thermal store

[hosegood]

I put the bypass on to try to solve this problem but it happens with or without.
The pump is on the slowest speed,its a Wilo 6 metre head, not the best I guess.
The air in the system seems very possible but I can not think where it is as there are only 2 rads which I have bled many times, how much air are we talking about?
The negative pressure in the top of the tank is also a possibility but how could I change that?
I can raise the vent pipes and I have just thought that maybe the F&E tank is too full, this might stop the overflow but it will not stop the suck down when the pump goes on but maybe that is OK?
Maybe I should get the other rads in place so I have a complete system.
Thanks again, if anyone thinks of anything else. I will let you know what happens
Laurie

 

[hosegood]

Also there is not a room stat just on and off rad valves and the lockshield are all fully open

 

[SimonH2]

The air in the system seems very possible but I can not think where it is as there are only 2 rads which I have bled many times, how much air are we talking about?

A lot more air that that would be required - so I think we can rule that out.

The negative pressure in the top of the tank is also a possibility but how could I change that?

I think that's a red herring - you can't change it as it's a function of the system design. The pressures at the top and bottom of the cylinder will be equal to the head available from the F&E tank +/- any dynamic pressure drops due to flow through the pipes. In the steady states, there is no flow in the pipes and so the pressures must be equal to the static heads.

The only way to get pressure differences around the tank is through dynamic effects. The only one I could think of here would be that if the CH return pipe was directly in line with the feed pipe from the F&E tank and there was a very high flow rate, then in theory it would be possible to create a slight pressure increase as the return from from teh CH loop travels across the store and "hits" the end of the feed pipe. But given the flow won't be high, and the store is fairly wide, I can't see this being an issue.

So really I cannot think of any realistic mechanism for what's happening.

Can you confirm my interpretation of your symptom description ?

When the pump starts, the F&E level drops - and stays at that lower level while the pump is running
When the pump stops, and not before, there is flow from the vent pipe back into the F&E tank and the level is restored

 

[hosegood]

Yes that's exactly what happens. What I find strange is that it only happens with the small rad, which is on the same level as the store.

These rads are 2.5 & 3 metres long and low. I got them second hand from a friend. I tested them using a garden hose to about 3 bar to check for leaks. I turned the hose off and when I undid the blanking nut the water pressure was amazing, made a hole in my tarmac. I guess that was air compressing. As the rads are so big maybe there is a lot of air trapped.
I have of course bled them and Iam trying to jack up the smaller one to get the air up to the bleed screw but so far no air.

The CH return pipe port is very slightly lower than the cold feed pipe on the tank.

As I said, on the small rad I have 15mm pipe both on the flow and return for the last 1.5 metres. Could this not cause a restriction on the return?
Thanks again for your thoughts.

 

[SimonH2]

Yes that's exactly what happens. What I find strange is that it only happens with the small rad, which is on the same level as the store.

These rads are 2.5 & 3 metres long and low. I got them second hand from a friend. I tested them using a garden hose to about 3 bar to check for leaks. I turned the hose off and when I undid the blanking nut the water pressure was amazing, made a hole in my tarmac. I guess that was air compressing. As the rads are so big maybe there is a lot of air trapped.

Well just hooking up a hose and not bleeding them would leave a lot of air, and once released that will give quite a gush. That why pressure vessels are pressure tested using a non-compressible liquid (usually water). Because it's non-complressible there is very little stored energy to be released should the vessel fail. If you use compressed air, then there's a lot of energy and a failure can result in very dangerous explosive rupture (blast of air which is dangerous in itself, and shrapnel flying about).

But sticking with your problem, I'm still thinking the only likely cause is air being compressed in the system. When it's compressed, the level in the F&E tank drops, when the pressure is released, the level recovers as the air expands back again.

I have of course bled them and Iam trying to jack up the smaller one to get the air up to the bleed screw but so far no air.

Are these standard rads ? They shouldn't contain all that much air once you've bled them.

As I said, on the small rad I have 15mm pipe both on the flow and return for the last 1.5 metres. Could this not cause a restriction on the return?

It shouldn't normally matter since water is non-compressible - so having a restriction on the return merely means more pressure in the rad. In terms of flow rates, it doesn't matter where any restrictions are places, they all just add up along the pipe. Eg, it doesn't matter whether you throttle flow with the valve on the flow end, or with the valve on the return end - the overall pressure drop in the circuit will be the same, but one will have a higher pressure in the rad than the other.

I take it the rad isn't ballooning at all

 

[hosegood]

What is a ballooning radiator?

They are steel column radiators. When we moved them they snake and bend as they are so heavy. I was always worried about breaking the joints between the fins. But they seem OK ( unless they are ballooning)[/list][/code][/quote]

 

[SimonH2]

What is a ballooning radiator?

There's probably some technical name for it in the trade, but I was thinking about panel radiators where the internal welds have failed. Without the front and back held together, they blow up like a balloon when pressurised.

They are steel column radiators. When we moved them they snake and bend as they are so heavy.

So quite a large volume then. And probably a significant residual air volume after bleeding.

If you have your bypass loop fully open and both rads turned on, does the problem still occur ?

 

[hosegood]

Yes it still happens
The only time it does not is when the smaller rad is off

 

[SimonH2]

Had another thought ...

Would I be correct in thinking that :

If you turn off the lockshield on the small rad but leave the control valve open - the problem is still there
If you close the control valve and open the lockshield - the problem isn't present
Assuming of course that the control valve is on the flow, and lockshield on the return.

Also, am I correct in thinking these are those "cast iron" segment type rads, which I assume are held together by long bolts/pipes threaded down the top and bottom ? I'm just wondering if the top tube is blocked, and so when you bleed it, it still contains a lot of air.

 

[hosegood]

Hi
I will try those combinations tomorrow, I have only been turning off the control valve when testing.
The rad is like a cast iron one so it could be blocked, although it gets hot all along the top, would it get hot if it had air in it?
I will post results tomorrow.
Thanks

 

[hosegood]

I had already raised the vent pipes as your suggestion which works fine, so no hot water in the F&E tank. But it is still sucking and returning via the feed pipe.
I tested all the combinations you suggested and sure enough it still happens (small rad) when the lockshield is closed and the control valve is open and not the other way round. That surely means air in the rad?
I was only testing it with the control valve before.
Does that make sense to you? If it does then thats really good news and thanks a million
Laurie

 

[SimonH2]

I tested all the combinations you suggested and sure enough it still happens (small rad) when the lockshield is closed and the control valve is open and not the other way round. That surely means air in the rad?

That certainly makes sense to me.

I can only guess at the internal construction of these rads, but I assume there's a perforated pipe runs along top and bottom which both keeps the segments together (by tension in the pipe) and keeps them in line (by the fit of the pipe in the holes in the cast segments). Firstly, there will always be some air in the top of each segment - that's just basic since the vent isn't right at the top of the space. Secondly, I could imagine it's possible for corrosion to start blocking the holes in the pipes and thus impede the ability to get the air out.
If (and that's a big if based on guesswork) that is the problem, then I'd imagine the only real answer would be to strip down and clean the radiator.

I wonder if there's anyone on here who's had one of these apart and could shed any light ?

 

[Andygasman2010]

Just to give my pennies worth... That whole store is a neutral point and the pump shouldn't have any bearing on the vent and feed. I think!

 

[SimonH2]

Wake up at the back
That was discussed early on in the thread.

What appears to be happening is that when the pump starts up, it pressurises air trapped in one of the rads and so lowers the level in the F&E tank. When the pump stops, the air expands back again, and the resulting rebound was pushing water over the top of the vent as well as back up the feed pipe.

The OP has raised the vent so the water no longer goes that way, but he's still shifting water in and out of the F&E tank every time the pump starts/stops.

Since he's already bled the rad, the consideration now is whether that's down to the design (I think it's one of the segmented cast iron jobs) or to internal blockage.

 

[Andygasman2010]

Ha oh yea sorry! Sounds like an interesting one. Also sounds like it's being over thought o me though, the level in the F&E tank will rise and fall with heat expansion anyway. I wouldn't worry about it. Any open system will pull a little oxygenated water in at some point, that's why cold feed/vent connections always block up.

on a side note... I always find that older boilers with F&E tappings straight onto the heat exchanger are much less prone to blocking because they are at a neutral point and aren't subject to pump pressure pulling in oxygenated water.

hope this helps!
andy