OB99W

P0420/catalyst efficiency links: http://www.motor.com/magazine/pdfs/102006_09.pdf http://www.catalyticconverter.org/news/news_page.cfm?Key=catalytic_converter-&News=120 http://www.import-car.com/Article/39019/Diagnostic.aspx

Ground clearance would go up, but by considerably less than an inch. The increase in width is 215mm minus 205mm, or 10mm. The ground clearance change would be 70% of that (from tire aspect ratio), or 7mm -- that's a bit more than 1/4''.

AT Oil Temp warning light usage started in 1995. Yes, on earlier models (like your '93) it's the ''POWER'' light that flashes on start-up if the TCM senses an electrically-related trans problem.

What number(s) is/(are) imprinted on the original bulbs?

Any loose-fitting vacuum hoses and other possible intake leaks should be dealt with. As I mentioned before, if there is any sign of gas in the vacuum hose at the fuel pressure regulator, its diaphragm might have a hole. Possible causes of P0172 include excessive fuel pressure, intake temperature/pressure sensor problems, engine coolant temperature sensor failure, and even loose engine ground connections. In rare cases, a hole or loose connection in the exhaust system can ''confuse'' the oxygen sensor. If you don't have the experience to safely/carefully check those things yourself, I'd suggest taking the car to a good independent Subaru specialist, or the dealer.

The low side might still be close to the low pressure cut-off switch threshold, but ... ... there's no question that refrigeration is all about temperature! It's good that you got it to the point where it's not short-cycling. I agree that waiting for a day that's warmer when you can recheck the pressures and overall operation under a more typical load is a good idea.

Measuring the resistance of O2 sensors (other than when checking the heater in ones that have it) isn't the right way to test them. The sensors generate a voltage when they're hot enough -- the voltage is highest when there's the least oxygen, and lowest when the most oxygen is present. The openings in the sensor can become obstructed, or the sensor element itself can become contaminated. Either could cause the sensor to not respond sufficiently (not enough voltage swing) or rapidly enough. Sometimes a prolonged hard run on the highway can burn off contamination if it's not ongoing -- that goes for the catalytic converter as well, by the way. A ''good'' way to kill an O2 sensor is to use silicone sealant that isn't ''sensor safe'' on anything that the engine oil comes in contact with or otherwise connects to the crankcase. Besides the exhaust, the sensor also needs to reference the outside air, so contamination of or any sealant applied to the outside surface can also damage it. As Log1call said, testing on the car is done with the sensor connected, and an oscilloscope or digital meter with such a function would be two ways of seeing what the sensor is putting out. With the system in closed loop, the voltage swings should easily be evident. You could ''play around'' with an O2 sensor when off the car by carefully clamping it in a vise (by the hex, not the body), attach a voltmeter, and heat the sensor tip with a propane torch. When left in the flame long enough to get it hot, the voltage should rise to about 0.9 volts (little oxygen). Remove the flame, and the voltage should drop to about 0.1 volts (detecting oxygen). Flick the flame past the tip and the voltage should switch between the two voltage extremes. By the way, although an analog/analogue voltmeter might make seeing the sensor voltage transitions easier, they (especially inexpensive ones) can have relatively low input impedance, which will load the output of the sensor and give inaccurate readings. It's best to use a meter or scope with a high impedance input.

I think we're getting into semantics, Mike . What I said was ''it's common enough for a marginally faulty front O2 sensor to not trigger a code for itself while still causing trouble'' (emphasis added). Perhaps what I should have said was ''it's common enough for a marginally faulty front O2 sensor to not cause a code to be triggered for itself while still causing trouble''. Of course, it's the ECU's programming that determines when parameters are far enough out of the ''normal'' range to trigger a code. Subaru refers to that as ''detecting criteria'' -- some of them are simple, and some complex, involving calculations using data from multiple sensors under varying conditions. For O2 sensors, a simple example could be that the voltage is too low or high (P0131, P0132), rather than swinging about a mid-point.

There are a few possibilities, so as already suggested, retrieve any trouble code(s). If one for the cam angle sensor shows up, it's likely that rather than the sensor being bad, the timing belt has failed.

While bgambino's problem may or may not be O2 sensor related, it's common enough for a marginally faulty front O2 sensor to not trigger a code for itself while still causing trouble. A multimeter can provide a rough idea what the O2 sensor is doing, and a sufficiently defective sensor might become obvious. However, a marginal one probably won't be easily detected that way. More-sophisticated equipment is usually needed for a reliable diagnosis. Actually, the voltage swing at the front O2 sensor should be much more rapid than that, which is one reason why a multimeter isn't that effective a tool for testing it. Just to clarify things, the voltage around which the swing should occur is about 0.45 volts. Yes, that's an ideal I can second. Unfortunately, some people don't have the knowledge or equipment to do a proper diagnosis (and some are apparently unwilling/unable to obtain them). Also, sufficient experience with certain symptoms may lead to the conclusion that a particular part is likely causing the problem.

As in the first link I posted above, http://www.aa1car.com/library/o2sensor.htm -- see the section ''OXYGEN SENSOR DIAGNOSIS''.

You're welcome. If the system was still under charge, then the drier should be fine. However, if the system was left open to atmosphere for a sufficient time, then a long vacuum application and possible drier replacement could be needed. Static pressure on low and high sides should be equal, so I assume the low also read around 75 PSI (not 25 PSI). Manifold gauges for use with particular refrigerants have corresponding temp scales that can be read to determine saturation temp for the pressure (and vice-versa). Your gauge set apparently is set up for C degrees, so of course you'd have to convert to/from F deg, but you don't need a chart if the temp scale is there. Just to make life easier, I found a chart online with both C and F temps versus vapor pressures for R134a and other refrigerants at http://www.chillers.com/PT%20charts%20for%20refrigerants.htm . The ''rule of thumb'' I gave earlier of course results in a ''ballpark'' figure -- actually, for 70F/21.1C the chart says R134a sat vapor pressure is 71.2 PSI(G). The ratio is a ''ballpark'' estimation of how many times greater the high side pressure should be than the low side when the system is operational. For example (only!), if the low side were 30 PSI, 8 times that would suggest a high side reading of 240 PSI. System load and other factors influence that, naturally. In case there's any confusion, the gauge temp scales, whether in C or F degs, in no way affect the pressure readings. As to the readings you last took, the rapid (5 second) equalization time implies no significant system blockages. With a 65 deg F ambient there wouldn't be much load, but the pressures seem to be dropping more than that would explain. I'd suggest running the system again when warmer, doors closed, windows and hood open, blower on high speed, and recheck pressures. If much lower than what you said in the first post of this thread (25L/200H), it would seem to point to a leak. I'd proceed by verifying whether there is a leak, and of course deal with anything found. Next, if the system had previously been open for any significant time, the refrigerant should probably be recaptured, and a vacuum pulled for long enough to ensure proper operation (need for a new drier depends on length of time system was open). With an evacuated system, adding back the proper charge should be a lot easier. A good shop will have equipment that does an air/moisture purge of the recovered refrigerant so that with recycling those contaminants don't wind up back in the system.

If I remember correctly, a hole in the condenser necessitated its replacement. Since the other work on the car took a bit of time, I assume that the A/C system was exposed to atmosphere for that period. Air and moisture can significantly impact A/C performance, so at the very least a vacuum should have been pulled for about as long as the tech said the entire job took. It's even possible that the desiccant in the drier got saturated, and the receiver/drier should have been replaced. Around 22 PSI, you're likely below the low pressure cut-off switch trip point, which could explain the short cycling. Based on a 10:1 ratio and 35-40 PSI on the low side, it implies a high side pressure of 350-400 PSI. That would be excessive. Under typical operating conditions with R134a, I'd expect the low side to be closer to 28-30 PSI. The high-low ratio, as a rough ''rule'', should be perhaps 7:1-8:1. It's important to keep in mind that the temperature and pressure of the refrigerant directly influence each other, so the ''correct'' pressure varies with temperature (among other factors). Static pressure for R134a, as a ''rule of thumb'' at typical operating ambient temps, is in PSI about equal to the temp in degrees F plus 10%. For example, at 80 deg F, R134a vapor pressure is about 88 PSI. If static pressure at the ambient temp read significantly lower than vapor pressure shown on a refrigerant chart, you could probably assume that the system was undercharged. However, if the reading is as high as expected, that doesn't ensure sufficient refrigerant. The system could have only vapor in it, or anything up to the correct charge (and beyond) with vapor plus a liquid component, and the static pressure would measure the same. (The vapor reaches an equilibrium pressure for the particular ambient temp.) Two questions: 1) After it's been off for at least several minutes, does the system short-cycle very soon after turning it on, or does it run longer at first and then get worse? 2) After running and short-cycling for a while, when the system is shut off how long does it take for the static pressure to reach equilibrium (that is, when both low and high side gauges read the same)?

The title of your thread is ''P0172'', but in the post you mention ''P0170''. Although there's some overlap of possible causes with those two codes (having to do with causes of excessive fuel pressure), knowing which is the correct one could lead to easier diagnosis, so please clarify. Also, can we assume we're discussing a Forester? How did the plugs you removed look -- were the firing tips sooty/blackened, tan/white, etc.? Is there any black smoke from the exhaust? Until we have more info, you could begin by checking for intake leaks and a ruptured fuel pressure regulator diaphragm (pull off its vacuum hose and see if there's any gas in it).

That's right, '96 was the only year that the 2.5L had HLAs. It was also the last year for HLAs in the 2.2L. Unlike 2.2's with screws and locknuts on the rockers, the '97-99 DOHC 2.5 requires selecting shims for clearance adjustment.

The ECU needs data from the front (upstream) O2 sensor in order to properly set the A/F ratio, and the sensor has to respond rapidly to changes in the exhaust. Lessening the sensitivity of that sensor will not only result in poorer performance/mileage -- it will also increase emissions (the catalytic converter will be less effective) and make the generation of code P0420 more likely. The rear (downstream) O2 sensor data is used by the ECU to determine if the catalytic converter is working well. Under normal operation that sensor responds more slowly and to a smaller degree than the front one. Reducing its exposure to the exhaust stream tends to reduce triggering of P0420. Although a ''lazy'' front O2 sensor is a common cause of P0420, it's hardly the only one. By using anti-foulers on the rear sensor, you might stop the P0420, but to the detriment of engine operation. If a bad front sensor is the problem, it's even possible that could reduce gas mileage enough so that eventually a new one would pay for itself in lower gas costs. Here are a couple of good links to explanation of O2 sensor operation: http://www.aa1car.com/library/o2sensor.htm http://www.autoshop101.com/forms/h37.pdf (slightly Toyota-biased) Some info on P0420 -- the second link gives several possible causes of the code: http://www.motor.com/magazine/pdfs/102006_09.pdf http://www.catalyticconverter.org/news/news_page.cfm?Key=catalytic_converter-&News=120

Typical compression pressure (assuming good engine condition and proper procedure) is about 185 psi. Bad rings or valve seating can obviously reduce that. So can the timing being off. If your engine should be closer to 200 psi, then a one-tooth error in the t-belt installation at the cam on the side that's low could reduce it to about the amount you're reporting. I know you've already checked and corrected the timing, but did you go by marks alone or did you do a tooth count? It's usually best to use the marks initially, but verify by counting teeth. As suggested, leakdown test results could be useful. If you didn't try it already, a wet compression test might also reveal something.

If you have leakage at the crank seal, it might be due to the oil pump problem previously mentioned. See http://endwrench.com/images/pdfs/LeakingFrontInfo.pdf .

You're welcome, and thanks for geting back with the results. Please keep us posted (so that I can decide whether to use the Trans-X on my own '99 if it develops the delayed engagement problem ).

Piston slap is a sound generated when the piston has excessive clearance in the cylinder bore. Generally, as the engine warms and expansion occurs, the noise diminishes. As davebugs said, that wouldn't typically be audible for just a few seconds, but rather minutes. For info on what's ''normal'' and what isn't, see http://endwrench.com/images/pdfs/OtherInfo.pdf .

Welcome to the forum. It would be easier to diagnose the problem if we had the trouble code, but the symptoms alone might point in a particular direction. Intermittently not being able to crank the engine when in Park probably means the inhibitor switch is acting up. That can cause other problems, if the ECU doesn't know what gear is selected. See http://endwrench.com/images/pdfs/Diagnostic.pdf .

Replacing any vacuum hose that doesn't fit snugly is a good starting point. The emissions label attached to the underside of the hood should show proper vacuum hose connections.

If the ATF was actually substantially overfilled, the fluid could have become aerated as the trans was running, and that alone might have been the cause of the ''slipping''. However, it's possible that drainback might have occured overnight. Accurately determining ATF level requires that the fluid be hot, the engine running, and the selector moved through all the gears and then back to Park. See http://www.ultimatesubaru.org/forum/showpost.php?p=850614 for details.

I'm not sure we'll ever know exactly what the original weirdness was, but http://endwrench.com/images/pdfs/07FebInter.EW.II.pdf could be a clue for some strange immobilizer behaviour.

In an auto trans, torque bind is caused by problems related to the tranfer clutch and/or ''C'' duty solenoid, and those parts are located in the trans rear extension housing. (Manuals use a viscous coupling.) However, I'm not going to get into the merits of an ATF additive to ''fix'' the problem. The search function will lead to enough info on the topic of torque bind to satisfy almost anyone.