CNY_Dave

I thought the whole point of having the shoulder bolt was that the shoulder was supposed to bottom... or aren't those shoulder bolts? My front flange completely disintegrated, not even enough for a split flange to grab, so on kine I just have strings going forwards from the rear flange bolts to the body.

I've found on similar setups you can get the nut started w/o having to sqeeze the springs.

Bump. Still can't view pics on opposedforces...

I have had bad driveshafts or bad inner cv joints have a wicked vibe that would come and go.

I often go a smidge over on the oil, I put in one quart and the castrol 5-quarts-plus-a-bit jug.

How well does an H6 drain? Or how much? I think I got only a gallon from the radiator drain, with the front up on jackstands.

I am of the ilk that think the swelling o-ring breaks the cap, so I would say you dodged this bullet...

You'd think the brake lines would be stainless. And the coolant pipoes. And the power steering pipes. And the trans cooler pipes. All of 'em are ticking time-bombs on a NY car.

Holly carp. Place of honor on the mantle for the hammer!

There are pics on the web of the AWD section, I'm not familiar with the '96. My '03 has a spring that holds back the parking pawl.

Yes, removal of power to the solenoid or failure of the solenoid and the clutchpack defaults to no lockup. I am not sure the filter is a 100% sign of what type, as the H6 4EATs have a remote filter and a cover where the filter usually goes.

If the rear output shaft stayed with the trans, just pull it off and you'll see the clutchpack, it stays with the output shaft.

Gen 2 is 2000-2004, yes? In 2004 the transfer duty solenoid signal was reversed, so right off the bat the AWD won;t work. I would be surprised if a gen3 auto trans would work in a gen2, usually just tcu compatability doesn't span an entire generation. For example, an '04 TCU is very different from an '03, and you can't swap it because the ECU is also different, and to swap the TCU and the ECU would mean replacing a hatfull of other stuff. Somewhere in there, gen3?, the 4eat was directly controlled by the computer instead of being a mix of TCU brains and valve-body mechanical/hydraulic brains. But, I'm far from an expert, oddities abound in the subaru world.

If there was no clutch pack, what was there? This is an automatic? was there a rear driveshaft?

TCU on my left-drive is a metal box that sits above and left of the brake pedal. Two connectors plug into it, right next to each other, one white and one grey on mine.

To get past this limitation, to a degree sufficient for steep slick hills, I put it in 1 or 2 and apply a small amount of throttle with the brakes on, as soon as the TCU sees off-idle throttle in 1 and 2 it applies a fair bit of 'lock' to the clutches, and there is absolutely no front-only wheelspin.

I have done a lot of snow driving, and briefly this winter I had a switch instead of the TCU controlling the transfer solenoid. I have never been in a situation in the snow, on road or in my yard, where the switch would have done a lick of good. Off-road, mud and rocks, where you can transition quickly from traction to none and back, I think a switch might be useful. I also learned that turning off the AWD when going down slick hill can be a good thing. I had my switch locked, and a snowy downhill with careful braking resulted in the rear end sliding out, because braking force from the front tires was also being transferred to the rear tires. With the TCU in control, throttle off = minimal AWD lockup.

Since the signal from the TCU to the FWD fuse is from the TCUs microcontroller, at some point it could have gone to 5V on that pin instead of 12V, or it could indicate a pretty severe current drain in the TCU. The TCU is easy to pull, pop the top off and take a look at. Some are repairable, a bad track can be replaced by wire, a bad driver can be replaced with one scavenged from another TCU, my '03 has several drivers of each type. Also, I found a chip that IO used to build my inverting one that could be used to replace one of the drivers, but you'd have to finagle a way to keep the 'sense' line happy, the driver chip incorporates a line that the microcontroller looks at to make sure the solenoid line isn't shorted or open.

The FWD fuse grounds a signal from the TCU, signalling the TCU to please apply +12V to the transfer solenoid (duty C solenoid). On my '03 if the trans light is flashing the TCU ignores the FWD fuse, and if the fault clears while driving the FWD fuse does nothing until the car is shut off and re-started. I do know the valve the duty C solenoid operates (the solenoid is actually a valve itself) depends on very clean fluid, as there are some pinhole-sized feed/bleed ports. The valve has it's own filter. If random bad things happen, could be the filter is missing or the valve has crud in it that will never flush out because the holes are so small.

Unless he's in the 2000+ year range, then it's just slide-back the axle and undo 4 bolts.

Any engine works most efficiently at about 80% of full load. Any throttled engine works more efficiently at wider throttle openings due to losses in 'pulling vacuum'. These are fundamentals that every IC engine engineer and scientist understand. Matching the engine's efficiency (turning fuel into power) and the vehicle's efficiency (MPG, what you care about) is the art to that science.

I never get under the jack, some rules do not get bent! But, I am almost always alone and the phone out of reach when working on the car.

I remember a research car built with no throttle, it had an insane constantly variable transmission (connected to the gas pedal) that could gear the car up enough to produce any desired slow speed.

Heat and vapor mist are a normal part of operation, just a tiny bit but over long time frames a little builds up.

If it's just a little bit of dark dust/crunge stuck in place, that's just normal hot oil mist/vapors from the vent condensing on the outside of the housing.