[thstone]

Here is my best understanding of the start up fuel injection control sequence:

The DME controls cold start using engine temp and rpm as the critical signals that regulate injector pluse time and ignition timing. Normally, the injectors will pluse longer to enrich the mixture along with maintaining the timing at near TDC to get a clean and smooth start.

If the engine fails to immediately start (as sensed by low rpm), then the ignition timing is advanced and the mixture further enriched to get the engine to start but not to the point where the engine would flood with fuel.

Once the engine starts, the DME will keep the mixture enriched for 5-30 seconds and continue to adjust the timing based mostly on engine temp. This enrichment is gradually reduced as the post start timer (time since engine start) reaches the 30-sec mark.

After post-start, the DME transitions to the warm-up phase where the engine is controlled by the stored map control feedback loop.

Of course, this is all happening in a closed-loop feedback system designed to start the engine smoothly and then maintain a smooth post-start idle.

If you were to stop the engine 10-20 seconds after start, then upon re-start it would begin this process all over. However, with the engine shut off in a somewhat rich condition, this should improve the re-start (since the mixture is already rich) but then the engine could run slightly "over rich" for 10-20 more seconds until the post start timer reaches 30-sec mark because the DME hasn't transitioned to using the O2 feeback system to maintain the mixture until the warm-up phase.

However, the amount of fuel that is injected to enrich the mixture at startup is only a teeny tiny amount more than normal and even a slightly over-enriched mix would still only contain slightly more fuel than a perfect mixture. Plus, a rich fuel mixture rarely (ever?) causes an engine to smoke.

Engine smoke is almost always related to oil in some way.