OB99W

The filter should be just to the rear of the intake manifold #3 runner, as shown in porcupine73's first pic.It's certainly there on my own '99 OBW.

I take no credit for the original picture -- it's your pic Skip, which I: 1) downloaded 2) converted to bitmap 3) "Painted" 4) cropped to reduce bandwidth 5) converted back to JPEG 6) uploaded I'm sure all the manipulation contributed to the blur, but if it got the job done...

Okay, so I'm not as good at this as porcupine73. Anyway, the text says "< T about here", if it can't be read.

From the pic, it looks like you have a vacuum hose running from the nipple on intake manifold #3 runner going directly to (I assume) the fuel pressure regulator. There's supposed to be a short piece of hose from that nipple to a T, then the hose from the T continuing to the regulator. The loose hose with the filter should be routed under the intake, and connected to the center nipple of the T. While you have the hood up, see if there's an emission control diagram pasted to it. It won't show the physical routing, but should give the hose schematic.

Great post!

Yes, its the oil cooler. Depending on where the leak is, it could be caused by the connector being loose, the O-ring being damaged, the filter mating surface on the cooler being scratched/dinged, etc.

I made that suggestion as well, over two months ago:http://www.ultimatesubaru.org/forum/showpost.php?p=672971 ...assuming that nothing else is the source of the rattle, as mentioned above.

There are some auto trans (and engine starting/running) problems that only show when it's cold. If the car was already warm, they may not be obvious or show at all. If you're seriously considering the car, try to test-drive it again from a cold start. As I said before, a significant number of bubbles in the fluid isn't a good situation. The fluid level has to be checked with the engine running. Although it's difficult to get a good reading when cold, do look for bubbles as soon as the car is started, and get an idea of the level. Drive the car and then check the fluid level when hot, car parked on a level surface. In order to get an accurate reading, pull the dipstick, wipe it, and leave it out for a minute or more so any fluid in the tube can return to the pan. Then reinsert the dipstick, pull it again, and look at both sides; the one with the lower indication is the one to go by. If the indicated level is low or even okay, but there are bubbles, the actual amount of ATF is probably low, and the fluid is somewhat aerated. Especially if it was low when cold, it may just need more fluid. Additives get depleted with usage, the fluid gets oxidized, etc., so if the car seems okay when cold and hot, it might be that a fluid change will be sufficient.

Welcome to the forum. ATF that's seen enough use will often turn brownish-orange, even if it doesn't smell "burned". Foaming of the fluid is never a good thing; it can be caused by either under-/over-filling, or perhaps due to an additive (which may also explain the strange smell). Did you check the fluid level? I can't say whether the car is otherwise worth the asking price, but I suspect the trans may have issues, even if things seemed normal during the test drive. Was the car cold when you started it, or was it warm from having been run earlier?

Also, to minimize the chance of damage to the brake line fittings, you might want to get a properly-sized tubing wrench if you don't already have one.

Or do a Google search on "bench bleeding"; there are a lot of online articles.

Sorry, but the viscous coupler (manual trans) is a sealed unit, unlike the clutch pack (auto).

Fortunately for you, '96 Subarus have a problem with the OBD-II I/M readiness monitors resetting every time the key is turned off. Therefore, as long as the CEL is off and there are no pending codes, you might pass. For later model years, monitors not being ready is in itself a reason to fail inspection.

Besides the already-mentioned valve clearance adjustments, there are other issues to consider mentioned in the article linked below. Pay particular attention to the possibility of switched fuel injector wiring -- it fits the scenario of rough idle but smooth at higher speeds and while under acceleration. http://www.endwrench.com/pdf/fuel/FtEmissionServiceS01.pdf

It may take a while after clearing before all those codes reappear, even if they're all still a problem. When the ECU goes into fail-safe (due to an active fault), it no longer senses or stores codes for additional faults. That means for more fault codes to be stored, the first fault has to be intermittent. The next time the ECU is active (no fault is occuring), it can sense additional faults. If a fault then occurs, but it's the same as before, the ECU will fail-safe and there will still be only one fault code in memory. If a different fault appears before the original one, its code will be stored in the second memory position, and the ECU will again go to fail-safe mode and sense no other faults. A similar sequence of events has to happen for a third code to be stored. That means the first two faults have to be intermittent, not presently occuring, and the "new" fault has to occur before either of the previous ones. Since you read three codes (the max that the ABS-2E ECU can store), for the reasons just explained it could take some time before all of them might be stored again. If the cause of any of the codes is repaired, it's then likely that any remaining problem will next store its code more rapidly. Therefore, to expedite the repair it's probably best to read the memory as soon after the ABS light comes on as practical, and deal with the cause of the particular code. Then, clear the memory and see what comes back. Something to keep in mind is that the sensor has to "see" all the teeth on the tone wheel. Anything that weakens or disrupts the signal can cause a fault to be sensed. That includes snow packed in the ring or corrosion damage to the sensor tip. I've found that if the sensors and rings are in good shape, the system is pretty tolerant of weather-related conditions, but marginal sensors allow fault codes to be easily generated. Because the environment the wheel sensors are in is pretty harsh, besides corrosion of the sensor tips, another common problem is that the sensor cable is prone to cracked insulation. Once that happens, water (and salt) can get in and damage the wire itself. If you find cracked insulation, the sensor may be unable to provide a proper signal. I don't know off-hand what operational differences there are between the auto and manual versions of the ABS ECU, but I wonder if the fact that you changed the trans has anything to do with the problems. When did they start occuring relative to when the trans swap was done? Sorry for the length -- I hope the above helps.

General AWD info: http://endwrench.com/images/pdfs/VarNov06EW.pdf

Don't panic yet. Check the CVs, just in case, especially if there's clicking. The fluid that can be replaced and might be helpful is ATF. As Skip has already mentioned, a complete change (flush) sometimes is the cure. A bad VLSD (assuming your car has one) could also cause certain binding symptoms, but it's not the first thing to consider.

Auto, turbo, so VLSD?

Sounds like "torque bind". Search here on that term for much more info.

Some food for thought: "Misfire" isn't the same as "failure of spark to jump the plug gap". Plug misfiring is an ignition problem, but P0304 indicates cylinder misfiring, which means that the mixture didn't properly ignite and produce expected power. While that could be caused by a bad spark, it's certainly not the only possibility. Because the ignition is wasted-spark type, given that the wires, plugs and coil were all changed, it's unlikely that any of them is the culprit especially since cylinder #3 (the one paired with #4 as far as ignition goes) isn't throwing a misfire code. Platinum plugs should be of the "double" type, because one plug of each wasted-spark pair runs with reversed polarity. Injectors can be swapped in position and a determination made if the problem switches cylinders. Unless I missed it, I haven't seen any indication that compression or leakdown testing was done. If it's checked and okay, and injection problems ruled out, it would make it that much more likely that valve adjustment is needed.

This is obviously some strange use of the word "fun" that I wasn't previously aware of. (Apologies to Douglas Adams)

Sorry, but it's highly unlikely if the HGs are bad on a phase-I EJ that the "conditioner" could withstand combustion-generated pressure. That sealant is sold by Subaru to deal with the loss of coolant on later (phase-II) engines, where the leakage is external and the pressure is only that of the cooling system.

When the HGs go bad on the Phase-I EJs, combustion byproducts wind up in the coolant. Since that's what you're apparently describing, unfortunately it appears you do have HG problems.

I've dealt with this kind of situation several times. The sticky "ick" (mainly sugar or corn syrup) may not have penetrated too far. The best solvent to disolve it is plain water, preferably warm. Even if the drink got beyond buttons, etc., it's probably best to start simply and try to remove as much of the goo as possible from the surface. Paper towels, wetted but not dripping, can loosen a surprising amount of the sugars without damaging anything. (Contact cleaners can be harmful to some plastics, and often won't disolve sugary substances.) You might park the car so the nose is uphill, making it less likely that any water will find its way further into the controls. A soft-bristled toothbrush, slightly wetted, can help get a bit deeper than paper towels. Keep changing wet towels, rinse the brush, and operate controls as you clean; absorb whats loosened with dry paper towels. When you can't see or feel any more stickiness, completely dry things and check operation; if you're lucky, that may be all that's needed.

Hmmm, that reads like a tacit admission by Subaru that even the newest head gaskets may not be up to the job. If I were you, I wouldn't accept rejection of the complaint by just one SoA rep. I'd suggest doing a Google search on "WWP-99 Service Campaign" to better understand the situation with the earlier Phase-II head gaskets. Armed with that, and the fact that they're still using "conditioner" with the new gaskets, you might be able to get Subaru to step up to the plate.