Does it have to learn to idle?

[Sqrly]

The capture is good and as expected for a hot idle vehicle. Your "observed" LTFT is the problem.... no way it should be sitting in that far negative. In fact, when the STFT swings negative, the sum is almost at coding (-25%)... running too rich.

So what could make it idle rich? Is there something to find with my scanner app?

I checked the pressure regulator vacuum line, there's no gas in it, not even the smell. The air filter was changed in march. I put in a stock one.

When I fixed the gas tank, because the cracked/leaking rollover valve had been replaced with a giant hose barb that was still leaking, I checked the charcol canister and it was dry.

I haven't changed the fuel filter yet, could a dirty one cause an issue like this?

Note: Same log as in post #78

 

[budwich]

For a speed density fuel control system, the MAP and intake air temp are the main inputs. My guess would be the MAP. You have shown a trace of the air temp which is somewhat easy to confirm based on "known" outside temperature athough that could be less than good as intake usually comes by a "hot engine component" (ie. rad). The MAP on the other hand would need a vacuum gage to confirm the readings. I think mooseman suggested pulling the map connection while the truck was running to see how it response. I don't recall, but it should stumble and maybe stall.

I think the other area is the throttle position sensor. The reading might indicate that the system is more open than it actually is... meaning less air is being taken in but the system "calculates" that it should be getting more according to the position. How to know... I don't know without actually using a "command function test tool".

 

[Sqrly]

For a speed density fuel control system, the MAP and intake air temp are the main inputs. My guess would be the MAP. You have shown a trace of the air temp which is somewhat easy to confirm based on "known" outside temperature athough that could be less than good as intake usually comes by a "hot engine component" (ie. rad). The MAP on the other hand would need a vacuum gage to confirm the readings. I think mooseman suggested pulling the map connection while the truck was running to see how it response. I don't recall, but it should stumble and maybe stall.

I think the other area is the throttle position sensor. The reading might indicate that the system is more open than it actually is... meaning less air is being taken in but the system "calculates" that it should be getting more according to the position. How to know... I don't know without actually using a "command function test tool".

How would I go about checking the map .... NVM took like 2 seconds to find a video. I don't have a vacuum pump.

 

[Sqrly]

I reread the entire thread.

Right after the 2nd TB Relearn (Post #24) when it was idling correctly with the defroster on and the AC compressor turning on and off the LTFT looked like this.



Now it's stalling again and it looks like this.



What is happening to make it change?

 

[TJBaker57]

Note: While stalled (not running) I had a "Check Gauges" light on and the DIC said "stop engine" and "low oil pressure".

This is not something to disregard.

This warning has to be triggered while the engine is running so it most likely happened shortly before or during the stall.

In any event, I would test the actual engine oil pressure with a test or aftermarket oil pressure gauge when the engine is HOT. Testing cold or semi warm is inadequate here.

I have a high mileage 4.2 and this is an issue for me. Thus I DO NOT idle my engine unless I am in the drivers seat monitoring the aftermarket engine oil pressure gauge I installed.

In case you haven't read this the factory engine oil pressure 'gauge' in the dashboard is a fake. It cannot actually read the true oil pressure.

Q: How do you get the app to display more then 3 seperate charts at once? I had to edit these captures together.

Do you mean during a review of data recording?

In this case I too only see a few graphs in one screen. But my screenshot function allows for paging down when capturing a screenshot.

 

[Sqrly]

This is not something to disregard.

This warning has to be triggered while the engine is running so it most likely happened shortly before or during the stall.

In any event, I would test the actual engine oil pressure with a test or aftermarket oil pressure gauge when the engine is HOT. Testing cold or semi warm is inadequate here.

I have a high mileage 4.2 and this is an issue for me. Thus I DO NOT idle my engine unless I am in the drivers seat monitoring the aftermarket engine oil pressure gauge I installed.

In case you haven't read this the factory engine oil pressure 'gauge' in the dashboard is a fake. It cannot actually read the true oil pressure.






Do you mean during a review of data recording?

In this case I too only see a few graphs in one screen. But my screenshot function allows for paging down when capturing a screenshot.

I've read all about the oil pressure guage and how to add a real one. Sounds like a pain in the arse.

I've never seen a screenshot function like that, nice.


This weekend I plan to pull all the sensors I can easily access and and inspect them and all their connections. I'll spray some ether on vacuum line connection points and around the intake to see if it alters the rpm. (Old tricks are the best tricks, no?)

Q: Can any of the sensors be damaged by brake clean? I do have a compressor to blow things out.

I'm also going to do another TB Relearn to see if the LTFT changes back to showing me several small increments of change again instead of the min/max behaviour I'm seeing currently.

I did order a WIX fuel filter, just because I haven't replaced it since repairing the tank, but it won't be here until Dec. I'm pretty confident it has some bits of pipe dope in it from the previous attempt at repairing the tank with a hose barb

@Mooseman You suggested logging a drive with the MAP unplugged. This suggestion seemed to be based on a log that didn't actually have any MAP info. After seeing logs that included the MAP, do you feel this excercise would still produce useful info?

 

[AmpOverload]

Right after the 2nd TB Relearn (Post #24) when it was idling correctly with the defroster on and the AC compressor turning on and off the LTFT looked like this.

View attachment 115029

Now it's stalling again and it looks like this.

View attachment 115030

What is happening to make it change?

Apologies for jumping in here... I don't intend to distract from the advice you're already receiving, but I think I have something useful to say.

Given that your problem is turning out to be hard to diagnose, I'd suggest that, at least initially, you start monitoring Mode $01/$22 PID $03 ("Fuel System Status"). Although adding an extra PID to the list you're recording is going to slightly lower the overall recording rate of the other PIDs, I'd assert that anytime you start looking at things like fuel trim (STFT & LTFT) and/or O2 sensor voltages, you really need to make sure that the fuel system is in "closed loop" mode and not "open loop" mode. Otherwise, you can wind up "chasing phantoms".

PID $03 reports 'fuel system status' as 1 of 5 possibilities (per SAE J1979):

1. 'OL' -- Open loop - has not yet satisfied conditions to go closed loop
2. 'CL' -- Closed loop - using oxygen sensor(s) as feedback for fuel control
3. 'OL-Drive' -- Open loop due to driving conditions (e.g., power enrichment, deceleration enleanment)
4. 'OL-Fault' -- Open loop - due to detected system fault
5. 'CL-Fault' -- Closed loop, but fault with at least one oxygen sensor - may be using single oxygen sensor for fuel control

Once you've been recording your fuel/O2-related data with PID $03 and have some confidence that the fuel system is operating normally and consistently (i.e. going from 'OL' to 'CL' and staying there, with possible, occasional transitions from 'CL' to 'OL-Drive' then back to 'CL', when appropriate), you can consider dropping PID $03 from the recordings to maximize recording rate once again.

 

[Sqrly]

Apologies for jumping in here... I don't intend to distract from the advice you're already receiving, but I think I have something useful to say.

Given that your problem is turning out to be hard to diagnose, I'd suggest that, at least initially, you start monitoring Mode $01/$22 PID $03 ("Fuel System Status"). Although adding an extra PID to the list you're recording is going to slightly lower the overall recording rate of the other PIDs, I'd assert that anytime you start looking at things like fuel trim (STFT & LTFT) and/or O2 sensor voltages, you really need to make sure that the fuel system is in "closed loop" mode and not "open loop" mode. Otherwise, you can wind up "chasing phantoms".

PID $03 reports 'fuel system status' as 1 of 5 possibilities (per SAE J1979):

1. 'OL' -- Open loop - has not yet satisfied conditions to go closed loop
2. 'CL' -- Closed loop - using oxygen sensor(s) as feedback for fuel control
3. 'OL-Drive' -- Open loop due to driving conditions (e.g., power enrichment, deceleration enleanment)
4. 'OL-Fault' -- Open loop - due to detected system fault
5. 'CL-Fault' -- Closed loop, but fault with at least one oxygen sensor - may be using single oxygen sensor for fuel control

Once you've been recording your fuel/O2-related data with PID $03 and have some confidence that the fuel system is operating normally and consistently (i.e. going from 'OL' to 'CL' and staying there, with possible, occasional transitions from 'CL' to 'OL-Drive' then back to 'CL', when appropriate), you can consider dropping PID $03 from the recordings to maximize recording rate once again.

Thank you for 'jumping in'. (Taking the time to read all this and offering your assistance)

I just took a look at my app (while connected to the vehicle) and I don't see a "Fuel Status" sensor in the live data section where I pick pid's to log.

I do see it in the "All Sensors" section. After finding it there I started the vehicle and it went from open to closed in less then a minute. It's just coincidence the coolant temp is displayed too.

I take it I should check the open/closed state while I am experiencing the stalling?

 

[AmpOverload]

I take it I should check the open/closed state while I am experiencing the stalling?

Actually, if it was me, I'd want a continuous recording of it, anytime you're recording fuel/O2-related data, not just a quick check from time-to-time on a page where it's dynamically updated. Otherwise, you risk having it change and then change back (i.e. to "CL") without your having detected it.

I just took a look at my app (while connected to the vehicle) and I don't see a "Fuel Status" sensor in the live data section where I pick pid's to log.

Unfortunately, I don't use 'Car Scanner Pro', so I don't know how to specify that PID for recording, but I'd expect there to be a way to do that, somehow. I'm sure @TJBaker57 will know how.

 

[TJBaker57]

I do have a compressor to blow things out.

I would be exceptionally cautious about blowing things with compressed air. You run the risk of forcing contaminents into areas where they can do damage. Keep a safe distance between the air discharge and electronic components of any sort.

I generally steer clear of brake clean unless I am cleaning brakes. MAF cleaner or other electronics cleaners are a better choice.

I'm also going to do another TB Relearn to see if the LTFT changes back

I am curious about your trim values myself. I am thinking the wide jumps are happening when the current trim cell changes due to some other value changing sufficiently.

I'd like for you to do something and post a screenshot or two here. I'll explain afterwards what we are looking at.

1: Turn the key to RUN but don't start the engine.

2: Open the Car Scanner app and connect to the vehicle.

3: Select "Settings", then "Terminal".

4: One line at a time enter the following and press "Send". The first 3 lines should each respond "OK". The rest should respond with a single line that begins with "6C F1 10" .

ATH1
ATS1
AT6C10F1
3C62
3C63
3C64
3C65
3C66
3C67
3C68
3C69
3C6A
3C6B

You should end up with something that looks like the following though you results will be different...

Capture your results and post them here in this thread. Then I will explain what this is.




On an unrelated note: while getting this sample screenshot I noticed the "PID scanner" button at the bottom of the terminal screen. Decided to have a look and holy crap! The developer has built into the app a PID Scanner! I wonder when this happened? One has to know what they are doing here to make it work but this is cool for those that do know.

 

[TJBaker57]

just took a look at my app (while connected to the vehicle) and I don't see a "Fuel Status" sensor in the live data section where I pick pid's to log.

We can add it as a custom sensor so it is available under the "live data" selection.

Since the "live data" graph cannot record text the result will be in the native form of the response which is a numeric value.

Not really a problem as there are only 6 possible values as displayed in this screenshot from the Wikipedia page of "Standard PIDs".




To add the PID where it can be chosen in "live data" choose "Settings", "Sensors" then tap the "+" button up top near the trash can. Now either scroll to where you see "New PID" listed or enter "new" at the search field.

When you see the "New PID" entry tap on that and make it look like this... (except for the obvious typo in the second line)....

 

[AmpOverload]

View attachment 115044

A minor footnote... I've never seen a reply to the "Fuel System Status" query return a value of zero, even with the engine off. This includes multiple GM vehicles, including a 2005 Chevy Avalanche. When the engine is off, the reported state has always been "Open loop - has not yet satisfied conditions to go closed loop".

 

[TJBaker57]

A minor footnote... I've never seen a reply to the "Fuel System Status" query return a value of zero, even with the engine off. This includes multiple GM vehicles, including a 2005 Chevy Avalanche. When the engine is off, the reported state has always been "Open loop - has not yet satisfied conditions to go closed loop".

I cannot comment on the textual portion. As we know the actual response is numeric. Currently the 2005 P10 PCM sitting here in the house test+drawer returns a 1 with no engine present.

 

[AmpOverload]

Sorry, I wasn't clear (due to a careless edit made before posting). I wanted to say:

When the engine is off, the reported state has always been 1 ("Open loop - has not yet satisfied conditions to go closed loop").

The text I used for the state corresponding to a value of 1 is right out of SAE J1979. Frankly, Wikipedia takes undeserved liberties in re-phrasing the meaning of that state to say "due to insufficient engine temperature". And I have no idea where they got the idea that "the motor is off" for value 0.

 

[Mooseman]

You suggested logging a drive with the MAP unplugged. This suggestion seemed to be based on a log that didn't actually have any MAP info. After seeing logs that included the MAP, do you feel this excercise would still produce useful info?

Maybe. This would take the MAP out of the equation if it's good (same stalling happens) or show it as defective (stalling stops).

Q: Can any of the sensors be damaged by brake clean?

I would use carb/throttle body cleaner instead. Brake cleaner can damage some rubber parts.

When I fixed the gas tank, because the cracked/leaking rollover valve had been replaced with a giant hose barb that was still leaking, I checked the charcol canister and it was dry.

This might be a significant clue. Even though the charcoal canister might "look" dry, it could still be saturated with fuel. The EVAP doesn't do the purge until the engine is warm. Maybe there is so much fuel vapor or liquid fuel that when it's purging, it makes the mixture so rich that it's pulling the LTFT and it has trouble regulating it when the AC comes on/off. One of the things that happens when you don't have that rollover valve (or fill valve) is that it overfills the tank, pukes fuel out the filler neck when it's full, and allows liquid fuel into the charcoal canister. I had this happen on my old Avalanche with a broken fill valve in the tank. The canister was full of fuel however I did not have any stalling or drivability issues but it might have been so full of fuel that nothing was getting through.

Maybe try to disconnect the purge valve or plug the line to see if this helps. I think it's worth exploring.

 

[TJBaker57]

I took my own 2002 4.2 TrailBlazer out for a short run to grab some data yesterday.

I can confirm the my 4.2 does change fuel trim cells at idle depending on the status of the A/C compressor. Also changes idle fuel trim cells whether in Park/Neutral vs a Drive gear.

In my case it really didn't matter as the trims in all my idle cells seem fairly similar.

I just set up some more PIDs for the cell values and plan on doing more recording today in hopes of gaining a better understanding of what the cell values in Data Blocks $63 to $6A represent.

Here is a sample from yesterday during the warmup idle. Though not nearly as severe as @Sqrly I also see a more negative short term fuel trim when the compressor is OFF and my trim cell is #20 (vs #21 with compressor ON in Park)

Similar conditions seen in the right side of these graphs while in Drive and switching between cells #18 and #19.




I also discovered that when the front defrost is enabled pressing the snowflake button to turn the compressor off is only temporary. The front defrost mode overrides this and turns the compressor back on.

 

[Mooseman]

Aren't you at a higher elevation? That could be why you're getting some negative FTs.

 

[TJBaker57]

Aren't you at a higher elevation? That could be why you're getting some negative FTs.

It is true, This recording is between 7500 feet and 8200 feet.

I'll readily admit I don't have a good working knowledge of this air/fuel ratio business and how it is supposed to work in a speed density system with no Mass Airflow Sensor.

And how would high altitude affect it.

My usual approach is to grab as much data as I can access and then see what I can learn from it.

 

[budwich]

OP, I would suggest that you turn off the environmental controls as your problem is unlikely to be related specifically to them.... they are just highlighting the issue which is a fueling control / mixture issue. Then, get your vehicle to a "normal temperature" (ie. closed system, running on O2, engine temp at 190ish). Then monitor the fuel status, the fuel trims, rpm at idle, and ECT in park. This should be a base line capture. Base on your previous work, it is likely to show very low LTFTs. IF true, then follow up with previous suggested tests, of disconnecting the hose going to the charcoal filter (and plug it). Run the monitors again to compare results. Go from the results.

 

[TJBaker57]

I will just toss this out here.

There are like 44 values of fuel trim saved in block memory. 23 for short term and 21 for long term.

Each set is used under different operating conditions some of which include A/C compressor engaged or disengaged.

Some of these values are updated during, and at the conclusion of each drive cycle.

A Tech 2 has a "special function" of a fuel trim reset where all of these saved values are reset to zero.

This can also be done directly without a Tech 2 via manually sending the proper instructions via serial terminal.

Additionally, if warranted, doing it manually provides the ability to return the values to the previous state if need be. A Tech 2 CANNOT do this.

 

[TJBaker57]

A Tech 2 has a "special function" of a fuel trim reset where all of these saved values are reset to zero.

I might clarify this just a bit. The values are actually saved with an offset to enable the recording of negative values. So in this context a "zero" trim is actually saved as hexadecimal 80 (decimal 128).

So the Tech II overwrites all the saved values to hexadecimal 80, the equivalent of a zero trim.

This is seen back in post #90 where the trim values are all "80". The particular PCM seen there is a 2005 P10 operating in a drawer in our kitchen !

 

[TJBaker57]

I'm still exploring and experimenting with the stored fuel trim cells, fuel trim index, and so on.

Yesterday I did more recording and captured data while refueling. For whatever reason I left the truck idling during the fueling.

Boy, that will cause some wild fuel trim data!

The Fuel Trim Index (FTI graph) plumetted when the gas cap was removed with the truck hot idling. Lots of vapors getting pulled in through the purge valve with no gas cap present.

 

[Mooseman]

So fueling while idling, which may have introduced extra fuel vapours, pulled both LTFT and STFT lower, especially the STFT. That's why I suggested disconnecting the purge valve, either electrically or the hose, in case that system is introducing an excessive amount of fuel vapours.

 

[TJBaker57]

The Fuel Trim Index, which includes 'Purge Learned Memory" in addition to the average Long and Short Term Trims, really dropped in my data log during refueling. And it is the "Fuel Trim Index" value which is used for codes like P0171 and the like.

I find it curious that my 4.2 can handle such a drop and display absolutely no ill effects. Idle was rock steady. If I hadn't seen the data I would never have known about the trim drops. And yet @Sqrly stalls.

Another curious thing I discovered is that in the fuel trim cell values for my 2002 4.2 there are no saved values for Long Term 'WOT' or Long Term idling with compressor OFF. Short Term values, yes. Long Term values, not saved.


I agree with the idea of eliminating the possibility of the EVAP system bringing in unintended fuel vapors as a test. Or at the very least record the data like I just did. Of course the data would not likely clearly show a stuck or leaking purge valve.

An EVAP system leak test should show that though and set a leak code. That is assuming the conditions for leak test(s) are being met and the test(s) are running.

 

[Sqrly]

I would be exceptionally cautious about blowing things with compressed air. You run the risk of forcing contaminents into areas where they can do damage. Keep a safe distance between the air discharge and electronic components of any sort.

I generally steer clear of brake clean unless I am cleaning brakes. MAF cleaner or other electronics cleaners are a better choice.





I am curious about your trim values myself. I am thinking the wide jumps are happening when the current trim cell changes due to some other value changing sufficiently.

I'd like for you to do something and post a screenshot or two here. I'll explain afterwards what we are looking at.

1: Turn the key to RUN but don't start the engine.

2: Open the Car Scanner app and connect to the vehicle.

3: Select "Settings", then "Terminal".

4: One line at a time enter the following and press "Send". The first 3 lines should each respond "OK". The rest should respond with a single line that begins with "6C F1 10" .

ATH1
ATS1
AT6C10F1
3C62
3C63
3C64
3C65
3C66
3C67
3C68
3C69
3C6A
3C6B

You should end up with something that looks like the following though you results will be different...

Capture your results and post them here in this thread. Then I will explain what this is.

View attachment 115043


On an unrelated note: while getting this sample screenshot I noticed the "PID scanner" button at the bottom of the terminal screen. Decided to have a look and holy crap! The developer has built into the app a PID Scanner! I wonder when this happened? One has to know what they are doing here to make it work but this is cool for those that do know.

Wow guys, lots of input, thank you.

We all have the week off for Thanksgiving here so I haven't been thinking much about the truck. I'm going out to change the fuel filter and popped in to check this thread as I was going to disconnect the battery while doing it and do the TB relearn again.

I added the Fuel Status Sensor, TY TJ. I will will do all the terminal commands as you specified before starting it and get a picture of the results.

This might be a significant clue. Even though the charcoal canister might "look" dry, it could still be saturated with fuel.

I also blew it out with air, all that came out was dry charcoal dust. I checked it because I knew it could get damaged after someone removed the rollover valve.

I had this issue on my 97 BMW R1100R, I removed it. (Bad design, the can was pyhysically lower then the tank)

Maybe try to disconnect the purge valve or plug the line to see if this helps. I think it's worth exploring.

Can I do that where the line hooks to the TB?

Do I need to plug the line from the tank too?

An EVAP system leak test should show that though and set a leak code. That is assuming the conditions for leak test(s) are being met and the test(s) are running.

Now that you mention it ...

I was getting a P0442 (Evap minor leak) before I dropped the tank and replaced the giant hose barb with the proper aftermarket rollover valve. I also replaced the entire filler neck assembly from the tank to the gas cap.

I did this back in May of this year. Since then the P0442 has popped up 3 times. I just clear it and check that the o-ring on my aftermarket locking gas cap is clean and make sure it is on tight.

If the purge valve is malfunctioning ..... I will definitely be looking closer at this.

My plan for today, after getting that info for TJ and changing the filter, is to let the vehicle warm up idling with the defrost on, which is when the issue occurs. Once I see the big idle drops I'll plug off the purge line and see if it changes anything.

Aren't you at a higher elevation? That could be why you're getting some negative FTs.

Yea, big difference. We're at an elevation of 131 feet here.

 

[Sqrly]

@TJBaker57

The typo in line 3 threw me. Luckily you added the screenshot too.

 

[Sqrly]

I started it cold and let it idle with the defrost until it stalled. Log looks the same as others I already posted.

I disconnected the purge line at the TB and plugged off the TB and the line. That made no difference at all.

@Mooseman

Disconnect the MAP and drive it?

I unplugged the map and started it and it instantly went into limp mode and it idled extremly rough, making the entire vehicle shake.

I thought maybe you meant disconnect it physically but leave it plugged in, same result.

I did remove and inspect it. It looks original and there's no gunk in the tiny hole.

I found a way to see the MAP PSI change in real time. While it was idling with with the defrost off and the scanner logging data, I disconnected the intake tube from the air filter and covered the hole with my hand. Just before it stalled I removed my hand and saw the PSI jump to just above 9 for a moment before the idle stabilized.

 

[TJBaker57]

@Sqrly

You have much better vision than do I !!

Transcribed the pertinent data from that screenshot, did some spreadsheet work and here is what I see.

But before the image let me explain what you have here.

The PCM stores values for short term and long term fuel trims in memory. It stores values for a wide range of operating conditions. These values are used and evaluated as you drive and updated during the drive as well as at key OFF.

I have noted a few of the operating conditions which I have observed/identified.



The sequence I had you enter reads out the stored values. First we read the short term values and then the long term values.

We didn't change anything, we only read out the values.

Two things strike me here.

The first is that there are no positive values stored at all. Every trim is in the negative.

Secondly is the amount of negative trim seen in a number of cells. Way greater than anything I would have expected.

I neglected to add headers so here is what you see:

Column I represents "Fuel Trim Cell ww xx yy:z" where:
'ww' is the number of the cell as reported by a Tech 2 AND the PID for "Fuel Trim Cell",
'xx' is ST means short term or LT for Long Term,
'yy' is the data block number,
'z' is the particular byte designation of the data block.

Column J is the hexadecimal value of the data byte

Column K is the result of the conversion from hex to decimal.


Temporarily removed image as I correct a thing.

OK, think I have it right now...

 

[TJBaker57]

@Sqrly

By comparison, I recorded my own cells on a 45 mile drive. The greatest negative value seen was -4 and the greatest positive was 20.

 

[Sqrly]

I thought that was what you were wanting to look at.

"By comparison" - My vehicle spends more time idling then it does moving while the engine is running.

An average day would consist of:
10 minutes warming up while idling.
A 3 mile round trip at no more then 35mph.
Parked for 5 hours.
A 1.5 mile trip then idling for 30 minutes
A 1.5 mile trip home.

Is it possible for an injector to leak fuel into the cylinder?

EDIT: Why yes, yes it is.

So I should probably check my fuel pressure and do a leak down test.

 

[TJBaker57]

@Sqrly

Clearly something is either making your fuel/air mix rich across most if not all operating conditions.

OR, something is making the PCM "think" the mixture is rich.



Something to consider,,,, We have the option of doing a fuel trim reset manually. It is very similar to what you just did to read the values. It just involves using a write instruction instead of the read instruction plus the values to write are provided in the message(s).

In considering that option,,, it may make the mixture overly rich until it relearns the corrections. I don't know the time it takes for such relearning to occur.

The Tech 2 provides the same fuel trim reset function but I have seen no guidance as to when the function should be used.

 

[Sqrly]

I borrowed a fuel pressure test kit from the local Autozone. I'll see what the pressure looks like in the next few days and if there's a leak somewhere.

How exactly are the fuel trims created?

Remember the vehicle had a severly clogged K&N air filter when I got it. I'm not exaggerating when I say it was "never" cleaned. It took a long time to accumulate that much dirt so it was air starved for a long time.

However, the stalling didn't start until it got cold outside, as opposed to daily temps of 90+. Does info from the MAT sensor effect which trim cell it uses and if so, is there a way to attribute a specific trim cell to a specific temperature range?

 

[TJBaker57]

I borrowed a fuel pressure test kit from the local Autozone. I'll see what the pressure looks like in the next few days and if there's a leak somewhere.

How exactly are the fuel trims created?

Remember the vehicle had a severly clogged K&N air filter when I got it. I'm not exaggerating when I say it was "never" cleaned. It took a long time to accumulate that much dirt so it was air starved for a long time.

However, the stalling didn't start until it got cold outside, as opposed to daily temps of 90+. Does info from the MAT sensor effect which trim cell it uses and if so, is there a way to attribute a specific trim cell to a specific temperature range?

This is something I have not seen documented in any great detail.

I have been looking at my recordings and seeking to correlate the current fuel trim in use to things like calculated airflow, engine rpm, A/C compressor ON vs OFF, throttle opening and so on.

I don't think ambient temperature comes into play in fuel trim cell selection.

However, in colder temperatures the engine may need more fuel at least until it is fully warmed up??

I found some information on sites like HP Tuners and the like regarding fuel trim cell selection criteria.

I once got a fuel pressure test kit from Autozone and in the state it was in (new) there was no way it could register any sort of pressure. The brass piece that is in the hose end to depress the schraeder had no hole in it !!

 

[Sqrly]

I sure do miss my Holley 3310-1 with 4 corner idle.

It was set up as 4 primaries. Talk about bad gas mileage LOL

 

[TJBaker57]

Here is an example of the sort of stuff I troll through seeking clues. Even if it doesn't directly apply to our P10 PCMs it does demonstrate some of the parameters that may be in play.

fuel trims on e38 ?

Fuel trims on this e38 are a bit confusing to me. Can someone explain what is going on? Car is an 09 g8 gt. Here is why I see in the editor (stock tune). LTFT boundries RPM low - 3968 mid - 8192 high - 8192

forum.hptuners.com

 

[TJBaker57]

@Sqrly

I had another thought....

When altering the fuel trim cell values, If we wanted to, we could selectively adjust certain fuel cells without affecting others. Just to see what effect it had and see if they go right back to where they are now.

A Tech 2 won't display them as far as I have seen. No menu item for it that I can find. And the Tech 2 only offers to reset everything.

When we are in the drivers seat we can elect to change, say just the cells that are highly negative. Like those 3 cells that are currently at -44.00. All 3 of those are for use when idling with the A/C compressor OFF.

 

[TJBaker57]

Maybe try to disconnect the purge valve or plug the line to see if this helps. I think it's worth exploring

Looking at the complete range of saved values of the fuel trim cells I see that all of the highly negative trims are during vehicle operating conditions with the highest intake manifold vacuum levels.

I 'think' this supports the notion that the likely source of an overly rich condition is the EVAP system. ???

On the 'flip' side there is the emissions system readiness data that appears to indicate that the onboard diagnostics have found no leaking purge valve or purge flow when not expected.

Disconnecting the vacuum line to the purge valve at the throttle body and plugging the port on the throttle body sure seems like a reasonable test.

Doing so, I would not be surprised if initially the stalling continued as the existing highly negative fuel trims would still be in effect at idle with A/C compressor disengaged.

If the purge valve 'IS' the source of the rich condition then the short and long term trim values should be seen to be moving towards the positive while operating with the purge valve vacuum line disconnected.

If I were doing this test I would NOT reset the saved fuel trims (don't disconnectect the battery) because this would mean you changed TWO parameters/conditions and thus introduce doubt as to which change was responsible for the observed results.

I have continued my exploration of these saved fuel trim cell values. Have made some key observations and gained some insights but will wait to post them up later.

 

[Sqrly]

Disconnecting the vacuum line to the purge valve at the throttle body and plugging the port on the throttle body sure seems like a reasonable test.

I tried that, see post #107 but .... only for a few minutes while idling in the garage and after it was warmed up and had started stalling.

Would it make a difference if I plugged it off and left it that way for a week of my short trips?

Also note: It will not run with the MAP disconnected. See #107. I mention it because it was suggested I try it so I assume they thought it would run that way.

FYI: I did the fuel pressure test. It has a rock steady 50 psi running, idling and revving the engine, and no bleed down at all after 10 minutes turned off.

 

[TJBaker57]

tried that, see post #107 but .... only for a few minutes while idling in the garage and after it was warmed up and had started stalling.

Would it make a difference if I plugged it off and left it that way for a week of my short trips?

Maybe. I am unsure what period of time would be needed, which is the thrust of this comment here...

Doing so, I would not be surprised if initially the stalling continued as the existing highly negative fuel trims would still be in effect at idle with A/C compressor disengaged.

I would advise monitoring the fuel trims during such an experiment to see if they trend towards the positive.


If the fuel trims do not seem to trend positive I might abandon the experiment in favor of a fuel trim reset.

 

[Sqrly]

Maybe. I am unsure what period of time would be needed, which is the thrust of this comment here...




I would advise monitoring the fuel trims during such an experiment to see if they trend towards the positive.


If the fuel trims do not seem to trend positive I might abandon the experiment in favor of a fuel trim reset.

Please clarify for me, does disconnecting the battery (for 30 minutes or more) reset the fuel trims?