auto tranny AWD convert to part time 4wd?

[herr_diesel]

OK, hear me out.

my 90 legacy auto wagon runs great. problem is my AWD seems to work only when it wants too. there is serious delay from when the front wheels break loose to the rear wheels kicking in. when they do, it's awesome. i usually end up doing a 360 or some other non-intentional rally move.

i've asked around and know that duty soleniod B or C controls this center clutch, and has been known to wear out.

here's my idea: can't i override whatever soleniod that is turning the center clutch on and off with a switch, ie i turn on and off 4WD when i feel it is needed. plus that useless manual switch would be a perfect candidate for it.

i can deal with the slushbox and the crappy shifting, but here in montana, i need AWD to work. what do y'all think...

[ciper]

I cant say I ever heard of anything like this happening.

To answer your question why not just wire up a fuse and switch to the FWD fuse under the hood?

I think I can speak for many of us that we would rather help you fix the problem then to disable the AWD.

The FWD fuse I speak of is on the passenger side right next to the windshield wiper motor.

[rallysam]

You could always just replace that solenoid (that's my smartass comment for the day )

This sorta reminds me of a thread somewhere on this board about wiring up that fuse so that the emergancy brake disabled AWD. That let the guy really lock up the rear wheels while cornering. I thought that was pretty friggen cool.

[rallysam]

herr_diesel: I think I understand. What I'm hearing is that you are trying to control that solonoid (instead of controlling the fuse) because you want to force those clutches to be engaged constantly - not just when the front wheels slip.

Something doesn't make sense to me. I thought those clutches were already engaged full-time??? Is it really true that NO power goes to the rear wheels until the clutches react to front-slippage????

[vrg3]

Well, *some* power goes to the rears, but the multiplate transfer clutch is partially slipping. When the solenoid activates, it locks the clutch.

If your solenoid is going bad, you simply won't be able to maintain the fluid pressure to keep the clutch locked, even if you override the TCU with a switch. You're going to need to replace the solenoid anyway.

[herr_diesel]

rallysam, you are with me. the roads here are almost always snow covered during the winter, and i don't see how AWD could be running all the time, like in the manuals. my dad's 91 legacy MT sedan works perfectly, and i'm convinced the MT's center diff is far superior than this infernal center clutch that turns on and off.

if i'm on a straight stretch of icy road, and i get on it, the revs kick up high as the front wheels break loose. then, abruptly i might add, the clutches engage (my understanding is the solenoid senses a difference in fluid pressure between the front and rear axles and this pressure differential causes the solenoid to trip, therefore engaging the rear axle) and once they do engage (if i don't spin out of control), and i continue to keep on the gas, you can even feel the rear axle "surge" ie switching on and off as i continue to accelerate. you can watch the tach go up and down as it goes from two wheel to four wheel.

i guess i should clarify some. i guess what i'm trying to do is the opposite of the fuse under the hood. i want to trip it ON when it gets slick. now i know those clutches weren't designed to be operated constantly for an extended amount of time, but i'm sure intermittant use would be fine.

i now the figure is supposed to be 90/10 for the AT torque split, but it sure seems like it's 100/0 until those clutches wake up and go to work. i can't imagine slipping as much as i do with even 10% going to the rear.

i've changed the filter IN the tranny (which was somewhat of a pain) and the external filter, changed all fluid, added some fancy schmancy additives, and it still has a significant delay/intermittany activation.

there are about a thousand subarus in this town, and having worked at a parts store, i asked everyone who owned AT early legacy's, and it sounded like i wasn't the only one with this issue.

i've got a line on two Justy's for sale, one w/ a blown motor, the other w/ a blown tranny. 2 for 1 deal, couple hundo $$. at least with those there is an actual switch to put it in 4wd. i'm so frustrated at this point, i'm ready to go that route or swap in a MT, which i've been wanting to do for some time now, and didn't think was realistic, that is, until i found this board!!

[herr_diesel]

vrg3:

but isn't the solenoid just a hydraulic switch? doesn't it take fluid pressure difference and converts it to a hi/lo electrical signal? i want to bypass it with a toggle switch activated by yours truly. rather than fluid pressure kicking on the clutches, it would be my switch inside (i still think the ever so useful MANUAL switch that resides on the shifter would be perfect).

i know the responsible thing would be to repair it correctly, but come on...i'm an engineer, what fun would that be?

[vrg3]

This solenoid isn't a hydraulic switch exactly. It is a hydraulic valve, controlled electrically. The Transmission Control Unit (TCU) is responsible for the signal sent to it.

The valve controls hydraulic pressure on the transfer clutch actuator Basically, when the valve is open, the clutch is activated and power goes to the rear wheels. When the valve is closed, it's not activated.

The trick is getting in-between results. The TCU constantly cycles the solenoid on (open) and off (closed), at some fixed frequency, probably something like 10 or 20 times a second. It varies the duty cycle to control how much pressure gets through. For example, if it wants 40% pressure and the frequency is 10 times a second, it turns the solenoid on for 4/100 of a second and then off for 5/100 of a second, repeatedly. A 40% duty cycle. It's not actually linear that way but it's pretty close..

The TCU controls this duty cycle based on the two vehicle speed sensors, what gear you're in, how hard you're accelerating, and how hard you're turning.

It sounds to me like your solenoid is sticking. If it had junk trapped in it, it would still be able to open and close, but it wouldn't be able to oscillate at the frequencies needed by the TCU.

If indeed your solenoid is dirty, I think you should clean it out, so it doesn't eventually become completely inoperable. Taking it out of the extension case is a pain, though, so maybe you could try just flushing your transmission fluid a few times (when was the last time you changed your ATF?), being sure to drive it and activate the solenoid between each flush.

But, in any case...

If your solenoid is actually functioning, you could indeed wire a switch to control it. Basically, just find the right wire to activate the solenoid and rig it up with a switch to connect it to a fused +12v source. My service manual is a little unclear about which pin on the TCU you want. If I were doing it, I'd probably just clip the wires down at the solenoid itself and run my own heavy-gauge (since that's a pretty strong solenoid, and the stock wiring was not meant to run it for a long time) wires down there.

[vrg3]

Sometimes I wish I still had an automatic Legacy lying around to try this kind of stuff with... When I had my old '91 Legacy I always wanted to rig that "MANUAL" switch up to be a "4WD" switch like on our old GL wagon.

This shouldn't be too hard to do. Good luck, Herr Diesel. Oh, by the way, do you drive a diesel of some type too?

[herr_diesel]

vrg3:

hey, thanks so much for the useful info. i've been scratching my head for many many moons about this, and then i stumbled on this site. it's great communicating with folks that are familiar with this stuff. it's amazing the number of theories that exist as to just how these cars work...

so let me get this straight...the clutch is controlled hydraulically? via the valve, which is controlled electronically... ok, that makes sense. i see your point about controlling a valve that might not be functioning properly...guess this would be a good way of isolating the problem to the solenoid.

my radiator is cracked, and i've tried fixing it with epoxy a number of times, but no go. i'm replacing it this weekend, and while i've got it in the shop, i'll crawl underneath it and do some inspecting. i've never pulled the back of the tranny off before, but i guess it's time for me to have yet another one of my cars disabled for no real reason...

as far as the diesel thing goes, i used to have a 99 TDI jetta. i cry everytime i think about that car. we bought it new off the lot. loved everything about that car. best mileage ever was 60.2 mpg. i could go over 800 miles on a tank. i personally think that is one of the best motors ever produced. but alas, we were young and poor, and the payments left only raman noodles for dinner...so we sold it and got the legacy. been a great car, other than this AWD thing... i wish subaru made a horizontally opposed diesel...that would sound sweet

can't wait for the US to adopt European diesel standards.

if i do override this thing with a switch...am i going to get error codes out of the TCU? just a thought...

[ciper]

For those of you wishing for a manual, you are really confused at which is the better AWD system. Either your car is not functioning correctly or something else....

I own multiple legacy plus soon two more subaru, all of which are AWD. No decision needed, Id always choose the center differential on the autos.

I have the most experience driving in low traction with my 90 wagon auto. The rear wheels never slip then engauge like has been explained. The only time this happens is when the amount of traction differs between the front and rear.

The amount of lockup is at its maximum level when traveling in first gear, and decreases as speed increases. The amount can be "manually" increased whenever the TCU detects slip.

The only time I have recreated the problem explained above is when the front wheels are on water/ice and the rears on dry pavement, then the amount of lag was VERY small.

The manual transmission center differential works like the rear lsd on the 91 turbo, think of it this way:

Take the rear diff out, on the side that the left tire was attached is now the output to the front wheels, the side that the right tire was attached is now the output to the rear.

You can see that when starting out all 4 wheels will get near equal torque, so if one slips the rest stay still, UNLIKE THE LOCKING ABILITY OF THE AUTO!!!

Sure the center diff will start to heat and increase the amount of maximum torque that can be applied, BUT you have another problem.
Say one front wheel was spinning, the speed of the front output shaft will be spinning at HALF speed since the other wheel is staying still. Or in other words you will have to spin the hell out of that one wheel before the center diff starts to react.

Sure I just explained worse case scenarios, but you can see that the AUTO can lock 50-50 while the manual is still only a limited slip. It would be the difference of drag racing with a locker and drag racing with just an lsd, and you should know that difference.

[vrg3]

Herr Diesel, I know exactly what you mean. I used to have an '84 Mercedes 300D. Diesel's so awesome. My friend just bought a '98 Jetta TDI and it reminds me so much of my old Benz. There's enough they have in common, being both German and diesel, that I get all teary-eyed when I think about it.

He's going to start trying to manufacture biodiesel when the weather warms up. That should run at least as well as the European fuel, even when blended with dino-diesel. We're going to improve the intercooling and try some electronic trickery to get better performance.

I don't know for sure about the error codes, but I guess it's possible the TCU could notice a missing Duty Solenoid C. Maybe you could replace it with any other old random solenoid or even a relay or some other kind of inductive load, just to trick it. It shouldn't be able to notice that the solenoid isn't actually controlling the multiplate transfer clutch.

[herr_diesel]

ciper:

so then my idea of manually overiding the center clutch (note i didn't use the word differential) with a switch would be the only way a legacy could be true 4 wheel drive(F and R axles locked together)...right?

here's another scenerio:

when i'm driving my dad's MT legacy and i hit the e-brake, all four wheels stop. when i do it in my AT, the ass end flings around just like a FWD car would.

at this point i'm sure my car ISN'T working as it should, because there's no one in their right mind that would think a rear axle engaging at 4000 rpm is a good idea. everyone i've spoken to says the engagement should be nearly intantaneous.

the main reason i'm not willing do it the "right" thing and fix the problem is i'm curious and lazy...it seems the logical fix without having to replace any real hardware, just a matter of cutting and splicing some wires.

i personally HATE this whole awd concept. i guess to me it represents how "disconnected" we have become from our cars in our society. back in the day, folks had to know so much more about mechanics because cars were so unreliable. nowadays, +200,000 miles out of a car is almost the norm, and manufacturers cater to the rich and clueless. i miss the days of a REAL lever that you had to reach down and pull, with ACTUAL linkages into the transfer case. simple yet so effective. so many people have no concept of when to go from 2 to 4 WD on a car/truck. i once met a couple that thought 4WD was for towing!!!! they wondered why they kept blowing out u-joints on their toyota 4-runner...

my dream would be a subaru with locking hubs...in the REAR!! and a big ass lever inside that said "4wd", plus a hi/lo range (why the hell did they stop doing that??). oh, and locking diffs front and rear...all controlled by me! no computers allowed!!

yea....that would make me happy. could you say i have control issues??

anyone have scans of the FSM for a 90 legacy L 4EAT electrical section?

thanks for letting me vent...

PS vgr3, have you checked out http://www.tdiclub.com/
TONS of info on hopping up TDI's...

[Legacy777]

herr,

Your scenario of pulling the e-brake does not justify anything, and is very damaging to the AWD system, Especially the MT equipped car.

The design on an automatic transmission will NOT produce the same results as a MT. There is no direct connection between front and rear wheels. Everything is done through hydraulics.

Go find some mud or snow or some other slick surface and gun it.....have someone outside the car and have them watch the front and rear wheels.

[IggDawg]

Check out this link:

http://forums.nasioc.com/forums/showthr ... did=227548

This guy was and still is developing a sort of DCCD for us auto guys. For the pre '02 WRXs it'll allow us to go from 100/0 to a 50/50 torque split on demand, or let the computer control it. he's also intergrating a TC lockup button. what does all this have to do with your problem? even though it won;t benefit WRXs, I've got him to keep developing the e-brake interlock switch taht'll disengage the clutch packs when you pull the e-brake. fun huh? anyways. with all this you can select the tranny to be in 50/50 whenever you want so there's no lag. you don't wanna drive around like that all the time tho. real hard on your diffs and tires when cornering.

PS - transfer cases haev teh sux0r. I'll take the multi-plate clutch anytime.

[herr_diesel]

777, little testy this morning?

the e-brake scenario was just to illustrate how the two different systems work, mostly for my own understanding. so, yes, they do justify something. and given the fact that all this was done on shear ice, the system didn't see any more stress than normal conditions...are you kidding me?

this has me thinking...does the TCU already have an e-brake overide? by this i mean open clutch when lever pulled? i guess this would make sense, because of the risk of damage.

there would HAVE to be a direct connection between F and R wheels, absolutely. but only when the center clutch is fully locked up ie. 50/50 torque split. and this is precisely what i am trying to create by having a switch to overide the TCU. i agree everything is done through hydraulics, but the hydraulics simply are a mechanism in which the valves and solendoids actuate the center clutch.

i have perfect testing conditions every single morning on my way to work. roads are ice-packed and snow covered here for months on end, and this is part of why i get so frustrated feeling my car turn on and off AWD all the time. i'm smart enough to know when it should be on or off, and am just looking for a simple fix...

IggDawg:

thanks so much for that info. this sounds like something i might be interested in...i'll be sure to watch that link for more info...[/quote]