Runaway Compressor Clutch

[1987 GMC Jimmy]

Problems, folks. My compressor clutch clicks on the instant you turn the key and cycles incessantly. The ECM burned again, albeit not completely, and the car runs okay after fixing my rat damage. I noticed a few days ago the clutch would not come on, and just a minute ago it was clicking so fast it sounded like a machine gun. I know there's wire that goes from the compressor to the ECM, and I was wondering if this erratic problem could have somehow overloaded the motherboard and fried the computer. It's seemingly the only thing that's affected because even though I cannot interface and get codes or live data, my sensors and mixture control work, and the car does not smell or act like it's in limp home mode. So in a nutshell, is this a feasible postulate, do I test for continuity and fix a potential break or search for a good ground? I'm lost here, and I can't keep buying these ECMs. Thanks.

 

[foamypirate]

Problems, folks. My compressor clutch clicks on the instant you turn the key and cycles incessantly. The ECM burned again, albeit not completely, and the car runs okay after fixing my rat damage. I noticed a few days ago the clutch would not come on, and just a minute ago it was clicking so fast it sounded like a machine gun. I know there's wire that goes from the compressor to the ECM, and I was wondering if this erratic problem could have somehow overloaded the motherboard and fried the computer. It's seemingly the only thing that's affected because even though I cannot interface and get codes or live data, my sensors and mixture control work, and the car does not smell or act like it's in limp home mode. So in a nutshell, is this a feasible postulate, do I test for continuity and fix a potential break or search for a good ground? I'm lost here, and I can't keep buying these ECMs. Thanks.

I'd be checking the wiring from the HVAC head, through the low pressure switch, to the A/C clutch. May have a break and vibration is causing it to turn the clutch on/off. The wire to the ECM is just so it knows to kick up the idle to compensate for the A/C load. It's possible the rapid cycling could be detrimental.

 

[1987 GMC Jimmy]

I'd be checking the wiring from the HVAC head, through the low pressure switch, to the A/C clutch. May have a break and vibration is causing it to turn the clutch on/off. The wire to the ECM is just so it knows to kick up the idle to compensate for the A/C load. It's possible the rapid cycling could be detrimental.

If it helps, my fan selector doesn't do max fan speed. Just the first three. I already know the HVAC controller has to come out. Where is the low pressure switch?

 

[yevgenievich]

If it helps, my fan selector doesn't do max fan speed. Just the first three. I already know the HVAC controller has to come out. Where is the low pressure switch?

It is mounted on the dryer. Signal from the ac controls comes to it, then follows around the back of the engine where it splits to tell the ecm it is on and continue to the ac clutch coil. The harness behind the engine is prone to damage if it gets loose. Mine would periodically short and blow a fuse untill i tied it away from rubbing area

 

[1987 GMC Jimmy]

I traced the wires out of the compressor. One's a ground, and it's solid. The other goes into a pressure switch mounted on the steering box. It doesn't make a whole lot of sense, but if that's it, then that's it. Whatever it is, it's half ass wired. It was taped, glued, and barely on there.

 

[1987 GMC Jimmy]

Here it is. @yevgenievich

 

[yevgenievich]

Is that on your 87 full size? That pressure switch would be used to bump up the idle when at full lock, but i did not think it was on the squares. Double check the wiring, as it does not make sense for ac signal to go through power steering pressure switch.

On the left in this picture is the ac pressure switch with plug unplugged (two green wires)

You must be registered for see images attach

 

[1987 GMC Jimmy]

Is that on your 87 full size? That pressure switch would be used to bump up the idle when at full lock, but i did not think it was on the squares. Double check the wiring, as it does not make sense for ac signal to go through power steering pressure switch.

On the left in this picture is the ac pressure switch with plug unplugged (two green wires)

You must be registered for see images attach

Nah, this is on my Caprice. At the moment, I'm two and a half hours from home without A/C and heat indices in the one hundred teens so I'm trying to get this fixed as soon as possible. The compressor is an aftermarket Delphi R134a compressor. Believe me, I checked several times. It has two wires coming out of it, and the connector looks like it's bridged with a diode in the middle. I've never seen anything like that. Anyways, one goes to a ground and the other one goes to the switch I showed a picture of. The wire that comes out of the switch goes to the low pressure switch on the accumulator, and then those go off to the HVAC controller and what not. I traced the whole shebang. I don't know what the point of plugging the compressor into a power steering switch is because all it's doing is just constantly clicking on and off, but I see the best way to remedy this, since I have the wiring already set up, is to pop the cap off the back of the compressor and install a high pressure switch, and plug that into it. I realize that I'll have to evacuate the refrigerant, but the o rings are leaky at the compressor anyways so two birds with one stone. I know I could just bypass directly to the low pressure switch, but I feel that it would be a safer, more modern touch to add the new switch since it does operate at higher pressures. What do you think? And where does the power steering pigtail come from because there doesn't seem to be anything else in that vicinity.

 

[RetroC10Sport]

The Firebird harness I am using in my truck has the PS switch, but I did not use it in my truck as I'm somewhat sure it's not needed since I used the truck's A/C harness (except where it goes to the Firebird's ECM). My A/C seems to work fine.

 

[yevgenievich]

Looks like power steering and ac use the same signal wire to trigger idle change with ecm. Diode would serve to isolate the two. I would disconnect the power steering switch for a test and then check/replace the diode. High pressure switch should not be needed if compressor has overpressure protection already.

 

[1987 GMC Jimmy]

I just don't understand how that power steering signal can be valid for the AC. The clutch cuts on and off from the moment you turn the key on to the moment you turn it off. That's with the air completely off. Wouldn't the appropriate reference make a difference in the function of the air conditioning, or should I look for something else all together to be messed up like a faulty low pressure switch or something? I can't conceive anything else sending additional power to the ECM because everything else is functioning properly and no fuses are popped.

 

[yevgenievich]

Ecm has limited number of inputs, so probably when gm needed to bump idle up with full lock on power steering, they used the same wire for both. Diode seperates the two circuits. The diode likely failed or there is another short somwhere. Also the pressure switch on power steering likely failed too

 

[1987 GMC Jimmy]

Ecm has limited number of inputs, so probably when gm needed to bump idle up with full lock on power steering, they used the same wire for both. Diode seperates the two circuits. The diode likely failed or there is another short somwhere. Also the pressure switch on power steering likely failed too

Okay, the wiring looks good so I'll wire a legitimate pigtail to the switch and see. I don't know how to test a diode unless there's some resisitivity procedure. Same with the switch. If I can crack out the voltmeter/ohmmeter and test everything, I'll be more than glad. Just to be safe, is there a procedure for the low pressure switch? The wiring looks good. I've got some dummy lights and a little plug or two that are a dead end, but all those wires look fine under the hood. @yevgenievich

 

[theblindchicken]

With a diode, wouldn't you be able to test the resistance through it in two directions?

If it flows the correct path (positive => negative) it should read out the resistance, but if you reverse (negative => positive) the leads, it should be an open circuit.

If you have similar values in both directions, the diode should be replaced.

 

[yevgenievich]

multimeter will have a diode test, but it could still be failing and not register on the meter. It is designed to flow current only in one direction, until the breakdown voltage is reached. Just disconnecting the power steering pressure switch from ac wiring would be the sure test. Low pressure switch on ac system would not cause the clutch to cycle if ac is actually off.