United States Air and Radiation EPA420-R-96-001
Environmental Protection November 1996
Agency
Office of Mobile Sources
vvEPA Exhaust Emission
Testing of Orbital
50cc Direct Injection
Two-Stroke Engine
-------
Publication No. EPA420-R-96-001
November 1996
Exhaust Emission Testing of Orbital 50cc
Direct Injection Two-Stroke Engine
Test Engineers
William Charmley, Engine Programs and Compliance Division
Carl Ryan, Test Service Division
Engine Test Technicians
Toni Bejma, Test Service Division
Arron Bohlke, Test Service Division
Engine Programs and Compliance Division
Office of Mobile Sources
Office of Air and Radiation
U.S. Environmental Protection Agency
NOTICE
Technical Reports do not necessarily represent final EPA decisions or positions.
The purpose in the release of such reports is to facilitate the exchange of
technical information and to inform the public of technical developments which
may form the basis for a final EPA decision, position, or regulatory action.
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
NATIONAL/VEHICLE AND FUEL EMISSIONS LABORATORY
2565 PLYMOUTH ROAD
ANN ARBOR, MICHIGAN 48105
OFFICE OF
AIR AND RADIATION
MEMORANDUM
SUBJECT: Exemption from Administrative Review
FROM: William Charmley, Mechanical Engineer
Nonroad Engine Programs Group
THRU: Paul Machiele, Manager
Nonroad Engine Programs G
TO: Chester France, Director
Engine Programs and Compliance Division
The attached report entitled "Exhaust Emission Testing of
Orbital 50cc Direct Injection Two-Stroke Engine" contains
emission results from a Piaggio 50cc two-stroke engine modified
by Orbital Engine Corporation to have a stratified charge
combustion process. The engine was tested using the Federal Test
Procedure for Small Spark- ignited Engines <19kW, utilizing Cycle
C ("handheld" engine cycle) .
The data from the direct injection (DI) two-stroke engine
shows that exhaust emissions of carbon monoxide and hydrocarbons
are substantially lower when compared to existing single-cylinder
four-stroke and charge scavenged two-stroke engines. The results
also/ indicate^ oxides of nitrogen emissions from the DI engine was
typical of exist ing""fouK- stroke engines but 2 to . 3 times higher
than existing charge scavenged two-stroke engines.
Since this report presents only factual emission data and
noncont rovers ial conclusions, and I would like to make copies
publicly available, your concurrence,' is requested to waive
administrative review according to the policj outline in 'the
directive of April 22, 1982.
Concurrence :
Nonconcurrence :
Attachment
Printed on Recycled Paper
-------
Table of Contents
1. Abstract 3
2. Introduction 3
3. Test Engine Description 3
4. Test Procedure/Set-up/Test Fuel 4
5. Test Results and Di'scussion 5
6. Conclusions 7
7. References 9
Appendix A - Modal Test Results 10
Appendix B - Data Graphs 11
Final Version, October, 1996
ORIBTAL2-WPD
-------
1. Abstract
This report documents the results of exhaust emission testing
from a single cylinder 50cc direct injection two-stroke spark-
ignited engine. The test engine was provided by Orbital Engine
Corp. The engine was tested using the EPA Phase 1 Small Nonroad
Spark-Ignited Engine testing protocol for Class 4 engines. Three
complete 2-mode steady state cycles were run, with emissions of
THC, NOx, CO, and CH4 measured. . The results of the testing are
.summarized in Table 1.
Weighted
Power (kW)
1.9
Weighted THG
(g/kW-hr)
1.5
Weighted NOx
(g/kW-hr)
3.4
Weighted CO
(g/kW-hr)
35.6
Weighted CH4
(g/kW-hr)
0.04
Table 1. Summary of 50cc DI Two-Stroke Emission Results
2. Introduction
EPA has several nonroad engine programs being developed which are
looking at various possible levels of emission controls. One
such area involves spark-ignited engines less than 19kW. EPA
arranged for the testing of a direct-injection two-stroke engine
at our facilities with Orbital Engine Corporation for the purpose
of determining potential emission levels from this technology.
This report documents how the emission tests were performed and
provides the emission results.
3. Test Engine Description
The test engine was a 50cc direct injection two-stroke designed
for use as a motor scooter engine. The engine was a basic
production Piaggio engine modified by Orbital to have a
stratified charger-combustion process with Orbital's Small Engine
Fuel Injection System (SEFIS). SEFIS is a direct-injection
process which eliminates the charge scavenging utilized by
conventional two-stroke engines (1),(2), (3). The engine was
mounted in a Piaggio motor scooter frame. The specifications for
the engine are summarized in Table 2.
Specification
Bore
Stroke
Displacement
Compression Ratio
Dimensions
40 .Omm
39.3mm
49.4cc
10.9:1 (Swept
Volume)
Final Version, October, 1996
ORIBTAL2.WPD
-------
Specification
Combustion Chamber
Design
Lubrication
Fuel Metering
Fuel Delivery
Ignition System
Electronic Control
Exhau s t Sy s t em
Rated Speed (rpm)
Rated Power (kW)
Electrical System
Dimensions
Containment design for
stratified combustion
Mechanical oil pump
Crankcase pressure driven Fuel
Metering Pump
Orbital SEFIS direct injection
nozzle
Capacitive Discharge Ignition
SEFIS single cylinder ECU
Standard Piaggio moped system
(de-tuned)
7000
2.1 (with de-tuned exhaust
system)
SEFIS interfaced with standard
Piaggio moped system and
battery
Table 2. Orbital 50cc Direct Injection Two-Stroke Engine
Specifications
4. Test Procedure/Set-up/Test Fuel
Test Procedure
The test procedures documented in 40CFR Part 90, Subparts D & E
were followed for^this test program. The test cycle followed was
the steady-state Cycle C found in 40 CFR §90.410. Cycle C is a
two mode test; mode 1 is maximum power, wide open throttle at
rated speed and mode 2 is a closed throttle idle. Cycle C is
identical to the two-mode cycle described in the Society of
Automotive Engineers Recommended Practice J1088 Cycle C. The
following parameters were monitored and recorded during the test
cycle; engine speed, engine torque, engine inlet air temperature,
cylinder head temperature, fuel temperature, exhaust gas
temperature, test cell temperature, test cell dew point, and test
cell barometric pressure. All of these measurements were
recorded as a iHz sampling rate during both the pre-test engine
warm-up and during the test modes.
Engine exhaust emissions were collected using a Constant Volume
Sampling (CVS) system utilizing a critical flow venturi with a
Final Version, October,,1996
ORIBTAL2.WPD
-------
nominal flow rate of 0.071 mVsec. (150 ftVmin.). Dilute
exhaust samples for total hydrocarbons (THC) were measured using
a continuous.heated flame ionization detector (H-FID) and dilute
oxides of nitrogen (NOx) were measured using a continuous heated
chemiluminescence analyzer (H-CLA). The continuous sample
readings were recorded at a iHz sample rate and averaged for each
mode. Carbon monoxide (CO), carbon dioxide (C02) , methane
(CH4) , and background samples for all constituents were
collected in tedlar sample bags. CO and C02 samples and
backgrounds were measured with non-dispersive infrared analyzers,
CH4 samples and backgrounds were measured with a processed gas
chromatograph methane analyzer, and background samples for THC
and NOx were analyzed with cold FID and CLA instruments
respectively.
Engine Set-up
The Orbital engine was tested at EPA's National Vehicle and Fuel
Emissions Laboratory in the Small Engine Test Cell. The engine
was coupled to an eddy current dynamometer used to control engine
speed. Engine throttle^position was set with a hand actuated
throttle controller. The dynamometer used was a Midwest Model 46
with a Dyne-Systems Dyne-Loc IV controller. A Morse Morflex 502-
R coupling was used to connect the engine crankshaft to the
dynamometer rotor. • The test engine was mounted on a machined
platform rigidly attached tjo the dynamometer test stand.
Test Fuel
(
The testing was performed using a certification quality test fuel
meeting the requirements pf. 40CFR86 .113-94 (a) . A high grade,
commercially available two-stroke lubrication oil was used for
all tests.
5. Test Results and Discussion
The emission results in this report resulted from three repeat
tests performed on the same day. The engine experienced no
difficulty during these tests. However, one complete test and
one partial test were attempted the previous day (8/29/95) .
During the second test of the first testing day the engines
electronic system lost power due to a malfunctioning battery.
Both tests run on August 29 were voided (EPA Test #95-8288 and
95-8289), the battery was replaced over night, and.the three
tests reported here were run the following day without problems.
The weighted emission results are summarized in Table 3, complete
modal results for all three tests are contained in Appendix A.
Final Version, October, 1996
ORIBTAL2.WPD
-------
Test Number
Total Hydrocarbons (g/kw-hr)
Methane (g/kW-hr)
Oxides of Nitrogen (g/kW-hr)
Carbon Monoxide (g/kW-hr)
Carbon Dioxide (g/kW-hr)
Weighted Power (kW)
95-8290
2.14
0.04
3.27
38.5
1,581
1.88
95-8291
1.29
0.04
3.36
36.3
1,608
1.89
95-8292
1.01
0.04
3.43
31.9
1,597
1.89
Average
1.48
0.04
3.35
35.6
1,595
1.89
Std. Dev.
0.59
0.00
0.08
3.4
14
0.01
Table 3. Weighted Emission Results for Individual Tests
The total engine run-time for the completion of these three tests
was approximately 65 minutes. At the completion of the tests an
initial review of the raw data indicated three successful tests
had -been completed. The Agency testing schedule, combined with
the Orbital staff timing Constraints, necessitated the removal of
the test engine from the test bed and a return of the engine to
Orbital.
One important trend that, later analysis of the data showed was
the substantial decrease in THC emissions and a large decrease in
CO emissions from test 95-8290 ("test 90") to test 95-8291 ("test
91"). Weighted THC emissions from test 91 and test 92 were 40%
and 53% lower than from t'est 90, respectively, and weighted CO
emissions decreased 6% and\17% from test 90. Changes in weighted
methane, NOx, carbon dioxide and power were all less than 5%.
Though additional testing could not be performed to confirm it,
the most reasonable explanation for the change in THC and CO
emissions is post-oxidatioi^ in the muffler. As indicated by the
very low CO concentration levels in Appendix A (all modes <0.5%
CO) the Orbital engine operates at lean air-fuel ratio, leaner ,
than stoichiometry, therefore ample oxygen exists in the exhaust
gas for post-oxidation. As noted in Table 2, the muffler
configuration used for the
-------
Parameter
Spark Plug Seat
Temp. (C) --
Exhaust Gas
Temp . ( C )
Muffler Surface
Temp. (C)
Test 95-8290
140
495
136
Test 95-8291
141
501
140
Test 95-8292
140
510
145
Table 4. Mode 1 Average Temperatures
Appendix B contains graphs for each test of continuous THC, NOx,
torque, exhaust gas temperature, muffler skin temperature, and
spark-plug seat temperature. Figures 1 and 2 correspond to test
95-8290, figures 3 and 4 to test 95-8291, and figures 5 and 6 to
test 95-8292. The "Mode Flag" trace in each graph indicates when
data is.being sampled during stabilization periods and during
modal periods. When mode flag equals zero, the engine is either
stabilizing or changing -speed/torque conditions, mode flag equal
to one indicates data is being recorded for mode l,and mode flag
equal to two indicates data is being recorded for mode 2.
An examination of each figure shows the same general pattern with
respect to stabilization o,f temperatures, THC, and NOx emissions.
Spark-plug seat temperature stabilizes quickly and remains
essentially constant during each mode. Muffler skin temperature
rises steadily during Mode 1 and drops steadily during mode 2,
with the same trend occurring but at a larger temperature
gradient for exhaust gas temperature. During Mode 1, THC
concentration decreases significantly throughout the mode, while
NOx emissions increase by a few ppm. The difference between the
three tests appears to be the continued heating up of the •
exhaust/muffler system. The three tests were essentially run
back-to-back. Though the exhaust system was not temperature
stabilized for any of the tests, the exhaust system started at a
higher temperature for each consecutive test. Post-oxidation is
one plausible explanation for the observed trends of lower THC
and CO emissions from consecutive tests.
The un-stabilized exhaust/muffler system condition was identified
too late to pursue additional testing to examine the true steady-
state emission results from the Orbital engine.
6. Conclusions
The SEFIS engine performed well during all three emission tests.
Engine speed and torque remained steady during all emission
modes The emission performance of the Orbital SEFIS system
demonstrated levels of THC and CO substantially below many
current technology single cylinder spark-ignition engines, both
Final Version, October, 1996
ORIBTAL2-WPD
-------
four-stroke and crankcase charge scavenged two-stroke engines.
NOx emission levels were comparable to current four-stroke engine
levels, but 2 to 3 times higher than existing charge scavenged
two-stroke designs. Test data indicated possible post-oxidation
of emissions which may have resulted in lower THC and CO values
if true steady state tests had been performed.
Final Version, October, 1996
ORIBTAL2.WPD
-------
7. References
1. S. Leighton, S. Ahern, "The Orbital Small Engine Fuel
Injection System (SEFIS) for Direct Injected Two Stroke
Cycle Engines." 5th Graz Two Wheeler Symposium, 1993, pages
28-38.
2. S. Leighton, S. Ahern, M. Southern, M. Cebis, "The
Orbital Combustion Process for Future Small Two-Stroke
Engines." IFF International Seminar, Paris, November 1993.
3. S. Leighton, M. Cebis, M. Southern, S. Ahern, and L.
Horner, "The OCP Small Engine Fuel Injection System for
Future Two-Stroke Marine Engines." Society of Automotive
Engineers Paper 941687, 1994.
Final Version, October, 1996
ORIBTAL2.WPD
-------
Appendix A - Modal Test Results
Parameter
Test Number
Engine Speed (rpm)
Engine Torque (N-m)
Test Cell Ambient Temp. (C)-
Engine Air Inlet Temp. (C)
Engine Exhaust Gas Temp. (C)
Muffler Surface Temp. (C)
Fuel Temp. (C)
Spark Plug Seat Temp. (C)
Dew Point Temperature (C)
Barometric Pressure (in Hg).
Raw Exhaust Concentrations
Total Hydrocarbons (ppmC)
Methane (ppmC)
Oxides of Nitrogen (ppm)
Carbon Monoxide (%)
Carbon Dioxide (%)
Exhaust Mass Flow Rates
Total Hydrocarbons (g/hr)
Methane (g/hr)
Oxides of Nitrogen (g/hr)
Carbon Monoxide (g/hr)
Carbon Dioxide (g/hr)
Calculated Fuel Flow (kg/hr)
CVS Dilution Ratio
Mode 1
95-8290
7,001
2.9
22.2
21.6
495
136
21.2
140
10.8
29.2
472
6.41
259
0.49%
12.5%
t
(4.0
Os 08
6.55
79.9
3,280
1.08
10.3
Mode 2
95-8290
1,850
0.2
21.8
21.5
166
85.5
21.1
114
10.7
29.2
5245
23.9
993
0.33%
8.6%
4.2
0.02
2.46
4.4
220
0.08
109
Mode 1
95-8291
7,001
2.9
22.0
21.5
501
140
20.9
141
10.8
29.2
254
6.66
264
0.45%
12.5%
2.2
0.08
6.81
75.5
3,348
1.09
10.2
Mode 2
95-8291
1,851
0.1
21.9
21.5
181
91.3
21.0
110
10.8
29.1
5821
28.1
888
0.34%
8.7%
4.5
0.03
2.13
4.9
214
0.07
112
Mode 1
95-8292
7,001
2.9
22.0
21.6
510
145
20.8
140
10.9
29.1
172
7.05
274
0.39%
12.6%
1.5
0.09
7.0
66.2
3,327
1.08
10.4
Mode 2
95-3292
1, 850
0 . 1
22.0
21.6
176
90.1
21.0
105
11.1
29.1
7364
32.8
766
0.35%
. 8.2%
5.5
0.03
1.77
5.4
196
0.07
117
Weighted Results
Test Number
Total Hydrocarbons (g/kW-hr)
Methane (g/kW-hr)
Oxides of Nitrogen (g/kW-hr)
Carbon Monoxide (g/kW-hr)
Carbon Dioxide (g/kW-hr)
Weighted Power (kW)
95-8290
2.14
0.04
3.27
38.5
1,581
1.88
95-8291
1.29
0.04
3.36
36.3
1,608
1.89
95-8292
1.01
0.04
3.43
31.9
1,597
1.89
Average
1.48
0.04
3.35
35.6
1,595
1.89
Final Version, October, 1996
ORIBTAL2.WPD
10
-------
Appendix B - Data Graphs
Final Version,1 Occober, 1996 .
ORIBTAL2.WPD 11
-------
FIGURE 1
Orbital 50cc Dl Engine - Test # 95-8290
800
900
1000
1100
1200
Time (sees)
1300
1400
1500
Mode Rag
"Torque (N-m) THC(ppmC)* • • NOx (ppm)
1600
FIGURE 2
600
800
Orbital 50cc Dl Engine - Test # 95-8290
900
1000
1100
1200
Time (sec)
1300
1400
1500
•Sparkplug Seat (C)
• Exhaust Gas (C)
•Muffler Surface (C)
•Mode Rag
1600
958290.XLS, Graphs
10/22/96
-------
FIGURE 3
Orbital 50cc 01 Engine - Test # 95-8291
Time (sees)
•Mode Rag '
•Torque (N-m)
-THC(ppmC)- " • NOx(ppm)
FIGURE 4
600
Orbital 50cc Dl Engine - Test # 95-8291
200 400 600 800
Time (sec)
1000
1200
1400
•SparKplug Seat Head (C)
- Exhaust Gas (C)
~ Muffler Surface (C)
^ Mode Rag
958291.XLS, Graphs
10/22/96
-------
FIGURES
5-
4.5-
4 •
r 35 -
o
"S 3-
o
~ 25 •
§
2" 1 5 -
5
n e .
.~~ A
,
1
1
1
JUAJM^tf
— (UWBOP*""
I
1
1
•^
1
.4
Orbital 50cc 01 Engine - Test # 95-8292
\.
Vi
— T"*
\
V
f ii M "^
0 100
^^^^^^^^— i^^^
•^^•^^^^^
\,
200
•^^~Mod(
-^
MMMtf*
1
300
a Flag
^V«
^*>»
i
\ I
*\ /
*\ f
*• ^*^
i
^V
J"
f
^>f^A-
VT
100
- 90
-80 |
• 70 x
O
• 60 TJ
C
O
' 50 o
An °-
O
i
-20 5
400 500 600 700 800 900
( Time (sees)
•Torq
nn IH m\ THC
(ppmC)" - - N0x(ppm)|
FIGURE 6
600.
Orbital SOcc 01 Engine - Test # 95-8292
100
200
300
400 500
Time (sec)
600
700
800
• Sparkplug Seat(C)
-Exhaust Gas (C)
- Muffler Surface (C)
^ Mode Flag
900
958292.XLS, Graphs
10/22/96
-------
OFFICE OF
AIR AND RADIATION
I AAiZ * UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
\ / . NATIONAL VEHICLE AND FUEL EMISSIONS LABORATORY
' ^olt 2565 PLYMOUTH ROAD
ANN ARBOR, MICHIGAN 48105
MEMORANDUM
SUBJECT: Exemption from Administrative Review
FROM: William Charmley, Mechanical Engineer
Nonroad Engine Programs Group
THRU: Paul Machiele, Manager
Nonroad Engine Programs G
TO: Chester France, Director
Engine Programs and Compliance Division
The attached report entitled "Exhaust Emission Testing of
Orbital 50cc Direct Injection Two-Stroke Engine" contains
emission results from a Piaggio 50cc two-stroke engine modified
by Orbital Engine Corporation to have a stratified charge
combustion process. The engine was tested using the Federal Test
Procedure for Small Spark- ignited Engines <19kW, utilizing Cycle
C ("handheld" engine cycle) .
The data from the direct injection (DI) two-stroke engine
shows that exhaust emissions of carbon monoxide and hydrocarbons
are substantially lower when .compared to existing single-cylinder
four-stroke and charge scavenged two-stroke engines. The results
also indicate oxides of nitrogen emissions from the DI engine was
typical of existing four-stroke engines but 2 to 3 times higher
than existing charge scavenged two-stroke engines.
Since this report presents only factual emission data and
noncontroversial conclusions, .. and I would like to make copies
publicly available, your concurrence is requested to waive
administrative review according to the policy outline in the
directive of April 22, 1982.
Concurrence :
Nonconcurrence :
Attachment
Printed on Recycled Paper
------- |