OB99W

Or, it could be determined less directly by logical process of elimination; disconnect the brake light switch and see what happens. If it's still rough (in the situation I described above), the likely culprit is determined (if not due to alternator load then vacuum leak). Maybe even unplug both the booster and the brake light switch (one at a time), since each could be contributing. Too bad I'd rather sit here and type than actually work on the car.

Perhaps, but perhaps not; a properly functioning trans torque converter shouldn't cause that to happen. My '99 OB has what I'd describe as a slightly rough/low idle (once the engine has warmed) when in drive and with my foot on the brake pedal. However, if I pull up on the emergency-parking brake and remove my foot from the service brake pedal, the idle improves substantially. In my case at least, I attribute the difference to two possible factors. One is that there may be a slight vacuum leak in the brake booster, and the other is that the brake lamps put an additional load on the electrical system that the alternator (and hence the engine) has to make up for. The "problem" isn't so bad that I've been motivated to determine the exact cause .

A few miles (enough to get the ATF up to operating temperature for a while) should be enough. If you haven't already seen it, take a look at this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=69474 My post in it (#9) gives an indication of how much of the old fluid gets changed if you drain and refill only the pan a few times.

Really; some people get to tell their car/deer encounter stories from a hospital bed, or not at all ! I have a love/hate relationship with my Subaru, but many of them do seem to protect the driver and passenger(s) in accidents, even if they get sacrificed in the process. I live in deer country; it's rare that a day goes by that I don't see at least a half-dozen or so right on my property. Most of the roads near me have posted speed limits of 40-55 MPH, some with fairly dense woods almost up to the road, providing little visibility for the deer or the driver. In the areas the deer habituate I've been known to drive below the speed limit, especially at dusk or after dark during rutting season, whether or not it annoys the drivers behind me. I've had a couple of deer near-misses (and no hits), and I attribute that to having a few extra feet to brake due to the lower speed. Obviously, that won't always save you, but so far...

Good link for general ABS info, Nipper. I've already mentioned the problem, but it bears repeating; as the article says: "...often times that the bolt that holds the sensor in is so rusty you can’t get that darn thing out without it becoming a production."

At least some of the Subaru gaps seem to average closer to 1mm. Attached is info about the 99 Outback; it uses the 5.3i ABS.

As I mentioned above, the tip normally projects about 3mm (beyond the cylindrical portion of the sensor pole piece). You're finding a 4-5mm gap; lets call it 4.5mm. Subtracting 3mm from 4.5mm leaves 1.5mm, which is very slightly larger than what the gap should be. The sensor tip is definitely missing. Code 56 is "Abnormal G Sensor Output". It's possible that the codes other than 22 and 24 (51 previously, 56 now) are byproducts of the malfunction due to the damaged wheel sensors. I can't be sure, but there's a fair chance that replacing the two damaged front wheel sensors will complete the repair.

Perhaps a point of interest -- Based on the full capacity of the 4EAT and typical drainage from the pan by just removing the plug (with the car level), I roughly calculated the percentage of old ATF replaced with new by multiple drain/fills. The following is in the form #-of-changes -- %-fresh-ATF: 1 -- 40% 2 -- 64% 3 -- 78% 4 -- 87% 5 -- 92% Naturally, the advantage continues to diminish as the number of changes increases. It seems obvious that if the original fluid isn't too bad, a few changes in succession might easily be sufficient. However, if the fluid hasn't been changed in 100k miles, it's probably prudent to just have a shop do a complete flush with new fluid. The '99 came with a Phase2 4EAT, using the external spin-on filter.

According to the AutoZone website: Component Locations - Flasher On your 1991 Subaru Loyale, the Component Locations - Flasher is: Front seating area, driver side, under dash, above brake pedal, mounted in steering column support bracket

If this ever happens again, and you suspect flooding, hold the gas pedal to the floor and crank. It should "unload", just like in the old carb days.

It's not just a matter of comfort. The A/C helps with defogging when the heater is in defrost mode. Whether that's important or not depends on the climate, in particular car interior moisture versus outside temperature. If fogging is a potential problem, then it makes sense to get the A/C seen to.

I don't have any info on code "46". However, since you got "42", I'd suggest checking the charging system before you do anything else, since low voltage can cause other codes to show. If that seems okay, look for possible loose/corroded connectors next.

How are the heater controls set? As already suggested, the A/C may operate during defrost usage; when cycling normally, it may not be very noticeable. However, if the refrigerant is low, the system may try to come on, sense the problem, and then immediately shut off. I suspect the cycling you're describing is caused by that, and the system needs recharging. If you find that defrost works well enough for you without moisture removal, the A/C could possibly be disabled rather than recharging it.

By the way, sometimes multiple ABS trouble codes can be caused by a charging system malfunction. Checking the battery voltage while the engine is running may be a clue to that possibility. EDIT: Of course, an excessive gap from sensor to tone wheel can cause insufficient voltage level for proper ABS functioning, and I just looked at the photos you linked to in your post #4 of this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=69239 It seems that the tip of the wheel speed sensor is missing (broken or corroded off). It's a flattened portion about 1.5mm wide and 3mm deep, and the sensor won't function without it. You'll need to replace the sensor; unfortunately, new cost even from the online discounters is typically over $100. Make sure you use plenty of rust penetrant before trying to remove the sensor mounting bolt, and a 6-point socket. If you're unlucky (don't ask how I know ), the bolt will break and you'll have to carefully drill out the remains to spare the threads.

They're diagnostic connectors; don't connect them unless you're "looking for trouble", so to speak.

Check for HG leaks, CV boot leaks, fuel tank filler pipe rust, torque bind due to viscous coupling failure, for some. The 2.5 is an interference design. If a seller claims the timing belt was changed, get proof, or to be safe you'll probably be replacing it.

Adding to my previous post: You might want to verify that all the codes are "real". I know that the weather in Washington has been bad lately, and perhaps wetting of some of the ABS system components has caused certain problems. It could help to temporarily clear the codes, and read them again when the ABS light comes on. Of course, the fuse will have to be reinserted. To clear the codes, connect the ground probe to pin 6 of the B82 diagnostic connector as you did previously to read the codes, and turn the ignition switch to "ON" (don't start the engine). The stored codes will again flash. Pull the probe from pin 6 and quickly reinsert it three times (apparently you have to complete that cycle within 12 seconds or so), leaving it in pin 6 the last time. If things are normal, you should hear the ABS unit go through a mechanical sequence (expected when codes are cleared). After turning the ignition to "OFF" and then "ON" once more (again, not starting the engine), the ABS lamp should flash only "11" if the trouble codes cleared. As soon as a problem recurs and the ABS light comes on after a drive, read the codes and let us know what you get.

I've got the 5.3i ABS on my '99. The manual says: 11 - Start code - Trouble code is shown after start code 22 - Abnormal ABS sensor signal - Front right ABS sensor 24 - Abnormal ABS sensor signal - Front left ABS sensor 51 - Abnormal valve relay The first code, 11, just tells you it's the beginning of the code sequence.

http://endwrench.com/images/pdfs/2.2SingleOverWin01.pdf http://endwrench.com/images/pdfs/2.2Liter.pdf http://endwrench.com/images/pdfs/TBeltEWWin05.pdf Motor Magazine series; starts with 2.2, then moves on to 2.5: http://www.motor.com/MAGAZINE/Pdf/072001_08.pdf http://www.motor.com/MAGAZINE/Pdf/082001_08.pdf http://www.motor.com/MAGAZINE/Pdf/092001_08.pdf http://www.motor.com/MAGAZINE/Pdf/102001_08.pdf http://www.motor.com/MAGAZINE/Pdf/112001_08.pdf

The spin-on filters, found on the phase2 4EATs, are externally mounted.

Nah, there's at least one other (First Edition, at that). Actually, I've read it's estimated that about 50 million computers are still running a legal version of Win 98. However, I did recently download Firefox 2.0, since IE6 has been giving me some grief. My philosophy is to keep things going until there either are no parts available, the economics make no sense, or they just don't do the job any more.

Exactly what do you mean by "the car temp sensor seems to warm up pretty fast ..."; are you saying that the temp gauge indication comes up quickly? A defective temp sensor can force the ECU to remain in open-loop mode. That will cause a continual rich idle and low-speed mixture that could result in poor gas mileage and high idle HC and CO. Did you use an OEM O2 sensor? Has anyone verified that it is responding correctly? A scan tool should probably be used to verify closed-loop operation after warmup and that all sensors are performing correctly. There are several possible causes of high idle HC/CO beyond the ones already mentioned. Something that might help point in the right direction is idle quality. Is the idle smooth or rough; when warmed up, does the exhaust pressure seem constant, or can you detect misses?

Nicely centered gear contact patches. So, does Flatuous Blather = All_Talk ? (For that matter, does "flatuous" = "fatuous"?) Happy New Year to you, too.

Actually, Dexron VI has been out for a while, and at least GM claims it's backward compatible with older hardware. If I were upgrading ATF, I'd want to do a full flush rather than trust that the fluids are completely compatible. See:http://www.imakenews.com/lng/e_article000384801.cfm?x=b11,0,w www.gm.com/automotive/gmpowertrain/transmissions/DEXRON-VI-Service_Fill_Release.doc I believe Dexron III is okay for the 4EAT (which is what the '05 2.5i has), but not for the 5EAT (found in the GT).

Some reading on 4EAT operation: http://endwrench.com/images/pdfs/4EAT.pdf http://www.endwrench.com/pdf/drivetrain/02Winter4EATDiagServ.pdf http://www.endwrench.com/pdf/drivetrain/02Winter4EATTransOp.pdf