CNY_Dave

For the fronts, you still have to disconnect enough so you can pull the axle out of the hub. I hear you can do this by disconnecting the strut from the knuckle, haven't done it that way myself yet. After you get the axle out, you need the 'cheap' harbor freight front-axle service tool or the expensive OTC hub-tamer (both screw-type presses) to press out the old bearing and press in the new bearing, and to press the hub out/in of the old/new bearing. I have heard either of those will work, when I did mine with a 20T press it took darn well near all 20 tons to press out the bearing. Not sure what the on-car screw tool can develop by way of force. Dave

This phenomenon has been studied and duplicated, it doesn't affect all of that year it seems. The rear end steps out without warning, with people having to crawl along on a slippery straight road getting passed by virtually every other type of vehicle. There is definitely an alignment component to it, a power transfer component would explain some aspects of the problem nicely. Dave

Make sure you add this to the end of the lengthy ghostwalking thread over on the other really really good subaru forum, subaruoutback . org Dave

Not sure what you mean by CV vs Axles... But what you are describing is the classic bad outer CV joint. Dave

So the VDC system was getting all freaked out, it thought you were sliding all over the place.

Any time the car is jacked up, the car regards it as an excellent time to shift, break, crack, or drop some part or another.

If the rim-rotor mounting surface is crusty (or rotor-hub) you can get a balance-like vibe that seems to come and go. Dave

Oxy-acet setups are pretty cheap, and small tanks are often purchased instead of rented. Then, get an air-acet sweating torch, and plumbing work becomes much easier. Won't overheat the joints like oxy-acet but much much much more heat than propane or mapp. Dave

The H6 does seem to normally turn over quite 'freely', it's those li'l' tiny 500CC cylinders and the extra two compression strokes spreading out the loading up of the starter. The fuel pump is a 'popular' problem on these critters, sometimes they just pee the gas right back into the tank. Maybe try turning the key on-off-on-off a bunch of times, then trying to start it- if it starts for a sec or coughs it is a good indicator it's not getting fuel. Of course a fuel pressure gauge is better. You could always disconnect the fuel line to/from the fuel pump, one person on the key and another with a coffee can. Also, don't forget to first try cranking it with the throttle held full open in case it's a flooded condition. If it is getting gas after a lot of crank8ing you should be abl;e to smell some gas at the exhaust pipe.

Yeah, hot is definitely an indicator it's time to limp home.

2 ways this could go wrong- 1) Tube gets sucked into the gear teeth and is pulled in rapidly enough or is tough enough the force is sufficient to damage a bearing or crack a housing or bearing boss 2) Ground up bits of plastic plug a passage starving a bearing or gear Most splash-only systems do have some passages where oil flows to vital bits after the oil has been flung to upper parts of the gearbox. Don't know if the front diff has any. If you can find a pic of a diff opened up you might be able to see if there's really any chance of the tube getting sucked into the gears.

Bad things can happen, but this really can't happen with this car. It can get very loose and wobble and pretty much chew the axle right into the bearing races and the housing, expensive and possible loss of control, so something to be avoided, but as you grind to a halt the wheel will still be attached.

If the wheel bearing is noisy but tight the chances for damaging the hub or the housing are low, in my opinion. If you drive on it and it gets loose and you don't notice, those other parts might take some damage. If there is any damage the hub will be the item damaged, most likely, you would have to do some really extreme stuff to damage the housing. If whoever replaced the CV joint didn't properly torque the axle nut, hammered the crap out of the axle to get it out, or allowed weight onto the bearing before the nut was torqued, the CV job could have caused the bearing to fail. My '03 LL bean left-front wheel bearing failed at 75k? 80K? 100K? I forget (thread is here somewhere) and I drove on it a looooong time with it making noise. It never got loose, never got hot, hub and housing were fine. Of course if it got loose I would probably notice and so I would drive on it longer than I'd be willing to let my wife. Whoever does the job should have the hub in hand in case it's needed. If there are any signs the inner race has been spinning on the hub, the hub must be replaced. I can't help much on the prices, did all the work myself. On mine it was indeed difficult to isolate the bad bearing, had to spin each wheel with a mechanics stethoscope on the strut spring. I'm now near 180k and it needs the other front or a rear, haven't isolated yet. Dave

OK, now that's like 3 times someone has started a thread on a topic that has recently been on Car Talk...

If it'll take the force... but it sometimes takes a lot of force. You could borrow a ball-joint or other press from autozone/advance/pep boys.

A bit of scuffing and a layer of even paper between layers of JB and it probably would have lasted forever!

I believe that what you are saying is correct but does not relate to the actual frequency of the noise. The way bearings make noise (wavy bearing surface or pitted bearing surface) it is much more like what I describe, given the bad diffs and noisy carrier bearings I have lived through. A wheel might go from 0 to 1440 RPM and the driveshaft would go from 0 to 6336 rpm, sweeping a much larger frequency range. Dave

Wheel bearings tend to more hum/moan than whine. Wheel bearing noise tone changes with the wheel rotation speed, at the same rate. 2x the speed = 2x the pitch Driveshafts and pumpkins have their tone change a lot more- a bit over 4x the change in speed as the shaft is turning 4.44 times faster than the wheels/axles. 2x the speed = 8x the pitch

If it's on-line, post it up!

Let me add I'm not trying to be a wongleflute (though I am gifted that way) but you are approaching this from a very specific angle that was taught and may have certain details, and I am coming at this knowing what is going on at the level of what principles are involved. I don't need to be a jet engine mechanic to know that the spinny bits up front compress the air, the air is heated by the compression, fuel is sprayed in,the fuel burns (O2 is consumed), and the hot gasses turn the spinny bits in the back of the engine, and those spinny bits are the ones that spin the front spinny bits. Likewise, I don't need to know exactly how the AF ratio may (or may not) fluctuate in order to make the converter operate optimally to know that the in converter O2 is consumed, and that an O2 sensor can only do one thing- measure the level of O2 (the O2 concentration).

Where did I say it fluctuates? I never did. And, the 'converter' most certainly does consume oxygen and certain compounds in the exhaust, that's its whole purpose. If you want to be pedantic and say it's not the converter that consumes the oxygen, but that the oxygen is consumed during a reaction inside the converter, that would indeed be more accurate, but needlessly so for this level of conversation. And, the second O2 sensor is most certainly there to measure how much oxygen was consumed in the cat (done by comparing the 2nd sensor's signal with the 1sts). If the actual catalyzation optimization mechanism is to have the mixture altered up and down a bit (utilizing the oxygen storage capability of the catalyst) and the second sensor sees a constant reading, that doesn't change the fact that it's monitoring the O2 level in the exhaust- it's comparing the O2 level against the 1st sensor. 1st sensor, it varies, 2nd sensor, it's constant. 3-way cat operation: 2NOx → xO2 + N2 2CO + O2 → 2CO2 Oxygen consumed (reacted) CxH2x+2 + [(3x+1)/2]O2 → xCO2 + (x+1)H2O Again, oxygen consumed. Dave

I think you are the one with a misconception about the way the system works, although I agree that this method is only a last resort in cases where it's a too-sensitive system and checks have been made for legitimate problems (or, the cat just plain doesn't work and you need to pass inspection where they don't 'sniff-test', heh). Dave

P0420 indicates that the after-converter O2 sensor is not seeing enough of a difference in reading from the before-converter O2 sensor, that the catalytic converter did not consume as much oxygen as expected, nothing more and nothing less. One of the causes could possibly be a lean/rich mixture problem (other than a small vacuum leak), but you would most likely get a code based on the failure of the computer to obtain the correct mixture. Dave

On the valves, do the H6 valves tend to loosen (wear) or tighten (valve seat recedes)? Dave

The LL bean H6 has the fancy stuff but no VDC. The specifically badged VDC model has the VDC. 142k for 5 grand? You're lucky we aren't in competition for it, I'd be hitting my savings acct. Dave