Sudden Inconsistent Results While MAF Tuning

[ApexClone]

First time tuner here. LS2 Corvette that I put a Texas Speed cam in last week (228/232 .600" 112LSA), the car also has longtubes with cats.

I've been doing a lot of research etc and everything has gone fine so far. I did some initial idle changes to get the car to idle (ignition timing, idle speed) based one what I saw for recommendations. Then I moved on to the MAF tune. My process is as follows:

Disable Dynamic Airflow
Disable O2 Readiness
Disable LTFTs
Disable COT
Disable DFCO and CFCO

Log EQ Ratio Error with very light acceleration to redline in usually 2nd gear. Once I had the MAF error close I moved on to VVE tuning. Same basic idea (except disabling MAF). That went well but wasn't 100% before I decided I wanted to recheck the MAF tune since I've learned a bunch in the past week from reading/hands on experience.

Here's where my trouble lies. Yesterday I went out to do more MAF tuning with the same process yet I started using more throttle (think 20-25% the first round, ~50% this round). I made some slight changes but nothing huge but now I was able to hit higher frequencies due to increased acceleration. I did try one full throttle pull during this. I made probably 4-5 ~20 minutes trips, making changes as necessary in this way. I thought I was all set, had maybe a couple trips left and I would be happy. Then I decided to try going back to very slight acceleration. All of a sudden my EQ Error was way off from 5650Hz and up (around 5% lean) however I coudl only get data up to 6600Hz again due to the light acceleration. This was repeatable. I made two trips with one adjustment in between and got it back close to 0% during very light acceleration. However, now if I go back to "normal" acceleration (~50% tps) its rich in those areas.

My main question is, why is it doing this? At the same frequency shouldn't the EQ Error be the same no matter how I'm accelerating? I just dont see what going from ~20% tps to ~50% tps would change to make that much of a difference.

Second, if I'm not messing something up, how should I tune it? I would think I would want to tune it for how I normally drive.

Third, I've read that I shouldn't get into PE during MAF tuning as it can skew results. If the data I'm logging is the error in commanded EQ vs actual EQ, why would that matter? The commanded EQ is changing when I hit PE so wouldn't that data be just as valid?

Last, E38 ECUs have a high and low MAF calibration table. The low ends at 5800Hz, the high begins at 5800Hz. Should the values be the same in both tables at that frequency? In the stock tune they were. I notice that while I'm logging the EQ Error, it seems as though 5800Hz on the high table is always getting data even when the frequency is clearly below 5800HZ (such as at idle). The count is still going up then and the data is changing for that cell. Should I just paste the value from the 5800Hz cell on the low table onto the 5800Hz cell on the high table?

Any help will be appreciated. This has gotten me pretty discouraged as I expected to be "done" with MAF Tuning tonight.

I attached a couple logs and tunes. They go in order - 02 MAF Tuning.hpt is the tune used in 03 MAF Tuning.hpl. Then 03 MAF Tuning.hpt is the tune used in 04 MAF Tuning.hpl. This is when I first noticed the sudden change noted above. When I made the changes in 03 MAF Tuning.hpt, the log for that on 04 MAF Tuning.hpl is when I went back to very slight acceleration and got much different (leaner) results at higher frequencies.

[blindsquirrel]

Post O2 Test & Catalyst Test still enabled, which'll do unpredictable things to the mixture in the background.

Yes, the corresponding cells between low & high MAF should be the same.

VVE is still used for transients in 'MAF-only', that could be part of what you're seeing if the VVE is off?

[ApexClone]

Post O2 Test & Catalyst Test still enabled, which'll do unpredictable things to the mixture in the background.

Could you explain when you mean by this? I have the O2 Readiness ECT table set at 400 across the board. The LTFT are set at 400 Min ECT and 1 Max ECT. I'm not getting any STFTs or LTFTs while I'm logging. I have the Cat Overtemp disabled. I couldn't find any other settings to might match what you're saying but there's a good chance I'm just overlooking them.

Yes, the corresponding cells between low & high MAF should be the same.

Thanks for confirming that.

VVE is still used for transients in 'MAF-only', that could be part of what you're seeing if the VVE is off?

Is there any way to completely disable VVE in that case?

Could you give me an indication of how I should drive when doing MAF Tuning? Is very slight acceleration from idle to redline correct or should I be driving more like I regularly drive? Should I be doing any WOT driving? I would think so if I'm going to hit the higher frequency cells.

[blindsquirrel]

screenshot.04-06-2022 09.16.46.png

[ApexClone]

This is all I'm showing in that tab.

Screenshot 2022-06-04 105154.png

[blindsquirrel]

It's been removed in current versions of the editor. Lawyers were involved.

[ApexClone]

Its always the damn lawyers.

I think I figured out my issue. I was datalogging too much with too many transitions. I did some more this morning with just doing a pull from idle to redline and got consistent results. Then did a pull at WOT to hit the higher cells and got consistent results. I believe I'll have it within 1% in a couple more runs.

[kingtal0n]

Hmm make sure you are looking at "advanced" settings and not basic. He might be joking... I can't tell...

[kingtal0n]

https://forum.hptuners.com/showthrea...l=1#post578480

Not sure if this will help you but hopefully with luck, ... good luck

[ApexClone]

Yeah I'm definitely running in advanced and I don't have those settings. Checked the navigator also.

Well, I thought I was all set after a few runs this morning but it has gone back to being inconsistent. "Pulls" back to back with no changes are not the same and I can't figure out why. Its only happening at low TPS slight acceleration. WOT is perfectly consistent. I attached a couple logs showing this inconsistency. Exact same tune. Make a run from low rpm to just before PE kicks in with very slight accel. Get to a turn around spot. Save the log. Turn around and repeat.

Is there anything else fuel/emissions wise that I could be overlooking to zero out? I read something about injector tip temp throwing off fueling right after writing a tune. I'm not sure if I can do something about that if its even a concern?

I'm going to pour through info tonight and hopefully get more consistent results in the morning.

[dhoagland]

Didn't look real close.....

But 2 logs I looked at have 20 - 80 counts per cell....

EDIT,,,, Sorry I had my filters looking at your data, I see you have upward of 1000 cell hits..

That isn't much data, especially in "Transient" state.

Do you have any filters for your data?

Yes Injector Temp Tip defiantly has an effect and for good reason
My injector tip temp will drive my trims 7% rich (I haven't got around to dialing them in)

Not trying to sound like an ass, but street tuning takes a lot of time. And if you are with in 5% +_- from day to day hour to hour your probably pretty darn good.
Even after a Dyno tune, nuts on, go log on the street and it will be different (Especially with under 100 cell hits)

You seem to be doing great, but you seem to want perfection in 3 logging secessions, and that is pretty unlikely...
Dial MAF, Then VVE, Then rinse and repeat

I started a log today, started lean. At the end everything was right around -3 trim (So yes, I probably need some start up tuning)

Look at my cell counts

[ApexClone]

When you are street tuning, how are you doing it? Are you driving as you normally would cruising around? Are you holding it at different RPMs (or MAF frequencies) for a long time at once?

At first I was doing ~20 minute drives with light accelerations to get lots of hits per cell. Then I saw an HP Academy video where, on the dyno (which I realize is easier to do this), he was starting at low MAF Hz, getting a couple hundred hits in a cell, then moving up, repeat. I thought that seemed easy so thats what I tried today.

I'm just not sure what the "right" way to do it is on the street. MAF tuning seems like it should be pretty straightforward since you're just looking at a single axis chart. Maybe I'm expecting it to go too quickly. Like I said, I'm new to this.

[dhoagland]

When you are street tuning, how are you doing it? Are you driving as you normally would cruising around? Are you holding it at different RPMs (or MAF frequencies) for a long time at once?

At first I was doing ~20 minute drives with light accelerations to get lots of hits per cell. Then I saw an HP Academy video where, on the dyno (which I realize is easier to do this), he was starting at low MAF Hz, getting a couple hundred hits in a cell, then moving up, repeat. I thought that seemed easy so thats what I tried today.

I'm just not sure what the "right" way to do it is on the street. MAF tuning seems like it should be pretty straightforward since you're just looking at a single axis chart. Maybe I'm expecting it to go too quickly. Like I said, I'm new to this.

I'm not an expert by any means, I struggled (got really frustrated) when I first started.
"The right way".... Many different opinions on that, LOL.. I didn't look to close, but what I saw your are making it happen.

I will do a mixture of holding gears slowly through the rpm range one by one accelerating (as you did today). Then go at a set speed and switch gears through the RPM range (so sometimes I'm at 25 MPH in 6th gear, very slowly increase up the RPM...
Then have normal driving, WITH VERY SLOW throttle movements. If I need to slow down I won't just let off the throttle, I'll apply the brake so I can SLOWLY release the throttle (Obviously this cant always happen)...
I also made an excel sheet where I put multiply logs and average them out... So you could try doing cell hits, then normal driving, then somewhat heavy footed driving and average them all out. (Once you get really close the difference will be minimal)...

Do you have filters in your graphs?
I have different graphs that I look at the same log with: Idle only, TP>20%, INJ TIP TEMP<140, INJ TIP TEMP>AND<160, INJ TIP TEMP>162, TP increase >5% in 200 milliseconds, ETC...
So for my log today. less than 140, about -3 fuel trim.. Between 140 and 160 about -6 fuel trim.. Above 162 about -10 fuel trim

Average is what you want.
Once you get that dialed then you can dial your PE, Tip in, Tip out, ETC...

Somebody might correct me, but getting exact results by running the same stretch of road is tough. Getting within 2% I'm told is attainable, but for me its over a minimum of a 30-45 min drive...

VVE is frustrating and much more time consuming, as you know... But same process...

[kingtal0n]

I would delete the maf if you can. I've tuned hundreds of maf cars and unless you are an maf expert, and know exactly how to setup the maf in a tube free from turbulence and vorticity and clear of bends and so forth it will be inconsistent. Also the smallest leak will create issues as well. So if you plan to keep the maf i would
1. pressure test to find leaks after the maf (fill the entire plumbing and intake manifold with 'boost' pressure)
2. Ensure the maf placement is ideal. Away from bends. Far from the engine as possible. Far from air filter. A nice straight section of tube with proper diameter. Check the maf conditions inside, its clean, its clear. No perturbations. Just try to visualize air flowing through it as smooth as possible.

[dhoagland]

So here are 3 logs I did tonight, SD, very disciplined driving...
1st,,, about 20 min into town
2nd,,, after about 10 min sitting at store, little heat soak, 25 min to restaurant
3rd,, about 35 min at restruranrt pretty good heat soak (Inj Temp 175)

See how it get rich with increased Injector temp. (which mean I need to dial that adder in)...

My point to you is you must make sure everything is exact if you expect except results

Why am I jacking with this???
I just installed a new flex fuel sensor, it is right on, my other was 5% off...
So I'm starting all over... MAF, VVE, Then all associated offsets...
Was the 5% off worth worrying about, probably not, but I want things as close to perfect as possible.

Hope this helps let you know, it take time

[dhoagland]

Here are logs if you want to dig into them for reference

[dhoagland]

Last one

[ApexClone]

Thanks, I'll take a look when I get some free time. I ended up getting good results Sunday and called it. I've moved on to the VVE table again, will probably circle back to MAF again at some point.

[PGA2B]

It's been removed in current versions of the editor. Lawyers were involved.
screenshot.04-06-2022 09.16.46.png

Damn- I hadn't even noticed.

[Dave_In_VA]

Do you have any filters for your data?

Yes Injector Temp Tip defiantly has an effect and for good reason
My injector tip temp will drive my trims 7% rich (I haven't got around to dialing them in)

cell counts.PNG

@dhoagland, your MAF Graphs appear to be filter by injector temp.

How do you do that?

Dave