WeatherManShawn's Tutorial: can someone look/help...??

[tatasta]

Hello,

I followed the new tutorial that Weathermanshawn posted up. The car ran pretty good with enabling the LTFTs etc. I am still lean as the engine warms but I can get that with a bit more work on A0008.

Please take a look at the attached tune and log file. My "A" map seems to have worked but the "B" looks like a miss. Is this related to the displacement issue mentioned before.

I would like to know what I need to adjust to get it in line with the tutorial.

AFR at WOT looks pretty good but, I did not log anything in LTFT, BEN, or Avg. What the he!L? Maybe something still disabled in tune?

I attached my calc.PID file for a check over.

Thanks in advance

[WeathermanShawn]

Tatasta:

Looks like you are not in Closed-Loop. I.E. You still have the car in Open-Loop. So it can't compute LTFTBENS, etc.

Just change B4205 to a normal ECT Temp (currently 140C). Its your choice on B0120. It can work with 400 or 4000 RPM.

Looks like the CALC.VE Table Formula is working. Its just that all your LTFTBENS are 1.000 right now, due to the car still being in Open-Loop..

Give it another shot. Changing B4205 to stock will do the trick..:cheers:

[tatasta]

I'll get it straight...
Thanks Shawn

[WeathermanShawn]

When you re-enable closed-loop, looks like you will plenty of LTFTBENS to work with. Looks like you may have modified your MAF Calibration Table B5001 at some time?

When you do the CALC.VE Table you will see your CYlAIR and DYNAIR come a lot closer..See attached chart.

Looks like you still are battling some KR?

Your CALC.VE PID looks fine..

Definitely not picking on you there..Some things I automatically see..:cheers:

[tatasta]

Thanks Shawn. Yes, B5001 has been changed because when I got the car, it had some crap aftermarket MAF on it. I sold it and went to a ported stock MAF. Even though the MAF is ported, should I return all of the values in B5001 to stock? I have spent a great deal of time matching actual data to that table and I think it is very close, not to mention when it was in CLMAF, I adjusted AFR with it some. Any advice on what to do with it?

Your not picking me apart on the KR. I spent last summer chasing that and found that there is a vibration that is causing the KR readings. They are consistent at RPM and conditions. Regardless of how much spark I pulled in that area or how high of octane I ran, the KR was always pretty much identical. There is no audible knock, ever.

Thank you for your help. I will log again tomorrow with the tune corrected and post up.

-Bob

[joecar]

Hi Robert,

We have new equations... see here for the new calc_pids.txt : showpost.php?p=117351&postcount=131 (http://forum.efilive.com/showpost.php?p=117351&postcount=131)

:)

[joecar]

Try with the current B5001, see how you go with LTFTBEN correction of it (turn on CL as Shawn said above).

[tatasta]

Hello,

I got the tune and pids right, I think, and took a few logs today. I have attached the original tune, the tune after applying the tutorial method and a large log file.

Please feel free to comment or pick me apart. I will say, the car ran very well.

Thanks

[WeathermanShawn]

I think that was generally good log run and resulting tune.

As expected I figured your MAF would show some LTFTBENS, which looks like they were all applied correctly.

I bet with another 1-2 logs both your DYNAIR and CYLAIR will 'converge'. As your MAF values 'settle' the CALC.VE Table values will also become more accurate.

Yours was a very good test case, as there was some deviation between the MAF Values and the VE Table % equivalent air prior to logging. I must say I am pretty impressed for a first run. I might just let your VE Table values stay as they are for now, but I bet with even one more log your Trims, MAF, and VE values will all converge. I would say you are 75-80% there.

Thanks for sharing your work. Please post up if you decide to 'rinse' it out one or two more times.

Congrats. Looks pretty good.

[tatasta]

I think that was generally good log run and resulting tune.

As expected I figured your MAF would show some LTFTBENS, which looks like they were all applied correctly.

I bet with another 1-2 logs both your DYNAIR and CYLAIR will 'converge'. As your MAF values 'settle' the CALC.VE Table values will also become more accurate.

Yours was a very good test case, as there was some deviation between the MAF Values and the VE Table % equivalent air prior to logging. I must say I am pretty impressed for a first run. I might just let your VE Table values stay as they are for now, but I bet with even one more log your Trims, MAF, and VE values will all converge. I would say you are 75-80% there.

Thanks for sharing your work. Please post up if you decide to 'rinse' it out one or two more times.

Congrats. Looks pretty good.

Thanks!! It was really nice to see so clearly what was happening from the log data and knowing what was to be fixed, both why and how. I am not close to being proficient with EFI Live but this certainly helped me to quit wandering and guessing with some of these basics that are so fricking elusive to us amateurs.

I finally feel confident enough to get my N2O dialed back in after a few more runs N/A and repeating this process.

That was bar none the quickest way I have used to clean things up.

Thank you very much for your help. I will continue to post my results.

Robert :grin:

[joecar]

+1 on what Shawn said, good log, LTFT's showed deviation from tables, good job... :cheers:

Did you use VE_Table[%] or VE_Table[g*K/kPa]...?

Thanks for posting.

[joecar]

...

That was bar none the quickest way I have used to clean things up.
...
Well said, the intent of the tutorial was to be able to get the tables corrected in a simple and quick manner... :cheers:

[joecar]

I'm going to change the title of this thread, if it's ok with everyone...

[tatasta]

I'm going to change the title of this thread, if it's ok with everyone...

Good with me. It was a poor choice of words.

[tatasta]

+1 on what Shawn said, good log, LTFT's showed deviation from tables, good job... :cheers:

Did you use VE_Table[%] or VE_Table[g*K/kPa]...?

Thanks for posting.

It was VE_Table[%].

[tatasta]

Should I smooth my new B0101 or leave it through the next several logs? It is pretty rough.

[RevGTO]

It's time for me to update also. Hope this isn't a hijack, but it seemed like the right place.

While I was able to bring a lot of my parameters into harmony, my commanded vs. logged AFR continued to be way off in the lean direction. I fooled around with my LM-1 settings and came to the conclusion that my WB was probably functioning properly.

Then I did a bunch of logs with my old OLMAF tune with the modified VE and MAF tables to eliminate narrowband contamination. Commanded vs. logged AFR was much closer. A little more MAF tweaking and it got very close. Then I went back to a CL tune and the AFR's were wildly off and lean again. LTFTBEN's varied quite a bit from log to log, too, but indicate a reduction in MAF values which would push AFR's further in the lean direction.

At this point, I'm stymied. I'm attaching one of my better logs for review and comment.

[joecar]

Should I smooth my new B0101 or leave it through the next several logs? It is pretty rough.I would push down a little on these two spikes (not completely flatten)... and leave the rest of the table as is.

$0.01

[WeathermanShawn]

Rev:

All I can say looking at your log is that I have never seen closed-loop AFR's so far from stoich. I am using a serial connection with my wideband and always average 14.63 AFR during closed-loop.

The method can not work if closed-loop is not functioning properly. What is odd about your log is that your LTFTBENS are pulling fuel (too rich), but your wideband says lean.

Either your narrowband O2's need replacing, or their is a problem with the wideband reading. Are you sure you are in closed-loop?

Again I have never seen closed-loop AFR's so far from stoich. It is a little puzzling..

[joecar]

It's time for me to update also. Hope this isn't a hijack, but it seemed like the right place.

While I was able to bring a lot of my parameters into harmony, my commanded vs. logged AFR continued to be way off in the lean direction. I fooled around with my LM-1 settings and came to the conclusion that my WB was probably functioning properly.

Then I did a bunch of logs with my old OLMAF tune with the modified VE and MAF tables to eliminate narrowband contamination. Commanded vs. logged AFR was much closer. A little more MAF tweaking and it got very close. Then I went back to a CL tune and the AFR's were wildly off and lean again. LTFTBEN's varied quite a bit from log to log, too, but indicate a reduction in MAF values which would push AFR's further in the lean direction.

At this point, I'm stymied. I'm attaching one of my better logs for review and comment.Can you also log HO2S11 and HO2S21 (you can remove say IAT to make room since we're using DYNAIRTMP_DMA now).

What value does your B3601 contain...?

Like Weatherman said, your measured AFR has an average value of 15.1...

You should see the measured AFR being equally balanced above/below the stoich AFR like this pic (its average value being stoich)...

[ in pic, to get the wideband AFR add 11.0 to AD1 ]

[WeathermanShawn]

Robert (AKA Tatasta):

Thanks again for sticking with it and your constructive feedback:cheers:..

What is interesting about your tune and log is that your LTFTBENS were indicating that it wanted 'more fuel' (+ LTFTBENS). So I think your car is overall going to be 'happier'. When I first started out H/C and purchased EFILive i was running +12-+15 LTFTBENS. Your car never runs very smooth when it is either adding or pulling fuel rapidly (opinion).

I notice your up there in elevation too (82 kPa's). Are you in Eastern Nevada?

Hopefully you can also mitigate your false KR. Feel free to look over my knock gain and knock retard recovery rates. Yes, they are fairly aggressive, but IMO it eliminates the 8-10 seconds of timing reduction from false knock. Real knock will come back. But be careful. It is hard to get good gas out West. 91 Octane is the max. I am not always sure there is as much of an altitude octane reduction as research shows. I.E., I still wish 93 octane was readily available.

RevGTO, keep trying. We will figure this out..

[joecar]

RevGTO,

I was just talking this over with Shawn, and he suggested several things:

- you have an LM1; if you have FlashScan V2 you should be using serial EQ/Lambda/AFR (do you have V1 or V2...?)

- did you recently do a free air calibration on the LM-1...?

- did you recently do a heater calibration on the Bosch sensor on the LM-1..?

[tatasta]

Robert (AKA Tatasta):

Thanks again for sticking with it and your constructive feedback:cheers:..

What is interesting about your tune and log is that your LTFTBENS were indicating that it wanted 'more fuel' (+ LTFTBENS). So I think your car is overall going to be 'happier'. When I first started out H/C and purchased EFILive i was running +12-+15 LTFTBENS. Your car never runs very smooth when it is either adding or pulling fuel rapidly (opinion).

I notice your up there in elevation too (82 kPa's). Are you in Eastern Nevada?

Hopefully you can also mitigate your false KR. Feel free to look over my knock gain and knock retard recovery rates. Yes, they are fairly aggressive, but IMO it eliminates the 8-10 seconds of timing reduction from false knock. Real knock will come back. But be careful. It is hard to get good gas out West. 91 Octane is the max. I am not always sure there is as much of an altitude octane reduction as research shows. I.E., I still wish 93 octane was readily available.

RevGTO, keep trying. We will figure this out..

Shawn (if you don't mind),

Yes, typical elevation here is 5100'-5500'. I am in NE Nevada (Elko County). 91 octane is the best available pretty much anywhere you look here. I have used toluene/marvel etc. up to a calculated 97 octane with no change to "logged KR". While on higher octane, I also pulled timing down to only 11* of advance in the affective area. The KR generally shows up when I roll on the throttle from a cruising situation to WOT (just like using a steep hill to log MAF/VE data when B0120 is set to 400 rpm), if that makes any sense.

From my experience with true KR, this lower to mid RPM KR with increasing TP and engine load is the easiest time to physically hear/feel knock. I have never experienced/detected the classic knock symptoms with this car. I have always been careful to get the best fuel available (Maverick fuel sux).

I will check out your knock settings/rates and compare to mine. I don't have alot of experience in those tables so looking over your settings will help me.

I currently have my trans pan off to change filter/oil. I will try to get back up and running by this afternoon (gotta work a little today) and log more data.

Also, thanks joecar for all of your help on this!

Thank you

[WeathermanShawn]

Shawn (first name ) is totally O.K.

Your KR does look false. Probably something with the tranny or suspension 'clunking' when you git the throttle. Trick is to keep it from pulling much timing, without affecting your knock sensitivity. It can be safely done. I have had both 'real knock' and false knock, so I have gotten fairly good at reading it. On my car I can hear pinging (real knock) if I turn off the stereo and roll down the window. False knock usually has one big event, but decreases as TPS opens. We can certainly address that after you get totally comfortable with your tune. It could be a separate Tutorial.

Perhaps the next very short note or Tutorial I may attempt is on how to filter logs to get the best data. Especially whether doing AUTOVE or CALC.VE Table. How to keep (or filter) for just steady throttle, using 3rd or 4th gear for airflow/fueling modeling, etc. So far I have seen not seen any appreciable IAT bias when utilizing the CALC.VE Table method. Therefore I am almost leaning to an Excel type merging of multiple log files. If you took 10 logs, your ability to filter VE Table spikes, etc would be greatly enhanced. One could also filter to a transient throttle of 0% per 100ms (very tight filter) to insure the best results. I realize this is straying a little from the main topic, but I know that is where most people get hung up on. For now, I would just say that filtering for your normal ECT range is critical for an accurate result, especially at Idle.

So for now, I will reiterate you are only 1-2 logs away from a 'perfect' CALC. VE Table. Once your MAF calibration and Trims are set, you can choose whether you keep LTFT's enabled. I would still recommend you keep closed-loop enabled, but the method can work with STFTS controlling the Trims.

Sorry for the book response. Maybe its time to start working on a short paper to describe filtering techniques that will help one iron out that VE Table..:grin:.

Later..

[tatasta]

Thanks for the reply. Not a long answer to me.I apprecite the info.
I am looking forward to running more log files. I will get work finished up and car back together to run again this afternoon.

Thanks again!

[RevGTO]

Shawn, Joe, thanks for the advice and comments. To answer your questions:
1. I did a free air calibration a couple months ago and it checked out ok. I could of course do it again.
2. I haven't calibrated the Bosch heater ... didn't know that that could be done/how to do it.
3. I'm running V1.
4. I've done some logs with the narrowband pids. I've attached one here.

With LT's it's always possible that I have an exhaust leak. Since installing them, I've never been able to get my fueling right; usually aggressive positive LTFT's were the problem. I swapped out my old catted Y for an ORY and that led to this round of tuning. I thought maybe if there was a leak with the old unit, the fresh install might correct it, but it's hard to know.

I can remember back a several years ago my AFR would hang close to stoich. That may have been before LT's, though. Attached is a log with narrowband pids.

[joecar]

To do heater cal:
-disconnect sensor from LM-1 controller,
-power on LM-1 for a few moments,
-power off,
-reconnect sensor,
-power on wait a minute or two.

[WeathermanShawn]

RevGTO:

One thing it looks like you might be in a Speed Density Closed-Loop Mode? I say this because I have never seen a MAF-enabled tune run exactly the same CYLAIR.DMA and DYNAIR.DMA unless the MAF has been disabled.

Its not that you can't perform the calculations without the MAF enabled, but are you sure you are throwing your LTFTBENS back into the MAF Calibration values? Based on your first log and LTFTBENS, I would have expected the MAF values to have lowered on the second run.

If you do indeed have the MAF enabled, and your CYLAIR and DYNAIR are that close, I would just tune it via the narrowbands and suspect your wideband is consistently reading too high. Do you have any logs where your open the throttle up and go into PE Mode? I would be curious to see your AFR under some load.

It appears from your log that the narrowbands are working properly. I would just tune it based on that. My hunch is that it is the wideband that is not correct. An exhaust leak usually goes the other way (+ LTFTBENS).

The other big question concerns your injectors. Are they stock?

Again, your narrowbands look like they are working. There is no logical reason why your AFR would not be at stoich at that point. Thats all I can think of on this end. Your tune may be near perfect (a little high on LTFTBENS), but a bad wideband reading could make this a long process.

Let us know..

[RevGTO]

RevGTO:

One thing it looks like you might be in a Speed Density Closed-Loop Mode? I say this because I have never seen a MAF-enabled tune run exactly the same CYLAIR.DMA and DYNAIR.DMA unless the MAF has been disabled. No, all my MAF fail settings, etc. are stock. Unless there is something wrong with my MAF, but no codes or anything like that.


Its not that you can't perform the calculations without the MAF enabled, but are you sure you are throwing your LTFTBENS back into the MAF Calibration values? Based on your first log and LTFTBENS, I would have expected the MAF values to have lowered on the second run. I pasted and multiplied and the values have changed, but it is by a very small amount. That's what I was talking about before when I mentioned that the LTFTBEN results tend to oscillate within a certain range from log to log ... sometimes higher, sometimes lower. I'm not trying to say that I did things right, but the base tune I started with was the result of a lot of trial and error MAF table tuning to get my AFR close to commanded on my OLMAF tune running from B3605.


If you do indeed have the MAF enabled, and your CYLAIR and DYNAIR are that close, I would just tune it via the narrowbands and suspect your wideband is consistently reading too high. Do you have any logs where your open the throttle up and go into PE Mode? I would be curious to see your AFR under some load.Unfortunately I didn't have my LM-1 hooked up to Flashscan when I went to the track. But based on the gauge, my PE was way lean - 12.8 commanded, 13.4 readout. I altered the PE tables to a 12.2 and I saw 12.6-12.8.


It appears from your log that the narrowbands are working properly. I would just tune it based on that. My hunch is that it is the wideband that is not correct. An exhaust leak usually goes the other way (+ LTFTBENS).

The other big question concerns your injectors. Are they stock?Yes, stock with stock tables.


Again, your narrowbands look like they are working. There is no logical reason why your AFR would not be at stoich at that point. Thats all I can think of on this end. Your tune may be near perfect (a little high on LTFTBENS), but a bad wideband reading could make this a long process.

Let us know.. But look at how close commanded/logged AFR got on this test run. But once the CL was enabled on the same tune, the results were way off again.

[joecar]

RevGTO, how many miles do you have on your NBO2's...?

[WeathermanShawn]

Rev:

I think what I am having trouble comprehending is I can not figure how you are able to have a commanded fuel so far from stoich. If I understand what you are doing, you are running a lean cruise or open-loop. How are you commanding 15.40 AFR with your OS? See though your wideband is closer to commanded, the actual values are just as 'lean' as all the other logs.

So what kind of tune are you running in your latest log that allows you to have that kind of commanded fuel? Can you post up your tune? See, I think we may be chasing something other than the technique here. I still say it is your wideband readings. If you hit PE (not in your log) and your actual AFR is leaner than commanded..that has nothing to do with the CALC.VE Table technique. At that point you are simply in open-loop and running lean. Closed-loop is not into play during PE mode.

Rev, I think if you could utilize a log that shows some PE and open-throttle and post up your tune, it might give us some more information to figure this out.

So, while true that your latest log shows actual AFR vs Commanded a lot closer..your wideband still is averaging 15.3 AFR..same as all your other logs. Try commanding like 14.0 (open-loop) and see if it still reads 15.3 AFR. What I am saying is that all your logs have shown ~ 15.3 AFR regardless of commanded. Doesn't that strike you as odd?

Let us know..

[WeathermanShawn]

RevGTO:

If you want you could E-mail your tune to either Joecar or myself. A lot of users are hesitant to post up a tune publicly (and sometimes with good reason).

We want to make sure you get it worked out. Though it does not look like it is the CALC.VE Table technique itself, the issues are preventing you from maximizing your tune.

Let us know if we can offer any further help..

[RevGTO]

I think what I am having trouble comprehending is I can not figure how you are able to have a commanded fuel so far from stoich. If I understand what you are doing, you are running a lean cruise or open-loop. How are you commanding 15.40 AFR with your OS? See though your wideband is closer to commanded, the actual values are just as 'lean' as all the other logs.

So what kind of tune are you running in your latest log that allows you to have that kind of commanded fuel? ..your wideband still is averaging 15.3 AFR..same as all your other logs. What I am saying is that all your logs have shown ~ 15.3 AFR regardless of commanded. Doesn't that strike you as odd?
Let us know..Good point. The commanded 15.4 comes from my modified B3605. I set it up that way to run a sort of "lean cruise" before the .cax download became available.

Two tunes are in effect here. I will PM them to you and Joe. I have had a couple more theories come to mind that I can test. One is that the CL tune does have the modified B3605 table. What would happend if I replaced that with stoich across the cruise range? If my AFR came down, that would indicate that somehow I'm not running fully from MAF as you suggested earlier.

I have some other ideas to isolate the issue but I'm going out of town for a couple of days and won't be available to test. I'll be back in touch.

[RevGTO]

RevGTO, how many miles do you have on your NBO2's...?They are used rears from a Corvette with supposedly about 40k on them. But I have tried new Densos, Bosch, etc., and they throw more codes than these, though these do throw codes.

[joecar]

With 40K on them they should be good... they should be good for 80-100K before they become slow.

[tatasta]

Shawn and Joe,

I logged a couple of runs today. Seems that things continue to get closer. DynAir and CylAir are noticeably closer. B0101 looks like the Alps. Does that matter? Or should I smooth it out without changing the values in the cells with no logged data. B5001 looks pretty good, too.

Bin 0003 has not been run yet and is based on logged data from log 0002. Bin 0002 followed log 0001.

Also, on Log 0002, the very last WOT pull is on N20. I did not intend to mix N20 into the log but, I forgot the nitrous master was on...

Thanks

[WeathermanShawn]

Robert:

You are definitely getting closer and closer. Its interesting to see how your CYLAIR and DYNAIR are starting to converge.

For now, I would just leave B0101 alone. I think as soon as you get the MAF Calibration down (zero out all the +LTFTBENS), then B0101 will make more sense. It will also help your WOT AFR..we need to get rid of the +LTFTBENS..

If you have not done it yet, perhaps it would be a good idea to reset your Trims. I take it you know how to do that? As soon as most of Trims are near zero, we can give you some hints on smoothing or at least applying stricter filtering to your data..and B0101.

Its looking better and better!

[tatasta]

Hey Shawn,

Thanks for the input. Actually, I'm not sure how to reset the trims. I am thinking it is in the DVT??? I am sure I can get it.

Is my MAF still being off causing the difference between commanded AFR and actual AFR at WOT? B3647 is set to 12.8:1 and I'm current ~ 13.2.

I will run again tomorrow or Saturday and post up.

[joecar]

Robert,

Like Shawn said, don't smooth it...

I might try this: leave the spikes alone, but bring the anti-spikes (the holes immediately next to the spikes) up a little (see attached, based on _0003.tun)... see what happens to these on the next iteration (and to LTFTBEN).

[tatasta]

Robert,

Like Shawn said, don't smooth it...

I might try this: leave the spikes alone, but bring the anti-spikes (the holes immediately next to the spikes) up a little (see attached, based on _0003.tun)... see what happens to these on the next iteration (and to LTFTBEN).

Got it. Thanks Joe.

[WeathermanShawn]

Yes, WOT AFR can not really be tuned to its ultimate accuracy until the MAF is fully calibrated. But, you are getting there fast. Also, carrying + LTFTBENS into PE or WOT can alter your fueling. The bottom line is that once all of your LTFTBENS are 'in line' everything gets easier.

Just remember during PE Mode or WOT your car is going open-loop. Sometimes the approach to calibrating WOT AFR is a little different than the 'non-PE Mode' portion of the tune.

We will get your B0101 straightened out. Whats important is that your actual airflows are getting very close.

I would bet another solid run or two..we will be there. If we need to reset Trims, we can walk you through it. Sometimes they react faster when you clear them out. Especially if you have a number of FTC Cells carry +LTFTBENS. I would hold off for now..Its interesting seeing your tune develop as is..:)

[joecar]

To reset trims to zero go into DVT.

[tatasta]

Joe or Shawn,

I have been contemplating adding a FAST 90/90 to my setup for a while now. In your opinion, would it be worth the money? I know money is subjective but, do you think it would be helpful at this elevation?

Just a thought.

Thanks

[WeathermanShawn]

I am pretty happy with 'just' the LS6 Intake. I think most of the airflow gains from the Fast 90 come >4000 Rpms..which makes sense. However I designed my combo to make all of its power prior to 6300 Rpms, so the Fast 90 would not benefit my combo. If you have the H/C and need that extra 6000-6800 rpm airflow..then it might be worth it.

Thats my opinion. It all depends on your combo and your goals...

[tatasta]

I am pretty happy with 'just' the LS6 Intake. I think most of the airflow gains from the Fast 90 come >4000 Rpms..which makes sense. However I designed my combo to make all of its power prior to 6300 Rpms, so the Fast 90 would not benefit my combo. If you have the H/C and need that extra 6000-6800 rpm airflow..then it might be worth it.

Thats my opinion. It all depends on your combo and your goals...

Thank you. That all makes perfect sense. I'm not too worried about rpms above 6300, since I shift at 6200. lol

Anyway, here is a log of Joe's tune with a smoothed B0101 and the resulting updated tune.

Car is running great!

[WeathermanShawn]

Robert:

I am really impressed how in 3-4 log runs you got MAF and Trims pretty much nailed. Notice how your CYLAIR and DYNAIR have converged quite nicely..

At some point I think that the LTFTBENS that you have in PE Mode may best be handled by resetting the fuel trims. That should really nail down the WOT portion of your tune. It looks like that 4-5 % + LTFTBEN has locked in one of your fuel cells. If you click under 'Calibration' on your tune (key turned to on), you can pull up the Long Term Fuel Trims matrix. You can hit 'reset', and that will clear out the fuel cells. They all should relearn quickly and will retain their accuracy.

See, a few of the fuel cells simply get locked (+ LTFTBENS). It is hard to keep adjusting the MAF to get rid of them. This technique makes it easier to clear them out.

Once you get PE mode 'normalized', you should see your VE Table values settle..and most of the spikes will be minimal.

Good job!:cheers:

[tatasta]

Shawn/Joe,

I ran my car today with my LTFTs enabled and still had some trouble with my WOT AFR. I was targeted at 12.8:1 but generally get low 12s, I disabled the LTFTs and ran the car again. The WOT AFR was more in line with my target.

One thing that sucks is that at idle, the car behaved better with the LTFTs enabled. I attached the logs both with and without trims. At WOT, there is significant positive trim. I reset the LTFTs prior to these runs.

What is likely to drive so much positive trim?

Thanks

[WeathermanShawn]

Robert:

You LTFT situation reminds me of a similar problem I had when I ran CLSD for a few weeks. If you break down your LTFT Cells in the Scan Tool, you will see that the WOT Fuel Cell (22) shares the same 'cell' as EVAP or Cold Idle.

I.E. when you first turn on your car on hit the cold idle parameters and EVAP kicks in..whatever +LTFT you have there carries over at WOT. I know it sounds bizarre, but I can post logs that show the same thing.

Here is how I got rid of that. You might want to change B0105 to 99 or 100% (from 95). Force that fuel cell to only be there at 100% TPS. You might want to change B0120 back to 4000 Rpms. It might be wanting more fuel at Idle..That VE Table could really come in handy at idle.

I agree about keeping LTFT enabled for a better idle. Mine is very smooth when I keep them active. You only other choice it to raise the MAF freq at cold idle to keep a negative fuel trim. You can log the FTC Fuel cells. If you need help on that let us know.

Thats the way I got rid of +LTFT's in FTC Cell #22. Once you get rid of that problem, working on nailing the Commanded Fuel vs actual is next. One note..Sometimes WOT AFR can vary per gear. I try to stick with 3rd or 4th (M6) depending on the local speed limit.

On a positive note your LTFTS, DYNAIR and CYLAIR and VE Table look good. Remember, I did not design GM's FTC matrix. Its a shame to have to figure out ways to outsmart it.

[tatasta]

Where would I find FTC Cell 22? I am not familiar with that.

Thanks for the help. I updated the settings you mentioned.

-Bob

[WeathermanShawn]

Robert:

I will have to review some of my logs to find a perfect example. However in the scan tool under "calibration' is where you find the LTFT FTC Matrix.

See attachment.

Also the Pid Selection for FTC Cells. If you look FTC WOT Cell is #22. The problem is that the emissions EVAP also 'shares' that cell. Perhaps one of our EFILive tuning geniuses will one day have a way of selectively clearing just one cell. It is one of the bizarre hazards of using LTFTS..

[WeathermanShawn]

If you get really crazy you can also log EVAPDC and see when EVAP is active. If you have good ears, roll down your window and listen for a loud 'whirring' or whistling sound when you drive. Thats EVAP. It is putting additional fuel back into the intake manifold (?). If you log STFTS, you can watch them skyrocket upward for about 3-5 seconds as the EVAP canister unloads its fuel vapors.

You will usually see it occur in FTC Cell 22..You need that cell 'clean' at WOT. It has messed with my WOT AFR more than any other single tuning issue!

[Wolfie]

Ya know.....
I've been waiting until Shawn's setup was "perfected" before I tried my hand at tuning the ve and maf again... So I finally get back to driving (was off for a month (heart operation)) and jumped in the van, and re-enabled the O2 sensors
(was running ol) and damn if my O2 sensors went bad.... the right bank reading 10x the left bank. So now I just had to order 2 new sensors (ordered NTK) and while I am at it, order a new bosch sensor from PLX.... someday... sigh...

[joecar]

Wolfie, hope you're doing good after your H.O... :cheers:

[WeathermanShawn]

Yea, Wolfie heal up. I mean you get a heart operation and new O2 sensors all at the same time..:grin:..

In all seriousness, hope you feel better! Thanks for your support on the Tutorial..

..Shawn..

[tatasta]

Ya know.....
I've been waiting until Shawn's setup was "perfected" before I tried my hand at tuning the ve and maf again... So I finally get back to driving (was off for a month (heart operation)) and jumped in the van, and re-enabled the O2 sensors
(was running ol) and damn if my O2 sensors went bad.... the right bank reading 10x the left bank. So now I just had to order 2 new sensors (ordered NTK) and while I am at it, order a new bosch sensor from PLX.... someday... sigh...

Welcome back to the world Wolfie. I am 6 weeks out of a c-spine surgery (neuro pain related). I was off for a month too but was out logging data on week two...lol. Take care.

[joecar]

Hope you're doing good too RT...:cheers: