Multiple differential bypass valves | Water Underfloor Heating Advice | Plumbers Forums
  • Welcome to PlumbersTalk.net

    Welcome to Plumbers' Talk | The new domain for UKPF / Plumbers Forums. Login with your existing details they should all work fine. Please checkout the PT Updates Forum

Welcome to the forum. Although you can post in any forum, the USA forum is here in case of local regs or laws

American Visitor?

Hey friend, we're detecting that you're an American visitor and want to thank you for coming to PlumbersTalk.net - Here is a link to the American Plumbing Forum. Though if you post in any other forum from your computer / phone it'll be marked with a little american flag so that other users can help from your neck of the woods. We hope this helps. And thanks once again.

Discuss Multiple differential bypass valves in the Water Underfloor Heating Advice area at Plumbers Forums

Messages
13
Is it ok to have more than one bypass valve on a system? Planning an UFH system (heat source supplied by air-water heat pump) with 2 manifolds, each with 6 circuits. Each manifold has its own motorized isolation valve connected to a room thermostat controller. Obviously the system needs a bypass in case both motorized valves are closed...however the heat pump specifies a minimum system pipe volume of 20 litres when all valves are closed...I've calculated each manifold leg has around 10 litres, so if there is only one bypass on one manifold it wouldn't be enough (since water would only flow through the one manifold with the bypass)... Any issue with installing a bypass on each manifold?? So two bypass valves? That way when all isolation valves close the system still has 20Ltrs
 
Any issue with installing a bypass on each manifold?
If I've understood you correctly you are proposing two independent 10 litre bypass loops in parallel.

This is not the same as a single 20 litre bypass loop because you can't be sure that the two bypass valves will always open simultaneously. Actually, you can be fairly sure they won't.
 
If I've understood you correctly you are proposing two independent 10 litre bypass loops in parallel.

This is not the same as a single 20 litre bypass loop because you can't be sure that the two bypass valves will always open simultaneously. Actually, you can be fairly sure they won't.
Yes you understood correctly. So this won't work then. Worth a try, I'll have to think of something else.
 
What if he uses a fixed flow bypass (someone on this forum suggested a Caleffi model) and isolated the manifolds with a 3 port valve that diverts to the bypass rather than the usual 2 port valve? To be honest, if there is a system pump with constant-ish head, a carefully adjusted lockshield would probably do, but then you need a meter to set it up! That would ensure he gets the 10 lpm per circuit he needs without too much expense.
 
What if he uses a fixed flow bypass (someone on this forum suggested a Caleffi model) and isolated the manifolds with a 3 port valve that diverts to the bypass rather than the usual 2 port valve? To be honest, if there is a system pump with constant-ish head, a carefully adjusted lockshield would probably do, but then you need a meter to set it up! That would ensure he gets the 10 lpm per circuit he needs without too much expense.
This doesn't seem relevant to the OPs problem, which is that his heat pump needs to be connected to a heat store with the heat capacity of at least 20 litres of water. In a big system you might get this from the water in the bypass circuit, but IIRC 20 litres is about 40 meters of 28mm copper tube. 20 litres is also the 'minimum', not necessarily the 'optimum', so bigger is probably better.

As @ShaunCorbs says, this is a job for a buffer tank.

People have got used to the relatively good controllability of natural gas boilers which can happily modulate down to a few kW and cycle six times an hour with a few meters of bypass loop for a buffer. With the exception of electric boilers, I think that most other systems require a significantly larger buffer store of some type.
 
Last edited:
This doesn't seem relevant to the OPs problem, which is that his heat pump needs to be connected to a heat store with the heat capacity of at least 20 litres of water. In a big system you might get this from the water in the bypass circuit, but IIRC 20 litres is about 40 meters of 28mm copper tube. 20 litres is also the 'minimum', not necessarily the 'optimum', so bigger is probably better.

As @ShaunCorbs says, this is a job for a buffer tank.

People have got used to the relatively good controllability of natural gas boilers which can happily modulate down to a few kW and cycle six times an hour with a few meters of bypass loop for a buffer. With the exception of electric boilers, I think that most other systems require a significantly larger buffer store of some type.
Oh I see. You mean he needs 20 lpm with an xx temperature drop rather than just a set flow?
 
Oh I see. You mean he needs 20 lpm with an xx temperature drop rather than just a set flow?
Here is how the system will be (albeit I have 2 manifolds, not shown on this pic). My issue was that with M1 & M2 closed, all flow is through the differential bypass valve (C), but in that case the volume (excluding the plumbing in the actual heat pump) must be a minimum of 20 ltrs. But doing the calcs the runs are so short that the pipe volume comes out to be around 10 ltrs.

1631793674582.png
 
Last edited:
BUT, I have to admit that this morning I realised that I've made a mistake...I forgot to include the volume of the manifolds, but worse still I forgot that the main feed and return pipes are 1"1/4...I did the calcs based on 1/2" PEX (which is the circuit pipe for the UFH) but obviously the manifold feeds are much larger! Doh. So taking that into account I have 30 ltrs ish, so I'm fine. But buffer tank is useful to know, might help out someone else in future.
 
Oh I see. You mean he needs 20 lpm with an xx temperature drop rather than just a set flow?
Nor really. The OP's question related to the volume, i.e. heat capacity (J / °C) of the store, not how rapidly energy must be transferred into it (J / s = W) which will be another constraint on the system design.
 
Nor really. The OP's question related to the volume, i.e. heat capacity (J / °C) of the store, not how rapidly energy must be transferred into it (J / s = W) which will be another constraint on the system design.
Ah, yes I missed that bit about volume. Now I get it.
 

Similar plumbing topics

Yes - the recommendation above is accurate for...
Replies
3
Views
2K
Behind the control panel the cylinder unit...
Replies
7
Views
1K
J
We’ve had a UFH system installed as part of...
Replies
0
Views
2K
JSlippy
J
Anyway, you can always increase the boiler...
Replies
16
Views
1K
Good point. My daughter rang me yesterday and...
Replies
20
Views
4K
Back
Top