Engine runs lean when hot

[V8hilux]

I have just taken a 36 minute log of what my engine does as it warms up and it shows a gradual leanout as the other sensors warm up.

The engine was held at revs for a while then idled for a while, engine turned off then restarted while the log continued recording.

I might have found a pattern in the tune that could be giving me the grief of leaning out with higher temps but not sure.

I selected moniter 2, the bigger one and selected:

wideband
manifold air temp
battery volts
rpm
injector pulse width

The part of the log that is interesting is between 10:35 and 18:55. (grab the slider and go back and forth between the 2 time points really fast while watching the lines)
It is a steady rpm section that seems to show a rising in manifold temp,a drop in battery volts, a rise in AFR, and a drop in injector pulse all at the same time which could explain a leanout. but i don't know as this is the first time I've really studied a log in the moniters part of tunerpro.

I'd love some feedback if you have the time to take a look please.

longlogethotoffhotrestart.xdl

[V8hilux]

I have fixed it.

[VL400]

For anyone else that may have similar issues what did you do do fix it?

[V8hilux]

I don't fully understand how I fixed it and doesn't make sense to me but here's what i did:

I grabbed a bag of frozen peas and a hair dryer.
The multi meter,
the laptop.

I checked all voltages at battery, coil, ecu wires with the engine cold.
Then I got the hair dryer and started heating up things to check for heat related connection problems.
I was looking for changes that could be affected by heat in the engine bay.
I heated the relays, the injector connectors, the alternator connectors and even used the dryer and a foot heater to get the whole alternator hot, I heated up the ignition module and other parts of the wiring throughout the engine bay while starting the car in between the heating up of parts to try to simulate the lean condition that happens when the engine bay gets hot.... all my tests were fruitless. Could not simulate it.

Then I used a foot heater and the hair dryer and heated the whole fuel rail, injectors, and fuel pressure regulator...No change in the afr's were noted except a slight lean out apon starting but went back to normal afr's within a couple of seconds.

I done all the same tests with my bags of frozen peas, beans, and frozen corn. Nothing changed.

So I wondered if the voltage compensations in my tune were the culprit:
I started car and noted the afr. then removed the alternator main output wire (big red one to battery) I saw a drop in afr, so I altered the "injector bias vs battery voltage" table so that the car delivered the same afr's with all accessories on, lights etc and then I did the same with the alternator wire connected and adjusted the table until I has reasonably consistent afr readings from 11.8 volts all the way through to 14.2 volts.

I let the engine get hot and then shut it down for about 10 minutes... sure enough upon restart it was running lean.

I was scratching my head... why.

As far as I could tell, I had eliminated most things in the engine bay like wiring, fuel pressure etc by heating them up and cooling them to no avail.

Then I thought about my intake manifold and the placement of the intake air temperature sensor. Could that be the problem?
I have an aftermarket come racing twin throttle body manifold that is very good at shedding runner heat compared to the standard banana manifold that comes on the 5l 304 holden v8 engine.

I have heard of some people in other forums (mainly rice) that frequently move their intake manifold temperature sensors but I didn't read so far as to why they do it and I was almost about to do the same. Instead.

I looked in 0se12p, and set my target on the "Charge temperature inverse temperature term vs Charge temperature" table and there was a pattern that was showing up every time the engine run lean after sitting for a while when hot.

The bubble tracing always showed that the cell in which the ecu was using in this table never gets above about 70C while driving around steady and under boost or heavy throttle it actually drops around the 50 mark.

After every hot shutdown of the engine and waiting 10 minutes and then re starting, the cell in this table is upwards of the 70 it normaly uses while driving.
So I simply punched in "20 into every cell in the "tempK" column above 70C and now the problem is gone.

I can't understand why this has fixed it but all i know is that it appears to be the problem causing my leanout after hot starts.

This was a big change to the table, the degrees K column "before" showed a gradual increasing number from top to bottom and in "tune" to the degrees C column on the left but now, I can't figure out why the engine runs good even though the drastic change to the higher temperature parts of that table.

If anyone could explain why this has fixed the problem I would appreciate it thanks.

[VL400]

Nice testing, am sure the mrs will be wondering why the peas and corn are all greasy now


What you have changed is what was suggested the issue was, heat soak causing a high charge temp - the ECU was over compensating for this and pulling too much fuel due to the different intake having different heating and cooling properties to a stock bananna setup. There are also post start AFR timeout items if you find that the lean issue only happens briefly after starting and then clears while charge temp is still high.

[blownvp]

had similar problems with my car.v6 vp twin screw whipple 10 psi no intercooler standard intake manifold and sensor in factory location.man i played around with the the charge air temp bias tables for years trying different values etc.in the end i found with my combo was to zero the whole charge bias table so it used the manifold air temp sensor only for correction and copy a v8 5l inverse charge temp table and tweak that slightly too suit.mine is still slightly lean on hot restart but not as bad as it use too be and the afrs seem too be more consistent driving confirmed with wideband.i run a full open loop tune at the moment.i think what was happening is the manifold was heat soaking so coolant temp and manifold temp would start the same then after driving the manifold temp would go higher than coolant and vary more with boost etc.i run an 82c thermostat.that inverse table doesnt have to be altered by much and it seems to make afrs change a fair bit.cheers darren.