Things that go bump in the night

[Guest Alphonsox]

The UFH seems to be working well in the new build with the minor drawback that we get serious water hammer issues as the room thermostat closes the manifold shut off valve. As the UFH heating is on E7 this usually occurs at some point in the early hours of the morning and wakes up everyone in the house. I have been informed that this is no longer acceptable .

I notice various companies selling water hammer arresters. - I do have a spare location on the UFH manifold that would allow me to fit one close to the valve.

Are these the right thing to use or are there better solutions ? Would I do better to just turn off the pumps while the valve closes ?

[dpmiller]

you're not using the microswitch atop the zone valve to control the pump?

[Guest Alphonsox]

5 minutes ago, dpmiller said:

you're not using the microswitch atop the zone valve to control the pump?

 

There are no zone valves - The whole UFH runs as a single zone with individual loop flow set by the mechanical flow meters on the manifold.

[dpmiller]

the thermal shutoff valve, in your diagram. And does the fan coil unit have a shutoff valve inside it?

[Guest Alphonsox]

The shut-off valve is currently a two wire version so no provision to switch the pump(s). I could replace this with a 4-wire version BUT I am currently running the UFH pump 24/7 to provide some redistribution of heat around the slab (no idea whether this is really required). The radiator has its own control system with timer and thermostat that is self contained. It expects a hot flow it can tap into whenever required and will not active if there is no hot water. The cylinder pump is therefore on most of the day as well (It turns off from 8am -4pm on the grounds that the area of the house it heats is largely unused in this time period)

[Jeremy Harris]

One option might be to fit an adjustable pressure bypass across the unused ports on the manifold.  I have one fitted across our ASHP flow and return and with a bit of tweaking they can be set to just open when the pressure increases above the set limit.  With luck it could be an easy fix.

[Guest Alphonsox]

Thanks - I'll take a look, although I dimly remember @Nickfromwales saying he hadn't had much luck with them for hammer suppression.

[Jeremy Harris]

The alternative may be to fit a very small pressure vessel close to the valve, to absorb the shock as it finally snaps shut, or simpler still the bodge that I fitted to our old washing machine outlet, a tee connected to a 500mm long blanked off flexi hose.  The dead end of the flexi, right next to the washing machine connection, stopped the pipes in the loft going bang every time the washing machine valve closed.

Edited December 29, 2017 by JSHarris

[Nickfromwales]

@Alphonsox

Is your schematic detail exactly as per your system? As in, you have your manifold flow and return on the left and then another return off the right lower connection ? There shouldn't be a third connection in any instance other than fill / drain. 

What type of motorised valve is it ? A gyneric one?

Also, is the UFH tmv teed in after the uvc ( buffer ) pump as per the drawing ? If so the additional pump velocity isn't doing you any favours here as the zone valve has both pumps fighting against it when it's trying to close. It should be teed off immediately from the buffer, or better still the uvc could have been ordered with a secondary ( HRC ) hot return tapping 2/3 of the way down the tank which would have stopped this issue completely. 

Can you confirm point 1 above to the letter please ! Ta. 

 

Another observation, your expansion vessel should be teed off the colder side of the buffer, eg where the return tapping is. At the moment you've got 2 pump isolation valves which could be closed inadvertently whilst the immersion is still on, say for the air rad to be serviced, plus I assume the air rad will also have localised isolation valves, so again the possibility of totally isolating the expansion from the buffer which is a big no-no in a sealed and pressurised system. 

Does the buffer / uvc have a Temp & PRV connected and terminated accordingly?

 

17 hours ago, Alphonsox said:

 

There are no zone valves - The whole UFH runs as a single zone with individual loop flow set by the mechanical flow meters on the manifold.

I believe @dpmiller is referring to the motorised valve that is causing the issue, not manifold actuators. . Having a 5-wire 2-port ZV would stop this problem normally as they are designed so the pump only spins when the valve is fully open. General rule is :-

• thernostat opens / closes the ZV's.
• ZV controls the pump and heating appliance. 
• micro switch ( across orange and grey wires ) turns the pump on, but only gets pressed closed when the actuator arm inside the ZV reaches 100% of it travel, eg  to the fully open position. Similarly, whilst closing, at 99% open the micro switch breaks again ( and shuts the pump off ) whilst it continues slowly to the fully closed position.
• that micro switch usually controls call for heat ( demand ) for the heat source too, where practical. 

[Guest Alphonsox]

47 minutes ago, Nickfromwales said:

@Alphonsox

Is your schematic detail exactly as per your system? As in, you have your manifold flow and return on the left and then another return off the right lower connection ? There shouldn't be a third connection in any instance other than fill / drain. 

What type of motorised valve is it ? A gyneric one?

Also, is the UFH tmv teed in after the uvc ( buffer ) pump as per the drawing ? If so the additional pump velocity isn't doing you any favours here as the zone valve has both pumps fighting against it when it's trying to close. It should be teed off immediately from the buffer, or better still the uvc could have been ordered with a secondary ( HRC ) hot return tapping 2/3 of the way down the tank which would have stopped this issue completely. 

Can you confirm point 1 above to the letter please ! Ta. 

 

 

Thanks for the replies Nick,

The schematic is drawn from photos but I believe it to be correct. The manifold set up is of the older Wunda type with a simple 2port radiator type TMV (same as @JSHarris). The return flow is definitely off the right lower manifold connection. The shut-off valve is a 240V generic wax type. The radiator is the other end of the building hence the UVC pump requirement, our plumber decided that feeding the UHF from after the UVC pump would also compensate for the piping between the UVC and the manifold.

 

Dodgy Pic follows

 

 

[Guest Alphonsox]

1 hour ago, Nickfromwales said:

Another observation, your expansion vessel should be teed off the colder side of the buffer, eg where the return tapping is. At the moment you've got 2 pump isolation valves which could be closed inadvertently whilst the immersion is still on, say for the air rad to be serviced, plus I assume the air rad will also have localised isolation valves, so again the possibility of totally isolating the expansion from the buffer which is a big no-no in a sealed and pressurised system. 

Does the buffer / uvc have a Temp & PRV connected and terminated accordingly?

 

The expansion vessel is definitely connected on the hot side, so yes there does appear to be the possibility of isolating the expansion vessel from the UVC. There is however a separate PRV and tundish on the UVC itself.

[Nickfromwales]

12 minutes ago, Alphonsox said:

 

The expansion vessel is definitely connected on the hot side, so yes there does appear to be the possibility of isolating the expansion vessel from the UVC. There is however a separate PRV and tundish on the UVC itself.

Was it 'signed off' ?

[Guest Alphonsox]

13 minutes ago, Nickfromwales said:

Was it 'signed off' ?

 

Northern Ireland requires that the installer is registered for UVC installation (he was), nothing more.

[Nickfromwales]

Well 'technically' it's a buffer tank in this instance, but still a water heater over 15L so G3 does kick in. I think your plumber has chosen the easier route, rather than the correct one. 

Connecting the much higher operating pressure T&PRV wasn't really necessary TBH as the EV would likely fail before they got to operate, but obviously the primary ( additinal equipment ) 3bar PRV would operate to protect the lot anyhoo. Only bit about that I really don't like is where they've been connected into the system. Having isolation valves between the heat source and its means of protection is just not allowed, that's a very basic discipline TBH. That's up to you to live with, but at least the 3bar PRV should be at the heat source. ( You can leave the kit physically where it all is and just run a 15mm pipe between them should you decide to do so  ). As long as you understand that the heater needs to be shut off in the event of the system being part or fully isolated for any reason, then it's your call. 

The UFH being teed off after the primary pump is a strange one, as surely that pump can operate for the air rad when the UFH is satisfied.  

Also, why is there no ZV to stop the air rad being heated when the UFH is calling ? Do they share a thermostat ?

Having two heating circuits flowing at different potentials, trying to share a single pump isn't a good arrangement imo, and the UFH most definitely didn't need the primary pump. Only the air rad needed that and the pump could be on a low setting too then. To properly achieve the connection of 'mixed potential circuits' the cylinder should have been a TS / buffer not an uvc ( so each heating discipline could be connected via dedicated separate tappings ) but I doubt you want to change that now. ?  

I wouldn't fit a bypass, instead id move that tee, and change the ZV to a 5-wire, using the micro switch to turn the manifold pump on and off. I'd fit a second ZV prior to the primary ( then air rad only pump ) and that should see any and all problems put to bed as the big man intended. 

Fitting a bypass is a quick fix but I expect it'll have side effects / other consequences. ?

The only downside I can see with the additinal return run off the manifolds is the possibility of it upsetting the balance between loops, by upsetting the potential of the flow and return feeding them. 

[Nickfromwales]

Oh, and I'd relocate the UFH ZV to the flow pipe before the TMV. 

[Guest Alphonsox]

1 hour ago, Nickfromwales said:

Well 'technically' it's a buffer tank in this instance, but still a water heater over 15L so G3 does kick in. I think your plumber has chosen the easier route, rather than the correct one. 

Connecting the much higher operating pressure T&PRV wasn't really necessary TBH as the EV would likely fail before they got to operate, but obviously the primary ( additinal equipment ) 3bar PRV would operate to protect the lot anyhoo. Only bit about that I really don't like is where they've been connected into the system. Having isolation valves between the heat source and its means of protection is just not allowed, that's a very basic discipline TBH. That's up to you to live with, but at least the 3bar PRV should be at the heat source. ( You can leave the kit physically where it all is and just run a 15mm pipe between them should you decide to do so  ). As long as you understand that the heater needs to be shut off in the event of the system being part or fully isolated for any reason, then it's your call. 

The UFH being teed off after the primary pump is a strange one, as surely that pump can operate for the air rad when the UFH is satisfied.  

Also, why is there no ZV to stop the air rad being heated when the UFH is calling ? Do they share a thermostat ?

Having two heating circuits flowing at different potentials, trying to share a single pump isn't a good arrangement imo, and the UFH most definitely didn't need the primary pump. Only the air rad needed that and the pump could be on a low setting too then. To properly achieve the connection of 'mixed potential circuits' the cylinder should have been a TS / buffer not an uvc ( so each heating discipline could be connected via dedicated separate tappings ) but I doubt you want to change that now. ?  

I wouldn't fit a bypass, instead id move that tee, and change the ZV to a 5-wire, using the micro switch to turn the manifold pump on and off. I'd fit a second ZV prior to the primary ( then air rad only pump ) and that should see any and all problems put to bed as the big man intended. 

Fitting a bypass is a quick fix but I expect it'll have side effects / other consequences. ?

The only downside I can see with the additinal return run off the manifolds is the possibility of it upsetting the balance between loops, by upsetting the potential of the flow and return feeding them. 

 

Thanks again for the detailed replay Nick - Lots to think about.

I fully understand the issue with the EV connection - I'll have a talk with my plumber next time he's in.

The air-rad is entirely self contained with its own internal timer and thermostat. It expects a hot water flow to be available at all times and has no means of calling for heat. This means that the UVC pump needs to be running at any time that the internal air-rad timer is set to be active. As the air-rad contains almost no water (0.6lts for our 1.5Kw version) it is almost  easier to leave the UVC pump running 24/7. The UFH ZV is set to come on only during the E7 period and only if the room temperature is below 18C (once every 2-3 days at the moment). The UVC pump is always running at this time. I can see this may well be over pressurising the manifold so moving the connection point looks like a solution.

I cant see how a thermal store would help, and I'm not sure after Jeremy's experiences I could be convinced to try one

Changing the ZV to a 5 wire and using this to turn off the manifold pump would help stop the hammer issue but would stop me running the pump to recirculate heat/solar gain around the floor. It's currently running 24/7 to allow this. What I really need to do is to kill the UFH pump (only) while the ZV closes, or run it constantly outside of the E7 period and intermittently under control of the ZV during E7.

I'm still unclear as to what you think is strange about the return flow set up on the manifold (but then you've seen many more manifolds than me !). As far as i'm concerned the set up is exactly as per Wunda instructions. What am I missing ?

 

 

 

[dpmiller]

in this scenario surely you wouldn't need the motor valve on the return and just run the UFH pump off the thermostats?

[Guest Alphonsox]

14 minutes ago, dpmiller said:

in this scenario surely you wouldn't need the motor valve on the return and just run the UFH pump off the thermostats?

 

There's effectively two scenarios I'm trying to deal with. Currently this is done as :-

Scenario1 - Slab Heating input - Only required during E7 period when room temperature < 18C. Pump = On, ZV =On

Scenario2 - Heat redistribution - counteract localised heating or cooling of slab due to solar gain or open doors. Pump=On, ZV = Off

 

In Scenario2 I do want the pump running 24/7, I cant rely on the room stat to tell me a particular part of the slab is too hot or too cold.

 

[Nickfromwales]

1 hour ago, dpmiller said:

in this scenario surely you wouldn't need the motor valve on the return and just run the UFH pump off the thermostats?

Nope. The pump for the air rad would cause cross circulation. UFH always needs isolating unless it's the only thing direct off the heat source, and thats only if it's an instant heater. A storage heater ( cylinder ) would still need a ZV to stop convection circulation occurring. 

[Nickfromwales]

@Alphonsox

Point conceded over the return coming off the third connection, just I've never seen that done. If Wunda do it that way then it's ok with me.  

5 hours ago, Alphonsox said:

I cant see how a thermal store would help, and I'm not sure after Jeremy's experiences I could be convinced to try one

Well you'd better go grab a stiff drink, because you've got one lol. Your repurposed UVC is filling the boots of a buffer / T/S, just its not designed to be there as your not using it for a 'mono-function' such as simply providing DHW. Jeremy's was an open pipe type instead of sealed like your cylinder, so suffered much higher losses.  

Where a purposefully chosen ( by design ) buffer / T/S would be better, would be where it had 2x tappings for the air rad flow and return, and then another 2 tappings for the UFH flow and return. As they're always at differing flow potentials, ( static for the air rad and dynamic for the UFH ), the two should never meet, let alone be teed together with a pump thrown in to further upset the balance. 

Changing the position of the UFH T will solve this as far as we can with this equipment. 

As for the constant circulation feature for the UFH loops, moving the UFH ZV as proposed, so it's correctly in the flow not the return, will still allow the UFH loops to circulate 24/7, just connect the room stat for heating to the ZV and connect the manifold pump to a timer, set accordingly ( timed / constant ). Pointless going to a 5-wire, but I'd drop the non generic wax ZV you have and instead fit a typical heating / DHW 2-port ZV  ( like this ) which will open and close without the hammer. 

I still don't get the air rad. If it doesn't call for heat, why does it have a stat ? What does the stat do?

[dpmiller]

Nick, the Smiths/ Dimplex style fancoils have an NO stat which inhibits the fan until the HE is warm. After that point control can be either external or internal.

[Guest Alphonsox]

1 hour ago, dpmiller said:

Nick, the Smiths/ Dimplex style fancoils have an NO stat which inhibits the fan until the HE is warm. After that point control can be either external or internal.

 

The one we have has a built in room stat which switches the fan on when the room temperature drops bellow the set level. In addition there is a 7 day timer that controls when this can happen.

[Nickfromwales]

5 hours ago, Alphonsox said:

 

The one we have has a built in room stat which switches the fan on when the room temperature drops bellow the set level. In addition there is a 7 day timer that controls when this can happen.

Well there's how to control the primary pump then ?

 

Does this need to be on often ? Is the UFH not 'cutting it'? Have you considered slightly raising the input temp to the UFH ? One degree can make a significant change. 

[Nickfromwales]

Does the air rad blow into unheated spaces or the the mvhr?

Sorry if you've answered elsewhere but theses a lot of folk on here now

 

[Guest Alphonsox]

Just now, Nickfromwales said:

Does this need to be on often ? Is the UFH not 'cutting it'? Have you considered slightly raising the input temp to the UFH ? One degree can make a significant change. 

 

The air-rad is in a separate granny annex that doesn't have UFH. Its got thick carpets and is occupied by someone who expects instant-ish heat when she turns up the dial. The air-rad works well in this scenario and warms the room up quickly when required. In the rest of the building the UFH sticks a slug of heat into the slab once every couple of days. This is all that is seems to be required to maintain a comfortable temperature, the slab is working very effectively as a big storage heater.

In fact I'm very happy with the operation of the whole system - apart from the water hammer issue.