Chevy Astro and GMC Safari Forum banner
1 - 12 of 12 Posts

·
Registered
Joined
·
15 Posts
Discussion Starter · #1 ·
My AC Compressor clutch doesn't engage unless I directly connect it to the battery. It has plenty of refrigerant and I don't believe it is the low or high pressure cutoff switches. Here is the behavior I am seeing.
  • I see voltage across the control inputs on the A/C enable relay.
  • If I jumper the switched circuit on the A/C enable relay socket while there is a diode in D101, it immediately blows the 10A mini fuse.
  • With a diode in at D101, both sockets of the connector going to the A/C Compressor Clutch show continuity to ground.
  • Without the D101 Diode, only the black wire shows continuity to ground.
Is the ground connection I'm seeing expected behavior for a diode or does this sound like a bad diode? Perhaps maybe a ground short somewhere between the relay and the diode? I must admit I am fuzzy on the expected behavior of the diode in this circuit and as I am typing this I am wondering if I got the connections wrong on the relay when jumpering it. I don't think I did but am not going to do it again until I get some more fuses. When I first started testing this, the 10a mini fuse was fine. I did not blow it till I jumpered the relay.

Any suggestions on how to further debug or fix this issue would be appreciated.

Thanks

Product Rectangle Schematic Font Slope
 

·
Registered
Joined
·
22 Posts
My AC Compressor clutch doesn't engage unless I directly connect it to the battery. It has plenty of refrigerant and I don't believe it is the low or high pressure cutoff switches. Here is the behavior I am seeing.
  • I see voltage across the control inputs on the A/C enable relay.
  • If I jumper the switched circuit on the A/C enable relay socket while there is a diode in D101, it immediately blows the 10A mini fuse.
  • With a diode in at D101, both sockets of the connector going to the A/C Compressor Clutch show continuity to ground.
  • Without the D101 Diode, only the black wire shows continuity to ground.
Is the ground connection I'm seeing expected behavior for a diode or does this sound like a bad diode? Perhaps maybe a ground short somewhere between the relay and the diode? I must admit I am fuzzy on the expected behavior of the diode in this circuit and as I am typing this I am wondering if I got the connections wrong on the relay when jumpering it. I don't think I did but am not going to do it again until I get some more fuses. When I first started testing this, the 10a mini fuse was fine. I did not blow it till I jumpered the relay.

Any suggestions on how to further debug or fix this issue would be appreciated.

Thanks

View attachment 285433
Have you tested the diode? They allow dc current to flow in one direction only. If current is flowing in both directions, it is bad.
That looks to be one problem, as it doesn't appear that with the diode in place, both sockets of the connector going to the A/C Compressor Clutch should show continuity to ground.
I'm not certain that is what is causing you to have to jump the relay in the first place however. That too, may be bad.
 

·
Registered
2000 Lifted 4x4 Astro 92 V8-350 Shorty
Joined
·
5,998 Posts
The diode is merely there to absorb the big SPIKE created by the clutch coil (collapsing field) when it is disengaged. This dampens the spike that would typically back-feed into the rest of the electronics. The system should still run without it (but needs it only for protection)

When tested, the diode will show continuity ONLY one direction, but not the other.
If it is blown.. it will short the system.

You should also be able to "temporarily test" the system without it.
If it's bad, you definitely want to replace it.
 

·
Registered
1998 LS AWD Forest Green metallic
Joined
·
1,323 Posts
Minor semantics issue...
A diode that is 'blown' usually means 'open circuit'.. That should not cause a short, to blow a fuse.
A diode can fail in a 'shorted' mode, and could cause a fuse to fail, but it is not nearly as common.
Rod J.
Issaquah, WA
 

·
Registered
2000 Lifted 4x4 Astro 92 V8-350 Shorty
Joined
·
5,998 Posts
Incorrect.. when diodes blow (in this type application), they typically pass signal in both directions (no longer functioning as a one-way device). This would suddenly pass signal to ground.
This diode (in this application) does not pass current (it is not in series with the supply voltage). In this spike dampening application, the failed diode would most likely become a short.

Simple test.. pull diode and see if A/C engages.
Or actually test the diode with a VOM meter.
 

·
Registered
1998 LS AWD Forest Green metallic
Joined
·
1,323 Posts
MM,
I will defer to you. I agree with most of what you are saying.
We use diodes regularly in industrial control circuitry on nearly all DC powered relay coils ( same principle as in the AC clutch circuit) .
A properly functioning diode in that circuit does need pass some small amount of current when the colapsing magnetic field creates a voltage spike, otherwise it could not do it's job of dampening the spike from the collapse of the magnetic field when the clutch is deactivated.
Sorry, I just don't like the word 'blow' in this context. Most folks think of a blown fuse when the term is used and that does not fit well with diode applications, being able to fail either open or shorted.
So if you say, "...when diodes fail (in this type of application, they typically short, so they conduct in both directions....), I think that is more easily understood...again it is just semantics.
Rod J
 
  • Like
Reactions: Mmusicman

·
Registered
Joined
·
15 Posts
Discussion Starter · #7 ·
Progress but it still isn't fixed. The diode is bad. It shows continuity between the pins no matter which way I test it. I am using the continuity setting which beeps when the probes are connected. The diode in the slot above shows no continuity either way so that is a bit perplexing. Either they both are bad in different ways or I need to learn more about testing diodes. My local Napa does not have replacements so at the current time I have just removed it and stopped blowing the 10a fuse.
  • I found that the connection to the compressor clutch is flakey (maybe the ground). With the relay jumpered I can get the clutch to engage by jiggling the connector.
  • I found a position that had the clutch consistently engaged with the relay jumpered but when I swapped the relay back in, the clutch does not engage even though the control circuit on the relay shows 13.1V.
  • With the plug not connected to the clutch my DMM shows 13.1 volts at the plug with the relay jumpered and fluctuating millivolts with the relay in place.
  • I replaced the relay with a new one bought at Napa and it still doesn't work.
  • I'll have a new plug for the clutch and 2 replacement diodes tomorrow. The diodes I found on amazon.

Is it possible the relay is "noticing" a poor ground or lack of diode on the switched circuit (that my DMM does not) and refuses to connect it? Any other suggestions on next steps?

Thanks

Circuit component Gas Control panel Cockpit Electrical wiring
 

·
Registered
2000 Lifted 4x4 Astro 92 V8-350 Shorty
Joined
·
5,998 Posts
It should ALSO be noted it appears the diode CAN accidentally be plugged in backwards.
It appears there is NO obvious visual polarity direction.
If it was plugged in backwards, it would have blown (shorted) it as well.

  • Again, the system WILL actually run without it (although it should be for testing only).
  • I tested with my diode removed and it worked fine.

It's certainly possible with the blown diode shorted to ground it may have also damaged the relay (or possibly something else). Make sure fuse to this circuit is good. Normally try swapping relays. If relay is not at fault, you'll have to do some trouble-shooting.

This diode is not normally carried by many LAPS.. but it is available on Amazon/Ebay.
I think it's rating is 1amp (nominal) since it does not carry any continuous current.
Be sure to plug new one back in correctly (polarity--see photos below)
Electronic component Rectangle Machine Circuit component Plastic

Rectangle Font Lego Gas Toy


Don't forget (according to schematic) the "A/C REQUEST LO" to compressor goes through the low-pressure switch on the accumulator/dryer. This connector can be jumped for testing purposes.
 

·
Registered
Joined
·
15 Posts
Discussion Starter · #9 ·
Thanks for the info and pictures.

I agree that it is hard to tell which way to plug the diode in. The white graphic on the back does seem to denote a direction for the diode but there is no indication in the fuse box as to which way to put it. The plastic body of the diode is keyed to the socket so that you cannot seat it the wrong direction but you can make an electrical connection without seating it. Probably best to pull the relay before inserting it if you don't remember which way it goes. That may very well be how mine got blown.

Took me awhile to figure out which diodes I needed to buy. Finally noticed a very hard to read part number molded into the back. GM #12135037. That is how I located them at Amazon. I actually took the van down to parts store without either diode installed so I can confirm that it runs without them.

Thanks again, hoping to get this fixed soon. It's supposed to be 98 here on Tuesday.
 

·
Registered
2000 Lifted 4x4 Astro 92 V8-350 Shorty
Joined
·
5,998 Posts
Again it WILL run without the diode. (for testing purposes)
If it's not working without it, then something else is wrong too

PS: Good to know the fuse block is "keyed" so it can't be put in wrong.
Also, good job in finding the elusive part number and replacement part
 

·
Registered
Joined
·
15 Posts
Discussion Starter · #11 ·
Progress, but not working yet. This has me really confused. I think I need to reassess some assumptions.

I've replaced the diode and the connector on the ac compressor clutch. I found this resource for testing diodes for those interested.

How to Test Diodes with a DMM

I have tested the diode and relay while not installed in the fuse box and both are fine. I have tried both a new and the existing relay. When I monitor the control voltage without the relay inserted and jump the switched contacts, the clutch engages and the control voltage stays at 13.1 volts. The moment I insert the relay, I get zero voltage in the control circuit. There is voltage across the hot side and battery negative, just not the ground contact for the relay.

I know the VCM switches the ground depending on the state of the high and low pressure switches. If this is what is happening then it seems that the VCM only switches the ground when the relay is inserted. I have tried disconnecting and jumping the connector for the low pressure switch but it makes no difference. I have not done so with the high pressure switch. Can this be tested the same way?

I have the service manual and it has a series of troubleshooting steps involving testing voltages of various contacts within the VCM connectors . I have not done so yet. Can anyone give me suggestions for unplugging the VCM plugs, they seem not to want to be removed without some force and I do not want to damage them.

Any suggestions are appreciated.
 

·
Registered
Joined
·
22 Posts
If I'm reading what you are doing right, remember voltage is an expression of the potential difference in electromotive force. This is measured in an open circuit. When you insert the relay, you should have a closed circuit, thus current running through it, negating a voltage at that point. At least so long as your air is turned on. If it's off, you should again have a voltage.
Looking at the schematic you posted, it shows the Low Pressure Switch to be "closed" as well as the High Pressure Switch. If you trace the route of current, the HPS would need to be closed to allow current to travel from Power Distribution Cell 10 in the fuse block, through the HVAC controller, through its self, to the VCM, to the relay, to the under hood fuse, back to the relay power switch, which would allow to go to the clutch, then to ground at Ground Distribution Cell 14
If the LPS is also closed, then if you trace the path, it goes directly to ground from the VCM, to Ground Distribution Cell 14.
Maybe that's what's throwing you off? I'm not entirely sure since it's pretty vague on the path of current through the VCM. The LPS could be the path to ground for current coming from the under hood fuse box, passing through the a/c enable relay.
Verify which is correct.
As far as connectors go, if they're warm they're softer and less likely to break, I've used small screwdrivers to push the lock tabs down and sometimes had to pull with pliers on the connectors, not the wires, and squeeze gently. Extra hands help if you have them.
Also, if you have good sharp probes on your multimeter, you can often insert the probes behind the connector, where the wire goes in at, without pulling them apart. If you test this way, be sure to test both sides of the two connector halves for each wire tested. (I hope that makes sense.)
I don't know if any of the previous 2 suggestions on connectors will help if you are really trying to work on the VCM.
 
1 - 12 of 12 Posts
Top