EJ25 overheating after rebuild

[98GT]

Here's the skinny. The engine in my son's 98 Legacy continues to overheat after we did a short block rebuild. I have done all that I can think of to do as to figuring out what is causing it.

The car warms up at idle-2000rpm just fine and cycles the fan and thermostat normally. But when I take the car out on the road, after several minutes, the gauge starts to climb and the upper hose gets swollen and really hot, while the lower hose stays very cool with no pressure on it. After turning off the car, you can hear coolant boiling in the engine and radiator, and it forces coolant out of the neck. into the overflow. Once it all cools down, it works like normal again until I drive it. Oh, and it starts running like crap once the temps go up. I've done several tests including looking for signs of a blown head gasket.

So. now I'm looking for some help, before I roll it out in the field and set fire to it. No, seriously, I'm thinking of burning it to the ground, and cutting my losses.

Here is what I have done:

Flushed the radiator and had good flow.

Replaced the new thermostat , with another and tested it before it went in.

Jacked the car up, and burped the air out of the system, which it did have a good burp after it was up to temp and I revved it a few time up to 3-4k rpm

Compression test. (Result: 190-200PSI)

Leak down test. (Result: Pass)

Coolant system pressure test with warm engine: (Results: System held pressure and had no sign of coolant in the cylinders)

 

The only thing I haven't done is replace the water pump, which was replaced just before he bought the car. I haven't changed it, because I'm thinking with the pressure in the upper part of the system, tells me the pump is working.

UPDATED EDIT: The solution to this problem was never found, as the car was traded off. However, there is a lot of good info here to guide you to a answer, if you're having the same problem,  just no final solution.

Here's a couple pics/video I managed to take

  Edit: The gold/brown stuff on the plugs, is anti seize, not oil

 

Edited November 18, 2018 by 98GT

[GeneralDisorder]

What brand is the thermostat?

[98GT]

Murray, I believe.

My boy bought a water pump today, but I'm still not convinced its the pump. I did spin it by hand when the belt was off, and It felt firm. Easy to turn, but no free spin.  Are these made like a boat water pump out of rubber, or are they metal fins like the older domestic pumps?

[GeneralDisorder]

You need a factory Subaru thermostat. They are made by Tama so if you find a supplier that sells those they will also work. ALL OTHERS WILL NOT WORK. That is almost certainly your problem. 

Your water pump is fine. The impeller is cast iron. 

GD

Edited November 9, 2018 by GeneralDisorder

[Mike104]

To make the above comment more obvious

[el_freddo]

+3 for genuine thermostat - every time!

Cheers 

Bennie

[98GT]

I'll call the dealer tomorrow and order one, and cross my fingers that it fixes the problem.

I'm not a long time Subaru owner, but I just don't get why these cars are so OEM dependent. And why, if things like the thermostats are so sub-par, the aftermarket isn't forced to pick up their game to make better parts.

Edited November 11, 2018 by 98GT

[GeneralDisorder]

Mostly it comes down to Subaru being a very small company and the aftermarket parts market doesn't care. They don't sell a lot of Subaru parts because all us professionals know better than to buy them, and that just makes the already small market even smaller and harder for them to get into. Gates and Stant have both tried selling rebranded TAMA thermostats and I guess it wasn't profitable because I don't see them available anymore. 

GD

[sparkyboy]

If you have time and the weather is suitable for you, remove the thermostat and see if the car functions normal otherwise. And someone can drive it if it's necessary.

I have experienced the same thing with my ej22 in my 86. These cooling systems are very sensitive to air bubbles and i found that using a vaccum filler provides the best results, but I have also found that the filling on an incline is effective. Occasionally squeezing the lower and upper hose so you can hear the jiggle valve will help remove any remaining air bubbles. 

Who puts the thermostat on the bottom like that? I understand the ej is very sensitive about spark plugs as well, ngk only bla bla bla 

[scar13]

I notice the 2 center plugs in the pic have white porcelain while the 2 outer ones are the proper tan color. the white ones(typically) were running hotter or leaner.

Do the t-stat. I'm curious, did you have the heads machined and did you use quality MLS head gaskets?

[scar13]

@sparkyboy-Honda uses the T-stat on the lower hose. 

[idosubaru]

On 11/11/2018 at 8:27 AM, 98GT said:

I'm not a long time Subaru owner, but I just don't get why these cars are so OEM dependent. And why, if things like the thermostats are so sub-par, the aftermarket isn't forced to pick up their game to make better parts.

the cars aren't OEM dependent - the aftermarket parts are just low quality.

The average consumer also keeps quality down.  If companies offered better parts for more money - very few people would buy them and it would be a capital loss. 

look no further than aftermarket axles - people keep buying trash axles, it's amazing. if a company started making high quality axles they'd cost a lot of money and no one would buy them.  those of us that care already have long term solutions to this issue and the rest would buy whatever is cheap. there might be a few sales - but they'd have a very low volume of sales not worth the tooling, manufacturing, listing, accounting, stocking of a new tier part.  shops wouldn't want to quote high prices to customers who don't know the difference and woudl call them "thieves" for quoting more than the shop down the road using cheap axles, so they'd have no incentive to use them.  then stores wouldn't carry them due to low volume, people would want whatever is in stock instead of waiting and thereby drive sales even lower. 

I don't own a global enterprise producing parts but even with small time business experience it's obvious this is a loose-loose endeavor. 

 

[98GT]

2 hours ago, scar13 said:

I notice the 2 center plugs in the pic have white porcelain while the 2 outer ones are the proper tan color. the white ones(typically) were running hotter or leaner.

Do the t-stat. I'm curious, did you have the heads machined and did you use quality MLS head gaskets?

Did the t-stat, and still over heating. It passed the first short test drive, on Monday, after I installed it. I was in the woods hunting yesterday, when my son took it for a longer drive, and it FAILED! However, he said the gauge pegged out on hot, then within 30 seconds, it dropped back to normal. I thought may be an air bubble passed by the sender, but he drove it again, and it did the same thing, however, he said the upper hose didn't swell this time. As soon as I get home from hunting this afternoon, I'm going to break out the digital thermometer again and get a couple more actual temp readings from the upper hose, the cylinders around the water jackets, AND the sending unit its self.

The plugs may have been a little lighter in color. The plugs have less than 100 miles on them. Not sure how long it takes to get a true reading.

As far as the heads, I did not have them machined, since the engine had no top end issues, and ran great until it spun a bearing. Plus, I checked them as well as the block for flatness. And yes, multi layer gaskets were used.

[1 Lucky Texan]

is coolant pushed out of the overflow when it overheats? is coolant pulled back in when the car cools?

[98GT]

1 hour ago, 1 Lucky Texan said:

is coolant pushed out of the overflow when it overheats? is coolant pulled back in when the car cools?

According to my son, it did not overflow or pull any from the tank this last "overheat". But before the thermostat swap, it was puking coolant.

Edited November 14, 2018 by 98GT

[whynot]

almost identical to what i went through. Spent 4 months trying everything under the sun, hoping it was not the gaskets.  it also would not run the overflow correctly. it would push into the overflow, but not pull from it. and kept building air in the cooling system over time.

after overheating again, and blowing water out of the top of the radiator again, I did them again, and used GD's method for machining at home. hope you have better luck than i did..

 

Edited November 14, 2018 by whynot

[golucky66]

Bypass the heater core. Restricted heater cores can reduce coolant flow to the back side of the t-stat not allowing it to open.

 

[98GT]

That's not a bad idea. The heat seems to work fine, so not sure it's blocked, but easy enough to give it a shot!

[1 Lucky Texan]

until this issue is fixed, never trust the o'flow bottle level to reflect the actual coolant level in the radiator - rad must be checked separately. Keep o'flow bottle at a normal level.

new rad cap might be good to try. several drive cycles could move a few ounces of coolant into the o'flow , over time if the cap or other problem isn't allowing those ounces to be pulled back, it could fill the o'flow....

if 'localized' boiling is the issue, fluid should be pulled back in. If bad headgasket is allowing combustion chamber gasses in the system, fluid may not be pulled back down to normal level in the o'flow.

 

put a coupla small zipties on the o'flow tube at the rad neck, take the o'flow bottle out, clean it by vigorously shaking a handful of ice cubes around in it, replace but cut the bottom of the tube at a 45* angle.

 

Edited November 15, 2018 by 1 Lucky Texan

[carfreak85]

54 minutes ago, 1 Lucky Texan said:

Put a coupla small zipties on the o'flow tube at the rad neck, take the o'flow bottle out, clean it by vigorously shaking a handful of ice cubes around in it, replace but cut the bottom of the tube at a 45* angle.

 

Hmmm, I like the ice cube idea, is that just to get some abrasive in there to clean out shmutz?

Also interested in the 45* cut on the tube.  What's the reasoning behind that?

[98GT]

The 45 prevents it from suctioning to the bottom of the overflow tank.

[98GT]

The overflow and cap are working properly. I've watched the level come up, then go back down several times.

Edited November 15, 2018 by 98GT

[1 Lucky Texan]

2 minutes ago, 98GT said:

The overflow and cap are working properly. I've watched the level come up, then go back down several times.

then if the HG is bad, there must be bubbling of cc gasses coming out of the cooling system when it's overheating - those bubbles 'should' change the color of a block test fluid sample from blue to green/yellow.

any bubbles?

[98GT]

1LT, I have not put any block test fluid. To be honest, I've never heard of it, and had to look it up. The first thing that came to mind was the ole blinker fluid trick lol.

I got home late last night, and was cold from being in sub freezing temps for two days, so I didn't even consider going out in a semi cold shop last night.

But I'm out here this morning, got the wood stove rolling and I'm fixin to tear into the shitbox again. Just to twist the knife in a festering wound, I found this when I came out this morning. Yeah. He's dreaming on the price, but at least I taught him to start high, because it's hard to get more that you ask for...

[1 Lucky Texan]

oh well, aside from intellectual curiosity of bystanders like myself, just getting away from a headache is sometimes the sensible thing to do.

 

honestly, I personally wouldn't have the persistence I see among many folks who post here if I were in their shoes. I mostly just throw out ideas - but, I'm not actually living with the frustration!

Edited November 15, 2018 by 1 Lucky Texan