McDave

For reference, there was a '93 Loyale wagon 4wd w/79,000 miles on the Austin Craigslist this morning. It sounded like it was the original owner with all service records and the original window sticker. He was asking $2,000 and it was gone within 2 hours. My own minty '93 wagon 2wd 3at w/165k miles cost me $1,300, though with new a/c, window tint, struts & mounts, and a few other odds and ends I have about $2,200 in it now. So hang in there. The deals are out there.

I just looked on one of my invoices and they sure are! But it must be a photo before louvers?

When you bleed your new master cylinder place a piece of 2x4 under the brake pedal so it can't go all the way to the floor. That may be what messed up your old mc. PS Bleed the mc at the lines first as I described in your other brake thread. Edit: Nevermind, I see you have a pressure bleeder now.

I just thought of something else. It sounds like once this new maf acted up it stayed messed up. Clean it with your cleaner and try again. If it's still messed up, put your old one back in.

Yes, try disconnecting the tps the next time it acts up. I understand funds are tight, but you might consider springing for a proper diagnostic scan, one that can read and store the data in movie mode while it's acting up. Then if the problem isn't obvious at the moment, it can be downloaded and studied later. Our shop charges 1.5 hrs for this and includes the labor to replace something simple like an O2 sensor or maf.

Have you checked the fluid level in the diff since then? Is it now overfull?

I've also seen O2 sensors short out intermittently and cause the engine to run so bad I was sure it needed a coil. I was reminded of this yesterday when a Dodge truck came in with an occasional miss and no CEL. While test driving it with the scanner monitoring for misfires, it started missing so bad I didn't think it would make it back to the shop. Oddly no misses were showing on the scanner. Fortunately it finally set a code for a shorted O2 sensor and I checked and saw the O2 voltage zero-ed out. Problem solved! I was also reminded that misfires can set codes that aren't directly related to the problem. A misfiring Chevy Vortec V8 is prone to set a MAF code just from the disruption of airflow, when the real problem is ignition related. With that in mind, I'm now less inclined to believe that you have a MAF problem, Mike. It would still be worthwhile to try another MAF if you can get one free, but then I would start looking in other directions, like an O2 sensor shorting out. Unplug it when the car is acting up to see how it runs. Edit: It would also be worthwhile to check the maf voltages and ground while it's acting up, if possible, before ruling out the maf. Check for ground the same way you checked running voltage, by backprobing the connector.

Lead is very good for engines, that's one reason why planes still use it. It's just not so good for the environment.

Apparently there is an electric (or maybe vacuum) solenoid on the side of the carb that opens the throttle with the a/c on. Unplug the solenoid and see if it idles lower. If it does, plug it back in and adjust it's adjustment screw to about 800 rpm.

Yes, the longer pump for a/c cars. You'll probably want to change the heater hose elbow (see photo) coming off the pump too. For that you will need to unbolt the compressor bracket. The hose needed is a generic 5/8" heater hose 90 degree elbow and can be bought at any auto parts store, though you will likely need to trim the ends of the new hose to match your current one.

That's some good ingenuity there! I hope you kept that 'tool'.

Is he mailing or emailing the info? Are you in Holland or Brazil? If it's by postal mail, a week is the minimum time frame for the info to arrive to either country. If you two are emailing, it's possible the emails are being blocked by spam filters, particularly if the emails are coming from Brazil. It's also possible they are on vacation.

Rotate the tires front to rear then drive it on various road surfaces, particularly the smoothest pavement you can find. I imagine if there is still any noise it will have moved to the rear.

It might be worth the $35 to get all the info on your motor... http://techinfo.subaru.com/html/ixSearchNkSm.jsp?level=3&model_year=1996&vehicle_line=Legacy&pid=207781

See GG, it's a viable technique even in the rust belt.

I would go back to the junk yard and tell him the one you bought failed after 10 minutes and you would like to exchange it. That shouldn't be a problem if he has another one. In the meantime, if you can get the car to act up again, try disconnecting the front O2 sensor and see if it straightens out.

I saw the photo of your tranny oil pump shaft still attached to the tc as you were pulling the engine w/tc still attached. I wonder now if your differential suddenly has two quarts of atf in it? Just a thought.

There are many O2 sensor codes that can mean many different things could be wrong, many that have nothing to do with the O2 sensors themselves. Getting a Subaru sensor is a good idea, but it may not solve the problem. If/when the code sets again please give us the code number so we can give you a better idea where to look next.

You're welcome, but it wasn't me.

What code are you getting, and is it the same code as before?

We know, but it's something that has to come apart when you replace the rack. I'm just trying to save you a few bucks if your tie rod ends are still in reusable condition.

If you have rust inside that joint you need a new one anyway. For the rest of the nation that doesn't salt their roads or drive on the beach, the method I described works great and is worth a try first if the tie rod end is still in good reusable shape. If it doesn't come off with a few solid whacks, then it's time to order new ends and hunt down a pickle fork (preferably with a pneumatic impact behind it).

Removing tie rod ends from the steering knuckle is actually very easy. Turn your steering wheel all the way one direction so the tie rod is out towards the fender. Remove the cotter pin and nut, then with a large hammer whack the side of the knuckle by the tie rod end. There is even a flat on the knuckle where you should aim for. After two or three good whacks the tie rod end will just fall off the knuckle without doing any damage to anything. Repeat for the other side. Of course if the ball and socket joint in the tie rod end is worn out, now would be the time to change it too. The reason this works is because the tie rod end is a tapered fit in a round hole in the knuckle. The hammer blow on the side of the knuckle is enough to jar it loose. Hitting down on the threaded end of the tie rod end just damages the threads and likely won't break it loose. And like Gary said, using a pickle fork will work, but it destroys the boot and ball and socket of the tie rod end. Note: This also works on ball joints that have a tapered end going into the knuckle/spindle. Just leave the nut on for a few threads so spring/strut pressure doesn't cause everything to fly apart when it breaks loose.

^ What he said. The spec is 145ft lbs and up to an additional 30 degrees to get a notch to line up for the cotter pin. When I did mine last weekend I used every bit of that extra 30 degrees, but no more.

Doh! Why didn't I think of that? How will you be able to tell when it sets the misfire codes if the cel is already on?