Water not returning from expansion tank to radiator

[Babord]

I am struggling with a recurring issue of coolant not returning from the expansion tank into the radiator after cooling. I move coolant back from the tank into the rad, warm the engine up with the rad cap off (I have also tried with the expansion tank cap also off at the same time), I wait to the air bubbles stop and then tighten everything up. But then after a drive, when I go to start the car the following day, the level in the expansion tank has risen towards the top (again, it has not returned to the rad). I have had a small leak from the left rear inlet manifold - when I pump 10 psi into the expansion tank I can see two different drips leaking onto the Vee. After pressuring this way, I also noted the level in the rad dropped a little.

Searching the forum it seems others have small leaks into the Vee, but don’t have the same issue as me. So, does that mean there must be an additional problem, as well as the inlet manifold? If, so what are the suggestions to test further? I am a bit nervous of changing the inlet manifold gasket as it seems this can be problematic, and I don’t want to make matters even worse!

Also, can someone confirm the normal level of coolant in the rad, after cooling down? If I remove the top brass plug (mark 2 Stag) should it be right to the top, or will flow still be ok if level with the bottom of the top-hose, or even a little lower?

[DJT]

If you are running the OE style expansion tank, are you sure the pressure cap is sealing correctly? There are some around which do not. Also, is the neck of the expansion bottle sealing where it joins the bottle? Again, there are some that do not.

[Lingen]

Babord.. Coolant not returning to the radiator could be because the expansion bottle is not seating properly. The pressure cap could be leaking, or even the brass top of the bottle may not be sealed to the plastic body..
The "Normal" level in the radiator can vary from car to car as it all depends how far down in to the radiator the metal overflow pipe goes, and it seems no two are the same.. Any air getting in to the system will collect in the radiator, and lower the level of the coolant till it reaches the bottom of this pipe, and then the just air will go into the bottle.
To be really pessimistic, it might be that a head gasket is leaking a bit. Have you checked the torque settings of the head studs and bolts lately?
Mike.

[sujitroy]

I'm assuming u have standard mk2 setup. Since you have a pressure tester as mentioned, you could brush soapy dishwashing liquid around the area between the plastic bottle and metal neck where the cap fits and the brass radiator plug if there is a leak, you should see bubbles. This is a common method used to find gas leaks.
Sujit

[sujitroy]

Another weak area is the exit of the plastic pipe from the over flow bottle, but you'd see water leaking out of that.

[Babord]

It would be nice think it is something as simple as the expansion bottle itself. As suggested, I will use the pressure tester tomorrow with some soapy water around metal neck area.
Torque settings of head studs and bolts were checked approx 2 months ago and were fine.
Lingen, thanks for the explanation of the rad level - one less thing to worry about, although the head gasket has been at the back of my mind. I should probably buy one of those gas test kits.
Thanks all - will return to the problem tomorrow.

[Goldstar]

If you’ve got a pressure tester with a gauge you shouldn’t need soapy water as the gauge will fall if you have a leak although the bubbles could be a good indicator of where the leak is occurring if you do have one.

[stagwingnut]

for what its worth .i bought a reconditined rad from one of our suppliers , my engine started to run hot with it installed,,very unusual for my car , i then put my old rad back in , run fine , so i investigated the recon rad only to find that the metal return pipe thats braized to top of rad ,was completely blocked with rust , i had to have a new pipe braized in it \, so worth checking yours is not bocked . i had to saw off the bend . then push a coat hanger down the straight section on top of rad , thats when i found the blockage ,regards graham

[MikeParker]

I don't think that the water bottle cap sealing is the issue. Mine leaks for sure and the liquid does return to the engine from the expansion tank when cold. I also run a standard setup. I have also at times run with the cap only loosely fitted just to avoid the system getting pressurised. In that state as well the water returns to the block when the engine cools. I can only think that you have quite a major leak somewhere. Your method of pressurising the system and looking for leaks is a good way to go.

[Babord]

Update - the expansion bottle does not appear to be leaking. I don't believe there is a "major" leak anywhere - just a small dripping leak from the rear of the inlet manifold - and the system appears to be losing very little coolant (after transferring from the bottle back to the rad, I don't often need to top up). Scratching my head..any ideas?

btw, radiator is almost new from EJ Wards, I don't believe there is an issue - there certainly wasn't when I flushed and tested it a few months ago when fitting - and I am not ready to "saw off the bend" to double check just yet!

[StagJonno]

Agree, very unlikely to be a big leak, otherwise it'd be easy to trace. Had a similar problem on 2 occasions. The first was a poorly sealing top hose to rad connection where the Kenlowe fan temp sensor was - sorted this using a formed silicon sleeve from Holdens. Second time was a pinhole in the brazing of the pipe feeding the expansion bottle where it enters the rad. Both very hard to spot and luck played a part.

LD Part (and others, no doubt) have a UV tracer fluid you could try putting in with the coolant. You'll need a UV torch, readily available on the likes of Amazon which also work as powerful white beam torches.

[Babord]

Ah, I also have an inline fan sensor (Revotec) fitted, so I will go and double check those connections. Do you have a link or photo of the "formed silicon sleeve"? I couldn't find anything that looked right on Holden.
I didn't know about UV tracer for coolant, so will go and source some. Thanks!

[StagJonno]

Here you go https://www.holden.co.uk/p/kenlowe_seal . A bit hidden! If you've got the Revotec thermostat type that fits inline in the top hose, it may not be appropriate - the Kenlowe one is in the form of a capiliary that is snuck into the top hose to rad connection.

Discovered the UV tracer after I'd sorted my issues. I hope it works and will be interested to hear how you get on.

[mjheathcote]

I would really question the original expansion bottle.
Mine failed around the neck, 35 years ago!
This was when I first fitted a turret mounted expansion header tank.
After all this time I would be amazed if any don't leak around thus area.

[ukbiggles]

After all this time I would be amazed if any don't leak around thus area.

I had this 'coolant not returning' problem recently. Replaced all hoses and clips to prevent air leaks but problem persisted. Tried pressure testing the expansion bottle and even bought one on For Sale and it was fine too. My problem turned out to be the worst problem. Head corrosion leading to exhaust gases getting into the cooling system. I sincerely hope your issue is not as drastic. Which is a long winded way of saying I have a couple of spare bottles available that have been tested and do not leak. ( I have a new cooling system now, not involving the original expansion bottle.)

Cheers

Gord