AC Cutting Out-- Long

[landtank]

Bill, he got the testing by unplugging the sensor from my post. If I'm reading the schematic correctly the switch is normally open. I've always assumed that in the wiring diagrams, the devices are shown in their normal state. That switch when connected energizes the engine coolant cut relay which is normally closed, again acording to the pic in the diagram. I've never had reason to actually meter this circuit but the circuits that I have, have been accurately portrayed in the diagram.

 

[Beowulf]

Anyone got an EWD for a '93 ? It would be helpful to Jared if they could post a scan of the circuits that Rick discussed; especially the coolant temp cut-out circuit. The alternative is to confirm that the '93 EWD circuit is the same as Rick's 96 and then maybe Rick could post a scan.

-B-

 

[Photoman]

I just went and checked the sensor and it is open. Rick is correct and I was the one that remembered it backwards. Sorry for the misinformation.

Bill

 

[sjpitts]

--NEW INFO--

I did some more tests. I started the 80 up this morning and let it idle with the AC on. The engine was cold and it was probably in the mid to high 80's. The engine idled for less than four minutes before the AC cut out.

I then pulled the AC coolant temperature sensor connector-- and the AC compressor started back up (it is a normally open switch). Plugging the connector back in caused the AC compressor to again cut out.

So clearly the cause of the compressor cutting out is the AC coolant temperature sensor. The next issue is whether or not it is actually overheating or if the coolant sensor itself is bad.

Like I said, the engine temp guage has never risen above half. On the other hand, half way is where the guage always sits and I know some have said that it is high. I recently towed a small trailer up eight thousand feet -- starting in mid afternoon with the temperature above 100 degrees and never experianced an increase in the temp guage.

The radiator, fan clutch, thermostat and belts are all new --- replaced one month before this problem started occuring. The belts on the water pump were loose, but I tightned them up and it did not effect the problem.

I think it is time to assume that it is really overheating. I intend to try and verify that with an ohm meter on the guage sensor, but I haven't done it yet.

So what are the potential causes for such an overheating? Note that it is not just overheating under high loads-- it does it at idle.

I did the Robbie HG bubble test-- this time running the car at 3000 RPM for a minute, and did not see a single solitary bubble. So hopefully we could rule out head gasket issues.

What else could cause this type of overheating?

Here is a potential list that I came up with-- do any sound like possibities?

1) bad water pump-- but I don't hear anything and they say that these are usually all or nothing things.
2) bad thermostat-- but it is brand new.
3) some restriction in cooling system caused by recently dislodged grey junk-- maybe in my cooling system overall I dislodged but did not remove some crud-- and that has recently created a restriction somewhere.
4) timing??? I haven't touched it and don't know anything about it
5) plugs? I have not touched-- I should do this
6) fuel delivery?

Anything else?

Where would you look first?

Jared

PS-- one more data point-- about the same time this started happening I had the check engine light come on-- code 28-- indicating a problem with the #2 oxygen sensor. I cleared the code, and it went away, only to come back a few weeks later. I cleared the code again, and it has not returned. I measured the impedence of the oxygen sensors and they seem to be in spec. Could this problem be related to the temp issue?

 

[cruiserdan]

Test the sensor, it may be bad. Put a meter on it and stick it in a pan of water. Heat the water till the circuit opens. I bet it may open before the water boils.

 

[alia176]

I don't know the '93 so I'm speaking in generalities here....while the vehicle is not running, check the condition of the hi and lo a/c press switch. Then run the engine and wait for the a/c to quit operating. Then check the state of these same press switches to see what happenend. I think the state of these shouldn't change in a normal a/c system. In your case, perhaps the hi press switch is tripping? Just guessing here!

Good luck.
Ali

 

[cary]

I'll throw in my two cents. I don't think you are overheating, I think you have a bad switch. Even with a seriously deficient cooling system, there is little chance the truck would overheat from dead cold by mearly idling. There simply is not enough heat coming out of the motor because it is being run under a minimal load. If you had stuck thermstat, the thing would overheat as soon as you drove short distance, so rule that out.

I think what is happening is the switch is bad, and as your thermostat opens, it gets hit with the warm coolant and causes the A/C to cut out.

Finally, you can check the actual temp of the coolant by using a mechanical guage through the radiator cap. This isn't quite the same as having it in the coolant passage but will give you a quick check.

 

[sjpitts]

I don't know the '93 so I'm speaking in generalities here....while the vehicle is not running, check the condition of the hi and lo a/c press switch. Then run the engine and wait for the a/c to quit operating. Then check the state of these same press switches to see what happenend. I think the state of these shouldn't change in a normal a/c system. In your case, perhaps the hi press switch is tripping? Just guessing here!

Good luck.
Ali

I am pretty sure that is is not the pressure switch, and is instead the coolant temp switch. As I said, when it cuts out I can get it restarted by unlugging the coolant temp switch. It then will quit again the moment I plug in the coolant temp switch.

Jared

 

[alia176]

Sorry, guess I didn't read everything as I should!

 

[Beowulf]

>> I think it is time to assume that it is really overheating. <<

Hang on there Jared! I'll agree with Dan & Cary that you've got a bad temp sensor or something in the wiring. If your truck had reached 226 (the A/C cut-off temp) then you would definitely know it was HOT, including the fact that the factory guage would be close to the red.

>> Could this [O2 sensor] problem be related to the temp issue? <<

Probably not.

-B-

 

[Rich]

I can share from personal experience, presuming that my AC cut-off sensor works correctly, that the factory water temp guage will be in the middle or just slightly above, and nowhere near the red, when the engine coolant temp is at 226.

That being said, I fully agree that Jared's ac problem is most likely the sensor and not overheating. From cold to overheat in a few minutes at idle would not be easy to accomplish.

 

[landtank]

I'd test the switch in the oil bath as the FSM states. This isn't hard to do and the only sure way to identify it as being the problem or not.

 

[Jwcool000]

i got a 93

Ok i just fixed my a/c a week ago, and i still have the r-12, so i got somethin called freeze-12 off ebay. worked great, but thats not the point.

when i was putting it in my cruiser i had to short/bypass a switch on the stage left side right at the front. this switch is an automatic shut-off incase frion gets too low. cuz it didnt take fluid till i bypassed that switch. so maybe its that switch that's being a little funky. try sticking a paper clip in the 2 larger holes after dissconnecting it. its low voltage so a paper clip will work. worked for me. but dont F**k it up, cuz i am not responsible. i am just tellin you what i did. so check that switch. its finicky

 

[sjpitts]

NEW INFO-- The saga continues

I pulled the AC coolant sensor sending unit for a test. This is actually pretty easy. I went to wally world and bought a cooking thermometer. It was a glass thermometer designed to use with cooking-- but it was less than 3 bucks-- so I am not sure how good it was. We can discuss this later.

I hooked up my digital multimeter to the sensor, and put it in a pan of water. The hardest part of this adventure was to get the sensor suspended in the water/oil without touching the sides of anything.

At room temperature the sensor measured OL-- open line, just like it should. I turned on the stove and heated the water. Here are the readings I got as the temperature rose:

150 degrees (F) -- 11.5 MegaOhms
175 -- 6.5 MOhms
200 -- 3.9 MOhms
212 (full rolling boil) 3.1 MOhms

Now I was unsure of what to make of this gradual decrease in resistance. This is still a pretty high resistance--I don't think you would get more than a few miliamps of current at 12 Volts with this resistance, and I don't know if that would be enough to trigger the relay that turns off the compressor. It also occurred to me that perhaps the steam created by this heating of water could be shorting the leads, creating a low conductivity path that registered on the multimeter.

Anyway, since I could not get the switch to switch in water, I decided I needed to do this in some oil. I chose canola as it is naturally cholesterol free.

I rigged it in the oil, and turned on the stove. I made the mistake of turning on the stove high and leaving. I came back to find the oil at 290 degrees, and a very low resistance --like 1.5 Ohms. So I turned off the stove and waited for it cool down. When the oil was cool again, I turned on the stove slightly and watched it heat up slowly.

This time around the resistance was given as OL until the oil reached 220 degrees. There was no gradual decrease in resistance this time -- at least at the lower temperatures. Finally, when it hit 220 degrees the multimeter started reading 25 Mega Ohms. The resistance continued to drop slowly-- at 239 degrees it measured 20 MOhms. Then at 240 degrees it switched --- to the low 1.5 Ohms.

I then turned off the stove and let it cool. The low resistance stayed until it dropped to about 230 degrees-- and then the resistance went back up.

I repeated this process again and it did the same thing.

So now I was left trying to explain what was going on. If anything, it was switching too high, and I could not explain why the sensor seemed to behave differently when it was heated in water then when it was heated in oil.

I came up with several possibilities. First, as I mentioned above, that heating the sensor in water creates steam that messes with resistance readings. Perhaps this can explain the gradual degrees in resistance when the water was heated. No steam with the oil, so this did not occur.

The second possibility is that by accidentally broke-- or fixed -- the sending unit-- by heating the thing to almost 300 degrees. Is this even possible? I say this for three reasons. First, there was no gradual decrease in resistance until it reached 220 degrees. Second, now the thing did not shut off until it reached 240 degrees. I should say that I am not confident that it was actually 240 degrees. It was after all, a three dollar wal-mart thermometer. Second, the thermometer says it needs to be immersed 2 inches for best accuracy, and the best I could do was less than an inch. So maybe the 240 degree reading was just a thermometer issue. Because other than that-- it seemed to work like it was supposed to. FINALLY-- get this-- I put the sensor back in the 80 filled it with coolant, and took it for a drive. I drove for a good 15 minutes and THE AC NEVER TURNED OFF. The problem seems to have gone away. More testing is required, but this was quite a shocker.

I would love to hear some other possibile explanations.

After all this I have decided just to call CDan and buy a new sending unit. It is the only way to get some piece of mind. With any luck, that will be able to confirm that it was just a sending unit problem all along.

Can I say that this has been one big pain?

Any comments?

Jared

 

[landtank]

Looks like the temp switch is working fine. Anything in the meg ohm range is as good as open. Water is a bad choice as the bubbles are hotter than the fluid and will give erratic results.

I originally missed that statement immersed in the second theory paragraph. But it most likely isn't a connection problem because that would provide an open in the circuit and disable the temp protection since a closed switch would have no effect through that bad connection.

Most likely wrenching on the switch torqued the unit just right to get it back working. This wouldn't be the first time that I've seen this sort of thing. Definately replace it as it probably won't last.

Good job!

 

[Beowulf]

>> Any comments? <<

I'm leaning toward a bad connection that was "fixed" by unplugging and plugging the sensor. I agree with your decision to replace it considering all the grief this has given you.

Keep us posted.
-B-

 

[sjpitts]

Just to update-- it has now been several days since I cooked my AC coolant temp sensor. The AC no longer shuts off-- it cools just fine-- and I am not having any noticable overheating problems.

I still can't really explain what happened. It really seems odd. My best guess is that I "fixed" a bad sensor or "broke" a good one. Lets hope it is the first one.

I still plan on replacing that sensor-- as soon as I get around to calling CDAN.

Jared

 

[landtank]

As I mentioned earlier, I've worked on various equipment almost all of them have some sort of thermal device whether switch, coupling, fuse or thermister, and more than once I've touched a bad device and had them maraculously start working. Almost never a permenant fix as they often fail at some point later on.

 

[sjpitts]

Sorry to resurrect a long dead thread, but I figured I would finish the story in case anyone ever reads this thread. To recap, last fall the AC was cutting off when the engine warmed up.

After trying a bunch of different things I was able to determine it was the AC coolant sensor that was turning off the AC. I figured this out by pulling the sensor connector and observing that the AC compressor started back up-- and it could be stopped by reconnecting the sensor.

I then tested the sensor by deep frying it in canola. During this process I accidently cooked it to almost 300 degrees. This actually fixed the problem. After this cooking the sensor appeared to turn on and off at the right temperatures. I reinstalled it, and there was no more AC cutting out.

Because I was not sure if I really fixed the sensor --in theory I could have just broke it-- I ordered a new one from CDAN.

A couple of days ago (months later) I finally got around to replacing the sensor. The problem remained fixed-- so I must not have been overheating, it was a sensor problem all along.

Conclusions-- 1) those AC coolant sensors do go bad. 2) It is easy diagnose-- just pull the connector when the AC has shut itself off and see if the AC starts back up (of course, actual overheating will cause the same result) 3) you might be able to fix your sensor by heating it to 300 degrees.

Jared

PS--Anyone want to buy a used AC coolant temperature sensor? Served medium well.

 

[gorister]

sjpitts,

Thanks for the follow up. I just started experiencing the exact same problem today. How much did the new switch cost you?