This weekend I did a compression test on the '92 (120k mile) engine, and I didn't know for sure how to do it - so I did it a bit wrong. First I didn't turn off the fuel by pulling the FI fuse. Second, I didn't open the throttle.

However - my results were roughly 250 across the four, and if they seem a bit high (or hella high), then see the second reason above. The first cylinder was actually 220 psi, and after i poured a cap of oil in, it went back up to 250.

I will be re-doing the test later on, but shouldn't it be about the same, lower but at the same ratios? Also - with one cylinder being low, and the oil curing it temporarily, I want to do a leakdown test of at least that cylinder. I have read to pump it up and listen at the intake, the crankcase breather, the exhaust, and the radiator for the air escaping. Are those the correct places to pinpoint the leak, and also what are some other symptoms I would see of leaks (ie. rings leaking, valves not seating, headgasket leak).

 

*poke*

 

yes 250 for a (stock) prelude is "hella" high. The helms maunal says that avg is supposed to be in the neighborhood of 185psi for H22. I get about 190 psi with approx 11.3:1 compression for my H22. That might sound low but I am also at 5500 feet elevation so I read about 20% lower than at sea level. I guess that since you drive a '92 that you have an H23? Keep in mind that your cam selection also affects your results.

Not opening your throtle should result in lower numbers not higher.

Something that you can try is starting you car, drive around the block to get the oil flowing and park it so that the engine is not too hot, then test. I have noticed some odd numbers after letting my car parked (for a couple of weeks) and there was a large variance in the readings. I guess that my piston rings were not well oiled. after driving around the block they ended up fine.

I have also noticed that a long roadtrip burned out a bunch of my carbon build up from in city drivingm, thus lowering my compression results.

the Pirate