cooling system question

[FFRWRX]

First, I did a very minor mod to my cooling system. I made a spacer plate to move the degas tank a little further from the engine. Yet another brilliant design, put a frame member right over a filler cap. But that's not why I moved it.

Screenshot 2023-08-05 at 6.08.48 PM.jpg

The reason I did it is that the hose that the arrow is pointing at was kinked...badly...maybe even completely closing it. Pretty sure this is all stock Subaru stuff. So moving the tank a little eased the kink in the tube.

Now, the question. My car is just legally on the road now, so I've gone for several short drives. Everything was fine...until it wasn't. The car was overheating quickly and badly. I didn't venture far from home, so not a huge issue. But really annoying. Then I remembered that I turned the heater on for the first time to see if it worked on a previous drive; not much heat out of it but I wasn't concerned. The A/C worked very well so I was pleased with that. So it occurred to me that the heater was filled with air and turning it on added a lot of air to the cooling system. After relocating the degass tank I checked the coolant in it....pretty much none. Filled it up and went for a drive. Of course it still over-heated, but when I got home and let it cool the tank was empty again. Filled it, started the engine and revved it a number of times, tank level very low again. Repeated that a number of times. Then let it warm up completely and turned it off. I'll check things when it cools completely.

Oh, the question (finally they say). The degas tank is supposed to refill itself if needed from the overflow tank. Everything is properly connected as far as I can see. But looking at the overflow tank, it is sealed; I can't see any vent hole in it. So how is the system supposed to suck fluid from that tank when it is sealed? It would have to be vented in order to allow that to happen. And that must be why my degass tank doesn't refill itself when cooled down. Must be vented somehow that I'm not seeing.

Rick

edit: After searching around I'm finding different names for the same components. The metal one I'm calling the degas tank is sometimes called the expansion tank, and even the overflow tank. I think expansion tank is the more common name. And maybe overflow tank for the plastic one?

[lance corsi]

Rick, can’t speak from experience but it sounds like you’re on the right path.

[mikeb75]

FFXWRX, can't see from the picture you posted, but is the outlet to the overflow tank part of the rad cap spout of the upper reservoir (degas tank)? From how I understand it, the upper reservoir's main purpose is to have a coolant source higher than the turbo. After running and engine shutdown, this coolant can still supply the hot turbo to allow it to cool down without the need for a post-run idle period.

If your coolant system has 2 radiator caps, the one on the reservoir should be the lower pressure one; stock caps are 1.3 bar for the radiator and 1.1 bar for the reservoir. When the pressure exceeds the rating on the reservoir cap coolant flows out to the overflow tank. Conversely, when the system begins to cool, as long as the pressure is still higher than the reservoir cap, it should siphon coolant back from the overflow tank - as long as the overflow tube is below the level of coolant in the overflow tank.

So I'd check 2 things: that you have the lower pressure cap on your reservoir and that the overflow tube goes all the way to the bottom of your overflow tank. Hope that helps.

[FFRWRX]

My radiator has a cap that basically seals it, no pressure relief. There is a fitting at the top of the rad with a hose going to the metal expansion tank. Makes sense about the turbo cooling, since I think the hose that was kinked was going to the turbo.

My metal expansion tank has a 1.3 bar cap on it. The tube in the overflow (plastic reservoir) does go below the fluid level. Are you suggesting that the plastic reservoir has pressure in it when the engine is running? Doesn't seem like a design that is supposed to be pressurized. I thought, when I wrote the first posting here, that the overflow tank was not vented. I don't see that it has a dedicated vent, but looking at it again the swivel connection for the hose is certainly not air-tight. So I don't think it can maintain pressure.

The cap on my metal expansion tank has the central valve to let coolant back in. But it never seems to do that. I checked it this morning after cooling overnight and the level was a little low. I topped it back up, went for a drive, and no overheating. So I think I finally got all the air out of the system. Not a big issue filling up the expansion tank when needed, but it seems that the plastic overflow tank doesn't do anything.

Rick

[mikeb75]

The plastic overflow tank is not pressurized - it's a relief for excess pressure in your coolant loop (gas or liquid). It will only return coolant to your upper reservoir by siphoning it- which is why it's important that the connecting tube is below the static coolant level in the overflow tank.

Glad you got the air out; these systems are notoriously difficult to do so (it's not just the 818, Subaru EJ's are also tricky).

[Sgt.Gator]

I need more pics of your entire cooling system to really understand what is going on.
Have you done Wayne's Burping mod? Post #8 in this thread.
https://thefactoryfiveforum.com/show...ighlight=Wayne

When we refill the cooling system we always raise the front of the car a few feet off the ground.

There are numerous cooling system threads.

https://thefactoryfiveforum.com/show...ighlight=Wayne

https://thefactoryfiveforum.com/show...ighlight=Wayne

https://thefactoryfiveforum.com/show...ighlight=Wayne

https://thefactoryfiveforum.com/show...ighlight=Wayne

And many more, but your answer is probably in the above.

[FFRWRX]

Yes, I've done the Wayne mod. And I've read most, if not all, the info on cooling systems and how to get all the air out. I have the outlet at the top of the metal expansion tank going to the plastic reservoir, below the coolant level. The expansion cap has the 1-way valve in it to let coolant back in. So while I do seem to have all the air out now, it doesn't suck the coolant from the plastic reservoir when it cools down. I did take the hose off the metal tank and suck some coolant from the plastic reservoir just to make sure everything was clear....tasty stuff.

Thinking about it as I write this. If the system is a bit low on coolant and when running gets up to just less than 1.3 bar, the system doesn't open at all. So when it cools down the only way it can suck coolant from the overflow is if it goes below atmospheric pressure, which it won't. If it does open up when running to spit out extra coolant, then closes and cools down, it will be below atmospheric pressure and suck in the coolant it lost. So maybe it makes sense that it won't suck in coolant from the overflow unless it lost some coolant while running. Mine never lost any (as far as I know) so it wouldn't replenish itself automatically.

Rick

[Sgt.Gator]

Maybe this will help. Jon at Flatirons had been plagued with cooling issues on their race car. I noticed he had the wrong radiator cap on his radiator. That brought about this video:


I see you have the Mishimoto cap on the burp tank like it should be. On my 818 the radiator cap is the oem cap that vents to an expansion tank next to the radiator up front. But there is no return back into the radiator, it's a one way valve only. If everything is staying cool there is nothing in my front radiator expansion tank.
I have a second expansion tank in the engine bay connected to the burp tank.
Here's my burp tank connections. Note the tee that goes down to Wayne's mod. The line from the nipple under the cap is going to he expansion tank located in the engine bay behind the driver's head.
oil and boost vac sensors 818.jpg

[J R Jones]

I watched a Flatirons video about Subaru oil pumps recently (Dave53 engine failure) and now this video on cooling systems. In engineering reports I expect conclusions and a summary of findings, but not with Flatirons.

Two successful cooling system experiences for your reference:
1) Roadracing SBF engines I used the SBC (Corvette) system as it worked well for them. It is simple, engine, radiator, coolant recovery tank.
https://www.amazon.com/DeWitts-Corve...4-dc28e6d374b3

Plumbing is the 5/8 (bottom) hose bib to the heater hose bib on the engine (high and forward on the engine). The 5/16 side hose bib to the radiator tank bib (no cap). The fitting under the standard two-way pressure cap went to my catch tank, and it could go to a standard recovery tank (bottom) bib, but I question if it would be required.
The aluminum tank is supplemental volume for the system. It bleeds air from the engine and the radiator and after shut-down, it feeds coolant back to the system. Simple

2) the Toyota MR2 mid engine cooling system is similarly simple with bleeders to purge air at the front. The radiator has a 3/8 hose bib with screw valve and no cap at the top of the tank.
The heater core also has a 3/8 hose bib and screw valve. Tigon hoses are attached to these two hose bibs to be stand pipes for (visually) purging air with the screw valves open.
The coolant recovery tank has a two-way cap and is at the top of the engine compartment with the bottom plumbed to the top of the engine.
Open the valves and recovery tank cap and fill the system. You can squeeze the radiator hoses to purge air or start the engine. When the bubbles stop appearing in the stand pipes and recovery tank, you are done.
This guy goes a step beyond by jacking up the rear of the car to encourage air purge to the rear:
https://www.youtube.com/watch?v=S4RuMaumw_0

These examples are more simple than the Subaru and 818. They work for me.
jim

[FFRWRX]

I don't have any issue with my cooling system now. My latest point was that it won't replenish the coolant level, if it is low in the expansion tank, if it hadn't spit anything out when running. Hot and cold cycles will get the air from the system and into the metal expansion tank, but won't draw coolant from the plastic overflow bottle.

Rick

[J R Jones]

Rick, I am guessing here without knowledge of the plumbing. It is harder or impossible to pull coolant into the expansion tank it the (refill) line is below the fluid level. The refill location should be in the air above the fluid and I expect that is best right below the (two way) cap.
BTW in the Flatiron discussion they mention two different pressure caps. I expect only the low pressure cap functions for pressure relief. Both will respond to negative pressure and draw whatever is available, air or coolant.
I would use one double function cap with a fluid recovery and if necessary a second cap for filling but no pressure function.
jim

[FFRWRX]

Both will respond to negative pressure and draw whatever is available

Yes, but there will be NO negative pressure unless coolant was lost when the system was hot. Heat it up, cool it down, and the pressure does not go negative.

[J R Jones]

I think I have been troubleshooting a problem that does not exist. If your expansion tank has coolant and some volume of air, the coolant can expand without blowing out into the overflow tank. Therefore it does not need to refill.
My cooling systems function that way, the overflow tank is static unless extreme circumstances displace coolant into the overflow, and back as the system cools. Larger expansion tanks are more forgiving in that regard.
jim

[driveslikejehu]

I'm pretty sure it does make negative pressure (remember, all that means is less pressure in the container than ambient) when it all cools down. With heat all fluid in the system expands, maybe goes to overflow, or maybe just right. When it cools, the fluids will reduce in volume creating a negative pressure that will suck fluid in from the overflow. My 818(all Subaru system) does this.
A related example is when I return a partially empty water bottle back to the fridge after reaching room temp. When I get it out, the thin plastic bottle has crushed in a bit.
I agree with Jones that it reaches an equilibrium with the right amount of air in the swirl pot; the air is compressed with fluid expansion and acts as a buffer so coolant isn't pumped in and out.
My $.02

[FFRWRX]

No. Take your partially empty water bottle and put it in the fridge. It crushes in a bit. Now take it back out and leave it on the counter until it comes back to room temperature. Does it stay crushed in? No. It is simple physics. Take sealed container (like a cooling system) heat it up and the pressure goes up. Cool it down and the pressure goes back down, but not below ambient temperature unless some coolant was lost in heating it up.

[driveslikejehu]

Ed Zactly...