Cooling System Problem

Cooling System Problem

Author
Discussion

cbjroms

Original Poster:

23 posts

153 months

Monday 14th September 2020
quotequote all
So I am running a Herald 13/60 engine in my kit car. The top of the radiator sits lower than the thermostat housing due to the shape of the bonnet. So, in the hose which connects the bottom of the radiator to the thermostat housing I have added a T piece, one limb of which is connected to an expansion tank. The expansion tank sits slightly higher than the top of the thermostat housing.

Just got the car on the road - 50 miles since first MOT. I was out in the car this afternoon and noticed that the temp guage was reading very high. First thought was that the temp controlled radiator fan had failed but that all checked out fine. Then I noticed that the radiator felt cooler than anything else and that the expansion tank was bubbling. I carefully removed the pressure cap from the expansion tank (it was full to the top) and lifted it higher above the engine. There was a gurgling noise, the level in the tank dropped back to normal, the radiator heated up and the fan kicked in.

I had presumed that with the exapnsion tank being higher than both the thermostat housing and radiator, any trapped air would make its way to the expansion tank. But today, the air seems to have collected in the radiator. What am I missing?

stevieturbo

17,214 posts

246 months

Monday 14th September 2020
quotequote all
Draw a schematic of exactly how it is plumbed.

cbjroms

Original Poster:

23 posts

153 months

Monday 14th September 2020
quotequote all
Thought some photos might be better than a schematic.

In photo below (looking from offside), 1 is the bottom radiator hose from the water pump. 2 is the radiator top hose to thermostat housing. 3 is the T piece and 4 is the radiator cap.



In photo below (looking from nearside), 5 is the thermostat housing and 6 is the hose that comes off the T piece and feeds the expansion tank.



In photo below (looking from nearside), 6 is the hose that connects the T piece to the expansion tank (7).



The radiator filler has a pressure cap and drain. The drain is sealed with a piece of hose which has a bolt epoxied into it.

The exspansion tank has a pressure cap, bottom feed (hose 6 connected) and 2 top drains, both drains have been sealed.

So what would cause the water to move to the expansion tank and leave the radiator full of air?

MexScort

54 posts

91 months

Monday 14th September 2020
quotequote all
Looking at your pics i would suggest you have a couple of things to change:

1. The ‘t’ piece for your expansion bottle should be in the rad outlet (lower) connection where temps and pressure are slightly lower.

2. You don't appear to have a degas connection. On your setup this can go from the top of the rad by the filler cap (the overflow), to the top of the expansion bottle - this time not the overflow but one near the top of the bottle. This will allow the system to release air to the tank, and is likely the cause of your issue, as it wouldn't have been able to.

3. In relation to point 2, when making this connection, you must fit a non pressure cap on the rad, its just a plain cap. This will leave the degas connection open permanently and overall pressure release will be determined by the expansion bottle cap.

Arnie Cunningham

3,754 posts

252 months

Monday 14th September 2020
quotequote all
Yes, needs a bleed valve or something similar at the high point.

stevieturbo

17,214 posts

246 months

Monday 14th September 2020
quotequote all
A schematic would have been easier lol..

As others have said, the expansion bottle should fill the system via the bottom hose on the radiator.

Given the downward journey from engine to rad....IMO there needs to be a bleed point from the stat housing to the header tank where your "7" is, instead of that bolted pipe.

As long as the cap on that radiator is a fairly strong spring and it's a single action cap...wouldnt worry too much about it.
It must not be one that allows air to be drawn in during cooldown.

And route fuel lines better removing strain from the fuel filter, and use correct clips for small diameter hose.




cbjroms

Original Poster:

23 posts

153 months

Tuesday 15th September 2020
quotequote all
Thanks everyone,

@stevieturbo says '....IMO there needs to be a bleed point from the stat housing to the header tank where your "7" is'.

and @Mexscort says '......degas connection. On your setup this can go from the top of the rad by the filler cap (the overflow), to the top of the expansion bottle'.

My radiator is from a Honda Civic and the filler 'vent' is always open (thats why I balnked it off) rather than open when the cap is under the right pressure.

So, if I fit a hose between the radiator vent and the top side (rather than filler neck vent) of the expansion bottle will this prevent air being trapped in the stat housing?

GreenV8S

30,129 posts

283 months

Tuesday 15th September 2020
quotequote all
cbjroms said:
So, if I fit a hose between the radiator vent and the top side (rather than filler neck vent) of the expansion bottle will this prevent air being trapped in the stat housing?
Since your stat housing is the highest point, you will have to hope that any air in there gets swept down to the radiator when the stat is open. It's not at all obvious that it will since the hose is going down hill, but that's your best bet. Venting is going to be a challenge, though. If that was me I'd be looking to put a vent line from the top of the top hose up to the header tank, or use any other output from the stat housing as a vent.

stevieturbo

17,214 posts

246 months

Wednesday 16th September 2020
quotequote all
cbjroms said:
and @Mexscort says '......degas connection. On your setup this can go from the top of the rad by the filler cap (the overflow), to the top of the expansion bottle'.

NO.

It does need a bleed/degas or whatever some might call it, but that is not it.



cbjroms

Original Poster:

23 posts

153 months

Wednesday 16th September 2020
quotequote all
@stevieturbo says 'It does need a bleed/degas or whatever some might call it, but that is not it.'.

So I could either:
a) drill/tap thermostat housing and connect a hose from here into the side (rather than bottom) of the explansion tank. Cannot see a suitable fitting for this on the Internet.
b) put an in inline radiator filler neck (with a recovery cap) into the top hose. If I used a longish flexi for the top hose then I would be able to vent the system by lifting the neck above the thermostat housing when I fill.
c) Put a second inline T piece in the top hose (as close to the thermostat housing as possible) and vent this into the sideof the expansion tank.

Am I on the right lines here?

MexScort

54 posts

91 months

Wednesday 16th September 2020
quotequote all
For what its worth, I'd still be adding a degas pipe in from the small top connection on the rad, to the expansion bottle to ensure the rad can can always expel air, and given you already have the connections for this at both ends, its pretty straight forward.

I do agree that the most likely place on your setup for an air lock though, is the stat housing given its location. So if you can add something here as well, that would be ideal. You should be able to 't' this connection in to the degas pipe from the rad to the expansion bottle, covering all bases.

GreenV8S

30,129 posts

283 months

Wednesday 16th September 2020
quotequote all
cbjroms said:
@stevieturbo says 'It does need a bleed/degas or whatever some might call it, but that is not it.'.

So I could either:
a) drill/tap thermostat housing and connect a hose from here into the side (rather than bottom) of the explansion tank. Cannot see a suitable fitting for this on the Internet.
b) put an in inline radiator filler neck (with a recovery cap) into the top hose. If I used a longish flexi for the top hose then I would be able to vent the system by lifting the neck above the thermostat housing when I fill.
c) Put a second inline T piece in the top hose (as close to the thermostat housing as possible) and vent this into the sideof the expansion tank.

Am I on the right lines here?
I would go for 'C', aiming for the vent outlet to be a small outlet coming out of the top hose as high as possible and preferably aligned with the direction of flow.

stevieturbo

17,214 posts

246 months

Wednesday 16th September 2020
quotequote all
cbjroms said:
@stevieturbo says 'It does need a bleed/degas or whatever some might call it, but that is not it.'.

So I could either:
a) drill/tap thermostat housing and connect a hose from here into the side (rather than bottom) of the explansion tank. Cannot see a suitable fitting for this on the Internet.
b) put an in inline radiator filler neck (with a recovery cap) into the top hose. If I used a longish flexi for the top hose then I would be able to vent the system by lifting the neck above the thermostat housing when I fill.
c) Put a second inline T piece in the top hose (as close to the thermostat housing as possible) and vent this into the sideof the expansion tank.

Am I on the right lines here?
a. very easy, just drill and tap it for an NPT fitting, or weld a suitable fitting on. Choices are many.

b/c, messy, why bother.

And the hose Mex seems to refer to on the rad is an overflow, it is not part of the pressurised system, and for this discussion is largely irrelevant. And as the path from stat housing to rad is naturally going to trap any air in the stat housing, it's pointless even worryng about the radiator itself.

cbjroms

Original Poster:

23 posts

153 months

Wednesday 16th September 2020
quotequote all
Thanks again everyone

cbjroms

Original Poster:

23 posts

153 months

Saturday 17th October 2020
quotequote all
I have changed the system so that hose 6 connects the 25mm spout on the bottom of the expansion bottle to a T piece in the bottom radiator hose (as per advice above).

Rather than fit an air bleed into the top radiator hose or thermostat housing, I remembered that my initial plan had been to unbolt the radiator and lift it above the thermostat housing when filling the cooling system.

So I filled the system with the radiator filler above the thermostat and ran the engine with the expansion bottle cap in place. As the engine got warm, I replaced the radiator cap and ran the engine until everything got nice and hot (temp gauge read normal). I then turned the engine off an allowed to cool, at which point I removed the radiator cap and found the radiator was not full to the top. So I loosened the expansion tank cap, water started to come out of the radiator filler and so I repaced the radiator cap. Went through this cycle a number of times (radiator was clearly getting very hot) before lowering the radiator down and bolting it back into position.

On road testing the car this afternoon the temperature gauge remained very low but when I got back and stopped in the garage the expansion tank starting venting water, all hoses felt hard (there seemed to be bubbling/boiling in some of the hoses) but the radiator was quite cool (thought this was due to have been out on the road and the air temp being quite low).I also noticed that a couple of new water leaks had developed and assumed that the cooling system was over pressurising? After the engine had cooled, I removed the radiator cap and it looks as if there is no water in the radiator.

So having bled the system, something seems to have caused the radiator to fill with air and its contents to spew out of the expansion tank. How can the system over-pressurise when the expansion tank cap can vent?


stevieturbo

17,214 posts

246 months

Saturday 17th October 2020
quotequote all
If you do not have a bleed at the stat like I said, you're creating a massive air trap....IMO it will not work, or at best it will be incredibly difficult to get to work.

And if you do not know what pressure there is in the system...you cannot know when the cap should vent...or if you have not tested the cap installed on the system, you do not know at what pressure it will vent.

I logged a friends car some years ago...we knew the HG had let go, but he wanted to get a couple of runs around the track at a lower boost anyway. The cooling system was recording over 30psi...but he disputed this, as the cap would have vented.

Well yes....if it was the correct cap for the neck on the custom header tank. Except it wasn't, and it was rammed tight shut.
It held fine for a couple of laps which was all he wanted anyway.

GreenV8S

30,129 posts

283 months

Saturday 17th October 2020
quotequote all
cbjroms said:
So having bled the system, something seems to have caused the radiator to fill with air and its contents to spew out of the expansion tank.
It can take a long time to chase air pockets out of the system, and even after you've done that there will be trace amounts of gases entering the system which need to be expelled. If your system needs such extreme measure to perform the initial vent, it isn't venting properly. Even if you carry on with that procedure over a few weeks and manage to expel all the air currently in the system, it will be prone to accumulate air pockets.

You need to ensure that all the air in the system gets carried out to the pressure cap where it can vent out. The engine itself will almost certainly have been designed to move air out to the manifold at the top of the engine and your challenge will be to get it out of there, and all the external pipes and components. It's not conceptually difficult to do, but to do it you need to have a clear understanding of the geometry of the plumbing.

tapkaJohnD

1,926 posts

203 months

Sunday 18th October 2020
quotequote all
cbjroms,

The original Triumph engine layout included as you may know a header tank mounted above the engine, as the radiator was at the level of the top of the engine. Yours is significantly lower! But perhaps a lesson from Triumph is to treat the header tank as a header, and not just a refill tank.

The original Spitfire radiator had no filler cap, but two large bore hoses to the top. One wnet to the thermostat housing, the other to the header tank, which itself was mounted slightly tilted down towards the rad, with the filler cap at the back.

Your rad doesn't have two top inlets, and we can't see your header, but perhaps teh Triumph layout might guide oyu?

John
PS what car is the engine in?

stevieturbo

17,214 posts

246 months

Sunday 18th October 2020
quotequote all
The above is a slightly unusual layout, but schematics do appear to show the right hose to be a bleed for the rad.
Not ideal, but certainly better than the OP is doing. It will still encourage air to go into the header, although still leaves a potential air trap at the stat housing which seems to be a high point.

https://mossmotors.com/cooling-system-1962-80

https://www.triumphexp.com/phile/8/61395/Spitfire_...

cbjroms

Original Poster:

23 posts

153 months

Sunday 18th October 2020
quotequote all
Really apprecaite this help - thank you everyone.

@stevieturbo - I hadnt realised that the air bleed needs to be always available (rather that just for filling). So I need a permanent and always open bleed from the highest point (ie thermostat hosusing) to the middle (ie above the water level but below the pressure cap) of the exampnsion tank. Thus, any air that finds its way into the thermostat housing will vent into the expansion tank and be replaced by the cooling system drawing water from the expansion tank?

@GreenV8s, the more I think about it the less certain I become about the geometry of the colling system. You say about getting the air to a presssure cap where it can vent out. Surely a pressure cap will only allow air to vent if it is either opened or has an open bleed to the expansion tank?

@tapkajohnD, the kit car is a Sammio (some photos here: http://www.madabout-kitcars.com/forum/showthread.p... I am not clear about the difference between a header tank and refill tank.Is it that my expansion tank does not have a feed from the top of the system and so it is only dealing with coolant exapnsion/contraction and not preventing airlocks?.