The water pump on my car was leaking, so I had it replaced.
The pump was leaking coolant on the serpentine belt, causing it to squeal, so I had the belt replaced, too. I also had the radiator hoses replaced due to age.
The car is a 1997 Monte Carlo with about 70k miles on it.
When I picked up the car, everything was fine. Drove 14 miles home, no problems.
The next morning, however, I started the car and drove it about 4 miles when the water temp gauge pegged out and the overheat light came on. Then, the temp dropped back to normal and stayed there, although it tended to fluctuate, whereas it used to stay pretty much in one place before.
I thought it might be the thermostat, so I had that replaced but it did not help. I also bled the air out of the system using the bleed valve on the thermostat housing.
Start the car cold, drive it a short distance, temp gauge pegs, light comes on, 30 seconds, temp falls rapidly to normal range and stays there with minor fluctuations. Can drive it all day after that.
At idle, it will not overheat. The fans seem to operate normally, kicking on and off when they should.
When I was 16 or so (A long long time ago) I installed a thermostat in my car backwards. It sure sounds a lot like that. It’s possible that an inexperienced mechanic was called to replace the thermostat that the original guys put in backwards, and just put it in how it was. An easy enough mistake.
Another less desirable possibility is gasses are escaping from the head and into the cooling system. But I’ll let the better mechanics of the SDMB come in for discussions on that unpleasant possibility.
The original thermostat seemed to operate normally. Winter or summer, rain or shine it would park itself at slightly less than halfway up the guage range.
It was only after the pump change that the problem started. There seems to be no indication of a head gasket leak (rough running, coolant loss, exhaust vapor, etc.)
What you say is possible if the mechanic, in changing the upper radiator hose, removed the thermostat housing as well (a distinct possibility) and put the original back in upside down. And then the subsequent mechanic replaced it, installing the replacement upside down.
It turned out to be a worn radiator cap that allowed some steam to escape, so I was losing about a cup or two of water per day. (I had just had a cooling system flush, so I “knew” that I had enough water in the system, and didn’t check that right off.) Eventually, the water level got low enough that, when sitting still, the temp gauge pegged out, but when I got the car’s speed up to 30MPH or so, the force of the air through the radiator was sufficient to bring the temp back down to the normal range.
Very alarming. And thank goodness that the fix was only an 8 dollar cap.
But, I could detect no steam escaping, and I monitored the coolant level in the overflow bottle. The hot and cold levels were normal, no coolant loss.
I am really leaning towards NurseCarmen’s response. When the first overheat occured, I looked at the engine and noticed that the thermostat housing, which is a tube-shaped casting about 6 inches long, was very clean compared to the rest of the engine. I think now that it WAS removed,to facilitate replacing the upper radiator hose, and the thermostat replaced upside down. I am going to take it in next week to have them check that out.
While I won’t say a backwards thermostat is impossible, I’m doubtful. Many designs are such that the thermostat cannot physically be installed backwards. If it were in backwards, I think it’s unlikely that it would ever open and allow the temp to return to normal.
My first thought is an air pocket in the cooling system. Often it’s difficult to get all the air out just using the bleeder fittings. Have you checked the coolant level in the radiator with the engine cold? Overflow jar levels are NOT reliable indicators of cooling system levels.
I agree. When I finally got around to checking the water level, it looked ok to the "eye’. (That means when I took the radiator cap off, and peered down the snout, I could see the top of the water surface a few inches down inside the radiator.)
However, when I added water, it just kept drinking and drinking, eventually about 2 liters went in. Yet, that water level near the cap fitting did not change at all, until the last few glugs of water. (I did not/could not fill it all in one gulp. I poured in a few cups and the water level would rise up to the top of the cap fitting. I would wait a few seconds, and the water level would drop back down as the water I just added worked its way into the system.)
Note: the owners manual for my car cautions against overfilling. Check yours.
Yes, I did remove the radiator cap with the engine cold, and the radiator was full.
I did bleed the system through the thermostat housing bleed valve. It’s possible I did not leave it open long enough because it was making a mess. There were some air bubbles at first, followed by a solid stream of coolant. There is supposed to be a second bleed valve at the 'heater outlet line" (?), but I was unable to locate it. That’s only in the Hayne’s book, the owners manual only mentions the one valve.
I’ve had coolant system work done on other cars I have owned, and there is usually some funny indications for awhile after the work is completed. I’ve put about 110 miles on this car since the pump change. Maybe not enough for air to be purged from the system?
Man, I hope they didn’t put in a faulty water pump. That would suck.
I considered that, though, and while possible, the system DOES cool the engine. It’s just that short time after starting the engine cold that the guage pegs, after that it’s (mostly) normal. Meaning there is some fluctuation, but it does stay in the normal operating range.
I did squeeze the top hose immediately after the overheat, and it was cold, meaning no flow from the radiator back to the engine. After the guage went back to normal, the hose was hot, which led me to consider the thermostat.