Expansion/Recovery Tank Plumbing Doesn't Seem to Function As Expected

[Pasha Saleh]

I am wondering if my Canton expansion/recovery tank setup is working properly. I am using a Canton 80-200 Fill Tank at the highest point in the cooling system and have the barbed fitting in the side plumbed to the bottom inlet of a Canton 80-201 Recovery tank placed down low alongside the ZF Transaxle. Radiator cap is 13psi on the fill tank and the recovery tank has its own cap that I am told vents outward but not inward.

The idea was to allow heated/expanded coolant to flow out the side nipple of the fill tank and down into the recovery tank, then suck back in as the engine cooled. I filled the recovery tank halfway with coolant and expected this level would rise when the car was warmed up, then drop back to half when cool. In practice, there is no level change whatsover.

Today, after a drive that got the coolant up to temp, I let the car cool down then opened the radiator cap on the fill tank, expecting the level to be right flush with the inlet to the side nipple leading to the recovery tank plumbing. Instead, a bunch of slightly pressurized (but cool) coolant flooded out. Why was it not forcing its way into the line leading to the recovery tank? I pulled off that line wondering if it was plugged and am able to blow in it and cause bubbles in the recovery tank but it takes a decent amount of lung power to do so. When I topped off the fill tank, I intentionally filled it past the level where the inlet to the side nipple is, again expecting it to flow in there like a sink or bathtub overflow. Instead, it filled right past it (see pic) and would have just poured out of the fill tank if I hadn't stopped. Why is this excess fluid not going into the overflow line leading to the recovery tank? Isn't that the whole point of this design?

Also, how can it ever function when the radiator cap rubber plunger goes down past the level of the side nipple, effectively blocking it? I called Canton Engineering and their own tech support couldn't answer this or explain why/how it works. Would this system only work if the pressure were so high that it was threatening to blow the radiator cap and then it just pushes the cap spring enough to vent some fluid into the overflow line? If so, what would ever cause it to "suck back in" to the fill tank once cooled off? The radiator cap spring and rubber plunger would again be blocking the side nipple/port so how could the coolant ever "recover" back into the fill tank?

I have either screwed up the plumbing of this somehow or am misunderstanding how the system is supposed to work!

 

[Joel K]

Pasha,

Is your overflow tank vented? It should be. If not the water from from the fill tank will not travel down to the recovery tank.

Also, water will only travel to the recovery tank when the pressure in the fill tank cap as well as the water level is exceeded.

 

[Pasha Saleh]

Thanks, Joel, for the quick response! Here is the overflow tank I am using: https://www.cantonracingproducts.co...num-recovery-coolant-tank-universal-2-qt.html

The description says the cap is vented and I can blow into the tube and get bubbles/flow although it takes some effort. Same effort when the tank is left off so it seems it must indeed be vented. If the coolant only travels to the recovery tank when the pressure exceeds the cap pressure, what would ever cause it to "recover" or draw back into the fill tank? Doing so seems impossible since once the system has cooled, the radiator cap doesn't have any pressure forcing the spring open so it will just block the side inlet port at the top of the fill tank. Is it meant to just be a one-way system, meaning it allows over pressure to spill into the recovery tank until that tank is eventually totally full?

 

[Joel K]

Cap may have a sintered stone as a vent so there might be a little resistance, but that should be fine.

So here are a couple scenarios on how it should work.

1)Say you start with a Fill tank at 3/4 full and the recovery half full. Now you run the engine and as the car warms up and the coolant expands, however if it does not expand enough to build enough pressure to trigger the cap no fluid will be sent to the recovery tank.

However when the fluid cools after you shut the car off, it will form a vacuum in the fill tank. That vacuum may open the fill tank cap, it does not take much vacuum to open it. Then fluid from the recovery tank will be drawn back up to the fill tank. Of course if there is no fluid in the line connecting the fill tank and recovery tank then it will just draw the fluid up the line. So in this scenario over time water moves from the recovery tank to the fill tank.

2)Say you start with a Fill tank 3/4 full and the recovery 1/4 full. Now you run the engine and as the car warms up the coolant expands, if it expands enough to build enough pressure to trigger the cap, fluid will be sent to the recovery tank.

Then when the fluid cools after you shut the car off, it will form a vacuum in the fill tank. That vacuum will open the fill tank cap, it does not take much vacuum to open it. Then fluid from the recovery tank will be drawn back up to the fill tank.

 

[Pasha Saleh]

Hmm. So does that mean that the radiator cap on the fill tank works in both directions? Meaning, at 13 psi positive pressure, it will open to let OUT the excess pressure in the cooling system. But under vacuum from below (as in your example with the 3/4 full fill tank that is cooling, it will also open INWARD to let IN fluid from the recovery tank? I thought it just worked one way, with the spring force keeping it "plugged" at 13psi and an overpressure (in excess of 13psi) pushing against the spring to let out the coolant. I'm not clear on how a vacuum would let fluid into the fill tank since it would work in the same direction as the spring force on the radiator cap which is acting to push the rubber plug into the filler neck of the fill tank and would seemingly just further plug it and block any influx of fluid from the recovery tank.

Maybe I'm overthinking this system...

Any harm in keeping the fill tank topped all the time?

 

[Jac Mac]

Your cap got two rubber rings ?, one that goes down & seals in the inner recess, other that seals on top face of filler neck, should have so that as engine cools it will suck coolant back up & into overflow tank. Metal to metal wont work. Ha Ha, two guys typing same reply!!

 

[Pasha Saleh]

Thanks, guys! I am missing something here. If you look at the radiator cap in the picture (standard Stant 13psi), it does indeed have the 2 rubber rings as Jac Mac describes. But the smaller diameter ring is so thick (see profile) that when plugging the filler neck, I would think it also completely blocks the overflow/standpipe hole at the side of the filler neck. Look how close that overflow hole is drilled to the bottom of the filler neck opening. It is barely 1/16" away -- certainly less than the thickness of the plug portion of the rad cap.

Therefore, under vacuum, I still don't see how any coolant could re-enter the fill tank from the overflow line since the thickness of the radiator cap "plunger" would be right up against that opening. I can see how positive pressure in the cooling system would "unport" the overflow port as the cap begins to vent but not the reverse.

Oh well, as long as I've got things rigged properly, I will just let it be!

 

[Joel K]

Hmm. So does that mean that the radiator cap on the fill tank works in both directions? Meaning, at 13 psi positive pressure, it will open to let OUT the excess pressure in the cooling system. But under vacuum from below (as in your example with the 3/4 full fill tank that is cooling, it will also open INWARD to let IN fluid from the recovery tank? I thought it just worked one way, with the spring force keeping it "plugged" at 13psi and an overpressure (in excess of 13psi) pushing against the spring to let out the coolant. I'm not clear on how a vacuum would let fluid into the fill tank since it would work in the same direction as the spring force on the radiator cap which is acting to push the rubber plug into the filler neck of the fill tank and would seemingly just further plug it and block any influx of fluid from the recovery tank.

Maybe I'm overthinking this system...

Any harm in keeping the fill tank topped all the time?

Yes, the cap works in both directions. Take the cap off and you can see how easy it is to pull down from the inside. Much harder to push open, since that requires 13 psi to Force fluid out. When the coolant cools creating a vacuum, it’s like drinking threw a straw and draws coolant up from the recovery tank.

 

[Joel K]

Thanks, guys! I am missing something here. If you look at the radiator cap in the picture (standard Stant 13psi), it does indeed have the 2 rubber rings as Jac Mac describes. But the smaller diameter ring is so thick (see profile) that when plugging the filler neck, I would think it also completely blocks the overflow/standpipe hole at the side of the filler neck. Look how close that overflow hole is drilled to the bottom of the filler neck opening. It is barely 1/16" away -- certainly less than the thickness of the plug portion of the rad cap.

Therefore, under vacuum, I still don't see how any coolant could re-enter the fill tank from the overflow line since the thickness of the radiator cap "plunger" would be right up against that opening. I can see how positive pressure in the cooling system would "unport" the overflow port as the cap begins to vent but not the reverse.

Oh well, as long as I've got things rigged properly, I will just let it be!

You should be able to just pull the inside metal center down opening up when vacuum forms.

 

[Joel K]

Here is a good video on how the cap works in both directions..,

 

[Pasha Saleh]

Got it! Schooled once again by the Forums! Thanks very much. Didn't realize the center section of the valve was hiding a second valve that opens inward. Great video. I ended up finding this diagram which clarified where the "vacuum" valve is:

 

[Ryan L]

I bought a ford radiator cap from that last of the AU falcons that used that style. think it was 13PSI that it was rated at. This is what ford used with the last of the Windsor engines they sold in Australia.
As for the overflow vent cap, you will need to check that it is functioning properly as well, and able to flow a vacuum as well as well as purge air when the pressure cap opens.

Mazda used a similar system with almost identical caps on the FD RX7. the caps were interchangeable across the two tanks, but functioned differently as you can imagine. This has been known to cause some issues with the cooling systems on these cars when they get mixed up.

 

[Kyle]

Just for anyone in the future. I’m not using an overflow tank. Only an expansion tank that is filled about halfway when cold. When hot it will only expand so much. As long as you have a solid system with no leaks and don’t let it overheat, you shouldn’t ever have to worry about volume. I’m sure track cars need a system but. I like to keep things as simple as possible for less failures.

 

[Terry Oxandale]

I believe he "recovery" diaphragm is simply a flapper valve that operates in one direction, blocking any flow when under pressure, but with no resistance to reverse recovery flow (lower pressure in the block drawing coolant back into the block). So I would ensure you have a "vented" cap on your tank, and not a "non-vented" cap. The smaller diagram should be loosely attached on a vented one (I believe).

 

[Pasha Saleh]

Just for anyone in the future. I’m not using an overflow tank. Only an expansion tank that is filled about halfway when cold. When hot it will only expand so much. As long as you have a solid system with no leaks and don’t let it overheat, you shouldn’t ever have to worry about volume. I’m sure track cars need a system but. I like to keep things as simple as possible for less failures.

This is good to know! Too late for me but explains why I am not seeing any overflow into the recovery tank. Must have a "good system" then!

What I do observe is that the coolant temp seldom gets about 70C (158F). I have a 180 degree thermostat in there so it would seem it never even opens. Any concerns I should have about this? I have heard too cool is not good. Using Webers.

 

[Lee Patterson]

I was of the understanding that an overflow tank was necessary as the expansion tank should be filled full with water with no air gap left. Which means a recovery tank is essential. At least that is how my Pantera operates best. What did the original 40 use? No overflow tank?