Coolant overflow "overflowing"

[Wxman]

2005 Impreza RS 86,000 miles - upon arrival at work a few minutes ago, I smelled antifreeze. My drive to work is 30 miles on the highway (about 65 MPH). Popped the hood and found the coolant overflow bottle was filled to the top and some coolant had spilled out. About 10 minutes had elapsed since I was driving by the time I checked for a leak, but the upper radiator hose was already soft. I removed the radiator cap and the radiator appeared to be full as well. Is 10 minutes too fast for the system to lose pressure after 30 minutes on the highway?

 

I have not added any coolant to the system. I had an oil change at the local dealer about 3 days ago (had a cheapo coupon), and I don't suspect that they would have overfilled the bottle. This was my third round trip to work since the work was done, but the first day for the problem.

 

Any ideas? I'm thinking maybe the thermostat or radiator cap, but I've never had a problem with either of those on a car of this age/mileage.

[davebugs]

Sounding like the classic DOHC HG issue. But that's not what your engine is.

 

But perhaps you're having the same problem.

 

I'd try a cap first. Thats what controls what goes to the overflow bottle.

 

I don't have much experience with your enigne. It may be a PITA to burp just liek the older 2.5's.

 

Do a search here for "burping". If you don't get all the air out the remaining pocket will likely cause the same thing to happen again even if you have fixed the problem.

[98sub2500leg]

I agree with Dave, if your coolant is overflowing,most likely to be HG failure. Your mileage is right in the curve which from my research is roughly between 50k miles-135k miles on the upper end. Mine started failing at 60k miles.

You might want to have some diagnostics done to feel better 1st, but you will find this info to be pretty accurate.

Diagnotics:

1. Warm up car, feel lower rad. hose, if upper is hot to touch(or normal) and the lower isn't, don't assume it is the thermostat.

2. Test the thermostat. Remove the lower rad. hose, pull out t-stat, put in a clean jar or similar add boiling hot water. It should typically open between 169-176 degF and fully open at 196 deg F. If your going to replace, use OEM, which has a ball & chain to weep coolant in the event the t-stat is stuck closed preventing block warpage or cracking.

3. Run a compression check each cylinder. This may or may not indicate a leak at cold though.

4. Run a vacuum check, although again this may or may not indicate a leak.

5. Have hydrocarbons(exhaust gases) tested in the coolant.

6. Test your pressure of your radiator.

7. Test or replace rad. cap.

 

 

I am in the process of changing out mine. I decided to go all OEM parts and since I have 107k miles, I decided to change everything, all seals, gaskets, compression check & machine heads, new hoses, clutch, plugs, wires...$$$$$

 

By the way, the engine ran as good as when I bought it when the HG's failed. I got an additional 48K miles after the original leak was detected 5 years ago, although I wouldn't recommend doing what I did. It only intermittently overheated until basically the HG's failed on all four cylinders. Another big indication is sludge buildup on the inside wall of the coolant resevoir. It starts out with a trace, then over time the sludge buildup gets heavier. Some say you will see bubbles come up with either the rad cap off or in the coolant resevoir, but I didn't experience that.

For me the problem was always intermittent, usually stated on only some steep hills. temp gauge would go to 3/4 of full scale. I would immediately turn off the engine and the problem would reset to normal temp. I later found that the lower rad hose was cold after driving many miles on it daily which means the t-stat was stuck closed but the temp gauge showed normal. This was after I had already installed a new one and I also tested the old one which was still good). If it wasn't for the weep hole design in the t-stat I most likely would have a cracked head or block.

Really, you can do the diag. to feel better, but sounds like you have the HG problem. One way or another if will eventually fail at some point.

I wish you the best of luck.

[Wxman]

Ugh...not the news I was hoping to hear. I also have a 2000 Outback. The HGs went on this one at 125K, but it was the typical external leaking.

 

I thought the Phase 2 2.5 generally only leaked externally (and my Impreza is supposed to be after the better gasket was used).

 

Anyway, when I drove the car home this morning (30 miles on the highway at 65 mph though I made a few stops along the way to check on things), I made the following observations:

 

- The car heated up normally (temp gauge in the middle of the range, just like last night) and showed no signs of overheating.

- The heat worked well (though I didn't need it).

- While a little of the coolant from last night had been sucked back into the radiator, the overflow bottle was still overfilled.

- While warm to the touch, the system never pressurized (the upper radiator hose was soft and you could remove the radiator cap).

 

I still hope it is simply the radiator cap, but it is not looking good. I wasn't thinking internal HG failure, so I didn't check the oil yet. Hopefully I didn't do damage if, in fact, it is the HG.

 

Thanks and I'll let you know how things turn out.

[OB99W]

I thought the Phase 2 2.5 generally only leaked externally (and my Impreza is supposed to be after the better gasket was used

It would not be typical for a recent Phase-2 2.5 to have an internal HG leak. As both you and davebugs mentioned, the radiator cap is something to consider first. Since the problem became obvious just a few days after service at a dealer, it might be related to that work. While doing an oil change, most techs will check other fluid levels. If the radiator cap was removed, and some contaminent got into the cap seal area, it might be the cause of your problem. If the seal is compromised only a little, it could take a few days of the coolant expanding into the reservoir and not returning to the radiator for it to overflow.

 

I'd start by thoroughly cleaning the cap, including all the ''nooks and crannies'', and check for any irregularities in the seal. Also, clean and inspect the radiator neck, especially sealing areas. There was a TSB on this topic...

 

...okay, I did a Google search and found a link: http://www.thel7.com/TSB/09-45-06.pdf .

[bratman18]

As I have posted before, I have this same issue on my 2000 RS. It's the head gaskets. I tried the radiator cap, and it wasn't that. I bet if you look under your car, the HG's will be a little wet. Just make sure you keep the level up in your radiator and you can get away with it for a while. I actually use a turkey baster to transfer coolant from the overflow back to the radiator, it works pretty well

[OB99W]

As I have posted before, I have this same issue on my 2000 RS. It's the head gaskets. I tried the radiator cap, and it wasn't that.[...]

A 2005 is less likely than your 2000 to have bad HGs, but for the sake of this discussion, let's assume that they're the likely problem. In fact, let's assume that HGs are 10 times more likely than the radiator cap to be the problem. However, HG replacement at a typical shop is about 100 times more than the cost of a radiator cap. It's certainly worthwhile cleaning or replacing the cap, especially given that Subaru TSB 09-45-06 (which I linked to above) covers this specific possibility.

[Wxman]

Just got off the phone with the dealer. They provided me with three findings:

 

- The thermostat was stuck. This was suggested to be the primary culprit why the system was not pressurizing.

 

- The radiator cap was also having issues.

 

- They also said that the "AC cooling fan" was disconnected. At first he made it sound like it was the radiator cooling fan (which I noticed was not running the other morning, probably due to the lack of coolant circulating). Not sure how this would have happened and they were the last shop with their hands on the engine. I didn't even realize there was another fan aside from the radiator cooling fan and the main cabin blower motor.

 

The mechanic said I was lucky the car didn't overheat (and he saw no visual signs of an overheat condition) since the coolant was not circulating properly. He also said me running the heat on high when I drove it home may have kept the engine cooler. Since the coolant was not circulating correctly, do you think the temp gauge gave an accurate reading (or could the engine have overheated with the gauge in the "normal" range)?

 

We'll see if this corrects things, but he was able to get the cooling system to pressurize after making these repairs and he thinks all is well. I find it odd that both the thermostat and radiator cap were having issues around the same time, but I will not argue if I can truly avoid a major HG repair.

 

Thanks again with your feedback to this point and we'll see how it goes!

[bratman18]

A 2005 is less likely than your 2000 to have bad HGs, but for the sake of this discussion, let's assume that they're the likely problem. In fact, let's assume that HGs are 10 times more likely than the radiator cap to be the problem. However, HG replacement at a typical shop is about 100 times more than the cost of a radiator cap. It's certainly worthwhile cleaning or replacing the cap, especially given that Subaru TSB 09-45-06 (which I linked to above) covers this specific possibility.

 

Agreed!!! +100

 

I wasn't trying to rule the small stuff out, but as we all know, the HG's seem to follow a pattern of failure. And the SOHC 2.5 in the 2000 is basically the same motor as in his '05.

 

Good luck and hopefully those items that were fixed were the problems!!! Keep us posted