Lean transition into PE and lean spot during WOT

[alexl226]

What do you know. I changed back to VE only and set pe to enable at 255Kpa, dumped in 1.22 for the values 100kpa + in the airflow gain table and the spike is completely gone. What is the downfall in doing this? my table does maxes out at 180kpa...


I'd be fine with leaving it VE only, but a few issues I couldnt ever fix when doing this:
1) cold start crazy idle swing. (goes lean rich lean rich for a number of cycles then levels out)
2) Tip in lean, tip out rich. Decel especially rich
3) Idle continues to get richer. If it sits for a few minutes could be down into the 12's. But Pulsewidth shows .9 which normally equates to lean/stoich.

Basically seems as though there is always some underlying fueling tables being used. Changes to VE seem to never be right. And overall it feels "sluggish" compared to using open and closed loop. Mileage was never good in pure VE also..

Attached is current tune and log. Airflow matching pe was me overlying the pe table in the airflow in hopes it may be defaulting to a table that uses that....

veonly pe disabled.hpl

airflow matching pe.hpl

07 Stock Suburban2_VEonly_AirflowGain.hpt

[Ranger6202]

Forgive me as I haven't read through this whole thread, but I had a similar problem with a turbo'd gen III Silverado. My fix was to change the dynamic zone # based on a lower MAP value. I would log and pay attention to what general MAP value my lean spikes would generally start to happen as it wouldn't follow the NB's. I then used that MAP value as my new dynamic air trigger to set ZONE 1 which would still give me closed loop control, but went more MAF based. This also transitioned into WOT a lot better if the MAP were to continually increase. My theory is that the predictive calculations in the default cruising/transition zones don't predict aggressively enough against increased airflow vs. smaller TPS% from a boosted setup. I wish I understood how the ZONES were setup in the relationship of MAP VS. MAF and predictive air calculations. As I would probably adjust my modes accordingly to that knowledge rather than just triggering good ol zone 1. The fix before this was to run straight VE or straight MAF. Triggering zone 1 at a lower map and TPS% let me run both without the lean spikes or crappy WOT transitions.
I have only worked on a few gen iv's and have not had to mess with any of the dynamic stuff yet. I see they have the zones boundaries and prediction coefficients, as well as the ve correction factor and of course the RPM triggers. By chance you might be able to determine which zone you are in that is causing the lean spikes or which zone transition into WOT is going lean, and possibly adjust the MAP prediction coefficient values for those zones alone. Or maybe I just like typing and making crap up as I go. Just a suggestion that might lead you to solve your problem. good luck, let me know what you think or if I should just disable my keyboard altogether to save others from my rambling.

[alexl226]

I up graded to the custom os so I have the classic ve tables. And it happens at different map values. The only thing consistent is it happens 2.xx seconds after pe is enabled. I do log fuel trim cell and when it happens it does bounce to trim cell 4 and 12 I believe..

[alexl226]

Forgive me as I haven't read through this whole thread, but I had a similar problem with a turbo'd gen III Silverado. My fix was to change the dynamic zone # based on a lower MAP value. I would log and pay attention to what general MAP value my lean spikes would generally start to happen as it wouldn't follow the NB's. I then used that MAP value as my new dynamic air trigger to set ZONE 1 which would still give me closed loop control, but went more MAF based. This also transitioned into WOT a lot better if the MAP were to continually increase. My theory is that the predictive calculations in the default cruising/transition zones don't predict aggressively enough against increased airflow vs. smaller TPS% from a boosted setup. I wish I understood how the ZONES were setup in the relationship of MAP VS. MAF and predictive air calculations. As I would probably adjust my modes accordingly to that knowledge rather than just triggering good ol zone 1. The fix before this was to run straight VE or straight MAF. Triggering zone 1 at a lower map and TPS% let me run both without the lean spikes or crappy WOT transitions.
I have only worked on a few gen iv's and have not had to mess with any of the dynamic stuff yet. I see they have the zones boundaries and prediction coefficients, as well as the ve correction factor and of course the RPM triggers. By chance you might be able to determine which zone you are in that is causing the lean spikes or which zone transition into WOT is going lean, and possibly adjust the MAP prediction coefficient values for those zones alone. Or maybe I just like typing and making crap up as I go. Just a suggestion that might lead you to solve your problem. good luck, let me know what you think or if I should just disable my keyboard altogether to save others from my rambling.

I've now read over this a number of times. And think I now grasp what you are saying. How did you log which zone you were in? and what logic did you use in changing them?

[Ranger6202]

I haven't logged what zone i'm in. I've tried finding out how, but haven't found where you can actually log it, maybe using the STATUS BIT DISPLAY? Generally you can figure out what zone you're in based on Map and RPM, but it won't help you know how the predictive calculations are done (ie 80%maf:20%SD, or 60%maf:40%SD, or whatever) I wish i knew the right people that could just read the predictive calculations in the logic and then somehow give us a caveman's description of each zone.
The only info about the zones i've gone off of is: https://www.hptuners.com/help/vcm_ed...ow_dynamic.htm
As you'll see Zone 1 is considered a WOT zone, and from what i have read it's 100% MAF based in all calculations. So knowing that, i trigger my WOT zone with my MAP increase to avoid lean spikes when cruising. Is there a better zone?....probably. Both Gen III's and IV's have the airflow mode table, i'm assuming this is where you can set the right zone per amount of airflow. But even then, i'm totally assuming everything as there isn't a whole lot of info out there on the zones and airflow modes.

[brentsten]

What do you know. I changed back to VE only and set pe to enable at 255Kpa, dumped in 1.22 for the values 100kpa + in the airflow gain table and the spike is completely gone. What is the downfall in doing this? my table does maxes out at 180kpa...


I'd be fine with leaving it VE only, but a few issues I couldnt ever fix when doing this:
1) cold start crazy idle swing. (goes lean rich lean rich for a number of cycles then levels out)
2) Tip in lean, tip out rich. Decel especially rich
3) Idle continues to get richer. If it sits for a few minutes could be down into the 12's. But Pulsewidth shows .9 which normally equates to lean/stoich.

Basically seems as though there is always some underlying fueling tables being used. Changes to VE seem to never be right. And overall it feels "sluggish" compared to using open and closed loop. Mileage was never good in pure VE also..

Attached is current tune and log. Airflow matching pe was me overlying the pe table in the airflow in hopes it may be defaulting to a table that uses that....

veonly pe disabled.hpl

airflow matching pe.hpl

07 Stock Suburban2_VEonly_AirflowGain.hpt

No downfall really.. It's preference whether you run VE, MAF or mixed. I've tried all 3 at the track and always went faster in MAF only, so that's how I run it all the time. Plus, no matter how many times I retuned VE, the transients were always erratic and mileage would suffer. If you're exceeding 180kpa MAP, it will use the EQ ratio in the 180kpa cell for anything higher.

[alexl226]

This is driving me crazy...

This is the factory setting for the zone map boundaries. In trying to understand how these work, I haven't seen any examples with different values for MAP zone 0-1 through different rpm zones...



If I flat lined the Map Zone 0-1 cells, and changed the map zone 4-5 to 120 kpa, is my understanding of the 29 zones correct?



Also, what is the proper way to deal with the prediction coefficients? I still have stock values, that vary in different zones.




And I've been looking into the transient stuff, such as evap and fuel mass gain impact, anyone mess with these? But at 2.xx seconds after a new commanded AFR, I would have to assume transient stuff is over....

[GHuggins]

Zero out the prediction coefficients as previously discussed in this section and your problem should be gone.

[alexl226]

Zero out the prediction coefficients as previously discussed in this section and your problem should be gone.

I gave that a go.

I'm still getting the spike randomly. It now appears to be more stable during transients while in PE.

Prior to changing coeff's to 0, for example, it would go really lean and fight hard to work back down to commanded during map increases. For example, if you were holding 120kpa then cracked the throttle more to 135kpa. It would pretty much hit stoich and sloooowly work its way back to commanded.

Current Tune: 07 Stock Suburban2_Backtostock2_pe2_maf2_boostPE1_spark3_4l80e2_1_map120_ltftDisabled_flexfueldi.hpt

2 Logs After Coeff Change to Zero: 000.hpl , 000_1.hpl

Frames 5140, 7300, 11500, 13500, 16500 (caused knock)....are spots where I entered boost in the first file. As you can see, fueling during this and the spike severity is inconsistent.

Config File: ~Burban Setup~~LAMBDA1~LeanDip_warmupcycles.cfg

[Xjas]

I stumbled on this thread after much searhing for a similar issue.
My car is not as modified as the others with this issue, as per my sig its a cam only 6L, stock fuel system including injectors, stock injector data and transient fueling.
I have a very repeatable rich spike 2 seconds after going WOT, regardless of what RPM or g/cyl or MAF airflow the engine is at the rich spike alway occurs ~2 seconds after going WOT or entering PE, although it is more noticeable at lower RPM, in the screen shot below I can noticeable feel the car stutter for a second.
This has me really perplexed, the injector pulse width average doesn't hardly change, neither does the airflow but the car goes pig rich for a second.

[Niemer]

Ha, leave it to the Australian to have an upside down spike.

It is interesting though, that yours is opposite of the others, mine included. Makes me wonder if there is hope yet.

[WS6HUMMER]

OLMAF WITH NO PREDICTION COEFF.hptlean spike.pngI think I may be victim to this problem, I've noticed a lean startup condition & this 2.1 second in PE transition. Could someone please look at my file and make sure I've not overlooked something? I've read this entire thread & tried everything mentioned. I'm dynoing the car tonight and the only thing I havent verified was IDC and a pressure drop. I'm on 3.0 so I've just now got IDC setup. Also in the screen shot you'll notice the delay, the AFR looks flat/blockish but actually starts at 14.69 and ends up at 13.40. If someone could just check my file for my sanity before I get it strapped down I'd really appreciate It. It's an 11 silverado E38.

lean spike.png
OLMAF WITH NO PREDICTION COEFF.hpt

[schpenxel]

File is tuner locked

[WS6HUMMER]

File is tuner locked

Very sorry about that, I unlocked & saved it but tuner lock was on so I guess it locked it again somehow. I fixed it and downloaded it to check myself.

[schpenxel]

Still showing locked (assuming you updated the attachment in the earlier post)

Make sure you cleared out slave and master tuner ID's and HPT Lock ID. I haven't messed with them enough to know which does what.. I think it's the "remote tuning" section you need to clear out though.

[WS6HUMMER]

OLMAF WITH NO PREDICTION COEFF.hpt

Still showing locked (assuming you updated the attachment in the earlier post)

Make sure you cleared out slave and master tuner ID's and HPT Lock ID. I haven't messed with them enough to know which does what.. I think it's the "remote tuning" section you need to clear out though.

Yeah showing unlocked on my end. I cleared out the master/slave like you said. Does the interface have to be attached for this?

[schpenxel]

I dunno then.. keeps saying

"HPT Lock Error. Unable to open file. This file is HPT locked by someone else. To be able to open this file, they must first unlock the file"

I don't think the interface has to be connected but I've never really messed with it.

[WS6HUMMER]

I dunno then.. keeps saying

I don't think the interface has to be connected but I've never really messed with it.

Thanks for your patience, let me know if the file I posted in post #136 is in fact unlocked.

[schpenxel]

yep, same thing

[WS6HUMMER]

yep, same thing

Ok I think I got it this time. It showed unlocked but still had a lock id code. What I did was double click below the lock ID # and make that blank, should be good now. Crossing my fingers lol.

OLMAF WITH NO PREDICTION COEFF.hpt