Leaning out under boost below 3k rpm

[Dirk37]

Hoping one of you turbo yamaha guys could help me out with a tuning issue.

Below about 3k rpm, if I get into boost it begins leaning out. At 0 psi it's dead on at 11.4:1 afr. But at 2-3 it goes up to 12:1, at 3-5 psi 13:1, and at 5+ I've seen 14:1 and above (I usually let off well before then). Above around 3k rpm the afrs are right where they should be.

It mostly occurs when I'm on the freeway (60mph at 2.8k) in 5th gear and step on it. The turbo spools and it starts leaning out until I let off. 4th gear is no problem. I also don't have the issue when its extremely cold out.

Any ideas of what may be going on?

 

[Toolman]

What does your stabilized fuel table look like?

 

[Dirk37]

Here it is

 

[Toolman]

You are commanding 14.6 at 80% load. Depending on how much TPS you have in 5th to see some boost, it's likely that the car is simply giving you what you command. Cruising in 5th you are seeing 11.4? Or do you mean once you are on the gas it drops to 11.4 but then starts climbing as the boost increases?

Also, if it only happens in 5th, just don't use 5th as a passing gear

You should try some low rpm pulls in 3rd and 4th as well, not WOT, just enough to see some boost build, and monitor your AFR.

 

[Dirk37]

Once I'm on the gas and it builds boost it will drop to 11.4 then increase from there. I'll upload a picture of the datalog so you can see. It's a really weird problem.

It happens in 3rd and 4th too but very rarely, usually because I'm above 3k when I hit boost in those gears.

I've been just been avoiding spooling in 5th but I'd like to figure it.

 

[Dirk37]

Here's a the datalog. Blue is commanded afr and red is actual afr. As you can see as boost builds it leans out. This is a 4th gear pull.

Here's a datalog of a pull without the issue. After comparing the two, in the first one you can see the pulsewidth doesn't increase with the maf, while in the second one it does. Not sure why that is. Perhaps my breakpoint or injector slopes are wrong.

 

[Dirk37]

I think I may have discovered the issue! So I looked at the datalogs again and traced the issue to the aircharge calculation. It does not increase with the maf like it should. I was reading through the ford GUFB documentation and there's a function called FN037 that caps the aircharge at certain rpms. Apparently this was causing issues with the supercoupe guys. http://eectuning.org/forums/viewtopic.php?p=71240 This function doesn't exist in the x2j strategy yet but I'm going to try and find it and fix it. I'll let you all know what I find. Here's the datalogs where you can cleary see the aircharge is dependent on the rpm and not the maf.

 

[Dirk37]

Alright I figured it out! There was a cap on the aircharge but it was much more complex than just an rpm vs cap. It's based on tps, act, and rpm which explains why it worked in the cold sometimes. Here's the thread on eectuning detailing the issue. http://eectuning.org/forums/viewtopic.php?f=8&t=21816&p=124265#p124265

Basically to remedy it you need to open up the raw bin file with a hex editor and replace A3763650 at offset 36AF with FFFFFFFF. That completely disabled the autism. I just tested it tonight and the car runs amazingly. Plus it spools way faster and doesn't lean out at all

 

[Toolman]

Who's strategy is this? Wondering if I need to look into doing this.

 

[Dirk37]

I'm using the coretuning x2j strategy, but it doesn't support this yet. I emailed them tonight with the info though so hopefully they'll add it in. I edited mine with a hex editor. What are you using for tuning? I can help you figure something out. I would also definitely recommend this. Makes a huge difference.

 

[Toolman]

Not familiar with core tuning.

Now that I do not have a street driven turbo SHO, I don't know that I will run into this issue, and it's probably not going tone an issue on my S/C car.

And I command open loop in my track car, and the turbo is much larger, so no real boost before 3k. Still, awesome work. Makes me think I want to put that little T4 on my 90 instead of the Vortech.

 

[Dirk37]

The limit still kicks in even when out of boost. Anytime there's more airflow at a certain throttle position than normally the limit will kick in and lean it out. I've found the car to be significantly more responsive now without it, and the turbo spools crazy fast compared to before.

If you do decide you want it let me know and I can figure out a way to make it work with whatever software you're using.

 

[Toolman]

Sounds good!

 

[decipha]

This is clearly a base fuel issue

Id highly recommend scaling perload from fn035 (prldsw==0) and disabling the stabilized fuel table fn1360

rescale the base fuel table so 90%-105% perlosad (100% perload == 0 vac / 0 boost ) commands a reasonable lambse ~0.875, then setup another perload value above that say 120% and set it to enrichment for boost ==0.805 lambda

make sure fn320a the open loop ect threshold is forcing open loop above 120% perload (~approx. 3psi of boost)

from what you describe it sounds like its still in closed loop in this transisitional period and your not correctly addressing the lambse for your fuel error

this is all described in great detail in my fuel write up

http://info.efidynotuning.com/fuel101.htm

im basing this on the assumption that the backflow table wasnt sufficient in fully resolving the issue??

 

[Dirk37]

Disabling the backflow limit fixed the issue. Fuel is spot on now.

 

[92sho16]

Not familiar with core tuning.

Now that I do not have a street driven turbo SHO, I don't know that I will run into this issue, and it's probably not going tone an issue on my S/C car.

And I command open loop in my track car, and the turbo is much larger, so no real boost before 3k. Still, awesome work. Makes me think I want to put that little T4 on my 90 instead of the Vortech.

Core tuning is Adam aka Popsracing and a few others.

 

[Dirk37]

I made a workaround for this air limit with Tunerpro and uploaded it here. Remove the .txt extension and then you can open it with Tunerpro RT. Once you've got it open in Tunerpro, download the bin from Emulator into Tunerpro, apply the patch, then upload it back to the emulator and you'll be good to go! Unfortunately every time you make a change with BE this will be overwritten so you'll have to repeat the process each time.