PHASE I FINAL REPORT
       HEAVY-DUTY ENGINE WORKING  GROUP
MOBILE SOURCE TECHNICAL ADVISORY SUBCOMMITTEE
       CLEAN AIR ACT ADVISORY COMMITTEE
                  APRIL 1997

-------
             PHASE I FINAL REPORT
       HEAVY-DUTY ENGINE WORKING  GROUP
MOBILE SOURCE TECHNICAL ADVISORY SUBCOMMITTEE
       CLEAN AIR ACT ADVISORY COMMITTEE
                  APRIL 1997

-------
                          Phase T Report

 This  report documents  the staV.u^ of the EPA-Joint Industry Heavy
 Duty  Engine Work Group  (HDEWG'}  activity at  the  transition from
 Phase I to  Phase  II of  our  cooperative program.   Based  on the
 results of  Phase  I testing,  the HDEWG has concluded  that the
 modified Caterpillar 3176 engine  at Southwest Research Institute is
 a  valid test  tool  for  evaluating  the  effects of  fuel property
 changes on NOx emissions at  FTP NOx+NMHC  level  of  2.5 g/bhp-hr.
 Based on  these  results  and  the  guidelines established  in the
 "Intent of Initial  Phase of Test Program" document,  we recommend
 moving  forward with  Phase II.

 Background

 The HDEWG was formed with the intent of contributing to  EPA's 1999
 technology review with  an  assessment  of the  relative  merits of
 achieving  the 2.5 g/bhp-hr NOx + NMHC  emission level by either:

  Engine-system only
  Engine-system + fuel modification

 The Mission Statement of the HDEWG is given as Attachment 1.   The
 first activity of the work group was to  tabulate a description of
 data  in the literature pertaining to  fuel  effects  on heavy duty
 engine  emissions for reference,  and to  determine if any  emissions-
 fuel effects data is published at the 2.0 g/bhp-hr NOx  level.  This
 tabulation,  given  as Attachment  2,  indicates that  no meaningful
 data appears in the literature at the 2.0-3.0 g/bhp-hr NOx  level  .
 The second work group activity was to design and conduct a  low NOx
 engine  and  fuel experimental  study in  four phases:

 Phase  I -- Validation of' "Transparent Test  Tool"  (2Q96-4Q96)
 Phase II -- Fuels Matrix and Engine Variables Testing   (4Q96-4Q97)
 Phase  III  -- Fuel Effects Validation   (4Q97-1Q98)
 Phase  IV  -- Report  (2Q98 target)

 Phase I Results

 The initial phase of  the   experimental study,  "Validation of  a
 Transparent  Test  Tool",  has  been  completed per  the  guidelines
 established in the "Intent of Initial Testing" document  (Attachment
 3) .   The  agreed upon  intent of this  initial phase  of the test
 program was  "to find out if the proposed "transparent  test  tool-"
 has a similar  overall  response  in  emission performance to  large
 changes in  fuel parameters  as the "black box" engines  - with all
 engines demonstrating  emissions capability  at  or  near  the 2.5
g/bhp-hr NOx + NMHC  level."

The engine being  proposed as   a  "transparent test  tool"  is  a
Caterpillar  3176   with  a cooled,  low   pressure  EGR.  The  engine
 features electronic unit injectors with  30 KSI injection pressure
capability.  The  engine operates  at 17 bar BMEP at rated, 22 bar at
peak torque.  Steady state 8 mode testing was employed because the

-------
 configuration  and control of  the  EGR system on  this  engine was
 incapable of generating representative r-   -sient emissions data.
 Though  there  is confidence that steady  .-late testing correlates
 well with transient NOx results, there is not the same confidence
 in  steady state  to transient particulate correlation.  Therefore,
 no  particulate  measurement was  prescribed  for  this  phase  of
 testing.   Phase III  of the test program  is  considered  to be the
 most appropriate place to evaluate particulate impact.

 Three fuels were tested in Phase I:  1)  a baseline commercial No. 2
 diesel fuel; 2} a cetane enhanced version of the baseline fuel; and
 3)  a high  (natural)  cetane,  low aromatic fuel.  Fuel properties,
 based on a  round-robin analysis by eight labs, are summarized in
 Table 1 and given in detail as Attachment 4.

 Six engine  manufacturers generated  FTP  and  a prescribed steady
 state  synthesis  (see Figure" 1) of  FTP  emissions data  on these
 fuels.  "Black  box"  research  engines which met  a  criteria  of
 operating at less than 3.0 g/bhp-hr NOx, and less  than 0.5 g/bhp-hr
 HC, and less than 0.15  g/bhp-hr particulate were used.  Lab-to-lab
 variability was not a significant factor  since only percent change
 versus the baseline  fuel  was being reported.  The use of percent
 change  data allowed  the  engine manufacturers to  protect their
 proprietary research engine performance data.  These engine tests
 were all run in EPA compliant  test facilities.

 Comparisons  of  "black  box"  engine  results with  those  of  the
 "transparent test tool" engine at SWRI are given  in Tables  2 and  3
 and are displayed  in the  accompanying graphics.   A more detailed
 listing of the "black box"  engine  data is given as Attachment 5.
 The  best  correlation  between  the  "black box"  engines  and the
 "transparent   test  tool"   was  on  NOx  emissions.   The  engine
 manufacturers  report a  7.6  percent  decrease in  NOx  from the
 baseline  fuel   to  the high cetane,  low  aromatic  fuel,  with  a
 standard deviation of  1.5 percent.   The  "transparent test tool"
 generated a 7.0 percent decrease in NOx for  the same  fuel  change.
 Hydrocarbon  results  compare  favorably  in  terms   of  order  of
 magnitude,  but due to the very  low levels being measured,  percent
 change and standard deviations  are  heavily influenced by extremely
 small differences.

 In  terms of the response  to cetane  enhanced baseline  fuel,  NOx for
 the black box engines increased by an average of  2.4  percent, with
 the  test  tool  generating a 3.4  percent  increase.   Hydrocarbon
measurements were  highly  variable again,  with black box  engines
 reporting an average 11 percent decrease  in hydrocarbon emissions,
as compared to no change with  the test tool.

Discussion of  Phase I Results:

While the primary objective of Phase  I  was  to validate the SWRI
Caterpillar-3176 as a "transparent test tool", it was also used  to
determine if the magnitude of fuel  effects on low emission engines
were significant enough to move to Phase  II  of the program.

-------
 The  magnitude of  the  NOx change  derived  from high  cetane,  low
 aromatic  fuel  of  7.6 percent correlates  with tl   findings of CRC
 VE10  project.   It is also in general  agreement  with the Chemkin
 (chemical  kinetics model)  prediction of  the change expected from
 the  aromatic or H/C ratio change discussed in  an  earlier HDEWG
 meeting.

 The negative or negligible impact of cetane  enhancement (0.5% alkyl
 nitrate} on  emissions observed in Phase I was unexpected.  "Black
 box"  results varied  from a 5.3  percent increase  to a 0.6 percent
 decrease  in  NOx.   Cetane enhancement is  understood to affect NOx
 formation  primarily by  reducing  the  amount of  fuel burned in the
 "premixed"  combustion  mode.    Premixed (as  opposed  to diffusion
 controlled)  combustion  yields  high  temperatures and pressures
 associated with high rates of NOx formation.  The combustion heat
 release measurements taken by SWRI,in Phase I  indicate that the
 amount of  fuel burned in  the premixed combustion mode was reduced
 by cetane  enhancement,  though by a smaller amount than expected.

 Previously published data (1)  suggests  that  increased cetane number
 tends to reduce gaseous emissions.  Effects  on particulate matter,
 "PM", are more variable,  with a number of  studies  showing little or
 no sensitivity of  PM emissions to increased cetane number.  Several
 of these (notably SAE 950251) and other studies, have  compared the
 effect  of natural vs.  boosted  cetane  number  on  emissions,  and
 concluded  that the effects  of  increased  cetane  number achieved
 either  through  changes  in  base fuel  composition   or  through
 additives  (alkyl nitrates, peroxides etc.)  are equivalent.  There
 is little  or no data in  the published literature which suggests
 that  either  fuel  bound  nitrogen or  nitrogen contained in alkyl
 nitrate ignition improvers makes a significant contribution to NOx
 emissions  from diesel engines.   However,  it should be noted that
 these observations derive from assessment of fuel  effects on heavy-
 duty  diesel  engines  {without EGR) calibrated to  meet US 1998 or
 earlier emissions standards.  The influence of cetane  (natural or
 boosted) on  engines designed to meet US 2004 emissions  standards,
 which - like the engine(s) used  in the Phase I tests  -  are likely
 to be equipped with EGR,  has  not  been fully evaluated.   Therefore,
 the engine results in  Phase I are inconsistent  with our current
 understanding of controlling phenomena, and it is  recommended that
 tests  conducted in  Phase II  of this study include  a thorough
 evaluation of cetane effects on gaseous emissions.

 References:
 1.  SAE:  902171,  922267,  930728,  932685,  932800, 941020, 950250,
 950251, 961074.

 The text of  the SWRI report  on Phase I is Attachment  6.

 Phase I Conclusions

The modified CAT 3176 engine at Southwest  Research Institute  (SWRI)
  responds to  fuel property  changes  in a manner  representative of
  the "black box" research  engines.   Therefore,  we have a valid

-------
   test  tool with which  to proceed  to  Phase  II.

Fuel effects on emissions on heavy duty diesel engines operating at
   the 2.0-3.0 g/bhp-hr NOx level are significant enough to warrant
   further  study in  Phase II.

During  the analysis of  the Phase  I  data,  a number  of  questions
   relative to the  impact and  interaction of engine EGR  rates and
   injection timing  with fuel properties were discussed.   Based on
   these questions,  EGR  rates  and  injection  timing  will  be
   evaluated in Phase II.

-------
Table 1. Phase I Test Fuel Properties.

Baseline - "At the
Pump"
Cetane Enhanced
Baseline
High Cetane, Low
Aromatic
Cetane Number
46
52
57
SFC Aromatics,
%m
26
26
15
Density, g/cm3
.8564
.8564
.8233
Sulfur, ppr- !
310
315
180
Table 2. Comparison of emission response to high cetane, low aromatic fuel.




Black
Box
Engines
SWRI
FTP NOx



-7.6%


-
8-Mode
NOx


-7.6%


-7.0%
FTPHC



-35%


-
8-Mode
HC


-26%


-13%
FTP
Particulate


-16%


-
8-Mode
Particulate
*

-16%


-
FTP Fuel
Consumpfi
on

-2.2%


-
8-Mode
Fuel
Consumpti
on
-0.5%


-1.1%
*2 engine sample
Table 3. Comparison of emission response to cetane enhancement.




Black
Box
Engines
SWRI
FTP NOx



+1.5%


-
8-Mode
NOx


+2.4%


+3.4%
FTPHC



-33%


-
8-Mode
HC


-11%


-0.3%
FTP
Particulate


+0.3%


-
8-Mode
Particulate


+0.5%

,
-
FTP Fuel
Consumpti
on

-0.5%


-
8-Mode
Fuel
Consumpti
on
+1 .0%


+0.4%
*2 engine sample
                               50
                    Engine Speed (%)
100
Figure 1. Weighting Factors of 8-Mode FTP Synthesis

-------
Measured HC Emission Effect - High Cetane, Low

Aromatic Fuel vs. Baseline Fuel

        0


       -10


       -20
     o -30
     D)
     O
     o>
       -40
-50


-60


-70


-80
             Transient
                  8-Mode
                EMA Black Box Engines
                                     Baseline Fuel
SWRI

8-Mode



CAT 3176
                                         Black = Average
                                                    lwnt»Tai4i7.ppl

-------
Measured NOx Emission Effect - High Cetane, Low
Aromatic Fuel vs. Baseline Fuel
       -2
     x
     O
     o>
     O>

     ro -6

     O
     +*

     Q) Q
     O -O

     0)
     a.


       -10
       -12
             Transient
8-Mode
             EMA Black Box Engines
                                     Baseline Fuel
                     Black = Average
SWRI

8-Mode


CAT 3176

-------
Measured Particulate Effect - High Cetane, Low
Aromatic Fuel vs. Baseline Fuel
                                   Baseline Fuel
                                               Black = Average
                   Transient
8-Mode
                      EMA Black Box Engines
                                                  bolK»7JZ14»7.|>flt

-------
Measured HC Emission Effect - Getane Enhance
Fuel vs. Baseline Fuel
   10
o
 o> -10
 D)
 o
 "c
 o
   -60
                      m
          Transient
                       8-Mode
              EMA Black Box Engines
                                       Baseline Fuel
                                               Black = Average
SWRI 8-Mode


  CAT 3176

-------
Measured NOx Emission Effect - Cetane Enhanced
Fuel vs. Baseline Fuel
           Transient
8-Mode
              EMA Black Box Engines
                                             Black = Average
                                      Baseline Fuel
SWRI
8-Mode

CAT 3176
                                                    ppt

-------
Measured Particulate Effect - Cetane Enhanced Fuel
vs. Baseline Fuel
   0)
   D)
   c
   CO

   O
   +rf
   c
   0
   U

   Q)
   Q_
\£.
in -
I U
8_
6_

4_
-


-
A
-4 -






































:,.-,.',:!: : :



•: '.'.'-.•:'.











_ ;:;;; ,_, |
Base




             Black = Average
              Transient
8-Mode
                     EMA Black Box Engines

-------
       ATTACHMENT I
HDE Work Group Mission Statement

-------
                             Attachment  1

    Heavy Duty Engine Work  Group Mission Statement

                             Final 4/11/96
The work group will be comprised of representatives each from the heavy-duty
engine manufacturers and the petroleum refining industry.  Representatives from
the Technical Advisory Subcommittee, other stakeholders, and EPA will also be
included as appropriate.

The work group is charged with developing emissions performance information
and with assessing the relative merits of approaches to achieve the 2.5 g/bhp-hr
NOx + NMHC emission level by either:

       1.     Engine-system only
       2.     Engine-system + fuel modification

It is recognized that the output of this work group will contribute to EPA's 1999
technology review and  that cost, cost effectiveness, lead time and related factors
will be considered.

The timely assessment process will include evaluation of existing data on low
emission heavy-duty diesel engines, acquisition and testing of low emission
production viable diesel fuel formulas on prototype heavy-duty diesel engines
with production intent technologies capable of approaching the 2.5 g/bhp hr NOx
+ NMHC emission level.

The results of this work and any recommendations from the work group on
follow-up activities will be presented to the EPA and the Technical Advisory
Committee for their review and approval.  The assessment should be completed
by October 31,  1997.

The work group will be co-chaired by a representative from the engine
manufacturers, the petroleum industry, and a "non-industry" representative from
the  advisory subcommittee.

-------
Suggested Approach for Test Plan Development

1.      Review current available information to direct.

       a.      Selection of engine NOx control technologies for evaluation.

       b.      Selection of fuel properties and range of values for evaluation.

2.      Design test program using appropriate test matrix to eliminate engine
       drifts and investigate non-linearity.

       a.      Engine-only emission control strategies using current low sulfur
              diesel certification fuel.

       b.      Engine-fuel emission control strategies using engines equipped
              with emission controls able to approach 2.5 g/bhp-hr NOx +
              NMHC (on current low sulfur diesel certification fuel) and a fuel
              matrix with production-feasible test fuels.

              1.     Develop engine baseline performance and emissions on
                    current  commercial low sulfur diesel fuel.

              2.     Evaluate the impact of test fuels on emission performance.

              3.     Assemble test results in report usable to facilitate economic
                    evaluation of strategies.

-------
        POSSIBLE STRUCTURE FOR EXPERIMENTAL PROGRAM
PHASE I   (1996-2Q97):

       Screen fuels with common engine at Southwest Research (Caterpillar
       3176 fitted with exhaust gas recirculation; previously used for SWRI
       clean diesel program).
PHASE 2   (2Q97):

       Evaluate specific candidate fuels with in-house experiments at individual
       engine manufacturer research laboratories.
      Engine manufacturers are currently planning a "round-robin" test with
      two fuels ("baseline" low-sulfur commercial diesel fuel; "low emission"
      low-aromatic, high cetane experimental fuel) to confirm that the response
      of the Caterpillar 3176 engine at SWRI to fuel modifications is
      representative of current proprietary low-emission research engines.

-------
     ATTACHMENT 2
Analysis of Current Literarture

-------
        DATA REVIEW (1990-1996) - FUEL EFFECTS ON EMISSIONS FROM PRE-EGR HEAVY DUTY DIESEL ENGINES

Commentary and conclusions from a representative (but not exhaustive) review of literature concerning fuel effects on heavy duty (non-EGR) diesel engines
(see attachment 1) and data supplied by GARB on the quality and composition of diesel fuel in California (see attachment 2) are as follows:

1.   Careful design of test fuel matrix is essential to minimise uncertainty in interpretation of results.

2.   Changes in both fuel quality and engine design/calibration can influence emissions (and other aspects of performance such as power and fuel econony),
    but In general fuel effects tend to be significantly smaller than engine effects.
       •  see SAE941020 and SAE950250 where fuel effects on emissions were assessed at two different engine timing settings.
       «  see SAE961074 where the effect of timing changes on NOx, Pm, power and fuel consumption was compared with changes in fuel quality

3.   Different engines have different emissions responses to the same change in fuel quality
       •  see SAE941020 and SAE950250 where 2 different engines were tested over the US transient cycle on the same fuel matrix
       •  see SAE961074 where 5 different engines were tested over the EU steady state test on the same fuel matrix (different to above)
    => assessment of the air quality impact of changing fuel quality should be based on the fuel property/emissions sensitivity of a representative range of
    engines

4.   The same engine will give different absolute levels of emissions and may give different responses to fuel changes when tested on the same fuel matrix
    over the US transient and the EU steady-state tests.  This coupled with different emissions responses amongst engines makes it difficult to draw
    comparisons between conclusions of US and European studies. The only published link between US and EU studies appears  to be limited testing of 3
    EPEFE fuels in a DDC Series 60 engine over the US transient cycle which suggests this engine is less sensitive (especially for NOx) to changes in fuel
    quality over the US test cycle than EPEFE engines over the EU test cycle
       •  see reference to EPEFE report annex 7 in attachment 1

5.   The majority of European studies pre-EPEFE focused on fuel effects on Pm. Since total aromatics were found to have no influence on Pm emissions
    (supported by a number of more recent US studies eg: SAE941020), but there was uncertainly regarding the influence of polyaromatics, then the latter
    was included as a key variable to be investigated in the EPEFE program.
    EPEFE showed that polyaromatics can influence both NOx and Pm. The effect of monaromattcs was not studied, hence there remains uncertainty
    regarding the relative importance on mono and polyaromatics on NOx emissions.

6.   The majority of US studies have focused on the effects on cetane and total aromatics on NOx and Pm emissions. The separate effects of mono and poly
    aromatics have not been systematically studied.
    European studies  have additionally included systematic study of other fuel properties - notably density, which has been found to have a significant
    influence on engine performance and emissions in a number of engines, caused by physical interaction with the fuel injection system - influencing dynamic
    timing and mass delivery but not affecting the combustion process itself.
                                                                                                                                   page I

-------
7.   Re-analysis of historic data in an attempt to investigate the significance of other fuel characteristics (eg: H/C ratio) is unlikely to lead to robust conclusions.
       •   Required information infrequently reported
       •   Uncertainty vs accuracy of data where it does exist (poor test method precision for H/C ratio)

8.   A wide range of diesel fuel formulations have been certified by CARS* to have equivalent or better emissions than a reference fuel containing 10% total
    aromatics. These fuel formulations include levels of total aromatics in excess of 27%, cetane numbers below 48 and sulphur levels up to 500ppm.
    * in line with requirements of section 2282(g) of Title 13, California Code of Regulations.
                                                                                                                                           page 2

-------
ATTACHMENT 1
I Source Engine Scope Cycle delta delta delta delta Comments
NOx Pm HC CO
SAE9O21?r
Ullman et al
SAE 912425
Lange
SAE 922267
McCarthy &
Slodowske
elal
SAE930728
Nikanjam
DDC Series 60 W protoype
11.1l;330hp
-------
1 Source Engine Scope Cycle delta delta delta delta Comments
NOx Pm HC CO
SAES32685
Unge&
Gaifingetal













SAE9028QO
Rosenthal et
al





SAE941020
Ullman et al



"





MB OM366L EURO 1
production engine
S; 177kWe2eOOrpm; TC;
1C; 01; mech inline pump

7.52g/kWhr NOx and
0 344g/kWhr Pm on 0.06%
S ceil fuel








Cummins N14*94
production engine
141; 3«*W (460hp)@?rpm

4 44g/brphr NOx and
O.OTSg/bhphr Pm on 0.05%
S cert fuel

DOC Series 60 94 prototype
11.1l;239t(Wei800rpm;
DI;TC;AC:EUI(std
calibration)

nominally 5g/bhphr NOx and
0. 1 g/brtphr Pm - no baseline
data on cert fuel given



12 fuel matrix
designed to study
fuel effects on
emissions.
especially NOx and
Pm










7 fuel matrix
designed to study
the influence of
H/C ratio, boiling
range and cetane
on emissions,
principally NOx and
Pm
1 1 fuel matrix
designed lo study
the effect of cetane,
total aromatic*
(SFC) and
oxygenates on
NOx, Pm, HC and
CO



EUECER49(13
mod* - steady
state)

Constant max
power for all fuels










US FTP
(transient) - hot
start only.

A ref fuel
transient
command cycle
used throughout
US FTP
(transient) - hot
and cold starts

A ref fuel
transient
command cycle
used throughout



no effect for
total (HPLC)
aromatlcs (28-
8%).

-0.358/kWhr
for+10CN(SO-
60)








-0.148g/bhphr
for -10% total
(FIA)
aromatlcs

-0. ICgJbfiphr
for+IOCN

-O.lg/bhphr for
-10% (30-20)
total aromatlcs


+OO39flJt>hphr
for +1% (0.25-
1.25) oxygen
-o.oeggAtiphr
for +1OCN (45-
551
•••••••••MP*i*J^B*^BV^Ba*a*ai
no effect for
total (HPLC)
aromatlcs (28-
8%).

-O.OCBg/kWhr
for-003%3
(0.06O.02)
-0.012flAWhr
for -C.02kg/l
density (0.84-
0.62)
-O.OIoVkWhr
for -2%
polyaromatics
0-1%)
-O02g/bfiphr
for -10% total
(FIA)
aromatlcs




no effect for-
10% (3O-20)
total aromattcs


•0.00eg/briphr
for +1% (0.25-
1 .25) oxygen
no effect for
HOCN (4&-5S)

P^B*^BPlW^BHi*JP^B*^Pi^iiPB























no effect for
-10% (30-20)
total aromatics


+aOQ6g/bhphr
for +1% (0.25-
1 .25) oxygen
no effect for
*1OCN (1&-55)

^^•^•••••••••••av^^^^^^vx























«O.O53g/bhphr
for -10% (30-
20) total
aromatics

•O.096g/bhphr
for +1 % (O.25-
1.25) oxygen
-O1g/bhphr for
*10CN (45-55)

Fuel effects determined by pairwise
comparisons within separate 3 sub-
matrices where parameters of interest
were decor related - plus regression
analysis of full 12 fuel test matrix











Significant Intercorrelatlon between
some properties, notably aromatics
and density - leading to uncertainty vs
assignment of emissions effects and
robustness of preferred models.



Aromatlcs measured by FIA and SFC
(MS and NMR given in full report).

Significant mtercorrelation between
total, mono, poly aromatics.






nb: 1 kW= 1.341 hp
                                                                                                                                                    page 4

-------
(Source Engine Scope Cycle delta delta delta delta Comments
NOx Pm HC CO
SAE94102D
Ullmanetal










SAE942022
Den Ouden
eta!















SAE960250
Spreen et al










DDC Series eo "94 prototype
11.1l;239kW@1800rpm;
Dl; TC; AC; EUI (timing
retarded from std calibration)

nominally 4g/bhphr NOx and
Pm slightly > 0. 1 g/bhphr - no
baseline data on cert fuel
given



5 different HD engines; all
01; TC; AC; mixture of mech
and electronic control;

pre EUR0 1; EURO land
USBI emissions calibrations












Navistar OTA466 ~94
prototype
7.61; 14SkW
-------
I Source Engine Scope Cycle delta delta delta delta Commenl^
NOx Pm HC CO •&.
SAE960250
Spreen et al










SAE060251
Ullman et al





Achen
Colloquium
Oct95









Navistar DTA466 "94
prototype
761; 143kW@2400rpm
TC; AC; 01 (3 deg in] retard);
oxidation catalyst

4.31 8/bhphr NOx and
0.12g/bhphr Pm on 'base'
fuel (0.06% S)



DDC Series 60 fXJ prototype
11.1l;242kWei800rpm
TC; AC; Ol; EUI

412g/bhpnr NOx and
O.O37g/bhphr Pm on 'base'
fuel
MAN D-0326LF production
engine
601; 169kWO2400rpm
TC; 1C; 01 mech Inline pump

nominally SgAcWhr NOx and
0 36g/kWhr Pm





1 1 fuel matrix
designed to study
the effect of cetane.
aromatlcs (SFC)
and oxygenates on
NOx. Pm, HC and
CO





1 1 fuel matrix
designed to study
the effect of cetane
on NOx. Pm. HC
and CO


13 fuel matrix
designed to study
fuel effects on
emissions -
principally Pm and
NOx






US FTP
(transient) - hot
and cold starts

A ref fuel
transient
command cycle
used throughout




US FTP
(transient) - hot
and cold starts
A ref fuel
transient
command cycte
used throuQhout
EUECER4B(13
mode -steady
state)

Constant max
power for all fuels






no effect for
-10% (30-20)
total aromatlcs


no effect for
+2% (0-2)
oxygen

-O.200g/bhphr
for +10CN (45-
55)
-0.164g/bnphr
for +10CN (45-
55)




•0.210/kWhr
for -10% (25-
15) total
(HPIC)
aromatics

-0.44g/kWrv
for +10CN (50-
60)



•OOOTg/bhphr
for -10% (30-
20) total
aromatlcs

•OOOSg/bhphr
for +2% (O-2)
oxygen

-0.013flA)hphr
for +10CN (45-
55)
-O.005g/bhphr
for +10CN (45-
55)




no effect for
total aromatlcs




-0.011 g/kWhr
for +10CN (50-
60)
-0066g/kWhr
for -0.15% S
(02-O.O5)
•WWm^^^"^^^^^^^™^"^^™l
no effect for
-10% (30-201
total aromatlcs


no effect for
+2% (0-2)
oxygen

-0.4B3g/bhphr
for +10CN (45-
55)
-0.06Bg/bhphr
for +10CN (45-
55)
















no effect for
-1O% (30-20)
total aromatlcs


no effect for
+2% (0-2)
oxygen

-1 .26g/bhphr
for +10CN (45-
S6>
•O.iaeo/bhprtr
for +10CN (45-
55)
















Aromatica measured by F . 3FC
(MS and NMR data given In .'• :
report).

Significant Intercorrelation between
total, mono, poly aromatics













The effects of density on emissions
were not studied In this program.










nb: 1 kW = 1 341 hp
                                                                                                                                         page 6

-------
I Source Engine Scope Cycle delta delta delta delta Comments
NOx Pm HC CO
SAE961074
Signer et al
& final
EPEFE
report










SAE961074
Signer et al
& final
EPEFE
report














5 prototype engines which
meet EU 1996 emissions
limits of 7g/KWhr NOx and
0.1Sg/kWhrPm











5 prototype engines which
meet EU 1996 emissions
limits of 7g/kWhr NOx and
0.15g/kWhr Pm:

8.69; 6cyl;
222kW@2300rpm TC; !C

2.81; 4cyl;
64.5kW@3600rpm; TC; 1C
6.9; 6cyl;
162kW@2400rpm; TC; 1C

10.961; 6cyl;
2SOkW@1900rpm; TC; 1C;
.
6.21; 6cyl;
125kW@2500rpm;TC;IC

1 1 fuel matrix
designed to study
the effects of
density,
potyaromatics, T95
and CN on NOx,
Pm, HC and CO








1 1 fuel matrix
designed to study
the effects of
density,
potyaromatics, T96
and CN on NOx,
Pm, HC and CO












EU ECE R48 (13
mode - steady
state)

Engines set to
reference power
and tested on all
fuels without
adjustment






EUECER49(13
mode - steady
state)

Engines set to
reference power
and tested on alt
fuels without
adjustment










-0.245g/kWhr
for -0.027kg/l
density (0.855-
0.828)

-0.114g/kWhr
for -7%
polyarom (8-1)

-0.040g/kWhr
for +8CN (50-
58)
-0.120g/kWhr
for-45C
T9S(370-325)
+0.060410
-0.6511 g/kWhr
for -Q.027kg/l
density (0.856-
0.828)

-0.0483to
-0.194g/kWhr
for -7%
polyarom (8-1)
+0.0235 to
-0.0702g/kWhr
for +8CN (50-
58)

-0.0027 to
•O.2421g/kWhr
for-45C
T95(370-325)
^^^^PHVMIV^^^^^^^^M*^
no effect for
-0.027kg/!
density (0.855-
0.828)

-O.OD5g/kWhr
for -7%
polyarom (8-1 )

no effect for
+8CN (50-58)

no effect for
-45CT95
pTO-325)
+0.0034 to
XX0077g/kWhr
for -0.027kg/l
density (0.855-
0.828)

-0.000410
•O.OOSSg/kWhr
for -7%
polyarom (8-1 )
+0.001 3 to
-0.001 5g/kWhr
for +8CN (50-
58)

+0.0131 to
-0.0099g/kWhr
for-45C
T96(370-325)
+OO34g/kWhr
for -0.027kg/l
density (0.855-
0.828)

-O.OQ9g/kWhr
for -7%
polyarom (8-1)

-0.014g/kWhr
for +8CN (50-
58)
+0.031 g/kWhr
for-45C
195(370-325)
+0053710
+0.014g/kWhr
for -0.027kg/l
density (0.855-
0.828)

-0.0048 to
•0.018gAWhr
for -7%
polyarom (8-1 )
-0006110
-0.024g/kWhr
for +8CN (50-
58)

+0.0622 to
+0.0005g/kWh
rfor-45C
195(370-325)
+OO3g/kWhr
for -0.027kg/l
density (0.855-
0.828)

no effect for
-7% polyarom
(8-1)

-0.061 g/kWhr
for +8CN (50-
58)
+0.039g/kWhr
for-45C
T95(370-325L
+0.0962 to
-0.033g/kWhr
for -O.027kg/l
density (0.855-
0.828)

+O.0165to
•O.0098g^W
hrfor-7%
polyarom (8-1 )
•O.010210
-0.1H4g/kW
hr for +8CN
(5f>58)

+0.0801 to
-0.0234g/kW
hrfor-45C
195(370-325)
Emissions effects quoted are derived
from the global regression model for
the fleet of 5 engines












Emissions effects quoted are
maxlumum and minimum effects
derived from regression model for
each of the 5 individual engines.















nb: 1 kW = 1.341 hp
                                                                                                                                             page 7

-------
(Source Engine Scope Cycle delta delta delta delta Comments
NO* Prn HC CO
SAE961074
and
EPEFE
Report -
table 10.2.2















EPEFE
Report -
Annex 07













5 prototype engines
(standard timing, fuel
delivery) which meet EU
1996 emissions limits of
7g/KWhrNOxand
O.ISg/kWhrPm:

8.65f; 6cyl;
222kW@2300rpm TC; 1C

2.81; 4cyt;
84.5M/v@3600rpm; TC; 1C

6.91; 6cyl;
162kW@2403rpm; TC; 1C

10.961; 6cyl;
250kW©19COrpm; TC; 1C;
6.21; 6cy1;
125kW©2500rpm; TC; 1C
DOC Series 60 '91 prototype
(at SwRi)













Two fuels
representing the
extremes of the
EPEFE matrix

-------
ATTACHMENT 2
                      California diesel fuel
           GARB compliance sample test results
uo
en _
OU
cc .
cetane **
number cn
Oil ~
AK -
15)
A(\ .
X
•

•
A *•
V X A *
X ••
A
A A X X
X XX

... .1 ,

. -J

. S<25
S 25-100
A S 100-300
x S 300-500

                              10    20

                             aromatics %m
30
            Data provided by GARB to EPA Heavy Duty Engine Working Group - Spring 1996
            nb: all samples deemed to be in compliance with C ARB regulations
                                                                  page1

-------
        ATTACHMENT 3









        HDE Work Group




Intent of Initial Phase of Test Program

-------
                               Attachment 3

               Heavy Duty Engine Work
           Intent of "Initial"  Phase of Test Program

The purpose of this document is to define the intent of the "initial" phase of
testing to be carried out by members  of the Heavy Duty Working Group.
Specific details of the test protocol (including engine test cycles, fuel analysis
and data reporting, etc.) Will be developed separately by representatives of the
key stakeholders (EMA, API, EPA).

1.     The primary objective of the "initial" phase of the test program is to
       validate the CAT 3176 of SwRI as a "transparent test tool" relative to
       current research ("black  box") engines in the laboratories of EMA
       members.  The intention is to find out if the "transparent test tool" has a
       similar overall response  in emission performance to changes in fuel
       parameters as the "black box" engines - with all engines demonstrating
       emissions  capability at or near the "2.5 g/bhp-hr NOx + HC" level.

       The second main objective of the "Initial" phase of testing is to establish
       the emissions sensitivity to extreme  fuel  modifications for a range of
       engines with such production  intent  technologies, and to establish options
       for the main body of work to be carried out by the Heavy Duty Engine
       Working Group.

2.     The "initial phase of testing will therefore involve testing  three fuels:  1)
       current commercial No.  2-D diesel, 2) an 'extreme' No.  2 diesel fuel -
       high cetane, low aromatics with all other properties allowed to 'float',
       and 3) cetane-enhanced current commercial No.  2  diesel.  These fuels
       will be evaluated with the CAT 3176 at SwRI and current EMA research
       ("black box") engines with "2.5 g/bhp-hr NOx + HC" production intent
       technologies.   (Fuel 3  is a cetane-enhanced version of Fuel 1 using cetane
       improver.)

       It is recognized by all parties  at the  outset that the results  of this "Initial"
       phase of testing is intended to define the boundaries for fuel effects "2.5
       g/bhp-hr NOx + HC" engines and not to define relative importance of
       any specific fuel parameters.

-------
3.      The Heavy Duty Engine Working Group will analyze the results of v'ais
       program and the agreed conclusions will form an interim progress uport
       to the Technical Advisory Subcommittee.

       These agreed conclusions, together with those arising from the data
       review process, will have significant implications for the direction of the
       next phase of work to be developed and carried out by the Heavy Duty
       Engine Working Group.

       For example:

              (1)    All engines show the same relative sensitivity to NOx, PM,
                    and HC to the fuel changes.  (Good transparent test tool)

                    (a)     if this sensitivity is insignificant, then further work
                            on  fuel effects will not be necessary

                    (b)     if this sensitivity is "significant" then further work
                            is required to understand the relative importance of
                            individual fuel parameters and to determine whether
                            all  engines respond in the same way (e.g., some
                            may be more sensitive to cetane, others to
                            aromatics, etc.)

                    Therefore...

                    (a)     use the CAT 3176 at SwRI to examine effects  of
                            individual fuel parameters in a jointly agreed to
                            fuels matrix

                    (b)     use the CAT 3176 at SwRI to examine effects  of
                            engine adjustments (e.g., different levels of EGR)
                     (c)     validate findings from (a) & (b) by testing a sub set
                            of the fuels matrix defined in (a) on a range of
                            "black box" engines.

              (2)    Engines show different sensitivities to NOx, PM, and HC
                     to the fuel changes - need to understand why and identify
                     those technologies with give the minimum emissions
                     sensitivity to fuel modification jjiot an acceptable
                     transparent _test tool).

-------
           ATTACHMENT 4
Round Robin Fuel Analysis Results - Phase  1

-------
                             PHASE1FS.XLS
HDDEWG Phase 1 Test Program

Test Fuel Inspection Data.
    12/4/96
R. SobotowsKi
PARAMETER
DENSITY® 15 C,g/cm3
API GRAVITY @ 60F, deg. API
CETANE NUMBER
CETANE INDEX
CETANE INDEX
DISTILLATION, F IBP
5%
10%
20%
30%
40%
50%
60%
70%
80%
90%
95%
FBP
NMR, mol % AROMATICS
SATURATES
MS, %m PARAFFINS
MONOCYCLOPARAFFINS
DICYCLOPARAFFINS
TRICYCLOPARAFFINS
ALKYL BENZENES
INDANES / TETRALINS
INDENES
NAPHTHALENE
NAPHTHALENES
ACENAPHTHENES
'ACENAPHTHALENES
TRICYCLIC AROMATICS
TOTAL AROMATICS
SFC AROMATICS, %m 1 -RING
2-RING
3-RING
TOTAL
CARBON, %m
HYDROGEN, %m
NITROGEN, ppm
SULFUR, ppm
HOC, gross, btu/lb
MJ/kg
MJ/dm3
HOC, net, MJ/kg
MJ/dm3
VISCOSITY @40C,mm2/s
@ 100C, mm2/s
CLOUD POINT, F
POUR POINT, F
FLASH POINT, F
ALKYL NITRATE CONTENT, %v
TEST
METHOD
ASTM D4052
ASTM D4052
ASTM D6 13
ASTM D976
ASTM D4737
ASTM D86
ASTM D5292
ASTM D2425
ASTM D5 186
ASTM D5291
ASTM D5291
ASTM 04629
ASTM D2622
ASTM D4809
ASTM D445
ASTM D2500
ASTM D97
ASTM D93
Nalco GC
AMOCO-1
0.8564
33.6
45.9
45.0
44.4
365
408
432
456
476
495
513
530
550
571
601
628
649
18.1
81.9
34.33
20.35
5.84
3.42
14.06
11.07
3.34
0.81
2.48
2.24
1.20
0.86
36.06
26.6
poly 9.1
35.7
87.11
13.06
85
310
19460
45.26
38.76
42.49
36.39
2.74
1.13
0
_i
160
0.047
AMOCO-2
0.8564
33.7
52.4
45.4
45.0
350
405
430
456
476
494
513
530
549
570
602
630
646
18.4
81.6
-
26.2
poly 8.9
35.1
86.99
13.17
438
315
19411
45.15
38.67
42.36
36.27
2.72
1.13
t
„(
161
0.553
W-407
0.8233
40.3
56.9
56.2
59.3
397
436
454
472
487
499
511
522
536
553
584
612
630
9.8
90.2
60.9
17.57
5.9
2.81
5.02
3.7
1.2
0.2
0.9
0.7
0.5
0.5
12.84
15.5
poly 4.5
20.0
86.09
13.84
28
180
19817
46.10
37.95
43.16
35.53
2.72
1.15
i
j
181
< 1ppm

-------
                                                           PHASE 1FS.XLS
Table 3:  Fuel W-407 Inspection Data.
HDDEWG Phase 1 Test Program
PARAMETER
DENSITY @ 15 C , g/cm3
API GRAVITY @ 60F, deg. API
CETANE NUMBER
CETANE INDEX
CETANE INDEX
DISTILLATION, F IBP
5%
10%
20V.
30%
40%
50%
60%
70%
80%
90%
96%
FBP
NMR, mol %C AROMATIC
SATURATE
MS, %m PARAFFINS
MONOCYCLOPARAFFINS
DICVCLOPARAFFINS
TRIG YCLOPARAF FINS
ALKYL BENZENES
INOANES / TETRALINS
INDENES
NAPHTHALENE
NAPHTHALENES
ACENAPHTHENES
ACENAPHTHALENES
TRICYCLIC AROMATICS
TOTAL AROMATICS
SFC AROMATICS, %m 1-RING
Z-RING
3-RING
TOTAL
CARBON, %m
HYDROGEN, %m
NITROGEN, ppm
SULFUR, ppm
HOC, gross, _ btu/lb
' MJ/kg
MJ/dm3
HOC, net, MJ/kg
MJ/dm3
VISCOSITY @40C, mm2/s
@ 100C, mm2/s
CLOUD POINT, F
POUR POINT, F
FLASH POINT, F
ALKYL NITRATE CONTENT. %v
TEST
METHOD
ASTM D4052
ASTM D4052
ASTMD613
ASTM D976
ASTM D4737
ASTM D86
ASTM D5292
ASTM D2425
ASTM DS186
ASTM D5291
ASTM 05291
ASTM D4629
ASTM D2622
ASTM D4809
ASTM 044S
ASTM D2500
ASTM D97
ASTM D93
ASTM D4046
ARCO
0.8228
40.4
57.5
56.6
-
397
440
457
512
594
634
635
"
•
15.8
poly 4.5
20.3
65.92
14.08
27
178
-
2.70
8
j
184
-
BP
0.8231
40.3
55.3
56.3
59.4
397
439
455
473
487
499
511
523
537
554
584
611
628
9.6
90.4
-
15.0
4.6
1.0
20.6
66.29
13.73
-
180
19948
46.40
38.19
43.43
35.79
2.74
1.17
6
0
185
	 "-$•**
MOBIL
0.8228
40.4
57.6(core)
-
-
401
459
513
583
€33
9.65
90.35
-
' Vt9^(««a)
85.63
14.18
'•-^~i-m
186
19811 (core
46.03
37.9
43.07
35.44
2.73
1.14
12
5
-
*&&3p«&
NAVISTAR
0.8236
40.3
56.2(swri)
56.0
56.9
393
451
469
488
493
509
524
537
553
581
636
«
-
^8,;Sf$w$l
86.82{swri
13.39(swr<
27(swri
190(swri
19780(swri
46.01
37.89
43.17
35.55
2.68
9(swri
' 5(swri
177
-
SHELL
0.8237
40.2
58.6
56.3
59.5
396
438
454
474
488
soo
512
524
537
555
584
611
628
10.0
90.0
60.87
17.57
5.91
2.81
5.02
3.70
1.24
0.24
0.86
0.70
0.54
0.54
12.84
17.4
poly 4.5
21.9
86.17
13.85
32
176
19792
46.04
37,92
43.10
35.50
2.73
1.14
S
i
184
r-^ijaa
SWRI
0.8234
40.3
55.1
-
-
407
432
454
473
489
500
512
523
537
555
585
609
624
*
-
14.7
poly 3.9
18.6
66.04
13.78
26
170
19756
45.95
37,84
43.03
35.43
2.73
1.14
10
5
183
< 1ppm *
76
PRODUCTS
COMPANY
0.8233
•
56.0
56.2
-
395
437
453
472
486
499
510
522
535
553
581
606
623
"
-
14.3
poly 4.1
18.4
55.70
13.90
27
17
•
2.7
h ' iOg
10

180
-
PHILLIPS
0.8235
•
58.5
55.8
-
388
432
448
470
485
497
507
519
532
550
577
601
629
"
-
-
86.15
13.85
-
186
-
-:-\\^9Q
j
5
174
-
AVERAGE
0.8233
40.3
56.9
S6.2
53.3
397
436
454
472
487
499
511
522
536
553
584
612
630
9.8
90.2
60.37
17.57
5.91
2.61
5.02
3.70
1.24
0.24
0.86
0.70
0.54
0.54
12.84
15.5
poly 4.5
20.0
86.09
13.84
28
180
19817
46.10
37.95
43.16
35.53
2.72
1.15

4
181
< 1ppm "
 ' Shell method. Claimed accuracy an order of magnitude better than D5291.




 ™ Nalco GC method
."..  Data not used in averaging !!!
  ' BP Infrared method

-------
                                                  PHASE1FS.XLS
                                                                                                    12/4/96
Table 1:  AMQCO-1 Fuel Inspection Data.
                                                            HDDEWG Phase 1 Test Program
PARAMETER
DENSITY @ 15 C , g/cm3
API GRAVITY @ 60F, deg. API
CETANE NUMBER
CETANE INDEX
CETANE INDEX
DISTILLATION, F IBP
5%
10%
20%
30%
40%
50%
60%
70%
80%
90%
95%
FBP
NMR, mol %C AROMATIC
SATURATE
MS, %m PARAFFINS
MONOCYCLOPARAFFINS
D1CYCLOPARAFFINS
TRICYCLOPARAFFINS
ALKYL BENZENES
INDANES/TETRALINS
tNDENES
NAPHTHALENE
NAPHTHALENES
ACENAPHTHENES
ACENAPHTHALENES
TRICYCLIC AROMATICS
TOTAL AROMATICS
SFC AROMATICS, %rn 1-RING
2 -RING
3 -RING
TOTAL
CARBON, Mm
HYDROGEN, %m
NITROGEN, ppm
SULFUR, ppm
HOC, gross, btu/lb
MJ/kg
MJ/dm3
HOC, net, MJ/kg
MJ/dm3
VISCOSITY @40C,mm2/s
@ 100C, mm2/s
CLOUD POINT, F
POUR POINT, F
FLASH POINT, F
ALKYL MTRATE CONTENT, %v
TEST
METHOD
ASTM D4052
ASTM D4052
ASTMD613
ASTM D976
ASTM D4737
ASTM D86
ASTM D5292
ASTM D2425
ASTM 051 86
ASTM DS291
ASTM DS291
ASTM D4629
ASTM D2622
ASTM D4809
ASTM D445
ASTM D2500
ASTM D97
ASTM D93
ASTM D4046
ARCO
0.8566
33.6
45.2
44.B
-
376
415
436
515
600
626
654
;
-
r;v%®
>;.>;:}«£
-
-
81
313
-
2.73
1.13
t
\
152
-
BP
0.8566
33.6
43.8
44.7
44.1
367
407
429
454
474
492
510
528
547
569
599
624
641
18.5
81.5
-
25.9
8.4
1.9
36.2
86.98
13.00
88
320
19590
45.57
39.03
42.8
36.67
2.76
1.15

-5
16
,094
MOBIL
0.6556
33.8
44.6{core)
-
-
367
436
515
606
646
16.63
83.37
-
25.7
9.1
1.7
36.5
87.53
13.38
-
333
19410(core
45.15
38.63
42.3
36.20
2.75
1.13
1 \ ; *€
-1
160
fcor&w*:
NAVISTAR
0.8566
33.6
49.5(swri)
44.7
43.7
361
427
454
473
492
510
528
548
565
593
646
;
-
_;^
66.80(swri
13.01(swri
83{swri
310{swri
19408(swri
45.14
38,67
42.38
36.30
2.70
0(swr
-6(swn
15
-
SHELL
0.8566
33.6
46.0
45.6
45.4
366
420
439
460
480
498
515
532
552
574
604
632
654
19.3
80.7
34.33
20.35
5.84
3.42
14.06
11.07
3.34
0.81
2.48
2.24
1.20
0.86
36.06
29.1
poly 8.54
37.6
87.117"
12.916"
82
283
19457
45.2
38.7
42.5
36.40
2.7
1.1
-4
-1
15
SWSfSWSSjiW
&H*S«*R!MW?
SWRI
0.8564
33.6
44.2
-
-
351
397
• 426
453
476
495
513
531
550
572
603
629
644
•
-
26.1
poly 7.54
33.6
87.10
13.00
84
300
19434
45.20
38.7
42.5
36.40
2.73
1.13
0
J
166
0.047 *
76
PRODUCTS
COMPANY
0.8564
33.7
48,3
45.2
-
367
403
430
458
478
496
514
532
551
572
602
629
656
•
-
26.2
poly 8.5
34.7
-
-
9
31
-
2.7
1.1


16
-
AVERAGE
0.8564
33.6
45.9
45.0
44.4
365
408
432
456
476
495
513
530
550
571
601
628
649
18.1
61.9
34.33
20.35
5.84
3.42
14.06
11.07
3.34
0.81
2.48
2.24
1.20
0.86
36,06
26.6
poly 9.1
35.7
87.11
13.06
85
310
19460
45.26
38.76
42.49
36.39
2.74
1.13

-5
160
0.047 "
 ' BP infrared method




 '* Nalco GC method



 '" Shell method. Claimed accuracy an order of magnitude better than D5291.
' Data not used in averaging !!l

-------
                                                  PHASE1FS.XLS
Tabls •;•: AMOCO-2 Fuel Inspection Data.
HDDEWG Phase 1 Test Program
PARAMETER
DENSITY @ 1S C , g/cm3
API GRAVITY @ 60F, deg. API
CETANE NUMBER
CETANE INDEX
CETANE INDEX
DISTILLATION, F IBP
6%
10%
20%
30%
40%
50%
60%
70%
80%
90%
95%
FBP
NMR, mol %C AROMATIC
SATURATE
MS, %m PARAFFINS
MONOCYCLOPARAFFINS
DICYCLOPARAFFINS
TRICYCLOPARAFFtNS
ALKYL BENZENES
INDANES/TETRALINS
INDENES
NAPHTHALENE
NAPHTHALENES
ACENAPHTHENES
ACENAPHTHALENES
TRICYCLIC AROMATICS
TOTAL AROMATICS
SFC AROMATICS, %m 1-RING
2 -RING
3 -RING
TOTAL
CARBON, %m
HYDROGEN, %m
NITROGEN, ppm
SULFUR, ppm
HOC, gross, btu/lb
MJ/kg
MJ/dm3
HOC, net, MJ/kg
MJ/dm3
VISCOSITY @ 40C, mm2/s
@ 100C, mm2/s
CLOUD POINT, F
POUR POINT, F
FLASH POINT, F
ALKYL NITRATE CONTENT, %V
TEST
METHOD
ASTM D4052
ASTM D4052
ASTMD613
ASTM D976
ASTM D4737
ASTM D86
ASTM DS292
ASTM D2425
ASTMDS186
ASTM D5291
ASTM D5291
ASTM D4629
ASTM D2622
ASTM D4809
ASTM D445
ASTM D2500
ASTM D97
ASTM D93
ASTM D4046
ARCO
0.8566
33.6
54.4
44.6
-
369
411
431
514
606
638
650
-
-
*-•??•&$
•• •••' •• 8 *
•••-;,!-••: ,*?<*
-
-
416
314
-
2.73
1.13
i
-5
162
-
BP
0.8567
33.6
52.9
45.1
43.9
342
406
426
454
474
492
510
528
547
568
598
624
639
18,1
81,9
-
26.1
8.6
1.1
35.8
86.96
13.07
412
320
19435
45.20
38.73
42.43
36.35
2.72
1.15

-11
169

MOBIL
0.8556
33.8
51 ,3(core)
-
-
338
428
515
605
642
18.79
81.21
-
26.1
8.9
1.4
36.4
87.37
13.42
-
365
19403(core
45.13
38.61
42.28
36.18
2.71
1.12
•.-•.-; ;, \iitg
-17
162
i$3@H
NAVISTAR
$$£14
•^.^sM
51 .6(swri)
46.7
46.2
358
430
454
476
494
512
527
547
' 568
595
647
_
-
vr^f*»*
87.24(swri
13.08{swri
480(swri
310(swri
19381(swri
45.08
38.38
42.30
36.02
2.71
0(swri
-6{swri
154

SHELL
0.8566
33.6
52.9
45.4
44.9
351
413
433
458
478
497
514
532
551
574
604
633
651
_
-
-
-
-
381
281
-
2.74
1.12
-
-
158
":.3£$m
SWRI
0.8565
33.6
52.0
-
-
341
397
429
455
475
494
513
531
549
570
602
628
645
-
-
26.2
poly 7.5
33.7
86.40
13.10
496
310
19427
45.19
38.70
42.4
36.32
2.73
1.12

-6
166
0.553 *
76
PRODUCTS
COMPANY
0.8565
33.7
51.5
45.1
-
353
396
433
459
477
495
513
531
550
572
601
629
650
_
-
26.2
poly 8.4
34.6
-
-
440
306
-
2.72
1.12
5
-1
154
-
AVERAGE
0.8564
33.7
52.4
45.4
45.0
350
405
430
456
476
494
513
530
549
570
602
630
646
18.4
81.6
-
26.2
poly 8.9
35.1
86.99
13.17
438
315
19411
45.15
38.67
42.36
36.27
2.72
1.13

-
161
0.553 "
 * BP infrared method
Data not used in averaging !!!
 " Nalco GC method

-------
        ATTACHMENT 5
Summary of EMA Engine Test Results

-------
 G. F. Keller 03/26/97
Engine Manufacturers Association
03:11 PM Revised Columns B&D

1


% CHANGE
from Baseline
Fuel



NOx










ENGINE MANUFACTURERS ASSOCIATION

I
TEST: Phillips DL 1500 Custom Blend

I
Convention: A negative change indicates a reduction vs. baseline.
| I



v. Baseline














!







1
ENGINE A B C DIE
F
1
! Percent Values ! !
Transient - 3 Hot -4.5 -9.3,' -9.6! -6.7! -11.7
% Sionilicance 99.9 1001 100J 100: 100
AVL t Mode -7.1 -7.9 -8 -9.91
% Significance ' 1 00 ! 1 00
-3.7
97.7
-5.3
99.9
PM iTransient-sHot -28.71 -15.7 1 -4.5! -9.6! -8.3! -30.9
ksiunifiance 1001 991 96.21 99.91 1001 99.9
! AVL 8 Mode -22.51 -9.61
% Siqnilfcance 99.8 '
HC iTr.n$ieni-3Hot -41.61 -17.9! -48.5! -15i -14.81 -70.5
i% Significance 100! 100 99.2 1 99.9! 97.3! 99.9
UvLiMode -37.6 i -9.8 j ! -18! -25.6! -39.7

NOx+NMHC

% Significance 100 1 100! 1001 99.9
T»ntieni - s Hot -8.2 1 -9.6! -14.1 -7.3!
% Significance 100! 100 I 99.9 !
I AVL i Mode -11.31 -7.91 -8.7:
ksignilicarce 100 1 !
-9.7
99.9
-12
99.9
Fuel Specific Emissions ] !
NOx/Gal
Trantient-3Hol
I AVL 1 Mode
PM/Gal

Trintlem - 3 Hot
-0.2 -13.4
-2.2 -12.2
-25.5
AVL a Mode -18.4
HC/Gal {Transient - 3 Hot -39.0
AVL 1 Mode -34.3
NOX+NMHO/GAL
Transient -3 Hot
1 AVL « Mode
Cycle Work {Transient - 3 Hot
1% significance
:AVLSMoOe
\% Significance
Observed Transient - 3 Hot
Fuel Consumption % significance
1 AVL a Mode
: % Signlf tance


-4.1
-6.7
-19.5

-21.6
-14.1
-13.6
-12.3
0.2 1 -0.2
52.8
-0.4

-4.3
97
0.3
99
-8.1 -10.91 -14.31 -7.1
I -13.31 -9.5
-2.9 -13.7! -10.31 -33.4
I
-47.6 -18.9i -17.4 1 -71.6
i -28.4
-12.6 -11.4
-42.4
-7.7
! -15.9
-1.1 -0.1 I -0.3
99.8 100 100
0 0
100 4
-1.11 -1.7 -3 -0.7
99.9 1 100| 96.3 99.5 95.9
-5;0






-0.9 1 4.8 -0.2
1001 100 0


-0.2
99.2


-2.2
97.6
-1.2
99.9


I

'


Averages
-7.6
99.6
-7.6
100.0
-16.3
99.2
-16.1
99.8
-34.7
99.4
-26.1
100.0
-9.8
99.9
-10.0
100.0

-9.0
-9.3
-17.6
-18.4
-36.0
-29.8
-9.9
-11.6
-0.3
91.5
-0.0
67.7
-2.2
98.2
-0.5
75.0
SD
2.9
0.9
1.5
0.0
10.1
1.4
6.5
0.0
20.8
1.0
11.4
0.0
2.3
0.0
1.7
0.0

4.7
4.3
10.0

19.4
10.3
3.5
3.8
0.4
17.3
0.2
45.0
1.2
1.7
3.1
43.3







Transient • 3 Hot
% Significance
AVL a Mode
•i. Significance
Tranttem - 3 Hot
% Significance
AVL B Mode
% Significance
Tranttem -3 Hot
% Significance
AVL a Mode
% Significance
Tranttem -3 Hot
% Significance
AVL a Mode


Transient - 3 Hot
AVL 6 Mode
Tranilent - 3 Hot
AVL 8 Mode
Tr»nt(*m-3Hot
AVL B Mode
Tranitont - 3 Hot
AVL 8 Mode
Trantiem -3 Hot
% Significance
AVL » Mods
% Significance
Trintienl - 3 Hot
% Significance
AVL 8 Mode
% Siqnificanc

NOi



PM



HC



NOi*NMHC




Noi Gat

PMGsl

HC/Gal

(NOX.NMHCJ/Gol

Cycle Wort



Fuel Consumed


k



.
gfk:\fueltest.wk4
           Pagel

-------
 G. F. Keller 03/26/97
Engine Manufacturers Association
03:11 PM Revised Columns B&D









t



ENGINE MANUFACTURERS ASSOCIATION

TEST: Amoc






o Cetane Enhanced v. ^Baseline















! I
% CHANGE Convention: A negative change indicates a reduction v. baseline.
from Baseline
Fuel


ENGINE ! A
!

B ; C


D
I



i ' i ;
E F
! j
' (percent Values
NOx 'Transient -3 Hoi !
6.21 -1.8 -1.9
'% Significance 93 .4 i 97 54.4
'AVL 8 Mode
i% SiamfearEn !
5.3 1 0.4;
100 i

0.4
56

4.4
I 96.6
-0.6 1 4.6
73
PM .Transient -3 Hot -5.5 1 -1.5! 11.3] 0.3
! % Significance £
lAVLsMode
% Significance
3.31 62 i 99.71 44
2.21

0.7
23
HC iTr.nn«it-3H« -23.9! -14.21 -53.5! -9.1
* Significance
100 1001 100
UvLiMoo* -22.3 20.71
•% Significance
NOX+NMHC jranslenl - 3 Hot
100 i
99
99.9
! -3.1
91.9


i -66.3
99.9
-8.3
92
3.1 -2.1 -7.8! -0.1
Insignificance :' 82.1 98 96.9!
AVL a Mode
I % Significance
Fuel Specific Emissions
NOx/Gal Transient . 3 Hot
1 AVL » Mode
PM/Gal

HC/Gal

[NOX+NMHC)/GAL

Cycle Work


Transient - 3 Hot
AVL V Mode
1.51 1.0!
100

8.1 -2 -3.4
6.4! 0.2
3.8! -1.7 9.5
33l

-32.4
99.9
-2.0
82.3
1.1
-2.6
96.1

0.5|

0.4

Tnmii.nt.3Hot -22.5 1 -1 4.4 j -54.31 -9
AVL « MOCM -21.41 20.4
Transient - 3 Hot
AVL 6 Mode
Transient -3 Hoi
% Significance 9
AVL a Mode
i% Significance
Observed
*uel Consumption
5.0 1 -2.4 1 -9.3
2.6 1 0.8!
0.8 1 0.1 -0.6
7.4I 65| 100
0.3 i 0.2 |
1 1001
Transient -3 Hot -3.2 0.2 i 1.6
X Significance i 86.0 90! 91.8
UvLlMode
% Significance
1.1 0.2:
98'
0.1

0
77
0
100
-1.3












4.1
4.6
-3.3

-66.4
-32.4
-2.2
-2.6
-0.1
89.8


i 0.2
95!
4.7
100


43.9
0

i
j • 1 ™"
Averages
1.5
79.5
2.4
91.0
0.3
78.2
1.5
23.0
-33.4
99.8
-10,6
97.3
-1.8
89.8
02
98.1

1.5
3.7
0.2
3.3
-33.3
-11.1
-1.8
0.3
0.0
85.8
-0.0
100.0
-0.5
81.3
1.0
99.0
SD
3.3
19.9
2.6
12.7
5.8
21.5
0.8
0.0
22.5
0.4
20.0
3.7
3.6
7.6
1.7
2.0

4.2
2.6
4.9
0.0
22.8
22.7
4.6
2.1
0.5
13.1
0.2
0.0
1.6
18.9
2.2
1.0

Transient • 3 Hot
% Significance
AVL a Mode
% Significance
Transient - 3 Hot
% Significance
AVL 8 Mode
% Significance
Transient - 3 Hot
% Significance
AVL B Mode
% Significance
Transient -3 Ho!
%Signiteance
AVLlUode


Trantient-JHot
AVL t Mode
Transient -3 Hot
AVL S Mode
TnntlMI-3Hot
AVL (Mode
Transmit • 3 Hot
AVL t Mode
Transient - 3 Hot
% Significance
AVL 1 Mode
% Significance
Transient - 3 Hot
% Significance
AVL 1 Mode
% Significance

NOi



PM



HC



NOx.NMHC




PM/Gal



(NOXtNMHCyGal



Cycle Work



Fuet Consumed



gfk:\fuertest.wk4
            Page 2

-------
           ATTACHMENT 6
SwRI Report on CAT 3176 Engine Test Results

-------
    MEASUREMENT OF GASF^US EMISSIONS
       FROM A CATERPILLAR 3176 ENGINE
  (WITH EGR) USING THREE DIFFERENT FUELS
                     FINAL REPORT
                         Phase I


                 SwRI Project No. 03-8286
                           and
                 SwRI Project No. 08-7601

                         Prepared for:
                 Engine Manufacturers Association
                   401 North Michigan Avenue
                    Chicago, IL 60611 -5267

                           and

                 Environmental Protection Agency
                     2565 Plymouth Road
                     Ann Arbor, Ml 48105

                         Prepared by:
                       Douglas Leone
                     Thomas W. Ryan, III


                         April 10,1997
Approved:
Nigel
Engine
 ice President
Vehicle Research Division
                                 In the event Client distributes any report
                                 issued by SwRI on this Project outside its
                                 own organization, such report shall be used
                                  in its entirety, unless SwRI approves a
                                 summary or abridgement for distribution.
                                                  \
                                                  I

-------
                               EXECUTIVE SUMMARY

       In July of 1995, the Environment Protection Agency (EPA), tne California Air Resources
Board (CARB), and the heavy-duty engine (HDE) manufacturers signed a Statement of Principles
(SOP) outlining a potential regulatory control program for 2004 and later model year HDEs. Central
to that agreement was a 2.4 g/Bhp-hr NMHC+NOX emission standard.  EPA formally proposed this
standard on June 27,  1996.

       The SOP  also included a provision that acknowledged that along with engine hardware
modifications, fuel composition has a significant effect on emissions, and that changes in the quality
and characteristics of diesel fuel may be needed to make the engine technology necessary to meet the
standards feasible and, otherwise, appropriate under the Clean Air Act.  The SOP identifies potential
evaluation of the emission sensitivity of several diesel fuel parameters.

       In response to these provisions, EPA established a HDE Working Group composed of API
and EMA members to work with the EPA in its assessment of the sensitivity of 2004-era diesel engine
technology to diesel fuel quality. The work described in this report forms part of the Phase 1 testing
program designed and conducted by this Group.

       The objective of the work described in this report was  to determine if a CAT 3176 engine
(with EGR) gave the same emissions response  when tested on  3 different diesel fuels as prototype
engines being developed by original equipment manufacturers (OEMs) to meet the 2004 SOP. Results
of the prototype engine  testing are described elsewhere.  Three fuels were tested:  a baseline
commercial No. 2 diesel fuel; a cetane enhanced version of the baseline fuel; an 'extreme' No. 2  diesel
fuel with high cetane and low aromatics. Identical AVL, steady state, 8 mode tests were run on all
fuels to estimate FTP NO,, HC, CO,  and BSFC based on weighted results.  Paniculate matter was
not measured because steady-state generated Pm is a poor predictor of FTP Pm.  The baseline fuel
was run three separate times and the other fuels were run twice each.

       The CAT 3176 engine was run in 1994 OEM configuration, except for the addition of a low
pressure loop exhaust gas recirculation (EGR) system.  EGR is a technology expected on 2004  diesel
engines. The level of EGR at the 8 discrete load/speed points was set to obtain an 8-mode weighted
result of approximately 2.7 g/bhp-hr NO, with the baseline fuel.

       Relative to the baseline fuel, the average 8-mode weighted results were as follows:

       •       The high-cetane, low aromatic fuel reduced NO^ by 7 percent, HCs by 12.7 percent
              and BSFC  by 1.1  percent.

       •       The cetane  enhanced baseline fuel increased NOX 3.4 percent but made no statistically
              significant change in BSFC or HCs.

       The results obtained with the high-cetane, low-aromatic fuel and the cetane enhanced baseline
fuel indicates that cetane number may not be the only factor affecting the NOX emissions from engines
employing this technology.  Future work should include examination of the importance of cetane
number, both naturally occurring as well as through the use of cetane improvers.

                                           iii

-------
                              ACKNOWLEDGEMENTS
             The authors would like to thank Christine Hobbs of Cummins Engine Company for
her help in coordinating this effort. Mike Ross, Mike Wood, and Jerry Chessher provided help in
setup and operation of the engine, test cell, and the emissions carts.  Janet Buckingham of SwRI
provided assistance with the statistical analysis.  Cherian Olikara helped in the initial setup of the
project. Lee Dodge helped edit the report, and Susie Schliesing helped in report preparation. We
would also like to thank Caterpillar for their support in this project and for allowing SwRI the use
of their CAT 3176 engine.
                                          IV

-------
                       TABLE OF CONTENTS

                                                             Page

1.0   BACKGROUND	    1
2.0   OBJECTIVE	    2
3.0   ENGINE CONFIGURATION	    3

3.1   Engine History	    3
3.2   Engine Overview	    3

4.0   EMA TEST PLAN AND EPA WORK ASSIGNMENT 	    4

4.1   Test Cell Modification (EPA Funded) 	    4
4.2   Test Engine Modification (EPA Funded)	    5
4.3   Engine Testing (EMA and EPA Funded)	    5
4.4   Fuel Purge Procedure 	   7

5.0   ENGINE TEST RESULTS AND ANALYSIS	    9
6.0   STATISTICAL ANALYSIS OF EMISSIONS RESULTS	    18
7.0   SUMMARY AND CONCLUSIONS	    22
8.0   RECOMMENDATIONS	    23
9.0   REFERENCES 	    24
APPENDIX A: SUMMARY EMISSIONS DATA FOR FUELS TESTED
APPENDIX B: LOW-SPEED DATA FOR FUELS TESTED
APPENDIX C: SUMMARY HIGH-SPEED DATA
APPENDIX D: HIGH-SPEED PLOTS FOR FUELS TESTED
APPENDIX E: EMA TABLE DATA

-------
                   LIST OF ILLUSTRATIONS

                                                        Page

1  WEIGHTED NOX VERSUS FUEL TYPE CAT 3176	    11

2  WEIGHTED HC VERSUS FUEL TYPE CAT 3176	    12

3  WEIGHTED BSFC VERSUS FUEL TYPE CAT 3176	    13

4  % CHANGE IN NOX FOR CHANGE IN FUEL CAT 3176	    14

5  % CHANGE IN HC FOR CHANGE IN FUEL CAT 3176	    15

6  % CHANGE IN BSFC FOR CHANGE INFUEL CAT 3176	    16

7  HIGH-SPEED DATA RESULTS FOR FUEL EFFECTS
   STUDY ON NOX CAT 3176 	    17

8  AVERAGE NOX BY FUEL	    19

9  AVERAGE HC BY FUEL	    20

10  AVERAGE BSFC BY FUEL  	    21
                            VI

-------
                       LIST OF TABLES




able                                                     Page




 1   LOW-SPEED MEASUREMENTS	    4




 2   CALIBRATION OR VERIFICATION SCHEDULE FOR CELL 07	    5




 3   AVL 8-MODE TEST MATRIX BASED UPON A 600 rpm IDLE SPEED ..    6




 4   ORDER OF AVL 8-MODE TESTING	    6




 5   TOLERANCES FOR CONTROLLED PARAMETERS  	    6




 6   ORDER OF FUELS TESTED  	    7




 7   FUEL ANALYSIS OF THE TEST FUELS 	    8




 8   AVL 8-MODE EGR SCHEDULE 	    9
                            VH

-------
                                 1.0 BACKGROUND
       In July of 1995, the Environment Protection Agency (EPA), the California Air Resources
Board (CARS), and the heavy-duty engine (HDE) manufactures signed a Statement of Principles
(SOP) outlining a potential regulatory control program for 2004 and later model year HDEs. Central
to that agreement was a 2.4 g/Bhp-hr NMHC+NOX emission standard, which is approximately a 50%
reduction in the NOX levels  from the 1998 standard.  EPA formally proposed this standard on
June 27, 1996.

       The SOP also  included a provision that acknowledged that along with engine hardware
modifications, fuel composition has a significant effect on emissions, and that changes in the quality
and characteristics of diesel fuel may be needed to make the engine technology necessary to meet the
standards feasible and, otherwise, appropriate under the Clean Air Act. The SOP identifies potential
evaluation of the emission sensitivity of several diesel fuel parameters.

       In response to these provisions, EPA established a HDE Working Group composed of API
and EMA members to work with the EPA in its assessment of the sensitivity of 2004-era diesel engine
technology to diesel fuel quality. The work described in this report forms part of the Phase 1 testing
program designed and conducted by this Group.

-------
                                  2.0 OBJECTIVE
       The objective of the work described in this report was to determine if a CAT 3176 engine
(with EGR) gave the same emissions response when tested on 3 different diesel fuels as prototype
engines being developed by original equipment manufacturers (OEMs) to meet the 2004 SOP.
Results of the prototype engine testing are described elsewhere.

-------
                           3.0 ENGINE CONFIGURATION
3.1    Engine History

       A Caterpillar 3176 engine was provided to SwRI to support the Clean Heavy-Duty Diesel
Engine (CHDDE) program. The engine was included in the CHDDE program since the Caterpillar
3176 engine incorporated the technology necessary to achieve 1994 on-highway emissions levels.
This specific engine was used throughout the CHDDE program with excellent results.

3.2    Engine Overview

       The Caterpillar 3176 engine is an in-line 6-cylinder engine that incorporated the following
design features:

             Rating of 261 kW (350 hp) @ 1800 RPM
             Peak torque of 1830 N-m (1350 Ib-ft) at 1200 RPM
       •       125 mm x 140 mm (4.92" x 5.51") bore and stroke
       •      Direct injection
       •      Electronically controlled unit injectors
             Mechanically actuated unit injectors
       •      Turbocharged with air to air aftercooling
       •      Uni-flow cylinder head configuration (exhaust and intake ports on same side of
             engine)
             Four (4) valves per cylinder
             Quiescent combustion system
       •      Shallow bowl, Mexican hat designed piston crater
       •      High top ring position
       •      Articulated piston with steel crown and aluminum skirt

SwRI has modified the Caterpillar 3176 engine to incorporate an EGR system designed by SwRJ.
This EGR system incorporates optional EGR cooling and is designed with very long lines. Therefore,
Caterpillar has the opinion that this system is suitable only for steady-state emissions testing. The
long EGR lines will compromise the transient operation of the engine.

       The Caterpillar 3176 engine is also representative of 1998 on-highway emissions levels with
the appropriate fuel injection timing.

-------
               4.0 JMA TEST PLAN AND EPA WORK ASSIGNMENT


       The following work was performed as a part of the project:

4.1    Test Cell Modification (EPA Funded)

       An intake air conditioning system was installed. This system controlled intake air temperature
to 25"C ± 3°C (77°F ± 5°F) and humidity to 75 grains water / Ib dry air ± 5 grains water / Ib dry air.
      Test cell calibration or verification was performed on all low-speed measurement devices. The
low-speed parameters to be measured/recorded are shown in Table 1. Table 2 shows the frequency
of calibration and verification during the engine testing.

                     TABLE 1. LOW-SPEED MEASUREMENTS
Run#
Year
Day
Barometer
Relative Humidity (outside)
Temperature (outside)
LFE in Temperature
Intake air Relative Humidity
Intake Air Dew Point Temp.
Intake Air Humidity ratio
Engine Speed
Torque
Brake Power
SAE Correction Factor
SAE Power
SAE BSFC
SAEBMEP
SAEBTE
A/F Ratio Dry (Mechanical)
Dry Air How
Fuel Flow
Coolant In Temp.
Coolant Out Temp.
Oil Sump Temp.
Fuel Inlet Temp.
Fuel Outlet Temp.
Compressor Out Temp,
Intake Manifold Temp.
Exh. Cylinder 1 Temp.
Exh, Cylinder 2 Temp.
Exh. Cylinder 3 Temp.
Exh. Cylinder 4 Temp.
Exh. Cylinder 5 Temp.
Exh. Cylinder 6 Temp.
Exh. Manifold Front Temp.
Exh. Manifold Back Temp.
Exhaust Stack
EGR Coolant Return Temp.
EGR Inlet Temp.
Oil Gallery Pressure
Fuel Pressure
Intake Manifold Pressure
Exhaust Back Pressure
Compressor Outlet Pressure
Exhaust Manifold Pressure
Compressor In Pressure
Emissions A/F Ratio (Spindt)
Emissions A/F Ratio (EPA)
Bosch Smoke
Carbon Monoxide
Carbon Dioxide
Oxygen
Unburned Hydrocarbons
Background Carbon Dioxide
Intake Carbon Dioxide
EGR











-------
      TABLE 2. CALIBRATION OR VERIFICATION SCHEDULE FOR CELL 07
ParameteL.^
Engine speed
Torque
Pressure transducers
Engine coolant out temperature (verified)
Fuel supply temperature (verified)
Rektive humidity sensor temperature (verified)
LFE temperature sensor (verified)
Cal. Schedule
Every day of testing
Every day of testing
Weekly
Weekly
Weekly
Weekly
Weekly
4.2    Test Engine Modification (EPA Funded)

       The turbocharger was replaced with a new turbo from Caterpillar.  During preliminary testing,
the new turbocharger was damaged by improper engine operation (due to a malfunctioning engine
speed sensor).  Therefore, the original turbocharger was installed and run on the engine, but its
performance was measured and considered acceptable compared to the new turbocharger.

       The EGR cooler was inspected for EGR contamination and its performance was verified for
control of EGR temperature to meet intake manifold temperature specifications. The engine oil and
filter were changed.

       The cylinder pressure and push rod transducers were removed, calibrated and reinstalled. The
dedicated high-speed data acquisition system was connected to the engine and checked out.  A
procedure to transfer high-speed data from the system to a PC for data analysis were determined.
The procedure for calculation of injection pressure on the PC was developed.  The high-speed data
were acquired in 0.5 crank angle degree intervals,  and averaged over 50 engine cycles. The cylinder
pressure data were analyzed by the first-law heat release analysis to estimate apparent heat release
rate. The estimated premised burn fraction was  computed from the heat release rate curve.

4.3    Engine Testing (EMA and EPA Funded)

       The  EGR calibration for the CAT 3176 engine was determined to achieve a transient
equivalent emissions value of 2.5 to 2.8 g/HP-hr NOX with the baseline fuel  The fuel injection timing
strategy was maintained at the 1994 configuration as supplied by Caterpillar. The transient emissions
bvel was estimated  by running an AVL 8-mode steady-state test and weighting the results.  At two
of the test points, the engine measurements were acquired three times to  determine significance of
the data. The 8 AVL test points and weighting factors are provided in Table 3. The AVL 8-mode
tests were run in the order shown in Table 4.  The tests were run while controlling several engine
parameters within specified tolerances. The controlled parameters along with tolerances are provided
in Table 5.

-------
TABLE 3. AVL 8-MODE TEST MA
                       BASED UPON A 600 rpm IDLE SPEED
AVL MODE #
1 (repeated three times)
2
3
4
5
6 (repeated three times)
7
8
% speed
0
11
21
32
100
95
95
89
% load
0
25
63
84
18
40
69
95
Target
rpm
600
732
852
984
1800
1740
1740
1668
Target
ft-lbs
24
175
554
962
184
425
733
1061
WF
0.4168
0.0755
0.0346
0.0398
0.1000
0.1244
0.1215
0.0874
      TABLE 4. ORDER OF AVL 8-MODE TESTING
I AVL Mode*
4 1 3
1 6 (3 times)
7
8
1 3
1 2
1 (3 times) U
TABLE 5. TOLERANCES FOR CONTROLLED PARAMETERS
Intake Restriction (can be checked at rated speed
only)
Exhaust Restriction (can be checked at rated speed
only)
Engine coolant temperature
Fuel supply temperature
Fresh Air temperature
Fresh Air humidity
Barometric pressure throughout day (if nearly done
with testing just note deviation)
Engine Speed
Torque
EGR
Intake Manifold
381 ± 51 mm Hp (15 ± 2 in 1^0)
76±8mmHg(3±0.3inHg)
88±3°C(190±5°F)
40±3°C(104±5°F)
25±3°C(77±5°F)
75 ± 5 grains HjO / Ib of dry air
± 25 mm Hg (± 1 in. Hg) of first day of testing
±2%
±2%
±2% of value
35°C ± 2°C (95 ± 4°F)

-------
       Once the baseline performance was checked, the '. ,?.ine was tested with various fuels over
identical AVL 8-mode tests as shown in Table 6. A totaV  "''_ :;e fuels were tested: the baseline fuel,
the cetane-enhanced baseline fuel and the high-cetane, low aromatic fuel. The baseline fuel was
tested a total of three times, the cetane-enhanced baseline and high-cetane, low-aromatic fuels were
both tested twice.  Fuel analysis for the three fuels is provided in Table 7.

                        TABLE 6.  ORDER OF FUELS TESTED
 Baseline
 High-Cetane, Low Aromatic
 Cetane Enchanced Baseline
 Baseline
 High-Cetane, Low Aromatic
 Enchanced Baseline
 Baseline
4.4    Fuel Purge Procedure

       Between tests on each fuel, a formal procedure was followed and documented for each
change in fuel. This procedure was as follows:

       1.     The fuel lines and day tank were drained.
       2.     The fuel filter was changed.
       3.     The new fuel was connected.
       4.     The engine was run for 10 minutes
       5.     The fuel lines and day tank were drained.
       6.     The fuel filter was replaced.

-------
TABLE 7. FUEL ANALYSIS OF THE TEST FUELS
PARAMETER
DENSITY @ 15 C, g/cm3
API GRAVITY @ 60F, deg. API
CETANE NUMBER
CETANE INDEX
CETANE INDEX
DISTILLATION, F IBP
5%
10%
20%
30%
40%
50%
60%
70%
80%
90%
95%
FBP
NMR, md% AROMATICS
SATURATES
MS, %m PARAFFINS
MONOCYCLOPARAFFINS
DICYCLOPARAFFINS
TRICYCLOPARAFFINS
ALKYL BENZENES
INDANES/TETPAUNS
INDENES
NAPHTHALENE
NAPHTHALENES
ACENAPHTHENES
ACENAPHTHALENES
TRICYCUC AROMATICS
TOTAL AROMATICS
SFC AROMATICS, %m 1-RING
2-RING
3-RtNG
TOTAL
CARBON, %m
HYDROGEN, %m
NITROGEN, ppm
SULFUR, ppm
HOC, gross, btu/lb
MJ/kg
MJ/dm3
HOC, net, MJ/kg
MJ/dm3
VISCOSITY 9 40C, mm2/s
© 100C, mm2/s
CLOUD POINT, F
POUR POINT, F
FLASH POINT, F
ALKYL NITRATE CONTENT, %v
TEST
METHOD
ASTM D4052
ASTM D4052
ASTM D613
ASTM D976 .
ASTM D4737
ASTM D86
ASTM D5292
ASTM D2425
ASTM D5186
ASTM D5291
ASTM D5291
ASTM D4629
ASTM D2622
ASTM D4809
ASTM D445
ASTM D2500
ASTM D97
ASTM D93
Nalco GC
BASELINE
FUEL
0.8564
33.6
45.9
45.0
44.4
365
408
432
456
476
495
513
530
550
571
601
628
649
18.1
81.9
34.33
20.35
5.84
3.42
14.06
11.07
3.34
0.81
2.46
2.24
1.20
0.83
36.06
26.6
poly 9.1
35.7
87.11
13.06
85
310
19460
45.26
38.76
42.49
36.39
2.74
1.13
0
-5
160
0.047
CETANE R '.NCED
BASE? ,EL
0.8564
33.7
52.4
45.4
45.0
350
405
430
456
476
494
513
530
549
570
602
630
646
18.4
81.6
•
26.2
poly 8.9
35.1
86.99
13.17
438
315
19411
45.15
38.67
42.33
36.27
2.72
1.13
2
-9
161
0.553
HIGH-CETANE, LOW
AROMATIC FUEL
0.8233
40.3
56.9
56.2
59.3
397
436
454
472
487
499
511
522
536
553
584
612
630
9.8
90.2
60.9
17.57
5.9
2.81
5.02
3.7
1.2
0.2
0.9
0.7
0.5
0.5
12.84
1S.S
poty4.5
20.0
86.09
13.84
28
180
19817
46.10
37.95
43.16
35.53
2.72
1.15
8
4
181
< 1ppm

-------
                    5.0 ENGINE TEST RESULTS AND ^ANALYSIS
       The first step in performing the testing was to determine the appropriate levels of EGR
required to produce a weighted AVL 8-Mode (transient equivalent) NOX emissions level of 2.5 to 2,8
g/HP-hr with the baseline fuel. The EGR schedule shown in Table 8 resulted in a weighted 8-Mode
result of 2.68 g/Bhp-hr NO^, and was selected based on previous experience at SwRI with this engine
configuration. The 2.8 gm/Bhp-hr level was achieved using a test fuel available at SwRI.  During the
two repeat tests of the baseline fuel, the weighted 8-Mode results were 2.75 and 2.72 g/HP-hr NOV

                      TABLE 8.  AVL 8 MODE EGR SCHEDULE
AVL Mode #
1
2
3
4
5
6
7
8
EGR %
50%
12.5%
0%
0%
20%
16%
6%
4%
       A summary of the weighted emissions results for all fuels tested is provided in Figures 1, 2,
and 3. Summary emissions data for individual modes is provided in Appendix A for all tests.  Low-
speed data is provided in Appendix B.  Summary modal high-speed data are provided in Appendix
C and the graphs in Appendix D. Appendix E is a compilation of the modal emissions data in EM A
specified format.

       Percentage change of the weighted NO,, unbumed HC and BSFC for the change in fuel
compared to the baseline fuel is shown in Figures 4, 5, and 6. The high-cetane, low-aromatic fuel
decreased the NOX approximately 7 percent, while the cetane-enhanced baseline fuel increased NOX
approximately 3.4 percent.  The high-cetane, low aromatic fuel decreased the HCs approximately 13
percent, whib the cetane-enhanced baseline fuel had a negligible effect on the HCs. The high-cetane,
low-aromatic fuel decreased the BSFC approximately 1.1 percent, while the cetane-enhanced baseline
fuel increased the BSFC approximately 0.4 percent.

       The increase in NO, noted for the cetane enhanced baseline fuel relative to the baseline fuel
was unexpected.  Previously published data ^ suggests that increased cetane number tends to
reduce gaseous emissions, including NOr Effects on Pm are more variable, with a number of studies
showing little or no sensitivity of Pm emissions to increased cetane number.  Several of these (notably

-------
SAE 950251) and other studies, have compared the effect of natural versus boosted cetane number
on emissions, and concluded that the effects of increased cetane number achieved either thr •; ;.i
changes in base fuel composition or through additives (alkyl nitrates, peroxides, etc.) are equivalent.
There is little or no data in the published literature, which suggests that either fuel bound nitrogen or
nitrogen contained in alkyl-nitrate ignition improvers make a significant contribution to NOX emissions
from diesel engines. However, it should be noted that these observations derive from assessment of
fuel effects on HD diesel engines (without EGR) calibrated to meet US 1998 or earlier emissions
standards. The influence of cetane (natural or boosted) on engines designed to meet US 2004
emissions standards, which like the engine(s) used in the Phase 1 tests are likely to be equipped with
EGR, has not been fully evaluated.  Therefore, it is recommended that tests conducted in Phase 2 of
this study include a thorough evaluation of cetane effects on gaseous emissions.

       The increase in cetane number for the cetane-enhanced fuel and the high-cetane, low-aromatic
fuel was expected to decrease the ignition delay and, therefore, reduce the premix burn fraction. The
premix burn produces a significant amount of the NOX and reduction in this burn should reduce NOX
emissions.  Shown in Figure 7 is the premix bum fraction computed for all the fuels. It is apparent
from this figure that the premix burn fraction for both the cetane-enhanced fuel and the  high-cetane,
low-aromatic fuel was reduced compared to the baseline fuel.  It is also clear that the  high-cetane,
low-aromatic fuel resulted in consistently lower premix burn fraction than the cetane-enhanced fuel.
                                            10

-------
u.uuu
OA*^n
.fr\J\J
OAC\C\

Oocri
. oijvj
^ n onn
^i U.oUU
Q.
g
. ^ n *5no
Q U.^UU
u.
(/) n "i en
QJ U. I3U
Omn
. I UU
On^n
.uou
Onnn
.uuu

































)77'
.Of*


















UJ
z
BASEL




? r
1 V















i
i







)*V7'
.O f Si

' - '

,
' "
'


,

"•- -'
^

; •;
-/



CM
UJ
Z
BASEL






























1 77'
/.Of*.









•, ,-, •.
'j^.
-
• ••

f t



eo
U
Z
BASEL




i r
5 C















t
i

u
2
<
U
5
C
3



\ OC(
);36(
-
,



•i ••
' ', :

'• - ^

, ,
^
'• %
^
,
:-' ,

"^^
I
O
PF
ei
\ °
>5
IS
: Li



> r
5 C















I



.
•
1
•




)occ
.36c















J

^\J
o
UJ 1-
IGH-CETAN
OW AROMA
r _i



Jr
i-















,








)-37^


/

*
'• ., -
•j


J.
*
'

\"
! r

i ,

^^
Q
LU
O
X
UJ
UJ
z
h-



l f
J v















I
!















-









i
Ol
O
UJ
o
\NE ENHAN
i—


























                  FUEL TYPE
                                               UJ
                                               O
LJ
O
FIGURE 1. WEIGHTED NOX VERSUS FUEL TYPE CAT 3176

-------
K)
U.vJV
Ooc
.^Ivl
09n
.zu
.c
A n 15 -
"J- U. 1  j" s "• •.


^
- - j \ ""
,

i i "
T- CM « •
m UJ LU O
±j -^ 2 . r r~
Z ± £ w \-
ui m iil i!
< I 2 g|
II
X -J








n 19^
U. I £.*J


' .-,
•= -:
- ^
>•_,_,
' '••
'
.-; .


CM
O
UJ H
Z <
< 5
H 0
s§
11
X -1







0.141


- - ';*
L-;
«|;
:ig;
' '''%},
o ""/••
;"Js"*
^ •"!•;
,; ^'
^5,}f-

T~
Q
LU
O
X
LU
LU
Z
i—
LU
O





0.141





""
'-
^ •
t




CM
Q
LU
0
X
LU
UJ
Z
•
ID
0





I





















                                              FUEL TYPE
                            FIGURE 2. WEIGHTED HC VERSUS FUEL TYPE CAT 3176

-------
3.00



2.50



2.00



1.



1.1



0.50



0.00









268
'















1
2.75




•
•
-
,








— i —
2.72



,

, j

•• ••








i


.51
-
•*
»
',
; '
> >








i

f\ cc



-
••.

I'









£.OU

i
;%
- ;

-
••f
'••









	 1 	
rroa

















           LU
           111
           CO
           <
           00
CM

LU
UJ
(0
<
ffl
CO           •
UJ          O
5        UJ P
_l        Z <
tu        < S
w        P o
<        LU or
m        oS
                                         = O
                                   FUEL TYPE
   (M
   i
   O
LU P
2 ^
< 2

LU §
9<


I!
O
HI
O

<

Z
UJ
UJ
                                                                 ti
                                                                 O
Q
LU
O
X
Z
LU
LU


I
LU
O
                FIGURE 3. WEIGHTED BSFC VERSUS FUEL TYPE CAT 3176

-------
UJ
O
o
  4.00


  2.00


  0.00


 -2.00


 -4.00


 -6.00


 -8.00


-10.00
                  HIGH-CETANE

                  LOW AROMATIC
                       -6.99
                                                CETANE ENHANCED
                                                        3.44
                 FIGURE 4. % CHANGE IN NOX FOR CHANGE IN FUEL CAT 3176

-------
UJ
o
<
o
  0.00
 -2.50
 -5.00
 -7.50
-10.00
-12.50
   .00
                HIGH-CETANE,
                LOW AROMATIC
                                                 CETANE ENHANCED
                                                    -0.28
                    -12.66
             FIGURE 5. % CHANGE IN HC FOR CHANGE IN FUEL CAT 3176

-------
      HIGH-CETANE,
      LOW AROMATIC
CETANE ENHANCED




LU

Z
X
o
 'f<'\

-1.08



FIGURE 6. % CHANGE IN BSFC FOR CHANGE IN FUEL CAT 3176

-------
  0.800
  0.700
D Baseline 1
  Baseline 2
          3
• Cetane Enh. 1
• Cetane Enh. 2
H Hi CN 1
  HICN2
o 0.500
2
                                    6783
                                     AVL Mode Number
        FIGURE?. HIGH-SPEED DATA RESULTS FOR FUEL EFFECTS STUDY ON NOX CAT 3176

-------
                6.0  STATISTICAL AN     SIS OF EMISSION RESULTS
       In order to compare the performance of the three tested fuels on emissions and BSFC, an
analysis of variance (ANOVA) statistical procedure was performed.  The ANOVA compares the
average weighted NOX, HC and BSFC between the three fuels to determine if they are significantly
different.  A total of seven tests were run:  three on the baseline fuel, two on the high-cetane, low-
aromatic fuel and two on the cetane enhanced baseline fuel. The ANOVA used fuel type as the only
independent factor.

       The results of the ANOVA for NO, indicate that there is a significant difference (p-value =
0.0013) in the average NOX among all three of the fuels tested.  P-values less than 0.05 demonstrate
a significant difference in the averages in at least two of the fuels. A post-hoc multiple comparison
procedure was then performed to determine which averages were  significantly different from the
others. The method used in this analysis was Fisher's least significant difference (LSD) procedure.
The results of the NO^ analysis are shown graphically in Figure 8. The average NO x is plotted for
each of the three fuels along with its 95 percent LSD confidence interval. If two of the averages are
statistically the same, their intervals will overlap 95 percent of the time., As can be seen in Figure 8,
all three fuels are significantly different from one another with respect to their average NOX. The
cetane enhanced baseline fuel had higher average NOX than the baseline, whereas the high-cetane,
low-aromatic fuel had  a lower average NOX.

       ANOVA results  for HC and BSFC were similar and are illustrated in Figures 9 and 10,
respectively. There was a significant difference in the average HC (p-value = 0.0467) and BSFC (p-
value = 0.0015) between the high-cetane, low-aromatic and the baseline fuels, but not between the
cetane enhanced baseline and the baseline fuel The 95 percent LSD intervals shown in Figures 9 and
10 demonstrate the statistically lower average HC and  BSFC for the high-cetane, low-aromatic fuel.
                                          18

-------
     2.9
              95% LSD CONFIDED   INTERVALS
     2.8
£    2.7
Q.
OQ
E
&   2.6
 x
O
     2.5
     2.4
                  J	1
                    HIGH-CETANE
                   LOW AROMATIC
          BASELINE
 CETANE
ENHANCED
                       FUEL
             FIGURES. AVERAGE NO* BY FUEL
                        19

-------
            95% ISO CONFIDENCE IN    ,VALS
    0.15
    0.14
S3
ut
O
0.13
    0.12
    0.11
                  i        r
                  I    	I
                    HIGH-CETANE
                   LOW AROMATIC
           BASELINE
                                CETANE
                               ENHANCED
                       FUEL
             FIGURE 9. AVERAGE HC BY FUEL
                        20

-------
   0.375
   0.373
u.  0.371
CO
ffi
   0.369
   0.367
             95% ISO CONFIDENCE INTERVALS
                    HIGH-CETANE

                   LOW AROMATIC
            BASELINE
                        FUEL
 CETANE

ENHANCED
              FIGURE 10. AVERAGE BSFC BY FUEL
                          21

-------
                         7.0  SUMMARY AND CONCLUSIONS


       The Caterpillar 3176 was utilized as a representative engine of future diesel engine technology
to test three fuels: a baseline fuel, a octane-enhanced baseline fuel, and a high-cetane, low-aromatic
fuel The engine was calibrated with a 1994-year fuel injection timing strategy. A low-pressure EGR
system was calibrated to provide an AVL 8-mode weighted (transient  emission predicted) of 2.7
g/hp-hr NOX.  An intake air conditioning system was installed to control intake air temperature and
humidity.

       The following conclusions were reached from the data acquired and  analyzed during the
performance of the project:

For high-cetane, low-aromatic fuel:

       •       Reduced NO^ 7 percent, reduced BSFC 1.1 percent, and reduced HCs 12.7 percent
              compared to the baseline fuel

       •       All changes are considered significant based upon the statistical analysis performed

       •       Reduced the measured p remix burn fraction compared to the baseline fuel

For the cetane-enhanced fuel:
              Increased NO^ 3.4 percent  compared to the baseline  fuel,  an increase that is
              considered statistically significant

              Had no significant impact on HCs or BSFC compared to the baseline fuel

              Reduced the premix bum faction compared to the baseline fuel
                                           22

-------
                               8.0  RECOMMENDATIONS
       The results obtained with the high-cetane, low-aromatic fuel and the cetane enhanced baseline
fuel indicates that cetane number is not the only fuel property that affects the NOX emissions from
advanced technology engines.  It is likely that this lack, or at least significant reduction, of the
sensitivity to cetane number is due to the fact that these engines have significantly reduced premixed
burn fraction relative to the old technology engines. Future work should include examination of the
effects of cetane number, both natural and additized, on the emissions from this type of engine.
                                            23

-------
                                 9.0 REFERENCES
1.      UUman, Terry L., "Investigation of the Effects of Fuel Composition on Heavy-Duty Diesel
       Engine Emissions," SAE Paper No. 892072, SAE International Fuels and Lubricants Meeting
       and Exposition, Baltimore, MA, Sept. 25-28, 1989.

2.      Ullman, Terry L., Robert L. Mason, and Daniel A. Montalvo, "Effects of Fuel Aromatics,
       Cetane  Number, and Cetane Improver on Emissions from a 1991 Prototype Heavy-Duty
       Diesel Engine," SAE Paper No. 902171, SAE International Fuels and Lubricants Meeting and
       Exposition, Tulsa, OK, Oct. 22-25, 1990.

3.      Ullman, Terry L., Kent B. Spreen, and Robert L. Mason, "Effects of Cetane Number, Cetane
       Improver, Aromatics, and Oxygenates on 1994 Heavy-Duty Diesel Engine Emissions," SAE
       Paper No. 941020, SAE International Congress & Exposition, Detroit, MI, Feb. 28 - March
       3, 1994.

4.      Spreen, Kent B., Terry L. Ullman, and Robert L.  Mason,. "Effects of Cetane Number,
       Aromatics, and Oxygenates on Emissions From a 1994 Heavy-Duty Diesel Engine with
       Exhaust Catalyst," SAE Paper No.  950250, SAE International Congress & Exposition,
       Detroit, MI, February 27 - March 2,1995.

5.      Ullman, Terry L., Kent B. Spreen, and Robert L. Mason, "Effects of Cetane Number on
       Emissions from a Prototype 1998 Heavy-Duty Diesel Engine," SAE Paper No. 950251, SAE
       International Congress & Exposition, Detroit, MI, February 27 - March 2, 1995.

6.      Ryan IE, Thomas W., Jirnefl Erwin, Robert L. Mason, and David S. Moulton," Relationships
       Between Fuel Properties and Composition and Diesel Engine Combustion Performance and
       Emissions," SAE Paper No. 941018, SAE International Congress & Exposition, Detroit, MI,
       Feb. 28 - March 3, 1994.
                                         24

-------