Jump to content
Ultimate Subaru Message Board

4EAT TCU Questions (96 Legacy Outback)


Recommended Posts

Greetings, everyone! I'm new to this forum but I'm a confirmed Subie owner having had an old Brat at one time, and now driving a '96 Outback with the 2.5L engine and a 4EAT. I bought it new in South Dakota and have driven literally from coast to coast & from Texas to North Dakota, and after 197K+ miles I'm just getting her "broken in." But I'm now getting one of those annoying ATF Temp flashing light messages from my TCU. Trouble is, I don't know how to interrogate the thing, so although I know something's wrong, I don't yet know what. This is a version 1 4EAT, having been produced early in '96. I've seen a handshake procedure elsewhere on the forum for interrogating the version 2 tranny controllers, but I don't know how to read a verson 1 box. What do I do??? THANKS!

Link to comment
Share on other sites

The attached files are from the diagnostics Section of '95 FSM for AT w. 2.2l (OBDII), I hope they would help.

AVK - This is great! This is exactly the data that I was looking for, and there appears to be no difference between the '95 and my '96. No surprise there. But I have another question. On page 13 of the Diagnostic Chart (section C - Onboard Diagnostics) it says to "move select lever to D and connect the diagnosis switch to ON." I don't have any idea what the diagnosis switch is. Whatizzit?? Is it the Diagnosis Connector 11 and Pins 12 from the Locations file? OK - I guess that's two questions. I'll stop now... :-)

Link to comment
Share on other sites

Josh: Your praise cost me only $20+download time at techinfo a year ago- much as I hate to admit! All I did was to give the files understandable names.

Wayne: I knew you'd ask. "Connecting the diagnostic switch" refers to joining a 1-pin connector to the specified pin on another one, with 6 pins. There's a footnote in the "code reading" section and an illustration in the "location" part. Please note that I didn't try this myself, but if you have any more questions, be sure to ask. To find the connectors, it's best to remove the trim panel under the dash and have a good light.

Add: I have looked at the wiring diagram and it shows the diagnostic connector as follows: It's black and looking into the plug with the latch facing up, there are two terminals in the top row, left and right of the latch, and four in the bottom row. The one to connect is no. 5, which is 3d from left (2d from right) in the bottom row. The other 1-pin terminal is simply on a ground wire (that picture shows there are two provided, which doesn't matter), so you can just as well connect the diag. terminal directly to the ground. I hope you find the same diagnostic terminal on your OB, otherwise you'd need the wiring diagram for your car.

Link to comment
Share on other sites

avk - Just so I'm sure I have it (and so I don't risk frying electronics in the process), let me try to re-state what you said. The illustration in the "locations" file shows a Diagnosis connector (Ill. #11) and Diagnosis terminals (#12) basically on the same harness. All I have to do is connect Pin 5 of the connector to either one of these terminals. Is that what you think this means? And you are sure that the two terminals are just connected straight to ground?

 

You can tell that I'm not near my car at the moment, so I can't verify that I've got the exact same configuration, but I'm guessing that they'd have had no reason to re-engineer this one between '95 and '96, especially since they would have been busy designing the Phase 2 TCU at that time....

Link to comment
Share on other sites

avk - Just so I'm sure I have it (and so I don't risk frying electronics in the process), let me try to re-state what you said. The illustration in the "locations" file shows a Diagnosis connector (Ill. #11) and Diagnosis terminals (#12) basically on the same harness. All I have to do is connect Pin 5 of the connector to either one of these terminals. Is that what you think this means? And you are sure that the two terminals are just connected straight to ground?

<...>

Wayne: yes, that's how I read it. There's actually another helpful file in the AT diagnostics section that I'm posting. It's a simplified diagram that shows a "diagnostic switch" between a pin on the TCU and the ground. There's no actual switch, but the schematics shows what's going on. Besides, codes for ABS and airbags are retrieved in a similar fashion: by grounding some pins on some connectors. You can double-check those single-pin plugs by measuring their resistance to the ground. The TCU pinout on this diagram (it's a '95 Impreza) might be different from yours, but it shouldn't matter if you still have the six-pin connector. There have to be some differences in TCU wiring between OB and Impreza: for example, my speed sensor is inside the cable-driven speedometer. But let's hope they kept the diagnostic connectors the same.

wiring.pdf

Link to comment
Share on other sites

Wayne: yes, that's how I read it. There's actually another helpful file in the AT diagnostics section that I'm posting. It's a simplified diagram that shows a "diagnostic switch" between a pin on the TCU and the ground. There's no actual switch, but the schematics shows what's going on.

Thanks!! I still am not near enough to my car to have a look at it (I've been driving our van to work until I get this issue resolved, and last night I got home after dark -- not a good time to work with your head behind the dash), but when I do I will post a message either confirming that my '96 Outback is the same as shown in your book, or if it is different I'll detail those differences. This forum is INCREDIBLE; I only wish I'd found you guys earlier!!!

Link to comment
Share on other sites

  • 4 weeks later...

Well, I am finally getting around to letting you all know that AVK's info is absolutely compatible with the '96 Legacy Outback Automatic! Thanks again for all the detailed data!!!! The reason it took me so long to get back to you is that the problem went away for a while before I had a chance to actually troubleshoot it.

 

But after about a week and a half it came back. Following AVK's indicated troubleshooting procedure, I got a code of 24 -- Duty Solenoid C. I checked voltage at the FWD fuse connector under the hood - 9.5 V is present. I inserted a fuse into that socket and drove the car a bit - the binding is still there. So it's either the solenoid itself or the wiring between the fuse and the solenoid that is bad. I don't have a good ohmmeter here at home and the spec resistance of the solenoid is only 17 ohms, which looks like a short on my meter. I'm seeing either a correct reading of 17 ohms, or a short -- I can't tell. The fuse I put in the FWD socket doesn't blow, so I'm probably seeing the correct resistance of the solenoid coil.

 

BTW, coincident with the problem coming back, I noticed some moderate to severe torque bind while negotiating a couple of switchbacks on a (paved) mountain road near here. But since the problem seems to be intermittent (at least for now), I have another question or two for you all. Just how is that solenoid connected to the TCU? I've found a wire that comes out of a metal fitting on the right side of the transfer clutch housing. Is that it? And how is that connected to the solenoid itself? Can I pull that metal fitting and check to see if I have continuity at that point? I want to know what I can do next (short of pulling the transfer housing off and yanking the solenoid -- that'll be my next weekend project if I can't fix this thing any other way!)

Link to comment
Share on other sites

Wayne: did you try to check the resistance according to the manual? Your trans. should have the same 16-pin plug (named T4 in the Impreza book, connecting to B9 -- even though the numbers themselves should be different on your OB). This is where the wiring for solenoid C passes through. The solenoid "pigtail" connects to the transm. wiring inside the case. The wire on the right side you mention is probably for the "speed sensor 1" (see the pic. in component location section). Get a good meter. RadioShack always has something on sale.

Good to know you could read the codes. To my knowledge you're the only one who posted with the results, although there were others who inquired how to do it, on USMB and elsewhere.

Link to comment
Share on other sites

I looked at the resistance at the FWD fuse socket - figuring that there was probably a straight run from there to the solenoid. Thanks for the info yet again -- I'll look for that 16 pin plug and I'll check right there! I'll keep you posted as to what I find.

 

 

Wayne: did you try to check the resistance according to the manual? Your trans. should have the same 16-pin plug (named T4 in the Impreza book, connecting to B9 -- even though the numbers themselves should be different on your OB). This is where the wiring for solenoid C passes through. The solenoid "pigtail" connects to the transm. wiring inside the case. The wire on the right side you mention is probably for the "speed sensor 1" (see the pic. in component location section). Get a good meter. RadioShack always has something on sale.

Good to know you could read the codes. To my knowledge you're the only one who posted with the results, although there were others who inquired how to do it, on USMB and elsewhere.

Link to comment
Share on other sites

I also thought the FWD fuse socket was directly in the solenoid circuit. But that diagram shows it's actually used to provide a control signal by grounding a pin on the TCU (together with the FWD warning light).

Hope this works out.

Link to comment
Share on other sites

Yeah, but that diagram is a little too sketchy. It shows the TCU as a black box, and while the fuse does provide continuity to ground, that still COULD be the drive voltage for the solenoid -- you're just completing the ground in the circuit instead of connecting the +12v side. It would really be overkill to use a 15Amp fuse just to pull a control signal to ground.

 

It could also be that the 9.5V I measured at the fuse pins is just a control signal, which when grounded switches a circuit in the TCU that then switches the relatively high current for the solenoid. I couldn't honestly say that I know for sure, just based on this schematic.

 

Still, the best way to tell exactly what's going on is to look both at the resistance of the solenoid at the other connector, and (if I can figure a way to do it) to monitor the actual solenoid drive current while the car is in operation. I'll have to think about how I want to do that....

 

This has now turned into next weekend's project, so I have a little more time to think about it before I proceed.

 

I also thought the FWD fuse socket was directly in the solenoid circuit. But that diagram shows it's actually used to provide a control signal by grounding a pin on the TCU (together with the FWD warning light).

Hope this works out.

Link to comment
Share on other sites

Wayne: I thought you can put the fuse of any rating in that socket because even the solenoid itself takes about 1 A if its resistance is as specified. What I saw on the diagram and didn't know before was that the fuse goes between TCU and the ground. The solenoid is operated by positive voltage and has a permanent ground of its own.

Connecting an ohmmeter in place of the fuse sends (small) current through both the solenoid and the TCU, so this would not give a correct reading but hopefully didn't cause any damage. Voltmeter is fine and it shows the open-circuit voltage.

I believe the 16-pin plug is one of the three sitting on a bracket to the lower right of the throttle body.

Link to comment
Share on other sites

The attached files are from the diagnostics Section of '95 FSM for AT w. 2.2l (OBDII), I hope they would help.
Those are good information documents. Do you have the section related to line pressure testing the tranny oil pump as far as operating pressures that are normal? I am still trying to sort through why my 94 legacy doesnot want to lock in at high speed. Thanks
Link to comment
Share on other sites

Those are good information documents. Do you have the section related to line pressure testing the tranny oil pump as far as operating pressures that are normal? I am still trying to sort through why my 94 legacy doesnot want to lock in at high speed. Thanks

I do have the pressure testing section. It would take my attachments slightly over the 1MB limit and it may be early to delete them, but I'd be glad to email it to you.

Link to comment
Share on other sites

Thanks. I agree with you; this is just completing the circuit for a control signal in the TCU, and any fuse should do. But some documentation that I've seen (although I can't remember where), suggests using a 15A fuse. It definitely is overkill.

 

More interesting observations: I haven't yet probed the connector that goes to the solenoid, but I have been driving the car with a fuse in the FWD socket for several days now, trying to figure out under just what conditions the solenoid problem manifests itself. I have found that the solenoid correctly locks into FWD mode when I have a fuse in the socket every single morning (when things are "cool") and generally stays there UNLESS the car is driven for extended periods (30+miles) at freeway speeds on very hot (98F+) days. Once the failure shows up, which happens while I'm driving, the FWD light goes off and the solenoid stays locked in AWD mode even with a fuse in the FWD socket. It stays that way until the car cools off fully (i.e. overnight). It's an interesting heat related failure, but I still can't figure out if it's the TCU, the connections to the solenoid, or the solenoid itself.

 

I'll do some more quantitative voltage testing over the weekend and I'll let you know what I find.

 

Wayne: I thought you can put the fuse of any rating in that socket because even the solenoid itself takes about 1 A if its resistance is as specified. What I saw on the diagram and didn't know before was that the fuse goes between TCU and the ground. The solenoid is operated by positive voltage and has a permanent ground of its own.

Connecting an ohmmeter in place of the fuse sends (small) current through both the solenoid and the TCU, so this would not give a correct reading but hopefully didn't cause any damage. Voltmeter is fine and it shows the open-circuit voltage.

I believe the 16-pin plug is one of the three sitting on a bracket to the lower right of the throttle body.

Link to comment
Share on other sites

  • 2 weeks later...

Hi Folks,

 

I've got a '96 legacy outback, 2.5L 4EAT automatic. Been reading and researching, since I have the binding during turn problem described in this thread. I was able to get the codes, finally, and the code was 24 which pointed to the duty c solenoid. Putting the fuse in the FWD socket under the hood does not light the FWD light, and the binding is still there. I suspect the 15 amp fuse is specified because its commonly used and is probably loaded as a spare.

 

So this means the duty c solenoid cannot release to FWD only, and is therefore bad, right?

 

I had some trouble identifying the diagnosis connector and the diagnosis terminal. As I subsequently found it, I'll describe it here and maybe someone will benefit. The diagnosis connector's lead wires are wrapped with black tape. The diagnosis terminals (there are two which are electrically connected) are black wires with small pin terminals on the end, which were wrapped in the black tape on the diagnosis connector's lead wires.

 

Looking at the diagnosis connector carefully, it has two thick wires, one white and one blue. It has three thin wires, first orange / white stripe, second blue / yellow stripe, third green / white stripe.

 

To connect for diagnosis, put the terminal of one of the black wires (doesn't matter) into the socket in the connector with the blue / yellow stripe wire coming out. As a check, the blue / yellow wire goes back to the TCM, pin c6 there on TCM terminal B53, which is a 20 pin. This agrees with the schematics.

 

After the black is connected to the blue / yellow (through the terminal) follow the diagnosis procedure, and the codes will flash repeatedly.

 

All this was possible with the 4 .pdf's found in page 1 of this thread, and with the .jpgs found at http://www.ultimatesubaru.org/forum/showthread.php?t=18356&highlight=binding

 

BTW, thanks very much for this excellent info.

 

Hopes this helps. Regards, Bryan

Link to comment
Share on other sites

Thanks, Brian! Your description is far and away better than any I've read here -- and that includes mine after I found the connector!! :grin:

 

Anyway, I've continued to work on my Duty Solenoid C problem. As you know I put a fuse in the FWD socket and the FWD light came on only for about 1/2 hour's driving; then it would go out and would stay out until the car cooled overnight. After about a week of this the FWD light refused to come on at all any more, whether there was a fuse in the socket or not.

 

Cheapskate that I am, I had to try all the other inexpensive fixes before I shelled out cash for a new Duty Solenoid C, so next I tried new synthetic AT fluid and a friction reducing additive for about a week. I didn't have the AT "flushed," but I did do 3 partial drain/refills by pulling the AT pan drain plug, draining about 3-4 quarts, then topping off and running the car in all gears for about 20 mins in between each drain/refill cycle. After all that, the FWD light stayed off. I figured the Solenoid had to be shot at that point, so I ordered a new one. It has taken about a week to arrive, and in that time I've still been using the car for average commuting (about 40 miles each way, mostly freeway driving).

 

Then, wouldn't you know it!!??!! The solenoid that I ordered arrived this afternoon and as I drove home from work today, before I had a chance to swap solenoids, the FWD light came back on (I'd left the FWD fuse in the whole time)!!! :D

 

I drove the car around a little more this evening and the light is staying on. I pulled the fuse for a little bit and the light went off, but the torque bind on turns is now much less severe than before. I popped the fuse in again just now, and the FWD light is on just as it is supposed to be. Moral of the saga -- CLEAN and FRESH AT fluid is essential for proper operation of this transfer clutch assembly!!

 

I'll probably still pull the transfer case and inspect (and possibly change out) the solenoid assembly tomorrow morning. I'd still like to know exactly what's going on in there! :o

Link to comment
Share on other sites

  • 3 weeks later...

My pleasure Wayne.

 

Thought I'd post the outcome of my adventures in binding while turning = duty c solenoid.

 

I called a couple of local Subaru specialists, and only one could follow along while I went through the diagnosis developed from the help I got here. The guy started reading me part #'s to help me get the stuff to do the job myself! How many garages would do that? Anyway, he quoted me $250 for the entire job, including parts! New solenoid, two small gaskets for the plate under the solenoid, and the transmission nose gasket. Needless to say, for that price I had him do the work.

 

My '96 legacy outback 2.5L 4EAT now works fine. The garage found the Y exhaust pipe rotted, so they welded on a new neck and replaced the exhaust donut. Total price for all this driving away was about $350. If anybody in VT or NY would like to know, I'd be very glad to recommend these guys.icon10.gif

 

So Wayne, I looked over your thread, and I didn't see if you found the trouble code for the duty c solenoid (code 24). The intermittent / heat related nature of your symptoms really makes me think it is electrical. If I recall, your FWD light doesn't light when the symptom shows up, which to me also points to electrical. Have you done the checks listed in the .pdf manual posted in this thread for the duty c solenoid connections? If you can get the symptom to repeat in the heat, maybe try one of those gel freezer packs on the TCM to see if cooling it down makes the symptom go away. The TCM has to switch electricity to the solenoid to get it to release, and there may be an electronic component (probably a transistor or an FET) that gives up when it gets hot. Back in my electronics days, we used freeze spray to pinpoint components failing on a PC board in much the same way. The heat could also affect an intermittent connector; there are several in the circuit.

 

The FWD fuse / switch is almost certainly on an isolated input of the TCM, because the TCM can sense failure modes through the output to the duty c solenoid. If adding the FWD fuse controlled the solenoid directly (we know one fuse leg is at ground), the TCM would not be able to sense the state of the duty c solenoid circuit, and in fact would sense a short to ground and light a trouble code when the fuse is used. The TCM diagnostics need to sense open, shorted, and resistance (normal). BTW, my FWD light didn't work with the fuse in because of my code 24 / duty c solenoid problem.

 

Let me know if you can't find the circuit diagrams to which I refer. I'm curious to hear where you're at with it.

 

Good Luck, Bryan

Link to comment
Share on other sites

Bryan,

 

Yeah, since I've been discussing parts of this "adventure in troubleshooting" over a couple of different threads, I forgot to explicitly indicate here that I DID get the Duty Solenoid C circuit fault message (24 out of the TCM) and I continue to do so whenever I feel evidence of torque bind. Here's what has happened in the past several weeks, just to bring you up to speed. I have driven the car with a fuse in the FWD socket most of this time. I have not yet swapped my duty solenoid, mostly because it will require some significant disassembly of the exhaust system in order to allow me to get to the transfer housing, and I'm saving all that work for later this year when I have enough money available to replace the catalytic converters, too. I have also recently been contending with a check engine error P0420, which inidcates that either my O2 sensors are not being read correctly or that the cats are bad. With 200+K miles on the car now and with both O2 sensors having been replaced within the past year, I suspect I need new cats. Anyway, I'll do all the mechanical disassembly at one time.

Back to the Solenoid C problem diagnosis - I have noticed that the proper switching of the solenoid (i.e. "proper" working of the FWD light with a fuse installed) does seem to be a function of temperature, but I haven't yet been able to definitively determine if it is an electrical problem with the TCM or a mechanical issue with the solenoid sticking. I have tried a can of freeze spray on the TCM, and that didn't seem to have any immediate effect on the problem. By that I mean I sprayed the TCM and chilled it down to the point where there was external frost on the box on a day when the FWD light was not coming on, and that did nothing in the short term to alleviate the problem. But, when I went out to the car later that day (maybe an hour or so later) the light came on as soon as I started the car, and stayed on during the 30 minute drives I made down to the store and back. I can't say for certain if this means that I needed time for thermal soakback to occur, or if this was just one of those times when the solenoid correctly latched up and stayed there. Lately, it has been too cool here in SoCal for the problem to show up at all -- so I may not be able to continue the troubleshooting until we get another spell of really hot weather. I'll still probably replace my solenoid, now that I have the parts here all ready to go. BTW - given the amount of labor involved in this repair, I think you were quoted a very reasonable price. You have a good repair shop there -- too bad I live so far away!!:)

 

 

 

My pleasure Wayne.

 

Thought I'd post the outcome of my adventures in binding while turning = duty c solenoid.

 

I called a couple of local Subaru specialists, and only one could follow along while I went through the diagnosis developed from the help I got here. The guy started reading me part #'s to help me get the stuff to do the job myself! How many garages would do that? Anyway, he quoted me $250 for the entire job, including parts! New solenoid, two small gaskets for the plate under the solenoid, and the transmission nose gasket. Needless to say, for that price I had him do the work.

 

My '96 legacy outback 2.5L 4EAT now works fine. The garage found the Y exhaust pipe rotted, so they welded on a new neck and replaced the exhaust donut. Total price for all this driving away was about $350. If anybody in VT or NY would like to know, I'd be very glad to recommend these guys.icon10.gif

 

So Wayne, I looked over your thread, and I didn't see if you found the trouble code for the duty c solenoid (code 24). The intermittent / heat related nature of your symptoms really makes me think it is electrical. If I recall, your FWD light doesn't light when the symptom shows up, which to me also points to electrical. Have you done the checks listed in the .pdf manual posted in this thread for the duty c solenoid connections? If you can get the symptom to repeat in the heat, maybe try one of those gel freezer packs on the TCM to see if cooling it down makes the symptom go away. The TCM has to switch electricity to the solenoid to get it to release, and there may be an electronic component (probably a transistor or an FET) that gives up when it gets hot. Back in my electronics days, we used freeze spray to pinpoint components failing on a PC board in much the same way. The heat could also affect an intermittent connector; there are several in the circuit.

 

The FWD fuse / switch is almost certainly on an isolated input of the TCM, because the TCM can sense failure modes through the output to the duty c solenoid. If adding the FWD fuse controlled the solenoid directly (we know one fuse leg is at ground), the TCM would not be able to sense the state of the duty c solenoid circuit, and in fact would sense a short to ground and light a trouble code when the fuse is used. The TCM diagnostics need to sense open, shorted, and resistance (normal). BTW, my FWD light didn't work with the fuse in because of my code 24 / duty c solenoid problem.

 

Let me know if you can't find the circuit diagrams to which I refer. I'm curious to hear where you're at with it.

 

Good Luck, Bryan

Link to comment
Share on other sites

  • 6 years later...

Hi guys.

 

How do you know which is pin 5 on that on the diagnosis connector (#11)? I see 4 wires: thick blue, thick white, thin white-with-blue-stripe and thin red-with-yellow-stripe. Which of those is pin 5?

 

(Near that connector are the 2 black solitary wires with blade-like connectors, which I take to be diagnosis terminal, #12.)

 

Thanks!

Link to comment
Share on other sites

i think it is left to right across the top and then left to right across the bottom. but in the long run it does not matter. they are all diagnostic pins. if the one you ground out does not flash the AT TEMP light, try another one. the other pins are for the air bag codes and the abs codes, etc .....

 

and yes the 2 single leads are the grounds to use.

Link to comment
Share on other sites

  • 1 month later...

Hello I have been reading this thread. Doing research.

 

I have a 95 Imp L with 2.2L dualport exhaust. OBDII. I installed a 94 JDM Legacy Twin Turbo 4EAT 4.444.

 

I am having the binding issue too. I also have shifting problems. In drive it stays in first. If you shift manually I can go 1st to 2nd hard then 3rd nothing and drive nothing. Any ideas????????

I was thinking that maybe I need a JDM TCU from the same car.

 

Any Ideas? I really appreciate it guys. I need to get it running so we have a way to get to the show. Jamie's loyales tranny just too a s**t.

 

Thanks Clay

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...