dreaded torque bind still not cured

[Bserk]

Hey guys, this is gonna be a little long, so thanks to anyone who bares with me.. My 96 Legacy Outback 2.5 AT. Posted a few times about numerous issues, Ive gotton great input and Ive solved/repaired everything EXCEPT this torque bind issue. When started, the AT temp light flashes 16 times and goes out. So, I was convinced from the start that it was an electrical problem. However, I pulled the engine and did the head gaskets, etc. which it also needed, planning to tackle the torque bind issue after I had it up and running. I drained the ATF and the torque converter while everything was apart. And after it was all back together, I drained and refilled the trans 4 times. The fluid is red and clear, and didnt really look too bad even at the start. I incorrectly thought that the AT temp light blinking 16 times meant a code 16, which according to my crappy aftermarket manual means the circuit for the solenoid 'C' valve in the tailshaft was bad. That controls the torque split, so it made sense. I also learned here and from a real great Subie dealer parts guy that this is a common failure, so I bought and installed the 'C' solenoid valve. Didnt fix the problem. While the tailshaft was off, I got to check out things, and the transfer clutch pak is in great shape, not burned or damaged at all, so it wasnt a completely wasted surgery...and hey, my C valve is new, so I wont have to worry about that later, huh?

 

Ive mentioned that I work at a Nissan shop, we have scanners that will access the OBD II on my Subie, but not trans codes. My junk manual explains how to enter a 'self diagnosis' mode using the trans. light, but I have had no luck doing it. I really would kill for a genuine Suby manual for this car..anyways, I pulled in some favors and got the car scanned at a local transmission shop and they come up with something along the lines of 'corrupt torque signal' and indicate a harness or AT control unit problem. I went back with this info and pinned checked the AT control unit harness and, if my crap manual is correct, everything points to the control unit itself being bad.

 

If I were 100% sure of this, I would bite the bullet and order a control unit. But at this point, Ive allready spent a ton more on this car than I planned to. It is to be a winter rat, and winter is damn near over! LOL! Ive been driving the car on snowy days on borrowed dealer plates. I freaking love this car, but I dont want to transfer plates and drive it full time until I can figure this out. I figure on slippery roads, I wont hurt too much. But I think it could be a bit dangerous in a fast turn, emergency situation...not to mention the strain on the drive train.

 

I have another favor owed, that Im going to try and pull in tomorrow...one of the local independent shops claims to have a wrecked 96 OB that I can 'borrow' the unit from. Its burried in the snow and he isnt even sure its a AT, so who knows.. (he is sure its a 2.5 dohc, so wouldn't that make it AT?)

So maybe tomorrow if Im lucky I will have an answer. But it anyone has any opinions on this, I'd love to hear them! Thanks if you read through this all! Doug

[nipper]

What makes you think you actually have torque bind. Forget the light. What is the car doing.

 

 

nipper

[WoodsWagon]

And if you do have tourque bind, does it go away when you put the FWD fuse in?

[johnceggleston]

i'm leaning towards a bad TCU but that's just because i want mine (95 leg w/ 96 AT) to be that simple. i have an intermitten AT TEMP light. mine seems to drive fine when cold, but fails (starts binding) when it warms up. and of course the confusing part is the light comes on at start up and blinks 16 times the NEXT time you start the car. you can check this by turning the car on and off 2 - 3 times. if the the fault is gone the light will not come on the 2nd time you start.

 

putting in the FWD fuse will tell you if the duty c is working correctly, but regardless of that, if you get binding something is wrong. DUH! when i put my fuse in, it runs FWD untill it warms up then binding. so either my duty c fails after it gets hot or some thing else does, wires, connection or TCU.

 

try driving in FWD and see if the binding comes back. (assuming the fuse puts it in FWD.) the TCU swap would be a great test.

 

this is the first i've heard about TCU problems.

[Bserk]

Ok, first Nipper: Yes, it does have torque bind. On bare pavement in a sharp turn, it feels just like a part-time 4X4 in 4WD.. My Nissan truck, for example.

 

As far as putting in the FWD fuse, I havent tried it since before I replaced the 'c' solenoid. I will do so tomorrow. (it did not go into FWD with the fuse installed before that, so I doubt it will be any different)

 

At lunch today, I shoveled in to the burried Outback at the other shop...only to find a standard transmission gear shift so that car is out! LOL....I have all the luck! Called around looking for another TCU and so far am coming up empty. But another call to Goldstein (the Subaru dealership that has been real helpful) informs me that they have sold TCU's for this car, so apparantly it isnt unheard of for them to fail. Hmmmmmm. Im going to keep looking for a used one, then decide on the next step. Thanks for the inputs again guys.

[nipper]

Ok, first Nipper: Yes, it does have torque bind. On bare pavement in a sharp turn, it feels just like a part-time 4X4 in 4WD.. My Nissan truck, for example.

 

As far as putting in the FWD fuse, I havent tried it since before I replaced the 'c' solenoid. I will do so tomorrow. (it did not go into FWD with the fuse installed before that, so I doubt it will be any different)

 

.

 

Wrong. :-p

 

you fixed the electrical failure that you had before. If the solenoid is dead, the FWD fuse will have no effect.

 

Also when you said you inspected the AWD unit, how did you inspect it?

 

You have another issue. you are pre 1997 1/2. After 1997 1/2 and all parts sold after that have been redesigned. Beofre that the seals and body would wear out and cause TB. It is very very possible that you need a new housing, even if your clutch pack looked fine.

 

nipper

[Bserk]

Wrong. :-p

 

you fixed the electrical failure that you had before. If the solenoid is dead, the FWD fuse will have no effect.

 

Also when you said you inspected the AWD unit, how did you inspect it?

 

You have another issue. you are pre 1997 1/2. After 1997 1/2 and all parts sold after that have been redesigned. Beofre that the seals and body would wear out and cause TB. It is very very possible that you need a new housing, even if your clutch pack looked fine.

 

nipper

 

Ive considered that, and Im no Subie expert (but getting there-LOL), but that would not explain the AT temp light flashing, after the codes were erased, and the car not driven? Im thinking I still have an electrical problem.. or can the light flashes indicate a mechanical failure, especially imediately? Before the car is even driven?

[nipper]

The light flashing is telling you that the last time the car was used started something was not right.

There is more then just one solenoid in the car (meaning you may have more then one failure). You need to read the code on that flashing light.

The light can read a limited amount of electromechanical faults. It can see if the solenoid is stuck open or closed due to a mechanical obstruction.

 

nipper

[Bserk]

Thankyou for the input, Nipper. I hope I dont sound thick here, but I do have a ton of experiance with electronically controlled transmissions, unnfortunately I dont own a scanner that reads Subaru trans codes or a decent Subie manual. Im aware that there are several solenoids, in fact other than the torque control circuit, the shift control circuits and solenoids in this thing are very similar to Nissan stuff. As I stated in the first post of this thread, a friend at a transmission shop scanned my car and came up with a code..I do not have the info in front of me, its at work, but it was a 'corrupted torque SIGNAL' code...and his snap-on scanner gives a summary of possible reasons for this code. It mentioned bad grounds and wiring problems between the TCU and the ECU, or the TCU itself. Now, believe me, Im not questioning your input. Thats why Im here! But from Nissan experiance only, this type of SIGNAL code would lead me to a electrical issue, esp. if the code was erased and imediately came back. I know in a Nissan (I hate to keep falling back on Nissan stuff, but its where Im most comfortable) if a solenoid was stuck, or there was a mechanical issue in the trans, it would set a code after the car was driven and the 'problem' occured. Where as a electrical problem would set a code right off the bat (unless it were a intermitant problem but thats another deal entirely) I hope this makes sense. thanks again

[OB99W]

[...]unnfortunately I dont own a scanner that reads Subaru trans codes or a decent Subie manual. [...]

I think you'll find this thread, and the PDF file links within it, very helpful:

http://www.ultimatesubaru.org/forum/showthread.php?t=19805

 

General 4EAT info is here:

http://endwrench.com/pdf/feb2004pdf/4EAT.pdf

 

I suggest downloading the PDFs, and reading the entire thread.

[nipper]

Thankyou for the input, Nipper. I hope I dont sound thick here, but I do have a ton of experiance with electronically controlled transmissions, unnfortunately I dont own a scanner that reads Subaru trans codes or a decent Subie manual. Im aware that there are several solenoids, in fact other than the torque control circuit, the shift control circuits and solenoids in this thing are very similar to Nissan stuff. As I stated in the first post of this thread, a friend at a transmission shop scanned my car and came up with a code..I do not have the info in front of me, its at work, but it was a 'corrupted torque SIGNAL' code...and his snap-on scanner gives a summary of possible reasons for this code. It mentioned bad grounds and wiring problems between the TCU and the ECU, or the TCU itself. Now, believe me, Im not questioning your input. Thats why Im here! But from Nissan experiance only, this type of SIGNAL code would lead me to a electrical issue, esp. if the code was erased and imediately came back. I know in a Nissan (I hate to keep falling back on Nissan stuff, but its where Im most comfortable) if a solenoid was stuck, or there was a mechanical issue in the trans, it would set a code after the car was driven and the 'problem' occured. Where as a electrical problem would set a code right off the bat (unless it were a intermitant problem but thats another deal entirely) I hope this makes sense. thanks again

 

BE HEALED HEATHEN!!!!!!!!!!

doth not speakith of thy evil nissan

 

 

SUbaru is not nissan. Subaru specifically states that the tranny trouble light goes on if there was a previous failure (fault) the last time the vehical was operated.

 

The endwrench will be very helpful.

 

nipper

[86BRATMAN]

Subaru is not nissan.

 

True. But a lot of electronis crap on them is. At least the older models

[gyuunuuya]

http://www.ultimatesubaru.org/forum/showthread.php?t=55651

 

 

 

You could pull out the tranfer clutches and see if they are welded like mine were. Would be free, only take you a couple hours, probably less becuase sounds like u got a garage with lots of tools, and a LIFT.

[Bserk]

Been a while. Sarah Subie got put on the 'back burner' as other projects requiring more attention came up. But I had to post again to give a belated THANKYOU to all of you, but specifically to OB99W....With those files, I was able to do the on-board diagnosis, and see that I still had the 'solenoid C' code stored. Back to the wiring diagrams and long story short, I determined that I had no continuity to the solenoid, and after some head scratching and pin checks, pulled the tailhousing back off the trans. (here is where I get really embarrased, because I may not be a Subaru tech,but I am a master Nissan tech..) When I re-installed the tailhousing, it did have a bit of resistance for the last quarter inch or so. Thinking it was just rust on the allignment pins, I used my trusty rubber hammer and pop...on it went. What I had done was catch the one-pin electrical connector for that solenoid on some inner projection of the trans case, and smashed it. (enter lots of expletives here, when I saw what I had done) So, weeks later, a easy connector repair, and back together.....and my Subie RUNS, DRIVES, TURNS....NO Torque bind, NO flashing AT Temp light. And hopefully, no more problems for a while. Its Spring. And i bought this car for a winter rat. LMAO! Been quite a struggle, but damn, Im glad i found this site. Big time heartfelt thankyou to all who gave me advice. Might not even licence the ole' girl at this point, but next Fall.... I'LL BE BACK!!! Awesome site, you'all. THANKYOU again

[nipper]

We love happy endings ....

 

nipper

[OB99W]

Been a while. Sarah Subie got put on the 'back burner' as other projects requiring more attention came up. But I had to post again to give a belated THANKYOU to all of you, but specifically to OB99W....With those files, I was able to do the on-board diagnosis, and see that I still had the 'solenoid C' code stored. Back to the wiring diagrams and long story short, I determined that I had no continuity to the solenoid, and after some head scratching and pin checks, pulled the tailhousing back off the trans. [...] So, weeks later, a easy connector repair, and back together.....and my Subie RUNS, DRIVES, TURNS....NO Torque bind, NO flashing AT Temp light. And hopefully, no more problems for a while. Its Spring. And i bought this car for a winter rat. LMAO! Been quite a struggle, but damn, Im glad i found this site. Big time heartfelt thankyou to all who gave me advice. Might not even licence the ole' girl at this point, but next Fall.... I'LL BE BACK!!! Awesome site, you'all. THANKYOU again

You're very welcome, glad we could help.

 

 

We love happy endings ....

Yes, we do.

 

We also appreciate people who get back to us with the outcome in general, even if just to say that the problem didn't get resolved or that it did but wasn't anything we suggested; it gives us a chance to rethink the problem or add to the "database". Some systems run better on a bit of negative feedback, but of course the positive kind is always nice.

[Gloyale]

Now let's hope that the TCU isn't affected now. I wouldn't think having the C solenoid circuit grounded would be good for it.

[Bserk]

Luckily, it wasnt grounded. The connector was broken and came apart, making an open circuit. Im driving her on dealer plates still tonight, and all is well!