G4x - trying to get Fuel Table tamed

[marches]

Took a drive around the neighborhood last night, freshened up the plugs as the car has been running rich.  At the beginning of the drive the car ran smoothly, still rich.  Takes a looong time for the Lambda sensor to give up the data!?!?  After the car warms up, there are a few lean backfires, within minutes after ( the last ~10 minutes of log ) the car will barely run.  Will sustain idle, any throttle input now makes the car nearly stall.  Was making minor adjustments to table, something else has happened!  Pre-Crank, Post Start, warm up are enabled and timed/aged out due to engine temp, accell enrichment is off, CLLambda Off.  Just trying to get the car to run well on it's own.  Does any keen eye see something obvious around the last 10 minutes?

Log:  https://drive.google.com/file/d/1YmYnPSjO0V46CeyZEdQXhZIkHcM6U7BW/view?usp=sharing

Tune file attached.  

 

Thanks!!

ColdStart_forShortDrive_x2.pclx

[marches]

Looks like a LOT of TPS noise as well, is that typical or do I need to go back and verify ground signal is DEFINITELY an ECU sensor ground?

6 cylinder, ITB's, COP.  No IAC, should be a pretty straightforward deployment.

[Adamw]

I see quite a few issues:

1. Your wideband is reading wrong, for the first 1/3 of the log the lambda is between 0.5 & 0.6, there is no way the engine would even run at that.
2. For ITB's you usually need to use the alpha-N strategy for fuel, the main fuel table should have TPS as the load axis.  Your lambda target table should use the same.
3. Before you can use TPS as your main load reference you will need to improve this signal.  It is hard to tell if it is a sensor issue, electrical issue or mechanical issue (backlash or no return spring etc).
4. The battery voltage only shows about 13.0V, so either the alternator is barely charging or you have a poor power supply to the ecu. 13.8-14.5V would be normal running voltage for most alternators.
5. Coolant temp calibration is wrong, you have ~200°C in some places.  
6. The coolant temp is very noisy and drops dramatically whenever the engine comes to an idle.  And not just a little so it's not related to waterflow or anything - in some places it drops from 200°C to 80°C in less than half a second, which wouldnt be possible even if you pulled the sensor out of the engine.  Interestingly the coolant temp gets nosier but the TPS gets quieter when this happens.  I think you likely have a serious ground issue somewhere.  

[marches]

Adam, thanks for the reply.  When ordering the ITB conversion kit with StormX, I was a bit suspicious of the single-wire temp sensor from a noise perspective.  This can be easily remedied as the switch is easily replaced and I have a few extra signal ground wires back in the engine bay.  I will inspect the ground situation a bit closer.  During the wiring phase, I tried to be vigilant in keeping sensor grounds separate from chassis grounds, but perhaps I missed something.  Not sure about the TPS though.  I have a lot of old logs, will go through them and see if this is a new issue or something recent.  The car is an air-cooled flat-six, so the ECT signal is coming from the single-wire temp sensor that is bathed in motor oil.  Thanks again!!

[Adamw]

Hmm, I remember on the last ITB porsche I tuned (I think it was a Jenvey kit), I had to cut one of the supplied linkages up and shift one of the levers around to get a better angle or pull or something.  That was a couple of years ago so it is pretty vague now but I think it was really touchy down low and didnt get full throttle either.  Im just mentioning that in case your TPS issue is mechanical rather than electrical.  So make sure the mechanical action looks good before wasting time on sensors and wires. 

Yeah, a single wire temp sensor is not a good option for an ecu.  The calibration must be wrong too though, no way that oil is 200°C.

[marches]

Update:  New two-wire sensor set up, still trying to get calibration correct.  However, the TPS is foiling me at every turn.  The sensor was grounded with MAP and Lambda sensor.  Have moved TPS to separate ground, at idle it seems quiet.  However, once the throttle is moved, RPM slightly increases, and the noise shows up!  Unplugging and reconnecting the sensor will arrest the noise until touching the throttle again.  Is this a bad TPS?  I think it is a BMW Throttle Position Sensor 4166000.  Any ideas folks?  Thanks!!

TPSNoiseProblem - 2023-04-17.llgx monday.pclx

[Adamw]

Two things to check:

1. Ensure there is some free play in all the linkages when the throttles are sitting on the idle stops.
2. Check the TPS is a tight fit on the "D" shaft, from memory the BMW ones arent spring loaded so any slop can cause movement like this.  Some nissan's suffer a similar problem and an easy fix for those is a blob of RTV silicone in the D hole before you fit the sensor.  

If those look ok then I would say this is a failed sensor.  TPS sensors from road cars spend the majority of their life moving around the idle area so that is where the track wears the most.  If there is some adjustment you maybe able to rotate the sensor a little to move off the worn area to confirm.