XFI, O2 Correction gets stuck.

[NY Twin Turbo]

Has this ever happened to you guys?

I'll power into a cell on the VE map, maintain a steady load point and hold that cell location until it stabilizes.

Then I'll make the correction based on the O2 correction. And then it will be good.

Then when I enter the cell under the same circumstances as before it will show O2 correction again. I know this part is normal, because it may be a little touch lean or rich in the AE. But this should only be a brief moment.

This should again clear out and the cell should go back to almost no correction again. But it does not. It keeps on correcting even if the correction it is making provides the wrong AF.

Then I lift and return to the cell another time........and now it's good again.

What makes it get stuck on correction? I never gave this much thought because it's somewhat minor. This happens mostly when I'm trying to tune light load areas in the cruising area of the VE table. This area is important to me because I spend a lot of time enjoying nice long drives in the summer.

Until recently, I never gave this much thought because it's somewhat minor.

 

[turbo nasty]

You can populate low load areas on the map in 4th, 3rd gear as well as 1st gear if in light throttle. Ones going to need more/less fuel than another due to gear, wind resistance, etc so that will cause it to correct. Another is if you correct it and dont save it then itll be whatever your tune was. I wish the XFI had a 1st/2nd gear maps and then 3rd/ 4th gear maps just for this reason. I would like to see those low load areas not change when cruising in 4th and then change again when in same area in lower gears.

 

[HighPSI]

Check your wbo2 gain tables

 

[NY Twin Turbo]

You can populate low load areas on the map in 4th, 3rd gear as well as 1st gear if in light throttle. Ones going to need more/less fuel than another due to gear, wind resistance, etc so that will cause it to correct. Another is if you correct it and dont save it then itll be whatever your tune was. I wish the XFI had a 1st/2nd gear maps and then 3rd/ 4th gear maps just for this reason. I would like to see those low load areas not change when cruising in 4th and then change again when in same area in lower gears.

But isn't the VE cell always the same no matter how you got there? Meaning, when in that particular cell on the map, aren't you always at that same load point? even if in a different gear?

On another note. I have taken advantage of the gears in order to enter areas of the map not easily reached otherwise. Having a manual 400 has given me the opportunity to tune in 3rd gear for high load areas at lower RPMs, and using 1st gear for light load areas at high RPMs. All within reason, of course. I'm not going to slam the throttle down from a stop in 3rd and I'm not going to rev it to 7,000 in 1st.

I always save before keying the car off. But changes made on the fly should go directly to the ECU, No?

When the O2 gets "stuck", it's not correcting an incorrect AF.........It's creating one. If I shut off the correction, the VE number is good. If I minimize the swing it's better. But wan't to allow at least 10 percent correction if needed. The thing is, I don't want it to get "stuck" 10 percent rich or lean.

 

[turbo nasty]

Rpm to kpa is your references so you can be in 3rd gear at the same point on the map as 1st gear but the engine is working harder in 3rd.

As far as saving corrections you can save all tables as one way or the self correct on its own. But....it does NOT write/save to memory until key off. And the save is done once the lights on the fast box go completely out. Many make the mistake of turning key off and then right back on. That wont save correctly as it takes a second or so. Also not waiting long enough can cause corruption in the files since its a mix of the old and the change that was attempting to be saved.

Your ve table corrections is based off your target af table not the ve table. VE table changes are showing the corrections needed to achieve your programmed target afr in the af table.

This is where the load comes in with the engine working harder in same part of ve table in say 3rd gear that it self corrected to achieve the target afr of same cell in 1st gear. Now the engine is in 3rd and the amount of fuel needed at that spot wont work for the target afr in that cell so youll see adding fuel.

Gotta have your plus and minus gains setup in afr tables as well

 

[NY Twin Turbo]

Check your wbo2 gain tables

I don't have the tune on this computer. But I think I have the gains set in the 10's around the cruise areas of the map (light loads around 2200 - 3200 RPM). Too slow you think? I also have the Dynamic delay set for 0.

 

[NY Twin Turbo]

Rpm to kpa is your references so you can be in 3rd gear at the same point on the map as 1st gear but the engine is working harder in 3rd.

As far as saving corrections you can save all tables as one way or the self correct on its own. But....it does NOT write/save to memory until key off. And the save is done once the lights on the fast box go completely out. Many make the mistake of turning key off and then right back on. That wont save correctly as it takes a second or so. Also not waiting long enough can cause corruption in the files since its a mix of the old and the change that was attempting to be saved.

Your ve table corrections is based off your target af table not the ve table. VE table changes are showing the corrections needed to achieve your programmed target afr in the af table.

This is where the load comes in with the engine working harder in same part of ve table in say 3rd gear that it self corrected to achieve the target afr of same cell in 1st gear. Now the engine is in 3rd and the amount of fuel needed at that spot wont work for the target afr in that cell so youll see adding fuel.

I think you miss understood me. I don't allow the XFI to self correct my VE map (learning mode is off). I make the changes to the VE map myself based on what O2 corrections are taking place to acheive what I have programmed into the AF map. So I know the proper VE number is mostly created by eliminating percentage of correction against my AF map. And this is all because we want the computer to have the right number to go on as it enters the cell.

So again, when I change a VE number, it goes directly into the tune as soon as I do it.

 

[turbo nasty]

If you make any change and dont save all tables then its not saved and further the changes does not hard save/rewrite until key off.

 

[NY Twin Turbo]

If you make any change and dont save all tables then its not saved and further the changes does not hard save/rewrite until key off.

Which still doesn't explain why the 02 correction gets "stuck".

Follow me for a second here. Again, if I enter a cell, with a known good VE number, and it is just a little rich or lean on AE, it will get stuck on correction, even if given the chance to stabilize. And when this 'getting stuck' is taking place, it might continue to throw positive correction at that cell even though the actual AF is still rich. And sometimes it gets stuck in the opposite direction, pulls out even though it's already lean.

So back to the original question. Why does it get stuck? Should I speed up the gains and minimize the swing?

 

[HighPSI]

Like I posted earlier, you may need to increase your WBO2 gain table. Thus is not to be confused with WBO2 correction limits

 

[NY Twin Turbo]

Like I posted earlier, you may need to increase your WBO2 gain table. Thus is not to be confused with WBO2 correction limits

I'll give it a shot. I figured around 10 on the gain table would be fast enough for that area of the map. But I guess it should recover from correction more quickly than it's doing. I'll go to the extreme and work my way back. I'll see what happens.

 

[HighPSI]

What firmware are you running? Also, try 15 on the gain. 10 is a little soft

 

[NY Twin Turbo]

What firmware are you running? Also, try 15 on the gain. 10 is a little soft

XFI 2.05.
Does this make a difference?

 

[HighPSI]

Yes. There was a minor bug with WBO2 gain on early, early stuff. You're fine

 

[turbo nasty]

XFI 2.05.
Does this make a difference?

What firmware? Should be a B number

 

[NY Twin Turbo]

What firmware? Should be a B number

What exactly is "firmware"? I can't play pac-man, or use a smart phone. The only thing I know about computers is how to navigate through C-Com and how to type to this list.

If you knew me like my other friends do, you would find it shocking that over the years I have become as proficient with XFI as I have! It's actually damn near a miracle.

 

[turbo nasty]

Lol iirc its in the drop down named computer info or something along those lines

 

[HighPSI]

The fact that you're running 2.05 tells me enough. The "bug" was fixed years ago. If you go to "View/Display Ecu Configuration" You can see what firmware you're running.

 

[NY Twin Turbo]

The fact that you're running 2.05 tells me enough. The "bug" was fixed years ago. If you go to "View/Display Ecu Configuration" You can see what firmware you're running.

Thanks Cal. I'll look this weekend. But I'll assume I don't even have to..
In a week or 2, I'll begin to drive it. I'll speed up the gain and let everyone know how it turned out.

 

[turbo nasty]

Only reason I posted to check firmware is at one point I had 2.05 but wrong firmware.