OB99W

It depends on the crack; repair of edge cracks is often less successful than those that are located elsewhere. You might want to check with your auto insurance company, since many of them will cover windshield crack repair with zero deductible.

Nipper, see this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=19805 The second post (by avk) in particular has links to some good info, but the entire thread is worth reading.

Sorry, I don't have info on 16 flashes of equal length and spacing. However, retrieving AT trouble codes for your OB is covered in the link and page numbers I mentioned above.

Uh Nipper, year and model ? If it's 4 groups of 4 flashes each, I believe that's an indication of a general electrical problem (not specific); you'd need to pull the code to get details. If yours is a 95-98 Legacy (or some similar model), you might be able to get someone to crawl under the dash and use connector B82 and ground probe to get the AT dash lamp to blink the code(s). See pages 12 and 13 of http://endwrench.com/images/pdfs/4EATPh1Win04.pdf

On the question of "How will my Sube die?": From http://www.williamson-labs.com/one-hoss-shay.htm "You see, of course, if you're not a dunce, How it went to pieces all at once,-- All at once, and nothing first,-- Just as bubbles do when they burst. " --Oliver Wendell Holmes Or perhaps "Not with a bang but a whimper." (Apologies to T.S. Eliot, who wasn't speaking of vehicles.) http://www.cs.umbc.edu/~evans/hollow.html

There are two basic types of ABS wheel speed sensors. The variable-reluctance kind are just a coil of wire wrapped around a magnet, and output an analog (sine wave) signal whose frequency and amplitude (tone and volume ) change with wheel speed. The other type is magneto-resistive, includes some electronic circuitry, and outputs a digital signal that changes in frequency with wheel speed, but has a fixed amplitude. Earlier ABS sensors tend to be of the variable-reluctance type (my own '99 OB has that). I don't know if Subaru has switched type in later model years. I guess we're now up to $.04 .

Maybe whoever decided to call it a tone wheel had some familiarity with old Hammond organs. See http://b3world.com/TechInfo/figure5_1.html .

If at all possible, an air compressor used for A/C service should have an air/oil separator, or be an oilless type; otherwise, there's the chance of contaminating what you just cleaned. An air/water separator would be advantageous as well.

Thanks for clarifying that. When I mention an "extended period of time", I'm thinking of anything significantly more than what's needed to open the system and replace a part or plug the opening(s), which usually doesn't take more than a few minutes.

Yes, but you should use the correct oil, and you haven't mentioned what model year you're working on, etc. I looked back at your previous posts; if we're talking about the '98 OB, use PAG oil or something compatible with it. The oil is distributed in the system as it runs, although much of it is in the compressor. Unless there was a significant loss of oil due to the leak (which usually doesn't happen, but do look for signs of that), there's probably no need to add oil. However, refrigerant oil is hygroscopic and if the system has been open for an extended period of time, it may have absorbed enough moisture to need replacement. You might find the links below useful. http://endwrench.com/images/pdfs/ACTechTips.pdf?PHPSESSID=f79afa671304403610485a3a591893f7 http://endwrench.com/images/pdfs/ACSystems.pdf?PHPSESSID=fb6fe5310b6c38d169e1b7629cb76948 http://endwrench.com/main.php?smPID=PHP::search_articles.php

From http://www.id-usa.com/how_to_faqs_retrofitting.asp : "8. Why does Interdynamics use Ester Oil instead of PAG Oil? While both lubricants are used with R-134a, Ester is believed to be better for Retrofit systems because it is compatible with the residual mineral oil left after evacuating a R-12 system. In addition, Ester oil is a preferred top-off oil because it is compatible with ALL PAG Oils and is much less hygroscopic..."

R-12 systems can use mineral-type refrigerant oil, R-134a should use PAG (polyalkylene glycol). I believe there are ester oils that are compatible with either, but check labels.

Yes, do oil the o-rings, but I'm not sure that motor oil or brake fluid are good choices. They're meant to be lubricated with refrigerant oil; if you don't have any, use something like this: http://www.nationalonlinetools.com/index.asp?PageAction=VIEWPROD&ProdID=28584

Okay, you get points for that . Since I now expect someone to ask what the diagnostics are for, as far as I know connector B82 and the probes are used to help with ABS, air bag, and/or 4EAT diagnosis, depending on the model year. For example, I last used mine to get the ABS dash light to blink the fault code, and then to reset (erase) the code in memory once the ABS repair was done.

They're factory wiring. The "bare" ones are ground probes. The connector and the probes are only supposed to be joined when you're running diagnostics. As I said before, they should be insulated with electrical tape and fastened to the harness so that you can't step on them (and to keep them from shorting to something else) !!

What do you see where the thumbnail picture(s) should be; a red "X", blank space, something else? At least that way you can't be accused of being "all thumbs" .

I hope you've plugged the openings; if the system is open to atmosphere for a while, the desiccant can absorb too much moisture and you may need a new receiver/dryer.

At least on my '99 OB, they're all different.

Appears to be diagnostic connector and probes. They're normally insulated with electrical tape and taped to the harness, high enough so that you can't accidentally stomp on them when aiming for the accelerator.

See http://www.ultimatesubaru.org/forum/showpost.php?p=464800

It's good that the torque wrench you're using agrees closely with a bending beam type (assuming that's what you mean by "pointer type"), which tend not to go out of calibration to any great degree. Do keep in mind that torque measurement is only an indirect indicator of bolt (stud) tension, which is what really determines the clamping force. Unfortunately, a torque measurement is very dependent on frictional forces. Lug nut torque specs are usually given for dry threads, but that also assumes the threads are clean and free of corrosion. It's possible for torque reading to be "correct", but for the tension to be too low, if there is any significant dirt, rust, etc. in the threads or on other contact areas. Since you're using steel wheels in the winter, I assume that you're changing to the Subaru OEM lug nuts when the alloy wheels are reinstalled. If the nuts used with the steel wheels have open ends, as they typically do, some rust may be forming at the exposed end of the lug, which may affect the performance of the Subaru nuts. I'm not saying that's the situation in your case, but it might not be a bad idea to take a wire brush to the lugs and make sure they're clean, for the reason I mentioned above. This may be obvious, but lug nuts should spin on fairly freely; any significant resistance may indicate thread damage that could cause torque measurement to be a false indicator of tension.

It's interesting that no leakage was found, but that apparently you're only covered for labor "if it does leak over the next few weeks". If there's a leak severe enough to be evident in a few weeks, it should be detectable now without too much difficulty. Seems like either the mechanic didn't do much/proper leak testing, the leak really is small, or the mechanic is hoping you won't notice a drop in A/C performance for at least a few weeks :-\ . If the charge doesn't hold for a reasonable time, and your "mechanic" isn't an A/C specialist, perhaps you could try to find one. Leaks can develop in lots of places (including pinholes in hoses, condenser, evaporator, etc.), but the most common locations are at fittings (o-ring seals) and the compressor shaft seal. Leaking refrigerant doesn't leave a visible clue, but it sometimes carries enough refrigerant oil with it to be noticeable; if you see an oily area around a connection, etc (may be dirty), that could be an indicator of a leak. Here's hoping you can keep your "cool" .

Glad to help. You've done about as much as you can with a multimeter; at least now you have an idea of what to expect when you take it in for service.

Thanks for the links, Carl. I'm in the 56-month "belt", which is apparently not only how long the average battery lasts according to the map, but also how long a car is free of significant rust (read "road salt belt") . Seems that by getting closer to 72 months, I'm seeing better than average battery life, even for my area. As I said previously, for me that happens with mid-line batteries that are correctly maintained.

Admittedly, my experience with about 6 years being typical life is based on a cooler climate (and good maintenance); Texas weather probably does significantly shorten battery life .