OB99W

As Skip said, it's under the #3 intake runner. Does this help, or make things worse? :

Nah, you're doing too good a job on yourself.

What, another variable? Sure, bad fuel (or a loose/corroded connection, or ...) could cause an intermittent problem. Oxygen sensors can be "poisoned" by contaminants; sometimes they'll burn off and operation will return to normal. (I'm not saying this is what happened, just that there are lots of things that could potentially go wrong, and that's why proper diagnosis is critical.) If the problem recurs, that's exactly what should be done. Can you change the "State" to "International"?

Subaru-recommended maintenance schedules can be found in the owner's manual, or at http://www.subaru.com/owners/schedules/index.jsp?from=topNav . Click on model year, then on the appropriate link at the bottom of the page. As to plug change interval for 2006, Subaru apparently specifies 60k for the Turbo or 3.0L, but only 30k for the N/A 4; that may or may not be conservative.

According to: http://www.tc.gc.ca/roadsafety/importation/VAFUS/list/VAFUS.pdf , most Subaru models (as well as those of some other manufacturers -- read the chart carefully) are both admissible and inadmissible! Maybe someone should explain the difference between the two words to whomever put that info together. Good luck; I have a feeling you may need it.

http://www.linkedin.com/pub/4/9B2/17BKotzwinkle would either be proud or confounded. Oh, and to keep this on-topic, although the problem could be due to insufficient voltage reaching the motor, I agree that lubrication is the first thing to investigate.

Speaking of waters, "across the pond" they sometimes don't have the same configuration as we do here, even for a particular model year. If anyone is in the know concerning what welly's vehicle is likely to have, it could be useful.

TOAST = Temperature Obviously Above Something Tolerable ? -or for the cynical- TOAST = Tapped Out, Applied Some Torching ?

Okay, this is going around in circles. Here's some info on A/F and MAF sensors, to help determine which we're talking about. Please don't me just because it's from Toyota. A/F (Air/Fuel) (O2) sensors: http://www.autoshop101.com/forms/h37.pdf http://www.autoshop101.com/forms/h44.pdf MAF (Mass Air flow) sensors: http://www.autoshop101.com/forms/h34.pdf

Yes, it's likely there has been confusion. The A/F (air/fuel) sensor is often referred to as an oxygen (O2) sensor, and is indeed in the exhaust stream. Actually, there are likely two of them on your car, one preceding the catalytic converter, and one following it (assuming, again, configuration like US vehicles). The first one provides data to the ECU, and helps it determine A/F ratio and consequently fuel injector pulsing, etc. The second is used to check up on how effective the catalyst is in the converter. Lack of lift or ramps is definitely more of an issue than if the MAF was the problem. On the other hand, an O2 sensor should be a lot less expensive than a MAF. Also, failure of that part on an '01 is fairly common, and replacement can solve some pretty poor running conditions.

You shouldn't need a pic. The only "lifting" needed should be that of the hood/bonnet -- find the air filter, and the next thing in the intake line should be the MAF. Before going further, would you note the part number, and post it here? It should be on a label, in the form "xxxxx AAxxxy" (where each "x" is a number, and the "y" may or may not be an additional character). Because you're "over there" (), your car might not be outfitted the same as ones we're used to "over here". About 1999 (depending on model, etc.) Subaru changed from thin-wire to thin-film MAFs; some of them were very unreliable, and there were "improved" parts issued. The part number, and whether there's a green-colored dot on the MAF, might help in determining which one your car has.

Since the definition for code 22 is "Knock Sensor or Circuit", it seems that code is set after a continuity check by the ECU, and wouldn't be affected by engine temp. Based on that, it wouldn't surprise me if a disconnected sensor immediately sets the same code.

'99: http://www.preamprod.com/ssas06/larger/ssas06_09.jpg

Those aren't the things that typically happen when the HGs go bad on the EJ engines. If you're saying that "air" is under pressure, then it's probably exhaust gases, not "air". You can often detect the exhaust smell, if there's enough gas. The cooling system on EJ engines needs to be carefully "burped", and even in a warm climate coolant/antifreeze (about 50% concentration) should be used. If you've done that, unfortunately it's likely at least one HG is bad, just perhaps not enough to be spewing coolant out the reservoir overflow all the time. If you get HG work done, both sides should be replaced with the latest OEM gaskets and the correct torqueing procedure should be used. At the same time, it would be prudent to change the timing belt (as well as idlers, tensioner, and water pump, unless you can verify that any were recently replaced). As WAWalker would likely tell you, make sure that valve adjustment isn't ignored .

No matter how good and "multi-vis" the oil, it's thinner when hot, and will get past small openings more readily than when cold and more viscous. Perhaps I'm missing your point, but valve seating shouldn't relate to oil getting into a cylinder. Valves in an engine that isn't running normally are at various degrees of being open -- however that's to either intake or exhaust manifold, not sources of oil unless something is really wrong.

Poor gas mileage and erratic idle speed (loping idle, if that's what it's doing) point to an overly rich mixture. As Skip mentioned, a CTS problem could cause this; if the ECU thinks the engine hasn't warmed, it will run "open loop" (default rich mixture) all the time. The CTS isn't the only possible cause, but I'd consider it first, given the symptoms.

Don't know for sure, but my guess would be that it does. As I understand it, one (the only???) of the functions of the knock sensor is to prevent engine damage cause by pre-ignition/detonation that could be brought on by running a lower octane gas than is recommended or by somehow having the timing advanced too much. Unless there is some other mechanism for retarding the spark timing when the engine is cold, these conditions could be present when the engine is cold (if a lower octane gas was used, for instance). I can't say it's universally true, but I know that many Subaru (and other manufacturers') ECUs default to limited ignition advance for cold start/run. Considering that detonation isn't typical on a cold engine with little spark advance, that wouldn't seem to present a problem. Based on GD's experience with the code only setting after engine warmup, and that being resolved when the knock sensor was replaced, it seems to verify that the sensor signal is ignored until warmup. However... I don't know if a missing knock sensor would immediately set a code, Cougar. Anyone know for sure, or want to disconnect one and report back?

Yes, Nipper, that link does cover a lot of basic info. Most of it is accurate, but there are a few errors and misconceptions. Some aren't too important (it's "Tesla", not "Teslar", and electromagnetic interference is "EMI", not "EMF", which is electromotive force). Some statements are misleading. An example from the section on point dwell is "The ratio of closed points to open is usually about 3:1.". While that ratio can vary depending on application, it's more typically about 2:1. Under "Common problems with a conventional ignition system are:" is "Timing belt (chain) wears and/or breaks "; well, as we know, that isn't limited by the type of ignition. Still, not a bad article, but I'd suggest reading it with a grain of salt.

Thanks for the kind words, Cougar.

Yes, I've got that info also. However, if you read ferret's FSM info, and compare it to jib's circumstances, it doesn't all fit neatly together.

Cougar has described the circuit connections, but I'd like to clarify the operation. "When the key is turned to Start and before the solenoid has had time to close", the "second coil" (pull-in winding) is energized (getting its ground through the starter windings), as is the hold-in winding ; that is, both windings are energized to accomplish pull-in. Since solenoid hold-in requires less force than pull-in, only the hold-in winding needs to remain energized once pull-in has occured, so the pull-in winding is depowered by being shorted out when the contacts make. As to the criteria for setting P1518, apparently Subaru isn't sure. Therefore, whether a bad solenoid can be the cause or not is something I won't take a stand on any longer. If the starter is cranking normally and the code is set, it's likely to just be caused by a bad connection to the ECU's sensing circuit. If the starter is misbehaving, then check the things that can cause that, and when fixed the code will probably go away. How's that?

Assuming that the car is really an '02, unless the code was misread, it seems a Subaru with OBD-II has set P1518.

I understand that the solenoid has two windings (or a tapped winding, based on the diagram). The external connection is for power to "pull-in". When the internal contacts make, that provides power to "hold-in" (but as you noted, it doesn't really hold if "pull-in" power is removed). However, jib said "I don't believe the starter is in an overrunning situation"; it should overrun if the solenoid sticks. What I find annoying is that different Subaru factory manuals apparently describe the conditions under which the code is set differently, and appear to also have diagnostic procedures that vary. It's not like this is a very complicated circuit. Oh well...

Just as long as it doesn't become "nitrous in, bottom end out".

Does the ECU monitor the knock sensor when the engine is cold (during "open-loop" running)?