Over the last 8-9 months I have been dealing with a really weird coolant issue with my 1998 Legacy Outback (2.5EJ DOHC). The car pushes coolant into the overflow tank…but it never returns into the system once the car cools down.

Amount of coolant pushed is directly proportional to how hard I drive the car. Under normal DD circumstances, it takes couple months to fill up the tank to a point where the temp needle starts climbing. If I cool the car down and refill the radiator, everything goes back to normal and the cycle continues.

Here are the things I have done to fix the issue thus far…

- Recent (< 1 month) compression/leak down: (1) 235 psi / 98% (2) 234 psi / 97% (3) 228 psi / 97% (4) 235 psi / 98%.
- Radiator pressure test – held 14 psi for 2 hours without any drop or visible leaks.
- Bled the system multiple times (personally and professionally).
- New radiator cap.
- New radiator hoses and clamps.
- New temperature sensor.
- New overflow tank and hose.
- Paid to have the water pump and thermostat inspected (latter was supposed to be replaced).

Misc. odds and ends...

- It will only do it while I drive it. I have idled the car for 3 hours under mid-day sun...nothing. Fans worked as expected and coolant level didn't rise.
When not under driving load, the temp needle drops about a notch but goes back to its normal place once I take off. This only happens when some coolant has already been pushed.
- The car does not smoke under any circumstances and it has a cat-less header back exhaust system.
- Coolant does not froth nor has signs of mixing with oil.
- There is no goopy jell under the oil cap.
- It does not eat oil beyond what’s expected of N/A high mileage Subaru.
- All fluids are Amsoil.

Engine aftermarket mods...

- Header, race pipe, muffler shop exhaust.
- GrimmSpeed lightened pulley.
- Denso iridium plugs and wires.
- A/C delete
- The end

After all of that the issue remains…dear lawd it is driving me insane!!! I finally decided to post on here for any ideas.

If you tell me “Head gasket!” please support your answer and/or how I can confirm your logic.

 

When I was pushing coolant into my overflow reservoir it was cylinder walls in 2 and 4 being cracked. Boost was compressing the cooling system and forcing it out of the overflow tank.

 

Would that issue go undetected by comp/leakdown? I rarely push the engine very hard (contrary to what some people might say )...I auto crossed it couple times, but besides that I really don't beat on it (high rev shifts, over revs, etc).

 

Possibly, I am not entirely sure. Depends on how bad the crack is I suppose.

 

Head Gasket!

I'll support my answer by stating it's head gasket simply because you've clearly answered that's what it is within your post by using just about every single test that has no chance of showing the signs of a small head gasket leak in fact I'd go as far as to say you'd have a better chance of finding rocking horse ****!

 

FYI I am taking the car to RBMS on Monday for them to look it over as well...

 

 

this is the right answer

 

Fixed.

It might not be at the head gasket, but this is definitely the mechanism of failure.

Some section of higher pressure has breached the limits of it's system and broken into the coolant system... and is expanding it every time you rev the motor up. Since you said there is no evidence of oil in the coolant, the only other option is air.

There are testing kits you can buy which will tell you if there are hydrocarbons (fuel) in a given fluid. Test your coolant and see if there is air/fuel mixture from inside the engine contaminating your coolant and you'll know you have a leak. Then pull the head gaskets and find it.


... I guess in theory it might be possible that you have a crankshaft case ventilation port plugged directly into your cooling system somewhere but that would be crazy.

 

Or, he can have Mert do a leakdown test and tell you EXACTLY what the problem is.

Jesus people do it right or don't do it at all. Sometimes this place is like watching retard kids at play.

 

He did a comp/leakdown test already and posted the results above.

Comp/leakdown will not identify a head gasket leak if it's small enough just like Mike stated. Nor will it tell him exactly what the problem is. It wasn't until Mert tore my motor down that he was able to find the cracked cylinder walls.

 

Possible test for exhaust gasses in your coolant.

 

Leakdown or O2 coolant test will not always work 100% of the time but rather 99%. There are hair line cracks or seepage through thinning of the head gaskets that are only exposed during high heat and high pressure conditions. I've seen it before. So don't be too surprised if the issue is not detrimental enough to show through the leak down, yet you are still experiencing the reoccurring issues.


As Mike already nailed on the head, most probably cause for this scenario are the head gaskets or at least a leak within the heads and cylinder block.

 

One other suggestion I received is a cracked/faulty heater core. My blower/heater have been working like ****; oh highest warmest setting it still feels like lowest lukewarm setting when it was working well. Probably a long shot but still...

So what about my trouble shooting has not been right? I started with smaller possibilities (cap, hoses, tank) and worked my way to more major ones (leak down, water pump, etc).

I have actually taken the car to very competent mechanics that have done good work for me in the past. It was just more convenient at the time and I was hoping for was a more straight forward answer...it is not exactly a complicated billion horsepower heavily modified car.

I actually knew, but totally forgot about this! Mert reminded me about and I will do one myself just in case before I drop it off. The kit is already in the mail.

If that's the case It is what it is. I will drive it until it is done for and in the meantime start saving for a another DD. I am over putting money into it.

 

Give me the car with the stoptech on them!