MAP @ idle changes and messes up lambda

[PeepPaadam]

I have a problem with Porsche 944 S2 (3.0 inline-4, 16 valves, common plenum intake, all factory hardware). Installed VEMS, started the car without issues and tuned it. It had good vacuum at idle (29-31 kPa at ~1000 rpm fully warmed up). But after stopping the car and starting it again vacuum at idle was suddenly 55-60 kpa and obviously lambda was @ 0.7.
Checked various tables etc, nothing that could cause this. Could it be related to tank venting valve? I configured it to be open over 800 rpm.

After driving a bit, MAP went to normal again. BTW, at WOT, lambda is normal because MAP is 100 kpa, like it should be.

After starting it next time, again 55-60 kpa @ idle.

Any ideas?

[Sprocket]

Tank purge valve should only be open on closed throttle above any idle condition when engine is warm. Basically an overrun condition, otherwise there is a possibility to implode the fuel tank with sustained vacuum.

Introduce a 1% throttle limit on the configured output

[PeepPaadam]

Setting up the tank venting to be open only during decel seemed to cure the issue, thanks!

[jrbe]

Your ecu should be on a dedicated vacuum port to the intake manifold with only the fuel pressure regulator reference port attached to the same line if used (so they get the same vacuum signal from the intake.)

Anything else that is T'd  off of that line could allow a vacuum "leak" large enough to shift what the map sensor sees through what's left of the vacuum through the vacuum line.

Also, be careful where the purge valve "dumps." If you have it connected to intake vacuum it can suck a lot of fuel vapor out of the tank. This can actually collapse the tank and can shift a/f quite a lot.

[mattias]

jrbe: very good tips. 

The purge valve can be connected to just before throttle on N/A engines, or before compressor inlet  (after filter)  on F/I engines.
Any pressure in the tank escapes and the tank can never be collapsed, any excess vapors consumed by the engine.

[Sprocket]

I asked some time ago for a dedicated evaporative emissions strategy. I was looking for a simple PWM output enabled on TPS, RPM and CLT thresholds, much the same as the factory Rover strategy. In most cases, there is a carbon canister that capturres the fuel vapors, and the gentle vacuum (pulsed purge solenoid) on overrun draws it out of the canister. The canister also has an oriface which allows a certain amount of air in during the purge so that its not just a slug of fuel entering the engine.