package.
A 15% increase in air consumption is
possible with an increase in compressor pressure ratio from 4 to 6.
Because of the cost, size, and difficulty in maintaining the heat
exchanger component, a somewhat less efficient simple cycle may be desir-
able.
Non-recuperated cycles require higher compressor pressure ratios
(typically of the order of 10-12) 'for optimum performance.
Recent advance-
ments in radial compressor technology have demonstrated that such pressure
ratios can be achieved in a single stage, but the present drawbacks of
such machinery include:
1.
2.
vertical speed lines with little surge free range
dependence upon variable compressor inlet guide vane~
3.
geometry for adequate surge margin during off-design operation
potential efficiency loss due to very small airflows and
4.
associated low Reynolds numbers
requirement for titanium parts, precise tolerances, and
s.
small clearances for realizing efficiency potential
bearing problems resu~ting from very high rotational
6.
speeds (over 100,000 rpm)
inertia effects on response time.
Since tortured flowpaths and added costs result if more than one stage of
radial compression is used, it is important that the above problems be
solved before the elimination of the regenerative type powerplant can be
considered.
Exploding the myth that centrifugal compressors are inherently low
in efficiency is another objective of extensive research.
Approximate 1968
state-of-the-art of small centrifugal compressors (23) is presented in
Figure 5.3.
100
-------
100
:> 90
tJ
C
OJ
.~
U ~J
.~ s::
~ Cj
11-1 U
~ 1-1 80
<:J
Q) P.
tJ1 I
nJ
,,.,
(jJ
If)
I 70
.j.)
~
O~.ctu 1 Stag0. crformanc
60
0 2 4 6 8 10 12
stage Pressure Ra tic
FIGURE 5.3
Present Performance of Small Cent.rifugal
(23)
Compressors .
101
-------
Comparison of various gas turbine concepts requires that careful
consideration be given to component efficiency values.
Compressor efficiency
characteristics corresponding to present state-of-the-art technology were
used in this study.
The incremental difference in BSFC for various levels
of compressor design efficiency for a representative range of compressor
pressure ratio (CPR) is presented in Figure 5.4 for a simple cycle with
86% turbine efficiency, and 20000F turbine inlet temperature. An error in
estimating compressor efficiency level of plus or minus 2% from the base
level of 80\ results in an incremental change in BSFC of .010 - .015 lbm/hp-hr,
o
depending upon the design CPR. For a turbine inlet temperature of 1800 F,
which is representative of current uncooled turbine blade temperatUEe limits,
the incremental change in BSFC is between .017 - .015 lbm!hp-hr.
In Figure
5.5 it is shown that this inability to predict compressor efficiency level
by a 2% margin affects the design cycle BSFC by approximately the same amount
as a change in turbine inlet temperature of lOOoF. These characteristics
point out the care that should be taken in estimating turbomachinery compon-
ent performance.
It has been mentioned that variable compressor geometry or preswirl
is necessary in the case of high pressure cycles to provide adequate surge
margin and optimum engine performance at part power.
The effect of variable
inlet guide vane geometry on compressor map performance characteristics is
illustrated in Figure 5.6, where a typical high CPR machine with variable
pre-swirl characteristics is shown.
The surge line has been effectively
shifted to higher pressure ratios and to lower flows, allowing equilibrium
operation at higher cycle temperatures in the normally unstable operating
region. Operation at higher T4' which may be limited by turbine outlet
(hib) over-temperature considerations, results in a significant part-load
efficiency improvement.
The variable geometry, however, affects the mechan-
ical and control system simplicity, cost, and reliability of the simple cycle
engines.
This type of variable geometry actuation system, however, is
considerably less costly, more reliable, imposes less severe component
efficiency degradation, and represents less of a development risk than the
variable power turbine geometry located in hot sections of a free turbine
engine.
The variable power turbine geometry will be discussed in follawing
sections.
102
-------
n .1r;
O.Of<
0.0(,
~~
r-:
I
P.:
.......
-3
0.0.1
,
U
f,..
If.
()
0.02
-o.o?
-0.04
0.70
. r '~0
~) ..-..
'I..~ (0 .-
-/0
o
0.72
0.74
0.84
0.86
0.76
0.78
0.80
0.82
Compressor Efficiency, ~c
!.T ;U~.,.E 5.4
Effect of Change in Compressor Component Efficiency on
Incremental Chanqe in Specific Fuel Consu.'1Iption (OSFC).
103
1'')1)
-------
.8
f = 0
R
'Ylc = .RO
l'lT = .86
T = 59°F
.7
~I
.t::
I
P.
~
...
.0
~:j
.6
o
T4 - F
Ihl)O
u
c..,
UJ
((j
OBSFC ~ Tlc
( Of(c = :.. 2%)
.5
. .4
6
8
10
12
14
Compressor Pressure Ratio
FIGU i.X 5.5
Degree of Uncertainty of Specific Fuel ConRumption
Resulting from Plus or Hinus Two Percentage Points
in Compressor Efficiency.
1(')/\
1256
-------
1,1
.1
T4/T1
/t Equi ibrium
Op('J!' ting
Line'
12
~ 10
P.
"-
N
H
p, .
100\ uiii1
o
..-1
+J
~
A
Q)
~,
::s
UJ
In
OJ
~
p..
6
:!
70%
o
40
60
80
100
Relative Corrected Compressor Flow (WVi i 6) - %Desiqn
FIGrm.E S. 6
Typical Compressor Characteristics of lIif"jh Pressur.e
Ratio Centrifw.T.,l Compressor with Varia.i..Jle Inlet
Guide Vane Pre-Swirl.
lns
11117
-------
Size effects of compressors with flow rates of 1-5 lbs/second do
not appear to represent a significant barrier with respect to performance.
UACL has published results of tests conducted on compressors sized for
2-2.5 lb/second (throttled to correspond to airflow sizes of 1.2 - 1.4
lb/second) which meet part speed efficiency predictions.
More information
is required, however to establish the effect of correct airflow sizes,
clearances (which cannot be scaled to small machinery sizes), and the
effects of very high rotational speeds on bearing requirements.
In conven-
tional aircraft gas turbines the power consumption of roller bearings is
a direct function of speed, and for speeds over 100,000 rpm (as required
for high pressure ratio compressors) this loss can be significant.
Develop-
ment of low-friction sliding type bearings for small turbomachines is
required.
5.2.3
Turbines
Turbines incornrnon usage for small turbomachinery include both axial
and radial inflow types.
Originally the axial flow turbines were preferred
for most applications, whereas radial turbines were utilized primarily in
turbochargers where efficiency requirements were relatively modest.
However,
current trends indicate a changing preference for the structurally superior
radial turbines in applications where low cost, configuration compatibility,
or particular performance characteristics favor this type (18).
As has been mentioned previously, the economic barrier is the most
formidable obstacle to mass production of turbine powered cars.
Low cost
manufactur~ng requirements dictate that turbine rotors be cast or forged
with minimum machining, and for low cost mass production something other
than investment casting will be needed.
Radial turbines have demonstrated
less sensitivity to casting tolerances and vibration than axial turbines.
Cast axial turbines can experience losses of the order of 5 to 10% of
efficiency potential if exact tolerances are not adhered to, which makes
efficiency of radial and axial stages competitive on an equal cost basis
. for the range of specific speeds of interest in small vehicular gas turbines.
In addition, radial stages are structurally and aerodynamically
capable of higher tip speeds and stage pressure ratios, which is important
for the higher pressure ratio cycles now considered for automotive engines
106
-------
due to the added cost of multi-axial stages necessary to accomplish the
same task.
Axial turbines typically show linear torque characteristics
over a wider range and less of a degradation in efficiency with reduced
tip speeds at constant pressure ratio (18).
These characteristics, along
with flowpath compatibility and lower stage work requirement, are important
factors in the choice ofaxials for free power turbines..
In general, the efficiency of radial gasifier turbines is largely
unaffected by variation in speed, and the use of a constant gasifier turbine
efficiency in preliminary cycleanalyses can be shown to be essentially
correct.
The stagnation enthalpy rise across the impeller of a radial
I
I
I.
t\h
- 0
c
without pre-swirl is:
2
m U .
. 2
:::;-
g J
c
where:
m :::; slip factor
U2 = impeller tip blade speed (fps)
g - constant in Newton's second
c -
law (ft-lbm/lbf-sec2)
compressor
J = energy conversion factor
(778 ft/lbf/BTU)
The gasifier turbine output head is:
i\ Co2
i\ h :::; 2g J T{ t
°t c
where:
Co = isentropic spouting velocity (fps)
If t = gasifier turbine adiabatic
efficiency (total to total)
Now,
!\h
o
c
= !\h
°t
or
2
ID 'U2
g J
c
C02
= 2g J 7l t
c
and for a 30-32 blade compressor and a radial turbine of equal tip speed,
m =.9 and
7) ,- 9
, I t--- .
so
2
~U/Co) = 1/2 (71 t/m) = l/2
and
U/Co = .707
107
-------
which is approximately the best efficiency design point of a zero-swirl
radial turbine and is relatively constant over the range of speeds and
pressure ratios of radial gasifier turbines.
In contrast to the operation of the gasifier turbine or single-
shaft turbines which operate over a limited output speed range (approxi-
(
mate1y 60-100% of design), the free power turbine can operate almost any-
where on its performance map in terms of pressure ratio and speed.
A
derivation of the axial power turbine relative efficiency relationship
at any operating point is possible and it is convenient to present the
results in terms of the hub spouting velocity or velocity ratio parameter
(Uh/Co) .
The linear torque characteristics of axial turbines is shown in
Figure 5.7, where torque at design turbine inlet temperature and 100% gas i-
fier speed is plotted as a function of hub velocity ratio.
The stall
torque is shown to be twice the design point torque and hub velocity ratio
is generally equal to .45 at the best efficiency point.
of a straight line for this relationship gives:
Writing the equation
T = Tv
.45 (Uh/Co) + 2 TD
where:
r = torque
script
point
Uh = hub wheel speed, fps
C = isentropic enthalpy drop
o
from inlet total to out-
let static,
\! 2g J h
c 0, ,
1.sentrop1.c
(ft-Ibf) and sub-
D refers to the design
From the definition of efficiency and horsepower/torque relationship, we get:
7l =.~
HP'd 1
1. ea
7LJ Tu _TqJh/CO)T{ D
= 1DtJD/1(V = TD UD - Tv (Uh/co)D
By combining with the linear relationship from above,
108
-------
0'
I:
'M
.p
'"
~~
~
r
OJ
::J
tJ'
~~
o
~
QJ
~.
'n
~J
'"
.-i
W
~,;
2.n
~J
,:
-M
o
UI
1.5
1.0
0.5
o
o
FICURE S. 7
Design
Best
@(U/
I
I
I
I
Ef~iciency
O)Hub = .45
.2
.4
(U/CO) hub
T4' CPR
.6
.8
1.0
Axinl I'o'Ner Turbine Torque Characteristics vs.
Hub Spouting Velocity at Design Turbine Temperature
and Compressor Pressure Ratio.
109
-------
7l T (Uh/Co)
7l Design - T D (Uh/Co) D
=
[2 TD- ~(Uh/CO) ]
TD (.45)
(Uh/CO) =
=
[(...1.-) - 1 (Uh/Co) ] (Uh/Co)
.45 (.45)2
and
7[/1) =(2)
, (Design .45
(Uh/Co) - (.~5)2 (Uh/Co)2
This equation gives the relationship of relative efficiency as a function
of velocity ratio for an axial free-turbine and is shown in Figure 5.8.
In this study radial turbine types have been used for gas generators
and single-shaft applications while axial turbines have been used for power
turbine stages.
Several of the possible disadvantages of radial turbines
are increased weight, larger outer diameter, and generally higher polar
moment of inertia, which should be minimized in an effort to reduce acceler-
ation lag between idle and maximum gasifier speeds.
However, weight and
frontal area are not generally critical considerations for vehicular
applications, and unless acceleration lag becomes an insurmountable oper-
ational problem with radial turbomachinery, it is concluded that the
advantages of radial turbines (particularly cost, system compatibility,
strength, and simplicity) outweigh the disadvantages. The radial turbines
chosen for analysis throughout this study have been designed for a total-to-
static adiabatic efficiency of about 85%. Figure 5.9 illustrates the
variation in the increment of BSFC caused by incremental change of turbine
efficiency for single shaft engines of 20000p turbine inlet temperature,
turbine and compressor efficiency at 86 and 80%, and a range of compressor
pressure ratios.
The effect of a two percent loss of turbine efficiency
has approximately twice the effect of a similar loss in compressor efficiency
(.shown in Figures 5.5 and 5.6), which was previously noted to be comparable
to a decrease in T4 of approximately lOOoF. Choice of a relatively con-
servative value of turbine efficiency (85%) reflects the consideration of
the importance of this effect.
Axial power turbine efficiency was also
assumed to be 85%, and a discussion of the effects of variable turbine
nozzle geometry on efficiency is included in the analysis of free turbine
engines in Section 5.5.
110
-------
\
1.0
.0
.6
:>,
U
c:
Q)
-.-I
U
-.-I
~-I
~
~
.4
.2
o
o
.2
.4
.6
.8
(U/Co) hub
FIGURE 5.8
~~ial Power Turbine Relative Efficiency V5.
Hub Spouting Velocity.
111
1.0
1259
-------
n.12
~I
~f:
I
{J.
:x:
" (1.\.8
,(4
,-1
lJ
["
~f)
()
(). 1)4
-0.04
--0.08
0.74
(). :;8
T
a
1\c = O. 8
T = 200 of
4 = 590 F
E "" 0
R
0./.0
0.16
o
0.7(,
('I. 7~J
0.8')
0.82
0.84
0.8G
0.88
0.)0
TUJ':'bine
Efficiency, ~T
FIGURE 5.9
Effect of C!angc in Turbine Efficiency on Gas Turbine
[ncre:ner.tal Chanrye in Specific Fuel Consumption
(OSF~-:) .
II?
-------
The turbine temperature limit for uncooled blades and vanes has been
determined to be about 1800oF, a temperature considered representative of
turbine superalloy materials currently available.
Cooled turbine blades
are gener~lly not considered for automotive application due to the additional
cost and complexity and the lack of experience by the gas turbine industry
in cooled radial turbine parts.
Cooling of turbines may prove attractiye,
however, as an alternative to the use of critical materials in the turbine
hot section, where a large percentage of the total quantity of cobalt and
nickel are required.
One of the primary benefits of the high pressure ratio,
non-recuperative cycles is the miniaturizing of components resulting from the
high specific air consumption and the' subsequent reduction of critical
material weight (16).
A moderate pressure ratio regenerative engine with
cooled radial turbine blades might be a better solution to both the critical
materials problems and the poor low speed fuel consumption; however, this
type powerplant will require additional development.
For this reason, air-
cooled turbine designs are not considered in this study.
5.2.4
Cycle Losses
Analysis and comparison of gas turbine cycles requires that careful
consideration be given to cycle pressure, leakage, and mechanical losses
appropriate to the particular design.
For preliminary performance analysis
of various cycles it has been convenient to account for intake pressure
losses (assumed less than 1%) in compressor efficiency and to assume that
exhaust pressure losses are a constant 1-1/2% for all cycles.
This implies
a rather careful component installation, which will be a must for any high
efficiency gas turbine automotive powerplant.
In heat recovery cycles the
pressure losses and leakage losses through the heat exchangers are a function
of heat exchanger type, size, and engine operating conditions.
Careful
attention has been given to this aspect of cycle analysis in evaluating
recuperative or regenerative engines.
Mechanical efficiency has been determined to be about 98% for all
cycles and is considered constant over the range of speed and power condi-
tions.
It is known that frictional losses are a direct function of speed
for engines with conventional roller bearings, and the wide variations in
design rotational speed between low pressure recuperative cycles (approximately
113
-------
60,000 rpm) and high pressure simple cycles (120,000 rpm) suggests a
two-fold increase in frictional losses with the latter type of engine.
Low-friction slider bearings can alleviate this problem in the high speed
design, but these bearings typically have large lubrication requirements
(whether it be air or oil), can be costly, and have high starting friction.
Slider bearings are used where necessary for competitive performance, thus
making the somewhat favorable value of 98% mechanical efficiency possible.
5.2.5
Exhaust Heat Recovery
Recuperative and regenerative engines are presently dominant in auto-
motive gas turbine powerplant applications due to the necessity for achieving
acceptable part load and idle fuel consumption.
A variety of heat recovery
means have been used or are currently under study and development for
automotive and truck-type applications, most of which favor the rotary
regenerator with either metallic or ceramic matrices.
Several varieties of
stationary recuperators have also progressed to field and/or flight develop-
ment status, but neither heat exchanger type has been field or customer
proven to the degree necessary to assure operational feasibility.
The rotary disk or drum regenerator usually uses a high surface
density matrix, and effectiveness values of the order of 90% are commonly
attained.
The major problems associated with these units are thermal dis-
tortion and sealin9.
The loss of seal integrity causes rapid deterioriation
in engine performance due to the cycle losses caused by the leakage from
hot and cold gas paths.
The high surface density of this type of heat
exchanger results in laminar flow which permits favorable effectiveness
at low pow.er and a less proAounced pressure drop (and thus power reduction)
at higher power levels.
Higher effectiveness also results in a lower
optimum design compressor pressure ratio for the cycle, simplifying the
development of this component and allowing the use of less costly materials.
Matrix costs for rotating heat exchangers are typically low but the seals
and rotational apparatus add to the total system cost.
stationary heat exchangers or recuperators can range from tubular
construction to intricate compact surface8.
A typical tuhular 3-pass
design represents a well proven heat exchanger design concept and should
present a minimum of fabrication and development problems.
Large bulk
114
-------
and excessive pressure drop characterize this type of heat exchanger.
More
advanced recuperator types such as compact plate/split-fin designs must
also be considered due to the much smaller size and reduced pressure losses
at high power levels with these types of designs.
Recuperators have certain advantages over rotary regenerators, such
as smaller leakage and no carryover losses, but they are often more difficult
to manufacture and are subject to burnout and fouling problems.
The regener-
ator experiences a periodic reversing of flow which purges the matrix and
reduces the fouling and burnout problems.
A feasibility study has also been performed with respect to a part-
time recuperator or regenerator system which utilizes a bypass concept.
This concept developed from the fact that heat recovery (regeneration or
recuperation) is really only required at the low power levels since for the
automotive application the basic simple cycle has acceptable fuel consump-
tion at the high power levels.
A variety of heat recovery types were
examined, including rotary regenerators and tubular and split/plate-fin
recuperators, to determine the size, cost, and performance advantages of
the bypass system.
This concept is described in more detail in the heat
recovery analysis of open cycle gas turbines, Section 5.5.
5.2.6
Operational Characteristics
Operation of the gas turbine powered vehicle is not expected to be
greatly different from the conventional reciprocating engine powered auto-
mobile in service today.
Engine power output is essentially determined by
the turbine inlet temperature (T4)' which is controlled by increasing or
decreasing the fuel flow to the combustor by means of a standard accelerator
pedal.
Fuel controls can be either electronic or hydromechanical, although
electronic controls are favored for reasons of low unit cost and durability
unless trouble is experienced in field operation.
The current Volkswagen
electronic fuel injectors combine almost all of the elements required for
a gas turbine fuel control.
Hydromechanical controls such as used on the
GM and Chrysler gas turbine engines are also available and in higher produc-
tion rates the cost can be brought down to a practical value.
Because low cost high temperature sensing devices do not have
durability or quick response, the turbine outlet temperature (TOT) for
U5
-------
single-shaft engines (or inter-turbine temperature for two-shaft engines)
is usually chosen as a more suitable control parameter.
This is true
because of the reduced gas temperature at this station and the fact that
a unique TOT exists for any operating point.
The TOT schedule at any
compressor speed is made to conform to the value of TIT corresponding to
optimum cycle operating conditions.
For any particular engine cycle the
turbine match, the surge margin, and the turbine outlet hub temperature
restrictions will determine the operating temperature at any output power.
Additional engine controls are necessary for utilization of various
variable geometry systems.
Vane settings are automatically scheduled as
a function of engine and ambient temperature conditions.
No driver response
is necessary for variable geometry actuation, but such additional systems
add complexity and are to be avoided unless absolutely necessary for
performance or operational stability.
Engine braking in gas turbines is dependent upon the particular
design type and the braking features included in the design. The single
shaft turbine engine has superior braking capability compared to any of
the two-shaft versions as a result of immediate absorption of excess power
by the compressor component upon release of the accelerator pedal.
In
two-shaft designs the power turbine is in a free-wheeling condition rather
than a retard position, and some technique of engine braking is required.
Variable power turbine geometry generally required for optimum part-load
performance in free-turbine engines is helpful in regard to braking because
nozzle vanes can be rotated to admit flow to the back side of the turbine
blades.
Dmmediate power reduction is provided with bypass features such
as the waste gate on the Continental J-65 or windows immediately upstream
of the power turbine in the Williams Research engine, both of which
divert gasifier turbine exhaust gases away from the power vehicle.
Another aid to braking incorporated into both the Williams engine and the
General Motors GT-309 truck turbine is the differential clutch arrangement
which couples the gasifier and output shafts for braking and power turbine
overspeed protection.
Two-shaft engines withoutsuch braking aids fail to
provide any retardation when the driver releases the throttle.
Dr i ver
technique is affected by this lack of retardation such that there is a
tendency to spend more time in free-wheeling and also very high throttle
116
-------
settings than would normally occur with engines with good braking
characteristics (17).
5.2.7
Power Transmission
The choice of specific engine type (single-shaft or free-turbine)
influences the transmission requirements for providing torque at the wheels
of the automobile.
The primary types of transmissions suitRble for
vehicular gas turbine powerplants include:
1-
. 2.
Mechanical or hydromechanical
Electrical
3.
Hydrostatic.
The single-shaft engine has a narrow useful speed range (approximately
60-100%) and an unfavorable torque-speed relationship (in contract to the free-
turbine engine which produces maximum torque at stalled output speed).
Vehicle acceleration considerations require more sophisticated transmission
output characteristics for both regenerative and non-regenerative single-
shaft engines.
The torque-speed characteristics of the medium (6-8) pressure
ratio regenerative cycle are somewhat more favorable because the torque
and power at any speed are generally only temperature limited whereas the
.torque of the simple cycle engine is probably surge margin limited.
However,
it is almost certain that either an infinitely variable hydrostatic trans-
mission or an 8 to 10 speed mechanical transmission with controlled slipping
clutch or torque converter is necessary to provide reasonable acceleration
characteristics.
Unfortunately, the development status of these more
sophisticated transmissions is not at the level where a definite judgement
can be made regarding a choice of transmission for a particular signle-shaft
engine design.
Prototype demonstration will ultimately be required to
establish the transmission choice.
The hydrostatic transmission is infinitely
variable, using positive displacement hydraulic components and technology
developed for aircraft hydraulic units.
This type of transmission has
relatively constant but low efficiency over a wide delivery range.
A trans-
mission concept rapidly gaining favor is one which combines the best
elements of hydrostatic and hydromechanical transmission, namely the dual
mode transmission.
In this concept torque is delivered entirely through the
hydraulic loop at low speeds where the infinitely variable hydraulic loop
117
-------
provides more favorable efficiency characteristics than the hydromechanical
type.
At higher output speeds a reduced percentage of power is transmitted
through the hydraulic loop and the more efficient mechanical drive take over.
Pure hydrostatic transmissions are typically 10 to 15 percent less
efficient than hydromechanical types, resulting in higher fuel consumption
and a larger, more expensive engine to delivery the required power.
A
hy~rostatic transmission does offer potential for transmission weight and
volume savings, and the infinitely variable feature can be effectively used
to provide improved response or to allow single-shaft or free-turbines to
run at their most economical speeds for a given output power.
Electrical transmission, which consist of. a d-c generator or a high
speed alternator coupled to the engine and electric motors attached to the
wheels or drive shaft, can be used with gas turbine engines.
Gas turbine
powered hauling trucks now use electric transmissions, but the major draw-
backs to their use in automobiles is the high initial cost and heavy weight.
Efficiency, however, is higher than hydrostatic and comparable to hydro-
mechanical.
Because of their superior torque-speed characteristics, free-turbine
engines probably require only a conventional 3 or 4 speed automatic gear
shift without the torque converter used in present I.C. engines.
Elimination
o~ the torque converter reduces the size and cost of this component and
partially accounts for the increased cost of the power turbine, which is
really no .more than a compressible fluid torque converter.
Both of these
devices multiply torque through slip, which means an inefficiency is intro-
duced to the system.
The power turbine component is lighter, smaller, and
more efficient than the converter machinery required for single-shaft engines,
and the choice ultimately depends upon weighing these factors against
additional cost and system complexity of the free-turbine engine with
variable turbine geometry features.
5.2.8
Gas Turbine Concepts
Having reviewed the basic characteristics of gas turbine components,
it is now pertinent to investigate the gas turbine concepts which are being
studied for vehicular application.
Various combinations of compressors,
118
-------
turbines, heat exchangers, and drive systems have been proposed for gas tur-
bine powerplants, however, suitable systems for automobiles require a simple,
low cost system with good efficiency over a broad power range.
The various
concepts investigated under this study are presented in Figure 5.10.
The
chart is divided to include both conventional (radial or axial compressors)
and unconventional (comprex) gas generator types.
The chart of Brayton
concepts is divided into essentially two sections, namely, closed (and semi-
closed) and open cycles.
Open cycle Brayton, or gas turbines, using atmospheric
air as a working fluid are currently the most popular choice for passenger
car applications due to the relative simplicity of this type of system.
Further breakdown of the open cycle gas turbine concepts into systems with
and without heat recovery and for single or two-shaft designs is shown in
Figure 5..10.
These concepts are fully discussed in Sections 5.4 and 5.6.
Particular emphasis has been given to the heat recovery concepts due to the
lack of agreement between various gas turbine designers concerning the most
suitable heat recovery means.
Several types of closed cycles have also been examined to determine
the ;easibility of such systems for automobile powerplants.
These include
the basic simple and recuperated closed cycles and the more complex
Ackeret-Keller cycles with intercooling between compressor stages and ~.
reheat between turbine stages.
The closed Brayton cycle study (Section 5.3)
also includes an investigation into the possibility of utilizing working
fluids other than air.
5.3
CLOSED CYCLE BRAYTON
In the closed cycle Brayton powerplant the working fluid is continu-
ously recycled.
Heat is supplied to the cycle by means of an external burner
through a heat exchanger and waste heat is removed from the working fluid
just prior to its return to the compressor inlet to bring its temperature
back to the initial cycle starting point temperature.
A schematic arrange-
ment of the basic recuperative closed cycle system is presented in Figure
5.11.
The basic recuperative system requires four heat exchangers, which
include the waste heat exchanger, burner loop air pre-heater, combustion
heat exchanger, and recuperator, in addition to an accumulator/compressor
tor working fluid density control.
Of these components only the recuperator
is required in a comparable open cycle system.
119
-------
FIGURE 5.10
Gas turbine cOltcepts for vehicular propulsion systems.
GAS TURBINE CONCEPTS
COMPREX GAS GENERATOR
CONVENTIONAL GAS GENERATOR
CLOSED CYCLE
OPEN CYCLE
I--'
:v
;:)
ACKERET-KELLER
CYCLE
BASIC CYCLE
REGEN. REGEN.
OR BYPASS S H1PLE OR BYPASS SIt-'I.PLE
RECUP. RECUP.
1-' ''''' ....... ~ .... ."... ~
I-'
\D ARGON, AIR, HELIUM, HYDROGEl;, FHEON, ETC. AIR
<3'
-------
L__,
J\mbient
Air
T=S4SoR
(ASoF)
FIGURE 5.11
Recupera tor
Waste Heat
Exchanger
(Cooler)
Combustion
Heat
Exchanger
Preheater
Combustor
Combustion
Products
Exhaust
Schematic of Arrangement of Recuperative Closed Cycle Brayton.
121
119<)
-------
Regardless of the added complexity of even the most basic closed
cycle, this concept has developed interest largely due to the following
potential advantages of the closed cycle:
1.
Favorable emission characteristics of external burners
designed for closed cycles as opposed to the less
favorable open cycle burner emissions.
2.
Relatively constant efficiency over a broad load range by
regulating control system density to control power output.
3.
Elimination of fouling of machinery and closed cycle heat
exchangers due to continuous recirculation of clean work-
4.
ing fluid.
Reduced size of all system components as a result of the
5.
increased cycle base pressure.
Increased heat transfer coefficients and lower pressure
6.
7.
drops for same size heat exchangers.
Possibility of using a greater variety of fuels.
possibility of using a working fluid other than
air if found desirable.
8.
Reduced noise due to isolation of turbomachinery
in closed loop.
The following closed Brayton cycles have been evaluated to determine
their feasibility in vehicular powerplant system:
5.3.1
1-
2.
Simple closed cycle.
Recuperative closed cycle.
3.
Ackeret-Keller closed cycle with one stage
inter cooling and one stage of reheat.
4.
5.
Recuperative closed cycle with one stage of intercooling.
Recuperative closed cycle with one stage of reheat.
Simple and Recuperative Closed Cycles
The simple (non-recuperative) closed Brayton cycle system requires
all of the components shown in Figure 5.11 except for the recuperator (or
regenerator) used for turbine exhaust heat recovery.
Control of base cycle
pressure (density) is the primary mode of power regulation and allows the
Because of
design efficiency to be maintained over a broad power range.
this characteristic, the cycle design parameters should be optimized to
122
-------
provide best efficiency at the design point.
Optimum design compressor
pressure ratio for minimum fuel consumption is presented in Figure 5.12
for simple and recuperative cycles (75% recuperator effectiveness) with
design turbine inlet temperatures ranging from 1300 to 2000oF. This figure
illustrates that optimum compressor pressure ratio for non-recuperative
closed cycles is much higher, necessitating either the development of very
high pressure ratio single stage compressors or dual compressor stages,
which necessarily add to system cost.
In addition, these high pressure
ratios increase the cycle specific air consumption, which when combined
with the utilization of higher working fluid density results in miniaturizing
I '
I
of compressor and turbine component size below practical limits set by speed,
clearance, and Reynold's number considerations.
Whereas this miniaturization
effect is beneficial in regard to large gas turbine units designed for
central station power supply, there is no apparent advantage in the case
of automotive powerplants where lower limits of size are easily reached
or exceeded.
The other consideration which virtually eliminates simple system or
the bypass recuperative cycle is the level of design BSFC'attainable with
these cycles as compared to the conventinal heat recovery type.
Figure
5.12 illustrates that the simple system is approximately 25% higher in
fuel consumption than a comparable recuperative cycle (t R=' 75) over the
range of turbine temperatures available with present materials. Both simple
and bypass concepts can be eliminated from further consideration for closed
cycle powerplant systems due to performance and size limitations.
It has been noted that external heat is supplied to the closed system
through a heat exchanger to bring the closed cycle gases up to maximum oper-
ating temperature.
Because of the stresses resulting from both the high
closed cycle working pressures and from the high combustion product and
working fluid temperatures, a tube-type design is suggested. Closed cycle
maximum turbine inlet temperatures (TIT) exceeding l3000F require utilization
of stainless steel alloy materials with high chromium, nickel, or cobalt
contents for the combustion heat exchanger.
The critical nature of these
materials in high production quantities with respect to both cost and
availability suggests that closed cycle turbine inlet temperatures for
large-scale vehicle powerplant production be limited to 1300oF. The import-
ance of reducing critical materials in the combustorjheater component is
123
-------
-------- --------
20
rr. 15
,.j
u
z
d
H 10
U)
IJJ
Q
-. rati.ve
.-, 5
. J
~;:
H
E4
p..
0 0
0.8
1
I
Max.CIIX Tem erature
Limit for U i1izatinn
of 300-Seri s
stainless 5 eel
0.7
0.6
I
I
25% I
I
I
Note:
=.80
c
''IT=.86
h=.97
Mec
P/P\==.07
sr.
~~
IX)
0.5
I
I
I
I
I
....J
U
Cr.
U)
CQ
0.4
0.3
1200
1400
1600
1800
2000
TURBINE ENTRY TEMPERATURE - of
FIGURE 5.12
BSFC at Optim\m Design Compressor Pressure Ratio
for Recuperative and Simple Cycle Gas Turbines.
124
1198
-------
reflected in the relative component weights of a closed cycle automotive
powerplant.
Such a powerplant is proposed (26) to have a total engine
weight of 547 lbs and a volume of 10.31 cu. ft., of which 29% of the
weight and 35% of the volume are attributed to the combustion heat exchanger.
'l'he maximum cycle temperature of this unit (1500oF) requires extensive usage
of stainless steel alloy materials in the combustion heat exchanger, and the
total consumption of nickel (assuming this engine were mass produced for
automotive applications) .would be prohibitive.
Cycle parameters for a typical recuperative closed Brayton cycle with
the 13000F maximum turbine inlet temperature limitation are presented in
Table 5-II.
Design compressor pressure ratio was optimized for the cycle
temperatures, assumed levels of component efficiency, and degree of recuper-
ation (75%) to provide maximum thermal efficiency at the design point.
Compressor efficiency (78%) was de-rated several points to account for
expected losses resulting from corrected airflows less than 1 lb/sec.
clearance effects and casting tolerance effects caused by the small rotating
machinery size.
Consideration of flow rate limits determined the maximum
practical base cycle pressure of 75 psia.
At the design power the engine produced 150 HP at a total thermal
efficiency of 15%.
By reducing the base pressure down to ambient level,
the design efficiency is retained down through 20% power.
Further reduction
of power is possible by either operating the closed cycle in a partial
vacuum or reducing turbine inlet temperature.
The latter method of power
control is illustrated by the part load fuel consumption characteristic
represented by the solid line in Figure 5.13.
The part load BSFC (dashed
line) of a recuperative open cycle engine representative of present tech-
nology is also shown. This engine is a fixed-shaft version with a maximum
turbine inlet temperature of 1800oF, CPR = 6.3, and plate-fin recuperator
designed for 70% effectiveness at approximately 60% output speed.
The comparison between open and closed cycles demonstrates that the
recuperative closed Brayton cycle (with air as a working fluid) exhibits
no efficiency or fuel consumption advantage over open cycle versions, even
at part load.
A slight advantage in near-idle BSFC could have been shown
for the closed cycle if power were regulated by reducing cycle pressure
below 14.7 psia.
This would require increased accumulator size and pump
125
-------
TABLE 5- II
CLOSED BRAYTON CYCLE DESIGN PARAMETERS
Recuperative
Closed
Cycle
A-K Cycle
1 stage
intercooling
and reheat
Recuperative
closed cycle
with 1 stage
intercooling
Recuperative
. closed cycle
with 1 stage
reheat
Turbine Inlet Temperature
(TIT) of 1300 1300 1300 1300
Compressor Pressure Ratio (CPR) 2.9 8.2 4.5 5.0
TIel = .7.8 11 cl = .78
Compressor Efficiency .78 .78
11 c2 = .78 1l c2 = .78
7ltl = .85 11 tl = .85
Turbine Efficiency (Constant) .85 .85
Tlt2 = .85 7(t2 = .85
Burner Combustion Efficiency .99 .99 .99 .99
Recuperator Effectiveness ( ER) .75 .75 .75 .75
€. CHX .75 .75 .75 .75
E PRE-HEATER .75 .75 .75 .75
€ WHX .82 .82 .82 .82
Maximum Combustion 0 2540 2540 2540 2540
Temp. F
(6p!p) R .007 (hot side) .007 .007 .007
.002 (cold side) .002 .002 .002
(6P!P)CHX .015 .015 .015 .015
(6P!P)WHX .005 .005 .005 .005
(6p/p) TURBINE .015 .015 .015 .015
EXHAUST
Net Brake Horsepower 150 150 150 150
Thermal Efficiency - % 15.0 (air) 17.4 17.0 15.5
Working Fluids Examined Air, Argon, Air Air Air
C02' NH3'
Freon, Helium,
Hydrogen, etc.
126
-------
Il':
:I:
I
~~
IJ:I
u
c..,
VJ
r.J
] . n
, -
\
ReduCEd N, T4
P1=14. 7 psia
\
Maximum T.I. 0
. Closed, Cycle: .=1300 F
\~ / CPR Z.9
t E.p = .75 @ Desi
\ I N= 00%, Reducing PI .
" P1=75 P
, P:
'IIi " O/CyCle, T.I.T. = 1800 F
CPR == 6.3
" ER = .70 0 60% N
......... -
r- .... - - --- ---
ia
1.4
1.2
1.0
n Power
0.8
0.6
0.4
o
30
60
90
120
150
BRAKE HORSEPOWER
FIGURE
5.13
Fuel Consumption of Open and Closed Cycle Brayton
Powerplants at Full and Part Load.
D7
1200
-------
capacity, however, and result in higher engine speeds at low power levels,
complicating the already difficult problem of low-power speed reduction in
the transmission.
Two factors are largely responsible for the generally lower efficiency
of the closed cycle unit:
1.
TUrbine inlet temperature was reduced to l3000F to limit
the critical materials content of the combustor/heater
2.
to a practical minimum.
Closed cycle working fluid temperature at the compressor
inlet is necessarily higher than ambient air temperature
when ambient air is used as the cooling medium for the
waste heat exchanger (WHX).
For the cycle described above the WHX effectiveness was 82% and
the starting temperature was 1650F. The effect of inlet air temperature
on thermal efficiency is presented in Figure 5.~4 for several typical'
Brayton cycles.
Although open cycles are shown, the effect of inlet
temperature would be similar for closed cycle versions.
A 75% recuperative
cycle with the design parameters shown is found to suffer a thermal effi-
ciency loss from 25.8% to 20.9% as a result of an inlet air temperature
increase from 8SoF to l650p. This amount of efficiency lost represents
approximately 20% in increased fuel consumption.
Although the closed Brayton cycle (with air as the working fluid)
was shown to be considerably less efficient than the open cycle, it does
have the potential of functioning with certain gases which have somewhat
better heat transfer properties than air and which might be used to improve
the thermal efficiency of the closed cycle powerplant.
It is more likely
on a cost-effective basis that these fluid properties (high specific heat
or fluid density, or both) could be used to more advantage by reducing the
size of costly and bulky heat exchangers. Only a limited improvement
in thermal efficiency (compared to the air cycle) is generated by the
higher effectiveness and lower pressure drops in heat exchangers, not
nearly enough to allow closed cycles to be competitive with the open cycle.
A detailed cost and performance study is required to determine whether heat
exchanger sizes can be reduced sufficiently by utilizing a working fluid
with good heat transfer properties to permit use of alloys with higher
128
-------
iJ
z
~
H ~
U c:
H Q)
c.... 0
c.... 1-4
~ 8.
H I
e;
2
~
:r:
E-<
NOTE: 71 = .84
c
"T = . 85
l1B = 1.0
(A P = 05
P R .
NOTE:
16 of is Li 't of COo
Ca ability 0 Air-Coo
WH 1n Close Brayton
ant
ed
cycles
30
26
22
18
14
T4 = 12 OOF
10
o
40
80
120
160
200
240
INLET AIR TEMPERATURE - of
FIGURE 5.14
Effect of Inlet Temperature on Thermal Efficiency
of Brayton Cycle Engines. (Ref. 2)
12'1
1201
-------
critical material content in the CHX.
It would then be possible to increase
maximum cycle temperature and allow closed cycle efficiency to be nearly
competitive with the best open cycles.
ered in this study.
This work, however, was not consid-
Some low density gases like helium have high specific heats and
provide small heat exchanger components along with much smaller turbomach-
inery components, thereby requiring more stages of compression to achieve
the same pressure ratio as the comparable air cycle.
The problems associated
with such very small machinery, includi.ng the requirement of delicate manu-
facturing techniques and extremely high rotational speeds, would seem to
eliminate these gases from consideration regardless of the other benefits
accrued with their usage.
Higher density gases such as carbon dioxide or
freon provide good heat transmission coefficients, which tends to reduce
the size of the heat exchangers; however, this benefit is largely erased
by the higher pressure drops which result from the increased density.
In
addition, some gases (freon, for example) require excessively small turbo-
machinery sizes and others (like C02) do not behave like an ideal gas at
high pressures. These are some of the reasons why air is used almost
exclusively as the working fluid in closed cycle central station powerplants.
The thermal efficiency of the recuperative closed cycle with a variety
of working fluids is presenteq in Figure 5.15.
Design compressor pressure
ratio was optimized for each fluid, and other design parameters were com-
parable to those selected for the air cycle (Table 5-II) .
Design point
efficiencies varied from 14.0 to ~6.2%.
Although no significant efficiency
is evident with working fluids other than air, there are benefits directly
attributable to improved heat transfer properties (HX size, cost, etc.).
Characteristics of the various possible working fluids for a closed Brayton
system are summari~ed in Table 5-III.
5.3.2
Ackeret-Keller Cycles
Assuming that the emission characteristics of open cycle gas turbines
can be solved, it appears that the closed cycle gas turbine, because of its
high initial cost, is justified from an economic standpoint only if it can
significantly improve thermal efficiency over a broad operating range. An
analysis of the recuperative Ackeret-Keller (A-K) cycle, shown schematically
130
-------
?-t 12
U
;t~
H
U +'
H s::
[" Q)
r., 0
1<1 ~
~~ 8.
I 8
,'.t
fr.
i.l
:1:
r...
20
H CO (~
8 (>L
ArgO~ Nil)
. » Air
I
HeliuTl
T = 3000F
1\
T = 6250R
1
'I c = .7[;
111 = .85
I ~
0 r.tona t i'mic G E = .75
flses R
0 Diato Inic Ga ~es
0 Polya omic ~ases
16
Freon
(F-12)
4
o
1
2
3
4
5
7
6
8
9
10
COMPRESSOR PRESSURE RATIO
FIGURE 5.15
Thermal Efficiency of Closed Cycle Brayton Engines
at Design Compressor Pressure Ratio for Various
Working Fluids.
131
1202
-------
TABU: 5-.£n
CHARACTERISTICS OF DIFFERENT WORKING FLUIDS IN CLOSED BRAYTON CYCLE SYSTEM
Specific Heat
Cp-BTU/lb OF
(@ 680F and
14.7 psial
Air
Argon
Anunonia
(NH3 )
......
W
N
I
Carbon Dioxide
(C02)
Freon
(F-12)
Helium
Hydrogen
(ReI + C02)
.241
.124
.523
.205
.244
1. 25
3.42
.755
k=E!.
CV
1.40
1.67
1. 32
1. 30
1.13
1.66
1.41
Specific
Gravity
(Air=1.00)
1.00
1.379
.596
1. 529
4.520
.138
.0695
.28
Advantages
Readily available; non-hazardous;
turbo-machinery design problems
minimal
High density fluid promises
compact rotating unit size
High spacific heat gives good
heat transmission qualities;
low density means low HX
pressure drops
Higher density than air gives
good heat transfer coefficients
Very dense and good heat trans-
fer qualities of fluid promise
compact unit size
Small heat exchanger size;
inert gas so no oxidation danger
exists for turbines or HX's at
high operating temperatures
Promises very compact heat
exchangers
Similar to helium
Disadvantages
Heat exchangers large in size
due to mediocre heat trans-
mission qualities
Low specific heat counter-balance
effect of high density on HX com-
ponent size; high pressure drops
in HX; costlier than air
More compression stages required
for same compression pressure
level and velocity triangles;
low density counter-balanOl8
effect of high Cp; costlier
Low specific and high pressure
drops heat counter-balance
beneficial effects of high
density; costlier; does not
behave like ideal gas at higher
base pressures (higher BSFC)
Higher optimum design compressor
pressure ratio (10) due to extre-
mely low k value; turbomachinery
size very small even without
high base pressure; costlier
Rotating machinery and apparatus
very small; more stages of com-
pression required (21) or higher
wheel speeds of turbomachine for
same aerodynamic design as air;
costlier; high leakage rates
Similar to helium; very hazardous
even more difficult to control
leakage
Similar to helium
-------
~
in Figure 5.16 with one stage of intercooling and reheat, as well as
recuperated cycles with (a) intercooling only, and (b) reheat only, was
performed in an effort to document the extent of possible performance
advantages of these more complex systems over the recuperative closed cycle
and the open Brayton cycle systems. Design cycle parameters for these three
variations of the Ackeret-Keller cycle are presented in Table 5-II. The
cycle characteristics, including temperature and efficiencies, are the same
as noted for the recuperative closed cycle previously noted.
Compressor
pressure ratios optimize at different values for each of the cycles, as
illustrated in Figure 5.17, where thermal efficiency is plotted as a function
of compressor pressure ratio (8). It should be pointed out that the relatively
high levels of theDmal efficiency (32-36%) indicated by Figure 5.17 are ~ot
representative of closed cycles operating at l650F compressor inlet temper-
atures and somewhat reduced flowpath efficiencies. This plot is useful,
however, in illustrating that an optimum compressor pressure ratio exists
for each type of cycle.
In this manner it was determined that the optimum
compressor pressure ratios for the closed A-K cycle of Table 5-II and
versions with intercooling and reheat only were 8.2, 4.5, and 5.0 respectively.
The design point the~al efficiency of the A-K cycle with intercooling and
reheat was ~7.4%, only marginally better than that of the basic recuperative
closed cyole (15.0%).
Full and part load fuel consumption of the recuperative closed Brayton
cycles is presented in Figure 5.~8. The reduction in BSFC made possible by
addition of intercool~ng and reheat was approximately 14%. Fuel consumption
of a comparable open cycle is an additional 30\ lower in BSFC than the A-K
cycle.
cycle}
It is concluded that the closed Brayton cycles (including the A-K
are impractical automotive powerplants for the following reasons:
1. Reduced efficiency relative to open cycle types due to
lower maximum cycle temperature '(~300oF) and higher
compressor inlet temperature (l650F).
2.
Increased cost and complexity of the closed system over the
open cycle system resulting from addition of at least three
heat exchangers plus a working fluid control system
{accumulator/compressor).
133
-------
Air
T=54SoR
Air
WHX
FIG\JIU': 5.16
Recuperato
Prc~hea ter
Combustor
CHXA
CHXB
Acker.et-Keller Type Closed Brayton Cycle ~ith
One str:tfJe of Intercooling and Reheat.
lVl
1.203
-------
~
CJ
><
CJ
...
><
u
z
~
H
U~
H C
.... II)
~ ~
~ ~
=-:J
p;
~
~
7-
~
><
<1;
~
:z:
r:rJ
g,
40
~ =.7
R
35
1 Reheat
...... .
1 ...... ......
nte.rCOOl'
In
3C
..
"-
..
T4 = 1350°
* At Standard
ient Temp.
2
2
4
8
6
CPR
F.H;URE 5.17
= .85
= .91
= .98
onditions
10
12
Thermal Efficiency of Recuperative Brayton Cycles vs.
Compressor Pressure Ratio for a Turbine Inlet Temperature
of 13500F and a 75% Effective Recuperator (Reference 8).
131)
1204
-------
".
1.6
,
i
.,
~\ No Interc 001 or Rehec t (CPR '-" 2. ~)
--.; - - 1-- --- --- ~--
~\.. 1 Reheat (CFR = 5.0)
~ IiIIo.......... 11 .,1 f("PC"L! C:\
---Iii ..:._-:. - --_..:. Iii> - ..: --
----
Acker.et...,Kel ,er'j 1 Xn ter. ooler & 1 ~ ~heat' (CPR=. 8
T4 - 13000p
., r ,.., "'.~ ---
R 62SoR
T ::
i;: 75 psi ~ @ m~S.PT.
.78
.85
'.2)
1.4
1.2
r.r~
a.:
I
~ ~ 1.0
ro
1/
()
~I
CJ)
(11
0.8
0.6
0.4
o
50
100
150
BRl\-,'<:E HORSEPOWER
FIGURE 5.18
Full ana P~rt Load Fuel Cons~ption of Recuperative
Closed Brayton Cycles with Maximum Turbine Entry
o
Ternpe:r.atur.es of 1300 P and 75% Recuperator Effectiveness.
136
120~
-------
3.
Reduced rotating machinery sizes and the high rotational
speeds which are due to the high base cycle pressures.
5.4
OPEN CYCLE GAS TURBINE
Sections 5.1 and 5.2 gave a comprehensive review of automotive gas
.turbine characteristics in general, while 5.3 discussed the characteristics
of the closed cycle gas turbine concept as an automotive powerplant.
It was
shown in this latter section that the lower efficiency and increased complex-
ity of the closed cycle as compared to that which is realizable with an open
cycle essentially eliminates the closed cycle as a contender for an auto-
motive type powerplant.
This section will now investigate the various open
cycle gas turbine concepts, in particular, those concepts which use various
types of heat recovery system to improve the part-load fuel consumption.
Various combinations of compressors, turbines, heat exchangers and
drive systems have been proposed for gas turbine powerplants in order to
attain improved efficiency over a broad operating range. This is especially
important for automotive vehicle applications where efficient operation
is required over a wide range of speeds and power levels. During the past
few years considerable effort has been directed towards establishing those
gas turbine parameters which will provide the best combination of perform-
ance and fuel economy.
Figure 5.19 shows the fuel consumption character-
istics for a simple cycle gas turbine (no heat recovery) as a function of
compressor pressure ratio.
As is well known, the fuel consumption decreases
both as turbine inlet temperature and compressor pressure ratio increase.
As was mentioned in Section 5.2, the maximum state-of-the-art turbine inlet
temperature for uncooled turbine blades is about IBOOoF (due to sulfidation
limits imposed by the turbine blade material) and Figure 5.19 shows that
the optimum pressure ratio for this turbine inlet temperature is about 15.
Therefore, if ~5 is chosen as the design compressor pressure ratio with an
IBOOoF maximum turbine inlet temperature, the specific fuel consumption
woudl be approximately .52 lbro/hp-hr.
More important, however, is the fact
that at lower power levels and therefore lower speeds both the compressor
pressure ratio and the turbine inlet temperature decrease resulting in much
higher fuel consumption values. The major point to be noted here is that
if the simple cycle is designed for minimum fuel consumption at maximum power
137
-------
>1
,1":
I
8<
,r,
",
~
~::
o
.,oj
p
p~
3
III
~:
8
.'~
(1)
~
I)
..oj
II.~
..oj
()
(1)
C)"
UJ
FIGURE 5.19
.~
I
Q.
..~
",
,q
,.1
r~
o
...j
~J
~
U)
I::
8
M
(1)
::J
Ii<
U
...j
II..,
...j
o
(1)
0,
UJ
FIGURE S. 20
n. 'j
Turbine Effic ency, llT =, .86
Compressor Ef iciency,¥tc =1.80
0.8
(1)
>4
~,
.II
,oj
>.,
Q)
(1,
F-
al
E-i
.',
()
0.7
0.6 1600
1BOO
0.5
2000
2200
. bOO
n.4
,j.)
al
....i
~
H
(1)
r~
.,oj
.n
"
'I
{ t
0.3
2
6
10
14
18
Comp~;essor P);essure Ratio
Generalized Simple G,'1S Turbine Performance Chari'1.cteristic',;.
n.')
O. E\
Heat
Turbine Effi :iency, 1\.T '" 0.n6
C()mpr8S~()r E :ficiency,rtc =-~ O.Bn
~changer Effe tiveness, = 0.75
0.7
. b.)() ,',
C'
01
~ I
:-,
.~
m
,.,
C)
, ),
rei
):00 'LI
E-i
~J
Q)
M
000 f~
H
(1)
20(' I~
'~f
400 ."
>.1
0'JJ ~
18
0.6
0.5
0.4
0.3
2
6
10
14
Compressor Pressure Ratio
Generi'llized Regenerative Gas Turbine Performance
Characteristics.
138
1261
-------
then the fuel consumption at part speed and power will be very poor.
In
contrast to Figure 5.19, Figure 5.20 shows the fuel consumption characterist-
ics for a regenerative (or recuperative) gas turbine also as a function of
compressor pressure ratio.
It can be seen that the regenerative cycle has
greatly different fuel consumption characteristics than the simple cycle in
that the minimum fuel consumption occurs at 8 much lower compress9r pressure
ratio. For the laDOor turbine inlet temperature cycle, and 75% regeneration,
the minimum fuel consumption is about .42 Ibm/hp-hr at a compressor pressure
ratio of about 6.
Unfortunately, as in the simple cycle, the fuel consump-
tion of the regenerative cycle also increases at lower power levels and speeds
where both the compressor pressure ratio and turbine inlet temperature
decrease.
However, as can be seen in Figures 5.19 and 5.20, fuel consumption
for the regenerative cycle is much less than that for the simple cycle at
both full and part power levels.
Therefore, from a fuel consumption standpoint, it appears that the
medium pressure ratio, regenerative (or recuperative) gas turbine cycle is
superior to the simple cycle.
In addition, as was pointed out in Section
5.2, the present state-of-the-art maximum pressure ratio for one stage radial
compressors (in the size required for automotive gas turbines) is about 7:1
(with or without pre-swirl) .
Furthermore, flow path, cost, and complexity
considerations have indicated the use of a one stage radial compressor
design for the most attractive gas turbine automotive powerplant.
There-
fore, due to the above characteristics, the one stage radial compressor,
regenerative (or recuperative) cycle was chosen as the basis for the most
feasible automotive gas turbine powerplant.
The cost and fuel economy advantage of the regenerative cycle over
the simple cycle, however, can only be realized with the proper design
and matching of the heat exchanger to the turbomachinery.
Recent develop-
ments in heat exchanger materials have done much to decrease both the size
and cost of these components for gas turbine applications.
The many types
of heat exchangers and their areas of application for various gas turbine
powerplants are represented in Figure 5.21.
This curve gives the approximate
cycle parameter boundaries for the various gas turbine heat exchanger
selections.
It can be seen that at very high turbine inlet temperatures
a ceramic type regenerator (rotary heat exchanger) is required while at very
139
-------
"-.
o
,
Qj
1-1
;:!
+J
n3
j..j
OJ
~
~
+J
tn
~
n3
.c.
&3
OJ
~
..-1
..Q
1-1
~
2-100
J\pprox. Temp.
Limits (Ext. Life)
~tP
2000
Ceramics
Refrac. Mtl.
1600
1200
400 Stnls
Mild Steel
R00
.--
-
m
400 [J
~
fZI
o
ReCllj). BOllndary
Pe~!en . Boundary
Tllbular Recllp.
Metallic Recup.
"\.b
Overall Turb. E f:f . =90%
~1rb. P.R. ~ 0.A5
Compo P.R.
1-
~ '<'-,s
% "';'
. <~
<\'},oO ~ ° 1>
IV ~ y.>
~((
r-..
~
'2>01. . 0
"\.~ :{.e,'2> ~~)..
CP~~ ~e
'2>-\>
~
-
-------
high pressure ratios a metal tubular recuperator (stationary heat exchanger)
is required. For the cycle parameters discussed above (turbine inlet
temperature = IBOOoF, maximum pressure ratio = 7) any of the heat exchanqer
types are possible, that is, a metal or ceramic regenerator or a tubular
or. plate-fin recuperator.
Thus, it appears that there are many heat exchanger-gas turbine
combinations possible and the problem now becomes one of determining the
heat exchanger which, when integrated with the turbomachinery, forms the
most compact, reliable, and low cost package.
In attempting to design a
heat exchanger that was compact, light weight, and low in cost, and which
also provides the high effectiveness and low pressure drop required by
the cycle thermodynamics, the concept of the bypass gas turbine was developed.
This concept developed from the fact that regeneration is really only required
at the low power levels since the basic simple cycle has acceptable fuel
consumption at the high power levels.
This is illustrated in Figures 5.19
and 5.20, where the specific fuel consumption of the simple cycle is about
.57 lbm/hp-hr and that of the 75% regenerated cycle about .42 Ibm/hp-hr at
a compressor pressure ratio of 7 and turbine inlet temperature of lBOOOp.
In addition, for the same turbomachinery or air-flow size a simple cycle
engine will be about 10% higher in maximum output power than the regennrativc
cycle uue to the heat exchanger pressure losses for the regenerative
cycle (48).
Thus, bypassing"the heat exchanger at full power appears to be
a most promising concept for reducing turbomachinery size and cost while
retaining acceptable fuel consumption at part load.
Further, the smaller
size regenerator designed specifically for operation at moderate speed road-
load levels, combined with its lower cost and weight, should make the overall
powerplant more suitable in an automotive application.
Having laid the groundwork for the regenerated gas turbine cycle
analysis, the difficult and time consuming task of matching the many possible
heat exchangers with the turbomachinery was undertaken.
Heat exchanger size,
weight, and cost, and powerplant output and fuel consumption were the major
characteristics to be considered.
The following heat exchanger types were
examined for both full time and part time (bypass) heat recovery systems:
L
2.
Tubular recuperator
Plate-fin recuperator
3.
Ceramic regenerator.
141
-------
The metallic regenerator was not investigated in any detail due to
the fact that its performance is comparable to the plate-fin recuperator
while its cost and complexity are somewhat greater.
Both recuperators and
regenerators were studied because both approaches are currently under
development by companies actively engaged in the automotive gas turbine field.
It should be emphasized, however, that none of the heat exchanger concepts,
except for the tubular recuperator, have been field or customer proven to
the degree where operational feasibility is assured.
'['he rotary disc or drum regenerator usually uses a high surface
density matrix, either metallic or ceramic.
The major problems associated
with these units are thermal distortion and sealing, the loss of seal
integrity causing rapid deterioration in engine performance.
The high
surface density results in laminar flow which permits favorable effective-
ness at low power and a less pronounced pressure drop (and thus power
reduction) at higher power levels.
The ceramic matrix costs are typically
low but the seals tend to be relatively expensive.
The recuperators (station-
ary heat exchangers) examined were of the tubular and plate-fin types and
counterflow, crossflow, and multi-pass arrangements were examined for both
types.
The tubular approach represents a well proven heat exchanger design
concept.
However, as will be shown, large bulk and excessive pressure drops
characterize this type of heat exchanger.
For the plate-fin design, several
compact matrices were examined with the plate/split-fin design proving best.
This type of recuperator is very compact, has relatively high heat transfer
characteristics, and does not produce excessive system pressure drops.
The
recuperators (both plate-fin and tubular) have certain advantages over the
rotary regenerators, such as no carryover losses, smaller leakage, and
no requirement for the complex and power absorbing rotary drive mechanism.
However, the rotary has the advantage of a certain degree of purging due
to the periodic reversing of flow without which, as in the recuperator,
burnout and fouling may occur.
Having discussed the pertinent heat
exchanger characteristics, it is now necessary to determine the turbomachin-
ery parameters which, when combined with these heat exchangers, will optimize
the powerplant design.
The regeneratiVe (or recuperative) cycle performance is a strong
function of pressure ratio and temperature. The maximum turbine inlet
temperature was determined to be 19000F and was set by sulfidation limits
142
-------
of the turbine blade material. The maximum turbine outlet temperature was
determined to be 11000F and was set by the Larson Miller stress rupture
characteristics of a 400 series stainless steel.
The design point compressor
pressure ratio was determined from Figure 5.2 where the locus of optimum
pressure ratios for lowest fuel consumption was determined for various turbine
inlet temperatures.
For the design point (turbine inlet temperature = 18000p)
the optimum compressor pressure ratio was about 6.0 (with 75% regeneration);
lower design pressure ratios than this will produce less feasible engines
for automotive gas turbines.
A single stage radial compressor was chosen
and the compressor map shown in Figure 5.22 represents a scaled version of
that reported by Kenny-Morris (9).
The compressor efficiency has been
discounted by 2 points to reflect diffusion and dump losses.
,
This map
has turbine inlet temperature lines overplotted and the design power output
(160 BHP) was established at a pressure ratio of 6.3 (on an 8SoF day) with
o 0
a turbine inlet and outlet temperature of 1800 F and 1100 F respectively.
The part load power control schedule is assumed to keep the turbine outlet
o
temperature constant at llOO F down to about 60% speed, which permits adequate
surge margin and prevents turbine hub and heat exchanger over temperatures
during acceleration.
Operation along this schedule (represented by dots on
Figure 5.22) results in the most efficient engine'operation.
The compressor outlet, turbine inlet and turbine outlet temperatures
which arc characteristic of this steady state operating schedule are shown in
Figure 5.23. At 100% power and speed, the turbine inlet temperature is
22600R and as power and speed are decreased, the turbine inlet temperature
must be decreased to keep the turbine outlet temperature constant at 15600R
and also to prevent the compressor from surging.
Both compressor and turbine
are limited to operation between 60 and 100% rotor speed, as speeds lower
than this result in poor acceleration characteristics.
As shown in Figure
5.23, the turbine outlet temperature is constant down to 60% speed (about
15% power) at which time it drops quite rapidly due to the rapid drop in
turbine inlet temperature. The turbine inlet and outlet temperatures at
idle (about 5 HP) are approximately 12000R and 10000R respectively. These
temperature paths will be used to determine heat exchanger performance
characteristics as a function of both speed and power.
The difference in performance between single-shaft and free-turbine
engines or other variations do not significantly affect the results of the
143
-------
7.0
6.0
5..0
U
H
Po.
..
0 4.0
-.-\
...
~
Q)
~
='
I/)
I/)
Q)
~ 3.0
Po.
2.0
1.0
.4
FIGURE 5.22
NOTE:
Stage P 1ytropic Ef iciency Red ced 2 %
Compres or Rotor Di eter, D = 4.75"
Nc c
Max. Co rected Spee , C. 100% = 87000 r.p;n
dP
o
\D
dP
o
co
dP
o
~
dP
o
r--
.6
1.2
1.4
.8
1.0
wVS
<7-
Corrected Airflow -
1b/sec.
Single Stage Centrifugal Compressor
Estimated Performance.
1411
E-to::
HO
E-t
rtI
..c::
U)
dP
o
o
..-t
1.6
1171
-------
p:;
o
2000
..
(II
~I
=='
+'
tIS
~
(II
~
(II
E-<
1500
1000
2500
5000
Percent Speed - lC) %
9 --
. ---
--0-
ompressor
utlct Temp.
20
60
80
100
40
Percent Power
FIGURE 5.23
Compressor Outlet, Turbine Inlet and Turbine Outlet
Temperatures as a Function of Power for the Sing1e-
Shaft Turbine.
145
1263
-------
regenerative (or bypass) study presented here.
Of course, the differences
do affect the economic feasibility of the engines, as will be shown in the
next section.
Thus, for simplicity, a single-shaft, one stage radial
turbine has been assumed for this analysis.
For the bypass concept, the valve is envisioned as being similar
to a waste gate butterfly valve used in certain turbochargers. Valve oper-
ating temperatures of over looofr will be encountered, which will require
the use of high alloy steels.
Valve leakage will not be a serious problem
due to the fact that the gases which leak are merely routed through the
alternate passage (bypass duct or heat exchanger) with only slight effects
on powerplant performance.
The most probable method of actuating the bypass
valve will be by a hydraulic servo associated with the vehicle transmission,
although solenoid or torque motor actuation is also a possibility.
A
schematic of the bypass engine concept is shown in Figure 5.24.
This
sketch does not show a bypass valve on the high pressure regenerator path
and, as will be shown later, this is due to the fact that the pressure drop
through this path of the regenerator remains relatively constant for all
power levels.
The three heat exchanger designs and the resulting power-
plant performance and design characteristics will not be presented as
evaluated both with and without the bypass concept.
5.4.1
Tubular Recuperator Concept
The first heat exchanger type evaluated was the well-proven tubular
design using a multipass crossflow configuration.
It was determined early
in the study that a full time recuperator of this type designed for operation
at all power levels (160 BHP maximum) would be prohibitively large.
To keep
the heat exchanger effectiveness and pressure drops at an acceptable level
required a heat exchanger with a total volume of about 3 ft3. In addition,
since the heat exchanger cost is related to its size, weight, and complexity,
the cost of this full time recuperator would certainly prohibit its use in
an automotive application.
Therefore, the only solution seemed to be to
design a compact tunular recuperator for operation only at the low power
levels and to bypass it at the high power levels.
In this way, the powerplant
becomes a combination of a simple cycle gas turbine for operation at high
power levels with a regenerative gas turbine cycle for operation at low'power
levels.
146
-------
4\6
I
RECUPERATIVE
FLOW PATHp
B
3
4
R
2
, NON-RECUPERATIVE
I FLOW PATH. ~
I .
L..._____---- - ,
EXHAUST 5
.I~ _6~
1
..
TO TRANSMISSION
C - compressor
R - recuperator
B - burner
T - turbine
V2- exhaust bypass valve
FIGURE 5.24
Schematic Flow Path Diagram for the By-Pass
Turbine Engine.
,
!i17
1070
-------
The tubular recuperator that was found to provide the best power-
plant performance in the most compact package is shown in Figure 5.25 with
the pertinent heat transfer characteristics given in Figure 5.26.
This
recuperator, a 3-pass cross flow arrangement, was designed for the flow rate
corresponding to 60% speed and an effectiveness of about .70.
A trade-off
between size and performance resulted in hot and cold side pressure drops
(~/P) of about .05 at the design point.
To determine the recuperator and
hence powerplant performance under all power conditions it was necessary to
determine the recuperator effectiveness and pressure drops as a function of
power output.
These characteristics are shown in Figure 5.27.
It becomes
evident from this figure that the bypass valve is only required on the hot
(or low pressure exhaust) side.
This is due to the fact that as the mass'
flow rate through the system goes up, so does the cold side absolute pressure,
thereby keeping ~P/F) relatively constant. However, the absolute pressure
c
on the hot side remains the same and consequently the (~P/P)H on that side
goes up almost as the square of the flow rate.
These recuperator character-
istics were incorporated into the turbomachinery cycle analysis and the
resulting powerplant performance characteristics are presented in Figures
5.28 and 5.29.
The curves on these figures represent the brake horsepower and brake
specific fuel consumption values for the simple cycle, the recuperative cycle
and the combination simple-recuperative, or bypass, cycle.
The advantage
of the bypass cycle is evident from the fact that the fuel consumption for
this cycle at 10% power is about 1.0 lbm/hp-hr, while at 100% power it is
about .67 lhm/hp-hr.
The fuel consumption for the simple cycle at 10% power
is almost 2.0 lbm/hp-hr, or about twice that of the bypass recuperative cycle,
while the fuel consumption at 100% power is about the same.
As mentioned
previously, the bypass valve would be similar to a waste gate butterfly
and it is apparent from Figure 5.29 that the best operating schedule would,
be for the valve to begin to open at about 50% power and be completely open
at approximately 85% power.
Only the exhaust gas side of the recuperator
is bypassed and the resultant power loss due to the full' ,power flow through
the air side of the recuperator is about 7 HP (Figure 5.28).
These results indicate that the tubular recuperator bypass concept
forms the basis for a compact, low-cost powerplant with good performance
148
-------
. ---
Ch~
~~
(~o $~~
(:\0 $C/ ~
~Q ~'~
~t: 0.
o~) -CJ~
/'
'1
4"
~-
~I}~
(~O Q<&t
Ci 1-.. 0
I)$ ""'110<&... Qt
:t-e)
~I}Ci
(.(': Q~t
~OIl1 ./"11
t
Q~L
Q-l.'
11$)
FIGURE 5.25
Bypass Gas Turbine Recuperator
149
1103
-------
M
"
N
~
j:1;
~
~
II)
:z:
0.050
0.040
~
0.030
0.020
0.010
0.008
0.006
0.004
0.3
Best Interpretation
Tube OD = .375 in.
4rh = 0.01237 ft.
Free flow area = (r =
Frontal area
0.200
Tube Wall Thir.kness ~ .025 in.
0.4
0.6
0.8 1.0
N 10-3
R x
FIGURL 5.26
~
exhaust
.
flow
.469"
.469"
Heat Transfer Area
Total Volume
2.0
4.0
10.0
6.0
8.0
(4rh G/14 )
Bypass Gas Turbine RecuperatQr Characteristics.
Exhaust Gas Side Heat Transfer Characteristics.
(Ref.: Fig. 10-14, Kays and London).
150
1102
-------
0.80
'II
..
(IJ 0.70
(IJ
QJ
c::
QJ
> 0.60
.ri
~ -
U
QJ
11-1
11-1
~ 0.50
0.40
(IJ
(IJ
S
QJ
~
(IJ
(IJ
QJ
~
Il.
0.10
Air Side
0.30
~Il.
0.20
-
./_---
o
o
20
40
60
80
100
Percent Power
FIGURE 5.27
Pressure Drop and Effectiveness Characteristics
for Tubular Recuperator as a Function of Power
Output.
151
-------
r
H,!)
]40
/
/
/
/
120
100
p,
::r:
r!I
~~
~
..
0 80
0..
Q)
(/)
~
0 .
:r:
(lJ
--.:
OJ
~I
r!I f;(J
40
20
o
f)
~~o
tiO
(-,0
80
100
Percent Power
FIGUPE ~~...: f3
Power Output Characteristics for Gas Turbine
Pn1Nerplant wit!! Tubular Recuperator.
152
1264
-------
1.4
1.2
.
\
.
).j 1..0
~
I
0.
~ '.
~
'"
e
..Q
~
u 0.8
~...
'I)
~
By- ass Re-
cup rative
Cycle
0.6
1 By-pass
I Recuper tive I mbination I Simple
0.4 Cy le Cycle Cycle
0 20 40 60 80 100
Peroent Power
FIGURE 5.29
BSFC Characteristics for Gas Turbine Powerplant
with and without Tubular Recuperator.
153
1265
-------
for all power conditions.
It should be noted here that the cycle calculations
were based on conservative, present state-of-the-art turbomachinery parameters
as described in Section 5.2, and that improved BSFC characteristics are possible
by using a less concervative approach.
For example, improved fuel consumption
characteristics are realizable by increasing either the turbine inlet temper-
ature or the maximum compressor pressure ratio. However, it was felt that
l8000p for maximcun turbine inlet temperature was a realistic value and that
the maximum compresso! pressure ratio of 6.3 was near enough to the present
state-of-the-art for a one stage radial compressor.
For the bypass concept
the a.r.qumcnt can also be made for usinq a. hiqher desiqn compressor pressure
ratio.
By assuming a two-stage radial compressor with backward swept blades
a compressor with improved surge margin can be obtained.
Such a compressor
wi th improved surge margin would allow part power operation at h~gher turbine
inlet temperatures than the single stage design, and these higher temperatures
require higher pressure ratios for optimum cycle performance.
This higher
pressure ratio design would provide better fuel consumption at both full power
(without the recuperator) and part power (with the recuperator).
The bypass
tubular recuperator design with an 8:1 compressor.. pressure ratio showed a
BSFC of about .60 Ibm/hp-hr for power levels ranging from 40% to 100% and a
very attractive BSFC of about .70 Ibm/hp-hr at 10% power.
However, prelimin-
ary investigations indicated that the one stage radial (with maximum pressure
ratio of between 6' and 7) provided the lowest cost, most compact powerplant
design.
Therefore, a selection of the single stage radial design was made in
which advantages in powerplant size and cost were obtained at the expense of
fuel consumption.
5.4.2
Plate-Fin Recuperator Concept
The next heat exchanger type to be evaluated was the metal, plate-fin
recuperator.
This recuperator concept required extensive investigation due
to the large number of matrix types available and the corresponding large
differences in friction factor and heat transfer coefficients.
Plain,
louvered, strip-fin, wavy-fin, and pin-fin surfaces were investig~ted with
respect to overall effect on powerplant size and performance; the results
indicated the strip-fin surface to be somewhat superior to the others.
This
surface, therefore, was used as the basic matrix for both full-time and part-
time (bypass) recuperators for the gas turbine cycle.
All cycle parameters
154
-------
(turbine inlet temperatures, compressor pressure ratios, etc.) were the same
as those used for the tubular recuperator analysis.
The strip-fin, plate-fin recuperator matrix is based on the data of
Figure 10-61 of Reference 2, presented here in Figure 5.30.
This particular
strip-fin matrix was chosen due to its characteristic fin pitch (18.82 finsl
inch) and plate spacing (.205 inch) which result in a favorable heat transfer
coefficient and an acceptable friction factor.
Using this type matrix the
arrangement which gives the highest effectiveness is a straight-through
counterflow as shown in Figure 5.31.
As shown there, the hot side free flow
area is twice as large as the cold side area which results in more equalized
pressure drops for both sides of the heat exchanger.
This flow distribution
maintains Reynolds numbers in the laminar (or sometimes transition) regime
for all operating conditions for both streams.
Several size heat exchangers using the aforementioned counterflow
matrix were examined.
As for the tubular recuperator analysis, the main
objective here was to determine the most compact heat exchanger-turbo-
machinery arrangement having the best performance.
To do this, several
heat exchanger sizes were investigated and the resulting powerplant per-
formance calculated.
To determine the recuperator, and hence, powerplant
performance under all operating conditions it was necessary to determine
the recuperator effectiveness and pressure drops as a function of power
output.
These characteristics are shown in Figures 5.32 and 5.33 for a
representative selection of possible recuperator designs (for both full
time and bypass operation).
These recuperator characteristics take into
account the varying compressor and turbine outlet temperatures as a function
of power output as shown in Figure 5.23.
Once again it is evident from the
pressure drop characteristics that the bypass valve is only required on the
hot (or low pressure exhaust) side.
These recuperator characteristics were
then incorporated into the turbomachinery cycle analysis and the resulting
powerplant performance characteristics are presented in Figures 5.34 and 5.35.
Examining Figures 5.32 through 5.35 in some detail brings out some
of the more important characteristics of the plate-fin type of heat exchanger.
First of all, Figure 5.32 shows that as the heat exchanger frontal area is
increased, the exhaust side pressure drop decreases in the same proportion
(for heat exchangers having the same length).
This is due to the fact that
the pressure drop is a function of the volume flow rate, which in turn is
155
-------
0.20
0.10
0.40
0.60
0.80
1.0
2.0
II-f
0.08
0.06
0.04
M
"
N
p::
Po.
Z
E-t
(/)
Z
II
0.02
.,...,
0.01
0.10
0.20
Reynolds Number, NR x 10-3 (4 r h G/ fA-)
FIGURE 5.)0
Heat Transfer and Friction Characteristics for
Plate fin Recuperator Matrix.
,
Fin pitch = 19.82/in.
Plate Spacing, b = 0.205 in.
Fin thickness = 0.004 in., nickel
Splitter thickness = 0.006 in.
Symmetrical Splitter
Fin Length flow direction = 0.125 in.
Fin area/total area = 0.841
Flow passage hyd. diarn., 4 rh = 0.005049 ft.
Heat transfer area/Vol. between plates, B=680 ft2/ft3
156
1172
-------
Air
'n
"""-- w
Exhaust
Out
Core
Geometry
'1°.01 in.
Air
Out
0.205(l\ir)
FIGt::RE 5.31
Gas Turbine, Plate-fin Recuperator.
Counterflow, Strip-fin, Plate-fin Heat Exchanger.
Surface = 1/8 - 19.82 D
157
1266
-------
P,--'P.
(].
.....
('.
II)
s
QJ
~
:;J
II)
Ir.
Q)
~I
P,
.28
Rcc\lp.Core Dimensions
.24
.20
.16
.12
o
x j x () in.
x H x 6 in.
12 x 10 in.
12 x 3 in.
/
/
/
/
./
./
-- ---
20
/
/
~\.
G~:i'
/
/
tJ
I
I
I
//
/
/
/
/
/
/
/
/
/
P
/
","
""
",
",
l\ir Si e
--- --
"",. ----
40
---
60
8U...
100
,...'
Percent Power
FIGURE 5.32
Pressure Loss Characteristics for Plate-fin
Recuperators as a Function of Power Level.
ISO
1267
-------
I .
.qn
.80
.70
....
w
,~
U)
U)
Q)
~~
Q)
~.> .60
. -.-I
~
0
"
'H
~
1>1
.50 Recup. ore
~ Dimensi ns
r o' 6 x x 6 in.
o 8 x x 6 in.
.40 ~ 12 x 12 x 3 in.
o 12 x 12 x 10 in.
.30
o
20
40
o
80
100
Percent Power
FIGURE 5.33
Effectiveness Characteristics for Plate-fin
Recuperators as a Function of Power Level.
159
1268
-------
160
140
120
100
H
:z:
I:Q
..
~
~ 80
&
CD
II)
~
g
CD
.~
~I 60
I:Q
40
Recllp. Cnr..
Dimem; inn::;
o 6 :x (, X (, in.
o 8 x A X 6 in.
~ 12 x 2 x 3 in.
() 12 X 2 x 10 in.
(IJ
.-4
20 ; I
o
20
40
60
80
Percent Power
FIGUPJ~ 5.34
Power Output Characteristics for Gas ~lrbine
Powerplant \vi.th Various Plate-fin Recuperators.
;1.60
luO
~
,.
1269
-------
1."
Rec:up. Dime sions
~im 1e cycle
\ 0 " x b x h in.
o
OR x B x h in.
l.1t
\ 6 12 x 12 x .3 in.
o 12 x 12 x 10 in.
1.2
~
~I
,(:
I
nl
.c:
In
"
r;
'~
LO
"
"
, )
i:t.
',)
:n
.R
"
.6
8
.4
.2
o
20
40
60
80
100
Percent Power
~"I',;um'; 5.35
BSFC Chn1:'acteristic5 fer Gas Turbine Po\\"erplant
with Various Plate-fin Hecuperators.
161
1270
-------
a function of the frontal cross sectional area. Figure 5.33 shows that the
larger the volume of the recuperator (all having the same strip-fin matr\x) ,
the greater is the resulting effectiveness.
Thus, the 12" x 12" x 10"
recuperator has the highest effectiveness value (.90 at 15% power).
The
recuperato:r:s chosen given a wide spread of powerplant performance character-
istics as -shown in Figures 5.34 and 5.35.
Figure 5.34 shows the power
penalty associated with the various recuperator designs for the entire
power range.
''he recuperators having the largest pressure drops experience
the greatest power losses.
It is apparent that to obtain full power with
any of the more compact recuperators
(6" X 6" x 6", 8" X 8" X 6") requires
bypassing the recuperator at or near the 100% power level.
In addition,
Figure 5.35 shows the fuel consumption penalty due to operating the more
compact recuperators at or near full power.
To examine the results presented in Figures 5.34 and 5.35 and to
conclude that one recuperator design is superior to the others for all gas
turbine powerplants would be a very meaningless task.
One must consider
not only the core size of the recuperator itself, but the size of the over-
all powerplant in conjunction with that recuperator.
Such things as mani-
folding, diffuser lengths, control valves, bypass ducts, etc., will greatly
affect both the size and cost of the total powerplant.
For example, the
turbine diffusion ducting required for the 12" x 12" face area recuperator
(144 sq. in.) is necessarily wider and somewhat longer than that required
for the 8" x 8" face area recuperator (64 sq. in.).
It is these things
which must be weighed against the performance curves of Figures 5.34 and
5.35 to determine the best powerplant with respect to size, cost, and per-
formance.
with these factors in mind, the various recuperator-turbine
powerplant combinations were analyzed and found to exhibit the following
characteristics.
The 6" x 6" X 6" recuperator powerplant experiences a large power
loss throughout the entire power range due to the large pressure drops on
both the air and gas side of the recuperator.
These. high power losses in
combination with the recuperators relatively low effectiveness result in
very poor BSFC characteristics.
The BSFC characteristics are only slightly
better than the simple cycle up to about 35% power and then become much
worse.
It would appear that this very compact recuperator does not improve
162
-------
the simple cycle fuel consumption enough to warrant its use with or without
the bypass concept.
The 8" x 8" X 6" recuperator powerplant experiences a moderate
pressure loss on the gas side at the high power levels while the air side
experiences an insignificant loss.
~he total pressure loss at full power
(without the bypass) is such that the simple cycle power output is decreased
by about 45 HP.
The effectiveness of this recuperator is high enough to
produce favorable BSFC characteristics at all power levels.
The BSFC for
the bypass recuperative concept is significantly better than the simple
cycle for the lower power range (100% better at 10% power), and approximately
equal to the simple cycle for the higher power range.
Therefore, it appears
that the bypass concept is attractive for this recuperator-powerplant com-
bination.
A minimum BSFC of about .70 lbm/hp-hr could be realized from
about 30 to 100% power with a full power capability of about 155 HP.
Only
the exhaust gas side of the recuperator would be bypassed (at about 80%
power), the power loss on the air side being only about 2 HP at full power.
The 75% increase in frontal area (and total volume) of this recuperator
over the previous one (6" x 6" x 6") has increased the powerplant performance
significantly while increasing the total powerplant volume (and cost) only
slightly.
The 12" x 12" x 3" recuperator powerplant exhibits very favorable
performance characteristics and does such without the need for a bypass.
The
power loss at full power is only about 7 HP which could be regained most
easily by slightly increasing the size of the turbomachinery. The BSFC is
below .60 lbm/hp-hr from about 40% to 100% power and is only .85 Ibm!hp-hr
at the 10% power level.
This is a great improvement over the simple cycle
performance; however, this improvement is paid for by the larger powerplant
volume.
The increase in volume is due not only to the core volume of the
recuperator but also the necessary counterflow manifolding and even more
important the required diffusing duct from the radial turbine to the recuper-
ator exhaust inlet.
The outlet area from the turbine is only about 15 sq.
inches and the exhaust must pass through this area, then through the necessary
diffusing duct, and finally be spread out over the face of the recuperator
t144 sq. in.) to insure against hot spots and uneven flow through the recup-
erator.
It seems likely that to obtain full advantage from the recuperator,
163
-------
vanes would be required to direct the exhaust flow to the outermost corners
of the face.
Of course, if the exhaust is diffused properly (to a Mach
number of about .2) the exhaust would tend to fill the entire recuperator
face area due to the back pressure caused by the recuperator.
However,
at low speeds where the back pressure is not large, the flow might not dis-
tribute itself evenly throughout the recuperator and the heat exchanger
effectiveness could suffer severely.
These are considerations which must
be taken into account as they may significantly influence both the size,
performance, and cost of the powerp1ant.
Finally, the 12" x 12" x 10" recuperator powerp1ant undoubtedly
gives the best fuel consumption due to its high effectiveness; however, it
suffers about a 20 HP loss at full power as compared to the simple cycle
due to its length and the resulting pressure losses.
This recuperator
is much larger than the others examined and the good performance is obtained
at the expense of size and cost.
This design exhibits an almost constant
BSFC of .50 Ibm/hp-hr from about 20% to 100% power and increased to only
.7 1bm/hp-hr at 10% power. The bypass concept would probably not be
attractive for this design as the horsepower lost could most easily be
regained by merely increasing the airflow size of the turbomachinery slightly.
Both full time and bypass concepts would require an exhaust diffusion duct
and flow guiding vanes similar to the 12" x 12" x 3" recuperator design.
In general, therefore, it appears that the recuperative, plate-fin
gas turbine exhibits the potential for a compact, high efficiency automotive
powerplant.
Any of the recuperators examined (except for perhaps the 6" x
6" X 6" design) are feasible and the particular choice depends on the trade-
off between size, cost and performance.
For best fuel economy, the obvious
choice would be the 12" x 12" x 10" design with no bypass, however, this
would dictate a larger p~werplant volume, weight, and cost. The most compact
design with the lowest initial cost would be the 8" x 8" X 6" recuperator
with an exhaust gas bypass.
This combination offers advantages with respect
to size, weight, and cost while giving somewhat poorer fuel consumption
than could be obtained with the larger heat exchanger designed for full power
operation without the bypass feature.
A schematic design of a single-shaft, bypass recuperative turbine
employing the 8" x 8" X 6" plate-fin recuperator is shown in Figure 5.36.
164
-------
Duct (Recup.-to-CoJl!bustor)
/'
compressor
Intake
Bypass
Valve
.'
, "" /' . Turbine
. .
_.-'
FIGURE 5.36 Schematic of a Single-Shaft Bypass-Recuperative Gas Turbine
Engine.
165
1193
-------
This engine would require about 2.50 cubic feet of envelope volume with a
one speed reduction gear box.
this volume are:
The components and assemblies included in
.1..
Radial compressor (4.75 in. tip dia., 6.3:1 max. press. ratio)
r~dial inflow turbine (5.25 in. tip dia.)
2.
:A.
Combustor
Speed reduction gear box (18:1)
4.
5.
8" X 8" X 6" recuperator core
6.
Exhaust bypass circuit
All scrolls and ducting.
7.
valve which would probably be operated as a function of throttle pedal
As mentioned previously, the bypass valve is a simple "butterfly"
position or could be actuated by automatic controls as a function of shaft
speed or pressure level.
The schematic design of the free-turbine bypass recuperative engine
employing the 8" x 8" X 6" recuperator is shown in Figure 5.37.
In this
design, the bypass recuperator has been offset from the engine centerline
to simplify the power takeoff.
The addition of the axial flow free-turbine
assembly and the necessity of two speed reduction gear boxes increases the
envelope volume to about 3.60 cubic feet.
included in this volume are:
The components and assemblies
l.
2.
Radial
compressor
(4.75 in. tip dia., 6.3:1 max. press. ratio)
3.
4.
Radial inflow gasifier (4.75 in. tip dia.)
Axial flow free turbine (6.6 in. tip dia.)
Combustor
5.
6.
Speed reduction gear box-drive (18:1)
Speed reduction gear box accessory (18 :1)
Recuperator (8 x 8 x 6 in. core)
7.
8.
Exhaust bypass circuit
All scrolls and ducting.
9.
To demonstrate the volume advantage of the bypass recuperative gas
turbine powerplant, it is compared with an internal combustion engine of
comparable output,in Figure 5.38.
A speed reduction gear box has been in-
eluded for the single-shaft turbine which reduces the output shaft speed
The six-cylinder internal combustion engine occupies
to about 4500 rpm.
166
'\
".
'"
.,
...
-------
.
Recuperator
Turbine
Compressor
FIGURE 5.:37
Schematic of a Bypass-Recuperative Free-Turbine Engine.
167
1194
-------
In-Line Six-Cylinder
Otto Cycle (I.C.)
250 CID, 160 HP
'"
Bypass Recuperativeiv
160 HP, Single Shaft
Gas Turbine
Speed Reduction
Gear Case
(Input = 80,000 RPM)
(Output = 4500 RPM)
FIGURE 5.38
Comparison of 160 HP Bypass Gas Turbine & Internal
Combustion Engines.
(Transmission for the Turbine Engine Nill be
Approximately 1.5 times the Size of that for
the Reciprocating Engine.)
163
1195
-------
approximately 9 cubic feet (at the minimum) as compared to about 2.50 cubic
feet for the single-shaft bppass recuperative turbine.
The transmission
for the single-shaft turbine, however, will be approximately 1.5 times the
size of that for the reciprocating engine.
Figures 5.39 and 5.40 are the photo-reductions of the full scale
preliminary design layouts for the single-shaft bypass recuperative engine.
It should be'noted that the scope of the pro~ram does not permit detailed
designs and therefore prime consideration was given to indicate the appli-
cation of the bypass recuperator concept.
Liberties were taken in turbine
designs with respect to actual blade optimization, bearings, lubrication,
fuel control systems and any other items which may be considered peripheral
to the presentation of the bypass concept.
The intent is to indicate a
method of improving the broad range performance of those turbines currently
under consideration for automotive application, without the addition of
excessive heat exchanger envelope volume.
In this, the single-shaft turbine, all power take-off drives have
been located forward of the compressor section as indicated in the left side
sectional view of the engine in Figure 5.39.
This arrangement facilitates
the mounting of the 8 x 8 x 6 inch bypass recuperator directly in the exhaust
gas flow path without the complication of turning this flow.
While the air
cleaners, starter and alternator are the only accessories indicated on this
drawing, it is clear from the dimensions (33.5 in. length, 18.65 in. height,
3
and 19 in. beam, and the resultant 5.5 ft envelope volume) that the bypass-
recuperative, single-shaft turbine does not represent a problem as regards
engine compartment space in the standard six-passenger vehicle.
A prelim-
inary weight investigation prepared for this compact single-shaft bypass-
recuperative design has indicated that the basic engine will weigh about
300 Ibs and the total propulsion system weight, including accessories, drive
line, and transmission, should not exceed 975 lbs.
This weight figure is in
line with the weight estimated for a similar advance recuperative design
proposed by United Aircraft (16). The United Aircraft design assumed a
o
higher turbine inlet temperature (2000 F) and resulted in a total propulsion
system weight of 905 lbs.
A comparable Otto cycle powerplant is estimated
to be about 100 Ibs heavier than the bypass-recuperative design, as shown by
the 1090 lb weight estimation of the 250 CID six-cylinder Chevrolet powerplant
presented in Table 4-II.(ECPE powerplant analysis).
It appears, therefore,
169
-------
i .:r-::-~
I.
!
!
I
t .
I
i
t
I~:.!.. _...:...:..
.'
. -,. ......_--
. .. - " .- . _l_':".-.
. ...
._, ..... 'no'
. . 'M......-'-.
'.':. "':-~." ..- '.~"::':'----'
'--0 .-.. ..... --
. ", .. -
. .
. ..
..
A
"''''
COMBUSTOR
~_. .:" TO THECOO..,.T""
18.6$
11.40
16.:500''''
-~~
SEC'IION A-A
,
~ ~
--_.-
'-- ----
------
- -'':'
._------ r-
00. . -.. ."
.. . .
...-..
.:... . I":
. .:-:-.---.-:--:'
,,-_....'..... ~.'_.._' :-.....--..
- ..-_u.._-...
.' .' .
,,11
r f i
[I i
I! I
I'
, .
,-1
I
-------
I
I
i
,
i
.i
~~~.:-
.,- ~..
ALTERNATOR
11.80
VIEW F-F
.- . ~ '.. ....~
. - - ~.
- '
-------~.- ----"-.-
-
~.
-,
. _. :!':'."'" :
,.
.,:F'-'
COMBUsTOR
AIR TO THE C0N!IUS1OR
Ra:U!'ERATOR
9.40
--
/
~
. ' ~.
- ,,'0. ...
"
". .....
, '
" ,
"
~_z: :':.. - .,.. . - .. ..". ". -:-- ::r:.~;~-' ". . ;. :.. - :.;'.: A-.
~:"'..J. <..,- - .....'-'-:~: ~;.''',., . . - ". ...
~' -. .~~}', . :~', .'-":",.' ,,~::,~-,--_.,,-,~~,~~,~,~_'_:'L.';.',:.~--~.'.{.'-,,~..:-r:~..: ..=,;:..-,:,~-.~.'.;-~:c_~,:;..'~::':'.~"'~';'_.~':"-:~-",~.~--,r.,.~._::~_.::..~.:..~,'-'~,-, - - - ,~., -',
:. ~:<~'. . :::::~~1~~fi~:;'~ ~ ~=.~<;~ .-' ~'~c ~~ -:z- ~--.'~=- '.>: "~;', ~.~~
.'
.'"
~:,,<-::-. .'
.. ,
. ....:-
j. .
~i~:"~ : ..:'::";-.~
. .;,.. - -.~ - "-::~;~;'~:"~~:<~.~':-. .. :"~~-:f,t'-;/ ~':';<:;'--.,f~T1~"~~f
- -.. -~---:';;:~,-q, .~:-. .- ~-'~~':~~-7-:-~~-'-:~"::;::-- .. --~- -. ,..~~~~,--:.-.-.:_~. -.. """..' . .
- - ,.."::. .........-;;._.-.~~ "':~!. . _.:~
."~
~.:-.p~..
--.
'.,
--
'T' .
.: '~f-"":~_',..
" ,'.. .."
~. '-
j", ,
."",,;:
'\t' .
. .
.'''.:',
"
..
:~t-'
.,.
"..
.r;;:
',i/!!"'- "
, .
. I'~.: .
'1"~~
~-~
;~'. .
'",+
"~.~ .
'.'
+~~;
.~£;" .
'Z"';;'"
"
.'
. '~{~
+~.
;~-:-
~;:'! .
, ..
, .:
'",
- '.
- .
1;-
.: :, i
~,-rz:=; 11
'-. _.~tJ
-------
that t1 small weight advantage is possible with an automotive gas turbine
powerplant, although in 20 years of development of an automotive gas turbine
no such advantage has been demonstrated.
In addition, automotive powerp1ant
weight is D0t of large concern in evaluation of competing systems except as
it affectr,; Gost and critical materials consumption.
The geometry of the bypass recuperator and associated ducting is
denoted in the side section of Figure 5.39 and the aft end view of Figure
5.40. Compressed air is ducted back along the left upper side of the engine
to the aft end inlet header of the recuperator where it is turned 1800 and
flows forward through alternate plate-fin passages in the recuperator.
the forward end of the recuperator, the preheated high pressure air is
At
collected in the outlet header and ducted forward to the combustor.
The
exhaust gases are diffused and flow aft to the recuperator where they are
allowed to either bypass the recuperator via the duct along its underside
or, with the bypass valve closed as indicated in Figure 5.39, flow aft
through alternate plate-fin passages in the recuperator.
Aft of the recup-
erator the exhaust gases, whether bypass or recuperated, dump into one
manifold to be finally ejected to the atmosphere.
It should be noted that for the design shown the compressor air and
some portion of the exhaust gases always pass through the recuperator.
The
bypass duct insures negligible exhaust system pressure drop at high power
level conditions.
The bypass is not necessary on the compressor air side
due to low pressure losses at any power condition.
Generally, it does not appear that the bypassrecuperator concept
would involve the degree of complication or bulkiness (ducting, etc.) which
might negate the favorable performance advantage OVer the simple cycle
turbine engine.
Relative to a fully recuperative turbine engine of compar-
able power output, the bypass concept does offer potential size and cost
advantages, particularly if tube type recuperators are utilized, while
providing comparable performance.
5.4.3
Ceramic Regenerator Concept
ator.
The last heat exchanger type evaluated was the ceramic rotary regener-
The engine turbomachinery and operating schedule are identical to that
172
-------
used for the tubular and plate-fin recuperator analysis.
A Cercor (trade-
mark of Corning Glass Works) disk. matrix was used having a frontal area of
1 ft2 and axial length of 2 inches, as shown in Figure 5.41. The regener-
ator design speed is 35 rpm and it exposes two-thirds of the face area to
the low pressure flow while the high pressure flow occupies the remainder
of the face area.
The regenerator seals are of the rubbing variety and are
spring and pressure loaded.
Leakage for this type of recuperator was deter-
mined to be about three percent; however, this may increase with time.
The
heat transfer and friction characteristics of the Cercor regenerator matrix
were obtained from reference 7 and are illustrated in Figure 5.42.
~e
regenerator matrix characteristics were combined with the turbomachinery
characteristics (see Figure 5.23) to give the regenerator effectiveness,
leakage loss, and pressure loss as a function of power output, as shown in
Figure 5.43.
A cursory study of the effect of leakage was made and it revealed
that BSFC increases on the order of 2% for each percent leakage and BHP
decreases at about 4% for each percent leakage.
These values were determined
at intermediate power levels.
These regenerator characteristics were then
incorporated into the turbomachinery cycle analysis and the resulting power-
plant performance characteristics are shown in Figure 5.44.
~e regenerative engine performance is quite similar to the perfor-
mance shown for the recuperative plate-fin engines.
There is only about a
17 HP loss associated with using the regenerator at full power and BSFC values
of less than .60 lbm/hp-hr are realizable from about 20 to 100% power level.
The relatively small power loss renders the bypass feature undesirable for
this engine configuration.
The engine performance is quite comparable to
the performance of the 12 x 12 x 3 inch plate-fin recuperative engine (without
the pass) of th~ preyious section.
To further illustrate how regenerator size
could influence the desirability of using a bypass, Figure 5.45 shows the
effect of regenerator size on engine performance.
As can be seen, the larger
the regenerator face area, and consequently the closer one approaches complete
regeneration, the lower the engine BSFC and the higher the power output.
fact, in the limit, for complete regeneration, an almost constant BSFC of
In
.4 lbm/hp-hr is obtainable throughout the entire power regime.
As the regen-
erator face area is made more realistic for an automotive application, however,
the BSFC gradually climbs and the horsepower decreases.
At a face area of
173
-------
!
Code 9690
Mounts
Core t.1atrix
FIGUEE 5.41
Gas Turbi:l8 C"unterf1ow Ceramic RDtary Regenerator.
Constant ;-opeed = 35 rpm
Fr6ntal Area, At = 1 ft2 (2/3 exh., 1/3 intake)
, r
Diameter, 0.0. = 13.5 in.
Length, L
') .
"" .l_n.
1711
1272
-------
0.100
0.080
4-1 0.060
0.040
0.020
M
........
N
~
P.
Z
E-4
(/)
Z
0.010
0.008
on
0.006
0.1
0.2
0.4
0.6
0.8 1.0
-3
Reynolds Number, Nm: x 10 (4 rh G/ tJ-)
F.IGURE 5.42
Glass Ceramic Regenerator Matrix Heat Transfer
and Fridtion Characteristics.
(Ref. :
Figur~ 6, Bayley and Rapley)
175
1175
-------
1-11 e
III .80
III
Q)
c.~
Q)
:>
.r!
+.1,-.
0\&1
G! --
I~
~
t4 .60
.')40
\..I
Q)
:>
0
~7 ~
~ .
uWt.
,tj II .030
~ III
(J)
~3
((j
~
ro
~
.020
.04
;T~
......
'11
I/)
.9 .02
<11
H
~
If}
UI
OJ ~
;..;
CI, Air Side
0
0 20 10 (,0 BO 100
Percent Power
f'IC~UHE ').43
Chp..rClcterist:ic:s of Cer,-'1.roic Regenerator as R
Function of Turbine P0wer Output.
176
1273
'; t
-------
I
100 .2
~. \-
:r::
1]:1
"-: }I
~'
(, 80 1.0
::01 ".,
Q! , ~
"
~ f .;
r :3
~i~
<11
:< :.",
,!"
~-' .'
(¥'. 60 ..~ ,)~
\
\ ...6
(,1)
- - -
, ,"
. ")
----
Simple CT~1
Regenerativ
C~rt;le
2.40
I
120
20
Q
o
20
.10
Percent Power
60
80
.2
100
FIGUF?;'~ :;. .14
1.4
~4
Power Out:::ut and '2:,PC Character-istics for Gas Tu;:~ine
(1 ft2 f~ce are~;
Powe}-plan.t: wi t'~, <4 Ritary CE>riUT'ic Regene:c?tor.
2 inch 1eng~,35 r~ill) .
" =,'-'=-,
1:::74
-------
1.8
1.6
1.4
~
.
~
. 1.2
p..
..c:
.........
.B
rl 1.0
'-"
u
~
U) .8
I:Q
.6
FIGURE 5.45
.4
'tI
-------
.5 ft2 (diameter = 10 inches) the BSFC is about .70 lbm/hp-hr from 25 to
100% power level and the engine experiences about a 30 hp power loss at full
power.
This compares favorably with the 8 x 8 x 6 inch plate-fin recuperative
engine and exhibits the same potential for using the bypass concept. However,
2
it appears that the better choice here would be to use the 1.0 ft face area
regenerator, and eliminate the need for bypass, obtaining more favorable
engine performance without a significant increase in total volume. The
2
horsepower loss (17 HP) for the 1.0 ft face area regenerator at full power
could be most easily regained by increasing the size of the turbomachinery
as necessary.
Further effort (engine schematics, design layouts, etc.) into the
regenerative engine concept was deemed unnecessary as several well known
automotive gas turbine research organizations have, and are now, devoting
much time and money to this effort.
From the results presented here, it
appears to be a very practical gas turbine powerplant for automotive appli-
cation, showing good performance and fuel consumption in a rather compact
package.
However, further development of the regenerative gas turbine is
required to attain a suitable automotive powerplant.
5.5
SINGLE-SHAFT AND FREE-TURBINE ANALYSIS
The majority of experimental automobile gas turbine engines built to
date have been of the free-turbine type.
However, the failure of this type
of engine to make any real progress toward the objective of a mass-produced
turbine powered vehicle has induced gas turbine manufacturers to re-examine
the myriad of possible cycle arrangements for a solution which may better
satisfy both the operational and economic requirements of a successful power-
plant system.
other arrangements which are primarily considered are the
basic single-shaft and various differential shaft cycles, all of which are
simply variations of the fixed or two-shaft types.
It has previously been noted that the fixed-shaft and ~ree-turbine
engines have greatly different torque-speed characteristics.
This is illus-
trated in Figure 5.46 where the maximum torque of typical free-turbine and
single-shaft engines is compared.
Whereas the single-shaft engine has negli-
gible torque below 60% output speeds and develops maximum torque only at rated
speed, the free-turbine develops maximum torque at stalled output speed equal
179
-------
;'! .0
1.5
,-0
~
0)
::I
0'
~
f?
Q)
:>
,.-I
+J
~
r-i
~
1.0
,0.5
o
o
.2
.4
.6
.8
Relative Output Speed - N/NO
.F'IGCHE
5.46 Comparison of '~aximum Torc:'cJe at- Speed Relationships
of Typical Free'~Tu;d)i:1e a.nd 3in;r1e-Shaft Er.gines
(19) .
180
Rating Point
1.0
1275
-------
to twice its t:nt"l,l.uc ilt the rating point.
At 60% of rated output speed, the
free-turb.i.nr: maximum torque is approximately 10 times the torque available
from the single-shaft engine.
Because of the limited torque at low speeds and the excessive time
required to accelerate the compressor from these low speeds (low power) to
full speed (where adequate power for acceleration is available), it is
necessary to operate the fixed-shaft engine in a speed range from about 60
to 100%.
The necessity of idling at high output speeds results in high
idle air and fuel consumption and therefore poor powerplant efficiency.
These are characteristics shared by all gas turbine types, including the two-
shaft versions, where the gasifier compressor must also operate between 60
and 100% speed.
The problem of low torque at near-idle output speeds is,
however, unique with the fixed-shaft engine and a more sophisticated trans-
mission (see Section 5.2) is required to provide acceptable vehicle acceler-
ation characteristics.
The free-turbine engine may be thought of as a fixed-shaft engine
with a compressible fluid torque converter attached to its own output shaft.
Output speeds from zero to 100% are possible and axial power turbine
efficiency varies with the h~b spouting velocity according to the relation-
ship developed in Section 5.2 (Figure 5.8).
The efficiency of this form of
torque converter is somewhat higher than the fluid torque converter used with
conventional automatic transmissions, although it is doubtful if a power
turbine assembly can be manufactured for the same cost.
In addition, the
free-turbine engine has certain matching characteristics which require some
form of variable geometry to provide optimum performance.
In Figure 5.47 the operating schedules of comparable free and fixed-
shaft englne designs are superimposed on the compressor map used for the per-
formance analysis of each.
As power is reduced, the operating line for the
free-turbine engine (with fixed geometry) is shown to drift farther away
from the surge line than that of the single-shaft engine.
This results from
the characteristic of the free-turbine engine to match to a lower cycle
temperature and pressure at any particular off-design compressor speed, and
lower cycle temperatures at part power result in lower efficiency.
Thus, it
is the primary interest of the various free-turbine matching schemes to
increase the part-load turbine inlet temperature and optimize the fuel con-
181
-------
~ Single S~art Engin
() Fixect- ometry Fre
Turbine
2260
7
6
5
!i
P.
0
o.-!
~ 4
C2
9 10 ~
<1.1 Mechani
~I 1
::3 Speed
en
III
<1.1
)oj
Il.
3
2
80%
70%
60%
\ 1
.4
.6
.8
1.0
1.2
1.4
1.6
Corrected AirFlow .- WiJ.~/ 61
- Lb/Sec.
FIGURE 5.47
Comparison at OperatinG, Characteristics of Comparable
Free-Turbine and Single-Shaft Enqines.
182
1276
I I ,
-------
sumption characteristics of the free-turbine engine.
Several means of accomplishing this goal are available ,.;,including:
1.
Variable geometry power turbine nozzles (VTN)
Differential or slipping clutches connecting the power
2.
1.
turbine shaft to the gasifier shaft
Variable compressor inlet pre-swirl.
Variable geometry turbine nozzles are proposed for many advanced I
engine designs.
This method of operation is illustrated by the schematic of
Figure 5.48 where axial power turbine nozzle vanes are shown to rotate about
a spanwise axis to set the proper stagger angle for any particular operating
regime (22).
The vanes can be rotated from the design position toward the
optimum part power settings or in the opposite direction as shown to admit
flow to the back side of ,the blade for a measure of braking.
For economy
or part-power operation the vane is rotated toward the closed position,
effectively reducing the nozzle flow area and allowing the power turbine
to match to a higher pressure ratio than the fixed geometry power turbine
could, thereby permitting higher turbine inlet temperatures (TIT) and
increased efficiency.
'l'he effect of this power-turbine nozzle area reduction is illustrated
in Figure 5.49.
As the load is reduced, the power turbine nozzle unchokes,
riding down the turbine flow characteristic quite rapidly as both pressure
ratio and corrected flow decrease.
During this period the gasifier turbine
remains choked, for the most part, since at lower power settings (lower TIT)
the gasifier turbine pressure ratio must be held nearly constant to supply
the power demanded by the compressor.
By reducing the power turbine nozzle
flow area 20% it is possible to match the power turbine to a higher pressure
ratio at a given corrected flow than that possible with fixed nozzles.
This
increase in power turbine pressure ratio is gained at the expense of gasifier
turbine pressure ratio.
Thus, it is necessary to increase the fuel flow and
TIT to raise the 'enthalpy of the combustion products entering the gasifier
turbine to match the turbine/compressor work.
The effect of increasing TIT
at any off-design compressor speed is, as previously stated, to increase the
part-load efficiency of the free-turbine engine.
The operating characteristics of similar free-turbine engines with
fixed and variable nozzle geometry features is presented in Figure 5.50.
At
183
-------
VARIABLE
NOZZlE
MODE
POWER
ECONOMY
BRAKING
FIGURE 5.48
Schematic of Variable Turbine Nozzle Vane Settings at
Different Operational Modes. (25)
18£\
1277
\ ,
tOr,
-------
1.1
1.
Fixed Turb ne Geome ry
:i' O.
o
r-i
Ct.
eo /'
Q) L.
+oJ
o
Q) /
H
~
o
(.)
Q)
~
.... 0.6
~
I'd
M
Q)
0::
0.
csign
Point
20% E;ffective Are Change in
Power Turrne 0..1.
"arta bre -,.ur ttrn e-r.e<"lme try
CPR ~ 6. J
T4- 18000F
Axial Power Turbl e
. ...
'~. 0
;4
P'r II""
..5 ''>b
FIG URE 5 .l~9
Effect of 20 percent reduction in power turbine
nozzle area on power turbine pressure ratio at
part load condition.
18S
1::n8
-------
() Fixed G ometry Ouer ting Point
<> Variabl Free Turb1 e Nozzle
~&bOR.
7
6
5
u
~
o
..-I
~ 4
&!
Q)
~
rn
rn
Q)
~
%
100
Meoh.
Speed
J
0%
2
1
0.4
0.6
0.8 1. 0
Corrected Airflow - W V'a1
°1
1.2
- Ib/sec.
1.4
1.6
FIG URE 5.50
Comparison of typical operating characteristics of com-
parable Fixed and Variable Geometry Free-Turbine engines.
18G
1279
-------
any particular compressor speed other than design, the variable geometry
8nqine operates at a higher turhine inlet temperature and pressure ratio.
T!-L~ 11TH charactr:r.istics shown are indicative of effective area reductions
~j f 'If> to 'WI!;.
By choice nf a la.r.
-------
105
~
s::
~ 100
~
8.
>.
()
~ 95
....
o
....
~
r;x.l
~ 90
....
.0
~
'"'
Q)
~ 85
a..
Q)
:>
'M
~
ro
oj 80
cc:
751. 0
FIG URE: 5.51
Re ative Area .~ 0%
.Des n
T10%
2.0 3.0
Turbine Total-Total Pressure Ratio
4.0
Effect of Variable Geometry on two-stage axial
turbine performance (18).
188
1280
-------
it a~;complishes this by means of well known and proven mechanical compon-
ents (24).
Basically, the variable coupling is used to transfer controlled
amounts of power from the gasifier shaft to the output shaft.
The gasifier
speed governor increases fuel flow to maintain relative constant compressor
speed, and the result is an increase in turbine inlet temperature at any
particular gasifier speed and therefore, an increase in cycle efficiency.
The effect of variable turbine nozzles (VTN) or differential clutch-
ing on fuel consumption is basically the same.
In properly designed engines
the efficiency losses attributable to variablenozzles is approximately equi-
valent to the losses resulting from the hydromechanical power transfer system.
Figure 5.52 shows the fuel consumption comparisons between fixed nozzle and
two VTN designs, one of which retains constant turbine inlet temperature
and the other constant exhaust gas temperature (22).
At approximately 30%
power (40 HP) the fuel consumption of the constant TIT engine is shown to
be approximately 35% lower than the SFC of the fixed geometry system,
whereas a 25% improvement is indicated for the engine which controls power
turbine exhaust temperature.
The latter is more indicative of the improve-
ment possible in a practical system, where at low compressor pressure ratios
the constant TIT engine would result in radial gasifier turbine hub over-
temperatures (radial turbines are stress limited at hub temperatures
o
around 1300 F).
Realistic improvement expected from variable geometry is
illustrated in Figure 5.53, which shows that the SFC of the variable
nozzle design is approximately 20% lower in fuel consumption than a com-
parable fixed geometry design (15) at the 30% load condition.
In short, variable geometry or similar matching methods are desirable
if not necessary free-turbine engine features.
The choice between variable
nozzles and variable coupling ultimately depends upon the success in develop-
ing variable nozzle systems competitive in the all important areas of cost
and reliability.
In addition, the ability to lock the shafts of differential
free-turbines provides favorable engine braking, power turbine overspeed
protection, and a means of handling gasifier turbine-driven accessory loads
(24) .
Variable nozzle systems are less bulky and lighter in weight, although
it has not been established that these are critical considerations for
automobile powerplant installations.
189
-------
1800
:z..
o
~ ,
~ G> 1600
~~
~
~ ~
:S~
~ ~ 1400
1200
1.4
Consta.n
"",,"
"","-
,,-
\ ."",,"
""'--
Tamb 0:: 8 F
Pam b ... 2 . 92 in.
-
--
Fixed Geometry
~ \
p..
.c 1.2
,c
~
rl \
~
o
..-I
~
p..
~
en
~
o
o
rl
~
1.0
0.8
()
..-I
fr1
..-I
()
G>
p..
tJ)
0.6
0,4
o
FIGURE 5.52
20
40
60 80 100
Output Shaft Power - hp
120
140
Fuel consumption characteristics for Free-Turbines with
Variable and Fixed Geometry (22).
190
l]f41
-------
? '
I... ")
~ 2.2
".
..,.
(;J
r:..,
i{3
'-,
u
r""
tB
1.8
u
rz.
~4
Q)
>
::
~ 1.4
G>
~
1.0
.0
FIGURE 5.53
VTN
NOZ:>;le G
eOmet
.t'y
20 40 60 80
Relative Output Horsepower - SHP/SHPMAX
Effect of Variable Turbine Nozzle G~ometry on fuel
consumption characteristics of Free-Turbine engine.
(1.5)
191
100
-------
Because power. turbine characteristics approximate those of the torque
converter in single-shaft engine systems, it can be stated that the performance
of optimized single and two-shaft engines is comparable.
For this reason it
was convenient to limit cycle analyses performed in this study (primarily
sections 5.3 and 5.4) to fixed-shaft engines, although this choice was not
intended to imply that the free-turbine has less merit.
Rather, this decision
was made to allow greater emphasis to be placed on gas turbine heat recovery
(recuperative or regenerative) analysis in general and the feasibility of the
bypass/recuperative or regenerative system in particular.
Choice of fixed-
shaft or free-turbine types does not significantly affect this analysis.
For
this study it is sufficient to say that although comparable in performance,
the choice in design type allows for trade-offs in areas of weight, size,
cost, and operational characteristics.
The single-shaft engine appears to offer potential benefits in the
area of cost.
It seems reasonable to assume that an acceptable transmission
can be provided well within the added cost of the power turbine-and-the engine
matching features required in free-turbine systems.
UARL (16) estimates
the direct cost of an 8 or 10 speed hydromechanical transmission with con-
trolled slipping clutch for the single-shaft engine to be no more than $50
more than that of the 3 or 4 speed (without a torque converter) transmission
required for the free-turbine engine.
The cost of the power turbine assembly
for the free-turbine is likely to surpass this figure several times.
The weight of the VTN free-turbine powerplant is potentially lower
than that of the single-shaft powerplant.
This results not only from the
more complex transmission required for the single-shaft system but also
from the necessity for a higher rated power to provide comparable vehicle
acceleration (due to the slope of the torque-speed curve).
Much of this
weight advantage is lost, however, if the free-turbine is of the differential-
clutch type.
Due to the fact that the free-turbine has superior torque at
low speeds, it is expected that vehicle acceleration goals can be met with
a somewhat less powerful engine than required with a single-shaft powerplant.
Therefore, the vehicle fuel economy with the free-turbine engine is expected
to be favorable, since road-load power requirements will be a larger percent-
age of maximum putput power for the smaller powerplant and the gas turbine
engine is more efficient at higher power levels.
192
-------
In summary, the characteristics of single-shaft and free-turbine
vehicle powerplants are compared in Table 5-IV.
5.6
COMPREX GAS TURBINE
The usual gas turbine is based on the compression, heat addition,
and expansion of a gas flowing through a compressor, combustor, and turbine.
Such engines deliver the maximum amount of power per unit volume and weight
of machinery.
While this approach is most advantageous for high output
engines for stationary power systems and for aircraft power systems' where
high power output per unit weight and volume are essential, the use of
corresponding levels of design sophistication for smaller automotive gas
turbines (lSO-160 liP) is not as feasible as the larger aircraft types.
The size of the rotating machinery is too small and correspondingly the
speeds and the resulting gear ratios are prohibitively high.
As an alter-
native, it would appear desirable to consider a form of elastic wave machine
in which compression and partial expansion are carried out in a single chamber,
a machine previously titled the Comprex.
The Comprex is basically composed
of several air passages and cells with shutters regulating the path of air
flow and consequently the power output.
It is proposed that the Comprex
gas generator would replace the conventional compressor and gasifier
turbine as the power source for a free-turbine type automotive powerplant.
Such a machine was invented by Brown-Boveri, the manufacturer of turbomach-
inery, diesel engines, compressors, etc., approximately 25 years ago.
It
was investigated for application in the aircraft engine field but the details
of the performance were such that the unit was sensitive to rotor speed.
Recently Brown-Boveri has visited several diesel engine manufacturers to
suggest its use as a turbocharger, since they had solved or alleviated the
sensitivity to rotor speed.
The name Comprex originates from the fact that the same rotating
machine compresses air and expands gas resulting in a much lower metal temp-
erature than is experienced in the turbine of the conventional gas generator.
Descriptions of Comprex operation are given in References 10-13.
of the rotating element of the Comprex is shown in Figure 5.54.
A sketch
The hot
expanding gas acts directly on the air to be compressed.
The action occurs
within the straight passages of the rotor.
The rotation serves only as a
means of connecting each channel with the desired port at the desired time.
193
-------
TABLE 5- IV
COMPARISON OF FREE-TURBINE AND SINGLE-SHAFT GAS TURBINE POWERPIANTS
system
Characteristics
Fixed Geometry
Free-turbine
Variable Nozzle
Free-turbine (VTN)
Differential
Free-turbine
Single-shaft
Turbine
Torque at low
output speeds
good
good
good
unfavorable
Transmission Required
3 or 4 speed gearbox; no torque converter
required
8 or 10 speed wi.th infini te-
ly variable hydrostatic or
controlled slipping cluth
Maximum Power Required
for Vehicle Acceleration
comparable
higher
Part Load Fuel Economy
low
comparable
fair good good
fair good excellent
small larger largest
low higher than VTN highest
highest lower than VTN lower" than free-turbine
comparable lowest
I-'
~ Vehicle Braking
none
Overspeed Protection
none
Size (including transmission)
small
Weight
lowest
Cost
lowest cost of
free-turbines
Use of Critical Material
-------
PI GURE 5.:#
Sketch of Comprex Rotor
----- .-~- ..----
-------~------_._.
IN:J~ AIR
1'(' OO"!'RlX
( lA'.-FR;:;.;lJ!,f.)
REX
ROTOfl
CAS
:c'T'. ~-. 5.55
S:-::lcmatic of JI.utomobi1e PmJerp1ant r.;sin9
C)r"'FJ~e:x Gas Genera.tor and Free Turbine.
J. f.).s
1206
-------
A schematic diagram of an automobile powerplant using a Comprex gas
generator is shown in Figure 5.55.
Hot gas from theccombustor is first used
to compress the ambient pressure intake air and is then discharged into a
free turbine to supply automotive power.
Valve timing is accomplished by
the motion of the rotor between two stator plates which have ports so
arranged as to control conditions of the cycle.
There are a large number
of passages in the rotor, all of which participate simultaneously in differ-
ent phases of the cycle.
The multitude of channels results in a continuous
flow of compressed air through the system.
Figure 5.56(a) shows a wave-flow diagram of an unwrapped rotor (11).
In order to follow the wave and flow phenomenon through one complete cycle,
consider first the channels at the top of the diagram.
These channels are
filled with ambient air at ambient pressure and temperature.
As the channels
move downward, the high pressure hot gas port on the right side is uncovered
and a shock wave followed by the interface is driven to the left.
wave travels much faster than the air and interface which follow.
The shock
The air
behind the shock is compressed to a level higher than ambient and this air
also possesses a velocity head, which if recovered, will provide additional
compression.
As the channels move downward, the high pressure air outlet port
opens at about the time the normal shock reaches the left end of the channel.
Then, with both ends of the channel open, the air and gas continue to move
to the left carried by the momentum already established.
Next, the high
pressure gas inlet port on the right closes causing an expansion wave to move
to the left, reducing the pressure of the hot gas and reducing its leftward
velocity to zero.
At the time the expansion wave and the interface reach
the left end of the channel, the left port closes.
At about the same time,
the low pressure gas outlet port opens on the right permitting the hot gas
to expand at high velocity out of the right end of the channel.
At this time,
an expansion wave moves through the hot gas towards the left end of the channel.
When the expansion wave reaches the left end of the channel, the low pressure
air inlet port opens and the momentum of the hot gas moving at high velocity
to the right causes a fresh charge of ambient air to be sucked into the chan-
nel.
When the channel becomes filled with ambient air, the port~ close and
a new cycle is initiated.
Figure 5.56(b) shows how the compression and expan-
sion waves and particle paths move back and forth across the channels.
196
-------
V'
..
I,' ;:
-. u_- , -_._---~
' ~. it ", '. ~~~.,~:~~~~- '
-.:~-:-:" '.~ ':. ..: ~~~ :.:.-
.&fO...:""..:!:'"_~_.~_._.. ,.. - -: ...w_..DU'hU
.~':-..~ ~:-:= ~::: ~ : :=- c.A)...
. ~!.~-'~~' ...--: :. -: =- : : - :
- CuT -J.~'-~..~:~[{';~~::;::
J ...-.........-..
. ~-~--~_.}--L..
'~4f.~t...,~..~.1. ._.7~ ....- .:
.~..t. .: :~L_-:-=. -- -
--I.. .
- - -- - - .
- ------
-.ow ..",..8'
""L I.
. - .. ~ ~~~"'..'''u..
- - =--=~.~ "'''O"T
f.'-":.. .:.- -- - ...... '
ua
o ".,., NUt,,-t AI' ,n ""
~O- P8U..,.' AI' ~I".
="10. Pun"" Ala "CWI"
-,,0.0- '."'1".' aA~ MW'.
. "I'" PCltwal ""'A".IIT
$"_. PI"'"'' 6A''''''
FIGURE 5..56 (a)
Comprex Wave-Flow
Diagram.
(Upper case - total cond.'
(LOwer case - static cond.'
P -
~,
S
,~
V' -
prEilssure
teniPera,~ure
Spe~ of ,sound
"
Mach no.
veloci ty ,
t '"
S =
V ==
P ==
Compressed
air
,t =
:T =
S =
,M =
v=
P =
~ =
. . " '.!1:~~
9;~S R , ':f:';~t:
9lsoR . :~~~
14'9gsPs! ::~~:~
., I ~~~,....
o ""'-:'i:.~:'1:-
11i9f~S'! ~~t~~.
- -1'10'\."...
3~. a i :~~.~~
5.).c:!a -gn j \:J«',,~
..., , 1,- .'
Shock Mach = 1.75
Shock Vel - ,2000 fps
F.t'.~t' '.ATIClE PilTH
..
H P 'fA OUT
" .
-:-
. H.P GAS '"
j
--
--
---
l P A"~ ,.. -
,
,
,
'"''
,
,
,
,
,
,
"
.
.............
-------
A preliminary calculation has been made to determine if the hot
compressed gas could, in fact, sUfficiently compress ambient air to keep
the cycle in operation.
The compressible flow charts of Reference 4 were
used in making the calculations, and the results are presented in Figure
5.57. It was assumed that the pressure
o
4 atm. and the temperature 2180 R.
air channel were 1 atm. and 5450R.
downstream of the combustion was
The initial conditions in the ambient
The calculations indicate that when the
hot gas port is opened, a shock wave will move to the left at a speed of
about 2000 fps followed by the compressed gas, interface, and hot gas
moving at a velocity of about 1100 fps.
The static pressure in the compressed
gas is 3.42 atm., but the total pressure is 5.13 atm. which should be more
than enough to offset duct losses and combustor losses.
Potential merits of an automotive powerplant using a Comprex gas
generator and free turbine are as follows:
1.
Gas generator is a low speed machine (on the order of 5000
rpm), therefore, stresses are much lower than in conventional
gas turbines.
Also, a conventional transmission could be used.
2.
Channels in gas generator alternately flow hot and cold gases,
so that high temperature alloys are not required.
3.
Cycle pressure ratio is not a function of rotational speed as
with gas turbine so that part-load efficiency could remain
high.
4.
System could be designed to "over scavenge" to provide
cooling for free turbine so that high temperature alloys
5.
f~r free turbine are not required.
system using high pressure ratio might not require heat
6.
exchangers, recuperators, regenerators, etc.
System is far simpler than Rankine systemdand eliminates
many of the' disadvantages of the current gas turbines.
It therefore appears that the Comprex gas generator has the potential
for being incorporated into a successful automotive powerplant.
5.7
CONCLUSIONS
The preceding investigation into various gas turbine powerplant concepts
for automotive application has generated the following conclusions:
198
-------
1.
In general, it appears that any automotive gas turbine power-
flant will be limited to certain maximum temperatures and
pr~ssures as determined by the availability and cost of
mat~rials.
The number and sizes of compressor, turbine and
heat exchanger Gomponcnts must be kept to a minimum to reduce
the very important initial cost.
For thi.s reason, (\ gas
turbine with the followi.ng characteristics (representing
present state-of-the-art technology) was deb~rmined to be
best suited for an automotive appli€ation:
a.
A maximum turbine inlet temperature of l8000p.
A one-stage radial compressor and turbine (with present
b.
2.
state-of-the-art maximum pressure ratio of about 7).
In general, the closed cycle gas turbine concepts do not appear
as feasible as the open cycles due to their lower efficiencies,
increased complexity, and cost.
This is t.rue for all the
Glosed cycles using both air and other exotic workillq fluids
~nd also includ~s the sophisticated hckeret-Keller cycle.
J.
For thc~ open cycle concepts, the simple.: cycle employing a single-
s'.:.a(fe radial compressor and limited to 18000F maximum turbine
inlet temperature was found to have unacceptable fuel consumption.
This led the way for the three heat recovery concepts, including
both recuperative and regenerative heat exchangers.
It was
shown that all three heat recovery concepts can offer similar fuel
consumption from idle to full power; however, trade-offs are
possible with regard to size, cost, and reliability in choosing
a system which best matches the particular'powerplant application.
The engine with the compact tubular recuperator requires a
scheduled bypass to achieve optimum performance.
This concept
provides good reliability and has been field tested and proven.
Its drawbacks are its large size and supposedly higher cost
than the plate-fin recuperator or ceramic regenerator.
The plate-fin recuperator and ceramic regenerator examined
showed similar performance characteristics, both giving very
good powerplant fuel consumption.
The cost of the compact
plate-fin recuperators and ceramic regenerators are comparable
199
-------
and the differences in cost in mass production is merely
speculation at the present time.
The plate-fin recuperator
is preferred over the ceramic regenerator due to the fact
that the plate-fin type- does not ~equire the complex and
costly drive system of the regenerator and is less suscept-
ible to seal and bearing problems.
The particular recuperator
chosen will depend on the available space and the associated heat
exchanger manufacturing cost; however, a very compact, low
initial-cost arrangement appears to be an 8 x 8 x 6 inch recup-
erator operating with an exhaust-side bypass valve.
This
powerpalnt package will provide the power of a medium pressure
ratio non-recuperative cycle for acceleration and high speed oper-
ation, while retaining the efficiency of a low pressure ratio
recuperative cycle at the lower road-load poer levels.
Although
smaller and less costly, this bypass arrangement offers some-
what poorer fuel consumption than can be obtained with a larger
heat exchanger designed for full power operation without the
bypass feature.
In effect, it appears that an attractive
automotive gas turbine powerplant can be developed based upon
the plate/split-fin recuperator (either bypass or full-time)
which minimizes the very important initial cost while retaining
acceptable fuel economy at road-load power levels.
4.
Subject to the demonstration of (a) a suitable transmission for
the single-shaft engine and (b) a reasonably priced power
turbine assembly, the choice between single-shaft or free-
turbine types allows for additional trade-offs in the areas of
weight, size, cost, and operational characteristics.
The
single-shaft engine with an appropriate transmission should
be less costly, though larger and heavier than the free-turbine.
The free-turbine, which has good torque at low speeds, should
be smaller in size and power for the same vehicle acceleration
characteristics.
5.
Finally, a very preliminary investigation of the concept of a
Comprex gasifier indicated that it has generally favorable char-
acteristics for application to a free-turbine type of automotive
installation.
More analytical studies 'are required to adequately
evaluate its relative merit compared to other turbine power systems
before consideration can be given to development of prototype
hardware.
200
-------
6.0
STIRLING ENGINE
6.1
INTROWC'l'ION
One of the engines to be considered for low emission automobile
powerplants is the advanced design stirling Engine.
The basic concept
for this powerplant dates back to various types of hot air engines
which were used extensively in the nineteenth century.
However, these
early engines did not mature into a practical pOweD)lant which could be
consid(~red to be competiti tve for stationary and mobile vehicle applica-
tioos until the Philips Research Laboratories in Eindhovcn, Notherlands,
initiated a scientific and engineering program shortly before World War
II.
Through this project, the Stirling engine progressed to a high
degree of refinement in the 1950's and 1960's.
Philips has developed several varieties of Stirling engines, and
several companies throughout the world have been licensed for design,
development, and manufacture of the powerplant.
Licensees include:
1.
General Motors Corporation in the United States
United Stirling in Sweden
2.
3.
M.A.N. in Germany
Ford Motor Company
4.
At one time .Japan was interested in the powerplant, but no activity has
bc~n reported i.n
r-eCf:llt years.
General Motors was extn~mely active in
the research and (k'volopment of Stirling engines, however, since 1970 General
Motors converted their license from an exclusive agreement for North
America into a non-exclusive agreement.
Ford Mo~or Company has recently
obtained a license for work on the Stirling engine.
Probably the most active organization (in addition to Philips in
the Netherlands) at the present time is United Stirling.
This company
has an aggressive program which is generally directed to heavy duty
vehicle and stationary Stirling engines.
One of their most promising
versions is a v-a engine.
This engine is effectively adapted for power
levels higher than the 150-160 HP class which is required for the OAP
representative automobile.
Of the several configurations of the modern Philips Stirling engine,
all have about the same thermodynamics, friction losses, and performance
with a given working fluid.
Engines have been fabricated for a wide range
of power levels, varying from single cylinder engines operating at 3 KW
(Allison and General Motors Research), to multi-cylinder engines operating
at 30, 40, 200, 400 HP (Philips).
2;J}~
-------
Three working fluids have been used for the Stirling engines: air,
hydrogen, and helium.
Air was used in pre-1940 era (hot air engine) and,
in general, the efficiency was of the order of 10% at low operating speeds.
Gne innovation introduced by Philips concerned the recognition of the import-
ance of (1) maintaininq high conductivity and specific heat of the working
fluid, (2) minimizing the void volume of the heater, the regenerator,
and the cooler, and (3) minimizing the fluid pressure losses in the closed
thermodynamic circuit. For vehicle applications, either helium or
hydrogen can be used as the working fluid, with hydrogen having the distinct
advantage vIi th respect to high speed operation, specific power output and
efficiency.
At a gross efficiency of 40% (which is often quoted as a practical
operating value) the engine attains almost 60 HP/liter on efficiency.
For
automobile engines, where size is of prime importance, hydrogen should be
favorably considered as the working fluid since a given power level can be
achieved in an engine having only 64% of the displacement and, therefore,
about 64% of the engine bulk volume of that required witt helium.
The
weight of the engine should decrease correspondingly.
'!'he effect of
working fluids on performance will be discussed at greater length in
Section 6.4.
6.2
THE PRINCIPLE OF 'I'HE STIRLING ENGINE
'rhe Stirling engine is a hot gas external combustion engine which
compresses a quantity of gas at low temperature and expands it at high
temperature.
Theoretically, the Stirling engine has the same ideal
efficiency as a Carnot cycle operating between the maximum and minimum
gas temperature.
The cycle consists of the following four processes:
1.
Constant temperature compression, during which
waste heat is rejected.
2.
Constant volume addition of heat which was stored
3.
in the regenerator from the preceding cycle.
Constant temperature addition of heat as the piston
4.
moves from minimum to maximum displacement.
Constant volume storage of heat as the working fluid
returns to the initial low temperature condition.
The gas temperature is changed periodically by causing a "displacer piston"
(hereafter simply called "displacer") to transfer the gas back and forth
between two spaces, one at a fixed high temperature, and the other at a
fixed low temperature (see Figure 6.1).
If the displacer is raised, the
gas will. flow from the hot space via the heater and cooler ducts into the
cold space; if the displacer is moved downwards, the gas will return to
,70::'
-------
the hot space along the same path.
During the first transfer stroke the
gas has to yield a large quantity of heat to the cooler; an equal quantity
of heat has to be taken up during the second stroke from the heater.
The
regenerator shown in Figure 6.1 is inserted between the heater duct and
cooler duct in order to prevent unnecessary wastage of this heat.
It is
a space filled with a compact, high density, porous material to which the
hot gas yields heat before entering the cooler; when the gas streams back,
it takes up the stored heat again prior to its entry into the heater.
The displacer system, which serves to move the gas through the
heating and cooling elements periodically, is combined with a power piston
(hereafter simply called the "piston") which compresses the gas while it
is in the cold space and allows it to expand while in the hot space (all
dead spaces in cooler, heater, etc. being disregarded).
Since compression
takes place at a lower temperature than expansion, a surplus of work
results.
Figure 6.2 shows four states of the cycle through which the
working fluid passes.
The fluid is assumed to undergo displacement as a
function of time as plotted in Figure 6.3.
The ordinate in band (E) repre-
sents the variation in the volume of the hot space, and the ordinate in band
(C) represents the variation in "the volume of the cold space.
The volume
variations are plotted separately in the lower part of the diagram.
Figure 6.4 shows the pressure volume diagram of the cycle represented by the
cyclic motion of Figure 6.2.
Referring to the T-8 diagram in Figure 6.4, the constant volume
heat input and output to and from the regenerator is shown to be equal
for the case where constant specific heat is assumed.
input to the regenerator,
Thus, for heat
QIV-I : Cv (T3 - TI)
and for heat recovered from the regenerator, where
QIII-I : Cv (T3 - T1)
where
T4 = T3
T2 : T1
The net cycle efficiency, then, becomes
r( : T3 S3 - T1 81
T3 53
or
T1
fl=l--
TJ
where
83 = 81
203
-------
F;tGURE 6.1
FIGURE 6.2
Heater
Regenerator
Cooler
Hot space
,,"
r ':
Power piston
!: t
Displacer
I.
I'
d. .
;1, ;
Iii' .
,I ,
! I
I':
"
j: II
i!.1
I' I
I: I
I' .
, '
'1:1:
':I,i
I' :
Cold space
~: . '
'i
1 ,
\; I
Principle of Displacer System.
As displacer strokes,
gas moves from hot to cold spaces via heater, regenerator,
and cooler and back.
I
i
!
II!
/I:
"1
",
II
III .
I;
IV
Principle of hot-gas Process.
Assume piston and Displacer
move discontinuously, then divide cycle into four stages.
I - Piston BDC, displacer TDC, gas in cold space
II - Displacer TDC, piston TDC, gas compressed @ low temp.
III - Piston TDC, displacer mid-stroke, gas displaced to hot space
IV - Piston BDC, displacer BDC, gas expanded.
..',0
204
1108
.~
,
-------
i--
I
i
FIGURE 6.3
I
I
I
I
I
I
I
II
I
III
IV
.... t
Discontinuous displacement of Piston (P) and Displacer (D)
plotted as function of time.
Band E = volume variation of hot space (VE)
Band C = volume var. of cold space (V )
c
Both are plotted separately at bottom.
205
1109
.," -J .
-------
T
FIGURE 6.4
(2) Consta
Volume
output
the re
P
II
III
II
(4) Const nt Volume Heat
Input to the Regenerator
IV
I
V
3
Heat Input to
Rege erator
II
Heat output from
Regenerator
S
P-V and T-S of Ideal Stirling Cycle.
206
.'
128,1
-------
This is the expression for Stirling and Camot efficiency.
They cu;e
equivale~t in the ideal case.
For the temperatures used by Philips
(1300oF a~d 6~oFi
~he ideal efficiency is of the order of 71%.
Including
friction and heater losses, actual brake efficiencies of 40% are achieved,
hence the actual performance is about 57% of the ideal.
This is approxi-
mately equivalent to the relative cycle efficiency of a diesel engine.
In a practical version of the engine, the movement of the piston
and displacer must be continuous, (not discontinuous, as they have been
assumed to be in Figure 6.4) with the continuous movements being obtained
with the aid of some kind of crank and connecting rod mechanism.
It will
then be impossible to distinguish any sharp transitions between the four
stages but this will not alter the principle of the cycle (or detract from
its efficiency).
These movements of piston and displacer are indicated in
Figure 6;5, in which the volume variations of the cold and hot spaces
have again been plotted separately.
The only essential condition for
obtaining a surplus of work is that the volume variation of the hot space
should have a phase lead with respect to that of the cold space.
This is
equivalent to requiring that the appropriate P-V diagram, shown in Figure
6.6 should be traced out in the clockwise direction.
6.3
DRIVE MECHANISM
There are various ways of making the piston and displacer perform
the desired movements.
Only those which are suitable for application to
automobiles will be discussed at this time.
since 1953, most of the
stirling engines have been based on the rhombic drive system which is
shown schematically in Figure 6.7 (Ref. 28).
This system was found to
work quite satisfactorily for the engines developed at Philips and at
General ~btors Research.
It is somew~at complex and the engine has a
high specific weight.
However, it did achieve major objectives for the
engine, among these being a mechanism coordinating the two pistons by a
linkage which resulted in essentially linear motion, and a method of
minimizing the need for lubrication.
The linkage mechanism eliminated
vibration, even in the case of a single cylinder engine, and although
complex, achieved 90% mechanical efficiency.
Engines using this mechanism
were built for both single cylinder and four cylinder versi.ons.
They
were fairly quiet and they ran satisfactorily, at least for the develop-
ment phases.
207
-------
FIGURE 6.5
D
v
c
I
I
I
I VE
~oc.
Continuous Motion of Piston and Displacer with Displacements
plotted as Function of Crank-angle ~).
Indistinguishable.
Stages of cycle now
208
1110
-------
,-/
'FIGURE 6.'6'
,--"
I .
I
p
t
.... V
The P-V Diagram for the Cycle Represented in
Figure 6.10.
209
1112.
-------
I I:.,
......
(6) {}ISPLACER PISTON~
HEATER (14)
(17) CYUNDER
REGENERATOR (15)
COOLER (16)
, '
( 1) pm.,'ER p I STaN
I'NTERCONNECT I NG DUCT
(2)
"
"
(13)
(5 )
"
SYNCHRONIZING
'GEARS (10')
(5' )
(9)
FIGURE p'. 7:
Schematic of Basic Rhombic Stirling Cycle Engine.
210
1024
-------
The rhombic drive mechanism allows the buffer pressure to be applied,
while ke(~ping the crankcase at atmospheric pressure.
This drive mechanism
also has the advantage of allowing even a single cylinder engine to be
completely r~lanced.
It consists of twin cranks and connecting rod mechanisms
offset from the central axis of the engine; the cranks rotate in opposite
directions and are coupled by two gears.
Fixed to the piston (1) by way of piston rod
(2) is a yoke (3).
One end of the yoke is linked by connecting rod (4) to crank (5), the other
end by connecting rod (4') to crank (5').
The displacer is actuated by
a precisely similar arrangement:
the displacer rod (7), which passes
through the hollow rod (2), has fixed to it a yoke (8), which is linked
to cranks (5) and (5') by connecting rods (9) and (9') respectively. If
(9) and (9') are given the same length as (4) and (4'), the two pairs of
connecting rods will form a rhombus, of which only the angles vary when the
system is in motion; it is for that reason that the name "rhomJDic drive"
has been adopted.
Gears (10) and (10') ensure exact symmetry of the system
at all times.
Since the two crankshafts are geared together, the entire
shaft output can be taken off either.
The symmetry of the system and the coaxial arrangement of piston
and displacer rods make it an easy matter to avoid pressurizing the crankcase.
The seal (11) for the displacer rod is inside the hollow piston rod.
One more
seal located around the piston rod (12) is all that is necessary to form a
comparatively small cylindrical chamber (13) under the piston, which is
separate from the crankcase.
the desired buffer pressure.
This "buffer space" can be filled with gas at
The minimum permissible volume of the buffer
space is determined only by the range within which it is desired to limit the
pressure variations.inside the chamber.
In a multi-cylinder engine the buffer
chambers can be inter-connected; this allows the volume of the individual
spaces to be made even smaller.
The piston and displacer movements of the
rhombic drive are displayed graphically in Figure 6.8.
There are several mechanisms in addition to the rhombic drive which
can be used to achieve the proper phasing of the gas flowing through the
heat exchangers.
One of these consists of a 4-cylinder, double-acting
engine using a swash-plate to convert reciprocating motion into rotary motion.
For this design, used by G. M. Research, one double-acting piston is used
for each of the four cylinders, and the heater tubes for all cylinders are
heated by the flame of a common combustor as shown in Figure 6.9.
Each of
the cylinders has a separate regenerator and waste heat exchanger (cooler),
as indicated schematically in Figure 6.10.
The single piston is hollow
LL..
-------
FIGUJC,;':; 6.8
I
...
":: 0
--.. -
The \.'olume Variations of VE' Hot Space,
and Vc' Cold Space, of a Hot-Gas Engine
Using a .Rhombic Dr: i ve Mechanism.
7.12
1284
-------
,. rv
......
w
......
N
00
U1
Combustion
Air in
........
Recuperator
Heater
(He Coils)
Combustor
o ,.cP--_c:.
~",..- -.,
o ,~..:.,c:)~co
O ~,iI'''---- ..
81 o=o"o;:::)~"
. ..,' d' -C3CDCSI>
-------
FIGURE 6..10
Schematic Representation of Double-Acting St.irling
Engine Flow.
2111
Regenerator
(Screen
Matrix)
Cooler
(Water-
Tube )
1286
-------
to minimize heat transfer from the hot space to the cold space.
A "roll
::jOe}:" seal is located on the piston rod below the piston to eliminate
fleM of gas to the crankcase or lubricant to the under-side of the piston.
It is essential that the pistons and rods be restrained to provide
essentially true axial motion so that, at the lower end of the pistons,
leakage ean be controlled by using a elose-clearancegas seal between the
pistons and cylinders, instead of the conventional piston ring and oil
lubricated bearing surface.
The desired linear motion is achieved by attaching the connecting
rod to a linear bearing device (the crossheads) which brackets the outer
diameter of the swash-plate and includes two socketed hemisphericaHbearings
for load transmission to the fore and aft faces of the swash-plate.
Wi th the
roll-sock seal on the piston rod (below the piston), the linear bearings
can be lubricated effectively with no lubricant contaminating the upper
gas circuit.
A 4-cylinder swash-plate of the type shown in Figure 6.9, designed
for a 160 HP automotive engine using hydrogen or helium as the working
fluid would have a length, of the basic engine, of 36 inches, with a
diameter of approximately 18 inches.
(With accessories, the engine height
and width are 24 x 22 inches.)
In comparison, General Motors designed
a 4-cylinder, 200 HP in-line stirling engine with the dimensions of the
cylinder block being approximately 47.2 inches long, 47 inches high, and
24.5 inches wide.
An advantage of the swash-plate configuration is that one combustor
can be designed to provide uniform distribution, simple preheater design,
and ease of gas recirculation.
Contrasted with this is the problem of
transferring heat to a multi-cylinder in-line engine without using
individual combustors for each cylinder.
Clearly, the system integration
problem is more complex in the case of the in-line engine.
6.4
GENERAL PERFORMANCE
The available data on the Stirling engine must be abstracted from
numerous documents in which the data are primarily used to tllustrate
characteristics rather than to present anyone specific complete design.
There appears to be no single source of data which can be adapted to an
engine for direct application to an automotive powerplant.
While data
presented here are qualitative in the sense of illustrating principles and
characteristics, extrapolated data will be presented, wherever possible,
to document the performance of a specific automotive engine.
215
-------
The Philips Stirling engine achieves a level of performance
'"hich avproximates that of a Diesel engine with respect to (1) the
maximum cycl~ efficiency, and (2) the generally flat efficiency character-
is tic relative to load and speed.
The Stirling engine does not use valves
and controls the power output by manipulation of the pressure level of
the working fluid.
To increase working fluid pressure for higher power,
or to decrease working fluid pressure for lower power, working fluid is
injected from or discharged into an external reservoir.
A similar
control system must be used for any closed cycle (including both reciprocating
and turbo-machinery).
The optimum compression ratio can be shown to be strongly influ-
8nced by the practical limits of maximum anu minimum cycle temperatures,
and the practical limit is mostly determined by the capacity of the heater
tubes and other hot sections of the engine to withstand the combustor
temp<:rature and the pressure of the working fluid inside the heater
o
tube~. Philips and General Motors Research have chosen 1300 F (approx.)
as the practical operating limit with heater tubes fabricated from heat
and creep resistant stainless steels.
o
up to, say, 1500 F con be used for the present types of engines, but that
level is generally considered to be beyond present metallurgical limits
Perhaps heater metal temperatures
and will require considerable research and testing before this level can
be considered.
. 0
With a limiting gas temperature of 1300 F, the compression ratio
can be established and with the maximum and minimum gas temperatures the
ideal cycle can be calculated.
For the temperatures currently used, the
compression ratio is found to be about 2:1.' At significantly higher com-
pression ratios, the effectiveness of the regenerator and the cycle
efficiency both decrease.
For the greater part, the following data in this section is repre-
sentative of experimental laboratory Stirling engines using hydrogen as the
working fluid, which is typical of Philips and General Motors engine
designs.
Under the optimum cycle conditions, the efficiency is about 37 to
38%, rather than the value of 40% which is often quoted in many general
discussions in the technical literature.
216
-------
As shown in Figure 6.11, the design point performance of a laboratory engine
operating with hydrogen at a coolant temperature of 60°F and a heater temper-
ature of 700°C (l/.92oF) is 31 BHP with 37% efficiency. The power output
can be reduced LI !:iO%, tv 15.5 BliP by reducing the heater temperature
° 'h
to 785 P, Wl.t a corresponding engine efficiency of 24%.
The Stir1.i.ng engine must be operated at relatively high heater
temperatures and relatively low cooler temperatures if high efficiency is to
be achieved. Whereas the maximum cycle temperature of an Otto cycle engine
will reach values somewhere of the order of 3500-4000oF, the Stirling
o
engine operates at a maximum cycle temperature of about 1300 F. It can do
this only because it is a n~generatj ve engine which operates at low pressure
ratio.
If the cycle pressure is used as the primary control method, a broad
power range can be achieved with only a modest effect upon cycle efficiency.
This is shown in Figure 6.12 where efficiency is plotted as a function of
engine speed for several levels of gas pressure.
As the operating gas
pressure is decreased from a value of a 1991 psi down to 1000 psi at 1500
RPM the peak efficiency drops from approximately 38% down to 34%.
For speeds
higher and lower than 1500 RPM the efficiency drops further.
The relation
of speed and maximum cycle pressure to the power output of a laboratory
type engine is presented in Figure 6.13.
These data show that, for this
engine operating with hydrogen as the working fluid, about 40 BHP was
attained at 1991 psi maximum gas pressure.
The 40 BHP laboratory engine is specifically designed with heaters
and coolers for the approximate speeds shown. Higher power levels at
higher speeds have been projected for Stirling engines operating
;:' 10,000 RPM, but the efficiency suffers, dropping down to about 27%.
Wi th-
in the limitations of the overall thermodynamics, many changes are possible
depending on the heater and cooler tubes, regenerators, working fluid,
type of heater, etc.
In all of these data, the performance is recorded at 600F coolant
temperature.
While such performance is representative of engines which
have abundance of cooling water, as would be the case £or a ship, a"much
higher temperature would be required for an engine which is used in an
217
-------
~
~
30
25
20
15
10
5
o
o
FIGURE 6.11
200
600
400
o
Heater Temperature - C
Brake HP and Efficiency plotted as Functions of the
Heater Temperature" for Stirling laboratory engine.
2 (32)
(N = 1500 rpm, Max. Press. = 140 kg/cm )
/.18
40
11
30, - %
o
o
800
1111
-------
40
I1dX. Press
Pei.
711
.----
30
~
,
>.
u
c: 20
Q)
-.-4
U /
-.-I
~
~
Q/
4J
C
Q)
U
~
Q/
c..
10
o
o
500
1000
1500
2000
2500
ENGINE SPEED,
- RPM
FIGURE 6.12
Efficiency of a small test engine as a function of engine
. . (32)
speed at various values of max~mum pressure.
219
1113
-------
.'
40
30
P<
:I:
~
20
10
()
o
FIGURY 6.13
Coolant Tperature =
OOF
500
1500
2000
2500
1000
Engine Speed, N - l~M
Brake HP of a Stirlin~ Test En~ine Plotted as a
Function of Engine Speed for Various Values of
Maximum Pressure with Hy.-'lrog<:'n as the Working
~, uid. (45)
220
1287
-------
automobile because the cooling water must be cooled by a radiator.
In
order to obtain an effective temperature difference in a radiator, a
co()lant temperature of the order of l300F is required. The effects of
c00lant temperature on both efficiency and power are significant. At
?Ooc (SHoE-') an efficiency on the order of 38% is indicated in Figure 6.14.
When the temperature is increased to 54.50C (1300F) the efficiency drops
to about 34%.
On this basis, the overall engine efficiency drops about
4 points from the laboratory engine performance level to that practical in
an automobile engine.
This corresponds to a loss in power of 10%.
. Typical heat balance data are plotted in Figures 6.15 and 6.16 for
a representative laboratory engine.
Figure 6.15 shows the effect of BMEP
on power output.
From 30% to 100% BMEP, the relative power i.ncreases from
30:); to '3(j~" rr:!specti vely.
Cooling water heat rejection, heater exhaust gas
and radiation losses, and friction losses also increase with power,
although the relative percentage of the first two remains about the same
while th~ latter decrr.ases.
The power output increases with BMEP up to,
and presumably, heyond 26 kg/sq. em., or 370 psi BMEP.
The overall performance of a representative laboratory Stirling engine is shown
in Figure 6.16.
The map presents BMEP as a function of engine speed for
a range of BSFC and BHP values.
It is to be noted that the minimum BSFC '
(.369 Ib/Bhp-hr) occurs in the region of 1100 to 1500 rpm for this engine.
As this chart is more or less representative of the operation of the Stirling
engine, performance data for a 160 BHP Stirling engine will be extrapolated
from these data.
6.4.1
Working Fluid
The working fluid which is used for the Stirling engine has a
significant effect on the power output and efficiency.
In general, the
low molecular weight gases are far superior, especially when it is desir-
able to operate the en~ine at high speeds.
While heat exchanger character-
istics and the heat transfer properties are very important in themselves,
the fluid dynamics and pressure drop of the gas flowing through the fluid
passages predominate for a high performance engine.
While any gas can be used, low fluid friction and high heat transfer
properties are so critical that most common working fluids are eliminated.
The original Stirling and all designs up to the period that the modern
stirline engine was developed by Philips, used air as the working fluid.
In the 1950's, a thorough fluid-dynamic and thermodynamic analysis resulted
in the realization that hydrogen had the best combination of properties.
Helium, having the next lowest molecular weight to hydrogen, was less
?21
-------
Temperature - of
40
50
120
140
80
100
30
r..
:r::
m
I I :~ I ,
~
~
~
r-----
_ .....
1\.
-
25
20
15
10
5
00
50
60
70
30
40
10
20
Inlet Temp. of Cooling Water
°c
FIGURE 6..14
BroKe Horsepower. and Efficiency Plotted as Functions of
Inlet Temperature of Cooling Water.
(N '" 1500 HPH, h;ax. Press. = 1991 psi)
Ref. 28.
222
40
30
1'\ - %
20
10
1288
-------
100
.j..J
iC
OJ
Frict lion
~ -
..
Exhau st G ses & Rad.iat ~ ,~
~ ----
.-
. "'-
,
__0. --_._. ---- _ .- -
C(.oli [19 Wi ter
~._'.. -
,
-
Brake Loac I I
I :
i
I
i i
!
tJ. B 12 ?O 24 2
80
.-
.-i
n:
~J
g
on
I.~
()
.j..)
r::
;J
40
')
~,
(1.1
C";
20
o
16
8
:2
1"\!\1F.J> .. Kq/cm
J:'IGUlir.; 6.15
Single Cylinder Test Engine. Heat Balance as a
Function of Mean Effective Pressure
(N == 1500 RPI'~). (45)
'/2"1
1289
-------
30
60°F Cool nt in
22 in3Dis lacement
00 .
"
,
)00
25
20
,
15
N
S
U
........ !XI
~ is
't1
I I
't1
!II
1-'.
1
5
,
'- ~
'''!'JII88-- ,
0.48::>
'--
0.5~
BSFC -. Ib/
o
500
1000
1500
2000
2500
o
3000
Engine Speed, N - I~M
FIGURE 6.16
Lines of Constant Specific Fuel Oonsumption and
Constant Brake Horsepower as a Function of Mean
Effective Pressure and Enqine Speed.
(Fuel
(45)
Consumption based on a Lower Heat Value of 10,000 Kcal/Kg) .
.
224
1290
-------
I
favorable but was superior to other gases.
Figure 6.17 compares the performance of a small, single cylinger,
laboratory engine employing air, helium and hydrogen as the working fluids.
As shown, air is almost as favorable as the other two in terms of maximum
efficiency, but hhe maximum efficiency with air occurs at so Iowa speed
that an engine so designed results in unacceptable horsepower-to-weight
ratios.
Helium is more effective, but hydrogen is definitely the
superior with respect to efficiency afid horsepower per unit volume dis-
placement.
At the elevated temperatures experienced in a Stirling cycle, hydrogen
diffuses through high temperature metals,and, ~hile the rate is not
intolerable for most applications, it cannot be used wherever a leakage
cannot be resupplied.
This is not a problem with helmmm or nitrogen.
To
date, no practical diffusion inhibitor has been used, but it is not
inconcei vable that one can be found.
The engine with hydrogen as the
working fluid has been chosen by Philips as the preferred working fluid
for an automobile powerplant.
Considering only the performance factors, it is difficult to come
to any other conclusion.
However, the safety and leakage loss factors
which are present with a hydrogen working fluid cannot be ignored.
Those
data available do not indicate what increase in fluid loss rate one should
expect as the mean pressure is doubled.
Nor is there any indication of the
relative degree of complexity associated with safety devices which may be
required to reduce the hazard in general public usage of hydrogen.
Indeed,
the hazard, if it exists, does not seem to have been very thoroughly
investigated.
Helium has been selected as the tentative working fluid for the
automotive Stirling engine analysis which follows.
This is done with full
recognition of the performance-weight penalty which results and is justiiiable
only in terms of least potential hazard.
6.4.2
Direct:and Indirect Heating
The extrapolated performance characteristics for a 4-cylinder auto-
motive Stirling engine with either direct or indirect heating are indicated
in Figure 6.18.
These data, generated by Philips, indicate reductions in
efficiency levels as compared with the Philips data of Figure 6.17, which are
based upon laboratory tests of small single cylinder engines.
It should
be noted that there are, in addition to configuration differences,
significant differences in maximum engine speed, pressure levels and
temperature levels.
r:;,
-------
*
~
~
>.
u
c::
I1J
"M
U
OM
4-1
4-1
c..1
54
50
.-.- ------
4(-;
I
I
I
i
" I'
i
I
1500
I
42
-------
38
i
I
J.-
I
I
I
I
I
I
I
j --...
I
!
_0_._.._- .
34
.. -"..-."- ....
30
o
Q.4
FI GURE 6..17
50 Bhp Cylinder
-----_.
-- _._--
-----
I
I
._._._---_._~_._-_.-
I
Helium
Air
-. ...... ...-
I
1- . - .-----.
I
I
I
I
---.." -
- -----_. ---
3000
750 RP:Ii
0.8
1 .,
-..<:
1.6
2.0
2.4
. -. .y-) I' .3
SpecJ.tJ.c Po,';er, 111.. In
(S\\1ept Volume)
Relation of Sin~le Cylinder Laboratory Stirling
. (28)
Engine Performance.
22G
1291
-------
!
>t 40
0
~
dJ
.01
0
.01
'.....
\i-4
~ '*
'cJ .10
OJ
+I
ttI
0
.01 H~
'tj
~
H 20
200
Indir.ct Heati 9
.... . Di.r.ec '. Heating
",
1::'0
:-.::
>~,
~
C:
p, 100 .. ------ ...--
OJ
tII
II
0
:r:: .......
He
'tj
OJ
+I 50
It!
U
.01
'tj
~
H
0
0 2000 4000. 6000 8000 10,000
Engine Speed - RPM
'FIGURE 6.18
Preii:ct~ Four Cylin~&r:Autdmotive.
Stirling Engine Performance with Helium or
Hydrooen Working Fluid.
(Ref. 28)
227
1292
-------
It is clear that the hea~)pipe displays an advantageous heat trans-
fer rate as compared to the design where the external tube surfaces are
heated with direct combustion gases.
A heat pipe employing liquid metal,
NaK or sodium as a heat transfer medium, not only increases the overall
heat transfer rate significantly but also results in less dead volume
inside the tubes and, therefore, improves the performance of the cycle.
The improvement in efficiency and allowable operating speed with indirect
heating is significant.
The problem of pumping the liquid metal has been conveniently solved
by Philips by using a heat pipe which transfers the heat through a com-
bination of convection and capillary pumping of the liquid metal rather
than us~ng a mechanical pump.
The liquid metal heat pipe has several
times the conductivity of an equivalent diameter of pure copper and should
be far superior to any other heat transfer device.
However, in terms of
performance, it seems best applied where the need for high specific power
outweighs economy.
It should also be noted that in addition to the
problems of stress reliability associated with the elevated pressures
(3000 psi mean), there will very likely be some degradation of seal life,
particularly if the higher pressures are combined with very high speeds.
6.5
AUTOMOTIVE ENGINE PERFORMANCE
The power and efficiency characteristics of an engine which is
suitable for use in an automobile is shown in Figure 6.19.
These data
were extrapolated from available sources and are modified to meet the
performance of the vehicle.
Design parameters are itemized in Table 6-I.
This engine operates at high peak pressures (3500 psi) over a pressure
range of 2:1 with a mean cycle pressure of approximately 3000 psia.
Helium is used as the working fluid due to the potential element of hazard
with hydrogen, but the expected performance levels associated with
hydrogen are noted where appropriate.
This results in an increase in
the peak pressure level of about 50% in order to attain the equivalent
power output and to compensate for the low~waste heat exchanger temper-
ature (60oF) used in the Philips data.
From the extrapolated performance data of Figure 6.19 the road
load fuel economy characteristics were approximated for the 160 HP
Stirling engine in a 4300 Ib automobile.
The road load power require-
ment for this 4300 Ib vehicle/engine combination is presented in
addition to the fuel mileage in Figure 6.20.
On the basis of performance
the modern Stirling engine is definitely competitive.
2?B
-------
TABLE 6-1
DESIGN PARAMETERS FOR THE STIRLING ENGINE
FOR THE EPA 6-PASSENGER AUTOMOBILE
Power
160 Bhp
Engine Speed (rated)
Maximum Cycl e Temperature
Working Fl uid
Minimum Cycle Temperature (Coolant water)
3090 RPM
13000 F
Helium
1300 F
Mean Cycle Pressure
3000 psi
I -'
Compression Ratio
2:1
Number of Cylinders
4
Bore
2.85 inches
stroke
1.825 inches
Displacement
46.57 in3
Cylinder Arrangement
Swash plate/Barre1 Config.
',--/
229
-------
. 1.00
0.8
H
::r:
I
n.
;:: (). 5
:Y.i
'...
,(.I
~
()
!'I
If, '1.4
CQ
200
150
..-......---
_.-._--
"
. .
Stirling SP-4 Helilun
46.57 CI , 13000p He ter,
130 'F' Coo er
--..-..-.-..---..
--
..._._-_._-~ .
---
-----
-_.--~ - .
-r
I
1-..
-----
100 -- -._--
~
if;
150 -...
a
O' .
1650
1000
hHY. r ,ss.
h~,i
1000
2000
3000
N - RPH
FIGURE 16..19 Estimated Perf0rmance for an Automotive Stirling Engine.
,230
-'- '-'-
1293
-------
160
140
120
100
A.
::I:
~
~
oS 80
'tj
ItS
~
60 30
40 20
t9
~
20 10
o
o
20
40
60
80
100
o
120
v - rnph
FIGURE ~. 2~
AutcmbtivEi.:YSt'.i1!'iin(~f;Edgine '~atl~Load Fuel Ni'1eage,
at 1,:ro~F -'G:dbUng:Wciter' rt(I:e~'~eratu1!'e. (47 CID,
160 BHP @"'700lr: RPM): He' W8rk~thg -F'I\iid, Direct H~ating.)
.231
-------
Th~. reader may, at first, be surprised at the low efficiencies,
as compared to previously presented Stirling performance, indicated
in Figures 6.19 and 6.20.
It must be kept in mind at all times that even
the most modern Stirling engines which.were actually tested were far
F.rom acceptable for those conditions peculiar to automotive application.
Changes dictated by safety and operational practicality (working fluids,
engine speeds, and minimum cycle temperatures) substantially reduce the
efficiency of the automotive Stirling engine in comparison to laboratory
and other special applications.
6.5.1
stirling Engine Emission Characteristics
The Stirling engine is an external combustion engine and has the
general characteristics of all external combustors.
In this sense, it
performs in a manner similar to the Rankine cycle, the closed cycle
Brayton cycle, and the External Combustion Piston Engine powerplants.
Differences in the characteristics of one versus another can be traced
to the specific characteristics of a particular design and to the state-
of-the-art of combustor design at the present time.
It is desireable to control smoke, odor, and emissions to a minimum
level over the operating spectrum.
The data presented are typical of
the General Motors Research Models 1036, 1036R, and GPU engines which are
rated at approximately 10 HP, the Philips Models Sl050 and S1210 (which
arc rated at 80 to 90 HP per cylinder) and the Model 3015 (which is
rated at 30 HP/cylinder) .
Detailed discussion relative to these engines
are presented in Reference 30.
Both the Philips and General Motors combustor designs use conven-
tional technology in the approach to combustor design.
A single fuel
nozzle and igniter is located centrally and kerosene, JP, or No.2 Diesel
fuel is sprayed into the combustor.
A preheater using heater exhaust
air is incorporated into the combustor design to provide primary air,
dome cooling air, and body cooling and secondary combustor air.
The
dome cooling air uses a series of tangential cooling slots which provide
a tangential swirl to improve combustion.
The maximum-to-average temper-
ature ratio was found to be approximately 1.04, indicating good distri-
bution of fuel and air.
In the stirling combustion system, burning takes place continuously
in a lean overall fuel ratio (30% or more excess air).
High peak tempera-
tures, excess oxygen, and adequate resident time provide for essentially
complete combustion of carbon compounds from carbon monoxide.
These
high pe~k temperatures also promote the. formation of nitric oxide.
A
232
-------
flame ionization detection (FID) meter was employed to measure total
exhaust hydrocarbons and continuous sampling techniques were used for
measuring the exhaust gas CO, NO, and N02 concentrations.
The exhaust emissions results are presented in the following three
forms:
1)
As functions of air-fuel ratio with a constant
2)
burner inlet air temperature.
As functions of burner inlet temperature at
3)
constant fuel-air ratio, and
As functions of engine load at constant air-fuel ratio.
loads.
The air-fuel ratio dependence was determined at both half and full
The concentrations of NO, CO, and hydrocarbons were converted from
the volume (ppm) to a mass basis (lbs/1000 Ibs/fuel) by using 28.8 units
as the average molecular weight of the exhaust gas.
All measured hydro-
carbon concentrations were considered to be ppm of n-hexane (C6H14).
three emissions of major concern, at half load, are expressed as a
The
function of air-fuel ratio in Figure 6.21 (Reference 30).
Me as uremen ts
both upstream and downstream of the preheater (hot side) are presented
for CO and NO.
The NO and CO were affected favorably by the introduction
of the preheater, to the extent of approximately 100 ppm at 20:1 air-fuel
ratio, however, this effect decreased to lesser values as the air-fuel
ratio increased.
Although no downstream hydrocarbon measurements were
taken, it is highly probable that the hydrocarbon measurements would follow
the same trend as CO concentrations--namely, a decrease with passage
through the preheater due to the added residence time.
The same data
are presented in terms of pounds of each component per 1000 lbs of fuel
in Figure 6.22.
As the air-fuel ratio increases, the temperature of the combustor
decreases and results in the corresponding decrease in the heat transferred
to the heater tubes of the engine.
This is clearly shown as a decrease
in efficiency as the air-fuel ratio increases from 20 to 40:1.
Whereas
the efficiency was 23% at 20:1 air-fuel ratio, it drops to 20% at 40:1
air-fuel ratio.
However, since the combustor should be operating at
approximately 25:1 air-fuel ratio most of the time, it appears that the
efficiency at half-load is about 23%.
Similar data are shown for full-load
233
-------
600 .0 't1
~
. .
~.
~
r) .5 p,
z 500 11
o
rcJ ()
c: ~
(\1 t1
O'
o 0
U ::1
400 .0 :'1
.......
.J..
p..
p,
300 .5
I
I
900
800
700
?-oo
100
Preheate
(hot 81 e)
bons
o
18
.5
.0
22
26 30
Air/Fuel Ratio
. I
0.5
38
42
34
0.8
I
0.7
I
0.6
0.4
Equivalence Ratio = 14.7/(A/F)
Effect of Mixture on Exhaust Concentrations
FIGURE 6.21
at Half Load. (30)
231
1295
-------
"
E'IGURE 6,. ~2
FIGUJ?,E 6.,23
Equivalence Ratio
0.7 0.6 0.5 0.4
24
;..,
Efficiency 0 ..
t:
20 Q)
,.; 20 'M
Q) Before preh ter u
::3 OM
~ ~
- - After pre 'ter ~
..Q 16 ~
,.;
0
0 ,.;
0 12 0.24 Q)
,.; ~
"
~ .
~
, 8 0.16
g 0
0
0
't1 ,.;
t: 4" 0.08 "
rd \0
0 U
u G .Q
0 ,.;
18 22 26 30 34 38 42
A/F
Effect of Mixture Ratio on Half Load Emissions,
at approximately l2000F burner inlet temperature. (30)
Before Preheater
After Prehea.ter
---
1200
1000
800 ~,Om
, t:
o
() ..Q
::-:~ ~
'd 600 3.0 u
t: 0
tU ~
't1
o ;..,
u ~n 2.0 ES
2:; ~
p.,
Pi ~.o p.,
o 0
18 20 22 24' 26'
A/I:'
Effect of Hixture Ratio on Full Load ExhaJ,lst
Concentrations at approximately l3550F burner
. (30)
~nlet temperature.
1~96
. ,235
-------
over a lesser air-fuel ratio.
While there is some scatter in the data
taken at 20 and 25:1 air-fuel ratio, there tends to be a somewhat higher
level of NO emissions at the lower and upper limits of air-fuel ratio,
and correspondingly higher numbers of hydrocarbons as would be expected.
These data are shown in Figure 6.23 (Reference 30).
The effect of combustor air inlet temperature on the th~ee emissions
of interest are shown in Figure 6.24 (Reference 30).
As would be expected,
the hydrocarbon and NO quantities decrease as inlet temperature increases
and the NO concentration increases quite drastically at the higher
temperatures.
Values of approximately 450-500 ppm are measured at 12~OoF.
similar data were obtained on a 30 HP Philips stirling engine.
The results are shown in iigure 6.25.
With no exhaust, i.e., circulation,
the Philips burner achieves comparatively lower emission levels than the
GM Research combustor, when both are operating at air-fuel ratios of 20:1.
While the GM Research data indicated approximately 900 ppm, the Philips
combustor shows about 300 ppm without circulation.
The effect of adding an inert diluent to the combustion air by
recirculating exhaust gases was determined for both GM Research and
. .
Philips combustors.
In both cases, part of the exhaust from the pre-
heater was returned to the suction side of the combustion air blower.
Because the recirculation line was not insulated, the exhaust gas inside
it was cooled several hundred degrees and resulted in descreased efficiency
with increasing circulation.
This is shown for the louvered combustion
and for the slit type combustor in Figure 6.25.
There is some difference between the NO and NO characteristics of
x
the two combustors which can be related to a basic characteristic of re-
circulation combustors.
The louvered combustor shows a relatively high
CO level, varying from 800 ppm at zero recirculation to a little below
400 ppm at 33% recirculation, increasing to about 800 ppm at maximum
recirculation.
The dip in CO emissions at 33% recirculation is probably
in error and the curve very likely is nearly flat.
The NO decreases
x
from about 300 ppm at zero recirculation to about 40 ppm at close to 100%
recirculation.
The slotted combustor shows much better emission character-
istics.
The CO increases from 80 ppm at zero recirculation to about 170 ppm
at 100% recirculation, representing only 20% of that encountered with the
louvered combustor at the highest point.
Similarly, the slotted combustor
shows a decreasing NO emission level with increasing recirculation,
x
with values dropping from about 110 ppm at zero recirculation to less than
40 ppm at maximum recirculation.
At low levels of recirculation, the
236
-------
--
F'I'.,:UP,r.; 6.24
-
lJeIDre .tTenea'ter
After Prehaater
6
4
~~
\D
U
......
---
25
20
.J:
.Ai
Q./
1200
:';ffect (;1 r:,.:h'~stion l\ir Tempera7 Uj~e on Exhaust
Conccntr.~ U,'-':lS '3.1: 1I./l" " 2S. (30)
~~
,.x f..!
;:..? Q.)
"-< c:
''J ~
2!=Q
C ~
S~
:r .£
~
p.,
FIGURE 6.25
15
",'1
~
Gl
",1
u
....,
~I
41
r,d
10
5
o
600
400
u
~
'-::J
~:
r!j
o
u
200
---
~..
~
p.
'j
o
400
800
R'!rnel
T O.~
Ai=: Inh',t amp.,.' r
}'"
,,,-
>t
()
c:
-------
slotted combustor records about half the values en~ountered with the louvered
design .
At maximum recirculation, both combustors have about the same
level of NO .
x
able effect on NO .
x
performance can be achieved by further research in combustor configuration.
Such improvements will be required to achieve the levels projected for the
In overall terms, the recirculation has a generally favor-
Very likely further improvements in emission
future gasoline engine.
1~e cooling effect of exhaust gas dilution results in about a 1%
loss in engine efficiency at 100% dilution.
This effect can be traced to
the lower combustor temperatures and reduced transfer of heat to the
heater tubes.
The highly efficient combustion in the Stirling burner has been
characterized by high temperatures which accelerate formation of nitrogen
oxides.
Because of this, the available data (circa 1968) on Stirling
emissions indicated relatively high oxides of nitrogen levels.
Three
methods of NO control are outlined below. They all achieve lower NO
x x
emissions by lowering the peak combustion temperatures at varying penalties
to overall thermal efficiency.
These methods are:
1.
Reducing the combustion air inlet temperature.
A direct method of lowering the flame temperature is
to reduce the air preheat.
This is accompanied hy a
decrease in the theoretical equilibrium nitric oxide
concentration for adiabatic combustion produces (29).
Although not in equilibrium, the concentration of NO
in the exhaust from a Stirling burner exhibits a
similar percentage decrease as the flame temperature
is reduced.
A substantial loss in efficiency occurs with
the preheater removed from the circuit as well as
an increase in hydrocarbon and carbon mono~ide emissions.
2.
Leaning the mixture ratio at full preheat.
The decrease in NO concentration with progressively
leaner mixtures results primarily from the NO formation
rate temperature dependence.
Lean air-fuel ratios
produce lower peak combustion temperatures and the
arguments for lower NO concentrations from method 1 follow
directly.
A secondary effect arises from the increased flow
rates through the burner at lean mixture ratios.
Because
the NO formation process is not in equilibrium, these high
238
-------
mass flows shorten the resider:~;:: time in which NO can be
formed.
These two effects are somewhat offset by the
greater oxygen concentration in the combustion ~roducts
with leaner mixtures, thereby tending to increase
the :"0 concentration.
Although the exhaust temperature is essentially unchanged,
thermal efficiency is adversely affected (Figure 6.24)
because the higher flow rates result in reduced preheater
performance and higher total exhaust enthalpy.
3.
Recirculating part of the cooled exhaust gases.
This method has a three-fold reducing effect on NO
formation.
First, the dilution effect lowers the flame
temperature and, consequently, the NO formation rate.
Secondly, the increased flow rate through the burner
decreases the residence time in a manner similar
to operating with a lean overall air-fuel ratio.
Finally,
the exhaust diluent, while not inert, contains a lower
percentage of oxygen than air.
The combustion products,
then, contain a smaller concentration of 02' further
reducing the formation rate of NO. Because of the
added heat losses in the recirculation lines, engine
efficiency decreases with increasing recirculation.
6.5.2
Noise Characteristics
In general, the automobile engine is sufficiently quiet for normal
operating requirements.
At idle, the sound level is relatively low.
Even
at high speed operating conditions, the engine noise tends to be over-
shadowed compared to tire noise, fan and exhaust noise.
engine is relatively quiet in comparison.
The Stirling
First, it is a closed cycle engine, hence, a very little internal
aerodynamic noise excapes.
Practically all aerodynamic noise is generated
by the fan for the external combustor.
Second, the elimination of abrupt
changes in cylinder pressure permits the design of an engine with a minimum
of mechanical noise.
Several Stirling engines have been made for terrestrial and space
applications and the experience with these engines is presented very well
239
-------
in the referenced reports.
One 4-cylLlder 400 BHP engine, which was tested
by the Naval Engineering Stat,ion at Annapolis, showed a significant reduc-
tion in structure and airborne noise as compared to the Diesel engine.
For
roughly comparable power levels and engi.ne speeds, the marine application
-3 2
showed a structure borne noise level ('!: 96 dbA (RE: 1 x 10 em/see) as
compared
with about 135 dbA for the f"luivalent Diesel (Figure ~.26) .
For
submarines, for which the tests were co~Jucted, and for other vehicles
which are sensitive to a structure-borne noise, this is a significant
difference and can be traced to a large degree to the inherently balanced
"Rhombic" drive which was used in this engine.
The swash-plate drive
should show similar characteristics.
Figure 6.27 shows that the Diesel
had a peak sound-pressure-level of about 112 db whereas the Stirling engine
had a peak of only 95 db, for a decrease of 17 db.
Sound measurements were made on a selfcontained ground power unit
(GPU-3) rated at three kilowatts continuous output (Figure 6.2S).
'l'he
engine design in this unit differs slightly from that in the dynamometer
i.nstallation previously described.
The GPU-3 has an integral preheater
and drives all its own, accessories, including an oil pump, nozzle atomi-
zinC} ,.air pump, radiator fan and combustion air blower.
Sound data were taken at the united states Army Mobility and
Equipment Research and Development Center (MERDC) in Stafford, Virginia.
The GPU-3 was operated at rated speed and load (3000 rpm, 3 KW) with
no acoustic baffling of any kind placed around it.
General Radio
equipment was used to make sound intensity measurements at varying
distances from the GPU-3 package.
Because noise level is a relative
matter, the Stirling engine test results are compared with results
from an internal combustion 3 KW Military Standard unit tested in ~1e
same manner.
Noise measurements of the two ground power units were made
on different days at the same location.
Since ambient sound levels
change from hour to hour, the ambients for both test days were averaged
for this comparison.
The results are shown in Figure 6.28, a plot of sound intensity
in decibels versus frequency in hertz at a radial distance of 100 feet.
The bars constitute an average sound level measured over a given octave
pass band.
(Each band or octave on the abscissa consists of a range of
240
-------
140
---'
130
120
N ]]0
U
lLJ
V)
........
V)
i ~
0:: ]00
" ~
: I ~
U
r<"'\ ~ "" ,". '" ........
I I ,I'... ~Y' ", " ...
I 0 , '" \ I
~ J -"'.1
I x 90
,
I
w
0::
,"
CD
0
"-..,/' < 80
, , ,
I "
't
70
60
50
10
50
100
500 1000
FREQUENCY, Hz
5000
j
i
~
~
~
~
"
FIGURE 6.2'6
Results of Naval Engineering Station Tests on a
4-cy/inder Stir! ing engine. 300 HP at 1200 rp~.
(Reference 311
~
,I
241
10000
1298
-------
120
a:::
~
~ 110
a:::
u
-"-
N
o
, g ]00
o a
iZ
:J ..
o w
'V') a:::
,~ gs 90
"~
en Z
. "'-
w
> ....J
~ W
~ >
g ~ 80
,.,
I ,
,
,
I
,
,
,
...
'"
ST 1 RLI ~~G
'...
.... ... - ~
j
w
a:::
::J
(/)
(/)
'w
'0::;
iO-
70
I
I
I
\
\
\
60 '
10
50
JOO
500 JOOO
FREQUENCY, Hz"
5.000
10000
FIGURE '~f?7
Airborne noise spectrum of Naval Engineering Station tests
on a 4-cyl indcr Stir I ing engine. 300 HP at 1200 rprn.
(Reference 31')
242
1299
-------
"--.
H
~
o
~i
C)
orl
~
N 40
a
o
a
a
'--'
D:I
Q
t:.:
orl 30
rl
Q)
:>
~
'd
c
::s
o
m
F'IGURE -6,..28
~ Military Standard 3kw Uni t
~ Stirling GPU-3(kw)
60
50
20
AVE; AGE AMBlE T
1 0 7~-
75~ J o. 300_1200~' 120~-~~8g~4800-
150 600 2400 9600
Octave Pass Band in Hertz
OV R
A'LL
Noise level of Stirling and Military Standard
(internal combustion) ground power units at
. (30)
100 feet d~stance.
243
1 ;;;:~:
-------
frequencies from a given value to twice that value.
All other frequencies
are filtered out.)
The column on the extreme right constitutes an "all
pass" reading in which the entire frequency range is monitored.
This
figure shows the Stirling ground power unit to be 21 db quieter than the
Military Standard unit when observed at 100 feet. It is appar~nt that
the Stirling power plant enjoys a significant advantage over more con~
ventional engines in the area of airborne sound.
6.5.3
Hazards
From a thermodynamic standpoint, the best working fluid for the
Stirling engine is gaseous hydrogen.
The next best fluid is gaseous helium,
but the use of helium results in an increase in powerplant size and
weight on the order of 30 to 40%.
Whereas helium is a safe and, inert
working fluid, hydrogen is a highly flammable gas which requires pre-
cautionary measures.
Characteristics that make hydrogen hazardous include the follmwing:
1.
The fuel-air flammability limits of hydrogen are
far broader than those of the petroleum fuels.
2.
The energy required to ignite hydrogen-air mixtures
is significantly less than gasoline-air mixtures.
The energy of static electricity sparks is qmite
sufficient, to ignite hydrogen-air mixtures.
3.
The flame speed in hydrogen-air mixtures is far
higher than in gasoline-air mixtures.
It could
potentially result in much greater explosive forces
upon the initiation of hydrogen air fires.
An additional factor to be considered is that hydrogen" at the
peak cycle temperature (1300oF) in the Stirling engine, is above the
auto-ignition temperature of hydrogen-air mixtures.
If a rupture should
occur in the high temperature portion of the system, it is possible that
the hydrogen would immediately ignite.
It can be argued that Philips and General Motors have had experience
with hydrogen in Stirling engines, generally without incident.
(There is
one exception which occurred at the u.S. Army Research and Development
Laboratories at Ft. Belvoir, Va. where a 10 HP General Motors Research
engine had a hydrogen~air explosion in the crankcase due to static
electrici ty. )
Nevertheless, it is questionable whether engines operated
under controlled laboratory conditions can be compared to putting 100,000,000
vehicles containing ,hydrogen in the hands of the general public.
244
-------
One method of reducing the hazard associated with using hydrogen
in the Stirling engine is to reduce the quantity of hydrogen by reducing
the size of the plumbing, manifolds, pumps, reserve bottle, etc., to a
minimum.
Calculations indicate that for the General Motors 4L23 150 HP
Stirling powerplant, the quantity of hydrogen required is as follows:
Hydrogen in englne only
Weight,
.....J!>..~.;-
0.030
Free Volume
@ 700F, cu. ft.
5.77
(at normal full load mean press. of 1500 psi)
Plumbing, manifolds, pumps, etc.
Reserve bottle
0.012
0.008
2.31
1.54
Total
0.050
9.62
If 0.050 Ib of hydrogen were completely burned, the total heat
release would be about 2500 BTU.
Although this amount of heat could
cause a rather violent explosion if mixed with a stoichiometric quantity
of air, no lasting fire could result.
Also, the probability of having a
stoichiometric mixture and an ignition source occur simultaneously appear
somewhat remote.
Therefore, in the case of the Stirling engine, it may
be possible to design a relatively safe powerplant using a hazardous working
flu id .
If the Stirling engine were to be chosen as the Optimum power-
plant for meeting future emission goals, it would probably be desir-
able to limit production until fire and explosion hazards can be determined
based on actual operating experience of Stirling engine powered motor
vehicles in the hands of the public.
Another safety precaution that would be necessary would be to vent
the ceilings of any garage structures housing vehicles containing hydrogen,
so that in caSe of a leak, the hydrogen would rise, pass through the
ceiling, and dissipate into the atmosphere.
6.6
AUTOMOBILE ENGINE DESIGN
Preliminary design studies of a 160 HP Stirling engine for use in
an automohile are shown in Figures 6.29, 6.30, 6.31, and 6.32.
The engine
has four cylinders arranged around and parallel to a central driveshaft
with a swash-plate to convert the reciprocating motion of the pistons into
rotary motion.
This mechanical design is similar to that used frequently
in hydraulic pumps and compact air compressors.
245
-------
;, .1
. I
,
I
I
i
.1
I
i
i
NOTE: ACCESSORY NOT SHOWN
EXH4UST MANIFOLD
,
'.
"
SECTION A-A
~._---_. ~
SELF LUBRICATING
BAND
II
I
FLvwHaL I
I
I
.'~i,;;. .::,C";'!~
-;:;.I..~~-
C-ENGINE
..
COLD SPACE'
PISTON
.~
35.60
-------
~.' . - - .--. .
--'"'-----=-..- <
...; .. ... . - .. .
.. ~,... f' . - .
,.
~ --.-..,.. .--.'--''''-'-~- _.~.--'-' ~- .
F~ _. ::-"'-:-:'.' ..?"-,--:--- -'~:.. .~-':~.-;-.r-.:'"' ~~-"-" '."~':':='-::":.~:~ -.==:~ ~-=-~~~ :-':f~~:7~: u": :'. '''''::'~' ...:..:.=?.._~~:~~ ~ --: -':."-=..:-
. " - '--. - ',-.-- ..... - ... ,-~";, ,.':';" . ~ -_.-._" - ~ . - ~ - .,..'
\
SECTION C-C
BURNER
~
SECTION B - a
- . _..~., .
:- -= ':.." -",- ...-- .~: .-:- . ' : "'7. -.
..--.- --
-.. ... '..-
. .... .
~. --.. - -- .- .. -. ,...
T
I
- MOT EXH:'UST GAS OUT
. .
rG.AS DUCT TO THE HEATER AIID
THE HOT SPACE OF THE CYLINDER
i I
SEIIEI.. GEAR
ACCESSORY DRIvE
II-SELT
FAN ~vE
1'...._--.. . 'C:'
-.-
".
_.,--~'~'"
FIGURE 6.30
~ MECHANICAL SYSTEMS I!\:C.
. NOGA PAR C
-j ~
I I
~ EI/GINE.STLOY-I-£AT~
247
~- ~ - p. 5j~:?£-46CU,"'i
...
-.":"2 :J
-------
r
I
I
i
II
INJECTOR
I. .
,
Ii'! '
8"',
j, '.. \
. . \
.-0-'1'
i :
I t---~ I i
I : : t <
.t=.~
GAS BOTTLE
..
t EIIG8IE
STAR~R
t'
E'!!7.E 6 - ;.:
~1oIEC~~L5Y:
"'"
I
S;!F'- ~.:: £'<:S-.;;E S':E VrEQ
-------
r~--
-
f .l'.~'::~"'~.~-~;::- ;-:-~-- -:: :::~~- - _:._;.~ -;.:;;;::::;:~_.~.".
:~~ <- .~'.,,",.':::. --~ -'--~ .....--" " ~- -
'.
. - - ." -_. ""
...-~~-. -~.:.~-~-=.:.".~-_:::_=-= ::.:;- : ~:'3~::~~"f~ ~: .:::.~
~...;:";'''.':-:-::::.-.:: .._-:-~.:_.
.-oF- -.
- -
.".. ...
-~ . - '.-.
. - -.- --~.-
--:-. ..~.
26.75
~
..
"
GAS
TO THE
~~ts
11.30
FAN
v- BEL T'S IDLER
2 PLACES .
WATER PUMP
B
A
r'
10.00
22.17
-.
~EAR END VI£W
:
'.
'.
1.
'.
249
-.
. '-. ;'1
I.:'~-."" 2":":'::::':'
!'IOTt: MTER ~TS NOT SHO\'IN
2300
10.00
'."
'.
':' ". ~:;,;:'~E- 46CU1N.
....
.... --:, ':t:::
-------
Each cylinder has a cluster of small diameter heater tubes (0.10 in.)
projecting forward from the head of each cylinder into a single combustor
which heats all four sets of tubes.
The regenerator and waste heat exchan- .
.ger elements of the. circuit, shown in Figures 6.29 ,and 6.30, are located
in four containers which are positioned slightly outboard and between
. alternate cylinders.
The hot gas is discharged from the heater section
into the hot space above the piston as the piston ~oves downward.
As the
swash-plate continues to rotate, the piston moves upward, the hot gas is
again discharged through the heater tubes and into the regenerator, which
conserves cycle heat, after which it flows down into the waste heat exchan-
ger, or cooler, and the cooled gas flows into the adjacent cylinder cold
space under the piston., and returns via the same path.
transferred to water or another suitable cooling fluid.
Waste heat is
The circuit from one cylinder to the adjacent cylinder is shown
schematically in Figure 6.10.
Both tubular heater and cooler are designed
for high effectiveness; the regenerator is similarly designed, but is shown
to be a capsule of closely compacted screen in an assembly which has welded
perforated plates at top and bottom to provide suitably low pressure drops.
The piston of the Stirling engine is hollow with three dams to mini-
I
I .
I
mize internal gas circulation which reduces the flow Qf heat from the hot
space to the cold space.
A small diameter bleed tube .is inserted through
the three baffles to balance the pressure in the cold space and the dome
of the piston to inhibit collapse of the high temperature cylinder due
to the combination of temperature and pressure (difference between peak
and average pressure) .
Effectively all of the useful work results from
the pressure difference acting on the top and bottom of pistons.
A small
clearance is maintained between the piston and cylinder since there is
essentially no lubrication between the surfaces but there may.be some
residual misalignment.
Since the region near the bottom of the piston
is relatively cool, a low temperature material. can be used; such' as a
"white metal", which is sufficiently soft to smear under contact,
especially when impregnated with carbon or graphite; or a form of teflon.
One of the reasons that a swash-plate design was chosen was that
it permitted a concept which facilitates linear motion of the piston.
The four pistons are attached to corresponding piston rods which, in
turn, are attached to cross-heads which restrain piston motion to
essentially pure linear motion.
Two hemispherical bearings are located
in each cross-head and ride on opposite sides of the swash-plate, and
are free to tilt and rotate in the cross-head socket to permit .proper
250
-------
contact
pressure on the swash-plate'.
Each hamispherical bearing is lub-
ricated on both sides and extensive tests at Philips and GMR have demon-
strated the suitability of such bearings for operation at high loads for
extended periods of time.
(Thermo Mechanical Systems Company also demon-
strated the suitability of bearings of a similar design under high bearing
loads with friction coefficients of the order of .0015 to 0.002.)
The
ball socket operates as a "squeeze"bearing, much like the piston pin for
a conventional reciprocating engine.
The flat on the opposite side of the
hemispherical bearing operates as a combination squeeze type bearing and
a sliding Kingsbury type thrust bearing, since the swash-plate provides
the relative motion corresponding to 4000 (or less) rpm.
While the lubricated contact between the hemispherical bearings
and the swash-plate results in forces acting at an angle, thus tending
to introduce radial forces which are alternately inward and outward, the
lubricated cross-head, with very small clearances, inhibits transmission
of other than linear movement to the pistons.
Additional support for
each piston rod is gained from the bearings acting within the structural
housing.
The inhibition of lubricant flow into the thermodynamic section is
accomplished by a combination of two seals per piston rod.
One is a
positive seal which is called a "roll-sock"; the other is a hydrodynamic
crown type seal.
The roll-sock and rod are designed such that as the piston
rod moves linearly, the space between the two sealing components is main-
tained at a constant volume (Reference 32).
The crown seals, which are
located at the forward end of the crankcase, consist of a slotted assembly
which is coated with babbitt on the inner surface and forced by a crown,
or slotted, spring to maintain contact with the connecting rod in such a
manner that the seal ring is slightly distorted, forming a very small wedge.
Experiments have shown that the lubricant is pumped upward by the wedge, as
would be expected by lubrication theory, from the crankcase into the constant
volume space maintained between the two seals.
While not shown in Figure .;
6.29, details are shown in Figure 18 of Reference 32, showing a small dia-
meter bleed port to drain the excess lubricant, which is pumped upward from
the crankcase, from the constant volume space between the two seals.
This approach has been found to be successful for Philips but there is some
evidence that the "roll-sock" encountered fatigue failures, thus making
the seal ineffective.
251
-------
An alternate design is used in which an external pump circulates
pressurized oil through the space between seals, thus providing a more
"elastic" liquid cushion, and for most conditions, a less uncertain upper
seal.
Theoretically, the original "crown" and "roll-sock" design
requires filling only once, and maintains the lubricant in the constant
volume space after the engine is stopped.
It should always remain
completely filled and, therefore, adequately resists the gas pressure
at all times.
The engine has a flywheel at the back end of the crankcase and the
power from the cylinders is taken out at this point through a clutch or
hydraulic transmission of conventional design.
The overall length of the
engine, including the diesel fuel (or kerosene) heater assembly is around
36 inches and the maximum burner width and height are 22 inches, respectively.
The accessory drive bevel gear is located on the forward end of the drive-
shaft, and two accessory drive shafts project radially outward between the
connecting rods to provide the necessary drive system.
1he general arrangement of essential accessories is indicated in
Figures 6.31 and 6.32.
Gear driven accessories include:
(1) oil pump,
(2) starter, and (3) helium pump-fuel pump. Belt driven accessories include
(1) water pump, and (2) blower. Neither the helium control system, the
air conditioner, nor the power steering is indicated.
Considerable flexibility in incorporating the accessories can be
exercised, and it appears that the entire 16 ft3 powerplant assembly can
be fitted in the engine envelope space available in a medium size vehicle
(about 21 ft3).
The preheater for the burner is shown schematically in Figure 6.30.
It is of the slotted type and consists of a series of involute plates.
burner exhaust flows outward through alternate passages while the cold
The
air flows inward through the intervening passages, picking up waste heat
and increasing the air temperature into the burner.
The effectiveness of
this heat exchanger is about 85%, which contributes significantly to the
overall thermal efficiency of the engine.
6.6.1
Materials and design specifications are tabulated in Table 6-11.
Weight Comparison
The swash-plate engine design generally has greater compactness
than most other reciprocating powerplants of the same displacement, which
results in lower weights.
As illustrated in Table 6-111, the Stirling
swash-plate design should weigh approximately 700 Ibs, including all of
252
-------
TABLE 6-II
GENERAJ~ AUTOMOTIVE STIRLING ENGINE DESIGN CRITERIA
TMSCO SWASH-PLATE
General
Rated power
Speed
160 HP
3000 rpm
Working fluid
Cylinder bore diameter
helium
2.850 inches
Stroke
1.825 inches
46.57 cu. inch
piston swept volume
Mean cycle pressure
3000 psi
Heater Section
Tube material
347 stainless
o
1300 F
Heater temperature
Regenerator-Cooler Section
Regenerator Sections (4)
Number of cartridges
Natrix material
1 (per sect.ion)
S5 310 screen
Cooler Sections (4)
Tube material
AMS 5577
Coolant type
Coolant flow
Water
Coolant in temperature
Coolant pressure
2 gpm
1500F
75 psi
Housing Material
AMS 5651 (310 S5)
Hain Cylinder
30re material
347 stainless
Drive Mechanism
Wobble Plate and Shaft
Main bearings journal OD
Connecting rod crosshead OD
8.275/2.000 inches
3.000 inches
W.P. pitch diameter
6.500 inches
253
-------
TABLE 6-11 (Continued)
W.P. bearing material
Hemispherical ball (8) OD
Ball material
Shaft material
Counterweight material
Connecting Rods
Length (crosshead-to-piston)
Material
Piston
Piston OD
Piston length
Seals
Shaft seal type
Shaft seal material
Piston seal type
Piston seal material
Crankcase
Main case.
OD
~Jall
Material
Main Bearing Supports
Number required
Material
Oil Pump
Type
Lubricant
Pressure
Flow
Filter type
Drive gear material
Pump gear material
Body material
254
25% leaded bronze
1.750 inch
AIS1 8620 steel
AMS 6415 (alloy steel)
AMS 4674
8.000 inches
AMS 6415 (alloy steel)
2.850 inches
6.500 inches
Ring loaded by spring
SAE No. 12 babbitt
Roll-sock
Viton (Dupont Fluoro-
Elastomer, -20
to +4000F)
10.300 inches
0.225 inch
AMS 6350
2
AMS 6350
Full pressure, gear
SAE No. 10 oil
200 psi
2.5 gpm
Full flow
AMS 6415 (alloy steel)
AMS 6272
AMS 5316
-------
Oil Helium Separator
Type
Filter material
Body material
Oil Cooler
Type
Material
TABLE 6-II (Continued)
255
Static
Wound cotton cord
AMS 5316
Tube
AMS 4015 (52S-0 Aluminum)
-------
TABLE 6-III
STIRLING ENGINE WEIGHT APPROXIMATION:
(Assuming Steel Throughout:
0.286 lb in3)
Engine Weight (including helium bottle,
compressor and controls)
700 lbs
Specific Weight
4.38 lb/hp
Envelope (including fan, starter, pumps,
accelerator, gear, case and engine)
16 ft 3
COMPARABLE (160 HP) OTTO. CYCLE ENGINE:
Bare Engine Weight
400 lbs
Specific Weight
2.50 lb/hp
Envelope (bare engine)
6 ft3
:->56
-------
those components peculiar to the Stirling cycle (helium bottle, helium
compressor, and controls).
One GM 3-cylinder Stirling engine has a specific
weight of 6.6 Ib/HP (using hydrogen), compared to 4.38 Ib/HP for the helium
swash-plate engine for the same power level (150-160 HP).
Also indicated in Table 6-111 is the bare engine weight of a 160 HP
six-cylinder spark ignition engine.
Considering all components which
are not included in the indicated weights of the Stirling and the S.I.
engines (starter, alternator, etc.) to be common to both, the S.l. engine
has a 300 Ib weight advantage.
Inspection of Figure 6.33, which represents the most recent data
available from Philips, reveals that, perhaps with hydrogen as the working
fluid the specific weight of the engine should be about 3.5 Ib!HP for the c
same maximum power and with some enhancement of overall efficiency.
would reduce the bare engine weight to 560 Ibs.
This
The weights projected for this Stirling engine design are reasonably
consistent with the data of Figure 6.33.
The general design of the Stirling
automotive engine is critically dependent upon the solution of primarily
structural problems.
Failure to resolve this difficulty, that of pressure
and heat resistant metals, would result in prohibitively high specific
weight.
As indicated in Figure 6.33, the best extrapolated specific weight
for a Stirling 160 HP engine using helium, with working pressures accept-
able to current metallurgy, is about 8.6 l~AfP.
6.7
Conclusions
1.
The Stirling engine is a relatively high efficiency power-
plant capable of overall performance close to that of the
2.
Diesel engine.
The Stirling engine external combustor has demonstrated very
low emission levels.
CO and HC levels are presently below the
1976 standards and various methods, such as exhaust gas
recirculation, are being tested to reduce the NO emissions.
x
Both hydrogen and helium are acceptable working fluids with
3.
hydrogen giving somewhat high efficiency (helium = 32% at
3000 rprn design speed, direct heating) while helium is
significantly less hazardous.
Helium is the preferred
working fluid until the hazards associated with the hydrogen
Stirling engine can be more clearly defined.
4.
The cost and complexity of the present Stirling engine make
257
-'. \
-------
too.
8.9 Working Heater Heating
Class . Type FIJ.iid Work. Press.
60 A Laboratory Helium 1500 psi Direct
B Laboratory Helium 3000 psi Direct
C Automotiv~~~e~ium 3000 psi Direct
D Automotive* 'HeliUm 3000 psi Indirec
~ 40 E Automotive* Hydrogen 3000 psi Direct
...... F Automotive* Hydrogen 3000 psi. InriiYQc
,Q
.-I Heater Working Temp. :::; 1380°1:'
Coo l.~w t 1n1et Temp. ., 131° F
+J
.C: *P.rojected Data
01
...
Q)
~ 20.,
()
'M
4-1
'M
U
Q)
PI
U)
10
8
'"
6 ~
'"
"
F "
4 "
"
'"
2
1
10
20
30
40
50 60 7080. iod
200
JOO
Shaft Horsepower
FIGURE Ei~:3:3
, . f. . h (28)
stirling Englne Specl lC We1g t.
258
1305
-------
it impractical for an automobile powerplant.
The use of hydrogen
working fluid, higher engine speeds, and a barrel engine con-
figuration are all means by which the Stirling engine automotive
characteristics can be improved.
259
-------
7.0
RANKINE AND OTHER CLOSED CYCLE WORKING FLUID HAZARDS
7.1
INTRODUCTION.
In recent years emphasis has been placed upon the Rankine cycle (and
other closed cycle systems such as the Stirling engine) as prime candidates
for reducing the emissions of vehicle powerplants.
The prime reason that
the closed cycle systems merit serious consideration is that they use ex-
ternal combustors in which the combustion process can be controlled to
produce relatively low values of emissions.
The specific objective of the work reported in this section was to
evaluate the potential hazards associated with Rankine cycle organic work-
ing fluids, and other candidate liquid and gaseous working fluids for low
emission closed cycle powerplants.
7.2
HAZARDS ASSOCIATED WITH CLOSED CYCLE WORKING FLUIDS
With the present interest in closed cycle powerplants (Rankine,
Stirling, etc.) for low emission automotive propulsion systems, consider-
ation must be given to the hazards associated with the working fluids
used in these new systems.
Working fluids will normally pass through
several environmental phases during the manufacture and use in the
standard passenger automobile.
These environmental phases include:
1.
Shipment of fluid raw materials
Manufacture of the fluid
2.
3.
4.
Shipment of the fluid
Manufacture of powerplant system and vehicle assembly
5.
Passenger vehicle under normal operation
Loss of working fluid during vehicle accidents
6.
8.
Maintenance and repair of powerplant and vehicle
Scrapping of vehicle or engine.
7.
7.3
VEHICLE ACCIDENTS
The National Safety Council publishes a comprehensive tabulation of
260
-------
of how people die in vehicle related accidents in the United states.
Per-
tinent statistics from Reference 38 are included in Attachment I.
During
1969, accidents involving motor-vehicles can be summarized as follows:
Numbers of Drivers Deaths and
Accidents (Vehicles) Injuries
Involved
Fatal 47,000 70,000 56,000
Nonfatal Injury (disabling) 1,300,000 2,400,000 2,000,000
Property damage 14,200,000 24,300,000
Total (rounded) 15,500,000 26,800,000 2,060,000
As seen above, there were 26,800,000 vehicles involved in accidents
in which property damage or injury resulted.
With 107,000,000 registered
vehicles in the U.S., this means that approximately 25 percent are in-
volved in accidents each year.
With this many vehicles involved in ac-
cidents, it appears worthwhile to examine the hazards associated with an
engine system using a flammable and/or toxic working fluid.
7.4
FLUID FLAMMABILITY HAZARD
About 600-700 persons die annually in motor-vehicle fires resulting
from accidents, according to the experience in Illinois tabulated by the
Illinois Department of Public Health (39).
National tabulations are not
made of deaths, and no records are kept of injuries or the total number
of fires; however, of the passenger cars involved in injury accidents, an
estimated 10,000 had fires, based on a study made by Cornell Automotive
Crash Injury Research.
This is a relative low incidence of fires consid-
ering the total number of vehicles involved in accidents.
However, there
is perhaps only 1/2 pound of fuel in the engine compartment (carburetor,
fuel pump, etc.) where there are a large number of potential ignition
sources in case of accident.
The remainder of the fuel is generally 10-
cated beneath the car and aft of the rear axle where there is a minimum
number of ignition sources in case of crash.
In a closed cycle alternate powerplant using a flammable working
fluid, the hazards involved might 'be likened to using a flammable liquid
261
-------
in the radiator and cooling system of the present automobile in addition
to the fuel tank and combustion system.
Instead of 1/2 lb. of flammable
liquid in the engine compartment, there would be 20-30 pounds.
Portions
of the fluid would exist under relatively high pressures and temperatures,
and the cooling system is frequently punctured in relatively minor crashes.
Even with added crash protection, it is quite likely that the death toll
due to crash-fires would increase manyfold if a flammable working fluid
was used in a closed cycle automotive powerplant.
The relationship of volatility and fire hazard has been the subject
of much controversy (1).
There is no simple scientific measure for fire
hazard, because it depends on particular circumstances.
However, the fol-
lowing may be stated as facts which have a bearing on the subject:
1.
The atmosphere over fuel in tanks may be explosive with
either gasoline, jet fuel, or diesel oil, depending on
temperature.
rich limits)
Por example, the explosive range (lean and
for gasoline is from -800F to OOF, for jet
o 0
60 F to 110 F, and for light diesel oil
These are approximate figures assuming
fuel (kerosene)
l400F to l800p.
the air space in the tank to be at sea-level pressure (40).
2.
In partly ventillated enclosed spaces, such as boat bilges
and aircraft-wing interiors, gasoline is usually more
likely to furnish a combustible atmosphere than are the
heavier fuels.
3.
Gasoline fires develop to great intensity almost immedi-
ately after ignition.
much more slowly (41).
Fires with heavier fuels develop
4.
The heavier fuels, having lower ignition temperatures,
ignite more readily than gasoline when spilled on hot
surfaces.
On the other hand, the vapor from spilled
gasoline are more likely to ignite from electric sparks
than is the vapor from heavier fuels, because of the greater
likelihood of a combustible mixture at the spark.
262
-------
From the aforementioned considerations it would appear that, under
most circumstances, increasing volatility tends to increase fire hazard,
and therefore, the theory that the use of diesel oil in place of gasoline
reduces fire hazard is generally correct.
However, experienGe with jet
airplanes, which use a fuel less volatile than gasoline, has not shown
any noticeably smaller incidence of fire after a crash than has been ex-
perienced with gasoline-powered aircraft (42).
Because of the large number of automobiles involved in accidents
each year (26,800,000), and because of the large number which are serious
enough to involve personal injury (2,400,000), it is believed that the
use of a flammable working fluid in a closed cycle engine would be in-
viting health hazards enormously greater than the current automotive
exhaust emission problem.
One possible exception might be where the
working fluid is a gas and the toral weight of fluid is only a fraction
of a pound as in the Stirling engine.
In this case, the total amount of
heat available by burning the working fluid might be sufficiently small
as to make the fire and explosion hazard an acceptable risk.
7.5
FLUID TOXICITY HAZARD
Before evaluating the toxic hazards associated with candidate closed
cycle working fluids, it is believed worthwhile to look at the hazards as-
sociated with current motor fuels.
The two principal hazards are oral in-
take and inhalation of gases and vapors.
Attachment I ,shows that about 2500 persons died in 1969 from poison-
ing from solids and liquids.
Details of the deaths in 1969 are not avail-
able, but 2506 such deaths in 1967 involved the following:
Sleeping pills
744
506
Barbiturates
Morphine
Alcohol
176
158
Aspirin
Other and unspecified drugs
153
281
263
-------
Lead 61
Industrial solvents 36
Petroleum products 31
Arsenic 31
Other and unspecified solids and liquids 329
Inasmuch as only 31 persons died from the oral intake of petroleum
products, and the number from motor fuels is probably less, it would ap-
pear that the hazards associated with the oral intake of current motor
fuels is very small.
Attachment I also shows that about 1700 persons died from poison-
ing by gases and vapors. Details of the deaths in 1969 are not avail-
able, but 1574 such deaths 0ccurred in 1967 as follow~:
Motor vehicle exhaust gas while vehicle was standing
Other carbon monoxide gas, largely due to defective
819
heating equipment
Utility gas
318
280
Other specified gases and vapors
Unspecified
144
13
The above tabulation does not indicate that any deaths occurred
from the inhalation of motor fuel gases or vapors.
However, 819 persons
accidentally died from breathing motor vehicle exhaust gas.
It could
well be that a major contribution of the Federal Exhaust Emission Pro-
gram will be to reduce the number of these deaths as the emission
standards for carbon monoxide become more stringent.
Attachment II presents the toxicity characteristics of some com-
mon petroleum products as tabulated by the Shell Oil Company.
It appears
that common motor fuels are not a significant toxic hazard to the motor-
ing public.
If a toxic working fluid were used in closed cycle motor vehicle
powerplants it again appears that vehicle accidents would be the most
critical environmental phase.
With 26,800,000 vehicles involved in ac-
264
-------
cidents, if they could be designed Sf) that only one in ten encountered a
break in the closed loop system, the£~ would still be something on the
order of 67,000,000 pounds of working fluid dumped on the ground and into
the ~tmosphere annually.
With a toxic working fluid, many persons trap-
ped in vehicle wreckage might perish that would otherwise survive.
In
addition, there would probably be much less willingness on the part of
bystandards to help the i.njured if there was fear of toxic poisoning.
It therefore appears that a working fluid with any greater toxicity than
that of gasoline should not be considered for automotive use.
Earlier in this discussion on hazards a list of environmental
phases were set forth through which closed cycle working fluids would
pass.
It appears that hazards associated with vehicle acci~~nts are the
< '
. .,'
most critical phase and this phase dictates that the fluid 'be nonfiam-
mable and nontoxic.
It is therefore concluded that a nonflammable, non-
toxic working fluid would not present any significant hazards in any of
the other environmental phases previously listed.
7.6
HAZARDS OF CANDIDATE CLOSED CYCLE WORKING FLUIDS
Candidate closed cycle Rankine working fluids include CP-27 (mono-
chlorobenzene), CP-34 (thiophene), Fluorinol 85 (trifluoroethanol), P-ID,
and AEF-78.
Data on these fluids are included in Attachments III, IV, V,
and VI.
It is recommended that the definition of fluid flammability and
toxicity be included to mean whether or not a fluid will burn or produce
toxic liquids or vapors if a system rupture occurs in either the hot or
cold portion of the closed loop.
Data from Aerojet (Attachment VI) on
AEF-78 indicate the toxicity hazard associated with the escape of the
high temperature fluid is not known.
Therefore, it is not yet ~own if
AEF-78 would be an acceptable working fluid.
Likewise, the Halocarbon
Products Corporation (Attachment IV) indicates that Fluorinol 85 has an
observable flash point but a very low heat of combustion (less than 3500/
BTU/lb) .
This indicates that the fluid could add some heat to an exist-
ing fire, but could not by itself sustain combustion.
It would therefore
appear that this fluid would be safe for use in a closed cycle system.
265
-------
Data on the flammability and toxicity characteristics of the vari-
ous candidate working fluids and other possible fluids and gases (for
both the Rankine and other closed cycles) are tabulated in Table 7-1.
From the safety standpoint, it is concluded that water, Fluorinol
85, P-ID, Freon, and the nonflammable gases all qualify as candidate work-
ing fluids (Freon, upon contact with hot surfaces, can break down into
phosgene, but there is no evidence that the rupture of an automotive air
conditioning system adds a hazard in automobile accidents).
The accept-
ability of AEF-78 is unknown at this time.
CP-27, CP-34, and hydrogen
(for the Stirling cycle) should be eliminated as candidate working fluids.
7.7
CONCLUSIONS
From the results of the study reported herein, the following con-
clusions are drawn:
1.
Each year approximately 26,000,000 vehicles are involved in
accidents resulting in property damage, injury, or fatalities.
2.
In a closed cycle powerplant, the working fluid must neces-
sarily circulate through various components of the engine
system where there is a relatively large potential for closed
system rupture and fire in case of a crash.
It is therefore
concluded that any candidate working fluid should not support
combustion nor be of an explosive nature.
3.
Toxicity to a large extent is relative.
Statistics of the
National Safety Council indicate that there is no significant
health hazard associated with the use of common motor fuels.
It is therefore concluded that any candidate closed cycle
working fluid possess no greater toxicity than current motor
fuels.
4.
The use of a flammable or toxic working fluid would be in-
viting a health hazard enormously greater than the current
automotive exhaust emission problem.
266
-------
5.
The National Safety Council reports that about 800 persons
accidentally die annually from breathing motor vehicle ex-
haust fumes from non-moving vehicles.
It could well be
that a major contribution of the Federal Exhaust Emission
Program will be to significantly reduce the number of these
deaths as the emission standards for carbon monoxide become
more stringent.
267
-------
TABLE 7-1
HAZARDS OF CANDIDATE CLOSED CYCLE WORKING FLUIDS
FLU! D FLAMMABLE TOXIC SUITABLE
Water no no yes
CP-27 (Attachment III) yes ? no
CP-34 (Attachment III) yes yes-vapor no
Fluorinol 85 (Attachment IV) apparently possible probably
no eye hazard yes
P-lD (Attachment V) no apparently probably
no yes
AEF-78 (Attachment VI) no unknown at unknown
high temp.
Freon no no yes
Hydrogen yes no no
Air, N2, A, He, etc. no no yes
268
-------
7.8
ATTACHMENTS
ATTACHMENT I
1969 NATIONAL SAFETY COUNCIL ACCIDENT FACTS
269
-------
How people died accidentally In 1969 ~ Firearms
Type of accident and age of victim
o..d. ChaD«" Popll..l.t10Q
n AU --'s T...l from l'Jt.a Death Ra~
115,800 0% 57.0
".
DII.TIIItAn:t 1111;'" KAI,.(I
.~
.0 . '. .". . ;
~: ~..~ .. - - - ~M:
H11IS- 7,000 8.tOO 24.100 1:4,000 U..8CIO 10.000 17,".)0
The tenn "accidents" covers
m08t deaths from violence, but
specifically e x cI u des homicides,
suicides and deaths in war opera-
tiOIl8.
.
..- H
..,.. 150
,-II
..0
D Poi808ing bJ selids aad liquids-
!! Motor-vehicle accidents__56,400
MUtt .'T!' t IIIOTr ~(I
10'
40
ISUtlt
;',200
-'I Falls
tV
-...J
o
',>00
11-" n.tlu
3,000 10,600
IIlrowninl
~:"1IIiiiii
", - , . ,... . . ,; .. ~
. ,', ,'. - , .\ "- -",,~:.:..~, ~.~ -' ~- >
to.:; '-'.' ,:0, ....>1" -- -. .
I ' .
o .-.' ,', '-. '
IC(- H "'14 fI-r4 n.... 4i~ U-JiI ~ i"8
IUlIS- 900 1,600 2.000 ',300 1.000 300 200
n Fires, burns, and
WI deaths associated with fires-
KAnla.Tlt ."on ICAl..ZI
..
J"
..
,. :.
. -'''',
M! H
HATIf- 800
"'_;:;:",M"""r~
"'01.4 ,~'P- '<'.4, '.
H It-" IH4 ...... n-w niffQ
700 400 1,1:00 1,000 '00 1,100
6
AC~IO(NT ~ACTS no I ..:';~ .
+ 2%
_&T1t UTI t 18011 IMoIUI
. ~ . .'~ '" .
:~,.~""',
'6 - '.
Ja..!... H '"'~' 1$-"" t..... 4H8 ...." nlnn
IflTIS- 110 40 1500 100 800 70 40
27.9
Includes deaths involving me-
chanically or electrically powered
highway-transport vehicles in
motion (except those on rails),
both on and off the highway or
atnet.
..;;i Machinery accidents
19,000
-5%
9.4
Includes deaths from falls from
one level to another or on the
same level, except falls in or
f~m railway-, road-, water-, or
lUl'-transport vehicles, or those
occurring in cataclysms.
.
--
IUTIf-
H
.0
....
150
II-It
500
n...
800
11-14
700
"""
100
n,nu
50
7,300
D Poisoning by gases aIId V8jJors- 1,700
-1 %
3.6
ICATM un' fllClT[ KALI:l
.
Includes all drownings (work
and nonwork) in boat accidents
and those resulting from swim-
minJil', playing in the wate~, or
falling in. Excludes drownings in
1l00ds and other cataclysms.
-'.',<
. -
.
..
I01Il-
7,100
-- All other types
-5%
3.5
MAn UTI' I8ClTt ICAf..Il
. :~
',.p",'~';'''''i,.,,_-- ).~"~,..,, 'F,,-
..~ .. ':'4." "7 . ~ ,- . (r~~ .
I.' . ..':-'..,..'?", ''''''''''''''' -
. . . - ,.. ' , , '
IU- H I-M ,,""" rH4 ..... ..." JUIt"
IUIII- 2.500 1.000 2.000 3.700 4,200 1,200 1,000
'tDeatiu ... 100,000 popalatloa I.. each ... poap.
Includes deaths from fires'
burns, and from injuries in con:
~grations-such as asphyxiation,
!&lis, and struck by falling ob-
Jects. Excludes burns from hot
objects or liquids.
Dnlh
Tout
a......
'~1'"
P...l.ad..
Dead. ..oat
2,&"
'%
1.1
Indudea d_ths in ftrearma 110-
cidllltb in reereatiOl\&I adiyWa
or OlD hame premises IUId a ~l
n11Dlher (less than S per ceDi)
from explosions of dynamite.
bon1hs, grenades, ek. Excludea
deaths in war operatiOIUl.
2,500
+ 4°A»
1.1
-Includes deaths from mediclnes.
as well as from commonly recog-
~fUttoisons. Xu.hroom and
sh poisoning deattw ale in-
cluded.. but fatal poiaonin&' tmm
spoDed fooda-botulJam, etc.---ue
classified as disease death&
2,000
-5%
1.0
IJIdndea deaths involving all
trpes of machinery. Nunf half
occur on farms m the course of
work; of these, ap)lroXimately
three-fourths invom tractors.
About one-third oecur in induatly.
Five per cent occur in the home.
+ 8%
I.'
Principally carbon monoxide
due to incomplete combustion. in-
vol'riDg cooking stoves, heatiBg
~ent and standing motor
veJdCIe& Excludes deaths in eon-
fiagratiOIl8, or associated witb
trall8JlOrt vehiclea in motion.
16,400
+1%
8.1
Most important types included
are: inhalation or ingestion of
food or other object, mechaniea1
sutl'oeation. air transportatio..,
blow by falling object, electric
current, raIlroad, excessive heat
,or cold, cataclysm.
to..tI.a per 100.000 ...-JatI-
ALl ACC,C("rs_-_-
.7
-------
t
MOTOR-VEHICLE ACCIDENTS, 1969
(See alao pqe Z for National Health Survey total8)
MOTOR -VEHICLE ACCIDENTS, 1950-19&9
Deaths- (See- page 12 for effect of ICD- Eighth Revision)
Injuries_- (Disabling beyond the day 01 accident)
Costs_- (For certain details 01 cost see page 5)
Motor-vehicle mileage__-
Death rate (per 100,000,000 miles 01 travel)
Registered vehicles in Ihe U.S_-_. --
Ucensed drivers in the U.S.-
56,400
2,000,000
$12,200,000,000
1,065,000,000,000
- _5.30
-- _107,000,000
_107.500,000
1950-19119
Deaths_- - . _375,000
InjurieL- --- - -- - _13,350,000
Costs_- - - - _$45.5 billion
Motor-vehicle mileage- - - _5,800 billion
11/61-11169
415,000
17,200,800
$89.6 billion
8,700 billion
1
I
I
,
l.twHn 1912 and 1969, motor.vehicl. death, FM' 10,000 regilt.red wlhid., ""er. re.
dUCld as per clnt, from 33 to about S. IMUeage data Wire not available in t 911.1
In 1912, thlr. we,. 3,100 fatalitl.. when the number of ragist.r.d vehicle. tololl.d only
950,000. '" 1969, ,hI" wlr. 56,400 'alomi.s, but "gl'frolio", u~a,...d to 107 million.
)Iotor-\"ehicle death and injury totals increased by more than one-fourth
in the decade of the sixtie" a" compared to the fiftit'~. The annual death toll
incrt'a~t'd almo~( without interruption to a high of 56..Hh1 from :H.76:~ in 1950.
Accidenllolals
Number 01
Accidents
Drt"'~r' ('/ehi:1e~)
InvoJ\jt!d
Accident costs nearly doubled in the sixties. dUt' to mOl'" a('ejdl'nts and
higher costs per accident.
JlotoJ'-vehicie mileage increased fifty per cent in the sixtie~ and the mill'agoe
death rate declined by sixteen per cent. The death rate per 100 million
n.hicle milcs was 7.59 in 1950 and declined steadily to 5.11; in 1%1; it then
increased to 5.70 in 1966. after which it declined to 5.30 ill 1969.
Registered vehicles more than doubled over the two decades to 107 million
in 1969 from about 49 million in 1950.
Licensed drivers increased nearly three-fourths over the two decades to
nearly 10S million in 1969 irom more than 62 million in 1950.
- -- - --
47,600
1.300,000
14,200,000
15,500,000
70,700
2,400,000
24,300.000
26,800.000
Acc.iden\ Totals
Sumber of Acddent'J
~-1939 1960-1969
:0
-...J
r-'
Fatal
Nonlatal injury --- (d'Slblin&) - -
Property da.magf ('Ot(lud,"& "o"diSlbtin~ inl"...ry ~~.nt.)
Tolal (rounded)- ------
FataL- -- - _325,000
Nonfatal injury- --- - _8,900,000
Propert~ damage - ___88,000,000
Total (rounded)-- 97,250,000
400,000
11,000,000
115,000,000
126,400,000
Dthen (Vehietes) Invohe4
1950,1969 1960-1969
450,000 580,000
14,600,000 19,800,000
150,000,000 200,000,000
165,000,000 220,000,000
'(~
...~.;~..!!t~~~ ~""~1"~"';;':~~>~:~~~""'..fJ.:"'~
;;,~~:,.}:-;. '{'~,.:.' :~"f: _.'.:.:"~'- .>~)S1~:~~?:t~~1~~
~1.~_Th~Mo~~~-~~~C_lCl~iT~~5:;ion t~~.~~~
. "~i.~ DEATHS andPEATH RATES d'''-..'.....,'';!?- 'Jlif":>".""'"
:~~~.~~..~~!..~.'1~~~:,'~;:~~:-;~~fj.,:~~~~'R~~
~-i
-------
How people died in molor-vehicle accidents, 1969
Type of accident and age of victim
~
All motor-vehicle accidents
..
eUTM ..TI t I~TI IUl..U
..
o.
.
AIf- 0-'
'UTIS- 2.100
1-" IS-M ,.14 41. tI-N
4,100 17, TOO ".100 11,300 .,100
"1"tI
3,200
~
o
~
n..-..th
r..tal
__56,400
I:rban
Rural
18,000
38,400
D.aoJo
I T.~
~ CoilisiOGS with fixed objects_- - - 4,200
I.h~n.f' t"opul..llu"
fr..". I-'hl\ UfOllth K,lt,.:-
toC"'" un. ...... tc.t..,AI
.
. . .
- . .
. .
. ~ "-.~"?'.
.......
;;.~
, - ~
+ 2 °/0
21.9
+ 3%
+ 2':0
.
m-
IlAm-
It.."
150
n","
100
IH4
',200
!HI
100
...
00
to"
100
,.."
'.OOC
Includes dea{hs im'olving me-
chanically or <'Iectrically powerE'd
highway-transport vehicles in
motion (except those on rails),
both on and off the highway or
street.
Collisions between motor vehicles___-- _24,000
"
10
.
1CI- .-,
m1lS- 100
"-M It... "....
7,100 6,500 5,""
.1-14
',850
UI"ft
1,400
....
~oo
~..)
~
~
. loncollision in roadway,
overturning, running off roadway
0"""" -.ent Illon KAU'1
to
It
. i .
. ' ,,-
~ . ,~
. .' -.-.. . .' . -.. 1i
. ~'"".'I"~" " . . - - .:.:<;'
. . , ~
. ;,.'
10
.
UE- .-,
tUlII- 300
,...
650
1$-24 n-t4 .......
7.000 4,'00 2,800
,1-"
150
nuna
'00
. Pedestrian accidents__-
DtATH ltaT! t C.OTII SCALlI
..
,.
10
42
..CCIDINT 'ACTS 1970 1IIIt1e8
Urban
Rural
6,000
1~,0(1(1
16,000 tt
Urban
Rural
3,200
12,800
CE Collisions with railroad trains
OtAf. 'UTtt C.Of'llG&\.t'
.
+7%
11.9
. -. - 4 . - , .'::'.~ '.~'
, -. . -'
. . " -." ~ - .
. .. - - '''''.-
. . . .~.
-- .. .
. .
-'~ c~~ -' ':,. - ':. - ~~ . ~~.
Includes d~aths from collisiol..
of two or more motor vehicles.
Motorized bicycles and scooters,
trolley buses, and farm tractors
or road machinery travelling on
highways are motor vehicles.
+ 9%
+ 6':;'
,
.u-
1U1IS-
...
50
1$-,.
"0
n~M
400
n-14
'00
n,.m
00
,'.
'00
.....
'10
~ ...-J.aue.
r.... 1968 n... a.-*
ft 1.1
l" rban 1,550 tt
Hural 2,65tJ tt
Includes deaths irl':n cl,llisil'Tls
with fixed obje~ts >uch a..< ,,-all!
and abutments, where thl.' c"lli-
sion occurred whit.. all wheels of
the vehicle were still ,'n thl' road.
(See comment OD noncollision.)
1,500 -6% 0.7
l7rban 430 -10~Q
Rural 1,0iO - 4<;"
Includes deaths from colJisions
of motor vehicles (moving or
stalled) and railroad yehicles at
public or private grade crossings.
In other types of accidents, elassi-
fication requires motor vehicle to
be in motion.
7.9
~ Collisions with bicycles_- ----
0.4
OIAT-.. .An. t I-on ilCAUl
.
tt
tt
. . ~ -. .~ ... -- . ~. ,-' ~,'''.~.:', - - :~:~.,~'!. ~;~~
. . . . '.::. :. ..'\~!~
. . .' . . , .~.~
'- ~-
Includes deaths in all types of
noncollision accidents. Classifica-
tion is according to first event. If
car runs off roadway and then
strikes fixed object, death is
charged as run off road accident.
. 9,800
Urban
Rural
.
&Cf-
!UTI!-
,,-"
150
"...
40
lS-f4
20
11'''£1
20
...
20
.."
550
.....
40
0%
4.9
I!I!
820
+ 3%
Urban
Rural
420
400
0%
+ 5%
Includes deaths of bicyclists
and motor-vehicle occupants from
collisions between bicycles and
motor vehicles on streets, hi¥h-
ways, private driveways, parking
lots, etc.
6,(00
3,400
0%
0%
Other collisions
<8111als, aDimaf.llnwn n"cIes, street cars)
Includes deaths from motor-vehicle collisions not ;;pecified in other cate-
gories above. Most of the deaths arORe out of accidents involving animals or
animal-drawn \'f"1icles. Deaths from accidents involving street cars are not yet
known for 1969.
Includes all deaths of persons
struck by motor vehicles, either
on or oft a street or highway, re-
gardless of the circumstances of
the accident.
tOeatlt. plr 100,000 po pula lion In ..uh 88e j!:roup.
t+T.'al. are Dot eornfl.rable 10 prewJou8 ,ear. See p...e "S.
.. Deatb n" ... leu than. 0.1.
-.-----.
80 -20%
*.
tDeal.he per 100,000 popalalloo.
- _IoIOTOIt.VEHICLL - --43
-------
ATTACHMENT II
HEALTH HAZARDS OF PETROLEUM PRODUCTS
273
-------
INDUSTRIAL HYGIENE BULLETIN
HEALTH HAZARDS
OF
PETROLEUM PRODUCTS
Fuels, Lubricants, Solvents
Waxes, Asphalts
Second Edition
January 1969
SHELL OIL COMPANY
PRODUCTS APPLICATION DEPARTMENT
50 West 50th Street
New York, New York, 10020
274
-------
INDUSTRIAL HYGIENE BULLETIN
INTRODUCTION
Petroleum products are used for a
variety of purposes. In order to insure optimum
safety it is essential that the user be aware of the
potential hazards involved and that he use the
proper precautions.
The fire and explosion hazard, par-
ticularly in the case of such products as gasoline,
liquefied petroleum. gas, and hydrocarbon sol-
vents, is generally quite well understood. How-
ever, hazards associated with toxicological
properties oftcn appear to be somewhat con-
fusing. One reads, for example, of children who
have died following the accidental swallowing of a
small amount of kerosine. although kerosine is
rated as "slightly toxic" or "practically non-
toxic" in acute toxicity evaluations. Certain
household labels required by the Federal Hazard-
ous Substances Act may appear to the uninitiated
to include statements which are inconsistent with
each other. Labels on household products contain-
ing ten per cent or more of "petroleum distillates"
must, for example, include both of the following
statements.
"Harmful or fatal if swallowed" and"!f
swallowed, do not induce vomiting".
The purpose of this bulletin is to pro-
vide the reader with a clearer concept of the
toxicological properties of petroleum products
and the hazards associated with their use.
ACUTE TOXICITY
Acute toxicity, i.e., the effect of a
single dose. is of primary importance as appli ~d to
petroleum products, since very few of these prod-
ucts givc risc to chronic toxicity on repcated
administration. A notable exception is benzene,
which is unique among petroleum products in
that it is capable of destroying blood-formiFlg
tissues.
The acute toxicity of a material is
usuatly expressed as the LD-SO, which is the
minimum dosage required to kill fifty per cent of
a group of test animals. The LD-SO, which is
usually expressed in milligrams or grams per kilo-
gram of body weight, is dependent upon the
method of application (usually either orally or
through the skin) and upon the test animal used.
Acute inhalation toxicity is normally expressed as
a lethal concentration (LC-SO), which is the
minimum concentration which will kill fifty per
cent of the test animals in a stated period of ex-
posure, for example. 4 hours.
A classification known as the Hodge-
Sterner scale is in common use iEt the industrial
field. It is a six-step scale ranging from "ex-
tremely tox.ic" to "relatively harmless". A some-
what different scale, known as the Gleason,
Gosselin, Hodge, is preferred by the Poison Con-
trol Centers (see later section). Both scales are
described in Table 1. It will be observed that the
former includes percutaneous an.d inhalation as .
well as oral toxicity, while the latter covers only
oral. Unfortunately, a given descriptive rating,
e.g., "practically non-toxic" does not correspond
to the same LD-SO on both scales. In describing
acute toxicity, except in very broad terms, it is
therefore necessary to designate which scale is
being used.
Petroleum products as a general rule
fall either in the "practically non-toxic", the
"relatively harmless" or the "slightly toxic" cate-
gory (Hodge-Sterner). Most of the highly paraf-
finic or naphthenic products are "relatively harm-
27-5
-------
2
less" or "practically non-toxic", while those
containing. high concentrations of aromatics are
more likely .to be rated "practically non-toxic" or
"slightly toxic". Thus, regardless of whether ad-
ministration is by month (oral toxicity) or through
the skin (percutaneous toxicity), the effects
of petroleum products are not particularly alarm-
ing. However, "light petroleum distillates"
(gasoline, kerosine, naphtha, mineral spirits, etc.),
while relatively inert when applied to the skin or
introduced into the digestive tract, are much
more dangerous when upirated in liquid form in-
to the lungs. The so-called aspiration' hazard is
discussed below.
Acute inhalation toxicity-.is likewise
seldom a problem. Except in the case of the more
volatile products (those with initial boiling points
of about 3000F or lower) it is generally impos-
sible to determine the LC-50, since the saturated
vapor at room temperature is not sufficiently con-
centrated to be toxic. A value which is of weater
practical significance than the LC-50 is the
Threshold Limit Value, or Maximum Allowable
Concentration as it was formerly called. Thresh-
old Limit Values represent concentrations which
can be tolerated by workmen throughout their
working hours, day after day. without adverse ef-
fect. They are discussed in greater detail in
another section.
ASPIRATION HAZARD
The most s'rious toxicological hazard
associated with petroleum products, particularly
with respect to the ingestion of light petroleum
distillates such as kerosine, charcoal lighter fluid.
etc., by children, is the aspir2tion hazard. Prod-
ucts which have little toxicological effect in the
digestive tract can, if aspirated in liquid form into
_the..lung:; cause a chemical pneumonitis which is
often fatal. The greatest danger in connection
with accidental ingestion of such products is not
the effect in the stomach, but the subsequent ef-
fect which may follow if, during gastric lavage or
induced vomiting, the patient aspirates some .of
the product into the lungs. Gerarde* has deter-
mined, on the basis of tests on rats in which
kerosine was aspirated directly into the lungs,
that the ratio of the intratracheal L0-50 to the
oral LD-50 is approximately 1 to 140. Thus, kero-
sine is 140 times as toxic in the lungs as in the
digestive tract of chis particular test animal.
Gerarde has studied the aspiration
toxicity of a large number of petroleum produ~ts,
and found that the mortality rate decreases with
increasing viscosity. Products having a viscosity
less than 50 Saybolt Universal seconds at 100 de-
grees Fahrenheit" possess the greatest hazard in
this respect. Those with viscosity values between
50 and 80 may be considered to be borderline,
and those with a viscosity above 80 are free from
the aspiration hazard.
Thus, products such as gasoline, kero-
sin e. 1 ighter fluid. mineral spirits. naphtha.
Stoddard solvent, and mineral seal oil are all
capable of causing serious effects if introduced in
liquid form into the lungs. This is the basis for the
requirement in the Federal Hazardous Substances
Act that household containers containing ten per
cent or more of such products bear the following
the following statements on their labels:
"DANGER! "
"Harmful or fatal if swallowed"
"If swallowed. do rIot i~duc(! I'orni til/g"
"Keep out of reach of children'J .
IRRiTATiON
Some degree of irritation may be ex-
perienced when liquid petroleum products arc
splashed into the eyes. Aromatics are appreciably
"H. W. Gerarde, Architeves of Environmental Health, March 1963,
Volume 6, pages 329-341.
. 276
-------
more irritating than the naphthenes and paraffms
of comparable boiling range. Any petroleum
product accidentally introduced into the eyes
should be removed by washing with a generous
quantity of water.
Because of their ability to dissolve
natural fats from the skin, most petroleum prod-
ucts can cause some degree of skin irritation, par-
ticularly if the exposure is prolonged. Aromatics
ar" somewhat more irritating to the skin than the
non-aromatic products, but the difference be-
tween aromatics and satW'ates is much less pro-
nounced than in the case of eye irritation.
prolonged skin contact with all liquid petroleum
products should, therefore, be avoided, and any
product which gets on the skin should be re-
moved by wiping, followed by washing with soap
and water.
THRESHOLD liMIT VALUES
The American Conference of Govern-
mental Industrial Hygienists (ACGIH) is a com-
mittee of experts, currently fourteen in number,
which meets annually to determine recommended
Threshold Limit Values (TL V's) and to prepare a
report for publication. TL V's are time-weighted
average concentrations believed to represent con-
ditions which can be tolerated by nearly all
workers, throughout the working week, without
adverse effect. In a few instances, indicated in the
list by a "C" designation, the value represents a
"ceiling" not to be exceeded.
The 1968 list, adopted at the 30th.
annual meeting, contains TL V's for approxi-
mately 400 materials. Table 2 (shown on pages 7-
10 contains that portion of the preface of
TL V's for 1968 which explains the meaning of
TL V's. It also shows all of the 1968 values on
petroleum products plus several others of general
interest for purposes of comparison. it will be
observed that the TL V's of light paraffinic,
naphthenic and olefinic products (e.g., propane,
3
hexane, cyclohexane and cyclohexene) are gen-
erally in the 300 to 1000 ppm range. Those of
aromatics, on the other hana, range from 0.2 ppm
for diphenyl, through 25 ppm for benzene, to
200 ppm for toluene.
FEDJERAIL HAZARDOUS SUBSTANCJES
The Federal Hazardous SubstaB1c~s
Act, originally called the Federal Hazardous Sub-
stances Labeling Act, was passed in 1960, largely
to protect children from injury or death through
misapplication of household products. It had
been estimated, prior to passage of the law, that
600,000 children swallowed household aids
annually, and that about 500 of these died.
Household products which are either
"corrosive", "extremely flammable" (flash point
below 200F) or "highly toxic" (ora! LD-50 of 50
milligrams or less per kilogram; skin (LD-SO of
200 milligrams or less per kilogram; LC-50 of 200
p P m 0 r 1 e s s ) must bear the signal word
"DANGER", together with the appropriate state-
ments describing the material al\d the hazards
involved.
Products which are either "toxic"
(oral LD-50 between 50 milligrams and 5 grams
per kilogram; skin LD-50 between 200 milligrams
and 2 grams per kilogram; LC-50 between 200
and 20,000 ppm), "irritating" or "flammable"
(flash point between 20 and 800P) must bear the
signal word lOW ARNHNG" or "CAUTION" 'to-
gether with a description of the material and the
hazard. The Federal Act describes test proce-
dures for determining acute toxicity and skin and
eye irritation.
A product which "generates pressW'e
through heat, decomposition or other means", or
which has been designated as a strong sensitizer or
a radioactive material also requires labeling. This
category is of little concern as far as petroleum
products are concerned except for components of
277
-------
4
products contained in pressurized containers, e.g.,
aerosols which contain propane or other petro-
leum fractions.
In addition, the law requires labels on
a number of specific products, "based on hUII1an
experience". Products containing carbon tetra-
chloride, diethylene glycol, ethylene glycol,
methyl alcohol, and turpentine are included. The
section of particular interest to the petroleum in-
dustry is the one which includes, under "products
requiring special labeling" , the following:
"Products containing five peT
cent OT mOTe of benzene and products
containing ten per cent or more of
toluene, xylene, or petToleum distillates
such as kerosine, mineral seal oil,
naph tl,a, gasoline, mineral spirits,
Stoddard solvent, and related petro-
leum distillates".
The reason for this is the aspiration hazard dis-
. cussed earlier. The labels on household products
covered in the "petroleum distillates" statement
must include "DANGER". "Harmful or fatal if
swallowed", "If swallowed, do not induce vomit-
ing", and "Call physician immediately", those
with a flash point of 800F or less must also bear
labels describing the fire hazard.
As a general rule, petroleum products
are not "extremely toxic"., "toxic". "corro-
sive" or "irritating" as these terms are defined in
the Federal Act.
POISON CONTROL CENTERS
There are in the United States and its
possessions 558 Poison Control Centers, whose
facilities make available to the medical profession
information on the prevention and treatment of
accidents involving poisonous and potentially
poisonous substances. This information is avail-
able to doctors, by telephone, on a round-the-
clock basis. The Centers are kept informed of
up-to-date information on various products, ob-
tained from manufacturers on a voluntary basis,
by the National Clearinghouse for Poison Control
Centers, which is a part of the Public Health Serv-
ice of the U.S. Department of Health, Education
and Welfare. These Centers are located in 48 of
the 50 states (all except Montana and Vermont)
and in the Canal Zone, District of Columbia,
Guam, Puerto Rico, and the Virgin Islands. Table
3 shows the location and telephone number of
twelve strategically located Centers.
Reports were received by the Clearing-
house from 395 of these Centers in 1967. The
number of cases of accidental ingestion reported
was 83,704, and almost 87% of these involved
children under 5. Medicines accounted for more
than half of the cases, with aspirin responsible for
21.5%. Petroleum products accounted for 4.7% in
the "all age group", and 4.6% among the children
under 5 years of age. Specific petroleum products
mentioned included gasoline, kerosine, lighter
fluid, solvents, thinners, and furniture polish.
Mortality data, which are more com-
plete than the figures quoted above since they are
derived from death certificates submitted to the
U.S. Public Health Service. showed 2,283 deaths
due to accidental poisoning in 1966. Aspirin and
petroleum products were involved in 7.7% and
1.8%, respectively. Children under 5 accounted
for only 15.1% of the deaths, but petroleum prod-
ucts were cited in 10.1% of the 15.1%.
The Appendix contains the informa-
tion on Shell's petroleum products which has
been sent to the various Centers by the Clearing-
house. It will be noted that the various products
have been classified into fifteen categories. This
was done for purposes of expediency, since it
would have been impractical for the National
Clearinghouse to issue a page for each individual
product. There is some variation in toxicity
within a given category, and the descriptions
278
-------
5
shown refer to the more toxic members. The sol-
vents, high aromatic, low flash, for example, are
shown as having an acute oral rating of "slightly
toxic" (Hodge-Sterner). This corresponds to a
range of LO-50 of 500 mg to 5 g per kg. However,
certain members of this category are known to
have LD-50 values greater than 5. Toluene, for
example, has an LD-50 of 7. The ratings shown in
the Appendix can therefore, in general, be reo
garded as conservative. More precise information
is available on many of the products listed in the
Appendix.
The Poison Control Centers have also
been supplied with data on the various Shell
branded specialty and anti-freeze products sold in
service stations, e.~, brake fluids, furniture
polish, SHELLZONE ,etc.
EMERGENCY
In the event of an emergency call for information involv-
ing a Shell product, it is recommended that the problem be referred
to a Poison Control Center, preferably through a call by the attend-
ing physician, since the information contained in the Appendix
contains medical terms which could be misinterpreted by a layman.
The physician should be given the exact name of the Shell product
and the telephone of a Poison Control Center (Table 3, page 11).
NOTE: Additional information may be obtained from:
Shell Oil Company
Products Application Department
Toxicology Section
50 West 50 Street
New York, New York 10020
Emergency information of a medical nature is available from:
C. H. Hine, MD.
San Francisco, California
Telephone: 415 UNderhill1-5494
279
-------
r-
6
TABLE 1
ACUTE TOXICTY SCALES
.,. or"'. . - . '. ,
'. . '. ,'- .~~- ~~., " ~" -' '-. ' ~. -
,~ " ,,r:., ',~; 1~~I.;.J.~Jj<~lltq) Ii"'..; "', ,,'.'
,~ ~.'i.-~~,/,<~ -~. --"~":.:' --.. rt~.:'-..~~ "t "t ,~~~; '.,1' .
-~"
~ L050
LD-50 LD-50 RATS. 4 HR..
RATS. ORAL RABBITS . S~ INHALATION PROBABLE LETHAL
(PER. KG (PER KG (PARTS PBR ORAL DOSE
RATING BODY WEIGHT) BODY WEIGHT) MILLION) FOR MAN
Exuemel, 1 mg or less 5 mg or less Less than 10 A taste; 1 grain
toxic
Highly 1 to 50 mg 5 to 43 mg 10 to 100 A teaspoonful;
toxic 4 ml.
Moderately 50 to 500 mg 44 to 340 mg 100 to 1000 An ounce; 30 g.
toxic
SliJbdy 500 mg to 5g 350 to 2810 mg 1000 to A pint; 250 g.
toxic 10,000
Practically 5 to 15 g. 2.82 to 10,000 to Aquan
Don-toxic 22.59 g. 100,000
Relatively 15 g. or more 22.6 g. or more Greater than More than a quart
harmle8s 100,000
"Hod" and St,m.r, A""rican Indultrial Hy,ifm, Association Quart.,ly, D.c_l"r 1949.
" . .::.~1"'" - . .' ...' "~
'"
'. (.,1 f';~...-J.)t"'-"!.'~~;,~Q.J;'E..I f.Vl)~. F:
:
.' . -f - 't; .
ORAL LD-50 PROBABLE LETHAL
(PER KG DOSE FOR
RA TlNG BODY WEIGHT) 150 LB. MAN
Super toxic Less than 5 mg A taste (less than
7 drops)
Exrremely 5 to 50 mg Between 7 drops and
toxic 1 teaspoonful
Very toxic 50 to 500 mg Between 1 teaspoonful
and 1 ounce
Moderatel y 500 mg to 5 g. Between 1 ounce
toric and 1 pint (or 1 pound)
Slightly 5 to 15 g. Between 1 pint
toric and 1 quart
Practically More than 15 g. More than 1 quart
non-toxic
"Clinical Toxicology 01 Commercial P,oducts. by Gleason,
Gosselin and Hodge, s#!cond editio1l, Williams and Wilkins,
Saltimor#!, 1963.
280
-------
NAItIB:
~ Jrl'm V"
:
SHELL OJL COMPANY PRODUCTS
Gasoline and related products
LIST OF SPECIFIC PRODUCl'S:
ABROSHELL@Turbine Fuel, JP-4
Aviation Gasoline
Gasoline
Marine Gasoline
Outboard Motor Fuel
Premium Gasoline
Shell Gasoline
Super Shell GaIoJine
Tractor Fuel
White Gasoline
DESCRIYnON:
Gasolines are complex mixtures of hydrocarbons, ranging from C4 (butane) through Cu. Most gasolincs
alao contain additives (anti'knock, anti-ox1dant, etc.), but the concentration is too low to contribute
significantly to the acute toxicity of the product.
TOXICITY :
The vapor pressure of gasoline is high enough to create a hazard if used in open containers in a poorly
ventilated area. .
Acute OrcJ1: Moderately toxic (Gleason, Gosselin, Hodge); slightly toxic (Hodge-Stemer)
Aspiration: Highly hazardous.
Inhalation: Slightly to moderately toxic.
Eye Irritation: Vapors: absent to moderately irritating, depending on concentration.
Liquid: moderately irritating.
Skin Irritation: Vapors: non-irritating. Liquid: moderately irritating.
SYMPTOMS AND FINDINGS:
Oral: Irritation of mucous membranes of throat, esophagus, and stomach. Stimulation fonowed by depres-
sion of the central nervous system. Cardiac irregularities of rhythm.
Aspiration: Severe lung irritation with coughing, gaging, dyspnea, substernal distress, and rapidly develop-
ing pulmonary edema. Later, signs of bronchopneumonia and pneumonitis. Acute onset of central
nervous system excitement fonowed by depression.
Inhalation: Irritation of upper respiratory tract. Central nenroull system stimulation fonowed by depression
of varying degrees ranging from dizziness. headache, and incoordination to anesthesia. coma, and
respiratory arrest. Cardiac arrhythmias are a da.rous complication.
TREATMENT:
Oral: Do not induce vomiting. Lavage carefully if apprecwble quantity was ingested. Guard agains~ aspira.
tion into lungs. Avoid use of epinephrine or rela~ed Dympathomimetic amines. Administer 2 to 4 0%. of
olive oil and 1 to 2 oz. of activated charcoal. Render supportive treatment if central nervous system
depression OCcurs Watch for possible bronchopneumonia or pulmonary edema.
Aspiration: Enforced bed rest. Admimster oxygen under slight positive pressure with antifoaming agent.
Keep air passages open. Administer broad spectrum antibiotics prophylactically, if in~~~~ See card
on Kerosine. . . .'
Inhal4t8cm: Maintain respiration. Perform cardiac resutlci~ation (e.g.. external cardiac massage) if cardiac
arrest occurs; defibrillation if ECG indicates ventricular fibrillation. Avoid administration of sym-
pathomimetic amines. Prevent self-injury if patient convulses or is disoriented.
Eyes: Wash with copious quantity of water.
sid,,: Remove gasoline by wiping, followed by washing with soap and water.
@ Registered trade_k.
2RJ.
13
.,
-------
.
,
18
NAJIl!:
I ~ ---
:
SHELL OIL COMPANY P~ODUCTS
Solvents, hWt aromatic, high flash (see list below;
MANUFAC1'UIUIR: SheH on Compeay, New York, New York
DESCRlYnON:
These products have aromatic contents pater than 33% and flash poinra of 80°F or greater. They are
simila.. to the corresponding low flash solftnts except that their vapor prelSure is somewhat lower. Diesel
(uel and furnace oi1 are included in this ca(egory beeaUN they often contain high concentrations of
aromatics.
TOXICITY:
Acute Oral:
Ii spiration:
Inhalation:
Eye Irrll4tkm:
Skin Irritation:
Moderately toxic (Gleason, Gosselin, Hodge); slightly toxic (Hodge-Sterner)
Hiahly hazardous.
Moderately toxic.
Slighdy to moderately irritating.
Moderately irritating.
SYMPTOMS AND FINDINGS:
Oral: GturomtutiMJ irritation: nausea, vomiting. and cramping. Central nervous system depression ranging
from mild headache to anesthesia, coma, and death. Pulmonary irritation secondary to exhalation of
solvent. Delayed effects may include anuria, dysuria, hematuria, and laboratory evidence of kidney
damage; hepatic tenderness; jaundice, liver enlargement, and other evidence ofliver damage.
Aspiration: Severe lung irritation with coughing, gaging, dyspnea, substernal distress, and rapidly develop-
ing pulmonary edema. Later, signs olbronchopneumonia and pneumonitis. Acute onset of central
nervous system excitement foHowed by depression.
Inhalation: Central nervous system depreuion of low grade characterized by headache and slight giddiness.
TREATMENT:
Oral: Do not induce vomiting. Lavage carefully (avoid aspiration) if appreciable quantity was ingested.
Adminiater 2 to 4 oz. of olive oU and 1 to 2 oz. of activated charcoal. Render supportive treatment
if central nervous system depression occurs. Watch (or delayed liver and kidney damage. Start
prophylactic liver and kidney protection program. .
Aspiration: Enforced bed rest. Administer oxygen under slight positive pressure with antifoaming agent.
Keep air passages open. Administer broad spectrum antibiotics prophylactically, i( indicated.
Inhalation: Analge.ics for headache. No specific treatment required.
Eyes: Wash with copious quantity of water.
Skin: Remove solvent by wiping, foHowed by washing with soap and water.
t
LIST OF SPECIFIC PRODUCTS:
Aromatic Solvent 29, 306
CYCLO-SOL@, 31, 32, 35 through, 38, 40
through 74, S-33, 6960
Diesel Fuel
DIESELEN~
ENSIs<8> Fluids (see lubricating oil section
for ENSIS oils)
Form Oil J-11
Furnace Oil
Heavy Aromatic Solvent
No.2 Fuel on
Mineral Spirits 80
NONA-SOL 85
Stove Oil
TS-28 (also 28M, 28R, 28W)
Xylene
@ Registered TrtJdemark.
,
292
-------
19
HAMB:
I r ra vr
SHELL OIL COMPANY PIWDUCTS
KerosiDe and related products (ICe list below)
MANUFAcnJRER: SheD Oil Company. New York. New York
DESCRJP110N :
This group includes kerosine and other fuels such as aviation turbine fuel. Products of sinoilar viacosity and
boiling range. such as certain intecticide bases and light lubricadna oil.. are also included.
TOXICITY :
Vapor pressures are sufficiently low so that vapor exposure is not normally a problem unless the liquid
product is at an elevated temperature. Some of the products contain additives. but these have little effect
upon toxicity in the concentrations used.
Acute Oral: Slightly toxic (Gleason, Gosselin, Hodge); practically non-toxic (Hodge-Sterner).
Aspiration: Highly hazardous.
InluJation: Due to low volatility. saturated vapors at room temperature are not toxic.
Slrin & Eye Irritation: Minimally irritatina.
SYMPTOMS AND FINDINGS:
Oral: Gastrointestinal tract irritation. Pulmonary tract irritation secondary to exhalation of vapors.
Aspiration: Severe lung irritation with coughing, gagging. dyspnea. substernal distress, and rapidly develop-
ing pulmonary edema. Later. signs of bronchopneumia and pneumonitis. Minimal central nervous
system depression.
Inhal4tion: Olfactory recognition, no significant nasal or respiratory tract irritation. No systemic effects at
room temperature.
TREATMENT:
Oral: Do not induce vomiting. Do not lavage. Administer 2 to 4 oz. of olive oil and 1 to 2 oz. of activated
charcoal.
Aspiration: Enforced bed rest. Administer oxygen under slight positive pressure with antifoaming agent.
Keep air passages open. Administer broad spectrum antibiotics prophylactically. ifindicated.
InluJation: No treatment required.
Eyes: Wash with copious quantity of water.
S"in: Remove product by wiping, fol/owed by washing with soap and water.
LIST OF SPECIFIC PRODUCTS:
AEROSHELL@. Turbine Fuel,
iPS. iPSA. }P6. }P7
Base on 3. 40. 185
Deodori~ed Spray Base
DlSPERSOL@
Egg Coating Oil
No.1 Fuel Oil
FUSUS@ Oil A
Heavy Solvent No.1. WTC
Honex Oil J -11
Hydraul Oil 11
Insecticide Base
Kerosine
Nylon Yarn Oil 11
,
Odorless Kerosine
Paraffin Base
PELLA@Oil911
Range Fuel
RoUex on HI, 211. 311.411
SheUdraw on 11
1300 Solvent
Special Fuel LFI-A
TEGULA@on 11
TELLUS@ Oil. 11
Textile Process Oil 1
Turbine Fuel 650. 650A. 640
UMf@Grade C
38S on
@ Registered trademark.
?B3
-------
ATTACHMENT III
CHARACTERISTICS OF CP-27 AND CP-34
284
-------
Monsanto
ORGANIC ~ DNIIIION
Monsanto Company
800 N. Lindbergh Boulevard
St. Louis. Missouri 63166
Phone: (314) 894-1000
December 15, 1970
Mr. Harvey W. Welsh
Thermo Mechanical Systems Co.
18345 Ventura Blvd.
Tarzana, California 91356
Dear Mr. Welsh:
Enclosed are data bulletins and information on toxicity
and safe handling of two Monsanto candidate thermo-
dynamic fluids:
CP-27 - monochlorobenzene thermodynamic grade
CP-34 - thiophene thermodynamic grade
This should provide the information requested in you~
letter of December 1, but if you find you need any ,
further information on Monsanto fluids, please feel
free to contact me.
( Verry t~-~ours,
" "1./ I
........ .. -' ,
- '- ~::: tc".. ~;~/»
Mchard Davis
Manager
Commercial Development
RD:ms
Bncs.
2eS
-------
.1:",.
f
-------
~'---.>
"q9b "
f - .
. i
.":'--_J
: -----,
~_Yi
.' ..- 1
: Q;E- t-
. -
~-=-~~~. 1
..-.% J.
. ,
: -!
. ;
--- - --- - J
i "-----~ I.
!'~!.
I~/:
~:...~.=j
'u""
~;;,;'.'.I~~
b \~/ L.'C"P)
6A£Il£L a.e.JJ.EMTAL~Y /'J. T HIi,Tlc:. ~ 1..1/ 1./2..$
CP-27
Chern1s try:
LUbricity:
Monochlorobenzene (treated)
No data
Thermal
Stability: MonaantQ th~rmodynam1c ~_de
monocITlorobenzene has a predicted
lifetime of several thousand ~ot
hours when liquid temperatures
n~ver exceed 333°C anQ moisture
and oxygen are excluded. This 18
based on Monsanto static tests'
. conducted at temperatures to 343°C
and times to 2000 hours. Commercial
monochlorobenzene 18 far less stable.
Toxicity:
The toxicology of monochlorobenzene
jca- reeMly a¥ai1a8-18. in s~d
texts and cQmpi~tions.
MONSANTO COMPANY
9/28/70'
MONSANTO COMPANY. 800 N9JfTH UNOSERGH BLVD. . ST. lOUIS, MO. 6&'166
287
:c. ':~_--:JJ'~W riN'~ PtJ ,~ Q:>.::rl~,\'{ftW) ,;311 ~'::1 rij,[t":5tX ;;i-,'t:j([f;i', ;..)ti ;;l'~ ~}I1,ruw.rr)t ~;a~ t1i;1O:~.ca.~I1C;;.: c;-r3J:'.
'j1I~i3~, LC':, %J~i; [fjt.[:,j\'El ,:'I~(,<;'; tJ'J~~~g ;?-:::::W-~I-;l$ !::J;,1~ Gr:::'~~l':~t1J;J-W D'~:f_'ri!r7i]ijf.:tmo;~ I~,J I;tt~~v;
-------
r'-- ~- . /1
: <..75fJ.-";
" I
: ~.- !
.J' ..'
. ~ -- t-. ~ '
;: ~
I '
I
I
I
I
!
0'#>$ I
I
I
i
j
I
I ,
.
.,
c'p' 34'
THER MOO YNA'M'IC FLU I D C.AND I DA TE (~[ol'){£#E ')
j
.'1
DEVEL OPMEN TAr. SYNTHETIC FLUIDS
II
'j
it
I
f
CP 34 is a stable organic fluid which has been,found
to p08~e~~, ur;11q~e thermQdrnU,"'.~-p!,operties potentially
useful in a s1mple Rank1ne cycle machine. The c$m-..
pressed vapor of CP 34 is capable of e~pa~~~o~~~*hdut
becoming excessivel;y satur~ted Qr supernea:;t.e.d..,. .,:t1.~~.:
a hypothetic, non-regenerative cycle, CP 34 gave'a'~pl'At~
dieted ettic iency 4~ grea,tftr tttan a, c'~~q~y .~upe~'~':1 ..
Rank1ne cycle fluid, Freon 12. While the ~xlmum
perm1ssible system temperature with CP 34 is a8 yet
undeterm1ned, it appears to be otthe order of2000p
greater than that of the Freon fluIds. Pertinent
1ntol'mat1on concerning CP 34,-1s tabulated below.
. erties
Temp. Viscosity Densitl Spec ific Heat'
JOF) -' C8) ig/ml. _(BTU/lb,")
-40 melting pt.
0 0.9,0 1.106 0.302
60 0.67 1.069 0.326
,100 0.53 1.044 0.342
150 0.42 1.U13 0.360
200 to33! ext 0..981 0.384
300 0.24 ext 0.920 0 . 426
400 Q.,19 ext o~859 0.460
Vapor PreQsure (mm) Temperature (OF)
160 183
100 86
10 10
1 -40
MONSANTO COMPANY. 800 NORTH LINDBERGH BLVD, . ST. LOUIS, MO. 63166
288
TO ASSIST YOU IN THE EYAlUATION OF THE ABOVE PROOUCT, WE HAVE OBTAINEO THE INFORMATION COlI,
TAINEO IN THIS BUUETIII FROM LITERATURE SOURClS AND I'R£LIMIIWIY INVESTIGATION IN OUR OWN
LABORATORIES, WHILE WE BELlM THE INFOfIIlATlON IS ACCURATE. WE' ASSUME NO RESPONSIBILITY AND
DISC&AIM AJfY LIABILITY INCURRED IN USING THESE DATA OR SUGBi;STIOIIS, ONLY QUALIFlEO TECHNiCAl
PI1IS01IIIn SIIOUI.D - WI11I TNIS PRODUCT UHnL SU8SEQUEIIT STUDIES JUSTin ITS USE IN. 'lOUR PROI).
UCTS OR PItOCD$ES, OBYIOUSLY NOTHIIIQ IN THIS BULI£IIN SHOUlD Sf: COHSTIIUlD AS A RFCOMIffil.
::~~;.; r:;JTnC .:.'-:'-',' k~J.;:)I;:'::C:' fi:T:;r:T:, '~,,:': I":".,J:I':'~ :..::~ 1'r3 'J:~i
-------
CP 34 - page 2
R1edel reduced vapor pressure equat10n -
In P = 6.031 In T - (0.1911)
(~ - 35 - T6 + 42 ln T)
Surface Tens10n
33.89 dynes/em at 69°P
32.37 dynes/em at 86°F
30.89 dynes/em at 104°p
Refract1ve Index
n20 =
D
1.5298
~5 :
1.5251
Flash Po1nt tQQ.Q) ~ ;1.0 p
Autogenous Ign1t1on Temperature 755°F
Thermodynam1c propert1es
Critical Constants
-
-
584.6°F
Vc
Zc
= 0.04341 ft3/1b.
= 0.258
Tc
Pc
= 791.73 psia
Estimated Thermal Conductivity and Viscosity
Thermal Conductivity Viscosity
Tem). ( BTU /h r . ft . ) ( 1b/hr .ft.)
of Liquid Vapor Liquid Vapor
200 0.0795 0.0188 0.7320 0.0594
300 0.0242 0.0678
400 0.0771 0.0299 0.2332 0.0758
500 0.0357 0.0836
600 0.1019 0.0417 0.1259 0.0910
(Calculated from Lennard-Jones parameters for 14.7 psia)
~89
-------
CP 34 - page 3
,
I
i .
Pluid Stabilit~
Based upon a series of bulk thermal stability tests, the
life predictions for CP 34 are as follows in terms of time
to reach a given level of conversion to decomposition products.
Temperature
600°F
550°F
5000p
450°F
~
~ .~
17
62
240
920
170
910
1,200
180,000
These are conservative predictions and are based on the fluid
remaining at that temperature constantlYi a situation more
severe than encountered in the actual practice.
Compatibilit~
Metal Corrosion Evaluation
(Test in confined vessel for 144 hours)
Metal
Corrosion Rate (calc.mils/yr)
Liquid' Vapor
@ 300°F
Aluminum 61S
Magnesium
Copper (oxygen
Mild Steel
free)
0.21
1.39
0.17
0.09
< 0 . 07
0.30
0.18
0.07
@ 500°F
Titanium
Incone1
0.09
<.0.08
0.04
0.15
@ 650°F
Stainless 316
(0.10
290
-------
\Ie oJ"" - ..,~t!lt= ....
. Elastomers
(FTS 791A Method 3603.4 - 168 hrs. at 158°F)
Hardness Swell (Vol. Percent)
Initial Final
1. Vlton 65 51 53 23.0 8.97
65 49 53 24.5 10.4
2. Buna N 60 46 56 140.1 1.99
60 45 57 141.1 2.12
3. Neoprene 61 44 67 127.0 13.9
61 45 68 127.0 . 13.4
4. Butyl 61 38 49 109.0 1.03
61 39 50 111.6 0.11
5. Silicone 65 60 63 42.3 3.21
65 63 64 42.9 3.22
6. EPR 66 52 72 74.0 8.22
66 52 73 70.3 8.31
(Numbers in parentheses indicate data obtained after
samples were air-dried for 24 hours at room temp.)
Fluids
Completely Soluble
Limited Solubility
Freon 11
Freon 12
Freon 114
Freon 22
Insoluble
Acetone
Benzene
Chlorinated ali-
phatics
Heptane
Hydrocarbon oils
Water
Lubricity Tests
CP 34
Water
Falex Test
Failure load (lbS.)
none up
to 4400
lbs.
500 Ibs.
291
-------
CP 34 - page 5
Shell Four-Ball Wear
Scar Diameter
1conditiona: 10 kg load, 1200 rpm, one hr.)
Steel-on-Steel
Steel-on-Bronze
CP 34
Water
0.48 mm~
1.20 mm.
1.13 mm.
1.53 mm.
Mollier Diagram
The Mollier Diagram on the next page has been drawn using
entropy and enthalpy data derived from the critical con-
stants and measured pressure-volume-temperature data.
The most outstanding characteristic of the curve is the
almost vertical saturated vapor portion.
Precautions
Although CP 34 is a flammable material, its use is not
considered hazardous if appropriate handling procedures
are observed. The liquid has a flash point of 60°F and
its vapors should not be exposed to open flame, spark
or other ignition sources. In the absence of an ignition
source, proper air-vapor mixtures would not be expected
to undergo autoignition below 750°F.
On the basis of preliminary evaluation, CP 34 is tentatively
classed as being only slightly toxic when ingested orally
or absorbed through the skin. It is considered to be a
moderate skin and eye irritant. Gross effect studies
indicate that the vapors of CP 34 may represent a severe
hazard when inhaled. Conse'quently , it should be used
only in well ventilated areas and care should be .
exercised to avoid breathing the vapors.
August, 1968
292
-------
90
I ! ~ ~ t:? :/ /
--1--- ~
----_. ~. ~ ~/
~o Vfc /
~ ~~ ~nn 1
-- ,--..A % ~
.- -- -/.~ 'r~" ... -r-- 7
L V ..L
./ ~P'''''
r ~V'.: ~ ,...- I
./ V / [,I'"
/ )~~ V ~
~O/
/ ~v ~ V / 400 ~
584 .6 Tc ~ J / ~ --
-- y
1~J. .1 PC 'rt ~V V
/ /
Z ..--... ./ '11:1\ ~
Sat Llrate :j V --- ;; ./
Liq\J lid Li rie / ~~ ;/- V /
L..-- P-/ u.
C / I '/ 300 V
7 ./ /
/'
V , /' / / £r
/ V .....q....,.
/ --/ .- .-._-. - - ~ /'
V
1/ / /
10' ./ ~vv ./
/' V /' ~
./ /
Sa ,,"urat ~d - V /' / SO.A' v
~ ....
Va por L ne ~ /, ,, .<'./ V V
- ./ ~
\ /' / /' V -;/
./ ".~ 00 F
/ ..,-" ,/" /'
" /" /" V '"
/' '\.~
:/
'< Oy . --..- ---
-- ~ ,,;
:/
~
..._.--.._~ 1-. - '. .----- .'-- r----
.--.--. - --....--.
MO LLIE R D I AGR AMI OR ~ P ~ 4
~ J
--_.-.. --.--
..~- -....----.- _..- . r'---' .. .- -. r- ----. --
550 F
140
130
120
110
100
80
70
60
50
- 40
~
"
~ 30
-
~ 20
~
ftJ
.t:
~ 10
r::
~
0
-10
-20
-30
-40
-50
-60
-.05 -.04 -.03 -.02 -.01
o
.01 .02 .03 .04
293
Fnt.ropl' (J\TU1lb/oR)
.05
.06
.07
.08
.09
-------
'1
.-- - ::::-...:-::..]:
"'.,. ,
(:'( f/') ! .
," ~ ",.~', ,
"~i I
. ,
" ---- -- ..-' .
------.........,
1.-1._:
L '- f /' :
,"" :
-': I
. '. ;
.
--..... ...-.~~
"---""-1
lrr:/: i
:,:~::.~-~ :
. '
. . . , .
. . .
'-. . .
. DE\lE.J~ ()Pk~~El;rTJ.! f..' oS VI1TJ/ E TIC fl.U IDS'
I
--. .. -....,
;i-""" ? '
)':~ i., '/'
'\4'" ~
~~~~,:~:. J
;2~. .J
-',"\/'.]
':}ji,) i~;:;~1ni:)
,...,". .~,j
CP-34
CJ1emistry:
Lubricity:
rrpermal
'roxici ty:
Flash Point:
Thiophene (treated)
See CP-34 bulletin.
data ava1lat)le..
No additional
. .
Stability: Monsanto,thermodynamic grad~
. thi6phene ha~ a predicted lifetime'
of several thousand hot riours
vlhen l1qgid temperatures never ex-
. cecd 288 C and moisture and.
oxygen arc excluded. This 1s
based on Nonsahto static tests con-
',ducted at temperatures to: 399°C and
times to l500hours., Commercial
" thiophene is far le~s'~tablc.
. 'See attachments.
Important. The flash point ,lj.sted
in the August 1968 Cr'-34 Bulletin,
(60°1<' by Cleveland Open Cup) is in'
error. It should be revised to'read
20° F by Tag Open Cup. '
".-
, . ,
. .
, ,
MONSANTO COMPANY. ,800 NORTH LlNDl3EHGH !'.;"V':/ '. Sl':'LOUIS. MO. 63165' ,
294'." .' .
'.; j ~ ,; rii ,-::
. '
. .' .
10 #..~SIS" Y'OJ IN tHE rv~U"T10N Of HII: A80\'[ J'RO£'l',!(;T, WE 'P,WE OBJ.\iN~C THE I~F01HAA!iQ,~ Co.~" .
'.:;"'L;J~:.r~;fi:::i~,:3 ,:,:, 'I.'':-::U.',' !.'~ I ;'f I ,'i', ::~,il';" ,'.:~
.!: ~.'>
j~."..::.~-,;,!:,'fJI':"~1 1-'; 1-'--_::,:~~~..,7~,
-------
FOR PERSONAL nLE 0.,
. ~~~~~~
REVIEW
API TOXICOLOGICAL
L TH 101'11 E:'-:E AND DEIU\'ATI\'P.R
SEPTE~I8ER 1948
Note: This review summarizes the best available informa-
tion . on the properties, characteristics, and toxicology of
thiophene and derivatives. It offers suggestions nnd tentative
recommendations pertaining to medical treatments, medical
examinations, 2nd precautionary measures for workers who
arc exposed to thiophene and derivatives. It was prepared at
the Harvard School of Public Health, Boston, Mass" under
the direction of Professor Philip Drinker. The review h3s
been accepted for publication by the Medical AdvisOI"Y Com-
mittee of the American Pets"oleum Institute. Anyone desir-
ing to submit additional information or proposed ch:mges £0'"
consideration prior to re-issu:mcc of this review is requested
to send them to the American Petroleum Institute.
This review was prepared by Marshall Clinton, :\1. D,
_,\\IERICi\~ PETHOLEI)\I JNSTITI :TE
J)EI'\HT:\IF:--;'T OF S\ITTY
50 \VEST 50TH S'JI!n:T
:":E\\' YOHK 2(}, ~. Y.
295
I'rit't. ?3 /', .
-------
1
.-- --.--.. ---
'--- ----.-v
,\I'J '11.\j( I II (),.I, ,\1. REVIEWS
."......---.-
- - -.------
I Hi"!'/lr:--:r\:,.!. ")LlU\'...r/\ ES
.-. -.#' -' .. -,- . '. ... .
.- . _.-
TO~IC()LOGIC:,\ L H F\I F\\' OF 'J IIIOPHENE AND DEI( I\'. \'1'1 VES '
I. Suhsl.lllcc
Thiul,ht"H' ,
F,.rmlll.l: C)'IIS, '
Sr ru< III rorl (01 mlll:1 : '
II( --ell
"
01
}.if .
..
I.
{ 11
'. /
, /
s
J\lvkt:ulltnd encountered as an important contaminant of
!)(,l'izene. It is insoluble in water, bu.t is readily soluble
in ~tlcohol, ether, benzene, and most hydrocarhons.
Thiophene is difficult to separate fcon1 benzene by
physical means because of their similar boiling points,
hut can be separated fairly readily after reaction of
the more reactive thiophene with other substances,
S'lth as mercury. Thiophene ran be obtained from
crude benzene, and now can be synthesized without
I,;xcessi\,{: Jifficu!ty. Thiophen(; is highly reactive, and
is readily nitrated, sulfonated, halogenated, or mer-
cmllted. It can he made to undergo ketone forma-
'(ion or aminomethylation without difficulty. It is
usually removed from, benzene by sulfuric-acid
trC';ltmcnt.
H r. Pr"hoblc Sources of Contact
Contact with thiophene may occur .due,. to leaks
occurril1.~ iJl the (Ollrse of its handling or manu-
facture, Contact may n;o;ult from handling crude
coal-tar ~nz~n.,) :1S this (f1f":liI1S lip tu 0.5 per cent
thiophene. It is !lot j'ossi!,!c to state tll<.' most proh.
Prtp,,.. under the auspices .,l the Subce.mmiUft for ~r-
i - ;"'\s.b~ CenO!llhaTi~U of Toric 51tb$t4l1cws i" 1he PeTroleulll
Tndldh7.
F'81{f1!S _r to bihliG,rapll)' on p 3,
ahle sources of exposure, as this SUbSl.'lll( is, L1)111.
mercially, relatively new, :H1d its uses arc ill-ddi lI~d
hut growing. '
IV. Toxicology
II General Considerations
A considerable amount of study has been de\'uteJ
to investigations of the toxicology and pharmacolob'J
of thiophene and its derivatives. Most of the latter
studies have been comparisons of the thio. homo-
logues of organic substances of known and, usually,
fairly marked pharmacological activity. Thiophene
in fairly high concentrations has, according to most
authors,4. :; an acute narcotic effect greater than equal
, concentrations of benzene. Flury, and Zernik report'
that the inhalation by mice of 2,900 ppm of thiophene
results in loss of consciousness and, in some instances,
death' whereas s'imilar concentrations of benzene
,
can be tolerated without difficulty. Concentrations
, of 8,700 ppm of thiophene caused death of mice in
20 min to SO min; whereas benzene produced no
such effect.
The acute toxic action of thiophene appears to be
exerted primarily on the central nervous system. It
has a selective action on the equiliiJrium centers of
the cerebrum and cerebellum, producing severe ataxia
fo119wing repeated injections.:;' r, Thiophene pro-
duces fairly diffuse changes in the cerebellum, p:1r-
ticularly the vermis, characterized by degeneration of
nerve cells in these areas. There mar be superim-
posed vascular changes.i. 8 The metaholism of thio.
phene is poorly undershx)d, although it is stated that
5 to 12 per cent is recovered in the urine in conjugated
form. However, there is no increase in the con-
jugated sulfates, and. the total sulfate ex(retion de-
creases aftet administration of thiophene.';
t.. Acute Effects.
As already noted, aClite exposure to high concl!'n-
'trations of thi')phenc results in nervo~ls,syst~m de.
pression. Repe.ltcd daily injection of 2 g of thiophene
in dngs rco;ulh in locomotor at:lxia and paralysis,:'
ralher sirnil:tr It) those 11l1tcd in sc\'crc (52 poisoning.
The dfcds of thiP!'lI(,fH.' \)\1 11\\11\:11\"; has "ot been
described.
296
-------
API rOXICOLOGICAL Rl:V\I:\\'S
TH'IOI'Hf.NE AND D.EHIVATJ\'L\
]
c. rhronic Effcch
NIl reports nil the chronic effccts of repe:1teJ or
pcolnn}.;cd exposure to low u>nccntratiuns of thio-
phenc arc avarlable.
d. Sole l.imits
b.crcmc conccntrations of thiophene are ohviously
jntulcrable, as they produce acute poisolling. No
in£.mnarion is available, howc\'er, on the e([ects of
repe.leed exposure to {esser concentrations, such 'as
100 to ] ,000 ppm. Therefore, 110 safe limits have
been or can be promulgateJ at present.
V. Treatment
No information 011 the possible therapy of thio-
phene poisoning is available.
VI. Examination~
The present state of knowleuge cOIKerning thio-
phe:ne does not permit the establishment of any
special pre-employment or periuJjc examinations.
It appears sensible to c::mploy only mcn in good health
to worJ: with thiophene and to re-examine them fre-
quently for possible c\'idenccs of blood dyscrasia or
neurologic uist urhances, but these measures may
penw: 'unnecessary.
VII. Precautionar)' McasurC5
Thiophene shoulJ be hanJled wit h extreme cate
in closed sysrems or with adequate \'cntil:Jtiun, until
its chronic toxicity is ~stabtishcJ or ->howl1 to h~
absent.
V III. Bibliography
I. C D. Hodgman .l1Jd II. N. Holm(,. If.Jllt/bonk fit
Chem;str)' alJd PbpifS, :~ 5th edn., Chemical Rubber
Publishing Co., Clcvel.1nd (1941).
2. P. Karrer, Organic" ClmlliJ/ry.. Nord(;mann Publishing
Corp., New York, 350, 700 (193ft).
3. Anon., Tbiophme Cht!/l/ic.1/s, Soconr-Y.\cuum Oil Co..
Inc., Research' and Development bboratories, New
York (1946).
4. F. Flury and F. Zernik, "Toxicity of Thiophene," Chem.
Zlg. 56, 149 (1932).
5. A. Christomanos, "Experimental ProJuction of Cere-
beHar Symptoms by Thiophenc," KIi'l. JI"o-chJ{b,.. 9,
2354 (1930),
6. A. Christom:lOos, "Action of Orb.wic Sulfur Com-
pounds on the Dog Organism: Action and Fate of Thio-
phene in the Metabolism of the Dog," Bin../'; WI. 7..
229, 24R (1930).
7. T. Upners. "Expcrim(;nt.1I Studies C)lkc:rning the Loo[
Action of Thiophene on the Cc:ntr.ll Ner\'ous System,"
Z. gn. N('IIrol. PJyrl,;;tI. 166, 6:n (! <)39).
R. A. Christom:lnos .II1J \X'. Scholz. ,. Electricity of To:-.i,'
Substan<:cs for the CClltr.tI Ncn'ous Systcm: Clinil.tI
and Pathological Studies of Thi'phcllc." Z. !:eJ. t\'"",.. {.
Psych;dt. 144,1 (1933).
297
-------
ATI'ACHMENT IV
CHARACTERISTICS OF FLUORINOL 85
298
-------
~
aI.ocax-bo:K1
PRODUCTS CORPORATION
92 8UALEWB COURT
. HACKENSACK, N. oJ. 07BOI
. TItL£F'HON£ 201 - 343-B703 . TELEX: 1334651
April 28, 1971
Thermo Mechanical Systems Company
7252 Remmet Avenue
Canoga Park, California 91303
APP '3 0 1911
Attention: Mr. Harvey W. Welsh
Gentlemen:
Enclosed are data sheets on Fluorinol 85 as requested in your letter of April 22.
This material is basically a fluorinated alcohol to which about 3% by \"'eight, of
water has been added.
We offer other Fluorinol fluids containing greater percentages of water. These'
fluids are designated by number representing approximately the mol % of the
fluorinated alcohol contained in each fluid. As an example, Fluorinol 85 contains
85 mol % of the fluorinated alcohol.
We will be glad to send samples and data on any of the Fluorinols you may want to
investigate. We have data on the following at this time:
Bolling Point
Freezing Point
Fluorinol 100
Fluorinol 85
Fluorinol 61
Fluorinol 51
Fluorinol 33
Fluorinol 21
164°F
166°F
170°F
172°F
176°F
178°F
-48.9°F
-82 of
-26 OF
- 9 OF
9 of
18 of
All Fluorinols containing more water than Fluorinol 85 have a melting point (ice point)
of -82°F due to selective freezing out of ice particles between the observed freezing
point and the eutectic formed at -82°F. In that temperature range a slurry is formed.
299
-------
-2-
Thermo Mechanical Systems Company
April 28, 1971
All Fluorinols also have an observable flash point but a very low heat of
combustion (less than 3500 BTU/LB). As a result, these materials will
not sustain combustion. We have not found an auto-ignition temperature
in any of these fluids, but we feel that it might be well to repeat these tests
if you are equipped to do so.
Toxicity data on trtfluoroethanol (Fluorinol 100) is attached hereto. As water
is added, it is assumed that inhalation and dermal toxicity is lowered.
Please let us mow if we can be of further assistance.
Very truly yours,
HAL~A~BO PRODUCTS CORPORATION
CC~-- ----
C. Conner
RGC:js
encl.
300
-------
I - - . - . - -. -- - - --
UNIVERSITY OF
MAR YLAND
SCHOOL OF PHARMAC1f
636 Wed LDm"ord Strut, Baltimore, Mo"Ls.d fJ_J
i)
DI!PART"£~T 0" PHARMACOLOGY AND TOXICOLOGY
EVALUATION OF TRIFLUOROETHANOL (TFE)
TOXICI'lY AND HAZARD
Introduction
TFE is toxic after oral ingestion and is an eye irritant.
Al though its
dermal and inhalation toxicities are significantly less, inhalation and skin
contact should be minimized.
If accidently ingested. it should be removed
from the stomach by careful lavage with water.
(Emesis should be avoided be-
cause of possible pulmonary aspiration).
There may be a latent perio~ prior
to the manifestation of symptoms during which time ethanol may be administered
to prevent conversion to toxic metabolites.
The symptoms and treatment should
parallel those of other metabolic uncoupling agents.
No human exposures to TFE which resulted in fatalities have been reported,
however, the lethal action of TFE has been extensively studied in animals.
Table 1 shows ,the LD50 data for various routes of administration.
TABLE 1
LETHAL DOSE LEVELS OF TRIFLUOROETHANOL
ROUTE SPECIES LDSO REFERENCE
-
Oral Rat 240 mg/kg Hazleton (1965)
Mouse 366 mg/kg Blake, et a1. (1967)
Intraperitoneal Mouse 350 mg/kg Blake, et al. (1965)
Mouse 195 mg/kg Airaksinen, et,al. (1970)
De nna 1 Rabbit 1680 mg/kg Hazleton (1965)
Inhalation Mouse 1.6 ml% 85.1 ppm
/10 min. Blake. et a1. (1967)
Rat 762 ppm/6 hr. Hazleton (1965)
In dogs, an intravenous dose of TFE of 400 mg/kg was lethal within 24 hours.
(Blake, et a1. 1967).
According to sta~dards established by the Manufacturing
30l
-------
2
Chemists Association (1970) the acute oral LD50 falls in the category of a toxic
substance.
Therefore if ingested) efforts should be made to remove it from the
stomach.
A series of subhypnotic doses of ethanol is antidotal for trifluoro-
ethanol. toxicity in animals (Blake et al.) 1967) and may be in humans.
Trifluoroethanol is not classified as ~oxic via the dermal or inhalation
pathways (MCA, 1970).Nor is it a primary skin irritant) but contact with the skin
should be avoided to prevent dermal absorption.
Spills are readily washed off
with water.
Inhalation of the vapors should be avoided by working in a well
ventilated area.
Animal studies have shown that trifluoroethanol causes severe eye damage
similar to many chemicals (such as isopropyl alcohol and toluene).
Immediate
washing of the eye with water should be used following any splashes.
Inhalation Toxicity
Based on the inhalation toxicity of TFE in mice and rats it appears that a
maximum atmospheric concentration in human environments should be suggested.
According to the generalJy recognized concept of TLV (Threshold Limit Value for
8 hour exposures during a S day week-40 hours) it is suggested that conc~ntrations
in excess of 50-100 ppm be avoided.
This level allows a safety factor of 10 in
translating results of animal studies to possible human exposures.
For the
purpose of comparison the TI.V values for some other common volatile liquids are
listed below:
!LV (ppm)
Ethanol
1000
Trichloroethylene
100
eel
4
10
Benzene
2S
Eye Irritation
When 0.1 ml of TFE was applied to the eye of rabbits it produced severe
irritation) corneal opacity, iritis, marked conjunctivitis and corneal damage.
30?,
-------
.;)
The effects were only slightly reduced by washing with water immediately after
the application (Hazle ton, 1965).
Four human eye splashes, followed by immediate
wash caused no discernable damage (Halocarbon Labs., personal communication, 1971).
, Skin Exposu re
There is no information available on the percutaneous absorption of TFE in
humans but considering its animal toxicity and other properties, the development
of systemic toxicity after external exposure is quite likely with sufficient
amounts.
TFE was found to produce death after high dose dermal application'
(24 hour) to abraded skin of rabbits and caused mild erythema on intact and abraded
skin (Hazleton, 1965).
Rapid decontamination should be advised in the event of
skin exposure.
Symptoms of Acute Poisoning
TFE produced rapid depression of the central nervous system at high doses
however its lethal action can be produced at sub-depressant dose levels.
There
is a latent period of 5 to 12 hours before overt symptoms appear.
This delay in
onset of signs increases the hazard associated with TFE usuage.
Animals' become
markedly depressed and weak with tremors and rapid labored breathi~g.
Salivation,
vomiting, lacrimation and fever also occur.
Bloody diarrhea has been observed.
Mechanism of 'Poisoning
Histopathologic findings at autopsy have not been conclusive.
The evidence
of liver toxicity (Rosenberg and Wahlstome, 1970) and pulmonary congestion (Hazle ton,
1965; Blake, et al., 1967) that has been found does not appear to explain the ob-
served symptoms.
The symptoms are similar to those seen with agents uncoupling oxidative phos-
phory1ation such as dinitrophenol (Blake, et al., 1967).
Moreover, TFE has been
found to lower the ATP/ADP ratio in mouse liver (Airaksinen et al., 1970).
The
toxicity does not appear to be the result of conversion to monofluoroacetate as
the trifluoromethy1 group is refractory to dehalogenation and sodium acetate
::so 3
-------
4
affords no protection (Blake, et al., 1967).
There is extensive evidence that the toxicity of TFE is at least partially
the result of ita metabolites.
Inhibitors of the oxidation of alcohols such as:
disulfiram (Antabuse), 3-amino-l,2,4-triazole, allopurinal (Zyloprim), ethanol,
4-iodo-pyrazole and isoniazid have all been shown to decrease the toxicity of
TFE.
Allopurinol, an inhibitor of xanthine oxidase, has also been found to syn-
ergize the lethal action Qf TFE.
A possible explanation for these findings is that
trifluoroacetaldehyde, (which is also toxic-Airaksinen et al., 1970) is formed
from TFE and is responsible for the toxicity.
Treabnents and Antidotes
Lacking an actual poisoning experience in humans, there is no established antidote
for TFE poisoning however on the basis of animal studies i~ would appeat that
the administration of inhibitors of TFE oxidation may be beneficial.
Considering
human safety and experimental efficacy, ethanol in sub-depressant doses appears to
be the best choice.
The antidoting should be continued until TFE and its metab-
olites are excreted which may take several days.
Because of the evidence that
TFE is an uncoupling agent 1t 1s also recommended that the usual trpatment for
Dinitrophenol poisoning be considered.
In general, this consists of measures to
reduce body temperature, assist respiration and maintain acid-base balance.
Since
human experience with TFE poisoning is lacking, t~eatment should be primarily
symptomatic and supportive and victims should be closely observed for several days.
D. A. Blake, Ph.D..and D. R. Brown, D. Sc.
April, 1971
304
-------
REFERENCES
Airaksinen, M.M., Rosenberg, R.H., and Tammisto, T. (1970) A Possible Mechanism
of Toxicity of Trifluoroethanol and Other Halothane Metabolites. Acta
Pharmacol. et Tox1col. 28: 299-304.
Blake, D.A., Rozman, R.S., Cascorbi, H.F., and Krantz, J.C., Jr. (1967)
tran8formation of Fluroxene-I. Metabolism in Mice and Dogs In Vivo.
Biochem. Pharmacol. ~: 1237-1248.
Bio-
Manufacturing Chemists Association (1970) Guide to Precautionary Labeling of
Hazardous Chemicals, 7th Ed. Manual L-l, Washington, D.C.
Hazleton Laboratories (1965) Acute Dermal Applications - Rabbits Trifluoroethanol.
Final Report to Halocarbon "Products Corporation, Hackensack, New Jersey.
Hazleton Laboratories (1965) Acute Oral Administration - Rats Acute Inhalation.
LC 50-Rats Trifluoroethanol. Ibid.
305
-------
, .
.,:,,<,b!~.'-~E!!~~i~,p.2n. ".RrqfJ ~.£t~4,..~,r4.?9f ~J~9!' .
, ',.' .~~'-\.~., 1."..~,<~", ;":'.,j>-' "':_~""''''''' .y ....,~~,-, .; .."' ...... . "..' .:fj. ~.':' '~~ ~" '!Ii. t.'. \ ",' ,'...11\. .- ,', " ".
".,\.'~ ."''':'._z_..,'_..''~.'''--~~:5.f~'' .''''',~''i''<.,.. '" ~J:.)......"., ."~- -';....~\":":-'~{'-"~f.'~'" ." .!':~::,.'-'"
82.' Burlews Court, J-Iackensack, N. J. 07601
t.. -
I._~-
r'f '-.
',~ .
, ,
! .
FLUORINOL 85
A Fluorocarbon Rankine Cycle Working Fluid
Molecular Weight
87.74
Boiiing Point at 14.7 psia
166°F
Freezing Point
-82°F
Ice Point (Melting Pt:)
-82°F
Flash Point
109°F
Fire Point
None
Critical Pressure
930 psia
Critical Temperature
465.5°F
Toxicity (MCA Standards)
Inhalation- non-toxic
Dermal - non-toxic
Oral - toxic
Flammability - wi~1 not sustain combustion
Co r r 0 s ion - non - cor r 0 s i v e wit h m 0 s t met a I s
Probable upper temperature limit 600°F
July 1970
306
1-
.. t
;' . ;.;.:' ~,
.t," <'!I'
'~ ,.
-------
ATTACHMENT V
CHARACTERISTICS OF P-1D
307
-------
:II
Specialty Chemicals Division
ALLIED CHEMICAL CORPORATION
P.O. Box 1087R, Morristown, New Jersey 07960
(201) 538-8000
August 11, 1971
Mr. Harvey W. Welsh
Thermo Mechanical Sys tems Co.
7252 Remmet Avenue
Canoga Park, California 91303
Dear Mr. Welsh:
In response to your request of June 9, I am en~losing
various technical data on our P-I-D fluid.
As you will see, the toxicology work has not been extensive
and. if your application would involve unusual release, this
area would need additional work.
We apologize for not replying sooner and hope that this
information is sufficient. If not please advise us of Y9ur
precise needs.
NCF/ndr
Atts.
308
-------
PHYSICAL AND THERMODYNAMIC
PROPERTIES OF
FLUOROCARBON P-lD
Allied Chemical Corporation
Specialty Chemicals Division
Morristown, New Jersey
309
-------
SELECTED PHYSICAL PROPERTIES OF P-lD
Molecular Weight
Chemical Formula
Normal Boiling Point
Freezing Point
Critical Temperature
Critical Pressure
Critic.al Vol\.UUe
Liquid Density at 75°F,g./cc.
Liquid Viscosity at 75°F
Liquid Viscosity at -40°F
Liquid Specific Heat @75°F
Heat of Vaporization @275°F
Liquid Thermal Conductivity @75°F
Vapor Pressure @75°F
Surface Tension @77°F
Solubility of Water @75°F
F1aumabi1ity
310
570
C10F220s
l35°C
275°F
-85°C
-121°F
244°C
470°F
172 psia
0.027 cu. ft.
1.75
1.2 c.s.
11.5 c.s.
0.24 BTU/lb./oF
28.6 BTU/lb.
0.066 BTU/hr./ft./oF
0.103 psia
13.8 dynes/cm.
45 p.p.m.
non-flammable
-------
TOXICITY
The following is part of a report from Rutgers University.
on toxicity tests on P-l-D.
PROCEDURE: Male Sprague-Dawley rats weighing 250-300 grams
were exposed in a 10-liter glass disiccator containing soda-lime
for carbon dioxide removal. The pressure within the desiccator
was reduced to about 660 mm Hg and then brought to atmospheric
pressure with oxygen; the pressure was now reduced to 520 mm Hg,
the calculated amount of sample vaporized in a carburetor external
to the desiccator and the vaporized sample in air slowly admitted
into the evacuated chamber. By this means the concentration of
oxygen in the chamber was 25 to 28% and the concentration of P-l-D
about 1.5%. The closed system was then put under a positive pres-
sure of about 1 mm Hg of oxygen, and the oxygen thereafter admit-
ted into the desiccator upon demand, that is, in response to its
use by the rats, the desiccator pressure falling by virtue of the
use of oxygen and the removal of carbon dioxide.
TEST PRODUCT P-l-D: This high boiling compound was vaporized
in an amount which saturated the atmosphere of the exposure chamber,
yielding a concentration of approximately 1.5%. Rats were exposed
for 6.3 hours; killed with ether 3 days later, no pathology was
seen; the lungs were normal in appearance and weight.
CONCLUSIONS: These initial results indicate the essential phy-
siological inertness, at the concentrations tested, of the per-.
fluorinated ether. Apart from the apparent lack of activity on
the central and autonomic nervous systems, there is also the fact
that, at least for the times of exposure used, these materials do
not appear to be pulmonary irritants.
Given concentrations and times which are unlikely to be exceeded
in practice, this compound would appear to present no inha14tional
toxic potential and no danger to personnel.
3U
-------
STABILITY
.High temperature stability tests were made on P-l-D using FC 75 as a control. The results o~
these tests are summarized below. These show that P-l-D has excellent stability at 700°F.
under the test conditions and is considerably better than FC-75.
TABLE I ;
RESULTS OF ntERHAL STABILln TESTS BASED ON ntE FLUORIDE ION ANALYSES OF P-lD AND FC-7S IN THE PRESENCE OF VARlOOS METALS
Temperature: 371'C (700'F). Time: To 126 Day.
Copper Stainless Steel Alum1nUIII Cold Rolled Steel
Sample Day. la Flu1d on Metal in Fluid on Metal .1n Fluid on Metal 10 Fluid on Metal
P-lD 7 4 710 S 203 ND 190 9 338
P-lD 20 61 Sl S1 190 ..- 1020
l-lD 84 81 149 203 176 14 406 203 6'00
1-1.0 126 40 740 S 116 14 338 108 2300
Fe-1S 8 15 46 181 510 70 460 878
Fe-1S 20 610 139 540 40e 810 148 1070 1430
FC-1S 84 1460 4150 1800 9500 8300 6300 3860 1SOO
FC-1S 126 3860 18.500 6100 12.000 3100 21.000 4630
w
~
tV Notes:
1. Tests were performed 10 an evacuated Pyrex tube conta1n1ua
0.741 P-lD or 0.54g FC-7S 1n the presence of a metal rod
of 1.26 102 area. Fluor1de lon 1. expre.sed 1n ppm ~ part.
by weight of fluoride found per part by weight of fluid.
2. Test. at 3n.C of P-l.O only gave 460 ppm of ,- after
126 days and 162 ppm of ,- after 213 day..
3. HI) - non-detectab1e.
Some preliminary data was developed at lOOO°F. which indicated substantial decomposition.
after 30 hours. Although all the decqmposition products were not identified, the fragments
found indicated rupture of the c-o bond.
-------
ATTACHMENT VI
CHARACTERISTICS OF AEF-78
313
-------
AEROJET LIQUID ROCKET COMPANY
P. O. BOX 1Saaa SACRAMENTO. CALIF'ORNIA 961113 . TELEPHONE (916) 355-0:12:1
28 July 1971
9612:0098
~UG.
5 '91\ .
Hr Robert A Yano
Project Engineer
Thermo Mechanical Systems Company
7252 Remmet Avenue
Canoga Park, California 91303
Dear Mr Yano:
Mr Alan H Kreeger has asked me to respond to your letter
of 2 July 1971 in which you requested various properties of two
Rankine cycle fluids: P-1D and AEF-78. Fluid AEF-78 is a recent
selection for our low-pollution organic Rankine cycle engine
program.
Table I summarizes some of the available properties of
P-ID and AEF-78. I would suggest that you contact Allied'
Chemical Company, Buffalo, New York, if you require more complete
data on P-1D. We are currently planning to determine several
additional properties of AEF-78, particularly in the general area
of health hazards and fluid/material compatibility. Table ~ does
not contain any specific infol~ation on performance characteristics
(Item 5), but perhaps we could supply required input if you would
be a bit more specific about the information you require.
Table II contains some of our preliminary thoughts on
environmental considerations. We would be interested in any
ments you might have in this area which might then enable us
respond more completely to your needs.
fluid
com-
to .
We are very much interested in your company's role in the
demonstration of the feasibility of the Rankine cycle engine as a
safe, low emission automotive power-plant and look forward to
further interaction with you. Please do not hesitate to communi-
cate with us should you require additional information.
Very truY11 yours,
....--~~ l / "
I :.'''' ( ,
\ ,kl../t' I './ - '0.~
....
r Ramon Garcia
Sr Engineering Specialist
Enc1:
(1)
(2)
Table I
Table II
314
A !)IVISION OF THE
AEROJET-OENERAL
CORPORATION 6
-------
TABLE I - ~~UID PROPERTIES
Propert.Y .F1uid .'\I;:K~:78 Fluid P-1D
1. Thermodynamic Characteristics 189°F 27SoF
Boiling point
Free7.ing point -103°F -121°F
Critical temperature 484°F 470°F
Critical pressure 443 psia 172 psia
Heat of combustion N/A N/A
2. FJ rc Hazards
Flash poin t None None
Fl rc poin t N/A *
Autoignition temperature N/A *
Explosive range N/A *
3. Heal th Hazards N/A (to be measured) *
4. Volume availability Present: 5-10 gal Present: 5-10 gal
Future: 100,000 - Future: *
1,000,000 lb/yr
5. Performance Characteristics High thermal stability High thermal stability
6. Corrosive Characteris tics Currently being measured *
7. Cost Present: $800/ga1 Present: $llOO/gal
Future: $100/gal Future: *
::ot£'s:
N/A - Not Available
* - Contact fluid manufacturer (Allied
Chemical Company) for data
315
-------
w
~
(J)
TABLE II - XOR.'L\L OR ABr\OR:-rAL ENVIRO~1-tENT LDtITS FOR ;.~.r! J ,\!.:;.-78
Environmental Phases
~ormal Environment
1.
Shipment of fluid raw
materials
Manufacture of fluid is covered by
patents held by supplier. we are
not aware of fluid raw materials
required or detailed manufacturing
procedure at this time.
2.
Manufacture of fluid
3.
Shipment of fluid
Fluid at ambient temperature is sealed
in a metal container with air occupy-
ing space above fluid.
4.
Manufacture of engine and
vehicle assembly
Fluid passages are flushed and cleaned.
Fluid is introduced into system and
vacuum condition is established.
s.
Normal passenger vehicle
operation
Fluid is contained in a hermetically
sealed system under vacuum. Fluid
temperature ranges from approximately
150°F to 700°F, while fluid pressure
level is 10 psia to 1000 psia.
~bnormal Environment
In the event of a leak, fluid at
ambient temperature would not react
with air (air is already present).
The fluid does not have a flash
point, and there would be no fire
hazard. The fluid has a low level
of toxicity comparable to "FREON"
refrigerant fluids and would not
pose a serious problem.
In the event of a leak, fluid at
high temperature and pressure could
be released to the atmosphere. The
fluid has no flash point, and there
is no fire hazard. Toxicity and
possible health hazards associated
with the high temperature fluid are
not known. Danger of personnel
coming into contact with hot fluid
is present.
-------
w
......
-..J
Table II continued
Environmental Phases
Normal Environment
6.
Vehicle accidents
Fluid is contained ~~thin system.
There is no hazard to personnel.
7.
Garage maintenance or
repair
Drain fluid at ambient temperature.
Flush system. Perform repair and
recharge fluid. Establish vacuum.
8.
Vehicle or engine
scrapping
Drain fluid at ambient temperature.
Flush system. Engine can be
scrapped. Fluid may be saved
for further use if decomposition
is slight or reprocessed for future
use.
Abnormal Environment
Fluid could lcak from system at
high temperature. There is no
fire hazard. Toxicity of high
temperature fluid is unknown.
" Danger of personnel coming" into
contact with hot fluid is present.
-------
8.0
POWERPLANT COMPARISON
8.1
INTRODUCTION
In any study of this nature, it is desirable to finally make a com-
parison between the various concepts investigated along with conventional
and Rankine cycle systems.
However, a rigorous comparison poses significant
problems.
For example, it is frequently QLgued that the automotive gas
turbine is lighter than the conventional spark ignition engine, therefore
placing the gas turbine in an advantageous position.
On the other hand,
the prime criteria i.n the manufacture of the conventional spark ignition
engine is minimum cost.
If weight was
of particular importance, the
block could be made of aluminum instead of cast iron, or the engine could
be made air-cooled as are engines for light aircraft.
Thus, in this case
of the conventional spark ignition engine, engine weight has been traded
to gain minimum cost.
Therefore, this section on powerplant comparisons
will attempt to cOmpare weight, cost, efficiency, etc. of each powerplant on
a generalized basis rather than trying to provide specific quantities for
each parameter.
8.2
CO?4P ARI SONS
Table 8-1 presents a comparison of the characteristics considered
most important in evaluating the various powerplants with respect to auto-
motive feasibility.
The spark ignition internal combustion (otto) engine,
witho~t extensive emission controls, is used as the reference powerplant
and the other candidate powerplants are compared to it.
The diesel, Wankel, stratified charge, and otto cycle engines are
all considered to be generally comparable as a class, all being internal
combustion engines.
The Wankel has the advantage of being smaller and
lighter, whereas, the diesel trades off weight and cost to gain fuel
economy.
On the other hand, the ECPE, Rankine, and Stirling, all being
I
external combustion engines*, will be somewhat worse with respect to specific
weight, cost, and complexity as compared to the Otto cycle while the gas
*External combustion engine, as used here, refers to
which the air and fuel mixing and combustion occurs
external to the power producing machinery.
a powerplant in
outside or
318
-------
TABLE 8- I
COMPARISON OF VARIOUS AUTOMOTIVE POWERPLANTS
En'Jine Type
Specific
Weight/lb/HP
Cost
$/HP
Complexity
Full-load
Thennal Eff.
Road-load Drive-
Thermal Eff. ability
Otto (5I-IC) Ref. Ref. Ref. Ref. Ref. Ref.
Stratified Charge Worse Worse Worse Better Better Equal
Diesel Worse Worse Worse Better Better Equal
Wankel Better Equal Better Equal Equal Equal
Gas Turbine (Regen) Equal Worse Equal Equal Worse Worse
Rankine Worse Worse Worse Equal Equal Equal
ECPE Worse Worse Worse Equal Equal Worse
Stir ling Worse Worse Worse Better Better Worse
319
-------
turbine should be comparable with respect to weight and complexity, but
significantly worse with respect to cost.
The complexity of the gas turbine
was considered to be comparable to the Otto engine, even though there
are far fewer moving parts in the gas turbine, due to the fact that the high
operating speeds and required close bearing, nozzle, and rotor tolerances
demands somewhat more complexity for the fewer moving parts that it does
have.
The fuel economy for the external c;<;..,bustion cycles is shown to be
equal to or better than that for the Otto cycle except for the part load
fuel consumption of the gas turbine.
Of course, there are indication~ that
some of the advanced gas turbine concepts, as studied in this report (such
as the plate-fin recuperative gas turbine), are comparable to the otto
cycle at hoth full and part power.
However, in general, the part load fuel
consumption of the turbine is worse than the Otto cycle and is shown in
Table 8-1 to indicate an area where continued improvement, both analytical
and experimental, is needed.
with respect to driveability, the external
combustion engines, in general, exhibit greater acceleration lag than
internal combustion engines except for the Rankine engines, which, after :-
startup, miantain a constant source of high temperature, high pressure
working fluid in the boiler.
With respect to startup, however, the Rankine
engine should require somewhat more time than any of the other engines
due to the requirement to fill the high temperature, high pressure boiler.
Thus, it appears that there is no single powerplant which is clearly
superior to the spark ignition internal combustion engine.
It would appear
that a strong case for any of the alternate powerpl~~ts can be made only if
such powerplant(s) can show significant advantages with respect to exhaust
emissions over that possible with the conventional spark ignition engine.
8.2.1
Spark Ignition (Otto Cycle) Engine
The spark ignition engine has been the prime mover for automobiles
for many years.
rapid response.
It has many advantages.
It provides immediate starting and
The weight, volume, and cost are reasonable, and the
demonstrated thermal efficiency is exceeded only by that of the diesel and
stirling engines.
The primary construction material is cast iron which is
both abundant and inexpensive.
320
-------
Table 8-II presents demonstrated exhaust emissions of various auto-
motive powerplants.
To date, the only engine that has met the 1976 emission
standards, is the Ford stratified charge engine.
Thermal Electron Corporation
(TECO) has demonstrated that they can meet the 1976 emission standards with
their Rankine burner, however, only the burner has been tested, and not the
burner-powerplant combination.
Whether the burner will perform as well when
combined with the powerplant, installed in a vehicle, and subjected to the
various driving cycles, is merely speculations at the present time.
The
Ford stratified charge engine is still in the R&D stage -and it may be some
time before it could go into mass production.
Nevertheless, the potential
for low emissions, with an internal combustion engine, has been proven.
DuPont has supplied the State of California with six Chevrolet
Impalas equipped with exhaust gas recirculation and exhaust gas thermal
reactors, and six standard Impalas for reference use.
The emission levels
for the cars equipped with controls are shown in Table 8-wI.
As can be
seen, the unburned hydrocargon and NO emissions approach the 1976 standards
x, .
while the CO emissions are significantly higher. These are full-sized
conventional passenger automobiles operating in every day service, and
demonstrate the emission gains which have been made with the conventional
Otto cycle engine.
A group of students from Wayne State university entered a vehicle
in the 1970 Clean Car Race which showed significant reduction in emissions.
The vehicle was a modified 1971 Ford Capri powered with a 302 cubic inch
V-8 engine.
The engine was fueled with unleaded gasoline and the car was
altered to reduce its weight and aerodynamic drag.
The emissions system
consisted basically of an exhaust gas recirculation system (EGR) and a
catalytic converter system.
A low overlap camshaft was installed in the
engine to improve the effectiveness of the EGR system.
Included in the
dual exhaust system were two pairs of catalytic converters to chemically
remove NO , carbon monoxide, and unburned hydrocarbons. A modified air
x
cleaner, carburetor and fuel system, along with a catalyst-temperature
modulated air injection system, were used to maintain maximum catalyst
efficiency at all engine operating conditions. The vehicle performance
and measured emissions were as follows:
321
-------
TABLE 8- II
DEMONSTRATED EMisSION LEVELS OF VARIOUS AUTOMOTIVE POWERPLANTS
Engine & Reference
Emissions, gm/mile
HC CO NOx
1976 Federal Standard
0.41
3.40
0.40
SPARK IGNITION I.C. ENGINE
1. Chevrolet Impala with DuPont Reactor (49) 0.46 10.05 0.60
2. Wayne State Clean Car Entry 0.49 5.38 0.43
3. Liq. Pet. Gas, Thompson, NAPCA (50) 1. 0- 2 . 4 1.2-4.2 1.7-3.3
4. Ford Stratified Charge (EPA Release Sep. '71) 0.37 0.93 0.33
DIESEL
1. Mercedes 220D, (51) 0.30 1.62 1.83
GAS TURBINE
1. Chrysler Gas Turbine Car (52) 0.91 7.04 1.86
2. Williams Research (53) 0.34 4.5 2.15
RANKINE
1. 1923 Doble (43) 1.4 3.9 2.0
2. SE-IOl Grand Prix (43) 1.0 8.0 2.2
3. SE-124 Chevelle (43) 0.3 1.0 1.7
322
-------
Acceleration, 0-60 mph:
Turnpike fuel economy:
7.5 sec.
24.5 mpg
Exhaust emissions (Test per Fed. Reg. 11/10/70)
- ] Lvdrocarbons :
0.49 gm/mi.
5.38 gm/mi.
0.43 gm/mi.
- cJ~bon monoxide:
- nitrogen oxides:
It is seen that the exhaust emissions are very close to the 1976 standards
while maintaining outstanding vehicle performance.
Another spark ignition engine which is a strong contender as an
automotive powerplant is the rotary engine.
Dr. Felix Wankel and a team
from NSU Notorenwerke A.G. Nekarsulm first publicly demonstrated the Wankel
rotary engine on January 19, 1960.
since that time substantial progress
has been made in the development of this unique engine.
Advantages of the
rotary engine over the reciprocating engine are as follows:
less vibration
1.
2.
simpler structure
ports replace valves eliminating the need for valve
3.
4.
operating gear
less weight and space are required for equal
horsepower output.
Some of the disadvantages currently facing the rotary engine are
as follows:
1.
the configuration of the casing makes it difficult
to manufacture
2.
there is a complex problem of sealing between the rotor
and the casing
3.
combustion and lubrication problems are more severe
than with reciprocating engines
4.
because of the large combustion chamber surface to
volume ratio, the engine emits relatively higher quantities
of unburned hydrocarbons arid carbon monoxide.
There is much optimism in the automotive industry that the above
disadvantages will be ove~come, and the smaller, lighter engine permits!
more room in the engine compartment for exhaust emission" controls.
If the
323
-------
disadvantages listed above are, in fact, overcome, the Wankel rotary engine
may well be the engine of the future.
8.2.2
Compression Ignition (Diesel Cycle) Engine
As shown in Table 8-1, diesel engines weigh. more and cost more per
horsepower, and are more complex, than Otto cycle engines.
However, diesels
provide significantly better fuel economy than Otto cycle engines and have
enjoyed wide acceptance in trucks and other vehicles where fuel costs are
considered a major operating expense.
Table 8-11 presents the results of
constant volume sampling (CVS) emission tests by Southwest Research Insti-"
tute on the Mercedes 220D diesel powered automobile.
The HC and CO emissions
of 0.30 and 1.62 grn/mile are well within the 1976 standards while the NO
x
emission of .1.83 grn/mile is still substantially higher than the standard.
The CO emissions are low because diesels inherently run at lean mixtures
while the HC emissions are low for that reason and because the combustion
process is usually completed before the fuel-air mixture can reach the
cylinder wall and be quenched.
In the future it may be possible to reduce
the oxides of nitrogen by using exhaust gas recirculation, lower compression
ratios, and prechamber combustion.
Diesels have the disadvantage of forming
aldehydes which create an obnoxious odor.
8.2.3
Gas Turbine
The gas turbine has proved to be an ideal propulsion system for high
speed aircraft, but as an automotive powerplant, it has not corne into
general use for a number of reasons.
Basically, there are three problem:
areas.
First, the gas turbine powerplant is characterized by acceptable
fuel consumption only at high turbine temperatures and pressure ratios
characteristics of high rotor speeds and power output.
To obtain acceptable
fuel consumption at part power and speed, high turbine tem~eratures must be
maintained in cambination with a very effective heat recovery system.
Second, high operating temperatures require turbine and combustor materials
having a high content of critical metals.
And third, the torque speed
characteristics of the automobile and the gas turbine engine are such that
either a complicated turbine arrangement or a complicated transmission are
required.
Perhaps the most critical of the above problem areas is the
324
-------
requirement for high temperature turbine materials.
Reference l6 indicates
that a 4:l pressure ratio gas turbine using a ceramic regenerator would
require about 2.03 Ib of cobalt per engine and 12.2 Ib of nickel per engine.
For an annual production of 10 million engines, the cobalt required would
be 156% of current U.S. consumption and the nickel required would be 38%
of current U.S. conswnption.
For this many engines, the cost of manufacture
per engine would go down significantly but the cost of material would
increase.
The ultimate cost per engine is difficult to estimate.
From the HC and CO (;mission standpoint, the gas turbine has some
definite advantages (Table 8-11).
The Williams Research engine in a
Volkswagen squareback vehicle produces 0.34 gm/mile of unburned hydro-
carbons, which is within the 1976 standards, and CO emissions of 4.5 gm/
mile, which is very close to the 1976 standard.
of NO is above the standard by a factor of 5.
x
However, the 2.15 gm/mile
It is difficult to speculate
when a solution to the NO problem will be found, but it is anticipated
:x
that eventually the problem will be solved.
The Williams Research engine produces 80 HP, weighs 400 pounds, and
completely fills the engine compartment of the AMC Hornet in which it is
installed (53).
The fuel economy
is less than that of a comparable piston
engine, and because the vehicle is underpowered, it requires 16 to 20
seconds to accelerate from 0 to GO mph.
However, advanced gas turbine
concf";pts (such as the hiqher pressure ratio plate-fin recuperative cycle.
shown in this study) have, on paper, shown much improved performance and
efficiency levels over that demonstrated in the Williams engine.
The
Williams engine itself can be improved by incorporating variable power
turbine nozzles and by increasing turbine inlet temperature.
It appears
that continued development is required on various gas turbine powerplant
concepts before the gas turbine can be considered to be an acceptable
automotive powerplant.
8.2.4
Rankine Cycle
The steam engine was a well established predecessor to the internal
combustion engine and was superceded by the internal combustion engine
because the latter engine was more efficient, more compact, less compli-
cated, and less expensive.
The advent of organic working fluids has improved.
the fuel economy of this type of closed cycle powerplant, but it has not
significantly improved the complexity, compactness, cost or driveability
325
-------
L--
I
of the system.
On cold starts, warmup time is required to build up a head
of steam, and to heat up steam pipes and expander to preclude condensation
in them.
In addition, many of the working fluids now being considered
are either toxic, flammable, freezable, unstable, corrosive, or very
expensive (several hundred dollars per charge).
Demonstrated emission levels of the Rankine powered automobiles are
shown in Table 8-II. Again, the NO emission levels are higher than the
x
1976 standard by a factor of about 5.
Recent combustion research sponsored
by EPA has indicated emission levels from the Rankine burner can be made to
meet the 1976 standard.
However, even if the burner-powerplant combination
can be made to meet the 1976 standards, it is doubtful that it can be made
into an acceptable automotive powerplant, for the reasons stated above.
8.2.5
External Combustion Piston Engine (ECPE)
As an automotive powerplant, the ECPE is similar in many respects
to the Rankine cycle engine.
Both will cost more, weigh more, and are more
complex than the conventional Otto cycle engine.
Both use external combus-
tOrs and have about the same potential for meeting the 1976 emission stand-
ards.
Because of the similarity of the operational characteristics and
the fact that comparable components exist in each of the ECPE and Rankine
systems, it is concluded that both are equally feasible as low emission
automotive powerplants.
The Rankine engine may be slightly superior with
regard to the problem of acceleration lag due to a continuous source of
high energy working fluid in the boiler.
However, the ECPE does not have
the freezing, lubrication, and separation of fluid problems of the Rankine
engine. It is judged that their overall costs should be comparable. In
comparison to the gas tUrbine engine, the ECPE promises to have favorable
part load fuel consumption and a lesser requirement for critical materials
(due to a much lower cycle temperature), although total system weight is
higher.
8.2.6
Stirline Engine
The Stirling engine is a relatively high efficiency powerplant, with
an efficiency falling close to the diesel cycle engine.
From the emissions
326
-------
standpoint, it has the advantage of an external burner so emissions should
be similar to the Rankine cycle engines.
On the other hand, the Stirling
engine introduces a degree of complexity far beyond that required for the
internal combustion engine.
It is ~stimated that the size, cost, and
weight of such an engine would be at least twice that of the existing
gasoline internal combustion engine.
Hydrogen is currently used as a working fluid in~ost Stirling
engines.
Hydrogen is a higly flammable gas and may pose a problem insofar
as hazards are concerned.
The fact that there is only a small quantity of
hydrogen used in each engine, however, tends to reduce the hazard.
Helium
can be used but the working pressure must be increased from about 2000 psi
to about 3000 to 3500 psi to maintain equivalent power output.
This higher
pressure involves major structural problems.
Although the Stir line engine
does have potential as a low-emission powerplant, it is believed that other
Angines being considered offer greater merit as an acceptable automotive
powerplant.
8.3
CONCLUSIONS
Based upon the comparisons of various powerplants reported in this
section, the following conclusions are drawn:
1.
The gas turbine inherently produces low values of HC and CO
emissions.
It does use significant quantities of critical
alloys and currently produces NO emissions higher than the
x
1976 standard by a factor of 4 or 5. Major advances in gas
turbine technology will have to be made for this engine to
2.
become an acceptable automotive powerplant.
EPA burner/vapor generator development programs have demon-
strated that the Rankine engine can meet the Federal 1976
NO emission standards.
x
a vehicle.
This remains to be demonstrated in
The Rankine engine is, however, costly and complex,
3.
and it may not be an acceptable automotive powerplant.
The External Combustion Piston Engine eliminates some of the
undesirable Rankine engine working fluid problems.
Because
the ECPE concept, like the Rankine, is more complex than
Otto cycle and gas turbine engines, it may not prove to be
327
-------
as acceptable a powerplant for vehicle applications.
Further
design and development work is required to ascertain the ECPE
potential.
4.
The Stirling engine has the potential of being a low-emission
powerplant, but it is complex and costly.
It is believed
5.
that other engines being considered offer greater merit.
Based on demonstrated emission levels, various versions of
the internal combustion engine (Wayne state Clean Car entry,
diesel engines, stratified charge concept, DuPont reactor
system, etc.) have significant leads over other proposed
powerplants.
It could 'well be that this lead will be difficult
to over come. .
328
-------
9.0
10.
REFERENCES
1.
Taylor, C.F., The Internal Combustion Engine in Theory
and Practice, Massachusetts Institute of Technology
Press, Cambridge, Mass.
Volumes I and II, 1966, 1968.
2.
Kays and London, Compact Heat Exchangers,
McGraw-Hill Book Co., New York, 1964.
3.
Engineering Handbook,
Wright Aeronautical Corporation, January, 1948.
4.
Smith and Stinson, Fuels and Combustion,
McGraw-Hill Book Co., New York, 1952.
5.
"Design and Performance Evaluation of the Variable
Displacement Engine", Thermo Mechanical Systems Co.,
U.S. Army TACOM (Contract: DAAE07-67-C-2852),
Report No. SR-IO, July 15, 1969.
6.
Welsh, H.W. and Riley, C.T., "The Variable Displacement
Engine:
An Advanced Concept Powerp:).ant", SAE Paper No.
710830, October 26, 1971.
7.
Bayley, F.J. and Rapley, C., "Heat Transfer and Pressure
Loss Characteristics of Matrices for Regenerative Heat
Exchangers", ASME 65-HT-35.
8.
Baumeister and Marks, Standard Handbook for Mechanical
Engineers, McGraw-Hill Book Co., New York, 1967.
9.
Morris, R.E. and Kenny, D.P., "High Pressure Centrifugal
Compressors for Small Gas Turbine Engines", AGARD,
Helicopter Propulsion Systems, Ottawa, 1968.
Meyer, Adolph, "Recent Development of Gas Turbines",
Mechanical Engineering, Volume 69, 1947.
329
-------
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
""\
Berchtold, Max, "The Comprex Diesel Supercharger",
Paper 63A, SAE Summer Meeting, June 1958.
Burri, Hans U., "Nonsteady Aerodynamics of the Comprex
Supercharger", ASME Paper 58-6TP-15, March 1958.
Foa, Joseph V., "Pressure Exc~ange, Applied Mechanics
Reviews", Volume 11, No. 12, December 1958.
Ames Research Staff, "Equations, Tables, and Charts
for Compressible Flow", NACA Report 1135, 1953.
AiResearch Manufacturing Co. of Phoenix, Arizona,
"Automobile Gas Turbine Optimum Configuration Selection
Study", Mid-Term Report to Office of Air Programs
(Contract No. 64-04-0012), August 25, 1971.
Wright, E. A., et aI, United Aircraft Research Lab-
oratories, Final Report (Contract No. EHS 70-115) ,
"Manufacturing Cost Study of Selected Gas Turbine
Automobile Engine Concepts", August, 1971.
Chew, N.B., "Gas Turbine Powered Trucks on the Job",
SAE Transactions 710269.
Benstein and Wood, "Applications and Performance Levels
of Radial Inflow Turbines", SAE Transactions No. 653D,
January 1963.
Wood, H.J., "Gas Turbines in Future Industrial Vehicles",
SAE Transaction No. 650480.
Shepherd, D.G., Principles of Turbomachinery,
The Macmillan Company, Ninth Printing, 1969.
Zucrow, Aircraft and Missile Propulsion,
Volume II, Wiley & Sons, Inc., New York, 1958.
330
-------
22.
sawyer, J.W., Gas Turbine Engineering Handbook,
Gas Turbine Publications Inc., Stanford, Conn. 1960.
2 J.
Dean, R.C., "On the Unresolved Fluid Dynamics of the
Cen tri fugal Compressor," NHiJE Publication
I\dvimced Centrifugal Compressors, fJP 1, 1971.
24.
'I'urenen, W. A., and Collman, ,J. S., "The General Motors
l
-------
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
Heffner, F.E., "Highlights from 6500 hours of Stirling
Engine Operation", SAE Paper No. 949D, Int'l Automotive
Engineering Congress, January 11-15, 1965.
Walker, G. and Khan, M.L, "Theoretical Performance of
Stirling Cycle Engines", SAE Paper No. 949A, Int'l
Automotive Engineering Congress, January 11-15, 1965.
Kirkley, D.W., "A Thermodynamic Analysis of the Stirling
Cycle and a Comparison with Experiment", SAE Paper No.
949B, Int'l Automotive Engineering Congress, January
11-15, 1965.
Creswick, F.A., "Thermal Design of Stirling Cycle
Machines", SAE Paper No. 949C, Int'l Automotive
Engineering Congress, January 11-15, 1965.
Accident Facts, National Safety Council, Chicago,
Illinois, 1970 Edition.
Accident Facts, National Safety Council, Chicago,
Illinois, 1968 Edition.
K1inkenberg and von der Mijn, "Electrostatics in the
--
Petroleum Industry", Elsevier Publishing Co.,
Amsterdam, London, New York, and Princeton, N.J. 1958.
Taylor and Taylor, "Crash Fire Tests with Diesel Oil",
Aviation, November 1930.
Scull, "Relation Between Inflammab1es ~nd Ignition
Sources in Aircraft Environments", NACA TN 2227,
December 1950.
Vickers, Mondt, Haverdink, and Wade, "General Motors'
Steam Powered Passenger Cars - Emissions, Fuel Economy,
and Performance", Paper presented at SAE Nat'l West
Coast Meeting, L.A., Calif. (700670), August 24-27,1970.
332
-------
44.
45.
4(, .
47.
48.
49.
50.
51.
52.
53.
54.
55.
Morgan, D.T., and Raymond, R.J., Thermo Electron Corp.
Report 1'10. TE4121-133-70 (Contract No. CPA 22-69-132) ,
"Conceptual Design, Rankine Cycle Power System with Organic
Working Fluid and Reciprocating Engine for Passenger
Vehicles", June, 1970.
Meijer, R.J., "The Philips Stirling Thermal Engine",
Philips Research Laboratories, Eindhoven, Netherland5.
Keenan and Kaye, Thermodynamic Properties of Air, 1945.
Keller, C., "The Escher-Wyss AI< Closed Cycle Turbine, Its
Actual Development and Future prospects", ASME November,
1946, pp 791.
McDonald, C.F., "A Oircumferentially Oriented Modular Gas
Turbine Recuperator", ASME Publication No. 68-GT-50,
presented March 17-21, 1968.
First Progress Report, Project CI., DuPont Reactor Vehicles,
California Air Resources Board, January 12, 1971.
Thomson, J.C., "NAPCA Findings on Gaseous Fuels",
Paper presented at LP-Gas Engine Fuel Symposium,
Detroit, October 21, 1970.
Springer, K.J., "Emissions from Gasoline and Diesel Powered
Mercedes 220 Passenger Car", Southwest Research Institute
Report AR-813, June 1971.
Korth, M. W. and Rose, A. H. Jr., "Emiss ions from a Gas Turbine
Automobile", SAE Paper 680402, May 1962.
Ludvigsen, K., "Williams Turbine Takes the Road", Motor
Trend, November 1971.
D'Alleva and Lowell Journal, SAE Volume 38, No.3, March 1936.
Newhall and Starkman, SAE Paper No. 670122.
333
-------
56.
57.
58.
~;9 .
60.
Newhall, H.K., and Shaked, S.M., "Kinetics of NOx
Formation in High Pressure Flames", February 27, 1970.
Meyers, J.H., and Pembleton, T.K., "Gas Turbine Power
for Earthmoving Equipment", SAE Paper 680251, April 1968.
Sunstrand Engineering Proposal *2294A-Pl, "400 HP Dual-
Mode Hydromechanical Truck Transmission".
October 1966.
New Mach Number Tables for Ram-Jet Flow Analyses,
Ordnance Aerophysics Laboratory, August, 1955.
Solar, "Low Emission Combustor/Vapor Generator for
Automotive Rankine Cycle Engines", Presented at Rankine
Coordination Meeting, Ann Arbor, Michigan, September
29-30, 1971.
334
-------
lO.O
APPENDIX I
OAP Vehicle Design Goals and Road Load Characteristics for the
Six-I:tissenger Automobile.
Thermo Mechanical Systems Co. has adhered as closely as possible to
the OAP design goals for the standard six.-passenger vehicle (Revision C -
May 28, 1971).
These goals are presented at the end of this section.
For
this study, less attention was directed towards the specific power ratings
of the various powerplants examined since they were not (and not intended
to be) designed in the detail necessary to accurately define the weight,
torque curves, etc., to determine acceleration rates, parasitic drive
losses, etc.
Usually, such characteristics are determined only after the
hardware is fabricated and a basic engine calibration can be measured.
Therefore, with respect to adhering to the design goals, considerable
latitude exists for the various powerplantssuch that the design parameter
of each can be "scaled. to satisfy all the design goals.
For example,
if the parasitic drive train losses for the ECPE vehicle should be larger
than those assumed, then the mean cycle pressure need merely be increased
over the nominal value to enable the vehicle to meet the design goals.
In addit.ion, the acceleration required for various power systems will vary
depending on its particular torque characteristics.
For instance, a high
torque Rankine cycle engine will require lower power levels to meet the
vehicle acceleration requirement than a low torque single-shaft gas turbine
engine. However, it has been determined that there is only a slight weight
penalty, for say, an 8-speed transmission relative to a 3-speed transmission,
and therefore, the torque curve of the engine can be made to satisfy the
acceleration requirements by using the required multi-speed transmissions.
. Therefore, it was decided to neglect secondary differences in power train
characteristics and assume similar drive train losses, transmission char-
acteristics, and acceleration rates for each of the powerplants examined,
which results in one peak power value.
The above approach is further justified as shown by Figure I-l
where the road load power as a function of speed is shown for various
weight vehicles.
The weights vary from a minimum of 3700 Ibs to a maxi-
mum of 5300 Ibs which includes up to 2600 Ibs for the powerplant, power
335
-------
- VEHICLE
120
WEIGHT
T . 5450R 5300 1bs
amb.=
V = 0 4300 Ibs
ambo
A. 100 3700 Ibs
0::
.
II)
Q)
~
.....
.....
~ ~ 80
~ +
A.
0::
~
~ t1.
f'tS
0 ~
0:: Q
Q 0 6
~ ~
Q)
H ,c;;
::>
0: II
~ ~
.
tJ'
Q)
Il: 4
o
40
60
80
100
2
Vehicle Speed - MPH
FIGUEE 1.1
Total Power Required due to Rolling Resistance
and Aerodynamic Drag as a Function of Vehicle Speed.
33Fi
1056
-------
train, payload, and test load (see "Yehicle Design Goals").
It can be
seen that the powerplant weight will have little effect on road load power.
From the "Yehicle Oesign Goals" it was determined that for a full size
vehicle (2700 lbs) with an engine plus power train weight of 1300 Ibs and
a standard passenger and test load of 300 Ibs, the design weight is 4300
Ibs.
The road load horsepower (aero + rolling) for this 4300 lb automobile
i ~-1 2Il~;o shown in Figure 1.1.
The power losses for the transmission, drive train, fans, and
intake and exhaust ducting were assumed to be as follows:
Fan and auxiliary losses
Transmission losses
6%
7%
Rear axle losses
Intake and Exhaust losses
4%
3%
20%
Therefore, the total estimated engine and drive-line power losses
were 20%.
Combining this with the average 4 HP accessory load (as given
in the design goals) and the aero and rolling HP requirement as shown in
Figure 1.1, gives:
p ~ p + p + P + 0.20 P
v RR . AJ) ACC m.
or
Pv = 1.25 (PRR + PAD) + ~.9
where:
= total vehicle power requirement
Py
PRR = road rolling resistance
= W/65 [1 + (1.4xlO-3) Y + (1.2X10-5)y2] /550
PAD = road aerodynamic resistance
= CoAf ~(y3/29)/550]
PACC= average accessory load = 4 HP
W = vehicle weight = 4300 lbs
V = vehicle speed (ft/sec)
COAf = air drag = 12 ft2
337
-------
Figure 1.2 shows the total vehicle power curve for the 4300 Ib vehicle
o
at ambient conditions of 14.7 psia and 545 R.
Finally, the peak power requirement for the vehicle was determined
from the vehicle design goals.
The most demanding maneuver was accelerating
from a standing start to 440 ft in 10.0 seconds.
The power required was
approximately 150-160 HP for the 4300 Ib veticle (including all auxiliary
and accessory requirements).
Therefore, the powerplant design power level
was determined to be 160 HP and, referring to Figure I.2, this gives a
maximum speed of about 107 mph.
338
-------
" " "
ROAD LOAD POWER REQUIREMENTS
160
140\
I
Co
..c
I 120'
-oJ
0::
a. 100\
...
0::
LLJ
w S
w 0
'-D a. 80,
o
20\
C1
LLJ. 9
0::
o
o
....
w
o
0'\
APCO VEHICLE DESIGN GOALS
" .
SIX PASSENGER AUTOMOBILE"
" "
VEHICLE TEST WEIGHT =: 4300 LBS 1
, . > .
P b '= 14.7 .
am .
Tamb. : = 545°R
PRL' =1.2S(PRR: PAD} +.4~9
20
40
60:
VEHICLE SPEED - MPH
100\ 107,
80
FIG URE 1. 2
I
I
I
I
I
I
I
I
I
I
120
-------
AIR POLLUTION CONTRC:. JFFICE
ADVANCED AUTOMOTIVE POWER SY~TEMS PROGRAM
"Vehicle Design GoalS - Six Passenger Automobile"
(Revision C - May 28, 1971 - 11 Pages)
The design goals presented below are intendeo to provide:
A common objective for prospective contractors.
Criteria for evaluating proposals and selecting a contractor.
Criteria for evaluating competitive power systems for entering
first generation system hardware.
Advisory criteria in such areas as rolling resistance, vehicle air
drag etc. are included to assist the contractor.
The derived criteria are based on typical characteristics of the class of
passenger automobiles with the largest market volume produced in the U. S.
during the model years 1969 and 1970. It is noted that emissions, volume
and most weight characteristics presented are maximum values while the
performance characteristics are intended as minimum values. Contractors
and prospective contractors who take exceptions must justify these exceptions
and relate these exceptions to the technical goals presented herein.
1.
Vehicle weight without propulsion system - Woo
Wo is the weight of the vehicle without the propulsion system and
includes, but is not limited to: body, frame, glass and trim,
suspension, service brakes, seats, upholstery, sound absorbing materials,
insulation, wheels (rims and tires), accessory ducting, dashboard
instruments and accessory wiring, battery, passenger compartment.
heating and cooling devices and all other components not included ~n
the propulsion system. It also includes accessories such as, the air
conditioner compressor, the power steering pump, and the power
brakes actuating device. .
Wo is fixed at 2700 1bs.
2.
Propulsion system weight - Wp'
W') includes the energy storage unit (including fuel and containment),
power converter (including both functional components and cant,",) 15)
and power transmitting components to the driven wheels. It a150
includes the exhaust system, pumps, motors, fans and fluids necessary
for op.racion of tho propull1on .ystem, and any propul.1on .y.tam
heating or cooling devices.
340
..,. .
""I""'~\-'~'\"~.':"""""""- ~ ~ ..........
.~:.< .
.. "
..~~~""'" ,.
''''''''.~ _.~,..V'.. ~~. ......-... ~;.f,;.
-------
. .
",
1
1
j.
Rev. C - May 28, 1971
-2-
The m;.n:imum allowable propulsion system weight, W , is 1600 lbs.
liO\vcver-:llght weight propulsion systems are high~y desired. .
(Equivalent 1970 propulsion system weight with a spark ignition
engine is 1300 lbs.)
3.
Vehicle curb weight - W
. c
Wc = Wo + Wp
The maximum allowable vehicle curb weight, Wcm, is 4300 lbs.
(2700 + 1600 max. = 4300)
4.
Vehicle test weight - Wt.
h\ = Wc + 300 lbs. Wt is the vehicle weight at which all accelerative
maneuvers, fuel economy and emissions are to be calculated. (Items 8c,
8D,8e).
The maximum allowable test weight, Wtm, is 4600 1bs.
max. + 300 = 4600).
(2700 + 1600
5.
Gross vehicle weight - Wg
Wg = Wc + 1000 Ips. Wg is the gross vehicle weight at which sustained
cruise grade velocity capability is to be calculated. (Item 8f). The
1000 lbs. load simulates a full load of passengers and baggage.
The maximum allowable gross vehicle weight, Wgm, is 5300 lbs.
1600 max. + 1000 = 5300). 'I.
(2700 +
6.
Propulsion system volume - Vp
VD includes all items identified under item 2. Vp shall be packagable
in such a way that the volume encroachment on either the passenger or
luggage compartment is not significantly different than today's (1970)
standard full size family car. The propulsion system shall not violate
the vehicle ground clearance lines as established by the. manufacturer
of the vehicle used for propulsion system/vehicle packaging. Additionall~
the propulsion system shall not violate the space allocated for wheel
jounce motions and vehicle steering. Necessary external appearance
(styling) changes will be minor in nature. Vp shall also be packagab1e
in such a way that the handling characteristics of the vehicle do not
depart significantly from a 1970 full size family car.
The maximum allowable volume assignable to the propulsion system,
Vpm' is 35 ft.3.
I
\
341
", - -----.---..
-~-.....~
-------
-...-
Rev. C - Ma1 28, 1971
-3-
7.
Emission Goals t
The v~hicle when t~sted fqr emissions in accordance with the procedure
outlined in the November 10, 1970 Federal Register shall have a
weight of Wt. The emission goals for the vehicle are:
Hydrocarbons* t
Carbon monoxide t
Oxides of nitrogen**
Particulates
0.14 grams/mile maximum
4.7 grams/mile maximum
0.4 grams/mile maximum
0.03 grams/mile maximum
, ,
*Total hydrocarbons (using 1972 measurement procedures)
plus total oxygenates. Total oxygenates including
aldehydes will not be more than 10 percent by weight
of the hydrocarbons or 0.014 grams/mile, whichever is
greater.
**measured or computed as N02'
t ror updated emission standards refer to EPA Prototype Vehicle Performance
Specification, Jan. 3, 1972, item 9, "Propul~ion Syst.em Emissions."
8.
Start up, Acceleration, and Grade Velocity Performance.
a.
Start up:
The vehicle must be capable of being tested in accordance, with
the procedure outlined in the November 10, 1970 Federal Register
wit~out special driver startup/warmup procedures.
The maximum time from: key on to reach 65 percent full power
is 45 sec. Ambient conditions are 14.7 psia pressure, GO°F
temperature.
Powerplant starting techniques in low ambient temperatures shall
be equivplent to or better than the typical automobile spa!k-
ignition engine. Conventional spark-ignition engines are deemed
satisfactory if after a 24 hour soak at -20°F the engine achieves
a self-sustaining idle condition without further driver input
within 25 seconds. No starting aids external to the normal v~hic."
system shall be needed for -20°F starts or higher temperatures.
. ..
I I
342
-------
.,
. .
,-..
-4-
. Rev. C - Ma\Y 28. 1971
b.
Idle operation conditions:
The fuel consumption rate at idle operating condition will not
exceed 14 percent of the fuel consumption rate at the maximum design
power condition. Recharging of energy storage systems is
exempted from this requirement. Air conditioning is off, the
power steering pUmp and power brake actuating device, if
directly engine driven, are being driven but are unloaded.
The torque at transmission output during idle operation (idle
creep torque) shall not exceed 40 foot-pounds, assuming conventional
rear axle ratios and tire sizes. This idle creep torque should
result in level r6ad operation in high gear which does not exceed
18 mph.
c.
Acceleration from a standing start:
The minimum distance to be covered in 10.0 sec. is 440 ft.
The maximum time to reach a velocity of 60 mph is 13.5 sec.
Ambient conditions are 14.7 psia, 850 F. Vehicle weight is Wt.
Acceleration is on a level grade and initiated with the engine
at the normal idle condition.
d.
Acceleration in merging traffic:
The maximum time to accelerate from a constant velocity
of 25 mph to a velocity of 70 mph is 15.0 sec. Time starts
when the throttle is depressed. Ambient conditions are 14.7
psia, 850 F. Vehicle weight is Wt, and acceleration is on
level grade.
e.
Acceleration, DOT High Speed Pass Maneuver:
The maximum time and maximum distance to go from an initial
velocity of 50 mph with the front of the automobile (18 foot
length assumed) 100 feet behind the back of a 55 foot truck
traveling at a constant 50 mph to a position where the back
of the automobile is 100 feet in front of the front of the 55
foot truck is, 15 sec. and 1400 ft. The entire maneuver takes
place in a traffic lane adjacent to the lane in which the truck
is operated. Vehicle will be accelerated until the maneuver is
completed or until a maximum speed of 80 mph is attained, which-
ever occurs first. Vehicle acceleration ceases when a speed of
80 mph is attained, the maneuver then being completed at a
constant 80 mph. (This does not imply a design requirement
limiting the maximum vehicle speed to 80 mph.) Time starts when
the throttle is depressed. Ambient conditions are 14.7 psia,
850 F. Vehicle weight 18 Wt, and acceleration is on level grade.
343
-------
..-.. .
.,
-5-
Rev. C - May 28, 1971
f.
Grade velocity:
The vehicle must be capable of starting from rest on a 30
"'percent grade and accelerating to 15 mph and sustaining it.
This is the steepest grade on which the vehicle is required
to operate in either the forward or reverse direction.
The minimum cruise velocity that can be continuously maintained
on a 5 percent grade with an accessory load of 4 hp shall be
not less than 60 mph.
The vehicle must be capable of achieving a velocity of 65 mph
up a 5 percent grade and maintaining this velocity for a
period of 180 seconds when preceded and followed by continuous
operation at 60 mph on the same grade (as above).
\
The vehicle must be capable of achieving a velocity of 70 mph
up a 5 percent grade and maintaining this velocity for a
period of 100 seconds when preceded and followed p.r, ~p'~~inuous
operation at 60 mph on the same grade (as above).
The minimum cruise velocity that can be continuously maintained
on a level road (zero grade) with an accessory load of 4 hp
shall be not less than 85 mph with a vehicle weight of Wt.
Amb.ient conditions for all grade specifications are 14.7 psia
850 F. Vehicle weight is Wg for all grade specifications
except the zero grade specification.
The vehicle must be capable of providing performance (Paragraphs
8c, 8d, 8e 8f)w1thDri5'percent of the stated 850 F values, when
operated at ambient temperatures from -200 F to 1050 F.
344
-4 - ..4.~. _.._.~- .
- . - "4. -_. ._4._.~._. --.. ... ~- .
"... 'T"" ...
. .. 4__.'_.-.._~_.--... _4.___.. ~'-'-...._._.__... ._._------- -
-------
. .
-6-
Rev. C - ~~ 28, 1971.
9.
Minimum vehicle range:
Minimum vehicle range without supp1ement1ng.,:th.e_'en~~ay- 8torag~:.
will be 200 miles. The minimum range shall be calculated for,
and applied to each of the two following modes: 1) A city-
suburban mode, and 2) a cruise. mode. '.
Mode 1:
~.
.HI.
Is the driving cycle which appears in the
November 10, 1970 Federal Register. For
vehicles whose performance does not depend
on the state of energy storage, the range
may be calculated for one cycle and ratioed
to 200 miles. For vehicles whose performance
does depend on the state of energy storage
the Federal driving cycle must be repeated
until 200 miles have been completed.
Mode 2:
Is a constant 70 mph cruise on a level road for
200 miles.
The vehicle weight for both modes shall be, initially, Wt. The
ambient conditions shall be a pressure of 14.7 psia, and temperatures
of 60° F, 85° F and 1050 F. The vehicle minimum range shall not
decrease by more than 5 percent at an ambient temperature of -20° F.
For hybrid vehicles, the energy level in tbe power augmenting device
at the completion of operation will be equivalent to tbe energy level
at the'begi?ning of operation. .
I i.
345
.. . '-' .
-------
~ ~..
" .
.. .....- -.-".-
-7-
Rev. C .. Ma.Y 28, 1911
10.
System thermal efficiency:
System thermal efficiency will be calculated by two methods:
. A.
A "fuel economy" figure based on 1) miles per gallon
(fuel type being specified) ap" 2) the number of Btu
per mile required to dr1ve the v~h1cle over the 1972
Federal driving cycle which appears in the November'
10, 1970 Federal Register. Fuel economy is based on
the fuel or other forms of energy delivered at the
vehicle. Vehicle weight is Wt.
B.
A "fuel economy" figure based on 1) miles per gallon
(fuel type being specified) and 2) the number of Btu
per mile required to drive the vehicle at constarit
speed, in still air, on level road, at speeds of 20,
30, 40, 50, 60, 70, and 80 mph. Fuel economy 1s based
on the fuel or other forms of energy delivered at the
vehicle. Vehicle weight is Wt.
In both cases, the system thermal efficiency shall be calculated
with sufficient electrical, power steering and power brake loads
in service to permit safe operation pf the automobile. Calculations
shall be made with and without air conditioning operating. The
ambient ~onditions are 14.7 psia and temperatures of 600 F, 850 F
and 1050 F. Calculations shall be made with heater operating at
ambient conditions of 14.7 psia and .300 F (18,000 Btu/hr).
11.
Air Drag Calculation:
The product of the drag coefficient, Cd, and the frontal area, Af'
is to be used in air drag calculations. The product CdAf has a
value of 12 ft2. The air density used in computations shall
correspond to the applicable ambient air temperature.
12.
Rolling Resistance:
Rolling resistance, R, is expressed in the equation
R = W/65 (1 + (1.4 x 10-3V) + (1.2 10-5V2)] lbs. V 1s the vehicle
velocity in ft/sec. W 1s the vehicle weight in Iba.
346
. . . ... .
.. . - ... 4r' ..
.~~ ~. ... ....
. ---- -;--- --- - ----- ----...
-------
.'.
. .
-8-
Rev. C - ~ 28, 1971
13.
Accessory power requirements:
. ,
The accessories are defined as ~ubsystems for driver assistance
and passenger convenience, not essential to sustaining the
engine operation and include: the air conditioning compressor,
the power steering pump, the a1tc~~3tor (except where required
to sustain operation), .and the POWI?'t' ,brakes actuating device.
The accessories also include a device for heating the passenger
compartment if the heating demand is not supplied by waste heat.
Auxiliaries are defined as those subsystems necessary for the
sustained operation of the engine, and include condensor fanes),
combustor fanes), fuel pumps, lube pumps, cooling fluid pumps,
working fluid pumps and the alternator when necessary for driving
electric motor drive~ fans or pumps.
The maximum intermittent accessory load, Paim,is'IO hp (plus the
heating load, if applicable). The maximum continuous accessory
load, Pacm, is 7.5 hp (plus the heating load if applicable). The
average accessory load, Paa' is 4 hp.
If accessories are driven at variable
apply. If the accessories are driven
Pacm will be reduced by 3 hp.
speeds, the above values
at constant speed, Paim and
347
C
-------
,
,.". :-'
. "':"..'
.'
- -..- .-... .-..
-9-
Rev. C - Mq 28, 1971
14.
Passenger comfort requirements:
Heating and air conditioning of the passenger compartment shall be
at a !ate equivalent to that provided in the present (1970) standard
full s1ze family car.
Present practice for maximum passenger compartment heating rate is
approximately 30,000 Btu/hr. For an air conditioning system at 1100 F
ambient, 800 F and 40% relative humidity air to the evaporator, the
rate 18 approximately 13,000 Btu/hr.
15.
'Propulsion system operating temperature range:
The propulsion system shall be operable within an expected ambient
temperature range of -400 to 1250 F.
16.
Operational life:
The mean operational life of the propulsion system should be
approximately equal to that of the present spark-ignition engine.
The mean operational life should be based on a mean vehicle life of
105,000 miles or ten years, whichever comes first.
The design lifetime of the propulsion system in normal operation will
be 3500 hours. Normal maintenance may include replacement of
accessable minor parts of the propulsion system via a usual maintenance
procedure, but the major parts of the system shall be designed for a
3500 hour minimum operation life.
The operational life of an engine shall be determined by structural or
functional failure causing repair and replacement costs exceeding the
cost of a new or rebuilt engine. (Functional failure is defined as
power degradation exceeding 25 percent or top vehicle speed degradation
exceeding 9 percent). .
348
~
.-.. --. r" .,......
..-.---..
\
\
-------
, . ~
.',
18.
, ,,-.
, \
. !
-10-
Rev. C, Mq 28, 1911
17. Noise standards:
(Air conditioner not operating)
"
a.
Maximum noise test:
The maximum noise generated by the vehicle shali not
exceed 77 dbA when measured in accordance with SAE J986a.
No~e that the noise level is 77 dbA whereas in the SAE
J986a the level 'is 86 dbA.
. ',1
b.
Low speed noise test:
The maximum noise generated by the vehicle shall not exceed
63 dbA when measured in accordance with SAB J986a except
that a constant vehicle velocity of 30 mph is used on the
pass-by, the vehicle being in high gear or the highest gear
in which it can be operated at that speed.
c.
Idle noise test:
The maximum noise generated by the vehicle shall not exceed
62 dbA when measured in accordance with SAE J986a except that
the engine is idling (clutch disengaged or in neutral gear)
and the vehicle passes by at a speed of less than 10 mph.
the microphone will be placed at 10 feet from the centerline
of the vehicle pass line.
'-)
Safety standards:.
The vehicle shall comply with all current Department of Transportation
Federal Motor Vehicle Safety Standards. Reference DOT/HS 820 083.
349
-------
. .
,
'f .'
.~
~.
~
----'j'
-11-
Rev. C - Mq 28, '1971
19.
Reliability and maintainability:
The reliability and maintainability of the vehicle shall equal or
exceed that of the spark-ignition automobile. The mean-time-between
failure should be maximized to reduce the number of unscheduled
service trips. All failure modes should not represent a serious
safety hazard during vehicle operation and servicing. Failure
propagat'ion should be minimized. The power plant should be designed
for ease of maintenance and repairs to minimize costs, maintenance
personnel education, and downtime. Parts requiring frequent servicing
shall be easily accessable.
20.
Cost of ownership:
The net cost of ownership of the vehicle shall be minimized for
ten years and 105,000 miles of operation. The net cost of ownership
includes initial purchase price (less scrap value), other fixed costs,
operating and maintenance costs. A target goal should be to not
exceed 110 percent of the average net cost of ownership of the present
standard size automobile with spark-ignition engine as determied by
the U..S. Department of Commerce 1969-70 statistics on such ownership.
~
350 '
- -- __0- -------- - _.-.. _...__. -. .-
. '-----'----'.'-*'-~--''''''~-'_.',-----'-------
---. ----.
!
-- ----.-.-
------- |