Need new ideas to find a coolant leak!!

[Jerry]

My Tr3A has had a coolent leak through two head gaskets. This last time I was very careful and still had a minor amount of coolent blowing white smoke out the tailpipe. I have pulled the engine apart and cannot find any water in the oil. Cannot find any viable cracks, the head was tested last time I put it on by the machine shop. I had a small bubble coming up the left front head bolt and pulled the bolt, put sealant on it and replaced. That leak quit. I am looking for other suggestions. I plan to take the block and head down to the machinst shop this week for some testing. Any other ideas?

Jerry

 

[Guest]

Fluorescent dye?

 

[kennypinkerton]

...and a blacklight.

 

[kennypinkerton]

are the surfaces perfectly flat and clean? block deck and head?
has the head been retorqued? (back the bolts off a quarter turn and retorque to spec - in proper order)

 

[Geo Hahn]

As you know, white smoke either means you've elected a new Pope or coolant leaking into a cylinder. Assuming the latter, UV dye probably isn't going to tell you much.

Do your liners stand proud of the deck of the block? The spec is in the book. Check all the way 'round on all 4 as the head surface might not be perfectly level.

If they do not project up to clamp the ring on the gasket you will geta leak there... I believe the height of the liners can be adjusted by selection of the 'figure-of-eight' gaskets which are available in various material.

 

[TRED]

I always take the time to retorqued heads on my motors. I have been very surprised by the difference in how the bolts felt on a retorque. I warm the motor up to operating temp and then let it cool down till its just warm and then retorque.

 

[piman]

Hello Jerry,

is the car using water?, because if not then white smoke can also be due to hydraulic fluid being drawn into the engine through a defective brake booster, if you have one?

If it is using water, a pressure test for porosity is another check that you can have done.

Alec

 

[Geo Hahn]

A DIY test is also possible if the head is still on. Take a large balloon-like device (available at your local drug store somewhere around the pharmacy -- or a rubber glove might work) and stretch it over the neck of the radiator (cap off, overflow blocked). Remove plugs and put each cylinder at TDC, car in gear, hand brake firmly on -- and use your air compressor with a fitting from your compression tester to blow air into the cylinder (the one at TDC). Do this for each cylinder -- the one(s) with the leak should inflate the balloon. Quite fun and even festive.

 

[sp53]

If the head gasket you used was brand new and the head was checked for leaks, I would start to think that there is a hairline crack in one of the sleeves. Does the car actually overheat when moving down the highway? Because, personally, I have never seen an interior crack not overheat after about a couple hours of operation. I would put some brand new plugs in it then I would take it out and drive it around awhile. After I got home, or broke down on the side of the road, I would pull the plugs. The plug that has a possible water problem should be a little different, usually cleaner with a kinda sticky look.
sp53

 

[Jerry]

Head was resurfaced prior to last installation. There is a loss of coolant and you can see it slowely drop after a few hours of running. No overheating and no water in the oil which would possibly eliminate the figure of eight leaks. I did try a pressure test with a air compressor into the spark plug hole. Wherever this leak is , it is small. I could not tell where the air was going. My first time using that test though. I guess, back to the machinist shop and have him test for cracks. One item that was visable was the leak at the front head bolt. I reset that bolt with sealer and retorqued after two hours of running. Still white smoke. There is no booster so it is not brake fluid.

What fun!

Jerry

 

[kennypinkerton]

could re-torquing have broken the sealant you just put in there?

 

[piman]

Hello Jerry,

it looks as though it is back to a proper pressure test. There could be some porosity, rather than a crack. I would not expect a slight weep from a head bolt (Not stud?) to get into the cylinder due to the bore seal of the head gasket.

Alec

 

[Alan_Myers]

Hi Jerry,

Are the cylinder sleeves protruding about .003" to .005" above the top surface (deck) of the block? As Geo mentioned, if not this could be the source of a leak like you describe.

If the sleeves don't protrude this amount, one solution is to install slightly thicker figure 8 gaskets under the sleeves. A more involved solution is to remove the sleeves (and everything else) and mill the top of the block.

 

[Geo Hahn]

Jerry -- I don't think you said whether you are using the stock head gasket or one of those solid copper jobs. If the latter, you might consider the much more forgiving former.

 

[dklawson]

I was going to suggest the dye also....

Since you believe the leak may be from the water jacket into the bores, consider investing in a leak-down tester (or building your own).