PDA

View Full Version : What delta are you guys seeing between commanded and actual AFR?



johnsZ06
April 16th, 2007, 11:38 AM
I tuned a car's WOT AFR the other day on a dynojet with wide band and I had to set the PE vs RPM to 11.8 to get 12.8 on the wideband. What are you guys seeing?

dc_justin
April 16th, 2007, 11:48 AM
I tuned a car's WOT AFR the other day on a dynojet with wide band and I had to set the PE vs RPM to 11.8 to get 12.8 on the wideband. What are you guys seeing?

I command 12.8 to get 12.8. :D

What I think you will find is that most here will "fix" the underlying table(s) that cause a disparity in commanded vs actual, rather than modify the PE table to get it where you want it. With it that far off, (11.8 to get 12.8), then I can only assume that after a bit of driving, the trims are going to trend positive, which will likely carry over into PE mode and richen it up a bit more, bringing it closer to 11.8 when all is said and done...

Another thing to be aware of is if you're having to command more fuel to get your target, then everything will be off on the tune. Spark reference, torque calculation (for automatic transmission cars), etc.

johnsZ06
April 16th, 2007, 11:59 AM
Well, the thing is I was kind of rushed to get the AFR within target because of limited pulls so I had to do what I had to do. The one thing that worried me was the car was getting around 8 degrees of KR which I couldn't explain unless it was bad gas. The car was using the basic spark map with about 22 degrees on top and was still pinging like crazy. The whole thing made me very tense but at least he won't detonate due to being lean.

I scanned the parameters of each pull and I couldn't see why he would be running any leaner vs. commanded. Of course, that's taking into consideration the wideband was working properly.

One thing I did note was the car had 122k miles on the clock and might of had a a lot of carbon buildup which might explain the KR.

dc_justin
April 16th, 2007, 12:43 PM
Well, the thing is I was kind of rushed to get the AFR within target because of limited pulls so I had to do what I had to do. The one thing that worried me was the car was getting around 8 degrees of KR which I couldn't explain unless it was bad gas. The car was using the basic spark map with about 22 degrees on top and was still pinging like crazy. The whole thing made me very tense but at least he won't detonate due to being lean.

I scanned the parameters of each pull and I couldn't see why he would be running any leaner vs. commanded. Of course, that's taking into consideration the wideband was working properly.

One thing I did note was the car had 122k miles on the clock and might of had a a lot of carbon buildup which might explain the KR.

So on the spark map it showed about 22* up top or you saw about 22* in the EFILive scanner?

johnsZ06
April 16th, 2007, 01:35 PM
I saw 22.5*.

johnsZ06
April 16th, 2007, 01:41 PM
Here's the last pull. When I opened the tune tool it showed 24* commanded.

SSpdDmon
April 16th, 2007, 01:49 PM
That to me looks like 30* commanded (22.5+7.5).

johnsZ06
April 16th, 2007, 02:12 PM
That to me looks like 30* commanded (22.5+7.5).
I know what you are saying but when I opened up the tune tool it was showing 24. So I'm a little confused as well.

dc_justin
April 16th, 2007, 02:18 PM
I know what you are saying but when I opened up the tune tool it was showing 24. So I'm a little confused as well.

What PID are you using to display airmass (g/cyl)?

dfe1
April 16th, 2007, 02:23 PM
I see a few things that are potential problems. The knock is long and repetitive indicating low quality gas or an excessive amount of oil in the combustion chamber. With 122,000 miles the rings, valve seals or both may be passing a lot of oil. What was engine coolant temp during the dyno runs? If it was 190+, that could be a contributing factor. I assume the PLX wide band was not connected since it's showing 16.8:1. Have you checked to make sure you don't have some spark being added through one of the coolant temp adjust tables? As SSpdDmon has pointed out, something is calling for 30 degrees of spark.

johnsZ06
April 16th, 2007, 02:23 PM
What PID are you using to display airmass (g/cyl)?

Actually, that PID isn't active. Would that make a difference on what I see on the spark map? Maybe I need some edumacation here. :lol:

johnsZ06
April 16th, 2007, 02:28 PM
I see a few things that are potential problems. The knock is long and repetitive indicating low quality gas or an excessive amount of oil in the combustion chamber. With 122,000 miles the rings, valve seals or both may be passing a lot of oil. What was engine coolant temp during the dyno runs? If it was 190+, that could be a contributing factor. I assume the PLX wide band was not connected since it's showing 16.8:1. Have you checked to make sure you don't have some spark being added through one of the coolant temp adjust tables? As SSpdDmon has pointed out, something is calling for 30 degrees of spark.
Temp was at 190 but was cooler on previous runs and still pulling lot's of timing.

PLX was not connected at the time.

I've looked for anything that might be adding (?) to the timing table but couldn't find anything.

I also looked at the low octane table and at that particular load and RPM was set at 20*, so where is the 30 degrees coming from?

dfe1
April 16th, 2007, 02:32 PM
Actually, that PID isn't active. Would that make a difference on what I see on the spark map? Maybe I need some edumacation here. :lol:
The PIDs you're using or not using shouldn't have any impact on other PIDs, unless it's a calculated PID. In that case the calculated PID would come up as invalid unless you had selected the other PIDs necessary for the calculation.

dc_justin
April 16th, 2007, 02:34 PM
Actually, that PID isn't active. Would that make a difference on what I see on the spark map? Maybe I need some edumacation here. :lol:

Absolutely. How do you know what area of the spark table you're referencing then?

Let's get back to the commanding 11.8 to get 12.8. That's approx 9% off of what it should be.
One of a few things could be happening. Could be an exhaust leak causing the WB to read incorrectly. If that's not the case, then either the PCM's airmass calculation is 9% on the low side, or the injector flow rate is not correct, either by incorrect settings in the table or via lower than expected fuel pressure. Let's assume for now that the IFR is dead on. This would mean that the airmass calculation is low, either from an inaccurate VE table if in SD, or from a MAF sensor that is not reporting accurately (in some way or another). This 9% low calculation doesn't just affect the amount of fuel supplied, it also affects spark reference, torque calculation and whatever other internal functions rely on airmass/airflow. It would result in the PCM referencing a column 9% lower on the spark table than it should for the actual amount of air entering, which would yield higher than expected spark timing.

Make sense?

johnsZ06
April 16th, 2007, 02:45 PM
Absolutely. How do you know what area of the spark table you're referencing then?

Let's get back to the commanding 11.8 to get 12.8. That's approx 9% off of what it should be.
One of a few things could be happening. Could be an exhaust leak causing the WB to read incorrectly. If that's not the case, then either the PCM's airmass calculation is 9% on the low side, or the injector flow rate is not correct, either by incorrect settings in the table or via lower than expected fuel pressure. Let's assume for now that the IFR is dead on. This would mean that the airmass calculation is low, either from an inaccurate VE table if in SD, or from a MAF sensor that is not reporting accurately (in some way or another). This 9% low calculation doesn't just affect the amount of fuel supplied, it also affects spark reference, torque calculation and whatever other internal functions rely on airmass/airflow. It would result in the PCM referencing a column 9% lower on the spark table than it should for the actual amount of air entering, which would yield higher than expected spark timing.

Make sense?
Absolutely makes sense. I know the IFR is same as stock so that shouldn't be a problem. I'm not sure about the fuel pressure, haven't checked it. The MAF table was tweaked (by me) to bring LTFT's in line because of an added low restriction air filter.

I guess the MAF table might be the cause of the disparity between the commanded and measured AFR! Okay, at least were getting somewhere!

Let me digest on this some more. I appreciate your inut dc_justin, but now it's time for us east coasters to go to bed. :wave:

johnsZ06
April 16th, 2007, 11:54 PM
I did notice that at WOT there is a 9% difference between MAF airfflow and calculated airflow so maybe that explains the PE delta. I still can't figure out how I'm loosing 7.5* when when commanded is 24 and I'm seeing 22.5. :nixweiss:

SSpdDmon
April 17th, 2007, 12:13 AM
If you want to know where the appropriate spark reference is, you need to log the following:

GM.CYLAIR_DMA when MAF is active
GM.DYNCYLAIR_DMA when in speed density

These are considered crucial for tuning in my opinion....just like RPM, coolant temp, etc. Any .pid ending in "_DMA" means it's a "Direct Memory Access" .pid (exclusive to EFILive :)). They are extracted from the operating RAM as the current operating parameters. More info is available in the scanner by hitting "F1" and scrolling to page 88 (90 of 192 in Adobe).

Doc
April 17th, 2007, 12:15 AM
The AutoVE tuning procedure works great. Fuel Trims? We don't need no stinking fuel trims.;)

johnsZ06
April 17th, 2007, 12:36 AM
If you want to know where the appropriate spark reference is, you need to log the following:

GM.CYLAIR_DMA when MAF is active
GM.DYNCYLAIR_DMA when in speed density

These are considered crucial for tuning in my opinion....just like RPM, coolant temp, etc. Any .pid ending in "_DMA" means it's a "Direct Memory Access" .pid (exclusive to EFILive :)). They are extracted from the operating RAM as the current operating parameters. More info is available in the scanner by hitting "F1" and scrolling to page 88 (90 of 192 in Adobe).

Actually, I was logging the GM_CYLAIR and was able to track the spark during playback.


The AutoVE tuning procedure works great. Fuel Trims? We don't need no stinking fuel trims.
True...very true. :)

SSpdDmon
April 17th, 2007, 12:55 AM
Actually, I had logged both and was able to track the spark during playback.

Well, it looks like something is not referencing right based on the two images you posted on the first page of this thread. The scanner appears to be highlighting a column in the tuning software way above what a N/A car would normally see. I would double check the .pids you're logging and make sure there's only one grams/cyl .pid selected (one of two I mentioned above depending on whether or not your MAF is active). Any more than one grams/cyl .pid is just eating up your logging bandwidth.

johnsZ06
April 17th, 2007, 01:08 AM
Well, it looks like something is not referencing right based on the two images you posted on the first page of this thread. The scanner appears to be highlighting a column in the tuning software way above what a N/A car would normally see. I would double check the .pids you're logging and make sure there's only one grams/cyl .pid selected (one of two I mentioned above depending on whether or not your MAF is active). Any more than one grams/cyl .pid is just eating up your logging bandwidth.

I was looking at the PIDS I had selected last night and I had logged quite a few in the AIR category, more than I need, but was left over from earlier tuning days.

I'll check it when I get home tonight.

When logging WOT pulls, what do you normally log as a minimum?

dc_justin
April 17th, 2007, 02:11 AM
If you want to know where the appropriate spark reference is, you need to log the following:

GM.CYLAIR_DMA when MAF is active
GM.DYNCYLAIR_DMA when in speed density

These are considered crucial for tuning in my opinion....just like RPM, coolant temp, etc. Any .pid ending in "_DMA" means it's a "Direct Memory Access" .pid (exclusive to EFILive :)). They are extracted from the operating RAM as the current operating parameters. More info is available in the scanner by hitting "F1" and scrolling to page 88 (90 of 192 in Adobe).

GM.DYNCYLAIR will equal CYLAIR_DMA when the MAF is enabled and DYNCYLAIR_DMA when MAF is disabled. Select it and it will always be correct, whereas the two _DMA ones may or may not.

SSpdDmon
April 17th, 2007, 02:25 AM
I was looking at the PIDS I had selected last night and I had logged quite a few in the AIR category, more than I need, but was left over from earlier tuning days.

I'll check it when I get home tonight.

When logging WOT pulls, what do you normally log as a minimum?
Well, your spark reference .pids that I mentioned earlier are under Tune...not Air. I log these DMA's for peace of mind and have the various assortments of .pids saved by speed density or MAF. So, I know I'll always have the right ones to log when I select the pre-sorted .pid file. That way, I know I get it 'straight from the horses mouth' every time. The .pids I usually log depend on the setup. I run open loop with a WBO2 and like to see the following:

SAE.TP
GM.CYLAIR_DMA (or GM.DYNCYLAIR_DMA for speed density)
SAE.MAF (or GM.DYNAIR for speed density)
CALC.BEN_* (your WBO2)
GM.AFR
SAE.ECT
SAE.RPM
EXT.AD1 (or EXT.AD2 if WBO2 is attached to AD2)
SAE.SPARKADV
SAE.IAT
SAE.MAP
GM.MAFFREQ (if MAF active)
GM.KR
SAE.VSS
CALC.AFR_* (your WBO2)

If you're not running open loop with a WBO2, some other .pids to consider are:
GM.HO2S11
SAE.LONGFT1
SAE.SHRTFT1

At any time, it's good to monitor the following as well:
GM.FTC
SAE.FUELSYS
GM.IBPW1
CALC.INJDC1

johnsZ06
April 17th, 2007, 03:42 AM
I never understood the difference between the DMA and non-DMA PIDs. Now I do so thanks for the lesson. :)

I guess I will play around this weekend with some of the DMA PIDS just to see what the difference is. This might be the excuse I need to go back and look at my own tune. :thankyou2:

johnsZ06
April 17th, 2007, 11:44 AM
Hey guys. I double checked the PIDS I had selected and the GM.DYNCYLAIR_DMA was one of them. It wasn't selected on the chart though. So with that in mind, does anyone have any ideas where the 7.5 * degrees of KR when 24.7 is commanded and 22.5* showing?

I guess I'm back to square one on this particular problem.

SSpdDmon
April 17th, 2007, 12:44 PM
Was this in SD or while the MAF was active? If the MAF was active, then you need the GM.CYLAIR_DMA .pid. That one is for speed density.

The other thing to look into is the spark adder .pids. You'll need to grab two seperate logs to see where all the adders are contributing to the delivered spark at the cylinder since they'll take up more than 24 channels. But, they'll help you track down which tables are influencing your timing faster than trying to guess about it.

johnsZ06
April 17th, 2007, 01:04 PM
Was this in SD or while the MAF was active? If the MAF was active, then you need the GM.CYLAIR_DMA .pid. That one is for speed density.

The other thing to look into is the spark adder .pids. You'll need to grab two seperate logs to see where all the adders are contributing to the delivered spark at the cylinder since they'll take up more than 24 channels. But, they'll help you track down which tables are influencing your timing faster than trying to guess about it.

I was in MAF but it should be close since I tuned in SD and calibrated the MAF to match calculated.

As far as spark adders, I looked at all the spark PIDS and can't see where spark is added.

SSpdDmon
April 17th, 2007, 02:56 PM
send me a copy of the current tune and a list of make/model/mods if you don't mind. maybe a second set of eyes would be helpful. if you have any logs from the current tune, that might help as well.

saj79 at aol dot com
Jeff

johnsZ06
April 17th, 2007, 11:19 PM
send me a copy of the current tune and a list of make/model/mods if you don't mind. maybe a second set of eyes would be helpful. if you have any logs from the current tune, that might help as well.

saj79 at aol dot com
Jeff

I'll do that when I get home tonight. Thanks for taking the time! :cheers:

SSpdDmon
April 18th, 2007, 11:19 AM
Email sent w/ new tune. :D

Chevy366
April 18th, 2007, 12:03 PM
Email sent w/ new tune. :D
Could you share your findings ?
Seeing same kind of spark issue , just wondering .

SSpdDmon
April 18th, 2007, 01:42 PM
There wasn't a whole lot changed in the tune. I made some minor adjustments based on my '02 f-body, a recommendation to log fewer channels to get a better sample rate (less than 25 channels), and suggested that the owner of the car clean his MAF sensor. I think that might be the main culprit as a dirty MAF can cause AFR problems as well as put you into a different timing cell. The other issue may have to do with the 122K miles on the odometer. Over time, carbon buildup in the cylinders can be responsible for some knock retard. But, that still doesn't explain the higher commanded timing that he was seeing like a dirty or uncalibrated MAF would.

johnsZ06
April 19th, 2007, 10:47 AM
Email sent w/ new tune. :D
Thank you kind sir! :cheers:

BTW, I know the guy drives this car like a granny so carbon buildup is probably a contributing factor to the KR.