Getting lean after a long trip

[lezsi]

I have a strange problem for ~two years now, not seems to be related to firmware version or the ECU itself.

I have a well adjusted smooth idle when starting the car and getting to working temperature.
This is the case when I'm doing short trips (like a few minutes) also.

But when I'm doing a long trip, like an hour on the highway, my idle getting ~10% leaner and stays that way for a while.
If I let the car cool for a few hours, my idle lambda goes back to normal again (still with the same temperatures).

I'm really curious about your opinion as I couldn't spot any difference in logged values other than lambda.

Here is a fresh log with 1.1.80:
http://quasar.dynaweb.hu/~lezsi/vems/v3.3_n002211-2010.07.31-16.25.24.vemslog

At around 1:00 I was in the city, idle lambda around 0.94
At the end of the log at 65:20 I get lambda values of 1.0-1.04
(EGO correction is off at idle)

I had no problem with ignition either case, the engine accepts both mixture, but I need to adjust it rich, to stay in a reasonable band when it gets lean after.


Engine is BMW M30 turbo, 60-2 primary trigger + Hall cam sync now (no cam sync before), 6x870cc siemens HighZ injectors now (1200cc lowZ before), Bosch 3x2 wasted spark.
I had the same effect with firmwares 1.1.26, 1.1.56, 1.1.78.
Tried more than two ECU boxes also.
 

[MWfire]

mat/tps table is wrong

[lezsi]

Good find, this was made for a non-IC setup two years back.
However I'm not sure that it is the cause of this problem as:
- The problem existed with 1.1.26 (without the MAT/TPS table)
- The idle air temperature was almost constant during the whole log.

The slight differences you see in MAT are related to intake manifold heatsoak.

[GintsK]

Also where is your MAT sensor located? And what sensor you use?

May be your fuel system suffers from overheating?

Gints

[lezsi]

MAT sensor is located above the engine in the alu intake, just like the water IC. 
Sensor is from VEMS webshop, the fast and no-heatsoak style (not in this position).

I also thought of fuel heating, but for a full 55L tank... I don't know.

[GintsK]

IMHO  better locate MAT sensor closer to valves as possible. Then it represents temperature closer to calculation model in ECU.

What is fuel flow/pressure? And what current goes to the pump?

[lezsi]

IMHO  better locate MAT sensor closer to valves as possible. Then it represents temperature closer to calculation model in ECU.

What is fuel flow/pressure? And what current goes to the pump?

MAT sensor is about on the latest position in common airflow, otherwise it should be on an individual intake runner (which is also hot)

I've two 255lph hp walbro fuel pumps feeding the custom made big rail on two separate 8mm lines.
Return is one 8mm line coming from a stock (VW/Audi?) 3.5 bar regulator.

[mattias]

Exactly where in that picture is the IAT sensor located?

[lugnuts]

Exactly where in that picture is the IAT sensor located?

there's a bosch MAT on the charge pipe, right after the 90 deg bend, on the right side of the pipe looking at the picture.

[gunni]

It should of course be after the cooler.

[z0tya]

before the throttle?

[gunni]

Normally close to the throttle where the airspeed is good , but in this case the air get´s cooled down after the throttle so the actual airtemp is going to be measured wrong. so it needs to be after the cooler on the way down a port. Maybe in the place the throttle body used to be

[z0tya]

Yes, and at 1:00 with shorter pw (2.3-2.5ms) richer than at 65:40 with 2.4-2.6ms. So measured air temp maybe wrong.

[lezsi]

Sorry for being so slow, I was on holiday.

The temp sensor right after the turbo is not for this IAT setup.
The actual sensor position is between cyl 3 and 4 intake runners, the green gasket area on the pic:


And this is my fuel rail:

On the left there are two input pipes and the reguator on the right gives the return. On far right there's a bleeding/measurement point which is not used generally.

[mattias]

I would fix this issue with the "MAT/TPS enrichment" table.
With a closed or near closed throttle, increase the enrichment until you reach lambda target. 
I've solved it this way on a couple of engines, usually BMW M30 due to the intake pipe going across the valve cover to the hot exhaust side.

There is usually no need to increase enrichment in the table for higher throttle openings, as the sensor does not heat soak in that case.