6L80 bad slip/flare shifting from 3-4

[TURBOSILVERADO11]

I have a 2011 turbo Silverado. Trucks transmission was rebuilt and only has about 5000 miles on the trans since the rebuilt. Truck has been running great until last night. It started slipping and flaring bad going into 4th gear. I reloaded the tune and did the reset preset after and that did not change anything. On the data log the clutches look really close to were they have been engaging and disengaging, only thing different is the huge timing dip where its not making the shift. The transmission fluid looks and smells fine on the dipstick. I'm not sure what to do, I don't want to hurt the transmission. Here's a data log, thanks for any help.4th slip.hpl

[hjtrbo]

Add 50% to your airmass A table. Only for a gentle test obviously. If it doesn't go away I'd say take it into a trans shop.

[GHuggins]

AS hjtrbo pointed out your engine torque is incredibly low while you're accelerating with light throttle. Most likely injectors not setup correctly in the tune and/or torque model not touched. Short transmission life is the result. Would recommend confirming injector data, which is most likely wrong then re-tuning and going from there. The airmass increase of 50% would be a good test, but transmission damage may have already been done. Could also be a sticking valve, so if the airmass doesn't fix it get a short log with just the issue.

[TransGo Robert]

Looks like the torque is not calculated correctly, it probably has to do with the fact that the Ethanol content is not calculated right. It says 79.2 %. I would reset that first to see if it takes care of it. Line pressure is not commanded high enough on the 3-4, that's why it breaks loose like that on the 3-4 when the 3-5-R releases and the PCS 3 has not yet ramped up all the way to max.

[TURBOSILVERADO11]

Thinks for the replies I just now saw them. I was messing with it last night and I added around 20% to the base pressure on the 3-4 shift and the 6-4 downshift. I did the reset preset and after about 20 to 25 miles it was back to normal. I have added to the virtual torque a couple of times already, I guess I need to add more? it defiantly has the right injector data. Here's the tune if ya'll want to take a look.
Hybrid Tune 15 line pressure.hpt

[TURBOSILVERADO11]

I just added 10% to the virtual torque, to ya'll think that's enough?
Hybrid Tune 16 virtual torque.hpt

[hjtrbo]

Monitor your trans slip pid very closely. I can't draw but these are some of the basic shapes for normal vs abnormal shifts.
Screenshot 2023-09-07 230546.png
Base shift pressure changes are generally pointless if your line pressure is already too low. Think garden tap 1 supplying 5 other garden taps. If tap 1 (aka line pressure) is low you can turn up the other 5 taps (base shift pressure) to your hearts content but it ain't gonna do shit. You need to find some way of getting your virtual torque back in line with reality. I don't know how to answer that part. Well, short of diving through the forum for data logs from good tunes of your same truck and comparing torque numbers per acceleration. Hopefully someone can answer you better.

[TURBOSILVERADO11]

Looks like the torque is not calculated correctly, it probably has to do with the fact that the Ethanol content is not calculated right. It says 79.2 %. I would reset that first to see if it takes care of it. Line pressure is not commanded high enough on the 3-4, that's why it breaks loose like that on the 3-4 when the 3-5-R releases and the PCS 3 has not yet ramped up all the way to max.

The e85 around here is always 83 to 85% so I don't know why it only shows 79.2%

[GHuggins]

Thinks for the replies I just now saw them. I was messing with it last night and I added around 20% to the base pressure on the 3-4 shift and the 6-4 downshift. I did the reset preset and after about 20 to 25 miles it was back to normal. I have added to the virtual torque a couple of times already, I guess I need to add more? it defiantly has the right injector data. Here's the tune if ya'll want to take a look.
Hybrid Tune 15 line pressure.hpt

What injectors?

To get the torque model right it's best to start with 100% gas, dial it in then dial in the E models.

[TURBOSILVERADO11]

What injectors?

To get the torque model right it's best to start with 100% gas, dial it in then dial in the E models.

Screenshot_20230908_001402_Samsung Internet.jpg

[TURBOSILVERADO11]

What injectors?

To get the torque model right it's best to start with 100% gas, dial it in then dial in the E models.

It's been on e85 for a little over 5 years now

[GHuggins]

FIC's using actual good lab flow data now. Might want to email them and let them know what you have then get the updated data for them. Think you might be surprised with how much the data has changed. It will say Calibrated Success in the docs if it's the updated data

Doesn't matter if it's been on E that long. Best to dial in the gas side first as the E content starts decreasing due to a bad batch or whatever so the gas side will properly blend with it.

It's not hard to dial in the torque model. You must have the air models both dialed in after fixing the injector data then you can log delivered torque from that and paste straight into the tables. You have to hand alter the map side after pasting the numbers into it because of the VE table throwing it out of whack - which it does BIG TIME - so it matches the airmass side, but other than that.

[TURBOSILVERADO11]

FIC's using actual good lab flow data now. Might want to email them and let them know what you have then get the updated data for them. Think you might be surprised with how much the data has changed. It will say Calibrated Success in the docs if it's the updated data

Doesn't matter if it's been on E that long. Best to dial in the gas side first as the E content starts decreasing due to a bad batch or whatever so the gas side will properly blend with it.

It's not hard to dial in the torque model. You must have the air models both dialed in after fixing the injector data then you can log delivered torque from that and paste straight into the tables. You have to hand alter the map side after pasting the numbers into it because of the VE table throwing it out of whack - which it does BIG TIME - so it matches the airmass side, but other than that.

I'll have to look at the email from fic, but has the injector data from them changed since last August/September? I have the flex fuel sensor turned on but the stoich is set to 9.8, would that mess with it a bunch?
Do you see anything in the tune thats way off thats going to hurt it or is there not much to adjust to get it right if the Injector data is good ?
Haven't dropped the pan but the trans fluid off the dip stick looks brand new and doesn't have any grey from clutches like the stock transmission had. So hopefully it's been caught in time before it's done any damage.

[NotSure]

It's not hard to dial in the torque model. You must have the air models both dialed in after fixing the injector data then you can log delivered torque from that and paste straight into the tables. You have to hand alter the map side after pasting the numbers into it because of the VE table throwing it out of whack - which it does BIG TIME - so it matches the airmass side, but other than that.

I've struggled with changing the VT tables. Where does the Delivered Torque figure come from, would it not be straight from the VT tables? I have experimented with modifying VT whenever the scanner gives me higher numbers than the VT table in a stock cam/PD blower setup (Gen4), and it resulted in jacking up the trans shifting. Rev match went away, and apparently due to having higher torque numbers in the lower spark VT tables, the ECM just completely quit trying to use torque management, even though I had TM set to stock and allowed up to -12 degrees. WOT shifts with NO spark reduction. I put VT back to stock and it began pulling max timing again during shifts. It just seems like if I put 1,000 lb/ft into the VT table for 1200 rpm, 300g airmass, 0 degrees spark, then that is what it would report; kind of a self-feeding loop.

[TransGo Robert]

The e85 around here is always 83 to 85% so I don't know why it only shows 79.2%

Oh you are running E85 then ignore what I wrote, I should stick to transmissions! I just assume you were using regular gas. We all know what happens when we "ass"ume.....

[TURBOSILVERADO11]

Oh you are running E85 then ignore what I wrote, I should stick to transmissions! I just assume you were using regular gas. We all know what happens when we "ass"ume.....

I probably should have listened everything that's done to the truck when I started the thread.

[GHuggins]

I've struggled with changing the VT tables. Where does the Delivered Torque figure come from, would it not be straight from the VT tables? I have experimented with modifying VT whenever the scanner gives me higher numbers than the VT table in a stock cam/PD blower setup (Gen4), and it resulted in jacking up the trans shifting. Rev match went away, and apparently due to having higher torque numbers in the lower spark VT tables, the ECM just completely quit trying to use torque management, even though I had TM set to stock and allowed up to -12 degrees. WOT shifts with NO spark reduction. I put VT back to stock and it began pulling max timing again during shifts. It just seems like if I put 1,000 lb/ft into the VT table for 1200 rpm, 300g airmass, 0 degrees spark, then that is what it would report; kind of a self-feeding loop.

Delivered is what the ecm is "altering" based off of airflow and the already set torque model. It should change from factory especially with anything that has changed the "mechanics" of the motor . Are you logging it correctly and for cam movement? I find with cam movement and mods it's usually the more retarded cam positions that really change. The map side is what controls shifting whereas the airmass side controls rev matching (lower part of the tables) and pressures.

I fought it for a long time, but it finally half way clicked after talking to one of the guys that helped develop the torque model's algo's. He basically told me that everything in the lower end of the tables are dialed in using a big electric motor to spin up the engine with a closed throttle. He never got into the details more than that and kept telling me you needed an engine dyno for any "math" verification even if I could come up with a math parameter to use. He also told me the models will shift 60Nm's for "learning". I've actually found it will report a lot more differently than even that and is what I rely on to dial them in. SO I thought and thought on it - how can I use what I have without making a math so to speak especially since you can't fully take the MAF out of the equation even when failed (probably important to remember that if doing SD tunes)... Then I realized I was over thinking it. Keep the stock values in the bottom row of the tables only on the MAF side. After the MAF and VE are dialed in log "delivered torque". Plug that straight back into the tables - yes it will stay close to the same on the MAF side of the model. The map side will greatly change. I then take the max torque setting from the MAF side and then raise the max map side up to it then interpolate the map side upward. It really is that simple. Hand editing is still required in a lot of the tables depending on how much data is acquired, but you can fix a lot of problems dialing it in like this.

[TURBOSILVERADO11]

Delivered is what the ecm is "altering" based off of airflow and the already set torque model. It should change from factory especially with anything that has changed the "mechanics" of the motor . Are you logging it correctly and for cam movement? I find with cam movement and mods it's usually the more retarded cam positions that really change. The map side is what controls shifting whereas the airmass side controls rev matching (lower part of the tables) and pressures.

I fought it for a long time, but it finally half way clicked after talking to one of the guys that helped develop the torque model's algo's. He basically told me that everything in the lower end of the tables are dialed in using a big electric motor to spin up the engine with a closed throttle. He never got into the details more than that and kept telling me you needed an engine dyno for any "math" verification even if I could come up with a math parameter to use. He also told me the models will shift 60Nm's for "learning". I've actually found it will report a lot more differently than even that and is what I rely on to dial them in. SO I thought and thought on it - how can I use what I have without making a math so to speak especially since you can't fully take the MAF out of the equation even when failed (probably important to remember that if doing SD tunes)... Then I realized I was over thinking it. Keep the stock values in the bottom row of the tables only on the MAF side. After the MAF and VE are dialed in log "delivered torque". Plug that straight back into the tables - yes it will stay close to the same on the MAF side of the model. The map side will greatly change. I then take the max torque setting from the MAF side and then raise the max map side up to it then interpolate the map side upward. It really is that simple. Hand editing is still required in a lot of the tables depending on how much data is acquired, but you can fix a lot of problems dialing it in like this.

Will it be harder to dial in the torque model with the MAF failed? The MAF sensor isn't even plugged in on the truck.

[GHuggins]

At that point you need to back calculate the MAF table - also possibly adding into your problem, which is why I made the point of stating the MAF table is still used in some weird way, even though it shouldn't be... Wonder if there isn't a setting somewhere that truly kills the MAF side of the model. I'm told there's not, but was told that on a lot of other things in the past that didn't hold water. Should also ask - what are you using for temp? I assume a breakout kit? If so then I hope you changed your bias tables to utilize the new location because even this is used for throttle and mbt calcs...

I think this is the correct formula. It converts VE to a MAF correction error (paste special half or whatever like normal) / 100 * ([50040.71] - [2311.71]) / [2311.71] + ([50114.156] + [50116.156])

[TURBOSILVERADO11]

At that point you need to back calculate the MAF table - also possibly adding into your problem, which is why I made the point of stating the MAF table is still used in some weird way, even though it shouldn't be... Wonder if there isn't a setting somewhere that truly kills the MAF side of the model. I'm told there's not, but was told that on a lot of other things in the past that didn't hold water. Should also ask - what are you using for temp? I assume a breakout kit? If so then I hope you changed your bias tables to utilize the new location because even this is used for throttle and mbt calcs...

I think this is the correct formula. It converts VE to a MAF correction error (paste special half or whatever like normal) / 100 * ([50040.71] - [2311.71]) / [2311.71] + ([50114.156] + [50116.156])

I relocated it to the intake behind the throttle body because of the meth injection and used the wire out of the MAF plug. How do you back calculate the MAF? Sorry to keep bothering you with all these questions but I've been trying to get into the 6s for the last 3 years and all this crap is happening now that boost weather is rolling in ��