1994 Pontiac Grand Prix, V6 3.1L 151K miles

When I start the car with the engine cold, it heats up all the way to the red, almost 260. So far, each time I begin to find someplace to pull over, and and then the temperature suddenly drops all the way down to about 120 or so (so it stays at the hottest temperature for about 30 seconds to a minute). The temperature continues to fluctuate though, getting up to 230, then down again, then up to about 220 and down, and so on until eventually it settles at about 180 and stays there as long as I'm on the freeway and keeping a fairly constant speed. After it finally settles down, the low coolant light comes on. However there is plenty of coolant, and in fact the coolant was just flushed. After it was flushed, the problem seemed to go away for about 2 days, but has since returned and is actually worse now after I took the car on a long cross country drive.

The water pump was replaced 8 months ago to fix a leak, but it hasn't been leaking since then. As the car heats up, the air blowing out the vent is cold, and when it finally cools down and the low coolant light comes on the air gets hot. The thermostat was replaced about a year ago. Any help would be very appreciated.

Coolant must be full at the radiator not just the reservoir. It's almost certainly low as the warning says and you need to know why also,

T
Tom Greenleaf - MetroWest Boston - USA
_________________________________________

http://www.autoacsystems.com/tomgreenleaf/

Thanks for your help. I filled the radiator directly instead of just the reservoir, and the low coolant light no longer appears, but the temperature continues to do the same thing; initially overheating to 260 then rapid fluctuation.

I typically turn the heater on full blast as soon as it starts overheating, but even with the engine temperature so high the air blowing is very cold. When it finally starts to cool down, there are a few blows of slightly warm air, but which go away when the temperature starts to heat up.

Are there any other ideas or suggestions? Thanks for your help!

That 260 has to go of course. There may be an air lock that won't purge out, thermostat may be bad, plugged up radiator and heater core and not so common in heating season but fans may not be coming on for the radiator.

What is probably happening is the spike to 260F is actually boiling even with system under pressure that's about the limit and the bubbling hot water/steam doesn't transfer heat - odd as that sounds.

You should check again that all is full when cold and start it up and feel the upper radiator hose. It should stay cold till you reach about the t-stat's setting - usually about 195F. It will slowly warm up and then be too hot to hold - be careful. Gance at gauge and note what it does. If it reads hot and hose really doesn't warm up then thermostat isn't opening. It may be hot right at the engine but radiator end would stay cool.

This could be a few things including the water pump impellers are corroded or broken somehow and not pumping the coolant very well.

Also feel heater hoses. One (smaller) is inlet and should be as warm as the engine temp. Outlet should still be warm but with heater on full blast would be notably cooler is blowing heat and if cold the flow could be so low that it loses all it's heat.

Shoot - this sounds like a lot to do but you really need to fix this or that temp will cause problems if it hasn't already. Some cars do have an attitude about getting all the air out. If you find it low again at radiator there's a problem and hate to say it could be the head gasket letting exhaust gasses into cooling system. Frequently if that's happening the upper hose can be felt for pressure when cold there's none and within a minute it builds up pressure which is too fast for normal engine heat to have done that.

You can check that stuff. How much are you apt to take on with this?

T
Tom Greenleaf - MetroWest Boston - USA
_________________________________________

http://www.autoacsystems.com/tomgreenleaf/

just to add to the perfect posts of tom....
the 3.1 engine is known for lower intake manifold gasket problems.....and head gaskets as well..
plus there is a air bleed on the pipe that runs along the top of the engine from the t-stat area to the surge tank/reservoir....when filling up with coolant-engine cold,not running,loosen bleeder and fill until coolant runs out of bleeder.....air will then be purged......close bleeder,run engine,adding coolant as necessary,until t-stat opens,assuming t-stat is ok,cap off,it should be ok,providing again t-stat,intake gasket and head gaskets are ok....
but if they never have been changed,good chance they will need to be.....
....check for an intermix of oil and coolant in the crankcase,can check inside oil fill cap,
also can check the egr for traces of it.....when bad the solenoids will sludge up,usually will set a mil light cuz solenoids won't operate properly on decel when pcm commands them...

just wanted to add my thoughts.....hope i helped....

Thanks both of you for your help. The other day I went out to follow Tom's suggestions, I let the car sit and warm up while I felt the hoses. On that instance, the car heated up to about 240 with the upper radiator hose still cold, then it dropped and the hose heated up and the temp actually didn't hardly fluctuate at all after that. Later I started the car again and took it for a drive, the car initially heated to about 230 then dropped and ran well. So I've been driving it quite a bit, and so far only about half the time I start it up will it even spike higher than 200, and the other half of the time it runs perfectly.

I guess this is good news; maybe there was just an air pocket in the radiator that needed some time after filling it up to purge out. I still can't help but think that there is some underlying problem causing the coolant to not circulate well. Anyway, thanks again for your help; we'll see how it runs from here out and I'll post again if it starts acting up again.

Sure seems like there was an air lock. You may still have some and some are a pest to get out all. Jim pointed out that. I've had some be so nasty I litterally jacked a car up till the rad cap was the highest thing in the car and that's where the air will go out to tank but not back if tank isn't empty. Oddly, the engine temp rises after you shut it off for a while. When you re-start at a few minutes the t-stat should open wide and you catch them them. Fluctuations will end when all is perfect,

T
Tom Greenleaf - MetroWest Boston - USA
_________________________________________

http://www.autoacsystems.com/tomgreenleaf/

Well here's the update. After a few days, the problem started up again. After topping off the radiator and letting it run for a bit, the problem would go away for a few days then return. I decided I wasn't going to get around taking it in to a shop. They looked at it, replaced the water pump which was supposedly corroded, put in a new thermostat, and it worked great for one more day, then the exact same problem returned.

I took it back to them, they said it was just another airlock, but they ran a test (something to do with hydrogen) to check the head gasket. It passed the test, but the mechanic said he had a suspicion that it might just have a small enough leak that it could fool the test, and told me to drive on it for a few more days and see what happens. Well, one day later, the temp is fluctuating again, although it hasn't let it get too hot yet.

If the problem continues tomorrow, I'll take it back, but it sounds like they're going to tell me there is a very tiny head gasket leak.

If that's the case, fixing a head gasket will cost more than the car is worth. The only other option I can come up with is to try some kind of sealant. Can anyone make any recommendations for a sealant? I know they aren't generally a great idea and don't always work, but as a last solution just to try and get another year or so out of this car I am willing to try it, especially if it's really just a very small leak. Has anyone used one of these sealants before?

Pre note: Exhaust gas testing can miss the leak. Pure gasoline that burns perfectly produces just CO2 and Water! Not much to detect.

I'm pretty sure this is getting combustion or exhaust gas pressure into the cooling system. This would be noticed a couple of ways. With engine cold feel the pressure like on the upper radiator hose which should be none. Start engine and it shouldn't built up pressure very fast and if so it's pressured up with exhaust/compression gasses that exceed the 15lb pressure cap and coolant would go to recovery tank early and perhaps not draw coolant back in so just the gas (vapor) is again in the cooling system screwing up the show. Perhaps not the first time but eventually you would see air bubbles going to the recovery tank as it runs or while warming up.

This could also get air in cooling system if a regular closed system pressure cap at the radiator is used and the dime size flapper at the bottom doesn't seal properly it wouldn't build up pressure at all but would let expanded coolant out to recovery tank and perhaps just draw back air when cooling or contracting. JUST THAT CAP COULD CAUSE ALL THIS AND I KEEP FORGETTING TO MENTION THAT - sorry. Three things must seal properly in a rad cap or it doesn't behave. Guck (nice word) in coolant can mess them up. If in any doubt just toss that cap for the exact proper replacement.

Sometimes a pressure test can put coolant backwards thru the leak and show up that way. Watch out as if it gets enough it could cause hydraulic lock where piston can't compress a liquid and could cause damage. A leak in a gasket might not leak with the 15lbs bust combustion pressures are out sight and may only leak with that much pressure and not go backwards.

Confused enough?

T
Tom Greenleaf - MetroWest Boston - USA
_________________________________________

http://www.autoacsystems.com/tomgreenleaf/

I am now almost sure that the air temperature outside is having some effect. I have noticed for a while the problem seems worse the colder it is outside. Now, the mechanic claims after replacing the water pump he can't get it to overheat in his shop. Although he did say his test now shows signs of a blown gasket. I drove it early this morning, with the temp outside about 10 F, and it overheated as usual. I went outside in the afternoon, air temp about 25 F to try Tom's suggestions, and everything worked perfectly with no overheating (naturally). I drove it around for about 20 miles with perfect performance. I don't have a garage unfortunately where I could test this more accurately.

If it is a head gasket as the mechanic claims, why does the temperature seem to affect it? And if I were to use some kind of sealant (since the car isn't worth the $1700 head gasket replacement), should I take the weather into account? Thanks for all your help by the way.

Ok Jakets15: I'll venture the guess that this indeed needs the head gasket job. Just know that these gaskets are sealing substantial metal parts keeping pressures, coolant, oil, and pressure of combustion where they belong and the metals expand and contract with temps and the gasket must comply to that. If every part of the engine warmed up and cooled down at the exact same rate this wouldn't be a problem in engines at all.


The job is expensive due to the time, tools, and skils needed for a sucessful fix - the parts aren't that expensive. There's also a risk factor of a flaw in the block or heads that could go undetected and the job fails - arggh for all involved!

You pointed out the most important factor that this exceeds the value of the car. Even if this is fixed the car is 14 or so years old and all the rest of it has over 150k on it and another big buck surprise could come along.

Are you willing to take a chance at some tricks that have some chance of buying you some time for little cost and just hope it works and go from there?

What do you want to do?

T
Tom Greenleaf - MetroWest Boston - USA
_________________________________________

http://www.autoacsystems.com/tomgreenleaf/

I have basically the same problem. I have a 1994 Pontiac Grand Prix GTP that is overheating...but I do believe there is air in the system...because I can see the bubbles in the overflow. I was wondering what the remedy for this problem is? I have already replaced the thermostat, all hoses...etc. But the car heats up (almost to the red) then it cools off and continues that same pattern. We recently did a radiator flust and changed coolant and the problem is still going on. I am at a loss about what to do,.. I love that car and will do anything to have it running right. Any ideas? All help welcome:)

You've found an old thread - best to start new and your own but no matter now. Contant bubbling in recovery tank especially with your observations points straight at needing the head gasket job.

If this just did that a couple times when you had system drained for any reason and quit it wouldn't be so suspect. Air is hard to purge out of some systems but will on it's own eventually and always best not to let that happen or allow overheats,

T
Tom Greenleaf - MetroWest Boston - USA
_________________________________________

http://www.autoacsystems.com/tomgreenleaf/