When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Resurrecting this old thread to post the trouble shooting I've done for my non-engaging compressor clutch. I'm lead to believe it's the relay in my head unit. Since I'm not the best at electrical testing, I just wanted to get opinions if I should test further, or just go ahead and try replacing the relay in the head unit (or buy a reconditioned head unit.)
Symptoms - AC clutch not engaging when ignition is on, AC button pressed on head unit and sliders are in appropriate positions. No cold air.
- Recently had leaking AC hose and expansion valve replaced by a shop, along with new receiver drier. They recharged with R12. No other known leaks in areas that are accessible. Evaporator leak is always a possibility, but I understand it's not extremely common.
- Drove car 3 different times since the leaks fixed and system recharged. AC was nice and cold. One day this week, started the car, and suddenly no cold air. It wasn't a gradual degradation of system performance - just very cold one day, blowing totally warm the next day.
- I do have an appointment to go back to the shop this week to find out if the R12 charge is holding. I don't have gauges, so can't determine that myself.
Testing done - Since the car is going on 30 years old, I know it's quite possible the recent repairs are fine and something else has simply decided to die a week after the fact. With that in mind I did the following tests.
- Checked for 12 volts from the lead that goes to the AC compressor. I disconnected the compressor wire from connector housing. Hooked positive lead of the multi-meter to the power wire. Grounded the negative multi-meter lead on one of the cross-brace bolts (nice and clean, bare metal).
Turned ignition on, AC button on and read no voltage.
- Next, tested the clutch directly. Wired the compressor directly to the 12 volt jump post inside the engine bay. Clutch engaged just like it should.
- Freeze switch test: Disconnected the 2 wires from the switch. Hooked multi meter to switch to measure resistance. No resistance. I believe that is good.
Tested the two wires that go to the freeze switch for power with ignition on and AC button engaged. Each wire was tested by hooking it to the positive lead of the multi-meter and the negative lead of the multi-meter hooked to the cross-brace bolt.
One wire did measure 12 volts. The other wire didn't read any voltage.
Wally's HVAC troubleshooting document says if one of the wires shows no voltage, it's most likely the head unit relay is bad.
At this point, do you think I should go ahead with my mechanic appointment to confirm the R12 charge is still holding, and then test the pressure switch if freon level is good, or just pull the trigger on installing a new head unit relay myself (or get another head unit).
If those connectors are taken off freeze switch then you will see 12V on hot side only with AC on - when attached to switch and normal no ice condition the switch is shorted (as you measured previously showing no resistance) and then you get 12V on terminal on AC clutch side (it goes to low pressure switch first though).
Measure voltage with connectors attached at the low pressure switch that is inline with clutch. If you have lost pressure then switch is open circuit and no power passed to clutch (I think fans stay off too). If pressure is ok then measure voltage at hot side of freeze switch with load attached eg headlight bulb. The head relay can look ok under no load but fail to pass current under load conditions. Ditto freeze switch which was my problem.
Thanks for the reply. I was testing something incorrectly. Did the following today...
- Re-tested the freeze switch properly. Got 12 volts from both terminals.
- Tested freeze switch with a headlight bulb hooked up to add load. Light was connected to the outgoing freeze switch terminal and grounded to cross-brace bolt. With light lit up, I measured 12 volts at the hot terminal of the freeze switch.
- Measured voltage at low pressure switch. Did receive 12 volts from one terminal. I then disconnected the two wires, bypassing the switch, and jumpered them together. The clutch did engage.
At this point, I guess there's no more diagnosing until I know the state of the freon charge. Fingers crossed that's it's not a leak.
At least you know the head relay and freeze switch are fine. It looks very much as if you have a leak somewhere and low pressure switch is opening.
When you bypass the switch so compressor runs, what do you see through the sight glass of the drier (after wiping clean)? With enough refrigerant you should see it flowing with maybe occasional bubble, but lots of bubbles when you turn AC off. If you see lots of bubbling when running then refrigerant is too low.
Site glass is nice and clean since since the receiver/drier is brand new, but I didn't actually start the engine & run the compressor with the low pressure switch bypassed. Was worried about damaging the compressor if the freon is low. Just tested to confirm clutch would engage when switch is bypassed.
How long can the compressor "safely" be run with low freon order to observe the freon flow in the glass - 30 seconds ? a minute...?
That's a good point Tom on running low - will defer to real AC experts here! Logically it can't be fatal to compressor if some refrigerant in system, or the sight glass test would not be used! But I've no idea what the advice is if you've lost all the refrigerant
Could be the clutch on the AC, subject to wear and it shows up first when it is hot.
To check, with the engine off, run 12v to the AC, you can run a jumper wire from the jump post. Then try to turn the clutch, the piece outside of the pulley. If it turns the clutch is open and definitely a . If it doesn't turn put a bar on it and try to turn it, not a 36" breaker, just enough to get some torque on it, if it turns the clutch could be a problem, if it doesn't turn, look else ware.
If the clutch is a problem, check if it moves when the 12v is applied, if it does thane you may be able to fix it by changing the spacing shims, I haven't taken a 928 clutch apart but if it is like most, remove the bolt in the end, remove the clutch, there should be a shim washer, replace it with a narrower one. After reassembly check that the clutch spins freely without 12v applied, if not the new shim is too narrow.
Those low-pressure switches get replaced a lot more often than they fail. Likely cause of the open safety switch is low pressure. And kudos to you for testing the loop with the spare headlight bulb as load. Doing that does give you valid positive results on the relay and freeze switch.
Didn't check to see if the clutch can spin or not once I applied the 12 volts directly. Merely checked to see if it snapped in place..assuming low freon since the 12 volts isn't getting to it in the first place unless the low pressure switch is bypassed. Since it's been mentioned, doesn't hurt for me to see if the clutch does grip when it's engaged.
Bob - Yeah, it probably is a leak. Now I just have to hope it's related to the work the shop did recently. If that's the case, they'll fix it. I already spoke with the owner. There's die in the system, so they should be able to spot the leak.
My worry is if it's not related to their work - especially if it turns out to be a leaking evaporator.
Between the cost of having a bad compressor replaced 3 years ago and having this recent work performed by a shop, I could have probably bought equipment to do all this stuff myself.
It's pretty easy to justify buying and/or renting the tools to do your own AC work if your only other option is a $pecialty $hop. It helps a lot if you have some education and experience with how everything plays together.
It's common for shops to find that "the evaporator is leaking" if they stuff a sniffer probe in the vent. They get to charge big dollars for the dash R&R for access. Reality is that evaporator leaks are only sometimes a problem with that symptom. The only true test is a pressure test, something that can be done with a gauge and a couple fittings in a fabricated block-off plate for the hose side of the expansion valve, at least on the later cars. If they put dye in the system, you can get a jump on their diagnosis with a good UV light, a darkened workspace, and maybe some glasses. Not a bad investment, and good for other cars too. Most folks put the dye in the system, have leaks but never see them because they are looking for a purple trace in daylight.
07-26-2015, 07:10 PM
Now I just have to hope it's related to the work the shop did recently. If that's the case, they'll fix it. I already spoke with the owner. There's die in the system, so they should be able to spot the leak. 
