Erratic Cooling System Behavior

[wikitjuggla]

Hello,

I have a 2000 Buick Century. About a month and a half ago it started overheating to about the 3/4 mark then falling back down quickly and erratically and the heat would stop working and with the radiator cap off, starting from a cold engine, within a couple minutes, coolant would come flying and pouring out and for the last few months the low coolant light comes on and off intermittently even when the coolant is full. I replaced the thermostat a few months back and flushed the nasty Dexcool gritty muddy sludge out of the system and replaced with the Prestone green stuff. After these problems started I replaced the water pump, didn't change a thing but then later that day, all the problems just stopped and everything started working again except for the low coolant light issue. Then the other day all the same problems started again. -10 below outside and within a couple minutes at idle from a cold engine, its overheating on the gage to 3/4 mark then falling back down almost like a broken gage. Once at operating temp still goes up to 3/4 only at idle and a little bit less than that during driving, all the while with no heat. Now if the thermostat i put in was getting stuck closed, I should still have heat. No oil in coolant, no coolant in oil, Not burning any coolant, not burning any oil. Loose about a quart of oil every 1000 miles or so, not entirely sure where it goes though. Pressure in the system with the cap on is building up enough to nearly overflow the reservoir tank but not enough to spill out of the overflow safety valve on the radiator cap and i am not leaking so much as an eyedrop of coolant. It's almost as if the coolant isn't flowing what so ever anywhere. This doesn't seem like a typical headgasket failure but pretending it is, The only thing i can come up with is it almost seems like a cylinder is pushing air into the coolant system and the air is moving against the directional flow of coolant and almost stopping coolant circulation all together. Just my thought. Don't know if thats even possible. I'd go ahead and replace the thermostat again just to rule it out but anyone who has done one on one of these engines knows how somewhat of a chore it is. As far as i can tell, i don't have any power or performance loss in the engine

Thank you for all your input

 

[Spriteman]

A quick check for a bad head gasket is revving the engine with the radiator cap off and see if the coolant comes shooting out. Sounds like you should do this test with a cold engine so you can remove the hot engine temp from the diagnostic equation. You can also do a compression test on each cylinder - this will also tell you if the head gasket is compromised.

Realizing it's freezing cold there but if you can do the following:
Drain all of the coolant from the engine, heater core, and radiator. Then flush out the each part of the cooling system with fresh water until it runs clean - this will make sure that there isn't a big blob of sludge in your cooling system. Another quick thing to do is drill a small 1/16 hole in the metal ring of the thermostat near the middle so it doesn't get covered up by the gasket. This will allow you to bleed air from the system on the engine side and make sure that the cooling system doesn't have any air in it.

Let us know what you find and we can go from there.

 

[wikitjuggla]

Sorry its kind of a long post but I already mentioned with the radiator cap off and starting from a cold engine, within a few minutes at idle coolant comes gushing out, Revving doesn't seem to make much of a difference and that is my one and only symptom of a bad head gasket. Already did a coolant flush a few months back using a backflush method and a bottle of that coolant flush cleaner. Thinkin about grabbing another bottle of that though and leaving it in the system for a few days of use. Back of the bottle said that was ok for really dirty systems. Still need to take it in for a compression test

 

[Spriteman]

Another possibility might be a collapsed radiator hose - the outside of the hose looks fine but on the inside there's no room for the coolant to pass through.

 

[wikitjuggla]

well if that was the case i wouldn't think coolant would come shooting out of the radiator and i should still have heat

 

[wikitjuggla]

When I start the car cold w/o radiator cap, within a minute or two, the coolant comes gushing out. I mean just pouring everywhere. If I had a hole in my head gasket big enough to projectile vomit my coolant out, shouldn't coolant be falling into the cylinder at some point and burning off? Because I can't smell burning coolant even if I stick my nose within inches of the tail pipe. With the radiator cap on the coolant does not build enough pressure to leak from the pressure release valve on the cap. It has gotten within a couple of inches of the top of the reservoir which is way far above the HOT fluid level line but thats it

 

[HotZ28]

When I start the car cold w/o radiator cap, within a minute or two, the coolant comes gushing out. I mean just pouring everywhere. If I had a hole in my head gasket big enough to projectile vomit my coolant out, shouldn't coolant be falling into the cylinder at some point and burning off? Because I can't smell burning coolant even if I stick my nose within inches of the tail pipe.

The cylinder pressure (psi) is much more than the radiator pressure, so therefore one pressure overrides the other. As suggested, do a compression test and leak down test, as well as a radiator leak down test to identify the problem.

 

[wikitjuggla]

wouldn't it draw coolant in on the intake stroke? or I suppose since thats the only stroke with a negative pressure in the cylinder coupled with high rpm makes the draw through a small gap non-existant. What is a leak down test? The shop told me a compression test won't tell them if the head gasket is bad but that doesn't really make any sense since it is pushing air into the coolant system, obviously the cylinder that is leaking would have reduced compression. They said only way to test for bad head gasket is with their thing that tests the coolant for exhaust fumes. I've seen those kits on the internet but do they work? I don't have tools for compression test and don't really feel like paying $150 or more dollars for it on a car thats probably toast

 

[HotZ28]

If the gasket is blown between the cylinder & the coolant passages, or if you have a cracked head located in the combustion cylinder, you would not ingest coolant into the cylinders. Pressure from the cylinder prevents that, and the cylinder pressure goes to the point of less resistance, i,e cooling system. A leak down test will show where the pressure is going, either through the rings, valves, or into the radiator. Once the cylinder is pressurized, you can here/see where the air is flowing. The compression test will tell the story if you compare cylinder to cylinder pressure, or just one low cylinder pressure reading. The problem with that is, a small leak (10 psi) may not identify the problem.

 

[wikitjuggla]

yeah I spose those caps are rated to blow at 15psi so only 10 psi could be anything from head gasket to sloppy piston rings. I will have to call around for that leak down test. Problem is, I live in a small college town with about 5 auto shops plus a gm dealership (which is overpriced and from what I hear terrible at repairs or conning people) and aside from the dealership, no one will touch this stuff with a 10 foot pole. Closest place is about 60 miles that I would have to drive it. At this point its still drivable and the temp doesn't go up too bad. More or less it seems to get "hot spots" near the temp reader. I'm pretty sure the air is traveling against the flow of coolant. Do those exhaust fume tests with the color changing liquid work?

 

[Spriteman]

I'm not sure but I think the guys at an Autozone type parts place could help you with a cylinder pressure test. They are usually helpful in the diagnosis area especially if one of them is a past mechanic.

 

[lewie2792]

I am having a similar problem. After trying two methods of bleeding the coolant system, I have good heat, but the temperature gauge surges up to toward
overheat, then I turn on the heat and temp gauge quickly drops down to normal range and stays there. Some other research points to head gasket.
Tried the block tester from Auto Zone (loan a tool). Surging coolant pretty much prevents successful testing. I don't want to spend the big money required on a 1999 Century with 124 K. I wonder if a head gasket sealer additive would work? I also have the catalytic converter check engine code. Back flow could be somewhat restricted but I don't see where that would affect the cooing system. Other than the coolant, I have no other symptoms of head gasket problems. Other than a head gasket additive, I am out of ideas. Regards, John

 

[wikitjuggla]

Thermostat. $10 fix

 

[tomjr]

Hi: I have a similar problem. Wonder if you have resolved your problem.
I have a 2004 Century with about 80K miles.
Runs great except. On a short trip I noted that Temp gage was almost full over with a red light. The gage fell back to about mid point but seemed to want to rise. Got home. Next day I found that most coolant was gone. Saw coolant on driveway. Noted coolant around the water pump. Replaced it. Still had problem.
Next I back-flushed and replaced rad cap, and thermostat. Still saw coolant under car. Removed the side panel behind the passenger front wheel. Ran car and took photos and video at many angles. It appeared that there was a steady narrow stream shooting out from behind the timing chain cover.
Read that coolant somehow runs in that area of the T C cover. Replacing that gasket is not a job that I would undertake.
Got gasket replaced. $1000 job. The leak is gone.
However, I am thinking that the leak was an EFFECT caused by another problem because, after repairs, the first time I took the car out on a drive the gage ran up again but not quite so far as I had my eye on it. It seemed that the gage fell back again to normal and seemed to move past the center digit. Prior to the first incident my gage always read about half way between the center digit and the mark to the left of center.
I drove to a auto store turned off the engine. Checked the coolant level and it was fine in the reservoir.
Went in and got a gallon of coolant just in case I had a further problem as I was 50 miles from home.
When I turned the engine ON I noted that the gage at first read very high for a few (10 or 15) seconds. It fell back to normal after a short time. When I got home the coolant checked ok and there were no leaks.
The next day I was able to replicate the problem. The gage went over the center mark and fell back.
I called the mech. that did the gasket job. He told me to run it until hot and attempt to release air from the system. I have not taken the car on a trip as yet. I did test it again in my driveway and carefully watched the gage increase. At the same time I kept a close look at the fans to see when they came on. Suddenly when the gage got close to the second mark past center the two fans came on. The gage quickly came back down to normal.
I studied the wiring of the cooling system fans, temp sensor, relays and the PCM (computer).
Here is what I am thinking after this study.
There are three relays in the circuit. There is one signal and an associated ground wire that goes to the PCM. And there are two (what I think are) control signals that are suppose to control the fans.
There is a LOW SPEED Relay and a Hi SPEED relay.
So it appears from the diagram that Fan 1 and 2 runs at LOW speed when the computer activates the low Speed relay located in the engine bay. A the HIGH speed operation occurs when the computer energizes the other two relays via the HI SPD signal from the computer.
Under low speed 12 volts run both fans in series. In high speed they each get 12 volt in parallel.
My Fans seem to work fine but I don't know if they are being turned on soon enough to show normal operation.
I can not determine if I have two speed operation. It seems that the computer is supposed to first call for low speed and if need be, hi speed.
Could this be a computer problem?
How can this kind of computer signal be tested?
Under what conditions does the computer operate the Low and the high speed of the fans?
Is there a way to get that kind of information?

I wonder if the originator of this topic might have had a similar issue to the one I describe above. Could he have had a relay or PCM problem?
Tom