Finding Coolant Leaks: Tips/Tricks?

[Spoolin98]

Hey guys/girls,
I have been on the search for a coolant leak for weeks now and I cannot seem to find where it is coming from. When the car idles it does not leak, even when I rev the motor from the throttle body at operating temp, still no leak. As soon as I take the car out and do a full boost pull and bring it back into the garage, there is a pool of coolant on top of the tranny. I have replaced both gaskets on the thermostat housing. I replaced several hoses. I bought a new radiator cap. I even used a tool to pressurize the coolant system with the car off, still could find no visible leaks. I feel around on everything, nothing feels wet. I am hoping someone has had a similar problem and could point me in a direction of what could be squirting coolant onto the tranny only? Does anyone have any tips/tricks for finding coolant leaks? Any help/comments are appreciated. Sorry for the long post. Thanks in advance.
91 Talon Tsi AWD

 

[MrBoxx]

You can use a UV coolant dye to help trace the leak's origin. They sell bottles of it at most auto parts supply stores. Usually they're in the chemical aisle, near the fuel, oil, and coolant additives.

A black-light is obviously required to get the most out of them.

 

[Spoolin98]

Eric,
Thanks for your help, thats a great idea, never thought about it. I will give it try and hopefully find the culprit.

 

[andymoraitis]

Eric has a great idea even though he uses HIS blacklight to check out his glow in the dark Britney Spears posters (j/k).

What you can do in the interim is look for any types of leaks around the coolant lines to the turbo if you have a wet bearing housing. I had a pesky problem like this and since Eric wouldn't let me borrow his blacklight (loser), I eventually traced it by feeling around. It used to only happen after boosting and it turned out that one of my coolant feeds to the turbo had a pin hole leak that would only spray under boost pressure.

Other than that, let us know what the blacklight test reveals.

Andy

 

[Spoolin98]

Eric has a great idea even though he uses HIS blacklight to check out his glow in the dark Britney Spears posters (j/k).

What you can do in the interim is look for any types of leaks around the coolant lines to the turbo if you have a wet bearing housing. I had a pesky problem like this and since Eric wouldn't let me borrow his blacklight (loser), I eventually traced it by feeling around. It used to only happen after boosting and it turned out that one of my coolant feeds to the turbo had a pin hole leak that would only spray under boost pressure.

Other than that, let us know what the blacklight test reveals.

Andy

Thats a low blow.
Do you think its possible for the turbo lines to blow coolant all the way over to the top of the tranny? And I don't see coolant anywhere else besides there, no trail to follow. I know the bearing housing is a mess, its leaking oil at the feed, so looking for a wet spot is pointless because its covered in oil.
Thanks Andy.

 

[toybreaker]

How are your motor mounts?

Is it possible that engine torque/movement is pulling on a radiator or heater hose, and making it leak only under boost/strain?

It happened to me when I use a chain store "cut to fit" (that was already too short )radiator hose I knew I shouldn't have. Fine idling etc...leaked under throttle...WTF

Please give the radiator and heater hose connections at the t-stat housing and water pipe a thorough visual inspection, and let us know what you find.

 

[BeerBaron]

I've had hose clamps cut my coolant line from being cranked down too tight. They are impossible to see or feel since the are right along the edge of the hose clamp. I'd take off the hose clamps and check the hoses around the edge of where the hose clamp sat.

 

[Spoolin98]

How are your motor mounts?

Is it possible that engine torque/movement is pulling on a radiator or heater hose, and making it leak only under boost/strain?

It happened to me when I use a chain store "cut to fit" (that was already too short )radiator hose I knew I shouldn't have. Fine idling etc...leaked under throttle...WTF

Please give the radiator and heater hose connections at the t-stat housing and water pipe a thorough visual inspection, and let us know what you find.

The motor mounts are brand new prothane mounts, so I doubt that is the problem.
I am def. going to give the lines a thorough check as soon as I get the motivation, I am so fed up with this car, its one thing after the next.

 

[Spoolin98]

I've had hose clamps cut my coolant line from being cranked down too tight. They are impossible to see or feel since the are right along the edge of the hose clamp. I'd take off the hose clamps and check the hoses around the edge of where the hose clamp sat.

I will keep that in mind, thanks.

 

[Spoolin98]

Has anyone heard anything negative about using dye in the coolant system? A mechanic at my brothers work, after hearing that I was going to use it, said definitely do not use dye. He never explained why? I told my brother to ask him why, I will keep everyone posted.

The thing I don't understand is I pressurized the coolant system with the car off, the radiator hoses were rock solid. My brother borrowed the tool from his mechanic at work. We pressurized it to what the Haynes manual said to, the mechanic said not to pass that because you risk blowing out water pump seals, which makes sense. But he did say to try putting the cap for the tool on, start the car and let it get to operating temp and see how much pressure it builds, and then retry pressurizing the system with the car off again. Does the coolant system stay at a certain pressure from idle all the way through full boost? Or does the pressure build with boost?

Thanks again for all your help guys/girls.

 

[andymoraitis]

Do you think its possible for the turbo lines to blow coolant all the way over to the top of the tranny? And I don't see coolant anywhere else besides there, no trail to follow.

It's possible, but mine never spat coolant onto the tranny. I'm just mentioning it so that you can cover all of the bases. Also, I think your fellow thread contributors are on to something with the coolant hoses being cranked too tight. The water neck isn't that far away from the transmission so that's a possibility as well.

As far as the dye, if Eric used it with no ill effects, then I'd trust his word. He is after all a wiseass, err I mean wiseman. Go ahead, it won't hurt. It might not be a bad idea to flush the cooling system when you're done though.

Keep us posted.

 

[Spoolin98]

It's possible, but mine never spat coolant onto the tranny. I'm just mentioning it so that you can cover all of the bases. Also, I think your fellow thread contributors are on to something with the coolant hoses being cranked too tight. The water neck isn't that far away from the transmission so that's a possibility as well.

As far as the dye, if Eric used it with no ill effects, then I'd trust his word. He is after all a wiseass, err I mean wiseman. Go ahead, it won't hurt. It might not be a bad idea to flush the cooling system when you're done though.

Keep us posted.

The only thing that has me confused is, wouldn't pressurizing the system find those leaks on hoses tightened down too much? And about the dye, my brother asked his mechanic why its bad, he said because it makes a mess throughout the cooling system and you have to flush it several times to remove all traces, no biggie. Thanks again.

 

[MrBoxx]

As far as the dye, if Eric used it with no ill effects, then I'd trust his word.

Well, actually I've never used it personally. I just see the bottles of it every day at work. The specific one I'm thinking of is a dye/sealant. I'm not a fan of radiator sealants (especially Bars Leaks) because of the risk of clogging the internal tubes of the radiator. But I know there are UV dyes available that don't also attempt to seal the system.

The mechanic is probably right, however. After you find the leak and fix it, it's probably a good idea to flush the system out.

Eric has a great idea even though he uses HIS blacklight to check out his glow in the dark Britney Spears posters

He is after all a wiseass, err I mean wiseman.

I'm speechless. I have no witty comeback. I'll think of one and get back to you.

 

[toybreaker]

My brother borrowed the tool from his mechanic at work. We pressurized it to what the Haynes manual said to, the mechanic said not to pass that because you risk blowing out water pump seals, which makes sense. But he did say to try putting the cap for the tool on, start the car and let it get to operating temp and see how much pressure it builds, and then retry pressurizing the system with the car off again. Does the coolant system stay at a certain pressure from idle all the way through full boost? Or does the pressure build with boost?

Thanks again for all your help guys/girls.

I would be very carefull about running the engine with a pump up style cooling system pressure checker. If it's anything like my snap-on one, there's no set pressure relief. Meaning, if the cooling system is full cold, and you start and warm the motor, the coolant expands, and the tester will not relieve the pressure, at least not until it gets to it's highest rated pressure...

You will find a leak, but that's only because the system will build 'leventy kerbillion psi.

I've seen it done...cost the shop a heater core...

I'm just saying please don't pump a full, cold cooling system up to 13 psi and then run the car 'till operating temperature It could build dangerous pressure, and rupture a hose, puking hot coolant all over you.

(And, yes I know people do this all the time,.. they seem to find a lot more leaks/bad heater cores... )

If you look at the tester's gauge face, you'll see it can obtain and hold as much as 25 psi. If it can take 25 psi, and you start cold with full coolant, can you see the problem that will occur as the coolant expands? This is how my tester works, it doesn't relieve any pressure until I let it, your's may vary.

These testers are meant to be pumped up and left on the not running engine, to check for leaks over time. You merely pump the system up to your desired pressure, and let it set awhile, and generally most leaks will show up.

You can use them on an already warm motor, just run it until it's at operating temp, Shut the motor off, wait a few moments till it's not too hot/pressurised and release the pressure, install the tester, and pressurise the system. You can even run the car with the tester attatched at this point, as the coolant is mostly done expanding. If you want to be absolutely sure you won't have a problem, you can pull the expansion chamber hose out of the chamber, so the system can't pull coolant out of the reservoir as it cools some from operating temp.

Testing the system at operating temp will find most cooling system leaks...

Unless of course it's a dsm

Then you have to find a 4% upgrade with a 15 mile an hour head wind on a 69* day,...and then it'll do it every time

As far as the cooling system pressure changing relative to boost, the answer is...

kind of...boost will put a lot of heat in a motor very quickly.

Boost makes heat, because you're making more power/doing more work, and heat potentially makes more pressure in the cooling system. This is because heat expands the coolant, and until the relief pressure of the cap is reached, more heat = more pressure.

As you're approaching operating temps,you could probably put the same amount of heat in a motor with 15 seconds of boost as you could with a couple more minutes of normal driving, so I can see how it would seem like boost is creating this problem.

If you just drive normal like, things don't get that hot.

If you drive it like it's a dsm, D ) it'll get hotter, quicker, and run the cooling system under a much higher load then you could ever get at idle.

Edit; had another thought...

It could also be that the leak could be from the overflow hose connection at the t-stat housing. When you pressurise the system, this hose isn't in the loop. The only time the overflow hose sees coolant is when the pressure relief in the cap opens, and bypasses some coolant to the expansion/overflow tank. ( Like a full boost pull) That little twist tie a 'mo bob doesn't really seal the hose all that well, a small clamp will do it much better. This hose doesn't really see pressure, just coolant to and from the expansion chamber, but it's right in the area that could put coolant onto the tranny...might be worth a look ....

 

[Spoolin98]

I would be very carefull about running the engine with a pump up style cooling system pressure checker. If it's anything like my snap-on one, there's no set pressure relief. Meaning, if the cooling system is full cold, and you start and warm the motor, the coolant expands, and the tester will not relieve the pressure, at least not until it gets to it's highest rated pressure...

You will find a leak, but that's only because the system will build 'leventy kerbillion psi.

I've seen it done...cost the shop a heater core...

I'm just saying please don't pump a full, cold cooling system up to 13 psi and then run the car 'till operating temperature It could build dangerous pressure, and rupture a hose, puking hot coolant all over you.

(And, yes I know people do this all the time,.. they seem to find a lot more leaks/bad heater cores... )

If you look at the tester's gauge face, you'll see it can obtain and hold as much as 25 psi. If it can take 25 psi, and you start cold with full coolant, can you see the problem that will occur as the coolant expands? This is how my tester works, it doesn't relieve any pressure until I let it, your's may vary.

These testers are meant to be pumped up and left on the not running engine, to check for leaks over time. You merely pump the system up to your desired pressure, and let it set awhile, and generally most leaks will show up.

You can use them on an already warm motor, just run it until it's at operating temp, Shut the motor off, wait a few moments till it's not too hot/pressurised and release the pressure, install the tester, and pressurise the system. You can even run the car with the tester attatched at this point, as the coolant is mostly done expanding. If you want to be absolutely sure you won't have a problem, you can pull the expansion chamber hose out of the chamber, so the system can't pull coolant out of the reservoir as it cools some from operating temp.

Testing the system at operating temp will find most cooling system leaks...

Unless of course it's a dsm

Then you have to find a 4% upgrade with a 15 mile an hour head wind on a 69* day,...and then it'll do it every time

As far as the cooling system pressure changing relative to boost, the answer is...

kind of...boost will put a lot of heat in a motor very quickly.

Boost makes heat, because you're making more power/doing more work, and heat potentially makes more pressure in the cooling system. This is because heat expands the coolant, and until the relief pressure of the cap is reached, more heat = more pressure.

As you're approaching operating temps,you could probably put the same amount of heat in a motor with 15 seconds of boost as you could with a couple more minutes of normal driving, so I can see how it would seem like boost is creating this problem.

If you just drive normal like, things don't get that hot.

If you drive it like it's a dsm, D ) it'll get hotter, quicker, and run the cooling system under a much higher load then you could ever get at idle.

Edit; had another thought...

It could also be that the leak could be from the overflow hose connection at the t-stat housing. When you pressurise the system, this hose isn't in the loop. The only time the overflow hose sees coolant is when the pressure relief in the cap opens, and bypasses some coolant to the expansion/overflow tank. ( Like a full boost pull) That little twist tie a 'mo bob doesn't really seal the hose all that well, a small clamp will do it much better. This hose doesn't really see pressure, just coolant to and from the expansion chamber, but it's right in the area that could put coolant onto the tranny...might be worth a look ....

I will keep all this in mind. I will not have time to do any work on this car till I get back from vacation, bu I will definitely keep everyone posted. I have one question for you, I was not planning on pumping the system to 13psi and then running the car. I was planning on leaving the system unpressurized, with the tool cap on, and seeing how much pressure the gauge builds until operating temperature. Do you see anything negative about that? And your definitely right about driving the car hard compared to normal. When i just drive the car easy, it doesn't leak, but under boost it leaks. Whats the fun in having a turbo car that you cant build boost in without it leaking? There is none, thats why I must find this leak. I'm going to try the dye if I cant find it by pressurizing it this time around. Thanks again everyone.

 

[andymoraitis]

I'm speechless. I have no witty comeback. I'll think of one and get back to you.

And I'm sure it'll be very painful for me since paybacks are well, you know!

 

[toybreaker]

I'm not exactly sure what would happen if you started with an unpressurisedcold system. If you monitor the pressure, and don't let it get over 14psi~ish, you'll be fine. The o.e. caps are rated at .9 bar, and a little more won't hurt anything, but you may have to let some out as the coolant expands.

Edit; I got to thinking about it, and I think you could pull the coolant transfer hose to the expansion tank up out of the coolant in the tank, when the motors at operating temp. As the motor cools down, it won't draw any coolant back out of the reservoir. That will leave room for the coolant to expand from cold, with the tester attatched. This may leave you with a low cooling system though...I don't know...

I think andy may be on to something with the pinhole leak on one of the little coolant transfer hoses to/from the turbo, and/or the throttle body. I could see how they would only spray under certain conditions...Maybe moving them around a little while the system is warm, and under pressure may replicate the conditions required for them to leak.