OB99W

As long as there's some point where the fins aren't bent (or you unbend them with another tool for a short distance), the comb usually can be inserted and then run the full length. Of course, if the condenser was smashed with a hammer, all bets are off.

Google "ac condenser fin comb". If you use one, be sure to match the tool to the condenser's fin count.

How does what you have compare with the label on the underside of the hood? Unless they're small, host them on another site and provide links.

I don't know exactly what you're seeing either, but if black, it could be carbon buildup. If metallic, could be due to detonation (running bad fuel, anyone move your knock sensor -- or replace it with a resistor?). If black with metal showing in spots, could be detonation impacts on a carbon layer.

Personally, I only "give up" when the other person wants to, they continually ignore my suggestions, I run out of ideas, or we resolve the problem. Naturally, I prefer the last of those.

As Fairtax4me is (I think) hinting at, if the front sensor wasn't a Subaru OEM one, that might be possible cause number 4 .

Not to distract from my post above, but did you check into what Gloyale mentioned in the last paragraph of post #35 in this thread? As he mentions, there can be crosstalk between the ignition cables and injector wiring if they're too close to each other. In fact, if the ignition cables have been replaced, it's important that the factory corrugated wire loom is in place -- unfortunately, sometimes it doesn't wind up on the replacement cables. See if your plug wires look like the attachment.

I'm sorry that the A/F sensor didn't resolve the stalling problem. I'm willing, if you are, to try a bit more diagnosis. We haven't looked at the live data since the FPR and A/F sensor were replaced. There may have been changes that would be revealing. If you feel like it: 1) Warm the engine to normal operating temperature 2) Use the 3130 and note all the live data parameter readings at idle 3) Watch them for a while, and if any change much, note that as well 4) Bring RPM up to 1500, and again note all live data parameters I'm interested in seeing all the data, although in particular I'm wondering where the "Calc Load %" and "Spark Adv" are at now. The spec range for engine load at idle is 1.6-2.9%, but previous readings indicated around 5% (about double the norm). Also, the ignition advance was retarded from normal and varying before. In addition, although lessened after the FPR change, the LTFT was still heading a bit negative, meaning the ECU was seeing the mixture as somewhat rich. You may wonder why I'm concentrating on those three parameters. The reason is because a vacuum leak will cause the MAP to be high, which in turn causes the ECU to calculate a higher load, resulting in ignition timing being retarded and mixture being enriched! I know that you checked and didn't find a vacuum leak, but perhaps there's one that escaped detection. So, that's why it would be good to know what the live data shows now.

Welcome to the forum. The diagnostic connector to locate is B82, 6-pin, black. If it and the two ground leads aren't hanging free above the accelerator pedal, then they probably haven't been used before, and are taped to the harness. You may find the following thread, and the PDF file links within, helpful: http://www.ultimatesubaru.org/forum/showthread.php?t=19805 You may also find this interesting: http://endwrench.com/pdf/feb2004pdf/4EAT.pdf See especially the "Trouble Codes" section, page 13, bottom left column, (page 8 of PDF).

I provided the link as response to the question, not as an endorsement. In my opinion, a properly performing original ignition system (good coil, wires with intact insulation and resistance in spec, clean plugs properly gapped and not overly worn, etc.) is all that's needed for reliable firing in an unmodified engine. Manufacturers do typically get this kind of thing right. In fact, increased spark voltage raises the chance of insulation breakdown or flashover if other parts aren't also upgraded. On the other hand, a modified engine might benefit from higher than stock spark voltage. In particular, raising compression requires higher voltage, all other things being equal. Of course, it's also possible that a poorly maintained engine might run better on a hotter spark, but a good tuneup would likely have the same effect.

http://sl-i.net/FORUM/viewtopic.php?f=39&t=9247&start=0

If you'd like to read the SRS codes, see: http://www.ultimatesubaru.org/forum/showthread.php?t=99872 The EndWrench article referred to in that thread is no longer accessible, but it's not really required. As long as you can find the B82 diagnosis connector and "diagnosis terminals" (ground leads), you can do the procedure. It's under the dash, above the accelerator pedal, but if not previously used it may be taped to the wiring harness. The thread includes a link to another post showing the B82 diagram, but I'll repeat it here: http://www.ultimatesubaru.org/forum/showpost.php?p=668884 Once you find the connector: 1) Turn the ignition switch to "ON", but don't start the engine. 2) Connect a ground lead to pin #1 of the B82 diagnostic connector 3) Note the long and short flashes of the airbag light Let us know what codes you get.

Or possibly just a bad connection of one of the cables at the battery. Sometimes corrosion gradually increases there, and some vibration or a temperature change is all it needs to get bad enough to not conduct enough current. I'd look at the battery connections first.

If a careful visual inspection doesn't turn up the cause of the problem, we can start troubleshooting with a basic check. Look under the dash near the accelerator pedal, and find the two green test mode electrical connectors. Couple them together. Turn the ignition switch ON. This should cause several things to cycle off and on, including the fans. Let us know what happens.

In the fan circuit, a black wire is a ground (common) connection. As mentioned, look carefully for other wire damage, and repair as needed. Does the second fan come on if you run the A/C?

To replace the damaged section, I'd use wire of the same gauge, soldered, and heat-shrinkable tubing to insulate the splices. What color is the insulation of the damaged wire? (It might give us a clue as to what happened.) Do you have a voltmeter, and know how to use it? If so, I can go into detail on how to troubleshoot the problem.

I'd be inclined to check for a pinched wire in the fan circuit.

You're welcome! With so many sensor circuit problems apparently resolved, it's possible that one of the multi-pin E/B (engine-to-bulkhead) connectors at the rear passenger side of the engine wasn't fully seated until you removed and reinserted it.

Unfortunately, not much changes. I gave the link to the previous thread, not in recognition of your dearly departed 2001, but because it related to its HG problems.

Is it a scan tool, or just a code reader? Either way, more info is better than less. You may have "fixed" the problem, or at least made it go away for a while. (More on this below.) I'd suggest a new air filter (okay, that's obvious ). To perhaps stop (or minimize) the oil contamination, clean or replace the PCV valve, and make sure the related tubing is free of sludge. P0108 Pressure sensor circuit high input P0113 Intake air temperature sensor circuit high input P0117 Engine coolant temperature sensor circuit low input P0325 Knock Sensor Circuit Malfunction These are all circuit faults, most likely due to bad connections. Your work on the connectors could have resolved any or all of them. If the problems never return, there's nothing more to do. If any of the symptoms and/or codes recur, let us know. It should then be possible to determine which connector(s) and pin(s) are the likely culprits.

As you found, TCU code 45 relates to the intake manifold pressure signal (the "baro", which I mentioned previously), and indicates an open or short in the circuit. This can be "real", or due to a problem in the TCU itself. The first thing to do is check the ECU for codes, even if the CEL isn't lit. Specifically, P0106/7/8 might show up. As you brought up before, it could be that there's a swap-related connection problem, but let's hold off on that until you see if the ECU is storing any codes.

Okay on the cam & crank sensors and reluctor condition, and thanks for posting the fuel trims. A quick explanation: As the engine runs, the value of fuel trim that the ECU uses to modify the injector output is the total of the STFT and the LTFT. Short-term fuel trim can vary considerably under normal conditions -- if it shows a definite trend, the ECU modifies the long-term trim, which allows the short-term to not run out of range. The problem is when the long-term value keeps climbing, and the short-term continues to cause it to move further in the same direction. Your figures show the LTFT now averaging around -1%, while previously it was getting up to about -12%. So it seems likely that the new FPR isn't allowing the mixture to get as rich as it was before, and the ECU therefore isn't having to correct as much. Typical failure of an A/F (O2) sensor causes it to both slow in response time and to narrow the range of the response. Often it will still see trends well enough to affect fuel trim, but not rapidly or sufficiently enough to give the ECU the instantaneous feedback it needs in order to properly adjust injector output as engine load demand changes. I'm hopeful that a new A/F sensor will allow the ECU to have tighter control of the mixture.

With that many codes (and especially having both P0122 & P0123, which conflict with each other), I'd suggest clearing them (reset the ECU) and see which return. While waiting for the codes to retrigger, don't do any troubleshooting involving disconnecting sensors, etc., while the engine is running, because that will trigger codes that can be misleading. You mentioned having taken the intake manifold off (or partially off? -- it wasn't clear). If it was removed, have you checked for vacuum leaks? Also, recheck any electrical connectors that were disconnected.

With a new FPR, the fuel pressure should be within specs. With the ECU reset, if the A/F sensor was responding well, the engine shouldn't run worse than it did previously. The poorer running is likely another indication that the A/F sensor should be replaced. Even with a sluggish A/F sensor, there will eventually be some adaptation to the new fuel pressure (fuel trims will change), but that's not a good reason to leave the old sensor in place. If you get the chance, it would be interesting to see what the STFT and LTFT are currently. A single live data reading, once the engine is warm, should be enough. There's another "chicken or egg" situation to consider. We checked the knock sensor and timing belt taughtness because ignition timing was erratic. Misfiring can cause that, but conversely problems with sensors can cause timing to be erratic and trigger misfiring. You could look at the crank and cam sensors, and get a basic look at the condition of the reluctors on the crank sprocket. First, check that the sensors (and mounting bracket for the cam one) are not loosely mounted -- spec is about 5 ft-lb. Then, remove each and make sure that the tip is clean. Look down the hole at the crank, and see if the reluctors seem rusty. R&R details are in the Fuel Injection section of the manual. I sense that you're becoming somewhat disheartened. I think I've said this before, but since there are no codes other than the misfire ones, the problem is subtle and not easy to pin down. A car with over 200k miles on it can have several sensors, etc., that are somewhat out of spec, but aren't bad enough to trigger codes. It may well be that the problem is due to some combination of marginal parts. The ECU does a good job of correcting for irregularities, and while it keeps the engine running with them, it also "obscures" some of the faults. We've ruled out several possible causes, and you've replaced a few parts that were determined to be bad. I'm hopeful that the last piece of this puzzle is near.

Thank you, Olnick. I like to contribute because it keeps my mind from wandering and confusing MTs with 4EATs, if you know what I mean.