OB99W

The wiring diagram I have shows the switches are illuminated on the '98 (aside from the orange indicators that show power is going to the seat heater). I just verified that on my own '99, so assuming the diagram is correct, the switches should be illuminated.

Not as far as I know; my own '99 OB has only a small green light above the driver's PW switch. The switches should be lit; they're powered via the instrument illumination circuit, so the bulbs may just need to be replaced. Seat heating is a separate issue; they're connected first through a 20A fuse and then the rear power supply relay. Does the rear accessory power outlet work?

If you don't want kids learning "new" words, don't have any around when you do this work. Also, a small first-aid kit (or at least some Band-Aids) might come in useful. No, I'm not kidding.

Congratulations!

Just in case there's air, here's what Subaru says about bleeding the cooling system: http://endwrench.com/images/pdfs/Ble...erEWFall05.pdf

Are you sure that it's not coming up to proper temperature, rather than that the gauge is giving you an incorrect indication?

Unfortunately, wiwrx hasn't told us if the coolant is really boiling or not (see my post #6 in this thread). Bubbles rising through a fluid, of course, isn't necessarily indicative of a boil; just ask the fish in an aerated aquarium .

A bad cap is only one place that an EVAP system leak can occur. For a starter, it might be worthwhile looking at this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=65045

Yeah, and as I'm sure you also know, there's an "igniter" between the ECU and the ignition coil (integrated into the coil pack on later models), etc. It's kind of amazing, but properly designed semiconductor-based switching circuits tend to be more reliable than the mechanical parts they replace. Of course, accidentally shorting something doesn't add to their life.

It can, although it may take some time and miles to significantly appear there. I'd check other places if you suspect coolant in the oil.

It's too bad you can't access End Wrench. Yes, the right side is passenger, although I must admit that location isn't as near where the circuit is used as it could be; maybe things are too tight in the Forester. Anyway, the article says it's under the dash, above the kick panel. It could be that the bulbs are out in the switches that don't light. If the radio/CD doesn't dim when using the control, whomever installed the unit didn't tie it into the "rheostat" circuit, which would explain why it's lit when other things aren't. If the aftermarket radio/CD was installed before you got the car, it would make me wonder if the wiring was damaged and that's why certain things weren't lit since you got the car. Best of luck; hopefully replacing the module (ummm, "rheostat" ) will fix at least the instrument and shifter lights.

You're welcome; let's see if we can clarify things a bit further. Let's make sure we're on the same page; I was only asking about the illumination of the radio dial and hazard switch "button", not whether the radio plays or the hazard lights flash. The reason I'm asking is because 83023FA000 controls more than the "dash lights, shifter light and heater/ Air control light" that you've mentioned. If those are the only lights out, 83023FA000 may be okay, but a wire in the harness may have burned, etc. If in addition to the lights you previously mentioned, all from the list below are out, then 83023FA000 is probably bad, so go ahead and replace it. (The End Wrench article that I previously provided a link to indicates it's located under the dash above the right kick panel on your Forester.) 1) Front Foglight Switch Illumination 2) Hazard Switch Illumination 3) Rear Defogger Switch Illumination 4) Cruise Control Main Switch Illumination 5) CD Illumination 6) Radio Illumination If not all of the above are out, which work and which don't? Please tell me by number, keeping in mind I'm only talking about lighting of the switch button, dial, etc.

It's probably less expensive (and easier to mount) a low-power-handling variable resistance device and have it control a higher-power-handling semiconductor circuit. A real rheostat that has to dissipate enough power is bulkier and could get sufficiently warm so that it can't be mounted just anywhere. EDIT: I just verified that some places refer to the Illumination Control Unit as a "rheostat" (I guess that's a throw-back to older times). On my own '99 Legacy OB, it's mounted to the back side of the lower dash panel (beneath the steering wheel), easily accessed by removing two screws at the lower edge and gently pulling at the top to free the clips. It's part number 83023FA000, and says "Illumi Con Unit" on it; the same unit is used on the '99 Forester, but the location is apparently different.

As far as I know, the "rheostat" on your '99 Forester doesn't directly control the light dimming; it connects to an "Illumination Control Module", which does the actual controlling. The control module is a more likely failure in my opinion, based on the improper bulb probably having drawn too much current. According to the wiring diagram I have, it appears to be a unit with wires colored green/yellow, orange/white, green/red, blue, violet, and black going to it. For '98-99 Foresters, here's info on where the modules, relays, etc., are located: http://endwrench.com/images/pdfs/ElectricComponInfo.pdf By the way, I assume that not just the instrument lights are out, but also several other things like the hazard switch, radio, etc aren't illuminated. Is that correct? I'm curious as to what part you were sold. Do you have the part number of the "rheostat", and a more-detailed description of it than you've already given?

Good advice. Here's what Subaru says, although I don't think it's the "final word" on the topic: http://endwrench.com/images/pdfs/BleedInsiderEWFall05.pdf

It's important to check both the radiator and coolant recovery tank levels. Certain types of cooling system leaks can cause the recovery tank to still have a reasonable fluid level, and yet the radiator (etc.) not to be full. In fact, if you find that your recovery tank level seems okay, but the radiator is down, checking for a leak is a good idea. This End Wrench article explains in more detail how the radiator can be down while the recovery tank still has coolant, and is a good read overall: http://endwrench.com/images/pdfs/CoolMay05EW.pdf

Normally, the "high" side of the bulb connects to 12V via the fuse (which of course protects some other interior illumination as well), and the bulb's "low" side goes to one contact of the switch. The other contact of the switch goes to ground, and when the switch closes (glove box door is opened) it therefore connects the bulb's "low" side to ground, completing the circuit. In your situation, when the switch closes it's apparently shorting the bulb's "high" side to ground. Since removing the bulb didn't make a difference, that pretty much leaves only a short directly across the bulb socket as the problem. The diagram I have shows the switch and socket as one assembly, so if examining it doesn't make the problem obvious, you could unplug it. Probably not. If a grounded wire came in contact with the "low" side of the bulb, it would just turn the glove box bulb on. If that grounded wire touched the bulb's "high" side, it would blow the fuse even without opening the door. I suppose that in a very unusual situation a ground could be close to touching the high side, and opening the door flex things enough to make contact... Check your private messages first.

I don't have any fresh ideas at the moment, but I do have an observation; you previously said "light didn't come on until months later, so no obvious connection." Does anyone here know if the airbag diagnostics could be "fooled" into ignoring the supposedly missing sensors for that long, and then have the system finally recognize the problem? Seems a bit strange to me; if someone fudged things, I'd usually expect that to be long-term (until corrected). Could it be a second airbag system issue in addition to the sensor question?

Yes, but I've got a question: Hey Nipper, does the ethanol that makes things so fast go into your "tank", or the OB's?

You're welcome; I'm glad to help. Yes, the End Wrench articles can be enlightening (thanks to Subaru), but errors do seem to creep in on rare occasions. If you ever read something there that doesn't make sense, hopefully someone here on USMB can clarify it.

The links below will provide some info on "wasted spark" ignition. There's also a part explaining why it's better to electrically disconnect a fuel injector rather than disabling ignition when checking cylinder balance. http://endwrench.com/images/pdfs/DirectIgnition.pdf http://endwrench.com/images/pdfs/IgnitionCoil.pdf http://endwrench.com/images/pdfs/IgnitionCoilSum04.pdf There are errors in the "DirectIgnition.pdf" info. In the "Ignition Coil Testing" section, the terminal numbers listed are reversed between the primary and secondary; however, the diagram is labeled correctly. The correct numbering relative to expected resistance is in "IgnitionCoil.pdf"; you can use the diagram from "DirectIgnition.pdf" to identify the terminal numbers.

See this article; includes info on retrieving codes pre-OBDII :http://endwrench.com/images/pdfs/EWPreOBDAug05.pdf

Which hopefully indicates that the fuel gauge, like many, is inaccurate. 3/4 of a tank is about 12 gallons; going 190 miles on that results in less than 16 mpg. Even city or local driving should yield at least 19 mpg, assuming properly inflated tires and reasonable driving. As has been mentioned before, it would be best to wait for a mileage determination based on a tank refill or 2, 3 or more of them.

I apologize for suggesting that you may not have installed the t-belt correctly. However, the symptoms you described do fit those of an incorrectly timed engine. Perhaps I'm alone in this, but "cough" implies to me that there is ignition of the mixture at least sometimes. A no-start condition and not even a "cough" suggests no ignition of mixture at all. That can be due to lack of spark, lack of air/fuel, lack of compression, or things not properly coinciding in time and space (otherwise known as engine timing problems). If this is not due to incorrect timing, then that could leave things like a bad ECU or wiring problems, damaged reluctor(s), etc.

Mike, congratulations on passing inspection. I have mixed feelings about OBD-II. We need a way to determine what the engine management systems are doing, but the DTCs can certainly be misleading at times, and manufacturers' data seems to have more than a reasonable number of errors. If we're taliking about using OBD-II controlling a MIL/CEL (and whether your state will let you drive the car based on that), I have stronger feelings. I'm sure there are people on the forum who'll disagree, but in my opinion a car that pollutes (significantly) more than necessary isn't something to ignore. I realize that some have made this into a political issue, but our health and that of everything on the planet is directly or indirectly related to things like auto emissions, and that shouldn't be a political football. Thanks for your feedback to the forum. I agree that it would be very useful if more participants got back with info on the outcome. If we got things right, it's nice to get a pat on the back, besides the obvious value of confirming an approach to anyone doing a future search. If we were mistaken or missed something, it certainly helps to know that as well.