Once I found this out, I want to know more so I jumped over here and found these threads talking about it:
To sum up all the previous posts... In order to achive "true" AWD (meaning no delay/tire slip before AWD is active) you can disable the duty C solenoid (which is in charge of the MPT clutch on the 4EAT) and essential lock the MPT clutch to a 50/50 split ratio. Now when you just put a switch in to disable the duty C solenoid, the TCU will throw a fit and cause the AT Temp light to blink something like 16 times. In order to prevent the TCU from throwing out codes you can build a resistor block to fool the TCU into thinking that the solenoid is still there by providing the same resistance the solenoid provides so you don't get the blinking light and the TCU throwing out codes.
Here's a diagram on the whole setup thanks to obsolete over at the rs25 forums for the diagram. It can also be found on page 5 in the first link. I take no credit for this:
resistorblock.gif 8.24K 64 downloads
The biggest downside to this is you might blow your 4EAT up because you are causing too much hydraulic pressure to build up when you disconnect duty C solenoid because the internal spool valve is also controlled by the duty C solenoid and when internal spool valve closes it cause pressure in the clutchplates to build up and blow up the tranny. I am summarizing very quickly here, but that is the jist of the all the threads.
So with all that in mind this is what I am thinking. I want to do all of this, but add another circuit in that would give a set amount of voltage to the duty C solenoid, say like 55% duty cycle (~55/45 split). This way the solenoid is not closed all the way, allowing the internal spool valve to remain partly open and not allowing pressure to build, solving most of the issues, but giving you instant 55/45 split allowing for better snow/mud/crap starts. I would just use a DPDT switch (or similar, not exactly sure what switch to use) and instead of connecting an LED up (maybe work on that later??) I would wire my new voltage limiting circuit to it instead. That way when I turn on the "diff lock" it would bypass the TCU, and activate my voltage limiting circuit at the same time
I don't want this to turn into a discussion on whether this is a good idea or not (the mod itself), I am going to try it, and if all else fails I am going to do the mod following the instructions in the first post and take the risk. I want this to be a discussion on whether or not this idea will work...
So I have 4 questions:
1. Am I on the right track here? The TCU sends a different voltage to the solenoid to control the split depending on conditions, so I should be able to do the same...
2. Does this idea solve the issues of building up pressure and the clutch plates failing?
3. Do I need a PWM controller circuit to reduce the voltage(PWM) to the duty C solenoid (much harder)? Or can I just use resistors to lower the voltage(non-PWM) to the duty C solenoid using the 12v from, say, the cigarette lighter (much easier)?
4. I am going to be doing this on my 1990 loyale turbo wagon w/4EAT, so does anybody know which wire goes from the TCU to the duty C soleniod on my car? I have the FSM, but it doesn't have anything for the 4EAT, and I can't find any wiring diagrams that I can read for my model (ae82), all I find is legacy and impreza stuff.
I am going to be consulting with a good friend of mine who is an EE, so he should be able to help me with all the heavy lifting (soldering, math, etc) to try to make this work. Also, if I do succeed I will post diagrams and pictures with part numbers and all that jazz. This mod is awesome, but having the added benefit of not completely locking the MPT clutch would make it perfect, but I only hope ... More to come!
Edit: I should add that I am not an EE in any way, shape or form. There is probably some technical stuff missing in here, but the concept is what's important
Edited by eulogious, 28 October 2009 - 01:20 PM.