Coolant Level Dropping in Radiator

[racer12306]

Whats odd though is that I will lose the approximate 1 ounce of coolant in around 2 minutes of pressurizing the cooling system with a Stant 12270 pressure tester used on a cold engine. I would think it would take a significant head gasket leak to push that much coolant into a cylinder in that short of a time period, but maybe it's possible.

After redoing the pressure test 3-4 times refilling the system each time, I wondered if there would be enough coolant in the cylinder to hydrolock the engine. I was able to turn it over by hand without issue and it started up fine afterwards. I get no overheating when driving the vehicle, and coolant temps stay within normal range 203-210F.

Our van never overheated, ran great until it didn't. The "it didn't" phase was a CEL.

Engines can take more liquid through their combustion process than we are likely to give them credit for.

If you're sure it's leaking and you can see the leak, you need to look where you can't readily see.

 

[explorer2030]

Have you checked the radiator closely for leaks?

I thought I looked pretty closely. I sprayed the tank seams on both sides of the radiator, the area around the filler neck, the petcock, and the inlet/outlet nipples with soapy water while pressurized to 18psi. I didn't see any bubbles forming or any liquid coolant seeping out.

 

[explorer2030]

Our van never overheated, ran great until it didn't. The "it didn't" phase was a CEL.

Engines can take more liquid through their combustion process than we are likely to give them credit for.

If you're sure it's leaking and you can see the leak, you need to look where you can't readily see.

Well I'm not sure if it's leaking internally, or if it could be a deep airlock that is compressing/getting partially pushed out each time I pressurize the system. I am doing another funnel bleed, but am noticing that some air bubbles start coming up around 215F. From what I remember the last time I bled the cooling system before this issue occurred, there were no bubbles at all once it was bled. I find it a little strange because the unpressurized 50/50 coolant shouldn't start to boil until 226F.

 

[Leo99]

I has this problem a few years ago with my Corolla. Borrowed the radiator pressure tester from Auto Zone. No leaks detected. Turns out it was a defective radiator cap.

 

[explorer2030]

I has this problem a few years ago with my Corolla. Borrowed the radiator pressure tester from Auto Zone. No leaks detected. Turns out it was a defective radiator cap.

My coolant level will drop to the bottom of the radiator neck after using the pressure tester to pressurize the system to 18 psi for a couple minutes. The radiator cap isn't installed during the test, so that would rule it out as the root cause. I also replaced the original radiator cap with a new Mopar cap about a year ago.

 

[SHOZ]

Both my Hyundais did this with as the OE spring clamps aged. Weeped out of the hose ends at different points as indicated by a stain but really let air in on cool down.

 

[KrisZ]

Did you use the bleeder screw at the thermostat housing? Get the engine hot and crack it open.

Other than that I would monitor the coolant lever in the overflow tank. See how fast that drops.

 

[dumkid]

my car does the same thing. if i open the radiator cap there's air in the engine, however the level in the expansion tank doesn't drop and does fluctuate with hot/cold cycles. before i replaced the radiator cap it would push coolant into the expansion tank and then suck air through the rad cap instead of sucking coolant out of the expansion tank.

 

[ripcord]

Both my Hyundais did this with as the OE spring clamps aged. Weeped out of the hose ends at different points as indicated by a stain but really let air in on cool down.

I had a similar issue once on a small block Ford. I noticed it was weeping at the hoses at the water pump and when tightening the clamps and then replacing with new ones didn't help, I pulled the hoses to look and found crusties and a little pitting on the water pump necks. I used sand paper to clean the surfaces and scrubbed the inside of the hoses out and that fixed it.

Also, Hylomar #2 sealant is great for sealing stubborn hose connection leaks... helps with getting tight hoses on and off, too.

 

[JamesBond]

First try a new oe cap. My civic pushes coolant into the expansion tank but then doesn’t suck it back into the radiator. If driven long enough it will overflow the tank and eventually overheat. I’ve decided it’s a head gasket problem. I take coolant from the overflow and put it back into the radiator every 100 miles or so.

 

[zzyzzx]

I has this problem a few years ago with my Corolla. Borrowed the radiator pressure tester from Auto Zone. No leaks detected. Turns out it was a defective radiator cap.

You can't test the cap with the pressure testing kit?

 

[Leo99]

You can't test the cap with the pressure testing kit?

I don't remember if the kit had that adapter and it was a new cap so I wasn't thinking it was the cap. Figured that I messed something up changing the hoses.

 

[explorer2030]

Did you use the bleeder screw at the thermostat housing? Get the engine hot and crack it open.

Other than that I would monitor the coolant lever in the overflow tank. See how fast that drops.

Yes I vacuum filled the cooling system, then did additional purging to get the air that remained with a spillproof funnel. I usually open/close the bleed screw after each cycle with the funnel. Overflow tank level isn't abnormal, it rises as the engine heats up and then goes back down as it cools.

 

[explorer2030]

Both my Hyundais did this with as the OE spring clamps aged. Weeped out of the hose ends at different points as indicated by a stain but really let air in on cool down.

I didn't notice any coolant residue on any of the hose ends, and the coolant level will also drop when the engine is cold without a heat/cool cycle if the cooling system is pressurized.

 

[explorer2030]

I has this problem a few years ago with my Corolla. Borrowed the radiator pressure tester from Auto Zone. No leaks detected. Turns out it was a defective radiator cap.

I replaced the cap about a year ago with a new Mopar cap, and it just passed a pressure test.

 

[SHOZ]

A good sealed cooling system will purge itself of any air. As the coolant heats up and is put under pressure the air will dissolve into the coolant, then when pressure is relived as it goes into the overflow the air undissolves.

 

[spasm3]

I would do a UOA and see if there is coolant in the oil and go from there.

 

[Fabulous50s]

Have you pulled plugs to see if any are a different color? Then pressurize the cooling system overnight and look into each plug hole for coolant with a borescope. That is always tye way I've done it.

 

[explorer2030]

Have you pulled plugs to see if any are a different color? Then pressurize the cooling system overnight and look into each plug hole for coolant with a borescope. That is always tye way I've done it.

I haven't pulled the plugs yet because it's a bit of a job to take off the intake plenum. This would be my next step if I can't find another resolution.

 

[D60]

My '19 JL has done something similar since new (now only has about 40k). I can fill the degas bottle to the fill line, but it'll drop maybe 3/8-1/2" below the line. I can refill again and it drops again. But if I leave it where it "wants" to be, it doesn't drop further.

Never understood it.....