Folks,

I suspect I have an internal coolant leak, so I need some confirmation. Anybody can suggest how to verify that the leak is internal (or external) and if internal then where exactly it goes. I can see no visible signs of external leak. At least not when the car is stationary.

Thanks

 

1) Check the potential coolant leak areas carefully. The leak could be small and dry up before you get a chance to see it. Ie.. look for stains.
2) Take a sample of your oil and send it to www.blackstonelabs.com They will email you a kit.
3) ARe you 100% it's not just going out of your overflow tube?
4) If you are unfamilar w/ any or all of the terms above, take the car to a qualified mechanic and have him look at it. Low coolant or coolant in the oil is a good receipe for engine failure in the near future.

 

Thanks for the suggestions.

1/ I will do #1 carefully this weekend. If coolant drips on the blacktop will it disperse and not leave any (noticable) trace (unlike oil)?

2/ www.blackstonelabs.com does not exists.

3/ the level of coolant drops in the reservoir. can it still come out of the overflow pipe in the reservoir?

 

1) Most coolant leaks will leak onto the engine, then ground.
2) http://www.blackstone-labs.com/
3) Yes, coolant goes back and forth from the radiator to the overflow tank.

 

An internal leak causes coolant to enter the combustion chamber, turn into steam, and exit through the tailpipe. This "steam" is not 100% water vapor, it contains traces of anti-freeze which has a distinctive "sweet" odor. Sniff your exhaust and make a judgement.

I do not dispute any of the advice given by other posters. Here is my method, and it may not be suitable for everyone as it relies on having a source of compressed air.

1) Remove all spark plugs.

2) Perform a compression test, dry and then wet. This may tell the story right away, and you need not proceed further. Instructions for performing a compression test are in the Chilton repair manual (page 3-58) and the Haynes repair manual (page 2C-7).

3) Remove the radiator fill cap.

4) Turn the engine so that #1 piston is at Top Dead Center on the compression stroke.

5) Thread an Air Hold fitting into the spark plug hole. An Air Hold is an inexpensive metal fitting which allows you to connect an air compressor to pressurize the cylinder. K-D Tools #901 is an example. I made mine from the shell of a used spark plug, a long nipple, and a compressed air quick-disconnect fitting.

6) Pressurize cylinder #1 to 100 psi, more or less. The exact pressure isn't important.

7) A hissing sound heard at the tailpipe indicates a leaking exhaust valve. A hissing sound heard at the engine air intake indicates a leaking intake valve. Bubbles seen at the radiator fill neck indicate a blown head gasket or a crack in an engine casting.

8) Repeat for cylinders #2-#6.

9) At the end of this process you will have a good idea of what the problem is, and where it is.

 

Thank you Daniel juts let me clarify some things:

1/ as far as I understood after I apply the pressure it must not drop if everything is fine. Correct ?

2/ I am not clear how I can control position of the crankshaft so that all valves are tightly closed at any given cylinder.

3/ what do leaking exhaust and intake valves have to do with coolant level dropping?

Thank you again.

 

The compressed air hose remains connected to the air hold so the pressure does not drop. The amount of pressure is not important. The route the compressed air takes to escape is the significant factor.

If you had a gauge on the air supply hose, and you turned off the air after pressurizing to 100 psi, then you would expect to see a gradual drop-off. Piston rings, even good ones, are not a perfect seal. The speed of the drop-off could tell you something about the condition of the rings but that is a different test.

You would do this in the same way as for measuring valve clearances. This is described in the Chilton repair manual (page 1-28) and the Haynes repair manual (page 2B-5). Regardless of the method described in the manuals, you can determine TDC on cylinders 1 and 4 with perfect accuracy by looking at the timing mark on the crankshaft pulley. You can determine TDC on other cylinders with reasonable accuracy by putting a 12-inch metal rod down into the cylinder through the spark plug hole and feel it rise and fall as the engine is turned. One reason for using a rod as long as 12 inches is that it cannot possibly fall into the cylinder, as might happen with a shorter rod.

The firing order of the VQ30DE engine is 1-2-3-4-5-6 so it may be convenient to perform the air hold test in that sequence. The rear cylinder bank is 1,3,5 (passenger side to driver side). The front cylinder bank is 2,4,6 (P-to-D).

Nothing, really. I mention that in the interest of completeness so the reader will understand all the things which can be learned from that one test.

 

Some easy methods are to look at your oil dipstick and see if coolant is mixing with the oil. Also check for white smoke coming from the tail pipe when the car is hot. Also open up the radiator cap and start the car and let it run. Do this when car is cold or warm. Coolant should stay calm and not shoot out.

 

This test is useful but inconclusive. A head gasket can rupture between a coolant passage and a combustion chamber. This results in an internal leak with loss of coolant out the tailpipe. It will not cause coolant to contaminate the engine oil.

 

Easy way to tell if it's a huge leak is white smoke coming from the exhaust. Cracked head or head gaskets.

 

Ok thank you Daniel, thank you guys. I like this method with checking for bubbles in the coolant when I apply some pressure to the cylinder. But this got me thinking... why cant I use cylinder pressure naturally occurring when spark fires and mixture in the combustion chamber expands. There seems to be more than enough pressure forming there to send the bubbles through the crack (or ruptured gasket) and back into the overflow reservoir. Should I be able to see any bubbles this way? or is there anything I am missing here? I understand that the pressure only stays in cylinder for a fraction of a second but since it occures hundreds of times per second is it not as good as steady pressure applied over the same time?

Also I noticed recently that not only do I loose coolant but also my oil vanishes with no visible external signs. Can it not be related to the gasket problem in the head(s)?

 

I still believe you should get an oil analysis so that you know exactly what's in there and what isn't. Plus, Daniel's tests.....

 

I would think that if you had a leak into a combustion chamber (with no evidence of coolant in the crankcase), then you should see a lot of bubbles in the radiator.

FWIW, if you are loosing coolant but can't find any leaks into the crankcase, on the ground or under the carpet (heater core), a bad radiator cap may be the culprit. Modern engines are designed to operate at relatively high temperatures (195 deg) and pressures (15 psi). If the cap won't hold the design pressure, you could very easily be loosing coolant in the form of steam.

 

Although it isn't very accurate, it is a very easy thing to do. Always do the easy stuff before the difficult ones. You may find the answer you need in one min. instead of half an hour. That's the way lazy people like me think.