So it took me longer than it should to diagnose this, but at least I'm on the right track.

It's not the transfer case, which is awesome

The encoder motor was staying engaged. The encoder motor (1 year old) moves just fine when voltage applied, forward and back. (tested on bench) Yes I have the little alignment tool that comes with a new encoder. So my biggest question would be is the encoder motor supposed to disengage on it's own after engagement or does the computer reverse it? I can't remember how much effort it took to manually move the encoder motor, but right now it's got some decent resistance to it.

 

I have verified that I was giving incorrect advice for AWD, so I have deleted the incorrect part of my post.

 

Do you think it's possible for an encoder to become so stiff, either it's defective or just cheaply made, that once the control voltage (brake circuit) is lifted, it doesn't return back to it's start position?

The AWD is super simple. The little stub off the TC is basically spring loaded and just pops right back to base position. Encoder is just as simple. I've read the 4 position is no joke and be a real pain if things get out of alignment with the encoder.

 

I just looked it up and the AWD encoder does not have the brake circuit and it seems very simple as you say. So I am probably full of it on this one. I'm going to have to back out as gracefully as possible. I hate giving bad advice. Is your transfer case sticking? You may be able to get a wrench on the encoder shaft to check (I know you can turn the shaft with a wrench on the 4-button).

 

Description from factory info says that TCCM moves the encoder motor back to RWD operation when difference in speed of front wheels becomes the same. So the motor must work both ways - not just spring loaded. See attachment. Hope that info helps in some way.

 

The transfer case is great. The shaft moves freely with a spring back to base position. But that encoder is super tight, and I obviously didn't know to get a baseline feel for the encoder before installing it. I'll order up another one and see how it goes. Thanks for the input.

By the way, you tried to talk sense into me awhile back and I would have been a WHOLE BUNCH BETTER OFF had I truly understood what you were trying to tell me. So don't be backing out of anything. You're about the only guy helping people here and you don't get enough kudos for all the help you provide.

 

With the AWD transfer case, the TCCM controls the engagement of the viscous clutch in both directions. It modulates the clutch engagement until speed sensor inputs return to similar readings (slippage goes to near zero) where it should then return to RWD mode as the two of you have pretty much determined already. I'm just confirming it for you.

What, specifically, is happening that makes you suspect something is wrong? Is the truck binding while turning? This is often an indication of fluid quality issues. It cannot be stressed enough that the NV136 transfer case requires GM AutoTrak 2 fluid. If you are experiencing issues, you may need to do a few flushes of fluid with some exercising of the viscous clutch between flushes to get any bad fluid worked out of the clutch packs.

 

I changed to the new fluid last year and there is no indication its the TC. When I was diagnosing why it was stuck in AWD, I pulled the encoder, which was a little tough because it was in "engage" mode and TC went right back to it's standard position while the encoder itself was turned to the engaged position.

So i put the encoder back into not engaged position (seemed very tight) with the alignment tool they send you with a new unit, and re-installed it. Everything was fine, until I took a fast left turn and noticed when I pulled into my driveway the AWD had engaged, but not disengaged. Now I knew it was the encoder stuck, so I just applied 12v to the wires that go to the encoder, first try was the right orientation, and it unlocked immediately.

I've got a new encoder on the way so we'll see if it makes a difference. I'll update this thread in case it happens to anyone else.

 

What I'm not sure of in the NV136 is whether there is any feedback as to whether it actually gets to 4wd/2wd mode and back again or not. From the wiring diagrams I have available to me, it would appear as though there is no way for the TCCM to know what mode the tcase is actually in. If there is some kind of additional resistance that stalls out the encoder motor and it doesn't make the transition back from 4hi, the TCCM might not know. All it cares about is that it engaged 4wd mode, the slip was handled, and it did what it needed to do to return the tcase to 2wd mode. If that is the case, hopefully a new encoder solves your problem.

The only other thing that I can think of that could cause it to engage unnecessarily is the TCCM itself. The encoder motor is nothing but a dumb electric motor that spins a planetary gearbox on the other end of a few other gears which is why it is quite difficult to turn the output shaft yourself. I have pictures of the internals of an NP231 encoder somewhere around the forum and have pulled apart all of them over the years. The motor/planetary setups are all very similar.

Have you tested the wiring between the encoder & TCCM? I'm wondering if there may be a problem with the current loop in one direction that doesn't affect the other. Given that the tcase input shaft is spring loaded, there would be more load to put it into 4wd mode than to take it out which makes it odd that the encoder would get stuck in 4wd mode which is why I'm kind of rambling on about this. It has me really scratching my head.

A full feature bi-directional scantool can access the TCCM and engage/disengage the transfer case which would be very helpful in diagnosing this issue.

 

my 99 Bravada was diagnosed the other day with a bad tccm because it showed an open circuit internal short at the rear shaft speed sensor that was not sending a signal to the PCM. the mechanic knew it was an internal short because he actually tested the entire harness from the speed sensor up to the tccm, essentially looking for a broken wire. the entire harness was live all the way. It would make sense that if the tccm was not able to compare shaft speeds that it would not return the encoder to 2 wheel drive. the advice I was given was replace the tccm first and not just junk the encoder. that is the battle I have come to so far.