CNY_Dave

There is typically one or more sense wires, maybe a sense-ground wire, and the shield wire. The shield wire will have its own pin (or maybe more than one) all the way back to the ECU. Any good mechanical conductive connection to the shield and another wire/piece of shield will work. It is OK if there is a short gap in the shielding.

Supporting the weight at the front center is a useful trick. It should look like it is just about to scrape the back of the bellhousing when seated, but measuring as suggested is best.

If you are lucky a piece of steel or brass line could be pressed into the hole, with or without redrilling the hole.

The sum of the inward force from the hammer plus the turning effort from the ramp inside the impact driver makes a huge difference. You just have to remember to put some 'turn' force into it then smack it with the hammer.

Heck, I've had to repair groundhog damage under my hood. Man those things can get mean when the dog chases 'em under the car...

I have not done the VTD swap, got laid off late January, started working a temp contracting engineering job in April, start my permanent job this coming Monday, and I sent my rear VTD housing to someone that had cracked theirs. It is guaranteed to work, the gears don't need ay input from the TCU, known issue is that the trans will throw a code because of a difference in the signal from the rear speed sensor. The VTD uses a small metal 'gear' to send the signal, the MPT uses the outside of the clutch basket, which has more 'teeth'. If the MPT front sensor has the same number of teeth on its 'sensor gear' as the rear on the MPT, I've wondered if you could just wire the front sensor to the front ands rear inputs. A daring soul could try getting a new rear sensor gear machined, or even boring a hole in the case and moving the rear VSS so it can getr a signal from the same-dimaeter fewer-plates sits-further-back VTD clutch basket. Some have said it may be possible to squelch the code with one of the ecu/tcu reader/programmer tools, like FreeSSM.

I think I recall an emissions equip problem that would fill the charcoal cansiter with gas when filling/driving, which would then cause a too-rich condition. Sound familiar to anyone?

Sounds like a foam carb float... soaks up gas and sinks and the engine floods. I wonder if the material in your float absorbs moisture? Maybe the float is finely density-tuned to sink when the brake fluid specific gravity is wrong?

"Manual" switch?

Cool beans. Our Forester is just a little 'tight' on parking-lot turns, maybe I'll try this if it starts to get worse.

If you can feel it but not see it- OK. If you can see it- it's getting sketchy.

Relatively few bucks for 2 pieces of steel line, some $$ for a bunch of hours on my back, and a swear-jar bill for $40,000 sounds about right.

I just did my '03 OBW rear brake lines, one let loose a few weeks ago.

There is usually a hole in the inhibitor switch where with the shifter in N (P?) the hole lines up with another hole and you can put a pin in. If the holes don't line up, it's misadjusted. If it has the holes.

You'll know if an H6 oil pan gasket leaks, in a big way, ours are far below the oil line! Makes me nervous.

Agree with the old-school vacuum gauge, systems are much more complex nowadays but it's still just a big air pump!

During my experiments, I recall you can go as high as 100 or several hundred ohms, maybe even 1kohm. Higher resistance limits the power dissipation, I had my dumping resistor inside the box with my pulse-inversion circuitry.

I've never come across what you describe, just the fuse which grounds a pin which tells the tcu to release the AWD. If the ATF Temp light isn't flashing, it haSN'T latched into a fault state.

And as far as I can tell, dumping to the resistor is optional. There is some discussion out there that you will kill the TCU if that connection isn't wired to *something* but it's just a solenoid driver, and also as soon as the AT Temnp light starts flashing the TCU stops trying to drive the solenoid. You do need to shut the key off to get it back to normal mode after disconnecting the solenoid. In fact, you could probably use a momentary-off switch in the circuit, push the button long enough to set the fault, then release and it probably won't try to use the solenoid again (unless it defaults to 'front wheel drive')..

It's pretty simple, a switch disconnects the trans from the TCU, and switches the TCU to a resistor to keep the TCU from thinkiing something has gone wrong. I temporarily had to use a switch ('04 trans AWD solenoid uses the opposite signal from all before that, and of course I put an '04 trans in my '03), and I can say if your AWD is working properly, there will be very little gain to adding the switch. And yes, locked on a hard surface would be bad, bad, bad. One real negative of locking it is down a slick hill, the rear wheels will slide because of the braking forces transmitted to them via the front wheels. Found that out going down a slick snowy hill...

Stainless steel ty-wraps are a good bet if you are looking to buy something... I just use plain tie-wire, it l;asts a surprisingly long time. (and it's C H E A P )

Repeated failure has to be one or more of: - cheap bearings - bearings that need to be greased before use - damaged by the install process (sloppy press usage, weight on bearing before torquing, etc) - hub worn/damaged - knuckle out-of-round BTW bearings from subaru look like they have almost no grease, and cheap stuff at that- I installed a front as-is from subaru many miles ago, no problems.

:popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn: :popcorn:

Needs an H6...

So the valve isn't 'tall' compared to what the shims are made to accomodate, that's good news. The bucket is seated properly and all that?