Our '06 Caravan with 108k miles on suddenly lost @ 1 quart of coolant out of the overflow bottle (radiator also low) in a 1-2 week period about 2 weeks ago. Since then, I've been trying to figure out what is wrong, keeping an eye on things. I've run a battery of tests. The level seems to drop slightly since the "big loss."


I wanted to post my results, which are somewhat contradictory to me, maybe someone can help me make sense of them.


1. Electrolysis -- I performed the recommended tests for electrolysis and found all voltage readings within "safe" ranges.

2. Visual inspection -- I have checked all hoses, clamps, the water pump weep hole and pulley area, and the visible parts of the radiator, with no luck. Overflow bottle is not leaking either. I have been keeping an eye on the tailpipe but it is clean, no soot or other problems, and with the changing temps in our area right now, it's difficult to say if the little bit of water and steam sometimes seen in the morning is normal, or related to the head gasket/coolant leak. I do not see the steam and leakage present once the system is warmed up.



3. Heater core -- based on no evidence of leakage of heater core (fog on windows, antifreeze smell in cab, loss of coolant into vehicle or on ground from condensate drain tubes) I am ruling out the heater core.


4. Combustion analysis test -- I was unable to get any results from a chemical combustion analysis test on the coolant -- although several attempts with a Lisle combustion analyzer ended with coolant getting sucked into the tester. I was able to do a "control" test using the tailpipe, so I know what I should have been seeing, and I didn't see it.


5. Coolant Pressure Test -- Using a Stant coolant pressure testing tool, I observed the following results:


A. In two "engine warm, not running" tests, the system held pressure acceptably for the recommended 2 minutes.
B. In the "engine cold, system full" test, where the engine sat overnight to stabilize coolant levels and I made sure everything was topped off, etc., the system held pressure.
C. In the "engine cold, test to idle," the pressure started rising somewhat rapidly after a minute or so to the 17-19 PSI range before I felt uncomfortable and shut the test down. The recommendation for this is REPLACE HEAD GASKET.
D. A repeat of this test done with the engine warm did not produce the same result, but the system may not have been properly full at this point due to my other tests.


6. Radiator Cap pressure test -- the @ 1.5 year old radiator cap leaked VERY slowly before stabilizing on 2-3 test attempts, but it always stayed within the @ 15-16 psi range that the system requires. I replaced it anyway with a tested good cap from the auto parts store.


7. Spark plug visual exam and compression/cylinder leakdown test -- I have not had a chance to perform these yet; this is our family car and gets driven every day, so I'm having to sneak my tests in around work, school, karate, etc. I have a compression tester and can do this if I have to.



8. Transmission cooler leakage -- no coolant present in transmission fluid, no unusual frothing or foaming.


Some service history: Water pump, hoses, thermostat, etc. all replaced @ 28k miles ago. Transmission serviced every 30k, last time was @ 18k miles ago at 90k. Oil changed every 3k with Mobil 1 Synthetic. I do all maintenance that I can here at home so I am familiar with the system, etc. enough to notice subtle changes. Nothing else about the vehicle seems abnormal, temp gauge reads fine, etc.


Last tune up was @ 76k miles, roughly 2 years ago. I was a Dodge Service Advisor for awhile and so I hit ALL the recommended milestones with my personal vehicle, force of habit I guess.



* * *



I guess what I'm asking is, if the head gasket leak is SEVERE enough to cause a rapid pressure increase at idle; yet NO COOLANT is present in the oil, and no evidence of coolant leaving through the exhaust system is present, WHERE is it going and am I correct in pulling the trigger on a head gasket job? Any other tests I should perform?


Am I looking at some kind of funky block cracks that are swelling shut when it heats up?


I've been a mechanic and do 90% of my own work, but this isn't a job I'm willing to tackle myself, with no covered garage area at our current home and no time to break it down or keep it down if bigger problems present.



I'm just trying to diagnose it to decide whether repair or replacement of the vehicle are prescribed here.


Next week I will call my trusted mechanic and get a second opinion as well.

 

I'm not familiar with a cold to idle pressure test though I've done pressure tests before. Reading up on it it does appear it indicates head gasket problems meaning an exhaust leak into the cooling system. If you could get the CA test to work it would confirm it. As I recall they make test strips you just dip in the coolant to detect it also. Might try that. Have you eliminated a possibility of a t-stat or dead fans? You've touched many of the bases mentioned in this article but maybe something will spark an idea. Just have to wade through the basic stuff: http://www.motor.com/article.asp?article_ID=1345

 

I will try to get some test strips, that might work.

I'm guessing that since I don't see bubbling in the coolant, it might be a small enough leak into exhaust that it is allowing coolant to be consumed but not excessively to the point that there's bubbling or coolant out the tailpipe.

The strips might be perfect because they would allow me to test for a minute amount.

The vehicle is not overheating, that's what's weird -- it's perfect, temp-wise.

T-stat was replaced at same time as water pump, etc., and the gasket shows no leaks or signs of evaporation of small leaks. Cooling fans working just great.

It's weird that everything is fine except the missing coolant. I'm pretty sure none of the local drunks are siphoning it out at night for a cheap thrill, LOL.

I'll check the article, thanks!