If you're still losing coolant then forget the scan tool, wherever the coolant is going is what you need to be finding out. The common leaks in my experience on these is either at the heater hose fittings on the fire wall or at the water junction on the driver's side of the head. The coolant reservior itself also leaks sometimes due to cracks along the bottom side.

Should there not be a leak at those locations then I would revisit testing for the head gasket. You said that you did an exhaust gas test, presumably you used combustion gas detection fluid and a tester? I've been bitten by those tests a couple times, head gaskets do not fail in any one specific way. My strategy for doing that test after having been bitten a couple times is to drive the car pretty hard after it's up to temp, get the engine under a lot of load however you can. Then, I immediately pull over and do my test. Use your brain, obviously you can't just rip the reservoir cap off a hot car, bleed the pressure off. It's actually ideal if the coolant is above boiling point, it'll help. Then do the test as soon as you're able to slap the tester on. I usually like to hop in the car and give it a lot of rpm as well, the more the better before I start pulling gas through the fluid. This method of testing will eliminate the circumstance where the leak only is occurring at the very top of the load range.

 

I have Mk i so you likely did not read it. The cars are well known for dropping the coolant level slowly over time by seeping very small amounts out through the endless number of hose ends. The coolant will evaporate at those small amounts and you will never find the locations unless you get under the car and look up at the bottoms of hose ends to see the stains from the leaks. The problem is worse if you changed hose using old school worm drive clamps, after time they will show to be loose as modern hose is more elastic and you absolutely must do at least one serious retightening of all the clamps a month of two later or then the seeping begins.

Had fits with both mine doing that, searching down those minute leaks was a task there until both quit doing that.

Not a bad idea to change relays too for cooling fans, they may work 99 times fine and stick on 100 to make car overheat then they go back to working 99 times again until next time.

 

Thanks both of you for the suggestions. I am in fact using the worm clamps. Leak detection is even more complicated by the cover on the bottom. I'll recheck for tightness and try to do the load test for the head gasket. If it comes to the gasket, how big a job is that?

 

Thanks both of you for the suggestions. I am in fact using the worm clamps. Leak detection is even more complicated by the cover on the bottom. I'll recheck for tightness and try to do the load test for the head gasket. If it comes to the gasket, how big a job is that?

Take the cover off for a bit while you're diagnosing and park it over cardboard or something that will stain.

Again, in my experience, I haven't seen head gasket failures super frequently so hopefully my suggestion reaffirms what you saw the first time but they're not invincible. Any single overheat event could be enough to do it.

An experienced mechanic would say that a head gasket on these is pretty easy. I did a full head overhaul in-car on my old zx3 and although the motor is a little different generationally I'd say it's maybe a 6/10 as far as difficulty. A mechanic would also just replace the motor whole though. They are way too cheap to do that big of a job on. In my area whole motors go for as cheap as $175 if you're really penny pinching.

Sent from my ASUS_AI2202 using Tapatalk

 

Mucho work with lots of room for error.

 

Apologies in advance for the long post, but I'm running out of ideas for this issue, which has plagued our son's 2014 SE for over 4 years now. Shortly after buying the car used from a dealer, my kids were driving it and got a high temperature warning. They pulled over and turned off the car, waited for it to cool down, and then drove home safely. I looked around and noticed that the radiator appeared to be leaking, good and bad news--good in that the probable cause of the overheating was obvious, bad that it appeared to be the radiator. I proceeded to replace the radiator, all the hoses and the thermostat. I also replaced the degas bottle, since it looked like it may have been cracked and for the price and given all the other components I was replacing, it seemed like a good idea.

The car seemed to run better, but at this point, I was keeping an eye on the coolant level in the degas bottle. Unfortunately, I noticed it was still dropping after a number of days. I have been using the general purpose 50/50 mixes in the car, not the Motorcraft Orange, since this is what was in the car when we bought it. I would top it up periodically and keep my eyes peeled for leaks, but never found any.

The car tended to overheat when not moving on hot days (drive-thrus were the most common); thus, I turned my attention to the fan. I replaced the control unit to see if that would help; it did not. I then took the car to my local shop for a full diagnosis, and they determined that the midi fuse (the 50A one) was blown. They replaced it, and the fan seemed to work again. We were moving from CA to CO at the time and the car was driven from NorCal to Colorado over the course of 3 days in the summer, from the Bay Area, through Utah, then across I-70. Temperatures were hot, but the car performed well, coolant levels were steady, and I thought maybe we had solved the problem.

Unfortunately, a few months later, the coolant level started dropping again as it had before. We kept an eye on it and topped up regularly. I was concerned though and took it to the local shop, where they performed pressure tests and other tests to look for leaks. They found none.

Over the past couple of months, my son reported overheating alerts on 2-3 occasions. He carries coolant with him in the car at all times, so he was well drilled in what to do. Just today, I performed my own pressure test of the system, but could not see leaks from any of the hoses, the radiator, the degas bottle or the thermostat. I also did an exhaust gas check to see if there was a head gasket issue or worse. That test passed as well. I ran the car with the heater full blast for 30 minutes to test the function of the water pump (per the shop manual, the outlet hose should be hot, which it was). We then ran the AC at max with car idling and my OBD reader connected to monitor ECT. The fan kicked on right away, but it eventually turned off. I think it turned off when the ECT was around 92C. As the car heated up, the fan came back on at 111C. We could not reproduce an overheating condition, although ambient temperature here is mild today (around 72F).

I was surprised that the fan turned off at all with max AC running, so I'm wondering if there's still an issue with the fan? I'm getting an advanced diag tool next week that will allow me to run the fan at all the different speed settings, so I will be checking that; however, I'm not sure how any of this would explain the coolant loss. Note that we are about 8000' above sea level, and while that affects things like engine compression, I'm not sure it should have any impact on the cooling system, and regardless, these issues were present in California as well, where we were near sea level.

I've read many of the other posts in this forum and while some of the symptoms are the same, the dropping coolant level in my degas bottle is the thing that's nagging at me and have not seen that covered in any of the other Mk. 3 posts. Any help/advice would be greatly appreciated!

Have you checked the water outlet itself maybe the gasket is worn out ?

 

Have you checked the water outlet itself maybe the gasket is worn out ?

Sorry, what do you mean by water outlet? I'm not familiar with that term..

 

This might help I'm no mechanic but I had a coolant leak on my 2010 recently and finally it was this little thing. Maybe it will help you I hope. Good luck

 

Thinking more about a head gasket issue, I just recently changed the oil (1 month ago) and there was zero contamination. There has never been any contamination on any of the oil changes I've done and the coolant is very clear and clean as well.

 

Oh wow, I didn't know about this part, either... thanks!

 

You should easily see a leak from that.

An experienced mech will NOT say the head gasket job is easy, it is for him (I've done countless) but he will have seen how many ways that easy work can be messed up. MANY head gasket jobs are flawed due to the worker skills there.

If you are leaking compression into the cooling system you can have a leak and still no oil or coolant transfer to wrong places.

 

Would a vacuum test show the failed head gasket more conclusively? If so, where do you attach the vacuum gauge?

 

A vacuum test has no worth there at all.

 

Rent a test kit to check for combustion debris in the coolant.

 

I did that; no color change, but I may try what BlueKoda suggests and re-run the test after the engine has been under load for a while.

 

Don't forget to check the water pump too, it's super easy to change out.

 

It passed the tests as specified in the manual (run heater full blast for 30 mins, outlet hose hot); but I do have a brand-new one in my parts box just in case.

 

Thermostat

 

I had a slow leak in my 2012 MK3, and it took a long time to find it. As BlueKoda mentioned above, mine was leaking at the heater hoses where they entered the firewall. The coolant would drip down the firewall and be absorbed by the foam rubber beneath the hoses.

 

I actually just had an overheating problem with my 2015. It only overheated when idling. It turned out to be a small fuse on the radiator fan. My mechanic initially replaced the radiator fan, and the coolant temp sensor, but the issue continued. They had to keep the car for another two days to try and figure it out. In the end they said there was a small fuse that was the culprit located on or near the fan. I hope this helps in some way. The issue only really started to be a problem in March when the ambient temp started rising.

 

Do you recall if this was the 50A midi fuse or the one in the fuse box?

 

If you're still losing coolant then forget the scan tool, wherever the coolant is going is what you need to be finding out. The common leaks in my experience on these is either at the heater hose fittings on the fire wall or at the water junction on the driver's side of the head. The coolant reservior itself also leaks sometimes due to cracks along the bottom side.

Should there not be a leak at those locations then I would revisit testing for the head gasket. You said that you did an exhaust gas test, presumably you used combustion gas detection fluid and a tester? I've been bitten by those tests a couple times, head gaskets do not fail in any one specific way. My strategy for doing that test after having been bitten a couple times is to drive the car pretty hard after it's up to temp, get the engine under a lot of load however you can. Then, I immediately pull over and do my test. Use your brain, obviously you can't just rip the reservoir cap off a hot car, bleed the pressure off. It's actually ideal if the coolant is above boiling point, it'll help. Then do the test as soon as you're able to slap the tester on. I usually like to hop in the car and give it a lot of rpm as well, the more the better before I start pulling gas through the fluid. This method of testing will eliminate the circumstance where the leak only is occurring at the very top of the load range.

Funny this came up. Tuesday while going to work my car ran up to the top of the temp gauge while in a drive thru(my other gauge said the temp was 251°f) and after checking the coolant i found it was low. Not having any on hand i just put water in it for now, 7 16.9 oz bottles later it was at the minimum on the overflow and got me home. I usually check underhood every thursday and last thursday nothing underhood showed signs of leaks.
So while i was filling it up i checked for where it could be leaking at since it was fine last week. And sure enough it was the heater hose connection in the firewall, all wet and leaking all over the wires underneath it once it got up to temp. I just got over 100k miles on it now.
I got new hoses and coolant at the dealer today and probably will fix that this weekend. I know its just the o rings that crap out but they werent that much.

 

Ok, I have an update. I bought a ThinkDiag OBD reader, which allows you to do diagnostic tests and read PCM codes. I did the fan test following the instructions in this video. The first time I tried it, I was able to get it to run at 18%, 28% and 38%. Once I tried to go to 48%, nothing. I re-ran the test and nothing at any %. I then started the car and turned on the A/C and the fan is now totally dead. I'm also getting the PCM codes in the attached photo. Does this mean the Bosch controller is bad? Now that it's not running at all, I'm wondering if the megafuse blew or the fan motor is dying/dead, since I could not get it to run above 38%. What do you all think? Fuses are cheap, new motors/controllers less so, but I want to fix it properly, of course...

 

Ok, another update: I bench-tested the fan relay and it's working properly. I replaced the fan control module with another I had on hand. I re-ran the test and it went all the way up to 78%, 3 speeds more than with the other controller. I'm not sure what the top % is supposed to be for this car, but this implied that there were 7 speeds for this motor. I stepped it back down to zero as well without issue. I am encouraged by this, but will be keeping an eye on things for a while. This would also rule out the midi fuse, so if it fails again, I have to assume it's likely the motor is dying. And while this may explain the overheating issue, I'm not sure it covers the coolant loss, though I'm sure there's some correlation there. I did also tighten all my hose clamps, so it's now time to monitor and see what happens. Since tightening the hoses, the degas bottle level looks steady, but the car has not been driven appreciably, either. Stay tuned....

 

Be aware that older reservoir caps can fray plastic threads enough that hand tighten is not enough. I have to take a big set of pliers to overtighten enough to seal so that the cap does not slow leak vapor only which will turn into lost coolant over time.

 

Some old timer advice: Overheating can cause weak hoses to leak at their connections (Plymouth Volare Slant Six). Corroded fan motor connections can cause fan failure and subsequent overheating and maybe even some fan failure codes (PT Cruiser... no post factory parts so I had to splice it through or make my own connection... I forget)). Good luck. When it comes to chasing down problems, my motto is, "Hunt it down and kill it."

 

Have you checked the water outlet itself maybe the gasket is worn out ?

That was it! I will post at the end of the thread with more details...thank you!!

 

Hey all--Coming back to this thread with what I hope is the final conclusion on this issue. As it turns out, short, in-town driving rarely resulted in any coolant loss. My son took the car on a 100 mile drive to Denver and back again and I noticed the coolant level was slightly low. I had added dye to the system a couple of weeks ago, so that whenever the coolant level dropped again, I was hoping I could pinpoint it, which is exactly what I did this morning. Turns out it was the water outlet, as The Noob had suggested in an earlier post on this thread. I have ordered the part from RockAuto and hopefully will have it installed shortly...fingers crossed that this will do it for this pesky problem!

 

If you're still losing coolant then forget the scan tool, wherever the coolant is going is what you need to be finding out. The common leaks in my experience on these is either at the heater hose fittings on the fire wall or at the water junction on the driver's side of the head.

 

newps,

At 8,000 ft, the barometric pressure is a lot lower than at sea level. Barometric pressure is just the weight of the atmosphere so higher elevations see less pressure. 14.696 psia at sea level and 10.91 psia at 8,000 ft. The absolute pressure (gauge pressure plus barometric pressure) is lower than at sea level and the boiling temperature is reduced. Water should boil at 92 C at 8,000 ft elevation. You can measure the boiling temperature on your stove.

My opinion is that the thermostat in your vehicle should open at a lower temperature than the standard t-stat. If your coolant is at normal (sea level) temperature when hot then the vapor pressure is higher (at elevation) and closer to the boiling point. Higher vapor pressure means higher loss by vapor loss and a slowly dropping level. The loss by vapor is slow because vapor has little mass. Under high load, the coolant may be boiling in the block in some spots and vapor greatly reduces heat transfer and two-phase flow is more difficult to move (coolant circulation rate decreases) so overheating occurs.

Of course, if the intake runner gaskets are deformed then coolant can sneak into the engine. Get some new gaskets and swap them out. But first I would try a cooler thermostat and an adjustment in the tune to open and close the t-stat at lower temperatures (about 15 F lower than stock) to move away from incipient boiling and to reduce the vapor pressure of the coolant.

HTH,

Chris

 

newps,

At 8,000 ft, the barometric pressure is a lot lower than at sea level. Barometric pressure is just the weight of the atmosphere so higher elevations see less pressure. 14.696 psia at sea level and 10.91 psia at 8,000 ft. The absolute pressure (gauge pressure plus barometric pressure) is lower than at sea level and the boiling temperature is reduced. Water should boil at 92 C at 8,000 ft elevation. You can measure the boiling temperature on your stove.

My opinion is that the thermostat in your vehicle should open at a lower temperature than the standard t-stat. If your coolant is at normal (sea level) temperature when hot then the vapor pressure is higher (at elevation) and closer to the boiling point. Higher vapor pressure means higher loss by vapor loss and a slowly dropping level. The loss by vapor is slow because vapor has little mass. Under high load, the coolant may be boiling in the block in some spots and vapor greatly reduces heat transfer and two-phase flow is more difficult to move (coolant circulation rate decreases) so overheating occurs.

Of course, if the intake runner gaskets are deformed then coolant can sneak into the engine. Get some new gaskets and swap them out. But first I would try a cooler thermostat and an adjustment in the tune to open and close the t-stat at lower temperatures (about 15 F lower than stock) to move away from incipient boiling and to reduce the vapor pressure of the coolant.

HTH,

Chris

Interesting. But isn't the cooling system a closed system and therefore not affected by the atmospheric pressure?

 

Changing stat by itself may be a mistake, the engines are pretty close to popping codes when the engine does not get hot enough on those. I would not unless the tune (operating software) was modified to change that threshold.