Losing a full tank of coolant every 450 miles, no symptoms...yet

[jmscdgn]

Dear Forum,

We are experiencing an unusual level of coolant loss in our 92 740T Wagon, with seemingly little side effects, and hoping the collective knowledge and experience here may be able to shed some light on the issue.

The issue:
We frequently drive Los Angeles to San Francisco on the 101 - around 450 miles. We begin with a cool engine and with coolant level topped up, to find that after 450 miles of driving and engine back to cool, the coolant has dropped by a full tank. This has happened every time for the past 4 - 5 drives up and down the coast.

The cooling system has been pressure tested by a local Volvo tech, and holds pressure.
We have hunted for a leak with UV dye/light and seen not a drop.
Spark plugs appear clean, and our MPG on the highway between cities is around 25 / 26.

The Head Gasket and water pump were replaced around 2500 miles ago, and the odometer is now approaching 140,000 miles.

Any insights or thoughts how best to proceed with troubleshooting this would be much appreciated.

Thanks for reading.

J

[Marvelous3]

Gotta come out somewhere. If it's not leaking then it's getting burned. 1992 had turbo's that are water cooled aren't they? Check for a leak off the water pipe where the cooler connects?

When you say full tank, you mean the expansion tank goes from the max level to empty or ???

[oldschoolvolvo]

Have you done a "block test" to check for combustion gasses in the coolant?
https://www.autozone.com/loan-a-tool...-gasket-tester
I know the headgasket is assumed good, but just a double check

Also, the coolant could be leaking from the turbo water jacket directly into the exhaust, so you would never see it during a pressure test or external leak check. If someone would drive behind you for a while they could watch for white puffs of smoke which would indicate burning coolant. The cause of this would be a internally cracked CHRA.

Also, you are not getting a wet carpet are you? If the heater core were leaking, I assume it would be apparent by now.

[2manyturbos]

^^ That is a cooling line that is often overlooked. The oil cooler has 2 hoses that connect to it. They never leave any trace of a leak since they are at the lowest point in the cooling system and can leak straight to the ground below the car.

[jmscdgn]

Thank you all for the quick and helpful replies.

Marvelous3 - Yes, the expansion tank. It drops from full to empty. When empty and looking through the filler hole i can see coolant in the hose that leads away from the reservoir, but the reservoir itself is dry. Hope that clarifies. The turbo is indeed water-cooled, but no sign of a leak from the outside plumbing.

oldschoolvolvo - The turbo water jacket is an interesting thought, we have never noticed - and have looked many times - any white puffs, and i replaced the CHRA around 6k ago with a new 'Mellet' brand version from the UK, it came well recommended. What would you recommend the best way to test this? Is there a chance that it may leak only when under boost? No wet carpet, thank goodness. We have ran a block test before, which was negative, but will try again to make sure.

2manyturbos - This is good point, though we changed these with new OE Volvo as a precaution about 4k miles ago, and seem to be leak free.

Thanks again, J

[oldschoolvolvo]

Originally Posted by jmscdgn

oldschoolvolvo - The turbo water jacket is an interesting thought, we have never noticed - and have looked many times - any white puffs, and i replaced the CHRA around 6k ago with a new 'Mellet' brand version from the UK, it came well recommended. What would you recommend the best way to test this? Is there a chance that it may leak only when under boost? No wet carpet, thank goodness. We have ran a block test before, which was negative, but will try again to make sure.

With a new CHRA unit, this theory seems unlikely.
Also seems unlikely that a crack would open up when the turbo gets hot when running vs. overnight when cold. But you could take the downpipe off, pressurize the system and let it sit overnight, then check inside the turbine housing for drips in the morning.

[TestPoint]

Pull the spark plugs looking for a steam cleaned cylinder. Easy enough to eliminate the head gasket.

[ZVOLV]

^^^good advice

Put a pressure tester on it and put the car on a lift and see where it's leaking out. There is likely an external leak.

[Janspeed]

a bit of a side note: you should never top up the expansion tan fully to the brim. Because as the name suggests, it is there to allow expansion of the coolant when it goes from cold to hot.

That said, the cooling system is designed as a closed pressurised system. It means the system needs to hold pressure. Even if you pressure test the system and you found it to be holding pressure just fine you could still have a problem. Because when you do the pressure test you connect the tester to the reservoir. To be able to do that you have to leave off the normal cap. And that cap needs to alsobe able to hold pressure (and act as a safety valve in case of OVERpressure). So this cap can be faulty as well. (not at all uncommon!) In which case the system can't build pressure and the expansion tank may overflow which would mean the system loses coolant.

In short: try a different known good cap,

[Mr. V]

What is a "CHRA?"

I'd never heard of that acronym.

[rb92673]

Originally Posted by Mr. V

What is a "CHRA?"

I'd never heard of that acronym.

I looked it up: Centre Housing Rotating Assembly

The middle portion of a turbo, not including the snails on both ends.

[Janspeed]

i read the OP again:
if the water pump has been changed then the cooling transfer pipe from the back of the engine to the pump had to have been taken off. That means the seal might not seal correctly anymore. When i did my water pump i couldn't make that pipe seal so is used some RTV under and around the seal, then i connected the pipe to the pump. Let he RTV dry before you fill the system back up.
Also, the water pump has a rubber seal between pump and cilinder head. This seal can fail but it could also be that the water pump has been installed incorrectly.
It needs to be push upward against the head in order to make that rubber seal do it's sealing job. (the pump casting has slotted holes for this exact purpose. The pump needs to be levered up against the head. then while holding the pump levered in place you need to tighten up the fastners. That should do the trick.

[122power]

I had a 85 245t leak on the exhaust port floor from a pinhole. That was fun to find. Radiator crimps good on end tanks? Hose nipples not cracked? Youd see rhose rhough i would think. Hose ends that go into firewall? Valve for heater ok? Outer edges of head gasket behind exhaust manifold?

[badvlvo]

No symptoms? Uh, no. You have a symptom, coolant loss. Now find the problem.

Pressure test the system, you will find your leak.

[Mr. V]

You might want to put dye in the coolant tank, run it awhile, then turn it off and using UV light look for evidence of leaks (the dye would be obvious to see).

I use the Autopro product.

[dl242gt]

Using the dye was the only way I was able to find the source of the bad head gasket on a friends 90 wagon. It was using coolant but was passing every test such as compression test, exhaust gas in the coolant test and a cooling system pressure test. Then after adding the dye the coolant loss source was revealed. It was coming out the back of the head only while driving at speed and left only a small trail as it ran down the back of the engine and came off the bell housing.

[jfs356]

Yes, that is an area that would be hard to see a coolant leak....

[TestPoint]

You have eliminated each and every one of the recommended issues with test or observations except for the head gasket. There the feedback was that it was replaced 2500 miles ago.

That alone is enough to make me question it. Why don't you pull the spark plugs as suggested and completely eliminate that.

[2manyturbos]

Originally Posted by badvlvo

No symptoms? Uh, no. You have a symptom, coolant loss. Now find the problem.

Pressure test the system, you will find your leak.

Not necessarily. I had a weird coolant loss on my LT1 Z28. I would never see any signs of a leak when parked, yet, it would lose coolant while driving it. Hooked up to a pressure tester it lost one PSI over a 14 hour period of sitting pressurized. My friend had the bright idea of letting all the pressure off the system and then starting the car with the tester hooked up. About 1 minute after the car started the pressure started to rise until it hit 25 psi when I shut it off. It wasn't even warmed up. That all happened in a minute or two. I turned out that it had a head gasket leak or crack somewhere that was essentially a one way valve. No coolant into the engine where it didn't belong. Combustion gasses were flowing into the cooling system once the engine warmed up enough for the crack/gasket to open up. The cooling system was pressurizing to the point of pushing coolant out of the cap/overflow tank while driving. I'm betting this car has a similar issue. I replaced a customer's head gasket on a 940 Turbo that was doing the same thing. It passed several pressure tests. We finally found the problem because of the white trail down the block where coolant would leak out the side of the head gasket behind the exhaust manifold. The coolant dye is a good idea to sort out if something like this is happening.

[quillc]

The original head on my car was cracked and leaking directly into the exhaust.

[badvlvo]

Originally Posted by 2manyturbos

Not necessarily. I had a weird coolant loss on my LT1 Z28. I would never see any signs of a leak when parked, yet, it would lose coolant while driving it. Hooked up to a pressure tester it lost one PSI over a 14 hour period of sitting pressurized. My friend had the bright idea of letting all the pressure off the system and then starting the car with the tester hooked up. About 1 minute after the car started the pressure started to rise until it hit 25 psi when I shut it off. It wasn't even warmed up. That all happened in a minute or two. I turned out that it had a head gasket leak or crack somewhere that was essentially a one way valve. No coolant into the engine where it didn't belong. Combustion gasses were flowing into the cooling system once the engine warmed up enough for the crack/gasket to open up. The cooling system was pressurizing to the point of pushing coolant out of the cap/overflow tank while driving. I'm betting this car has a similar issue. I replaced a customer's head gasket on a 940 Turbo that was doing the same thing. It passed several pressure tests. We finally found the problem because of the white trail down the block where coolant would leak out the side of the head gasket behind the exhaust manifold. The coolant dye is a good idea to sort out if something like this is happening.

Still coolant loss is a symptom of the problem. Regardless of where, it shouldn’t be losing or consuming coolant. If it hasnt been properly pressure tested it should be.

[2manyturbos]

Originally Posted by badvlvo

Still coolant loss is a symptom of the problem. Regardless of where, it shouldn’t be losing or consuming coolant. If it hasnt been properly pressure tested it should be.

I agree 100%. My point was simply a pressure test may not find the problem. The same thing can occur if the overflow bottle cap is the problem. You replace the cap with your pressure tester and the results say everything is fine. Meanwhile, the actual problem is sitting on the work bench while you are running the test.

[TestPoint]

Abandoned by OP.

[badvlvo]

Originally Posted by TestPoint

Abandoned by OP.

Happens often when work becomes involved and it isn’t magically diagnosed online.

[lummert]

i had a heater shutoff valve on my 88 765 that didn't leak at idle.