OB99W

Cookie, you're revealing your (approximate) age, and since I know what you're referring to, I suppose mine is now showing as well . "Steam cleaning" was a bit easier when just removing the air cleaner exposed a nice vertical carb throat (okay, you may have had to open the choke). I've never tried it on a more modern car, but maybe a spray bottle able to put out a large enough volume of water might work. Besides regulating the amount you use, it helps if you develop the knack of modulating the throttle at the same time. At least water doesn't tend to smoke up the whole neighborhood like the chemical cleaners usually do.

The holes in the crank sprocket are pretty near the crank's axis, and symmetrically located. Therefore, even if the "pins" aren't exactly placed (obviously, they're close enough for them to engage the holes) there shouldn't be much change in balance. In fact, the piece of pulley missing at the edge would likely have more of an effect. The timing mark may not be able to be relied on as precisely. However, that doesn't affect actual engine operation. Proper tightening of the crank pulley bolt is what's supposed to stop relative motion between the crank and pulley. The keyways and Woodruff key are there primarily for location purposes; that's why things get messy if the crank bolt loosens. Probably not an issue, but I never say "never". So that brings us around to the "ticking"/"tapping" noise. I don't know offhand what's causing it, either; knowing the apparent area the noise is coming from might help with a diagnosis. Perhaps just tightening the pulley bolt to spec might help, if there's some relative movement between the pulley/crank/pins/sprocket. Things might even change once the A/C and other belts are reinstalled and tensioned. Of course, if the noise was there before the crank pulley work, that's another issue. It could even relate to the question of whether the timing belt is properly installed.

Thanks; here's the correct link for the 2nd part (I also edited my post above, and gave you credit ).http://www.motor.com/MAGAZINE/Pdf/082001_08.pdf

It's likely that the problems are related, and that you have not correctly installed the timing belt. See: http://www.ultimatesubaru.org/forum/showthread.php?t=64923

There's an entire series from Motor magazine covering both the 2.2 and 2.5 timing belts. It includes some very specific claims about loose crank pulley bolts and the relationship to previous timing belt work; I'm sure you can guess what they said, but I'd suggest reading all the articles, as there is some good info in general. See: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 Someone whose Subaru has a loose crank pulley with related damage might consider printing out the articles and presenting them to a mechanic who did the previous timing belt work. It would seem that the owner shouldn't have to deal with the problem. Of course, whether the same mechanic can be trusted at this point is another story. There used to be a kit to repair the damage, similar in manner to what you're suggesting. Unfortunately, I can't seem to find a link to that now. Anyone else have info on this?

There are a few possibilities. One thing to determine is if the coolant is really "boiling", or if there are air or exhaust gas bubbles but normal coolant temp. If truly boiling, that would mean that at least some of the coolant (it can be at a localized area of the engine) reached a very high temperature; even pure water under only atmospheric pressure needs to get to 212F/100C degrees to boil. If hot enough, there should be other indications (gauge, etc.). If the coolant is doing its job, the boiling point gets elevated quite a bit, and if the radiator pressure cap works, it's also raised considerably. If you think it's really boiling, make sure the pressure cap is good and that you used a correct coolant/water mixture; a concentration much higher than about 50% is less effective, including in heat transfer. As to the apparently high oil level -- engine oil does expand somewhat with heat, and that will show on the dipstick if the engine is hot enough. If there's an internal leak, coolant in the oil will eventually make the oil appear milky on the dipstick; if bad enough, it forms an emulsion like mayonnaise. Naturally, draining some oil may also allow you to see if the crankcase has more than just oil in it. If the bubbles are exhaust gas, I guess I don't have to say more...

A rev limiter shouldn't shut the engine down, but the car will feel gutless if you hit the limit and don't/can't upshift. There would be nothing "safe" about an engine shutting down the way described (unless the rev limiter wasn't working, and the alternative was something blowing up).

Xrturbo, thanks for the "heads up". Here's a related previous post: http://www.ultimatesubaru.org/forum/showthread.php?t=58846 (No, I'm not 99obw, although we have similar screen names and locations.)

Indeed.

You say the noise occurs at about 40 mph; what happens to the noise if you drive at 40 in various gears?

It's possible it was damaged during the t-belt work; the crank pulley bolt torque is quite high, and someone may have used an inappropriate method of restraining the crank pulley from turning. For reference (you may not want to get this deeply involved):http://endwrench.com/images/pdfs/TBeltEWWin05.pdf http://endwrench.com/images/pdfs/2.5Timing.pdf

The codes can be dealt with later; right now you need to deal with the crank pulley situation. It's very likely that either the pulley bolt is loose (check that first), or the pulley is coming apart (it's 2 pieces of metal with rubber isolation between them). That's probably why the A/C belt suddenly loosened and the pulley seems "not well balanced". If the pulley bolt is loose and remains so, it can destroy the front end of the crank (keyway, etc.). If the crank pulley itself is bad, it can do damage to other parts. Either way, don't delay in resolving the problem. Once you do so, deal with the other problems.

Unfortunately, cylinders 2 and 4 are not a wasted spark pair (they don't fire from the same coil in the pack). The spark-paired cylinders are 1&2 and 3&4; 1&2 are nearer the front, 3&4 nearer the firewall. I'm not sure exactly what is incorrect in your post, a typo or understanding of the cylinder numbering. See the following for coil info: http://endwrench.com/images/pdfs/IgnitionCoil.pdf http://endwrench.com/images/pdfs/DirectIgnition.pdf http://endwrench.com/images/pdfs/IgnitionCoilSum04.pdf I noticed that 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. Just to be sure the engine is timed correctly, etc., the following may be useful: http://endwrench.com/images/pdfs/2.2Liter.pdf http://endwrench.com/images/pdfs/TBeltEWWin05.pdf The info on which mark (notch) on the crank to use is particularly important. EDIT: Lmdew may have good info. If the above doesn't help, let us kow if you have any more to go on.

There's always some risk when attaching used electrical/electronic parts; heck, even new parts can have problems. However, in most cases involving an ECU/ECM, the risk is damage to it from external factors, and not nearly as much so to external parts from problems with the ECU. Of course, if you connect the used ECU, and anything seems grossly wrong, you shouldn't continue to run it. Minor irregularities in running are probably to be expected until the ECU "learns" your engine, etc.

Did you use an OEM sensor for the replacement? Had any other work been done on the engine immediately preceeding the P0133 code showing with the original sensor? Do you have to add engine oil between changes, or is there even just a noticable drop on the dipstick? If so, how much? Have you noticed any change in gas mileage or performance since the code appeared?

If the problems are ones that arise from manufacturing defects and the like (not owner abuse), and the dealer has been given a reasonable number of opportunities to make repairs but is unsuccessful, you may have a shot at resolution under your state's laws. For CA, see these:http://ag.ca.gov/consumers/general/lemon.htm http://autopedia.com/html/LemonLaw/CA_lemonlaw.html

The 3-wire sensors usually have an output/signal wire and 2 more for the heater, with the ground connection made via the shell. Bosch O2 sensors typically have the two heater wires the same color (white), since the heater has no polarity and it doesn't matter which is which. The black would then be the output/signal. How did you match the colors? As has been suggested earlier, going OEM for the O2 sensor is a good idea because OEM is more likely to be an exact match for the ECU requirements. That doesn't mean that the Bosch unit is the culprit, though. An O2 sensor whose output doesn't oscillate isn't necessarily faulty. For example, if the engine is running very rich, the O2 sensor output will be used by the ECU to try to lean the mixture, but if the rich condition is due to sufficiently leaky injectors, the ECU can't correct enough, and the O2 sensor can't swing the other way.

The glove compartment light is turned on when the switch completes a path to ground. The problem you describe, assuming the wiring diagram I have is accurate, can only be caused by a short in a very limited area. I suspect that the short is right across the lamp socket, since if that's the case the hot side of the lighting circuit would get grounded ("shorted out") when the switch closes. Such a short could be as simple as the bulb itself being bad; I've seen situations where a piece of broken filament lodges in a manner which can cause that, or vibration moves the filament contact/support wires causing them to touch. So, the first thing I'd suggest is to remove the bulb, and see if the problem goes away. If it does, get a new bulb. If not, look closely at the socket and see if there's an obvious short there.

It seems that significant differences triggered the "Phase-2" label. The last page of the following has some pics for IDing:http://www.endwrench.com/pdf/drivetrain/02Winter4EATTransOp.pdf

I believe it relates to this: http://www.ultimatesubaru.org/forum/showthread.php?t=64167

The Phase-2 4EAT wasn't available until MY 99, so a '96 should have a Phase-1. http://endwrench.com/images/pdfs/4EAatPh2Win04.pdf

Good question.

Welcome. Subaru says the '98 OB will run on 87 octane, and the engine controls include a knock sensor that will retard the timing a bit if needed. Some people claim that they get somewhat better performance and/or gas mileage with the next higher grade. I'd suggest that you run a few tankfuls of 87 and calculate the mpg. If you're satisfied with performance and mileage, continue with the 87; if not, move up a grade, run through a few more tanks, and see if the mileage or performance improve (I suspect they won't, or if they do won't be worth the additional expense). Contrary to what some may say, higher grades of gas provide little or no advantage if an engine doesn't require the higher octane.

Yes, if things are really rich, both the HC and CO would be high. It could even help explain why the idle numbers look worse than at cruise. Of course, if the cat saw much of that for long enough, it's probably not doing its job any more, compounding the problem. Amen.

Yes, a bad ECU coolant temp sensor can do that. So can a stuck open thermostat (not common), or a missing one (someone may have pulled it); might not be obvious until the weather gets colder. High HC could be due to a burned or otherwise leaking exhaust valve; a compression test should reveal that. By the way, is this car a 2WD? I'm asking because cruise emissions were quoted, which is a dyno test, and few test stations have 4-wheel dynos.