Help to diagnose 8L90 that won't move in any selected range

I have an L86 swapped into a Jeep with an 8L90 (RPM Extreme harness). Both came from a wrecked donor '18 Sierra 1500 Denali so the engine and trans have always been together.

After swap trans would not engage when shifting to RNDL(1-8). Fluid levels were good. After some checks it was determined the internal pump wasn't making pressure so the pump was changed, along with new torque converter, and internal harness. New complete seal kit was used and system was changed over to Mobil 1 LVP HD. When the transmission connector is unplugged (round plug) from the transmission, it pulls hard in manual valve body override (reverse and manual 1). When the transmission is plugged into the harness it won't engage.

I swapped in a new TCM (T87a grrr) and downloaded the GM TIS2/SPS calibrations into the new TCM. I am able to successfully go through the fast learn process in GM Tools TIS2/SPS (Not HPT). The transmission will bark the tires during the relearn with my foot on the brake so I know the transmission is strong. However, when I cycle the ignition off to allow the TCM to drop relearn commands and adopt the new fast learn calibrations the transmission still does nothing in reverse and drive. In GM GDS2 I can see the commands from the range selector commanding the voltage changes in the transmission, and there are no MILs on TCM.

My setup right now is bone ass stock E92 with VATS disabled in the ECM (ECM has all the latest software from GM as of 12/29/20). I have swapped in a new locked T87A so it is very stock with latest software and GM calibrations for VIN and trans TUN. I do still have my HPT unlocked T87A TCM.

Has anybody ever encountered a stubborn 8L90e that will not move when commanded through ranges? Any help is appreciated. I have access to TIS2Web, GDS2, SPS, and HP Tuners. Thanks!

I saw a procedure in GDS (6L80E) that said after the scanner gives you the completed status. Return your car to park, shut off the engine. Then in the scanner close the test page, then close the program. Disconnect the scan tool from the data link connector. Do not turn the ignition on for a minimum of 2 minutes (I usually wait for 5)...

If it still doesn't work, Reflash with your TIS2Web file... Do not perform the relearn just to see if it moves...

Attachment 106150

Hope this helps. Happy New Year! Even happier when you can smoke the tires at will:)

Quote:

---

Originally Posted by IARLLC

Attachment 106150

Hope this helps. Happy New Year! Even happier when you can smoke the tires at will:)

---

Thanks for the reply! I have already read and attempted this procedure twice.

Quote:

---

Originally Posted by markric

I am able to successfully go through the fast learn process in GM Tools TIS2/SPS (Not HPT).

---

Also this... https://gearsmagazine.com/magazine/8...ew-technology/

I will try again using GDS. I guess if that fails I can try HPT.

Quote:

---

Originally Posted by dhoagland

If it still doesn't work, Reflash with your TIS2Web file... Do not perform the relearn just to see if it moves...

---

Thanks dhoagland! I will attempt a reflash again to see if the cali sticks after the relearn.

Keep us posted
Good Luck, HAPPY NEW YEAR!!!

Are you exiting the fast learn process correctly? This issue can happen of that is not followed correctly. I've done it. I usually complete mine with gds2 but sometimes use sps if I happen to have that open and the temp is above 165. Both of them have you place it in park then exit the control function (relearn). Shut vehicle off open and close driver's door. Wait 3 minutes so the tcm can go to sleep and reset. Start and recheck for issues.

Quote:

---

Originally Posted by HardH20

Are you exiting the fast learn process correctly?

---

Thanks for the reply HardH20! You present a tricky question here that I am not sure I can answer and have second guessed myself on this exact thing. This engine is in a Jeep, so there is not a "true" connection to a BCM that monitors door entry/exit. Aside from the open/close door procedure, I believe I have followed the on screen and documented procedures. I am really hoping that is not the issue, because I will have to figure out how to trick my ECM to think a BCM saw a door open and close. Once it warms up a little bit I will try the suggested relearn and calibrations suggested above. The Jeep doesn't move so it is not much fun to try and heat the trans fluid to 165 when it's 40 degrees outside.

A little history before this gets too far with BCM talk. These engine/trans combos (LT/8L) have been swapped into numerous Jeeps without issue. Mine is the statistic POS that will not work properly. After the swap, the trans range selector never would try to move the vehicle even though all the range commands and clutch solenoid voltage changes can be seen in GDS and HPT. This engine and trans have always been together from a donor truck, so ECM and TCM were VIN married with good calis. After having the trans pump replaced is when I started chasing relearn and possible TCM headaches. I am hoping that someone has been able to magically take an 8L from non-responsive (through normal ECM-TCM comms) to working. I have been working on this trans issue since August '20 with very little difference in everything that has been tried. Rebuild was definitely needed due to cracked stator plate and bad trans pump (very low pressure), and now the trans is wanting to pull. There is not a tune on the ECM or TCM other than VATS delete, so it's basically stock. ECM has current E92 updates via SPS. TCM has current T87A updates via SPS.

I kinda think that the instructions regarding door open and close is the issue. If the TCM cannot finish the process it just throws the info out the window. Kinda like key off to power down the PCM after writing a tune to it. I hope there is another way. I wonder...if the need is for the TCM to go to sleep what about disconnecting the battery after the process finishes?

Worse comes to worse, you could try a used T87A or even T87 out of a similar rig with no programming. I know for a fact that T87 can be swapped between rigs without an issue. Of course there is a slight difference from not programming the solenoid info but they run and self-learn the shift properties just about as good as ones that have the calibrations in them. Whenever possible I run all the processes with SPS but in a swap situation like yours, I might not feel the need....plus if you stick a T87 out of a 2015.5 or 2016 8 speed you can tune it:)

When the ignition is turned off R.A.P. (retained accessory power) stays active and will keep powering several modules and devices in the vehicle. Opening the driver's door opens (turns off the rap relay) and the modules begin to go to sleep.

I'd have to see the diagram but I think that shutting the ignition off after fast learn and then removing power from the ignition fuse to the tcm would accomplish the same result. A 2 minute time out ( module goes to sleep and restarts at next key on) may fix your issue. I can look at digrams tomorrow and get a better idea of how the circuits are built for those and how rap functions with it

Thanks HardH20!
So I can confirm that I never have accessory power to TCM pin 51. Most race or aftermarket 8L harnesses don't include power to that pin, and if they do they splice directly from pin 35 (run crank power). I wonder if that is the RAP accessory voltage pin for serial data? So far I have not seen a need for the pin 51 to have 12+ voltage. I can run a backpin jumper over form 35 if that looks like what the RAP is using and it will cut ignition accessory power when the key is turned to the off position (run crank power at 0v). I do have TCM pin 66 (12+ batt) run through a blade fuse, however it seems like cutting batt+ all together would just turn off the TCM and not allow it to "sleep".

If somebody with GDS2, or a multimeter on TCM pin 51, and a GM truck, can monitor that accessory voltage data on the K71 module during a key-off/door-open/door-close event I would really appreciate it. It would be interesting to see what that pin voltage is doing. Again, I do not have a GM vehicle with a BCM, so I am unable to test, and I don't know what the desired behavior should be. Thanks for all the assist!

I had some time to heat the trans up to temp (165-7ish) and try a few things...

GDS2 would never let me perform a fast relearn. Once the checklist screen would appear and validate the trans temp, it would immediately change to a screen that said "An error has occurred, please exit". I attempted this multiple times with the same result.
This occurred with both the HPT unlocked T87A, and the GM locked T87A. I could not find a way to get past the error in GDS2.

SPS allowed me perform a fast relearn when GDS2 would not.
Using HPT unlocked TCM, I performed an SPS TUN calibration and powered off the vehicle for the recommended 30 seconds. After that I performed a fast relearn using SPS. Once the relearn was successfully completed, I put the vehicle in park, exited the relearn program (Finish button), and turned off the ignition and waited for 5 minutes. Once the Jeep was restarted it did not have any response to Reverse or Drive(D/M).
After the trans cooled off a bit...
Using GM locked TCM, I performed an SPS TUN calibration and powered off the vehicle for the recommended 30 seconds. After that I performed a fast relearn using SPS. Once the relearn was successfully completed, I put the vehicle in park, exited the relearn program (Finish button), and turned off the ignition and waited for 5 minutes. Once the Jeep was restarted it did not have any response to Reverse or Drive(D/M).

I had to quit working on the Jeep to go get gas lol. It took an hour to bring the transmission temp up to 165 and I didn't have enough gas to keep testing. My poor neighbors... I had a 3 inch Kooks exhaust pointed right at their houses for over 2 hours straight... oh well.

Quote:

---

Originally Posted by dhoagland

Keep us posted
Good Luck, HAPPY NEW YEAR!!!

---

dhoagland, I was able to try the SPS recal, SPS fast relearn, SPS recal, process and the transmission was still unresponsive. I am beginning to wonder if I have ECM to TCM communication issues. I know that SPS talks directly to the TCM during a fast relearn and engages the solenoids and clutch pressures. However, without SPS I get nothing. Something is starting to wreak of ECM communication failure here. No MILs at all. Man this sucks. Thanks again for the suggestion. I will apply my basic HPT tune, upload the tune and logs here tomorrow and see what can be tweaked to get this stupid thing to work.

HARDH20 and I both point at the need to power the TCM down after the relearn for the relearn to finish successfully. My guess is that most swap wiring harnesses don't take into account the fastlearn process just as most swap motor harnesses don't take into account crank relearn processes (even though it has a major affect on GEN5 rigs). They probably just expect that people won't do the process...which is ok as long as nothing has changed. The TCM will learn how to run over time (though they seem to learn bad downshifting behavior if you don't preform throttle closed coastdowns kinda often. I have a 2017 Denali with the 8 speed in my shop right now but no time. Hopefully, tonight I will have a chance to watch the power down process as you suggest. We'll see.

Looking at the pin outs....pin 51 is for circuit 5985 accessory wake-up serial data. Pin 66 is main module power and pin 35 is ignition voltage from kr73 main ignition relay. I'm willing to bet that pin 35 voltage is cut when r.a.p. is deactivated by opening the door. The module will still have power and can reset/ put itself to sleep - wait 5 minutes.

During 2020 I put a GEN4 LY6/6L90E/NQH combo into an old 2 door S10 Blazer. I ran the BCM, steering column TCCM and cluster in order to have Auto4x4 and factory upshift/downshift on the shifter. I have not hooked up any of the door switches/latches so I was curious if the modules would go to sleep. I have never timed it but they do go to sleep after a few minutes. No noticeable drain. She is a fun little beast!

I have not looked at the schematics yet but from what HardH2o is saying I'd think that pin 35 could be wired to lose voltage when you turn off the key.

I have not had a chance to check the 2017 Denali yet. As much as I hate to admit it, I was stuck on a big Mercedes project until 5AM. Maybe tomorrow I can scan this Denali as it goes to sleep.

Thanks for the time IARLLC! I have an 09 Benz C350... working on that thing is a biatch. I got slammed with home life and haven't been in the garage for a few days, except to fix other peeps problems. Hoping to get out there tonight. I will post HPT most recent tune, logs of shift range changes while engine is running (including all A8 PIDS). Also hoping to get some scope/meter checks on serial data between ECM and TCM.

I really appreciate any and all input on this issue! My list is long for diagnostics, but I appreciate your diagnostic input, so I can double check my list.

Sorry for the long delay. I finally got a chance to watch the Denali TCM as I turned it off and opened the door.

2 screenshots:
1. With Ignition/key on
2. With Ignition/key off (door still closed)
As you look toward the bottom of the middle row:
Power Mode
Ignition Accessory Signal
Ignition 1 Signal
Ignition Voltage
Control Module Voltage

What I saw is that Power Mode, Ignition 1 Signal, Ignition Voltage, and Control Module Voltage all turn off as soon as you shut it down but RAP keeps Ignition Accessory Signal awake. I did not take another screenshot because Ignition Accessory Signal stays on until about 15 seconds after opening the door or of course you can wait 10 minutes for RAP to shut down. Soon after RAP turns off the Ignition Accessory Signal the TCM shuts down and loses contact (no screenshot). I did not plug into the pins to see but I'd be pretty sure that pins 66 and 35 both shut down when you turn the ignition off. But the TCM is expecting to get told to go to sleep by a CAN message on pin 51. I am not sure what that CAN signal looks like. I am not sure what your harness does with pin 51. Obviously, the TCM wakes up so probably pin 51 goes to something that it can be unplugged from to make it sleep sooner than later.

Hope this helps.

Quote:

---

Originally Posted by IARLLC

What I saw is that Power Mode, Ignition 1 Signal, Ignition Voltage, and Control Module Voltage all turn off as soon as you shut it down but RAP keeps Ignition Accessory Signal awake.

---

Thank you IARLLC! I have stepped away from the project for a while. I will try to mimic this behavior on pin 51 in a couple of weeks. Right now my pin 51 is not used, nor powered. I will need to splice my 35 to 51 to get the accessory power to at least mirror the run position. If I get ambitious I can locate a slow ign off signal on the Jeep TIPM and use that. The Jeep will be traveling to Alabama next month to get some professional eyes on it. Hoping that will be the final shop stop to get this thing up and running.

We'll look forward to hearing how it goes.