98 Caravan 3.8 AWD bubbling from overflow

[dx92beater]

Hi,

I think that I might have posted something similar to this before. Before you guys start pointing fingers at the Head Gasket, please READ THIS POST CAREFULLY.

I am getting bubbling from the coolant overflow.

The bubbling only happens when the radiator fans are not on and stops instantly when the fans come on.

Here is some history. I've had this problem since I bought the vehicle used.

I observed the problem for a few months and could never find a leak until I looked underneath the vehicle and could see that someone JB welded the rear head around one of the coolant ports and the jb weld was flaking off. I replaced the head and head gasket with a remanufactured one. I got it all back together and I am still experiencing the same problem.

So parts replaced are:

radiator cap (twice)
thermostat
head/head gasket
intake manifold gasket

To me it seems that the fan is not turning on soon enough....this is all that I can think of.

Thanks

 

[Gabe]

The first two things that come to my mind are an incorrect antifreeze mix or a problem with the coolant temperature sensor.

The temperature sensor is listed as a top seller by RockAuto.

 

[dx92beater]

Thanks for your thoughts...antifreeze is 50/50. Forgot to mention that I replaced the coolant temperature sensor already.

Sorry

 

[Texan4Life]

combustion gases
coolant boiling (not enough pressure and/or not enough antifreeze)
air trapped from coolant change

maybe a suction leak?

I'm stumped

 

[dx92beater]

I know you are stumped...just reviewing what you are saying.

If it is combustion gasses...then I figure it would be all the time and if that is not right, the fan coming on would not stop the gasses.

Suction leak?

Air trapped from coolant change...possible, but the air is never ending in supply

Originally Posted By: Texan4Life
combustion gases
coolant boiling (not enough pressure and/or not enough antifreeze)
air trapped from coolant change

maybe a suction leak?

I'm stumped

 

[dx92beater]

Well I took the thermostat out to eliminate temperature as being a variable in seeing if this a cracked head/head gasket issue. I wanted to see if the champaigne bubbles would start out at a lower temperature. Indeed they have!

Man, I am ready to throw in the towel.

 

[mechtech2]

You are supposed to get flow TO the overflow when things get hot.
Normal expansion is the reason.
Are you seeing air bubbles for sure? Boiling fluid? Constant?
A head may be cracked in all sorts of places, warped, and the block may be warped. Did you reuse TTL bolts?

 

[dx92beater]

Yes, it isn't boiling fluid. It is now constant. Removed the thermostat to be sure it was at all times and it sure is.

I have been trying to make a homemade pressure tester but decided to run regulated compressed air to the radiator at 16 psi (what the cap is rated at).

Noticed a leak at the heater bypass hose.

Wonder if this could be causing this. Gotta wonder why it won't leak under normal circumstances.

Thoughts...

Thanks so far!

 

[dx92beater]

I meant the hose from the oil housing to the water pump....anyways tightened it up....leak persists....too much time and money into van...bye bye

 

[yonyon]

You need to know if the bubbling gas is exhaust or boiling coolant. Auto parts stores sell test kits for this that use a color change to indicate presence of combustion gasses. I think they're called chemical head gasket test kits. I've never used one, but I've worked with guys who say they give a clear and definitive answer every time. If your local auto parts stores don't sell such kits you could also bring your car to a repair shop that has an exhaust gas analyser and have them check it for you. Typically this would be pretty much any place that does emissions work or tuning work.

It's possible your bubbles are just boiling coolant. Built up rust/scale or sealer or some sort of blockage in coolant flow such as from a foreign object or a damaged or improperly installed gasket can cause hot spots. If it's just buildup making a hotspot you can probably fix the whole problem with a chemical flush.

Or you could maybe just assume you have a cracked head and dump it, yeah. It feels good when something that's so very frustrating is gone.

 

[johnsmellsalot]

Maybe you didn't clean the mating surfaces between the block and head good enough and they aren't sealing properly.

 

[doitmyself]

My daughter had bubbling into the overflow with her Sebring (RIP-the car). Turned out to be a bad water pump (disentigrated vanes/no leakage out the weep hole).

I can't figure out the stuff going on with the radiator fans/bubbling stops.

Good luck.

 

[dx92beater]

LOL @ yonyon....went to harbor freight yesterday as a last ditch effort and got a leakdown tester. The tool is poor quality though...can only test at 15 psi and is broken already from being pegged.

If I do a leakdown test and don't get bubbles from the radiator is it definitively not the cylinders/head gasket?

The stuff with the fan/bubbling stops was my imagination...check a few posts above. I pulled the thermostat out and it bubbles all the time. What was happening was that it wasn't bubbling until the thermostat opened (and that happened pretty close to the time the fan turned so I thought I had that part figured out).

Interesting on the water pump! Another thought was that the felpro intake manifold required RTV whereas the factory one did not. Felpro's directions were very much off during the headgasket process, so I decided to ignore their RTV instructions too.

 

[dx92beater]

Alright people, still trying to work this. Some good news and some bad news along with some questions to see if I am reading into this right:

1) I put compressed air directly into the each cylinder and put each cylinder at TDC.

2) One of the cylinders shot tons of air out of the intake manifold!

3) No bubbles came from the radiator during pressure testing at all.

So....

Questions:

1) Does the above test rule out the possibility of a bad head/head gasket forcing air into the cooling system with absolute certainty?

2) Does the above air leaking through intake manifold mean that the cause of the issue is a leaking intake valve?? I can't see it being anything else.

As for some history, this is remanufactured head....so maybe the intake valve isn't seated right?

Thanks

 

[dx92beater]

Intake leak not really an intake leak...was not at TDC apparently

The only question that remains is question number 1.

 

[antiqueshell]

You'll get the answer for Question number 1 if you get a coolant test strip kit from the auto parts store that tests for combustion gas contamination.

 

[dx92beater]

I have read that they are not all that reliable and that you have to do crazy things like pop the radiator cap when it is running.

So in other words, a leak down test is not 100% reliable?

Originally Posted By: antiqueshell
You'll get the answer for Question number 1 if you get a coolant test strip kit from the auto parts store that tests for combustion gas contamination.

 

[antiqueshell]

dxbeater:

Nobody wants to find out the bad news that their head is cracked or the HG is done, but it seems like you know the answer before you do the conclusive tests. As for the cooling system combustion gases test you can usually just have the rad cap off first then warm the car up to operating temps. They may not be 100% accurate (nothing ever is) but they are pretty accurate.

 

[dx92beater]

antiqueshell....not talking about the accuracy of the strips...want to know about the accuracy of the leakdown test...willing to try it after I get the rest of the engine back together but still want to know why a leakdown test is not conclusive or if it is that way I don't need to waste time on the strips....astounding that no one seems to know...I mean before the strips became mainstream, did people consider the leakdown test to be conclusive?

Originally Posted By: antiqueshell
dxbeater:

Nobody wants to find out the bad news that their head is cracked or the HG is done, but it seems like you know the answer before you do the conclusive tests. As for the cooling system combustion gases test you can usually just have the rad cap off first then warm the car up to operating temps. They may not be 100% accurate (nothing ever is) but they are pretty accurate.

 

[yonyon]

No, the leakdown test as performaed isn't conclusive if only for the reason that one cylinder wasn't tested at all. That cylinder is the one that had an open intake valve and allowed air to blow out through the in door.

Now, how about the pressures involved? Are they appropriate? I don't know offhand the compression ratio of your engine, but I think gasoline engines tend to use compression ratios in the neighborhood of 8.0:1 to 11:1. That means that cylinder pressure during cranking can be as high as 11 bar which is roughly 160 psi. If 160 psi is possible before the engine even starts, then once it fires up and has combustion going on the pressures could be just a tit-bit higher. How much air pressure did you feed in during your leakdown test? Did you perform the test with the engine at full operating temperature?