Overheating Continues

[TomsSVX]

Unless there is a coolant jacket that is clogged or anything else blocking the flow... I don't see what it could be. Hoses don't cause overheating if they are not broken or clogged

Tom

[shotgunslade]

Said a hose was cracked and had intermittent leakage. Was running a quart low on coolant, even though to me, the coolant level in the expansion tank looked OK. Said they ran it and let it idle for hours, checked the tank with HC sniffer, etc. No evidence at all of head gasket problem. Will know more after I pick up the car today.

[SVXRide]

Dan,
"Leting it run" and "pushing it hard" are very, very different conditions. If there was an "intermittent leak" I would have thought you'd have seen coolant sprayed somewhere in the engine bay....
Anyway, I hope I'm wrong and your overheating issue is fixed
-Bill

[dcarrb]

Again, I'm no mechanic by any stretch, but I believe coolant loss is a consequence of your troubles, NOT the cause. If the coolant level gets too low due to a leak, the car gets hot and stays there 'til you shut down; the temperature would not spike upward and then return to normal as you've described. You've got a problem with flow.

Forgive me if I missed this, but you've replaced the thermostat and water pump?

dcb

[shotgunslade]

Coleman Subaru checked out the car fully. No head gasket problems, water pump is OK. Seems that when I had the car tech-inspected by a local Princeton (non-Subaru) shop 6 months ago, they noted that I needed radiator hoses replaced and they did that. Seems they didn't tighten them well, both top and bottom, so they were both leaking. So I was losing coolant, but also seems, somehow, I was also getting air into the system, and evidently the temperature gauge spikes happened when a large air bubble was passing by the sensor. Also, may be why the coolant level didn't look so low in the overflow tank. Anyway, got it back and temperature gauge is dead level, no spikes at all. Took it up to 6000 rpm, needle dead level. When I was having problems, the spikes and the heaviness of my foot never seemed connected, anyway. So, they either fixed it or put in an hidden stop on the temp gauge.

[Earthworm]

Maybe they glued the gauge in position

How much did they charge you to tighten the clamps on the hoses?

How's your radiator cap working?

[shotgunslade]

They charged me $165 for the complete cooling system check out, drain and flush, and .. tighten the belts.

Don't know about the radiator cap. How do I test that?

[Earthworm]

Well make sure the spring is....springy?

I was thinking about how air can get inside a pressurized system. My only answer would be if it became a vacuum. The only way it could become a vacuum is after the engine is shut off the radiator cap becomes a 1-way valve allowing coolant to enter the overflow but not return. This is where the hoses come into play. If they're loose then under a vacuum they may allow air to enter the system.

I have no idea if this hypothetical situation is possible but I thought I'd throw it out there anyway.

[Hocrest]

I'm having the exact problem that Shotgun was having. The thermostat was replaced not too long ago with a Subaru one. I ordered a cap, thinking that may be the problem, but I haven't gotten to the dealership to pick it up yet...

[oab_au]

Originally Posted by shotgunslade

Said a hose was cracked and had intermittent leakage. Was running a quart low on coolant, even though to me, the coolant level in the expansion tank looked OK. Said they ran it and let it idle for hours, checked the tank with HC sniffer, etc. No evidence at all of head gasket problem. Will know more after I pick up the car today.

OK Dan I think you have found the cause. As you say they found a leak from the replaced hose.

What I reckon has happened, is when the engine got hot, and built up pressure, it pushed some of the water out the leak. When it cooled down, instead of drawing the water back from the expansion tank, it pulled air in the leak. This continued till it lost enough water to heat up, if the pump wasn't going fast enough, the water didn't circulate.

Moto of the story is to check the water in the radiator, every once in a while, not just the expansion tank. To do this you have to do it, in the morning before the engine starts. Open the cap, if the water is OK it will be full to the brim, if not add water to the radiator, replace the cap.

If you have had the water drained and are filling it back up. Check the radiator the next morning, as air will come up out of the block, and heater pipes. It is better when filling the radiator after a job, to jack the front up, so that the air comes out easier.

Harvey.

PS. DONT OPEN THE CAP WHEN THE ENGINE IS LOT

[shotgunslade]

Harvey:

Your observation about adding water to the radiator is very interesting. I did open the radiator cap a week ago, when the car was cool. Noticed that the coolant level wasn't full to the top in the radiator, then looked at the coolant level in the expansion tank and it was OK, and then thought that maybe that was the way it's supposed to be. So, the issue is that air can get into the coolant system, making perhaps large bubbles within the radiator, and the lack of coolant will not be reflected in the expansion tank.

I know that Bill remarked that he had experienced boiling out of his expansion tank at the AutoX after a run (not during). That behavior is similar to what I have experienced. This would indicate that when the engine is at low rev's or idle, the water pump is not turning over enough to force enough cooling water through all the passages, especially if, while running at high rev's it forced some air bubbles into the engine water passages. So, when you slow down after running the engine hard for a while, and the coolant flow rate drops, the coolant adjacent to those bubbles boils, because that area of the engine isn't adequately cooled.

Anyway, no overheating at all since the shop visit, which is important since overheating was always a relatively low-rev phenomenon. The engine has never been hot when I was running the p** out of it. Only afterward, or recently, just driving down the street. But, not now.

So, lesson to be learned (Thanks Harvey) Check the coolant level in the radiator (after it cools down), not the expansion tank

[SVXRide]

Dan,
Glad to hear things worked out. Where, exactly, were the leaks?
-Bill
p.s. I've installed a small billet overflow tank between the radiator and the stock overflow tank, as I couldn't get the fitting on the stock tank to keep from leaking while driving the p*** out of the car

[shotgunslade]

Bill:

Neither hose was replaced, so I suspect that the hoses were just improperly tightened. The hoses had been replaced by Tiger Garage, here in Princeton, when I took the car in on a Friday to fill out the required 3rd party tech inspection sheet for a PDA track day the following day. They noted the radiator hoses were cracked and I told them to replace them. I think the leaks were slow, because Coleman actually left the car idling for a number of hours and they said there was a small puddle on the floor.

Anyway, now I know to check the coolant level at the radiator, so minor leaks will not cause a major problem.

[Earthworm]

Originally Posted by SVXRide

p.s. I've installed a small billet overflow tank between the radiator and the stock overflow tank, as I couldn't get the fitting on the stock tank to keep from leaking while driving the p*** out of the car

What else have you tried? I get this happening to me during (after?) autox runs and I would like it to stop!

[shotgunslade]

Removing the grill helped mine, along with adding Water Wetter to the coolant. Also, driving around 20mph or so for 5-10 minutes after the run, rather than just shutting it down. Now will open radiator cap after the engine cools down and top off with water every time, .