A/C relay won't kick on

hear

Can't type
Original poster
Lifetime Supporting Member
Joined
Oct 18, 2020
Messages
6,054
Location
Prosper, TX
2000 TJ, we've done every diagnostic we can come up with, including following the troubleshooting guide in the FSM. Relay works, jumping the relay turns on the compressor, it blows cold so everything is good there. We've got continuity from the HVAC switch/computer/relay/hi&low pressure switches/etc. Finally stumbled into the fact that if we move the harness just the right way, we can make the compressor kick on. Specifically tugging on the low pressure switch pigtail where it joins into the main harness along the firewall makes it go on & off.

Now we've got the main part of the harness out of the jeep and torn apart, but I can't find any problematic wiring, at least on first pass visual inspection. I've had several harnesses out because of a suspected wiring problem and every time I've found multiple bare wires which were obviously the problem. I'm not intentionally neglecting anything but I've focused on the light green wire, which is the "a/c turn on wire" and the blue/white & blue/orange wires on the low pressure switch but haven't found anything yet.

I guess I'm looking for tips & tricks for finding a short...I'm normally really good at this but every challenge is unique.
 
First thing I would check is the harness inside at the control panel. TJs had a problem with them melting the connectors.

The switch was the first thing my dude replaced. But as I said, the switch isn’t the problem, it will only kick on if I tug on the harness in the engine bay.
 
Sounds like you may have a “stretchy” wire for the LP switch circuit. Insulation is in tact but the copper is broken inside.

so far haven't been able to find it, although it almost has to be isolated to about a 12" section of the harness. If I can't find it I may just replace the entire wire, although if that doesn't manage to fix it it's going to be extremely annoying to put the harness back in only to find it still not working.

Seems like a continuity test while tugging on that wire should reveal the issue but so far no luck.
 
  • Like
Reactions: Rickyd
Is there a splice inside the section of harness? That would seem to be the most likely spot for a break in the conductor. The FSM lists splices...
 
there is a splice on the light green wire which comes off the HVAC switch and goes to the high pressure switch + PCM, but guess I need to understand the exact wire routing to see if that's in the area of the problem. We got the harness out and opened up, but I don't recall if we saw that splice or not. I always supsect those factory splices as a potential problem but thus far it's never been that. Probably because those factory splices are way better than mine, lol.
 
THAT is a useful tool! Have you used it? Probably would be challenging to use now that we have the harness out...

Not that particular one. I have used them on both AC and DC circuits. My most challenging situation was that I had a break in an underground line powering my well pump house. The line is 200' long, and I was not looking forward to digging up the entire 200'. Using something similar, I pinpointed the break, dug down, and the break was right there. I was surprised it worked so well! The DC one I have is set up for phone and network cables, but can be used for this with a little adaptation. I don't see why you couldn't use it with the harness out, but it will only work if the open circuit is constant. I'm sure there are YouTube videos showing how to use it.
 
  • Like
Reactions: Rickyd
well I figure it's detecting the current in the wire, but if it's all unplugged there won't be any current. I guess if there's a short there won't be any current either. I guess I have no idea how this works then. I'll have to just buy it.
 
The simple explanation: it turns the conductor you're testing into an antenna and the probe is a radio receiver. You have to place the probe tip very close to the conductor to read the radio signal, and when you reach the break, the probe stops receiving the signal.
 
Well however it works it will be here tomorrow.
 
there’s a short in there somewhere. I can’t find it. When it was in the jeep, pulling on the low pressure pigtail would cause the compressor to kick on. Low pressure switch consists of the blue/orange & blue/white wires. I can tug & push & twist the harness but I can’t make it break continuity on those wires.

The other wire that factors in is the light green wire which is essentially the “turn on” wire. I can’t make it break continuity either, from the big gray plug all the way to the high pressure switch.

There is a splice, but it’s not in the section of the harness I have here on the table. It’s in the section that runs along the top of the firewall and goes over to the gray PCM connector. That section of harness is factory zip tied and all out of harms way, I don’t know how it could’ve been degraded, but I can’t find it in this section.

57A06EAF-ECF1-4A74-A49B-FF3B5969EF06.webp
 
Those are junk. They turn the whole wire bundle into and antenna.

Unplug the battery. Unplug the the pig tail that's causing your break. Short the suspected broken wire to ground. Remove the relay, with a test light connected to battery positive, touch the female pin (where the relay plugs in). The light should light. If it doesn't, start wiggling. When you find your intermittent break, your light will flicker.

You can do the same thing with a multimeter set to beep on continuity. That beep just annoys the piss out of me.

Edit: Good lord. I didn't realize you pulled the harness for this.
 
Unplug the battery. Unplug the the pig tail that's causing your break. Short the suspected broken wire to ground. Remove the relay, with a test light connected to battery positive, touch the female pin (where the relay plugs in). The light should light. If it doesn't, start wiggling. When you find your intermittent break, your light will flicker.

I'm all about a more efficient approach. So take a probe and stick it in the cavity I think is the problem, and then ground it. Test light to battery, and the other side to Which pin on the relay are you thinking? Because the circuit I believe is the problem doesn't go through the relay. My (current & potentially flawed) understanding is that the A/C Select Input tells the PCM that you want the A/C on, and then the A/C Switch Sense tells the PCM if it's ok to turn on, and if both conditions are met it will fire the A/C Relay Control. The section of harness that I'm tugging on to get it to work is the Low Pressure switch, which isn't part of the relay circuit except through the PCM.

I'm not saying your method won't work, but it's not clear to me how it would work.


1686201864525.webp



Edit: Good lord. I didn't realize you pulled the harness for this.
99.99% I was going to need to remove the harness to actually be able to repair whatever I find, and I've reached the point where I would rather disassemble stuff than fight with it and end up doing a lousy job.
 
Have t you checked the connector itself? I’ve seen them corrode at the wire termination end to the point where the crumble when you touch them. That is typical hidden behind a seal and pressed into the housing, so it’s difficult to see without pulling the pins on the connector body.
 
  • Like
Reactions: Rickyd
I'm all about a more efficient approach. So take a probe and stick it in the cavity I think is the problem, and then ground it. Test light to battery, and the other side to Which pin on the relay are you thinking? Because the circuit I believe is the problem doesn't go through the relay. My (current & potentially flawed) understanding is that the A/C Select Input tells the PCM that you want the A/C on, and then the A/C Switch Sense tells the PCM if it's ok to turn on, and if both conditions are met it will fire the A/C Relay Control. The section of harness that I'm tugging on to get it to work is the Low Pressure switch, which isn't part of the relay circuit except through the PCM.

I'm not saying your method won't work, but it's not clear to me how it would work.


View attachment 430871



99.99% I was going to need to remove the harness to actually be able to repair whatever I find, and I've reached the point where I would rather disassemble stuff than fight with it and end up doing a lousy job.

Forget the relay then. Everything is unplugged, so pick one side of the leg to shoot power through and the other side to ground so all you’re doing is making a loop. You’re turning the unplugged wire into a ground so a test light will light.

For instance, put power to C104 and ground C170. Then get to wiggling, tapping, stretching, etc. When the light flickers or goes out, you’re onto something. If that doesn’t work, put power to C170 and ground C3/C23. Rinse and repeat.

Don’t use an LED test light. Use an incandescent bulb. Either a test light or rig up a headlight. You want to load the circuit. Multimeter continuity checks will get you into trouble, because a wire with all but one broken stand will pass a multimeter continuity check, but the light bulb won’t light because that one strand can’t carry enough current.

Also, don’t jam your test light into any female terminals. Based on your previous posts, I doubt you have any jumper wire sets. You’ll need to either backprobe or something. You’re going to need to rig this up. Don’t spread these terminals or you’ll hate your life.
 
Last edited:
Forget the relay then. Everything is unplugged, so pick one side of the leg to shoot power through and the other side to ground so all you’re doing is making a loop. You’re turning the unplugged wire into a ground so a test light will light.

For instance, put power to C104 and ground C170. Then get to wiggling, tapping, stretching, etc. When the light flickers or goes out, you’re onto something. If that doesn’t work, put power to C170 and ground C3/C23. Rinse and repeat.

Don’t use an LED test light. Use an incandescent bulb. Either a test light or rig up a headlight. You want to load the circuit. Multimeter continuity checks will get you into trouble, because a wire with all but one broken stand will pass a multimeter continuity check, but the light bulb won’t light because that one strand can’t carry enough current.

Also, don’t jam your test light into any female terminals. Based on your previous posts, I doubt you have any jumper wire sets. You’ll need to either backprobe or something. You’re going to need to rig this up. Don’t spread these terminals or you’ll hate your life.

Thanks, that technique makes sense I hadn’t considered putting power through the unplugged harness.

And I’m not sure what made you think I was a Neanderthal, but I do have the right sorts of test probes to not damage the connectors and back probe when necessary.
 
Update: since we already had the harness out, I went through it with a fine tooth comb but couldn't find anything wrong in the harness. In the interest of making progress we put the harness back in for more integration tests, and lo and behold, A/C worked 100%. I hate it when you don't really do anything to fix a problem, because deep down I know it's not fixed. But it's his wife's jeep so for the time being she's satisfied that she has A/C as it's starting to get balls hot here.