Design and Simulation of Two Stroke Engines

Chapter 7 - Reduction of Fuel Consumption and Exhaust Emissions
start and duration of the metering of fuel into the injector body are referred to as "start of fuel"
(SOF) and "fuel duration" (FD), respectively. Similarly, the start and duration of the airassisted
injection process are referred to as "start of air" (SOA) and "air duration" (AD),
respectively.
The engine is also tested with the "perfect carburetor," i.e., an electronic fuel injector
(EFT) mounted in the inlet tract so that the engine is homogeneously charged at all times. The
word "perfect" is used in the sense that the operator can dial in any air-to-fuel ratio desired.
The engine is tested at full load at 3000 rpm, and at 1600 rpm at part load, with both
stratified and homogeneous charging, i.e., ABI or EFI installed; the results of these tests are
found in Figs. 7.52-7.55 and 7.56-7.59, respectively, over a very wide range of trapped airfuel
ratios, AFRt. Ignition timing is always set at MBT (minimum advance for best torque).
For the full load tests, stratified charging is recorded for three SOA timings, namely 110°,
125° and 140° atdc, and the air duration is 2 ms. For the part load tests, stratified charging is
recorded for three SOA timings, namely 215°, 230° and 245° atdc, and the air duration is 5 ms.
In the entire test series, presented in Figs. 7.52-7.59, the profile of the relevant performance
parameter for homogeneous charging can be compared with previous test results in
Figs. 7.3-7.8 and theoretical results in Figs. 7.13-7.18. However, here in Figs. 7.55 and 7.59 is
also shown the brake specific emissions of oxides of nitrogen with respect to air-to-fuel ratio,
which can be compared for profile with theoretical computer simulations in Fig. A4.3.
During full load (wot) operation, the SOA timing preceded exhaust port closure by 1 IS­
MS 0 . However, the SOA timings do not include any allowance for air solenoid delay. Tests
conducted with a Hall effect sensor on the injector show that the poppet valve opening is
delayed by approximately 2.5 ms. Therefore, the actual poppet valve opening occurred
approximately 45° later than the nominal SOA timing. Nevertheless, it is clear that during
this test a substantial proportion of the fuel is injected prior to exhaust port closure. The wot
results, presented in Figs. 7.52-7.55, show that the air-assisted direct-injection system pro-
03
CL
.*:
c£
E
600 -,
500
400
QUB 500
RESEARCH ENGINE
AIR-ASSISTED FUEL INJECTION
10 12
-
— i
14
—r-
16
TRAPPED AIR-TO-FUEL RATIO
SOA125
^SOAHO
—i
18
Fig. 7.52 Effect of fueling on bmep at full load, 3000 rpm.
523