jarl

I would expect the '98 OBW to be exactly the same as the '98 Legacy wagon. The changes are mostly in the suspension and spacers, aren't they?

How???

Ok... I'ver read your original post again, and I think we all are reading something that is not explicitly said. At least in my case, I think I've made my mind after reading the title ("battery grounded on engine"). What I realized is the following: had the battery really have grounded on the engine, the battery cables involved would have melted, and the battery would be no more. The fact you mention "it was black in a certain area" means it may have probably arced somehow at some point, but it doesn't necessarily mean that was your problem. In the same tune, I doubt any computer has been compromised. If you saw a cable glowing inside the car, I think *THAT* was -and probably continues being- your problem. Let's stress this out: IF YOU HAVE NOT CORRECTED THAT, YOUR CAR MAY BURN TO THE GROUND ANY MINUTE . I'm not saying you have not corrected it, but given the possible consequences... Now: do you know which cable was the one behaving like a bulb? Anything inside the car that is connected to the (+) side of the battery most probably runs through a fuse, and I doubt any cable would start glowing before a fuse is blown (unless it's a fusible link, but I don't think there's any in there). Can you take a picture of the cable? A (+) cable arcing to ground would probably lower the voltage on the ECU enough to kill the car (which will in turn light a few lights on the dashboard)

From the link posted before: "EDIT: Just had another thought... Try disconnecting the two fuel-tank sensors, and apply a known resistance of more than 80 ohms (say 100 ohms) across the wiring terminals (and wait 10-minutes too). This might fool the combination-meter into thinking that it should illuminate the low-fuel lamp" Shouldn't a resistor of infinity ohms work as well? If so, just disconnect the connector for the sensor and run the car for 10 minutes Edit: never mind... the FSM for the '99 OBW ("wiring diagram", Pg. 47) shows a switch connected to terminal a9 of the dashboard. Pg. 73 of the same manual shows something different, but still according to the diagram the light should turn on if you connect terminal a9 of the dashboard to ground (I would use a fuse or aresistor just in case). A dumb design if you ask me (it's not fail safe), but...

First of all: if you appreciate your friendship, stop using your friend's shop. Nothing good will result from that. I've seen in this forum reports of ECUs going berzek after being without power for a few seconds, but working correctly after disconnecting the battery for a couple of hours. Hopefully that's your problem. Now... do I understand correctly that when you looked down the hand brake mounting area you saw an overheated cable? If the cable that rubbed against the engine was the one from the battery to the starter, it has drawn hundreds of amps (have you ever melted a wrench between the battery + and ground?). I'm surprised the battery is still working, but at least in theory the ECU / transmission control module should have been spared (the current would flow through the shorter path, which excludes your electronics) Finally, if the positive cable took that amount of amps, the ground points took those as well. Did you check the grounding cables to make sure nothing is fried there?

Timing kit from ebay (including water pump), OEM stuff from online dealer (subaruonlineparts worked for me). There are several threads that include parts numbers and all relevant information

Ditto

19mm

Did you check the FSM? It has the procedure (albeit using "special tools") and all the torques. PD: about the video... man, how I hate when someone can't talk without shouting... Edit: I found this... the thread at NASIOC is missing the images: http://www.scribd.com/doc/43786961/Wheel-Bearing-Replacement-Guru-s-Guide-W-pics-NASIOC

Sorry for the stupid question, but I thought the OEM sensors were Denso. Weren't they?

I would be interested in a description of the (electrical) tests as well, but the link I posted has an excellent description of how to *really* test both sensors and the converter. You need an OBD2 logger, though (the poster on that thread bought a Tactrix cable, I'm checking to see if I can do something similar with my ebay OBD2 thingie) Check that thread, or if you don't have much time check post #65.

There's an excellent thread at NASIOC about a guy diagnosing his P0420 with the help of a Subaru technician: http://forums.nasioc.com/forums/showthread.php?t=2210139&highlight=oxygen&page=3 I don't know if you have the tools to diagnose this, but even taking the car to a dealer to have it diagnosed sounds like a sensible alternative...

That's the weird part: the code is for a "bad O2 sensor circuit", not for a bad O2 sensor. From what I have found, this means the sensor's heater is not working as it's supposed to, but the O2 sensor is producing a "legal" output. This seems to be backed by the fact the gas mileage is not particularly bad -I think-. I'm more concerned with the catalytic converter melting or something like that... Now back to our normal programming: white smoke

I have noticed a little white cloud behind my car, on which I recently installed a 2.2 with ~90K miles. I blamed it on the cold weather, but while waiting at a red light this week I realized most other cars were not producing anything similar Today (very cold) I asked my wife to start the car while I played close attention to the exhaust. When she started the engine, there wasn't any significant amount of white smoke. Then I asked her to bring the engine to 3K RPM, and then I saw the smoke coming out. It didn't smell like coolant at all (no sweet smell or anything like that), but rather like gas... lots of it. The EJ22 came from a car that was totaled several years ago so I don't know much about it's history. My question is: can the excess fuel create this white cloud somehow, or should I assume a head gasket is blown? How can I diagnose this? BTW: the check engine light is on, with a code about the oxygen sensor circuit being bad, so I assume this is the cause of the excess fuel. Is it a fair assumption?

Mods: This is a duplicate posting (1st one was 10 minutes earlier). Is it possible to combine the two?

If your car has the stubby antenna you may have an issue with the antenna preamp (I think you car has one). That would translate into poor reception, so the tuner would work only as you get close to the radio stations.

Uhm... I don't see any mention of a coolant hose that goes to the coolant "bridge". Also, for any reader with a newer AT, there's the EGR pipe as well.

I think you can buy the equivalent to a FSM in electronic form directly from Subaru.

Well... any "symptoms" you may describe? Do you think the hatch is unlocked? On my OBW the link between the hatch handle and the lock had rusted out, but the place where I bought it from had removed the internal cover somehow. Nothing was broken, so I assume it should be possible to do something similar in your case.

Yup... and THEN twist

Uhmm... this is a 4+ y.o. thread... not sure you'll get much. In any case, I would expect torsion bars to fail in torsion. From the pictures on this thread, not a single one has failed in torsion (it's a very easy to recognize failure). I guess the manufacturing process left some residual stresses at that particular point or something like that. Very interesting...

If you are going to be working on the engine you should consider an engine stand (~$35 at harbor freight). If not, I have seen people using tires to avoid damaging the oil pan.

From what I have read here, the problem has to do with corrosion "welding" the tip of the torque converter shaft to the hole where it goes (end of the crankshaft). On my car it wasn't an issue, but the engine had been pulled 50K miles ago or so. If in doubt, you may try to break the two parts loose (after removing the 4 bolts to the flex plate). If you can turn the engine without turning the TC you should be OK.

Crap... I'll have to wait six months before recommending it again The thing is the Turtle Wax kit is the cheapest one to include a sealant. The 3M was somewhere around $12 or so and didn't include a sealant. I think 3M sells a sealant separately from any kit. Maybe that's the best option