If I drive 20 minutes, I have to add coolant after it cools down. The coolant level is always low after a drive longer than 10 minutes.

I just bought an air compressor and I have a leakdown tester on order, so I should be able next weekend to get some air in that cylinder and see.
The symptoms don't match the coolant loss.
If you are loose coolant from a location other than the coolant tank while the engine is running, I seriously doubt it's going into a cylinder. The pressure in the cylinder is many 1,000s of Psi. This always allows combustion products into the coolant. Not coolant into the cylinder, as this only happens after shutdown.
More tests are needed I think.
 
The symptoms don't match the coolant loss.
If you are loose coolant from a location other than the coolant tank while the engine is running, I seriously doubt it's going into a cylinder. The pressure in the cylinder is many 1,000s of Psi. This always allows combustion products into the coolant. Not coolant into the cylinder, as this only happens after shutdown.
More tests are needed I think.
Would disagree mate, he said that the cap was ok to 16 and then it increases pressure above this when revving (he stopped at 19 at a couple of thousand rpm - and that was without the engine under load.. There is no evidence at all for coolant into the cylinder (quite the opposite). A small leak in the cylinder to coolant most often is only one way at the beginning due to the huge pressure differential. Also the bubbles in the previous test would confirm it - AND the low 3 pot reading. AND the over-pressurising when revving.
IF coolant was getting into the oil - and there is absolutely no evidence at all of that - the oil level would rise slightly and a rapid emulsion would form.
Hence all absolutely totally classic of one way (minor path) symptoms of HGF with cylinder to water gallery - one way - not at all in any respect cylinder to oil passages or coolant to cylinder.
 
Last edited:
Well today I did a few tests with the expansion tank cap removed.

I drove two trips of 20 minutes, let the car cool a few minutes and then removed the expansion tank cap.


If I let the car idle after that trip and without the cap, the coolant level keeps rising and rising until it over flows.

Furthermore, I was about to try another combustion gas test but as you can see in this video




There are bubbles (seen in the tester even though I am not sucking in air) and pressure in the expansion tank.

I am wondering if I need to buy a cylinder leak down tester at this point as I am thinking it is pretty confirmed.

Opinions ?

Thanks again

Martin

Hi Martin, yes, that would confirm it for me. I wouldn't bother with a leak down tester (although a handy tool to have and a compressor is a tool I wouldnt live without) - you may find not enough pressure can be applied with a leak down to simulate., however - regarding the main point - you have MORE than enough symptoms to diagnose.
If you want to confirm coolant loss from the expansion tank due to pressure rise then put a container on the outlet to catch the spill off.
However, just do your bubble test again and give it some revs
 
Last edited:
Would disagree mate, he said that the cap was ok to 16 and then it increases pressure above this when revving (he stopped at 19 at a couple of thousand rpm - and that was without the engine under load.. There is no evidence at all for coolant into the cylinder (quite the opposite). A small leak in the cylinder to coolant most often is only one way at the beginning due to the huge pressure differential. Also the bubbles in the previous test would confirm it - AND the low 3 pot reading. AND the over-pressurising when revving.
IF coolant was getting into the oil - and there is absolutely no evidence at all of that - the oil level would rise slightly and a rapid emulsion would form.
Hence all absolutely totally classic of one way (minor path) symptoms of HGF with cylinder to water gallery - one way - not at all in any respect cylinder to oil passages or coolant to cylinder.

I understand what you are saying Joe and I do agree, kind of.
However the OP is able to drive the vehicle without it spewing coolant out of the cap. If there is a sufficient sized leak into the coolant side of the HG for a compression test to show a 40Psi drop. Trust me, with the engine running, the expansion tank will literally explode with the pressure of combustion. You definitely couldn't drive the vehicle 20 clicks and only loose a few ML of coolant.
 
I understand what you are saying Joe and I do agree, kind of.
However the OP is able to drive the vehicle without it spewing coolant out of the cap. If there is a sufficient sized leak into the coolant side of the HG for a compression test to show a 40Psi drop. Trust me, with the engine running, the expansion tank will literally explode with the pressure of combustion. You definitely couldn't drive the vehicle 20 clicks and only loose a few ML of coolant.
Hi Nodge mate,
I also understand where you are coning from but have a slightly different take on the results. as follows.
He is losing a lot more than a few Ml though - about 1/4 of a litre in 50Km - (that is a lot !) - that is about 2 PINTS of fluid in 100 miles.
He is also getting the bubbling back through the coolant. A simple test would be to put a container to catch the over-spill from the coolant expansion tank.Also, the coolant cap is set at a blow off of around 16PSI, the escaping gasses are quite capable of passing through the fluid and out of the cap without blowing the tank. It doesn't take a huge leak to actually drop the compression test down. In a small leak, the dynamic compression in the cylinder is often higher (than the relative difference would show) with injected fuel and this can partially seal the leak - a little bit like a tiny bit of oil assisting the seal of a worn bore. There is also no evidence at all of emulsification.
Again the simple test for the op is to watch the bubbles at idle - cap off engine cold - and then rev the engine and observe (I still do not believe he has done this - only at idle). Also there is a distinct odour to combustion gasses. Any sustained bubbling is indicative of an issue, and certainly and increase in bubbling with increase in RPM is definitely indicative of an HG call to me. ?
Interested in your thoughts.
Joe
 
Hi Nodge mate,
I also understand where you are coning from but have a slightly different take on the results. as follows.
He is losing a lot more than a few Ml though - about 1/4 of a litre in 50Km - (that is a lot !) - that is about 2 PINTS of fluid in 100 miles.
He is also getting the bubbling back through the coolant. A simple test would be to put a container to catch the over-spill from the coolant expansion tank.Also, the coolant cap is set at a blow off of around 16PSI, the escaping gasses are quite capable of passing through the fluid and out of the cap without blowing the tank. It doesn't take a huge leak to actually drop the compression test down. In a small leak, the dynamic compression in the cylinder is often higher (than the relative difference would show) with injected fuel and this can partially seal the leak - a little bit like a tiny bit of oil assisting the seal of a worn bore. There is also no evidence at all of emulsification.
Again the simple test for the op is to watch the bubbles at idle - cap off engine cold - and then rev the engine and observe (I still do not believe he has done this - only at idle). Also there is a distinct odour to combustion gasses. Any sustained bubbling is indicative of an issue, and certainly and increase in bubbling with increase in RPM is definitely indicative of an HG call to me. ?
Interested in your thoughts.
Joe

I kind of agree and what you say is valid for most engines. However I have been caught out by similar bubbling in the coolant before on the KV6. I was asked to give a second opinion and possibly repair a suspected HG leak.
On this KV6, I could find no external leak to speak of and the pressure cap wasn't blowing, but it was loosing similar amounts of coolant as the OPs car. I didn't do a compression test iirc, but the HG was blamed for the coolant loss by the garage that looked at it before me. After I'd removed the engine for a strip down, I was taking the thermostat and link pipe off, when I spotted a small knick in the link pipe O ring. The link pipe is on the suction side of the water pump. So a small hole here will draw air into the coolant jacket and out the top bleeder pipes, but this is only when the cap is off. With the cap fitted, coolant pressure would stop air being drawn in and small drip would have been evident under the pipe, but as the engine is over 100°C this drip simply evaporated off almost instantly, so I didn't spot it

I spoke to the owner, suggesting that the HG might not be the issue. But as he had got all the parts, said replace the HGs anyway. I rebuilt the engine and fitted a new thermostat and it's pipework.
The engine ran perfectly for the next 5 years that I know of.
So I'm generally more couscous before jumping on the "it's the head gaskets mate" waggon. ;)
 
Last edited:
I kind of agree and what you say is valid for most engines. However I have been caught out by similar bubbling in the coolant before on the KV6. I was asked to give a second opinion and possibly repair a suspected HG leak.
On this KV6, I could find no external leak to speak of and the pressure cap wasn't blowing, but it was loosing similar amounts of coolant as the OPs car. I didn't do a compression test iirc, but the HG was blamed for the coolant loss by the garage that looked at it before me. After I'd removed the engine for a strip down, I was taking the thermostat and link pipe off, when I spotted a small nick in the link pipe O ring. The link pipe is on the suction side of the water pump. So a small hole here will draw air into the collant jacket and out the top bleeder pipes, but only when the cap is off. With the cap fitted, a small drip would have been evident under the pipe, but as the engine is over 100°C this drip simply evaporated off almost instantly, so I didn't spot it

I spoke to the owner, suggesting that the HG might not be the issue. But as he had got all the parts, said replace the HGs anyway. I rebuilt the engine and fitted a new thermostat and it's pipework.
The engine ran perfectly for the next 5 years that I know of.
So I'm generally more couscous before jumping on the "it's the head gaskets mate" waggon. ;)
I can quite understand your hesitancy after an episode like that ! - we have all been 'burnt' when #making the call# :( - what you say is quite enlightening and logical, and we can only go off what someone has said and NOT something we seen or done, which makes thing extremely hard to pin down.
One thing to consider is that he was testing with a coolant sealed pressure tester - when getting the pressure increase he also had bubbles. He also said he let it run (on a sealed coolant pressure tester) to 19PSI before stopping. Even at this stage bubbles were evident ! - so we have a pressure increase due to bubbling in the coolant system on a close circuit (which should never arise) and would not, IMO be caused by an intake suction issue although It is not clear if coolant loss was noted at other areas. I fully appreciate the implication of what you are saying re air intake on the 'suction (low pressure) side but in this case (only with the info I have seen) - it does not appear to be such a case.
We do have some option though.
Using a catch tank will show coolant output from the coolant header - and not elsewhere. A definitive test is to rev at cold and literally smell the escaping bubbles - if definite combustion smell I would have no hesitancy at making the call without fear of being 'burnt'. I would also make the call on fluid loss (with all the other symptoms) via the coolant expansion tank.
Regards Nodge,
Joe
ps, how close are you to the 'pointy end' of the UK - :) - I spent 8 years of my life on the Isles of Scilly. .. (and yes, I have heard all the jokes:rolleyes:)
 
I can quite understand your hesitancy after an episode like that ! - we have all been 'burnt' when #making the call# :( - what you say is quite enlightening and logical.
We do have some option though.
Using a catch tank will show coolant output from the coolant header - and not elsewhere. A definitive test is to rev at cold and literally smell the escaping bubbles - if definite combustion smell I would have no hesitancy at making the call without fear of being 'burnt'. I would also make the call on fluid loss (with all the other symptoms) via the coolant expansion tank.
Regards Nodge,
Joe
ps, how close are you to the 'pointy end' of the UK - :) - I spent 8 years of my life on the Isles of Scilly. .. (and yes, I have heard all the jokes:rolleyes:)

On rereading the OPs tests. I suspect that this could well be the HG. I would like to use my endoscope in the low cylinder, to see if residual coolant pressure is allowing coolant into the combustion chamber after shut down. Seeing any coolant seeping in would be conclusive.

I live around 35 miles from the pointy end Joe. My sister lives on St Martins IOS. It's a lovely place in the summer. I can't say I'd like to be there in the winter though.
 
On rereading the OPs tests. I suspect that this could well be the HG. I would like to use my endoscope in the low cylinder, to see if residual coolant pressure is allowing coolant into the combustion chamber after shut down. Seeing any coolant seeping in would be conclusive.

I live around 35 miles from the pointy end Joe. My sister lives on St Martins IOS. It's a lovely place in the summer. I can't say I'd like to be there in the winter though.
Hello Nodge,
Which endoscope do you use ? I have a cheapo ebay chink one - it is a superb image and light - just no strength in the cable - ie - no method of setting the camera / cable to the correct angle - just a 'floppy' cable - I added some 'bailing wire' and taped it to the cable and it is now excellent.
The winters are great on the IOS ... you just spend them going from the Mermaid to the Bishop and Wolfe (via the Atlantic hotel bar), then to the 'Club' :) :)
Seriously it is a lovely place in the winter.
How long has she been on St. Martins ?
My dad was a school teacher there (Primary) - I did all my secondary schooling there - age 11 to 16 - the IOS secondary was the absolute top in the first schools league tables ..... mind you - our WHOLE 5th year at school was 5 boys and 14 girls - what more can you say...:oops:
I worked on the inter island boats for several years - also gig racing crew for a year. (so was my bro)
(that was in the early 70's - left IOS secondary school in 1973) - sorry , but again I digress.... doh...
One of the boats I worked on for a long time was the 'Black Swan' which is now in Falmouth... I was shocked when I saw her in 2004 when we were awaiting the biscay crossing in our 'yott' - as I hadn't seen her since the mid 70's - last reports I had prior to Falmouth were that she had been sold to the Island Hotel on Tresco, then had sunk in a gale... I though she was a write off...

Did your sister marry a Goddard by any chance :) ?

edit - re winter on IOS - yes I do agree if in St. Martins - begger all to do -- :(

If on St. Mary's though - way hey - a life of debauchery and out of mind experiences....:D

Me gads sir !.. :rolleyes: - I certainly had lots and lots and lots of that ;) - well, it WAS the 70's :eek:
 
Last edited:
Hi Joe. My endoscope is a £5.99 Ebay special. I had the same floppy cable problem. Some soft iron wire taped to the cable sorted that out.

My sister moved to St Martins in 1992. She now owns the camp site there. She married a lad from the mainland and moved there for a different way of life.
I guess it works for her. For me, there's not nearly enough stimulus for my active mind!!
 
Wow thanks to you both for spending lots of time on this issue.

@Nodge68 to test your idea of air coming in via the Thermostat piping, at what PSI do you think I should test the cooling system cold and not running ? Kaiser on another forum suggested going up to 32 psi.

@Joe_H I did try to look for bubles with the cap off and the car idling. I could not see any. The only way I can see anything is with the combustion gas tester just like in the video. If I do this test with the car warm, the expansion bottle overflows as the level keeps and keeps rising over and eventually over where the cap would be.


The supposition I currently have is:
There is a small leak allowing combustion gases into the colant. These gases are slowly raising the pressure in the coolant system until such time as the expansion tank cap release pressure, loosing a small amount of coolant at the same time. And the cycles continues. How do I attached something to the expansion tank to catch this suspected expelled coolant at 16 psi ?


Now if this is the case, coolant should be displaced away from the leak and thus I am speculating that I should get a hotter area near one of my cylinder liner as the liner would have less coolant to cool with.

I want to run as many and as varied tests as possible so that if I ever take the cylinder head(s) off, this information can help others.
 
Pumping the cooling system up to 30Psi won't harm anything that's in good condition. The pipes should hold 50Psi without an issue. The tank will be the weakest part of the system and so will fail first.
 
@Nodge68

I am trying to do a leakdown test. I first removed the gasoline pump relay and then removed the side cover to get access at the crankshaft pulley. I then removed the spark plug of cylinder 3. I found TDC using a piece of a metal cloth hanger and I found the compression stroke using a kids small balloon on the end of my leakdown tester hose.

The issue I have is the moment I put more then 30-35 psi the crankshaft pulley moves. I tried putting a breaker bar on the 22 mm bolt but my concern is if I put too much psi of air it would remove the crankshaft bolt pulley.

Are my fears real ? Or should I brace that bolt better ?
 
If you can find the exact top dead centre TDC. Then the rod will be pushing straight down on the crank pin. This is best described as "rocking" at TDC. At the moment, you must be just off TDC so the piston is doing what it's designed to do and rotating the crank. Holding the pulley nut might work, but only if the crank is turning backwards. This would efficiently be trying to tighten the pulley bolt. So you need to tweak the crank round sufficiently to hold the piston at a "rocking" TDC. Or if that proves impossible to find. Allow the engine to try to rotate backwards and hold the pulley bolt still.
 
If you can find the exact top dead centre TDC. Then the rod will be pushing straight down on the crank pin. This is best described as "rocking" at TDC. At the moment, you must be just off TDC so the piston is doing what it's designed to do and rotating the crank. Holding the pulley nut might work, but only if the crank is turning backwards. This would efficiently be trying to tighten the pulley bolt. So you need to tweak the crank round sufficiently to hold the piston at a "rocking" TDC. Or if that proves impossible to find. Allow the engine to try to rotate backwards and hold the pulley bolt still.

Thanks! You indeed confirmed what I experienced. I was not quite at TDC.


The leakdown tester with 90ish psi of air applied did not show any bubbles in the coolant expansion tank.

Another test I ran is I tested the cooling system at 22 psi, no leaks.



Pictures of air gauge for leakdown test.

image.jpeg


Compressor air gauge

image.jpeg


Leakdown tester air gauge
 
Hi, Zefrench, I believe you are simply not getting enough pressure. I
You had bubbles in the coolant at idle so, at when cold, start the engine with reservoir cap removed and observe the bubbles in the res'. Now increase the revs of the engine and observe the bubbles.
During this test do not attach any from of pressure devices etc, purely by observation, a slight oveflow of fluid - especially if warm - is normal. It is the bubbles you are looking for.

The fact that your system holds pressure at 22 psi, a compression test shows a low pot, the pressure in the coolant rises at idle with a coolant pressure tester (you stopped the test at 19 - above the blow of pressure and quite rightly so ! - but, as you observed, the pressurisation of the system continued to rise !), you have constant and observable bubbles in the system, and, you lose coolant is totally indicative of HGF.
That is enough information to make the diagnosis conclusive IMO.

If you really want to - you can also attach a container bottle to the overflow of the expansion to confirm fluid is actually coming from the expansion.(but this means little as you KNOW it is over pressurising hence coming from the internal compression of the cylinders.

You should also be able to see the increase in bubbles as you rev the engine (cold) - this is totally abnormal.
Hope that helps. I appreciate you want a belst and braces diagnosis, - but you actually have one.
Joe
 
Last edited:
I kind of agree and what you say is valid for most engines. However I have been caught out by similar bubbling in the coolant before on the KV6. I was asked to give a second opinion and possibly repair a suspected HG leak.
On this KV6, I could find no external leak to speak of and the pressure cap wasn't blowing, but it was loosing similar amounts of coolant as the OPs car. I didn't do a compression test iirc, but the HG was blamed for the coolant loss by the garage that looked at it before me. After I'd removed the engine for a strip down, I was taking the thermostat and link pipe off, when I spotted a small knick in the link pipe O ring. The link pipe is on the suction side of the water pump. So a small hole here will draw air into the coolant jacket and out the top bleeder pipes, but this is only when the cap is off. With the cap fitted, coolant pressure would stop air being drawn in and small drip would have been evident under the pipe, but as the engine is over 100°C this drip simply evaporated off almost instantly, so I didn't spot it

I spoke to the owner, suggesting that the HG might not be the issue. But as he had got all the parts, said replace the HGs anyway. I rebuilt the engine and fitted a new thermostat and it's pipework.
The engine ran perfectly for the next 5 years that I know of.
So I'm generally more couscous before jumping on the "it's the head gaskets mate" waggon. ;)

Looking more closely I saw a "small water puddle" near the thermostat in the V of the engine, no idea how old it is. I am wondering how I could test for leaks in the straight pipe at the pump or thermostat as this might fit my situation.

I am thinking that because when I came back from driving 20 minutes, waited a bit, removed the cap, the coolant level would raise very very quickly, yet the moment I would stop the car, I would hear a sound similar to water rushing where air used to be.

I know I can remove the inlet manifolds to get to this, but how would I test this is not happening, when I think have good pressure with the coolant pressure tester,
 
On my V6 I had a leak from the front intake manifold gasket. The water passages pass through the gasket at the lower outer corners.
The leak on mine saw coolant coming out at high pressure which pooled in the vee of the block under the thermostat.
Here is a pic of the water passage with the manifold off. You can see the remains of the black aftermarket gasket next to the bolt of the alternator mount. It failed at the bottom edge and was very difficult to see as the leak was about the diameter of a sewing needle.
 
Last edited:
Looking more closely I saw a "small water puddle" near the thermostat in the V of the engine, no idea how old it is. I am wondering how I could test for leaks in the straight pipe at the pump or thermostat as this might fit my situation.

I am thinking that because when I came back from driving 20 minutes, waited a bit, removed the cap, the coolant level would raise very very quickly, yet the moment I would stop the car, I would hear a sound similar to water rushing where air used to be.

I know I can remove the inlet manifolds to get to this, but how would I test this is not happening, when I think have good pressure with the coolant pressure tester,
One way to isolate if water is coming from cap or otherwise is to fit a 'fluid indicator' to the expansion tank. It could be kitchen towel /paper (not sue what the correct name is in Canada) around the expansion tank and screw top area to catch any coolant / fluid that may be being ejected from there. The colouring in the antifreeze will help, also the paper, even if it dried (unlikely) would show obvious evidence of having been saturated.

The other worrying part for the alternative leak point (Which would be just that would super :) - well kind of ;) - depending WHERE it was) is that the pressure rose to at least 19psi (before you intervened and stopped the testing)with your coolant pressure tester on. (hence a sealed system) with pressure continuing to rise it would indicate an internal pressurisation of a closed system which can only be coming from the combustion process. the leak may be elsewhere, however the pressure is still rising above normal.
When you observed bubbles with the pressure tester on you are effectively testing against a pressurised circuit also (I presume the system was pressurised when you observed the bubbles ?)
Nodge raises some very interesting points but I do not see how the coolant pressure continued rise could be occuring.
Again, best of luck. You are doing the correct thing in trying to postively identify the issue as the work involved in HGF is major.
Joe
 
One way to isolate if water is coming from cap or otherwise is to fit a 'fluid indicator' to the expansion tank. It could be kitchen towel /paper (not sue what the correct name is in Canada) around the expansion tank and screw top area to catch any coolant / fluid that may be being ejected from there. The colouring in the antifreeze will help, also the paper, even if it dried (unlikely) would show obvious evidence of having been wet.
(I presume the system was pressurised when you observed the bubbles ?)
Nodge raises some very interesting points but I do not see how the coolant pressure continued rise could be occuring.
Again, best of luck. You are doing the correct thing in trying to postively identify the issue as the work involved in HGF is major.
Joe

The stat link pipe can only allow air to enter the system when there is no pressure (cap off).
When the system is above atmospheric pressure, a bad link pipe seal can allow coolant to leak out, and potentially evaporate away, unseen.
 

Similar threads