Engine Overheating / Losing Coolant (4th dealership visit)

Head gasket issues will push coolant into the overflow and the reservoir will be overflowing. Opposite of a coolant leak which will bring the reservoir tank almost empty.

Cooling Pressure test........usually will pickup the problem IF there is a leak, but again it's a dead easy test to do.

There is one aspect not mentioned. Most shops don't actually know how to refill the cooling system after a coolant flush and will introduce a **** ton of air into the system. This shows up as an overheat. It'll also fail the pressure test because air can leak thru the hoses but water won't, hence the pressure tester shows a pressure decay, but it's not coolant that's leaking.

My Juke had the same symptoms. Basically the engine would go into protect mode due to engine overheat after long extended drives. I checked the overflow/refill coolant reservoir and sure enough I was dry as a bone, never touched the cooling system so it should have never been low. I refill and drive and the repeated thermal syphoning should purge any air in the system. Normally on other vehicles this takes 2-3 drive cycles at most and all the air is out. The Juke is definitely different, it took almost 10 days of repeated drive/refill before the reservoir would hold the level at the exact same spot.

jet said:

Thank you for all the ideas, everyone. I've forwarded these comments on to the service advisor in case they don't find anything with the UV dye. Fingers crossed...

@macgyver - "3 times to fix it or its a lemon" - where can I find out more about this?

@pboglio - "after a coolant flush and will introduce a **** ton of air into the system" - This is what we thought happened after the first dealership visit. They just refilled and made sure there was no air in the system when I left the second time.

Click to expand...

The problem with the flush is they do tend to add air into the system. Normally, a little gas get's purged into the refill tank when the hot coolant pushes past the radiator cap into the overflow/refill tank. The cooling system is designed to purge this entraining gas on every hot shut down and let it "settle" out in the overflow tank.
That gas is a problem for heat exchange and which is why the thermostats have the jiggly doodad that bleeds trapped gas past it to avoid trapping it in the system.
It doesn't take but a minute amount of air trapped in the cooling system to cause "engine knock" in only a small section of the cylinder. This is why you can "overheat" locally, the end result being engine "knock", but the bulk coolant temps will look right on the money.

Anyway, then the engine cooling jacket cools down on engine shutdown creating a vacuum because of the high thermal contraction of the coolant, I want to say 1/2 liter or so from memory or whatever the overflow tank capacity and it's dependent on total coolant volume. Anything that can interfere with this vacuum draw is a problem. I do both a pressure test to 15 psi or so thru the radiator cap via an adapter and also a vacuum draw test. These check for obvious "leaks" but if there is air packed into the system, tough to tell unless it takes forever to "pump" the trapped air against the cooling jackets using a mityvac like trying to pump a car tire up. The difference is a pressure "decay" test which will show a drop in pressure, meaning a coolant OR trapped air leak thru a hose or whatever, like a car tire leak.

Everyone that thinks they have a head gasket issue......usually doesn't. You have a headgasket issue if you do a cylinder leakdown test with 100 psi and you can physically see air bubbles blowing up in the radiator cap neck. In some cases the head physically lifts under power because combustion gas can peak at 2000 psi or higher, and so it gets trickier to detect. In that case the radiator literally overflows and it is painfully obvious it's a head gasket leak. Either way, you don't need a combustion gas test of the coolant to know what's going on.

It's mind bending because the radiator overflow tank is full yet the engine is low on coolant, basically it all gets pushed into the overflow/refill instead of the ground like a normal "leak". It stays full because the headgasket leak both allows combustion gas to drive the coolant into the overflow and puts so much air in the cooling jackets there simply is no way to thermally siphon it back into the radiator when everything cools back down. I highly doubt this is the case with your vehicle.

Look for the obvious problem. A recent coolant flush is the first indicator. My Juke problem I never did find the root cause but it was fixed by the method I described earlier of repeated refill in the overflow tank, then check the level, then refill, and so on. Never let the refill tank go dry or you will suction air again into the sealed cooling system and then the entire process needs to start over again. I start the process by filling thru the radiator cap till it tops off while engine COLD, then I fill the overflow tank to its cold level to give room for the thermal expansion during hot operation. Some idle the car to purge using the water pump to drive the air out but it'll purge naturally if the system is "air" tight anyway. Then I keep checking the overflow tank and always top that off from then on if it keeps dipping low on hot operation. After like 2 weeks I finally stabilized the fluid levels.

When things are working perfectly you can crack the radiator cap when the engine is stone cold and hear a "sucking" sound which is the vacuum seal breaking and the coolant should form a meniscus which means zero air pockets and proper thermal syphoning. When there is a problem you see a huge air cavity in the radiator cap neck.

This is my *******/engineering hack way of diagnosing cooling system issues, NOT Nissan's recommended method but it works.

Sorry to hear about it.

I'd request a Cylinder compression test & Leak down test. It would be extremely difficult to pass these (2) tests and have a bad motor. They should be providing this to you as proof it needs a rebuild. Coolant issues are notorious for being difficult to detect. I'm not gonna try and say we can diagnose it over the internet, but I've experienced maybe 5 different ways a cooling system can "leak" and none required an engine rebuild. They do take an enormous amount of time to track down though.

Anyone with experience can tell you trapped air will cause huge problems with overheat. The fact it's "leaking" during a pressure test doesn't necessarily mean you are "losing" coolant. As I tried to explain before, the trapped air can leak out thru the hoses, etc. but the coolant may not, causing a failed pressure test. I had this same issue diagnosing my water valve design that had trapped air when I did a water pressure decay/leak test. We would pressurize the valve, then shutoff and trap that "pressure" and wait for it to dropoff. The pressure would collapse yet not a "drop" off water was found to be leaking or dripping out. Several fellow engineers and lab technicians were just mind boggled by what was going on until I realized the trapped air leakage, then it made sense. Sure enough, we had trapped a ton of air inside the water valve and couldn't purge it. It doesn't take but less than a teaspoon of volume movement to completely collapse the water pressure because it is incompressible at these pressures.

Anyway, so we designed a fixture to flip the water valve on it's back and get every last ounce of air out. Finally it would hold 65 psi water pressure for 1 full day fully submerged with water which was plenty good for what it needed to do. The seals just could not seal perfectly against air, yet once the air was completely removed they sealed up perfectly and passed the water pressure/leak test. I'll say this again, very few hoses or even O-rings have ZERO leakage against a gas, doesn't exist and its a myth. See where I'm going with this, trapped air can fail your coolant pressure test and make it look like the coolant is being eaten up by the motor. If your motor was really eating up the coolant the engine oil would look milky or you would be pushing a cloud of white smoke out the back.

If you are passing: Cylinder Compression test, Cylinder leak down tests, no white smoke, clean looking motor oil, not pushing coolant into the overflow tank, no air bubbles in the radiator refill neck, etc., then I find it difficult to believe you need an engine rebuild. All of that stuff is maybe 2 hours worth of diagnostics. At this point you MAY have warped the head, but you need data to go off of.

Given the amount of times they screwed up the diagnosis, it looks like you need a different mechanic.

Gene

RESOLVED

The last dealership I took it to did a ton of testing and the issue seems to be resolved. Here are the notes from the last visit:

"Found coolant full with dye installed. Pressure tested cooling system for external and internal leaks. Removed ignition coils and spark plugs to check for headgasket problem - headgasket is good. Removed turbocharger ducting to inspect for internal coolant leak with scope - no leak found. Test drove vehicle about 20 miles and allowed engine to idle for several hours checking for coolant leaks during testing. Contacted Nissan Tech Assistance and they instructed tech to fill coolant and mark level, return vehicle to customer and have customer return after driving 1000 miles to check coolant level. When vehicle returned after driving 1000 miles found coolant about 1/2 quart low. Rechecked for coolant leaks and found coolant leaking at housing on rear of cylinder head just above transmission. Replaced housing and gasket to correct and refilled coolant."

I drove another 3k miles and brought it back for an oil change. They retested the coolant levels and checked for leaks and found no issues. Best of all, the warranty company covered those repairs!

Thanks to everyone here for your input!