1995 Chevy Cavalier Overheating and fan isn't running

The temp sensor sends a signal to the powertrain control module which starts the fan. You can go ahead and change the sensor as long as you have it. For minimum coolant loss follow this proceedure. With the engine cold, remove the resevior cap to release any pressure and put the cap back on. Have the new sensor ready to install and pull out the old one in one motion and screw in the new one in a second motion. This way you will only lose a cup or two of coolant. Normally I could paste a wiring diagram to this post, but my program is down until Thursday. If you have, or have access to, a fax I can send you a copy. Please advise. Jim

[quote:2d7e54b2b4="James R. Weizel"]The temp sensor sends a signal to the powertrain control module which starts the fan.

...

Normally I could paste a wiring diagram to this post, but my program is down until Thursday. [/quote:2d7e54b2b4]

I bought a week of eautorepair.net and already found the wiring diagram on there. It is as you say. I was assuming wrong.

The information on the site says it will turn on around 100 degC, which would make the sensor 177 ohms. When I measured it in a somewhat hot engine I got 270 ohms, so that may be the problem after all? It also says "ECT sensor may have shifted out of calibration and should be replaced". I'll try replacing it.

[quote:2d7e54b2b4]You can go ahead and change the sensor as long as you have it. For minimum coolant loss follow this proceedure. With the engine cold, remove the resevior cap to release any pressure and put the cap back on. Have the new sensor ready to install and pull out the old one in one motion and screw in the new one in a second motion.[/quote:2d7e54b2b4]

Hah. Good idea.

[quote:2d7e54b2b4]This way you will only lose a cup or two of coolant.[/quote:2d7e54b2b4]

I'm not so much worried about losing coolant as making sure there's no air in the system, though.

How do I replace this sensor without letting air in? Or how do I bleed the air out after? The system has apparently boiled over previously without my knowledge, so I'm worried there would still be air in the system from that, too.

If don't have a bleeder port on the thermostat housing,which I'm sure you don't, don't worry about air in the system. When the motor warms up and the thermostat opens, any air will go into the overflow tank.

Replaced the sensor this morning and drove it around more than 5 miles in 35 degF weather. The temperature did not go much above the middle, but the fan did not come on, either. Maybe I will test the old thermostat in boiling water and see if it matches the resistance from the chart.

It seems I can drive it around like this, at least.

GOOD, keep me posted, Jim

I tested the old thermistor and the resistance values are correct for boiling water and ice water, so I think the reason it overheated was actually because of air in the system. Everything else has been identical since then, but it hasn't overheated. It probably leaked until empty, but the low coolant light didn't come on immediately, and then when I filled it up, there was still air in the system so it overheated and boiled over? But since then it's been fine. The low coolant light often stays on even when there's coolant in there, so I wouldn't be surprised if it didn't come on when empty. I should probably get a new sensor?

Yes, Iwould advise a new sensor. The perpetual problem with warning lights versus gauges is, a light will tell you what you have just done, ie, over heated your engine, run out of oil etc. and by the time the light comes on, it's too late. A gauge, on the other hand, will tell you what you are about to do, giving you a little warning.

I'm still having problems with the system overheating or mysteriously emptying of coolant.

We've checked the thermostat in a pan of boiling water and the recovery tank pressure cap by pressurizing the tank to 16 psi and both are working.

We've run water through every part of the engine until it comes out clear (was dirty brown with rust and stop-leak), run water both directions through the radiator and it seems to be flowing well.

The radiator hose gets warm after getting up to operating temperature, so it's flowing through the radiator.

I don't see any leaks. The only time I ever see water/coolant coming out of the engine, it's coming out of the recovery tank cap.

The temperature gauge goes up steadily until it gets to operating temperature and then levels off, then after 10 or 15 minutes of driving it overheats, spraying out from under the pressure cap. If I open the bleeder valve at this point air sprays out. If I open the recovery tank cap, water sprays out at first, but then when the spray stops and I finally open it, it's completely empty, like the water was being pushed out by air/steam?

The cooling fan turns on if I turn on the A/C, so the connection to the computer is fine, but I've never seen it come on otherwise, even once when overheating (didn't look other times).

I already replaced the fan relay and temperature sensor, and I tested the old temperature sensor in boiling and ice water, and the resistance was correct with the old one, too.

I really have no idea what's causing the problem. I drove for 400 miles without any problems, but then drove 20 miles with water in the system and it overheated. Maybe the emptying of the tank is just from me opening the cap while it's hot, which lowers the pressure and allows it to boil? That might explain why it only happens sometimes.

Actually the fan is working. Drove it around with cardboard over the radiator and the fan finally kicked on, with the temperature gauge still in the normal range. So that's not it. A little water was coming out the cap of the recovery tank, so maybe that's the problem after all?

Let's try a little experiment. With the engine cold and the cooling system completely full, and radiator cap "off". Start the engine and let it warm up, the whole time you are doing this, watch the fill neck on the radiator for excessive bubbling and churning of the coolant. We're looking for a bad headgasket or a cracked head.Let me know.

I don't know what the fill neck is. This is the kind of system where the recovery tank has a pressure cap, and there's no cap on the radiator itself that I know of.

I flushed it and filled it again yesterday, then drove it for 7 hours and noticed no problems. I stopped after a few hours and bled a little air, added a little more coolant, but that's it. It seemed to be working fine, and I figured I just needed to let the air out better before driving it. I was ready to blame the problems on myself again for leaving air in the system which let it boil over.

But then today I drove 30 minutes to work and parked it, and when I came back a few hours later and twisted the recovery tank cap, it was still under pressure and sprayed out coolant. I opened the bleeder valve and a lot of air came out. I had to add another tank and a half of coolant to fill it back up. Then I drove 30 minutes home, drove up a steep hill (heard gurgling from the heater core, as if there was still air trapped in it), and opened the hood while the engine was still running. I didn't see any coolant leaking out anywhere. I turned off the engine and then heard a spurting noise. I ran over and looked and coolant was pouring out from the recovery tank pressure cap.

So when I turn off the engine, there's a sudden increase in pressurized gas, which then pushes all the coolant out the recovery tank? That's not good, is it?

Lets go back to the test I mentioned in my previous response. Again, making sure the system is full as you can get it. With the cap off the reservoir, let it warm up while watching for bubbles at the fill cap. It's sounding more and more like a head gasket problem.

[quote:a2aee48846="James R. Weizel"]Lets go back to the test I mentioned in my previous response. Again, making sure the system is full as you can get it. With the cap off the reservoir, let it warm up while watching for bubbles at the fill cap. It's sounding more and more like a head gasket problem.[/quote:a2aee48846]

Ok, I had time to try that today.

At first, the coolant level in the recovery tank rose as the engine ran, but it's not because the coolant is getting hot and expanding; it's because of air entering the system and pushing the coolant out as it runs?

I ran the engine without a cap on the recovery tank until the coolant had risen to near the top of the tank, and then shut it down. The engine wasn't up to temperature yet. I then bled the air out of the bleeder valve and the coolant level went right back down to where it was. I did not see any bubbles or foam in the coolant tank. I did this two or three more times, with the level rising to overflowing, and then shutting it down and bleeding the air out to get it back to normal level.

It got closer to normal temperature. The radiator hose was warm, so the thermostat was open.

Then each time I turned off the engine, the coolant level would suddenly rise up and overflow the tank. It didn't shoot out or anything, just rose up and overflowed within several seconds. The level of coolant "pulses", too, which might just be from sloshing around somewhere in the system?

Then if I open the bleeder valve, "air" comes out and the level goes back down to what it was before I started it. I tried to smell the air to see if it smelled like gasoline/exhaust, but I couldn't tell anything besides coolant smell.

So then I started it again, and it went up to normal operating temperature (slight above halfway on the gauge, where the temperature usually stabilizes), and the coolant level didn't rise this time. The level stayed pretty constant and I let it run for a while, significantly longer than the previous times as it was warming up. The thermostat was definitely open now.

The level never got too high, so eventually I shut down the engine, and THEN the coolant level rose quickly and big bubbles entered the recovery tank. There were never bubbles any of the previous times. Then I opened the bleeder valve again and approximately the same amount of air came out as other times.

I don't see any oil floating in the coolant, and I don't see any foam in the oil fill tube. We checked the spark plugs several weeks ago and didn't notice anything too strange. The oil level does not seem to rise as the coolant decreases. I have not actually taken out the oil plug to see if there's coolant in the bottom of the pan, though. Need to find a place I can jack it up.

I don't mean to cry wolf, but the air has to be coming from somewhere. With all the bleeding you have done, there should be no air left in the system at all. Keep in mind, the cooling system has a max of 15 lbs pressure where compression is 120 lbs plus. A slight headgasket leak or small crack in the cylinder head will pass pressure into the cooling system, but very little back into a cylinder. A test for this theory is to get a cooling system pressure tester. They are a small unit with a radiator cap attached to a small hand pump with a short hose. With the engine as warm as you can stand it, pull the spark plugs, pressurize the cooling to 15 lbs and let it sit overnight. If the pressure drops within the first hour, pump it up again. Then, in the morning, crank it over to see if any coolant comes out of the plug holes.

[quote:ee4917220f="James R. Weizel"]I don't mean to cry wolf, but the air has to be coming from somewhere.[/quote:ee4917220f]

Yes, but where? Is the cylinder the only possibility? I've had the head gasket replaced in 2006 already.

[quote:ee4917220f]Keep in mind, the cooling system has a max of 15 lbs pressure where compression is 120 lbs plus. A slight headgasket leak or small crack in the cylinder head will pass pressure into the cooling system, but very little back into a cylinder.[/quote:ee4917220f]

What are the long-term effects of this?

Also, why would it not have problems when driving for long distances?

I can borrow a pressure tester from AutoZone for free, apparently.

I guess I'll try that next.

Long term affects, if I'm correct about a gasket or head problem, will be very little other than losing coolant and possibly overheating. I say that as long as coolant isn't getting into the oil. Antifreeze just "destroys" the lubrication ability of motor oil.

Ok, that's good to hear. Last time I had a head gasket leak it was pretty bad. After the first mechanic lied about fixing it, it eventually broke through into a big leak and I had smoke pouring out the exhaust, etc. This time it doesn't seem so bad.

Instead of wasting $1000 on a repair that only lasts two years, I think I'll try one of the magic snake oil liquids to seal the hole this time. :) I don't think my car's worth a real fix anymore.

But first I'll get a pressure tester and confirm that that's what's actually wrong.

(After driving it around a bit this weekend, I checked this morning and there was only a little air to be bled, and the coolant level hadn't dropped. After driving to work, I saw that it had been pushed out the recovery cap again.)

[quote:8b03dcc92c="James R. Weizel"]A test for this theory is to get a cooling system pressure tester. They are a small unit with a radiator cap attached to a small hand pump with a short hose.[/quote:8b03dcc92c]

Got it. I'll have to do it tomorrow or Sunday though, and it's going to be cold and snowy Saturday night to Sunday morning.

[quote:8b03dcc92c]With the engine as warm as you can stand it,[/quote:8b03dcc92c]

Why does the engine have to be warm?

[quote:8b03dcc92c]pull the spark plugs, pressurize the cooling to 15 lbs and let it sit overnight. If the pressure drops within the first hour, pump it up again. Then, in the morning, crank it over to see if any coolant comes out of the plug holes.[/quote:8b03dcc92c]

Is there any danger involved with this? If lots of coolant leaks into the cylinder can it hydrolock even with the spark plugs out?

I like that, "magic snake oil liquid". We used to call it "a miracle in a can". Just keep an eye on the color of your engine oil. If it starts to look like a chocolate malt, you in trouble. Antifreeze in oil will take the bearings out in a heartbeat.

The engine should be warm so a crack in the head or head gasket is at it's most "open". As far as hydrolocking, with the plups out, any coolant in the cylinders will blow out the plug holes.

[quote:738c6b7603="James R. Weizel"]I like that, "magic snake oil liquid". We used to call it "a miracle in a can".[/quote:738c6b7603]

I see a lot of naysayers online who have never actually tried it, and a number of people who have tried it and claim that it has worked for them, so I think it's worth a shot. :)

I'm wondering if this leak is caused by the mechanic in 2006 not replacing the torque-to-yield head bolts.

I guess if you want something done right...

This page says it's actually unlikely that a pressure test will be conclusive for a head gasket leak (as opposed to some external leak) because of the much higher pressures involved in pushing exhaust into the coolant, and because gases are molecularly smaller than coolant.

In fact, now that I look at it, this pressure tester doesn't have any kind of threaded attachment for going onto my recovery tank. :(

I replaced a head gasket only once without changing the headbolts trying to save a customer a few bucks, and I wound up doing it again for free. As far as the pressure differential shown on the page, It's a 100% true. that's why you want to do the pressure test starting with a warm engine. The test still may not show anything, but that doesn't rule out a gasket problem.

[quote:dcc2af07c7="James R. Weizel"]I replaced a head gasket only once without changing the headbolts trying to save a customer a few bucks, and I wound up doing it again for free.[/quote:dcc2af07c7]

I should have confronted the guy about it as soon as I noticed, but I figured he knew more than me...

Would it be stupid to try to tighten the bolts more now? Would probably break them or something, right?

[quote:dcc2af07c7]As far as the pressure differential shown on the page, It's a 100% true. that's why you want to do the pressure test starting with a warm engine. The test still may not show anything, but that doesn't rule out a gasket problem.[/quote:dcc2af07c7]

Oh, ok. So when the metal expands it opens the gap bigger? Seems like it would have the opposite effect.

And I'm not going to cause damage to the aluminum head by unscrewing the plugs while it's hot and expanded, am I?

You would think that heat would make the leak tighterbut, the head is aluminum and the engine block is cast iron which expand and contract at different rates. If the head isn't torqued evenly all the way around it's going to move around on the block. this movement is what causes most leaks. In a "gross" overheat situation, the expansion of the aluminum head will actually stretch the head bolts and eventually cause a gasket leak. I would never warranty a head gasket replacement unless I had the head milled before I reinstalled it 'cause aluminum is so prone to warpage.

Indeed, the pressure tester doesn't fit, so I gave up on that. My symptoms sound exactly like a combustion leak, though, so I'm just going to assume that's what it is.

Also I was reading that a block tester would be better for combustion leaks, and is also available at Auto Zone

but I figure it would be better to do that *after* pouring this stuff in, to see if it actually works? So I just dumped it in. I took off the upper radiator hose with a bucket pushed underneath to catch the coolant, but NOTHING CAME OUT. :( Just some drips from the engine.

There was coolant visible in the hose, though, so it wasn't completely empty. You're supposed to put the stuff directly into the radiator or upper hose, so I let some coolant out the stopcock to make room for it in the hose. The coolant that came out did look a little orangeish/brownish, but also had black dust in it. The black dust sticks to a magnet, so it must be rusting from having air in the system? :( I don't remember seeing much rust dust when I drained it and flushed it over Christmas, so I guess it's been getting worse.

So I poured in the Bar's Leaks Head Gasket Fix, which doesn't require you to drain the coolant first. It has copper-colored sparkly particles in it, which I suppose is pixie dust? :) It's a little viscous. I put the hose back on and filled up the recovery tank, opened the bleeder valve, filled up the tank, etc.

Then it said to idle the engine for 15 minutes, so I did. The thermostat didn't open fully in 15 minutes, though. I'm not sure if it's supposed to. If it is, the instructions should have said "until the thermostat opens" instead of "for 15 minutes". But I'll follow the directions and see what happens. I'm waiting for it to cool off, then I am supposed to top it off and drive around like normal. I emailed back and forth with the company and they swear that people have had success with this stuff and combustion leaks.

"If you are losing coolant, but there are no signs of leaks, you could have a blown head gasket. The best way to test for this problem is with a combustion leak test on the radiator. This is accomplished using a block tester. This is a kit that performs a chemical test on the vapors in the radiator. Blue tester fluid is added to the plastic container on the tester. If the fluid turns yellow during the test, then exhaust gasses are present in the radiator."


"A block tester, often called a combustion leak tester, is placed in the radiator filler neck. The engine is started and the test bulb is squeezed and then released. This will pull air from the radiator through the test fluid. The fluid in the block tester is normally blue. The chemicals in the exhaust gases cause a reaction in the test fluid, changing its color. A combustion leak will turn the fluid yellow. If the fluid remains blue, there is no combustion leak."

let me know how it goes.

The accounts I saw said the stuff worked within a few minutes, and it hasn't worked for me after a day, so I guess it's not working? I'll give it a few days, maybe this stuff works differently because it is allowed to mix with coolant? But I'm not optimistic.

After driving it today, I opened the bleeder valve while the engine was still running. A little air came out, but then coolant. Then I shut down the engine and did it again right away. This time lots of air came out. So it's really only pushing air into the coolant when the engine shuts down. What would cause that?

The air is probably trapper in the cooling system by the water pump and when you shut it off, gravity allows it to come to the top.

[quote:7a5d12c536="James R. Weizel"]The air is probably trapper in the cooling system by the water pump and when you shut it off, gravity allows it to come to the top.[/quote:7a5d12c536]

It seems to increase a lot in pressure, though. When I tested it with the cap off the recovery tank, the level became steady and stayed there for a long time, and then when I turned off the engine, it suddenly increased and surged up over the top of the tank. Like it's only leaking into the coolant right when I shut it down.

Also, even after I bleed the air out at the bleeder valve, there's still gurgling in the heater core. Does that valve not get all the air out?

Alright. Now I have the block tester, which is a little clear tube and you put pale blue liquid in it and then suck air into the tube from the cooling system. If the air has exhaust in it, the liquid changes to yellow. I'll have to do it tomorrow during the day.

In the instructions, it says to identify the cylinder by running the engine with one spark plug wire removed (and purging all the coolant between tests). I didn't know you could even do that.

The fluid did not turn yellow. It's very pale blue, though, so if the yellow is pale too it's hard to tell. But it sure looked blue after I tested it. Then I put the tube right near the tailpipe and it still stayed blue, so I don't know what's up. I might try again.

It's pushing coolant out even more now than it was before, I think, and it looks kind of brownish. :(

The "brownish" is rust being churned up in your engine block.

The pressure tester I borrowed didn't work. The block tester I borrowed didn't work. The Head Gasket Fix apparently didn't work (though they say it can take up to three weeks), and now it's having trouble starting.

I took it to a mechanic. They'll look at it tomorrow morning and then ask if they can charge me incredible amounts of money to fix it.

They say it's definitely the head gasket ("the corner under the thermostat housing - that's the corner that always goes on the cavaliers").

$950 to fix it for real. :'(

He agreed with me that it failed because the previous mechanic didn't replace the head bolts. He even said they were "loose"!

Now I wonder if I could have "fixed" it just by retorquing the bolts.

I guess it's good that I got it fixed for real, but ohhhhhhhhhh $1000 hurts.

The car's been misfiring and idling poorly ever since I got it fixed. This isn't just the computer relearning, is it? I've driven it 40 minutes a day for a week and a half.