Threw a spare radiator in the civic and it’s back up and running again.
Instead of just going ahead and fixing the passenger door window, I went and tinkered with the AC again for some reason. As if a 23 year old compressor will magically start working right
Started by getting my roommate’s help locating and attaching gauges. The low side valve appeared to be leaking, so we used one of those “airlock chamber” tools to replace it.. And the replacement appears to be leaking slowly too. Anyway, both sides read about 190psi, which we thought was pretty high. Went to turn on the AC with the engine running and.. nothing. AC compressor won’t engage. We test the relay, and then confirm 12V at the little plastic connector by the radiator. It has 12V, so looks like the compressor is shot.
After some searching I learn about the thermal cutoff protector, and went back out to test that. It appear to work fine: 0.1 ohm connectivity. Tested the clutch coil and it came back 3.4 ohm, which is in spec from the service manual. Next I cleared the power steering pump away from the top of the compressor, and then wired in a bench power supply to the compressor clutch coil. Set it to 12V and it started pulling 3.8 amps (which is appropriate for 3.4 ohm coil resistance). At first it looked like the clutch was totally dead, but after some fiddling I noticed that it was actually engaging but seemed pretty weak.
So I gave up and brought the compressor up to the top of the engine bay to take the clutch assembly apart.
It looks to me like the clutch material is pretty much worn through.
And then cleaned the clutch faces and put it back together. Now when testing with the bench power supply the clutch grabs hard. I put the compressor back on the engine and reattached the belt.
Next I took a look at the AC radiator at the front of the engine bay. About 50% of the fins were bent sideways and likely blocking all or most airflow. I took a dental pick and manually straightened most of them out:
With that out of the way I tried doing the factory AC test again. Set the idle to 1500rpm and left the AC running for 10 minutes with a thermometer/humidity sensor by the vent. Here are the results:
If I filled out that chart right (and I have no reason to suspect that I did) then it looks like the system is low on refrigerant. I tried putting a little more in, but didn’t notice any real difference in the cold air coming out of the vents.
So that concludes the AC adventure for today. I have something good enough to manage condensation when it rains, but not strong enough to keep the car cool on an Alabama summer day.