MeanMini 's cooling system anomaly: coolant push to overflow

[contraption22]

My Lebaron has never really kept the radiator full, it always seems to push it out and not pull it back in. However, it only seems to do this until the level is a couple inches down from the neck, after that seems to stop losing it.

My Jeep was doing this. Every time I would pull the cap to check it, it was down a bit. I was afraid it was consuming coolant. Turns out that the hose to the overflow was a loose fit on the replacement radiator. When the coolant would expand, it would push out past the cap into the reservoir, but instead of being drawn back in as it cooled, air would get drawn in. I put clamps on the hose at the filler neck and the reservoir and it was never an issue again.

[wheming]

And as Gary said, that drawing in of air (when the system goes into vacuum when cooling) can happen at any fitting/hose clamp.

[johnl]

Great thread. It's elementary, but it bears saying, purge the system in order to get a baseline coolant level, or at least a stable level, so that you know exactly where the coolant level was at. That is, sometimes the level that appears through the radiator cap changes just because air bubbles are freed from somewhere during operation and that sets up questions about a leak, unless you've done several purges.

[wheming]

Good reminder John.
I'm sure I'm not the only one who dislikes how the minivan heater core is so high in the system. I'm sure that Brian had purged all the air out when he performed his work, and i know I tried to be sure of the same when i replaced the radiator.
I think one thing that wold be handy is some vent ports on the heater core lines.

[contraption22]

Can you add a length of heater hose to a new high point in the system? Use one of those heater hose flush n fill kits to add a fill cap at the high point.

[mopar-tech]

Can you add a length of heater hose to a new high point in the system? Use one of those heater hose flush n fill kits to add a fill cap at the high point.

I always just cracked a hose open at the highest point while it was running or jack up the front end.

[wheming]

I always just cracked a hose open at the highest point while it was running or jack up the front end.

I've done both of these too.
I like the idea of a vent valve there at the high point.
There are automatic (mechanical) air purgers, they are often on reheat type industrial water systems. They will allow air to escape, but prevent fluid loss. I considered trying something like that at some point. Of course, once the air is out, then you don't need one. Unless you are getting air in the system somehow.

[mopar-tech]

If you are getting good heat out of the core I would not worry about it Wayne. I never saw a vehicle that failed to clear the heater core after running it briskly down the road..

[crusty shadow]

Like others stated above, you got a leak of some type somewhere.
pressure test the cooling system, build to cap pressure, and let it sit for 2-3 hours.
any external leaks except for the overflow hose and overflow tank will show themselves by then.
if the pressure drops noticeably but you have no external leaks there is another type of cooling system pressure test that will 100% confirm if its a headgasket or not.

connect cooling system pressure tester , pressurize the system to 2-3 psi, just enough to move the needle off zero. disable the ignition coil and injectors then crank the engine for 4-5 seconds. if the pressure tester gauge went up you have a leaking head gasket or a crack in the block or cylinder head.

connect cooling system pressure tester same as above, pressurize system to 2-3 psi. remove 1 spark plug and rotate to TDC for that cylinder, use a leak down tester or a fitting connected to shop air and pressurize the cylinder for 30 minutes. if the cooling system pressure tester gauge goes up that cylinder is leaking. repeat the test for the other 3 cylinders to pinpoint which ones are leaking. this test takes a bit longer to run than the cranking test but it works really well to pinpoint small or hard to find leaks. I've found head gasket leaks with this method that did not show up using the color changing fluid block tester method.

For getting air out of the system i cant speak highly enough about the vacuum cooling system fillers. Quick, easy and it gets all the air out of even hard to bleed, no need to burp air out for 45 minutes after, 5-10 minutes from empty to full and you're done. if you decide to buy one dont get one that has the expanding rubber cone type radiator cap mount, get one that has a positive lock setup like a cooling system pressure tester.
this is the one i have, i love it http://www.cornwelltools.com/webcat/...Refiller.html# when i used it to fill my shadow from dead dry empty took 10 minutes start to finish, full to the brim , no air burped out even 6 months later.

[wheming]

Thanks for the tips. I haven't seen that tool before.

I believe all those tests you described have been performed in some manner or other.
I'll try and get an update tonight and see if there has been any developments.

[wheming]

So it has been a while since posting to do with this troubleshooting.
Here was the lastes from the last update regarding Mini's cooling system.

I sent Brian the waterpump cartridge that i had modified ala Ed Peters recommendation for the older style pump. The reverse rotating pump has a different impellar shape, yet i still clipped every-other blade, taking the six blade impellar down to a 3 blade. (For anyone that may remember back a few years ago, i couldn't ever find definitive opinion on whether the mod was also meant for the later common block reverse rotating pumps.) During the build we removed the underdrive pulley to start off back to a baseline.

It appears the theory of these bubbles being cavitation was true. When Brian tested the system with the modified pump, the bubbles were significantly reduced, and eventually after a couple heat cycles all but disappeared. Also, the pump was pulling better suction than the 6 blade impellar.
Brian found a couple coolant hoses that were leaking during a pressure test and tightened those. On temperature cycling, the coolant was getting drawn back in now.
The (combustion gases) block test was negative, and he tested several times, different ways to be sure.

I think with these developments and some road use testing he was ready to proceed with the remaining work plan and call the coolant issue handled.

[GLHS60]

Wayne, you have provided some valuable info for sure!!
Appreciate your sharing.

Thanks
Randy

[wheming]

Wayne, you have provided some valuable info for sure!!
Appreciate your sharing.

Thanks
Randy

No problem Randy! I hope it helps someone else. Might be a good alternative than having to use the older pump.

[GLHNSLHT2]

did you take any pics of the modified pump blades???

[wheming]

did you take any pics of the modified pump blades???

Just found some, not easy to find when i have over 5500 pics on my phone right now!

I debated whether to take off the entire blade near the hub, or to just take off the "paddle". In the end i went with the latter.

[cordes]

Thanks a ton for this info. Please let us know how it idles with the blades removed.

[wheming]

Brian said it seemed to have good flow at idle. And it maintained temp at idle. The test will be idling in NC summertime weather. But, I can always bump the idle rpm slightly. And, the stock crank pulley is still on.

[johnl]

So . . . less blades, less cavitation.

[contraption22]

Fascinating. Cavitation at idle?

[Pat]

Interesting indeed, but I think its far more likely that the pump was drawing in some air (leak) on the suction side of the pump causing the air bubbles/pockets.