Subaru Impreza GC8 & RS Forum & Community banner

321 - 340 of 346 Posts

·
Registered
'02 Forester, '15 WRX, '69 360
Joined
·
7 Posts
left to right wheel slip? Like torque steer? Ideally you shouldn't have any as that's the one key feature of subaru's "asymmetrical" awd.
No, not torque steer. Like having a front or rear diff lock. The factory rear limited slip is basically useless. As far as I know, there is no way to put a real diff lock into the front or rear. That is a huge downside when you're doing hard core off-roading.
 

·
Registered
sprinkleboat
Joined
·
18 Posts
put the 4eat center diff into a 5mt and make a diy rally dif so you get 50/50 lock w/ handbrake deactivation.


probably already been asked but if not it could be rad.
 

·
Registered
1995 EZ30 Coupe
Joined
·
718 Posts
If I'm not mistaken, the 4EAT uses a clutch pack instead of a center differential. A solenoid tightens a band around said clutches to engage the rear output shaft, that's why we can control it with a switch. The 5MT has actual mechanical differentials, no electronics involved, different mechanism entirely

Also, Dirty Bucket is not trying to lock the front and rear together, the 5MT has a 50/50 split already. Rather he wants to lock the front differential itself, in order to lock the left and right wheels together. Same with the rear. Normally the differential will let one side spin and the other side does nothing, unless you have a Limited Slip Diff.

I say put Torsens in the front, center, and rear diffs, see how that goes!
 

·
Banned
2001 Impreza 2.5RS+T 2001 Legacy GT
Joined
·
1,236 Posts
I say put Torsens in the front, center, and rear diffs, see how that goes!
You'll go where ever the front wheels are pointed!! :projectsti.com:
 

·
Registered
1995 EZ30 Coupe
Joined
·
718 Posts
It’s hard to justify spending $1200 on an LSD when I’ve just about spent that purchasing the entire vehicle.
I've heard welded diffs aren't THAT bad... :blol:

Actually I do intend to try welding one of my spare rear diffs sooner or later. If it's tolerable I'll weld the front too.
 

·
Registered
'02 Forester, '15 WRX, '69 360
Joined
·
7 Posts
welcome to the world of modding these cars.
True. I guess I did spend about that on the lift and tires, so what's another $1200?

I've heard welded diffs aren't THAT bad... :blol:

Actually I do intend to try welding one of my spare rear diffs sooner or later. If it's tolerable I'll weld the front too.
I have considered that. Would definitely solve the issue, but it's my daily, so I don't want it to munch tires or be difficult to manage. I did read a thread about someone who managed to put an ARB air locker in the rear, but it really seemed like a pain. Involved making custom CVs and grinding out a bunch of the diff housing. Would be awesome though.

What kind of LSD would be the best for an off-road application? Helical, Torsen, or viscous?
 

·
Registered
'02 Forester, '15 WRX, '69 360
Joined
·
7 Posts
I can't remember where I read it, but I read that a Torsen or helical (can't remember) is as good as an open differential when there is one wheel off the ground. It needs the resistance to work properly. Wish I could remember which one.
 

·
Registered
'02 Forester, '15 WRX, '69 360
Joined
·
7 Posts
So, I just read something interesting. For anyone with a true off-road Subaru, there is a company in the process of making a true locking rear diff. Should be out as early as June. I plan on picking one up as soon as I can.
Here is the link
Facebook
 

·
Registered
2001 Impreza Outback Sport
Joined
·
54 Posts
So, I just read something interesting. For anyone with a true off-road Subaru, there is a company in the process of making a true locking rear diff. Should be out as early as June. I plan on picking one up as soon as I can.
Here is the link
Facebook
Whoa, that's sweet. Wonder how much it'll be???
 

·
Registered
1995 Impreza L & 1997 OBS
Joined
·
2,258 Posts
I'm not electronically inclined in the least. I was wondering if anyone on here would be able to build one of these units that I could then install into my car. I would gladly pay for labor and parts.
 

·
Registered
Joined
·
12 Posts
Step 1 - Build the dummy load:

A quick look at the wiring diagrams and diagnostic sections for the 4EAT shows this page:


Side Note: I've included pictures of all the relevant info you'll need, but you really should get a copy of the Factory Service Manual (FSM) for yourself - they're free, AND there's a download link at the top of this forum.
http://www.rs25.com/forums/showthread.php?t=73249

If you have the FSM, the relevant information can be found here:
DIAGNOSTICS SECTION/Automatic Transmission and Differential/Diagnostics/Transmission Control Module (TCM) I-O Signal.pdf
/page3

Over on the far right we see that the solenoid presents between 10-17 ohms of resistance.


Consequently, this is the number that the TCM expects to see when the solenoid is functioning properly. Above or below that number by too much and the TCM will throw a fault code. This means that we need to install a resistor that behaves in a similar way. The easiest way to do this is to go buy a 10-25 watt, 10-17 ohm resistor. Unfortunately you won't find one of those down at your local Radio Shack - that's just too specific - you're going to need to build a custom setup. Now, there are a bazillion ways to wire up resistors to get the results you need. I'm not going to go into gory details about ohms law, and all the options unless asked to do so. Suffice it to say that if you buy what I listed above, and build it as I instruct, you'll end up with a dead-sexy 10watt, 17ohm resistor.

Let's build it -
First I like to check each of the resistors to make sure they are working properly. Set your multimeter to check impedance (ohm symbol - looks like a horseshoe) and test each resistor by touching the probes to the leads of the resistors (polarity doesn't matter here). Ideally you'll see a number close to the rating of the resistor - in our case 50 and 100 ohms.
If everything checks out, grab your perfboard, and mount your resistors by pushing the leads through the holes and gently bending them to hold everything in place. Now would be a good time to add some little pieces of heat shrink tubing to the exposed leads on the resistors (see later pics). Mock everything up and make sure you're happy before moving on.




You'll notice I have my resistors mounted in a set of heat sinks. Again, you don't have to do this - I'm a bit obsessive and tend to overbuild things, plus I had them laying around.

At this point I like to mark the locations of the resistor leads, and in my case the heat sinks. After you've marked everything - disassemble or at least pull everything away from the board an inch or two so you can get the glue gun in there.


Now take your hot glue gun and put a bead where each resistor body is going to rest on the perfboard and quickly push the resistor back in place. This will firmly attach the resistor to the perfboard so it won't vibrate out over time. Repeat for the rest of your resistors and whatever else you may have.

Now it's time to solder up the electrical connections on the back side. Now's a good time for a quick electronics tidbit: We're wiring these resistors in what's known as a "parallel circuit" as opposed to a "series circuit". I'll skip the mumbo jumbo, and math about why, and just say that I like the idea of having multiple resistors to share the load. These things get pretty warm during use, and more surface area to dissipate heat is always a good thing.

What I did was to cut and strip three separate pieces of wire for each set of leads - you can just run one continuous uninsulated wire as well - just make sure you insulate it against shorting out somehow - even electrical tape is better than nothing. Regardless of you wire choice run a wire up one side, and down the other. Now nolder each of the resistor leads to the wire and you're done.
Do NOT bridge the leads of any single resistor - just connect all the leads on each side. When you're done soldering, snip the excess resistor leads - you don't need anything past the actual solder joint.





Now it's time to check the overall impedance of your finely crafted dummy load. Just like before, put the probes of your multimeter on the leads of the first resistor. Only this time, since the resistors are all part of the same circuit, you should get a cumulative reading of all the resistors. If you've done it right, and your resistors were within their specs, the number should be very close to 17ohms.


If so, your dummy load is complete - congratulations. Have a bite to eat and a soda to celebrate, because now it's time to head out to the car - which only means lots of contortions under the dashboard.

Next - how to run wire in a car....upside down!
Why are all these photos blurry ?
 

·
Registered
Joined
·
6 Posts
My apologies as this thread is old but such a cool idea. I’m having trouble implementing, however.
I’ve got a 99 Legacy, automatic. My tcm/tcu has two connectors and I found my fsm and all seemed well. I find a White wire with Black stripe on the B54 connector in pin A15, and I go to test. My fsm seems to confirm this is the AWD solenoid. So with FWD fuse in (using spare 10A fuse), I get 9.5 volts on this wire. Great so far.

Then with fuse out , 1st gear and throttle I get 6.5 volts (yes, six point five). Now not so great. Obviously expecting less than 0.5 volts. My diagram seems to match what I’m confirming visually. And clearly there is a voltage change depending upon whether the FWD fuse is there or not on this wire. I even tested the wire next to it in pin A14 because some of the diagrams assume you’re looking at the connector from the pinhole side vs the wire side, just to be safe. Still no luck confirming greater than 8.5 v with fwd fuse and less than 0.5v in 1st gear, no fuse, and open throttle.
Im about to go test every pin unless there are ideas...? Why 6.5 volts? What does that mean?

Fyi, I do have an engine swap that was done by a subie tech who owned the car before me. It is a 2.2 , I think from an outback or Impreza sport. It’s a 2.2 that shares design similarities with the 2.5s; even had to do head gasket fix, but I digress. I don’t have all the details on this engine, obviously, but I wasn’t too concerned being that the transmission diagrams seemed to match what I saw. And I don’t think there was a tranny swap. Hopefully there isn’t some weird custom...stuff going on with my tcm/tcu set up.
I ask for the great wisdom of this forum. Thanks. I’m going to get some pictures soon.
 

·
Registered
1995 EZ30 Coupe
Joined
·
718 Posts
When that solenoid gets full voltage (12) it's in FWD mode, when it gets no voltage (<0.5) it's fully locked AWD.
My understanding is that there will be SOME power transferred to the rear upon acceleration, but not full AWD lock. So you would end up somewhere between 0 and 12 volts.
I could be wrong, but it sounds to me like your trans is acting normally.

Edit: I would expect the harder you accelerate, the further(farther?) the voltage will drop. Is this not the case?
 

·
Registered
Joined
·
6 Posts
When that solenoid gets full voltage (12) it's in FWD mode, when it gets no voltage (<0.5) it's fully locked AWD.
My understanding is that there will be SOME power transferred to the rear upon acceleration, but not full AWD lock. So you would end up somewhere between 0 and 12 volts.
I could be wrong, but it sounds to me like your trans is acting normally.

Edit: I would expect the harder you accelerate, the further(farther?) the voltage will drop. Is this not the case?
I would agree that my trans seems to act normally in every day use and I appreciate the response.

But the concept of this diff lock device seems to put forth that there should be full lock of the awd solenoid when implemented. I also think the general consensus in this thread is that the solenoid will lock when no voltage is applied. The FSM diagnostic section seems to state I should see less than 0.5 volts when in 1st gear and a good deal of throttle.

To your question, no, the voltage did not drop with more acceleration. I went past 4K rpms and no real change from 6.5 volts on the wire that SEEMS to be correct.

I’m just not sure why I cannot confirm the right wire to cut when it seems so straight forward in this forum and the fsm literature. I had to take my car off the jack stands but I guess I’ll just need to start testing all the wires, since the fsm test doesn’t hold up, even though the wiring diagram seems to. I think the logical conclusion is I just don’t have the right wire yet. Much appreciated. I’m open to other ideas of course. Thanks.
 

·
Registered
1995 EZ30 Coupe
Joined
·
718 Posts
Did you do these tests with the ignition on but the engine off? It sounds to me like you were running the vehicle on the jack stands, which would likely account for your variances.
 
321 - 340 of 346 Posts
Top