Reverse Circulation when HW On

[woodpeckerx]

Trying to sort out a reverse circulation when HW on issue, three tee rule not followed in any way it seems. S plan system, valves are not letting by.

Issue 1 - HW On, upstairs rads get burning out.
15mm return (presume from all upstairs rads except towel rail) is teed into the 22mm return about 1m AFTER the cylinder return.

Added a Reliance 15mm check valve in the return, fixes burning hot rads but valve chatters under some pressures so no good as a permanent fix.

Issue 2 - HW On, bathroom towel rail reverse circulating eventually starts heating the downstairs rads probably due to 3 below.
Rad feed is 15mm teed from the 22mm just below the CH valve, return is plumbed into the END of a reducing tee at the cylinder return tee. Its a 15mm-22mm-22mm tee, towel rad (only rad) in the 15mm, cylinder in the 22mm tee centre and end of tee back to boiler return.
Not sure why this reverse circulates?

Issue 3 - 22mm Return Tee at the boiler itself.
The 22mm return from the upstairs (return from upstairs rad loop, towel rail and cylinder) goes into the boiler return via a 22mm tee just above the boiler. The return from the 22mm downstairs rad loop is connected to the centre of the tee at the boiler, ie it is around 4 meters AFTER the cylinder return.
Eventually reverse circulation via the bathroom towel rail starts to heat the downstairs rads although if the towel rail is closed the downstairs loop doesn't *appear* to reverse circulate despite the tee position at the boiler.

Ideas to fix this (properly), don't think check valves will be the answer due to noise and lifespan of the valves.

Thanks.

 

[fixitflav]

15mm return (presume from all upstairs rads except towel rail) is teed into the 22mm return about 1m AFTER the cylinder return.

That's the problem. The pipework needs re-routing

 

[woodpeckerx]

That's the problem. The pipework needs re-routing

Thanks for replying.

I can re-route the upstairs rad common return and join it to the towel rad return which enters at the reducing T but the towel rad is already also suffering from reverse circulation so would that work?

Also is the return tee at the boiler ok or does that need changing?

 

[fixitflav]

Thanks for replying.

I can re-route the upstairs rad common return and join it to the towel rad return which enters at the reducing T but the towel rad is already also suffering from reverse circulation so would that work?

Also is the return tee at the boiler ok or does that need changing?

Can you do a sketch of the layout, showing relative connections? All the rad returns must be commoned before the HW cylinder return joins.
It's possible the towel rail is in parallel with the HW cylinder, rather than the rads. If so the return from that should be commoned with the cylinder return before teeing into the rads return.

 

[woodpeckerx]

Can you do a sketch of the layout, showing relative connections? All the rad returns must be commoned before the HW cylinder return joins.
It's possible the towel rail is in parallel with the HW cylinder, rather than the rads. If so the return from that should be commoned with the cylinder return before teeing into the rads return.

Quick sketch, tee 1 needs re-routing, for now I have a check valve in the 15mm return.

Towel rad (returns into tee 2) is still getting burning hot when HW on, reverse flowing. This reverse flow eventually starts to make its way to the downstairs rads via the flow as the reverse flow has effectively bypassed the CH valve.

Is tee 3 ok? This breaks the rule of all rad returns being connected together and being before the cylinder return?

 

[fixitflav]

Is tee 3 ok? This breaks the rule of all rad returns being connected together and being before the cylinder return?

I tried to add flow arrows to your pic but couldn't get it to copy so I could post it. But I'm sure you can work it out, the cylinder return flow splits at Tee 1, part of it going through the upstairs rads the wrong way, then through the downstairs rads the right way. The flows recombine at Tee 3.

Tee 3 is OK, just needs the cylinder return connection changing to downstream (to the left) of Tee 1.

 

[woodpeckerx]

I tried to add flow arrows to your pic but couldn't get it to copy so I could post it. But I'm sure you can work it out, the cylinder return flow splits at Tee 1, part of it going through the upstairs rads the wrong way, then through the downstairs rads the right way. The flows recombine at Tee 3.

Tee 3 is OK, just needs the cylinder return connection changing to downstream (to the left) of Tee 1.

Thanks, I see what you are saying.

This might be a silly question but would this have the same effect; Moving the 15mm rad return from tee 1 and teeing into the 15mm return from the towel rad, ie pushing the 15mm rad return upstream instead of moving the cylinder return downstream?

I was going to do it that way, but when I found the towel rad was reverse flowing decided it might not work?

With the check valve in place on the 15mm return at tee 1, what would cause the reverse flow through the towel rad?

 

[Daveydub]

On hw only, most return water will take the direct route from tee 2 to tee 1 but some will take the more complicated journey, backwards through the towel rail and forwards through every rad and check valve in the house. Probably not noticeable at rads but turn them all off bar one and you will notice. Moving the towel rail flow to the primary flow would give you all year round towel rail and stop this problem. Will deal with the other problem separately or it gets a bit much.

 

[Daveydub]

Or fit a check valve to towel rail circuit and towel rail will work only with ch

 

[fixitflav]

Looking at it again, I made a mistake in earlier post. Also need to move the downstairs rads return to before the cylinder return tees in.

This might be a silly question but would this have the same effect; Moving the 15mm rad return from tee 1 and teeing into the 15mm return from the towel rad, ie pushing the 15mm rad return upstream instead of moving the cylinder return downstream?

Yes, as long as all the rads returns are commoned before the cylinder return.

With the check valve in place on the 15mm return at tee 1, what would cause the reverse flow through the towel rad?

All I can think of is the check valve not seating or fitted wrong way round! I suppose the towel rad gets hot when CH called?

 

[woodpeckerx]

All I can think of is the check valve not seating or fitted wrong way round! I suppose the towel rad gets hot when CH called?

Valve is correct way around, no detectable reverse flow through it.

Towel rad works fine when CH on, feed side hot first. With HW on gets hot with reverse circulation, return hot first.

Is it something to do with the towel rad loop being a very small loop with only that one rad on it?

 

[fixitflav]

Valve is correct way around, no detectable reverse flow through it.

Not with you. I thought the towel rad is getting hot when HW only called. That's the only way you could detect reverse flow, or lack of.

 

[woodpeckerx]

Not with you. I thought the towel rad is getting hot when HW only called. That's the only way you could detect reverse flow, or lack of.

The towel rad gets hot with HW on but with reverse flow, return side gets hot first. It also gets hot when CH on with correct flow, feed side hot first.

This reverse flow ends up making its way to the downstairs rads eventually.

The check valve is in the return from the upstairs rads (these now don't reverse flow but valve is noisy under some pressures).

The towel rad is not on the upstairs loop, its a single rad on a short loop quite close to the cylinder (right of my diagram).

Is there any reason why this towel rad would reverse flow when HW is on?

 

[fixitflav]

The towel rad gets hot with HW on but with reverse flow, return side gets hot first. It also gets hot when CH on with correct flow, feed side hot first.

The check valve is in the return from the upstairs rads (these now don't reverse flow but valve is noisy under some pressures).

The towel rad is not on the upstairs loop, its a single rad on a short loop quite close to the cylinder (right of my diagram).

OK, I thought the check valve was in the towel rad circuit.

Is there any reason why this towel rad would reverse flow when HW is on?

Yes, the cylinder return flow can split at Tee 2, part going through the towel rad the wrong way. Repiping as suggested should cure it

 

[woodpeckerx]

OK, I thought the check valve was in the towel rad circuit.

Yes, the cylinder return flow can split at Tee 2, part going through the towel rad the wrong way. Repiping as suggested should cure it

I will re-pipe, can I do this by moving the upstairs return (with valve removed) to the right of tee 2 and tee the return with the towel rad, or do I need to move the cylinder return down. Does either way work?