Ufh manifold, blending valve with ASHP.

[Nickfromwales]

5 minutes ago, Pete said:

Yes it is a low temp install so are you saying it is pointless installing the existing manifold ?

No. This is a Wunda manifold, all stainless.

 

Note the air vent is not on the end after the bore reduces, as is with other cheaper brands.

 

The cheaper ones promote air NOT getting to the air vent eg the air has to go 'downhill' before it gets to be released.

 

I am saying, pointless asking someone to supply a suitable blending set when they don't sell one.  

 

Ask Wunda to replace the manual vents with bottle vents so the air escapes automatically, or just buy them from a 3rd party and fit yourself.

Edited July 18, 2020 by Nickfromwales

[Pete]

2 minutes ago, Nickfromwales said:

No. This is a Wunda manifold, all stainless.

 

Note the air vent is not on the end after the bore reduces, as is with other cheaper brands.

 

The cheaper ones promote air NOT getting to the air vent eg the air has to go 'downhill' before it gets to be released.

 

I am saying, pointless asking someone to supply a suitable blending set when they don't sell one.  

Ok thanks, what options do I have to avoid having to get another manifold please?

[Nickfromwales]

39 minutes ago, Pete said:

Ok thanks, what options do I have to avoid having to get another manifold please?

I think the Ivar set and 3rd party manifolds mate together, but IIRC you have to ditch the red and blue isolation valves due to the conflicting male / female 1" BSP fittings. Order what you need and do a 'dry run' to see what does / doesn't marry up.

In the past, where it's been a 'hybrid' arrangement, I just added standard gate valves before the primary plumbing hits the pump / blending set.

Best to check compatibility and go from there. 

Edited July 18, 2020 by Nickfromwales
typo

[pdf27]

On 12/07/2020 at 14:49, Nickfromwales said:

Yup.

The valve is actuated by the TRV style head and that gets its reference via a capillary wire that is connected to a thermo-probe ( which is inserted into a pocket to detect the flow temp after the pump, eg after return water has been blended back in already ) and the TRV head will only open if the flow temp is less than the setting on the TRV head. The valve is normally fully open, so it promotes great full flow rates when the UFH starts up and asks for heat, and only starts to close as the temp set at the probe meets the temp set on the head.

There is no 'cold' to speak of, just the available return water and whatever temp that is.

Just to make sure I've fully understood this:

• The TRV and pump are purely there as a safety measure in case the heat pump thermostat fails leading to the floor overheating and being damaged. If the TRV is below temperature (including for cooling) it will always be fully open.
• Normal (gas, older ASHP, etc. systems) can't really turn the flow temperature down low enough so the recirculation and TRV is required for comfort.
• If it's a safety thing only on a system where the heat source can modulate to low temperatures, any old TRV will be fine as precision of switching temperature will have no impact on comfort.
• With the valve fully open, 100% of the flow goes into the manifold and around the floor. The ASHP and manifold pumps run in series.
• With the valve fully open, turning on the pump by itself will try to force water through the ASHP (not acceptable). I like what Jeremy has done using the circulation pump to even out temperatures between rooms, but this only works with a buffer vessel rather than direct from the ASHP.

The question I have is if this is a real risk for a small heat pump as opposed to a boiler. The small 3.2 kW Panasonic ones have a 9 l/min circulating pump - that's 23.1 kJ added per litre of water. As 4.2 kJ is required to heat 1 litre by 1°C, that's a temperature increase of 5°C. Getting to temperatures (>55°C) which might damage the floor requires a double fault: water thermostat and pump sensor bringing flow down to 1 l/min,  or for the room and water thermostats to fail and the whole house to be at >50°C.

[Nickfromwales]

On 02/08/2020 at 19:09, pdf27 said:

If it's a safety thing only on a system where the heat source can modulate to low temperatures, any old TRV will be fine as precision of switching temperature will have no impact on comfort.

No. 
It’s a safety thing full stop. Forget modulation, it’s to prevent the max heat delivery ever getting into the slab. 
Not any old TRV, as these have capillary wires coming off them which go to a reference sensor probe ( look at the pics for the curly silver wire ) so must be made fit for purpose. Regular TRV’s reference room temp to actuate, these take reference from the probe.   
 

On 02/08/2020 at 19:09, pdf27 said:

The TRV and pump are purely there as a safety measure

No. 
The manifold pump is there for circulation of the water through the loops. The ASHP pump is there to circulate heat back and forth the manifold > ASHP only and should never be relied upon for UFH flow.

 

On 02/08/2020 at 19:09, pdf27 said:

With the valve fully open, 100% of the flow goes into the manifold and around the floor. The ASHP and manifold pumps run in series.

Yup. Not quite 100% but the majority of it. 
 

On 02/08/2020 at 19:09, pdf27 said:

With the valve fully open, turning on the pump by itself will try to force water through the ASHP (not acceptable). I like what Jeremy has done using the circulation pump to even out temperatures between rooms, but this only works with a buffer vessel rather than direct from the ASHP.

No. 
The flow to the manifold from the ASHP will be arrested by a 2-port zone valve fitted before the manifold in the flow pipe. With that shut you can run the manifold pump to do as per Jeremy did and Bobs your uncle and Fannys your aunt. ? Water will always follow the path of least resistance, so the pump will just happily recirculate through the pump / blending set arrangement. 
 

The answer is, you need a manifold pump and blending set ( TMV ) full stop. Convincing yourself that the risk is minimal is academic as you can still get that small HP over 50oC, regardless of how long it takes to get there. Scenario could be that just one loop is open and flowing vs the load of the entire heated area, so compound failures are what drives these initiatives, guidelines and regulations.

 

Now shoot off and buy your TMV ?!!

[pdf27]

54 minutes ago, Nickfromwales said:

It’s a safety thing full stop. Forget modulation, it’s to prevent the max heat delivery ever getting into the slab. 
Not any old TRV, as these have capillary wires coming off them which go to a reference sensor probe ( look at the pics for the curly silver wire ) so must be made fit for purpose. Regular TRV’s reference room temp to actuate, these take reference from the probe.   

I think I'm more confused than I was earlier. Does this therefore mean that all regular TRVs are unsafe as they won't protect the slab?

 

54 minutes ago, Nickfromwales said:

The manifold pump is there for circulation of the water through the loops. The ASHP pump is there to circulate heat back and forth the manifold > ASHP only and should never be relied upon for UFH flow.

How come? Are they just cheaping out on the ASHP water pumps?

 

54 minutes ago, Nickfromwales said:

The flow to the manifold from the ASHP will be arrested by a 2-port zone valve fitted before the manifold in the flow pipe. With that shut you can run the manifold pump to do as per Jeremy did and Bobs your uncle and Fannys your aunt. ? Water will always follow the path of least resistance, so the pump will just happily recirculate through the pump / blending set arrangement.

Is there an additional bypass valve within the TRV? If not then presumably it would be trying to send 100% of the flow through the ASHP and be blocked by the zone valve.

Edit: at least some of them do which is starting to make more sense.

 

 

54 minutes ago, Nickfromwales said:

he answer is, you need a manifold pump and blending set ( TMV ) full stop. Convincing yourself that the risk is minimal is academic as you can still get that small HP over 50oC, regardless of how long it takes to get there. Scenario could be that just one loop is open and flowing vs the load of the entire heated area, so compound failures are what drives these initiatives, guidelines and regulations.

Sadly, that's kind of my day job - among other things, ensuring double fault cases are treated differently to single fault and we don't try to prevent them unless it's safety critical (as in, everybody dies).

 

54 minutes ago, Nickfromwales said:

No shoot off and buy your TMV ?!!

I'm still short of a house to put it in at the moment .

Edited August 2, 2020 by pdf27

[Nickfromwales]

On 02/08/2020 at 21:19, pdf27 said:

I think I'm more confused than I was earlier. Does this therefore mean that all regular TRVs are unsafe as they won't protect the slab?

I've spotted why  

TRV = Thermostatic Radiator Valve ( big white knob with graduated numbers on it ) typically found on the end of a domestic radiator

TMV / TBV  = Thermostatic Mixing / Blending Valve ( the valve that mixes flow and return ( hot and cool water )) together to reduce the delivered water temp

[Nickfromwales]

On 02/08/2020 at 21:19, pdf27 said:

How come? Are they just cheaping out on the ASHP water pumps?

No. They're serving two separate bodies of water with dissimilar hydraulic resistances. The manifold pump see's high and low resistance as a constant operating environment due to loops opening / closing in varying numbers at varying times. The ASHP circulation pump must see a constant in order to satisfy the mandatory ( factory fitted / supplied in most instances ) flow switch, which tells the heat pump that all is well and it can fire up safe in the knowledge that the flow rate is sufficient to convey the produced heat away effectively.

Edited August 3, 2020 by Nickfromwales

[joth]

On 02/08/2020 at 20:33, Nickfromwales said:

The answer is, you need a manifold pump and blending set ( TMV ) full stop. Convincing yourself that the risk is minimal is academic as you can still get that small HP over 50oC, regardless of how long it takes to get there. Scenario could be that just one loop is open and flowing vs the load of the entire heated area, so compound failures are what drives these initiatives, guidelines and regulations.

 

 

Right, the ASHP we're using (Ecodan 8.5kw) controller has an output to manage an electrically actuated mixing valve, see snippet from the MIs copied below. It has thermistors on the flow and return to the UFH so in priciple can be as good as a thermastic analog valve, failure modes excepted.

 

 

 

This is quite a nice setup, as it means the ASHP can do the right thing to actively manage cooling as well as heating temperature for two emitters running simultaneously at different target temps, and also in principle could close off (set to bypass) the mixer valve when the zone is not being actively driven, which would allow the zone pump to overrun too (i.e. support 24/7 UFH recirculation as discussed above).

 

Now assuming we don't trust this setup to protect our expensive floor finishes, yet would still like to have that motorized mixer valve for the fine control it gives, would it make sense to use both a blending TMV pumpset and the motorized valve? We'd put the TMV on the UFH side of the motorized valve (i.e. integrated with the pump).
For this exception situation only use-case, using a standard (less expensive) TMV from wunda  that only goes down to 30ºC maybe ~OK as this would only close off in error case anyway?

 

(Still can't help it feels OTT.  The alternative safety measure that comes to mind is to put an additional thermistor in the floor screed or on the UFH flow pipe, and have it trip a relay to cut out the pump if it ever exceeds a set point).

 

Bonus question: recommend me an appropriate motorized valve? ? (2L+N connections)

 

Cheers

Edited September 17, 2020 by joth

[Nickfromwales]

12 hours ago, joth said:

Bonus question: recommend me an appropriate motorized valve? ? (2L+N connections)

https://www.jtmplumbing.co.uk/central-heating-controls-valves-c436/esbe-valves-controls-solid-fuel-products-c509/mixing-valve-c511/esbe-valves-actuators-3-point-p19117/s20737?utm_source=google&utm_medium=cpc&utm_term=esbe-valves-esbe-ara652-3-point-spdt-230v-6nm-60s-actuator-esbe-valves-esbe-valves-esbe-ara652-3-point-spdt-230vac-6nm-60s-actuator-12101700&utm_campaign=product%2Blisting%2Bads&cid=GBP&gclid=EAIaIQobChMIha-U0qrw6wIVgc_tCh0Q0gfiEAQYASABEgJx4fD_BwE
 

Liking the above setup too. Going to ask their tech support a few questions to get a proper heads up. 

[joth]

3 hours ago, Nickfromwales said:

16 hours ago, joth said:

Bonus question: recommend me an appropriate motorized valve? ? (2L+N connections)

https://www.jtmplumbing.co.uk/central-heating-controls-valves-c436/esbe-valves-controls-solid-fuel-products-c509/mixing-valve-c511/esbe-valves-actuators-3-point-p19117/s20737?utm_source=google&utm_medium=cpc&utm_term=esbe-valves-esbe-ara652-3-point-spdt-230v-6nm-60s-actuator-esbe-valves-esbe-valves-esbe-ara652-3-point-spdt-230vac-6nm-60s-actuator-12101700&utm_campaign=product%2Blisting%2Bads&cid=GBP&gclid=EAIaIQobChMIha-U0qrw6wIVgc_tCh0Q0gfiEAQYASABEgJx4fD_BwE
 

 

Noice, thank you sir.

Looks like ESBE ARA662 3-Point SPDT 230v 6NM 120S Actuator is the one: mains operated and 2minute open->close run time, which is the "default" and recommended setting in the FTC5 controller. (MI Page 44: "Set the Running time according to the specifications of the actuator of each mixing valve. It is recommended to set the interval to 2 minutes that is a default value. With the interval set longer, it could take longer to warm up a room.")

 

3 hours ago, Nickfromwales said:

Liking the above setup too. Going to ask their tech support a few questions to get a proper heads up. 

? do update us with what you find.

 

The FTC5 looks pretty comprehensive and well thought through. Possibly a brain-ache to setup, but I am hopeful it's going to suit us really well.

 

 

 

[joth]

On 17/09/2020 at 03:33, joth said:

(Still can't help it feels OTT.  The alternative safety measure that comes to mind is to put an additional thermistor in the floor screed or on the UFH flow pipe, and have it trip a relay to cut out the pump if it ever exceeds a set point).

 

FWIW this is exactly the route I ended up taking. I have a 1-wire thermistor in each room with wood (bamboo) finishes, sending data to loxone, and it cuts off the call-for-heat if the temperature gets too high.

It also can dynamically adjust the flow temp, e.g. use a high flow temp when the screed is cold and gradually back it off as the screed warms up towards the limit. I have all the logic for that implemented, but it's pretty flaky controlling the Ecodan FTC6 via their MELcloud Wifi interface, it would be much better done over the modbus interface but unfortunately you can't have both interface modules on a single FTC6 so meh, I just don't bother with this level of flow temperature control feedback. ... if i left the ashp on 24/7 it wouldn't be interesting anyway, but the nature of overnight cheap rate (and solar PV) does make it more worthwhile to back off the temperature then blast it as hot as possible during the cheap rate window

 

[HughF]

Very interesting discussion, I've learnt a lot, all of which can be applied to my system as I plan for UFH in the extension and new, upsized rads everywhere else.

 

Goes against what I've read in JCs book, but makes perfect sense, so I'll go with the blending valve set recommended here...

[sean1933]

Seems to be a running theme here despite the age of this thread.

I have Wunda manifolds installed which are already in situ with the standard mixing valves (and Wilo pumps) - my fault as since install we've ditched gas and gone to ASHP (various reasons).

Our ASHP (Mitsi Ecodan 14kw with FTC-6 controller) install is starting on the 20th Feb. Installer prev said we dont need the mixers, but at the time I didnt dive into why and he was relaxed about leaving them in if I wanted.

I note Pete and Nick, you both say with ASHP we need the lower temp mixers. It might be because I'm tired but I'm struggling to get my head around all the reasons. I trust the advice just dont understand! Aside from the protecting overheating issue for the floor type (surely 35deg is safe enough - we have UFH downstairs in slab and upstairs overboarded, NO RADS), if the ASHP was running in weather compensation mode why do we need mixers that can get to 20deg?) I am minded that I only have a short window to buy different mixers if they are a must, but at the same time I have 4 manifolds (400sqm house) so not a small cost! (ESBE T4's @ £132 is a bit cheaper than the IVAR's thankfully - assuming they are up to the job and will work with the Wilo pumps from Wunda).
   https://underfloorparts.co.uk/product/esbe-t4-water-underfloor-heating-blending-valve/

 

Appreciate I'm going off topic slightly but due to size of the house/number of zones our ASHP installer has specced for a low loss header. Not something I've seen many others mention so interested on if its contentious or should I just trust its justified? With a low loss header do we need additional pumps to push water from the LLH to the post manifold/TRV pumps (required due to TRV's I believe)? I ask because they are showing on the schematic... meaning in total 7 pumps - 1 when ASHP hot runs into the house, 2 after LLH (per floor (2 x 2 manifolds) then the 4x post TRV's @ manifolds. Seems an awful lot to me! Diagram attached, note it doesnt include any of the UFH manifolds as thats outside their contract and 'my problem', and the HWSR shown has been deleted as we're going for a short run manifold setup.

 

Having typed all this out I have now read the entire post and seen Joth's comments, I wonder if I've come full circle and am further confused! Bearing in mind we're using the same ASHP, FTC6 controller (albeit bigger/more loops/LLH etc) and eventually loxone too, would this do away with the needfor mixers entirely? If so, does this also do away with the need for pumps at the manifolds entirely as the post LLH pumps will suffice? HELP!

 

Really appreciate the technical help as always!

Edited February 10, 2023 by sean1933
add image

[PeterW]

Not quite sure why the LLH is where it is if you’ve not got blending valves as it’s there to provide a balancing point - as it stands you are using the manifolds as isolation points by the looks of it ..? Not quite sure why you have a pair of pumps off the LLH - could easily get away with a 28mm circulation loop and single pump between all manifolds from LLH and returning back to LLH with each manifold pulling from it. Not sure who specced or designed the UFH though as you run the risk of pulling cooler water into each floor manifold as the closest / least resistive loops may scavenge the hotter water first with no blending valves. 
 

Could be a candidate for auto balancing actuators but how many loops do you have..??

[Nickfromwales]

Swing-check non return valves after the LLH would kill most issues off, but the low volume of water in it ( <15L typically ) may not solve any short-cycling issues. I'd use a 50L or 100L buffer in place of that.

I'd then move that to front of house, instead of the last component before the UFH manifolds, so the duct heater / cooler also didn't promote short-cycling eg all of the downstream heating applications benefitted from it.

DHW needs to go around the buffer / LLH of course, and not via it.

Not a good idea to have no hydraulic separation between the ASHP and the duct heater / cooler as it looks like it's all set up so the heating loads can run independently.  

[Rich123]

Hi Russel , it depends on the application. The ASHP should have a weather compensation on it , therefore the flow temp will be determined by the ASHP.  Having a blending valve installed won’t reall change anything as long as you keep it set to the maximum temperature of the ASHP.  If you have an ufloor design it may be based on a flow temp of 45 degrees for example . It could be useful as a high limit so I’d install it regardless, it won’t do anything if the temperature from ASHP is lower as it will operate fully open.

[sean1933]

Thanks Gents, insightful as always.

 

I've fed this back into the design guys and he's going to move some things around. Particularly the duct cooler also coming off the LLH/buffer. He didnt appreciate previously that I will want it running at the same time as UFH  - assumed it was for cooling only and that I wouldn't cool through some of my UFH (he's not a fan). Will also be ditching some of the pumps as unnecessary so close to the ground floor manifolds.

[Nickfromwales]

As your cooling will be indiscriminate, you will need to choose the cold flow temp well. Eg the fan coil would like much colder water than the screed / slab could tolerate, but you wont have a means of getting true ( linear ) control over that. 

My current project has a Stiebel Eltron ASHP and all of its ( bloody expensive but bloody good ) controls and mixing stations, so I can choose the cold temp for the slab and a much colder temp for the Brink AHU ( heater / cooler to MVHR ).

Keep an eye on condensation, and super insulate all the joints / valves with pieces of self-adhesive neoprene tape and then a very thick ( 25mm wall ) Armaflex neoprene pipe insulation over every piece of pipe that will be doing cooling. A length of much larger ( 54mm or more ) Armaflex can be used to make jackets for the isolation valves, with a small slot made for the lever handle to still be accessible.  

[SuperJohnG]

Dear lord....I'd totally forgot about this thread. 

 

Wunda manifolds in place have been for nearly two years...I was about to but the pumps etc from wunda. I assume I literally buy an IVAR one and hey presto I've got everything I need? 

[dpmiller]

an IVAR pump? do they make them themselves or just reprice and rebrand?

[SuperJohnG]

15 hours ago, dpmiller said:

an IVAR pump? do they make them themselves or just reprice and rebrand?

No IVAR blending set, it includes a grundfos pump

[ectoplasmosis]

Yet another thread dredge…

 

I’ve just had a 7kW Vaillant Arotherm Plus system installed as part of a Victorian house retrofit, with 3 loops of UFH on the ground floor only (rads everywhere else). Design flow temp: 45degC @ -2degC OAT. 
 

The UFH was specced and installed by a separate plumber to the ASHP installer. They installed a “Reliance HEAT970366” thermostatic blending valve, which says the following in its specs: “Hot temperature supply range: 60°C – 85°C”: https://underfloortradesupplies.co.uk/product/reliance-water-controls-thermoguard-heat970366-ufh-blending-mixer-valve/

 

What does that spec mean? Have I got a blending valve that’s unsuitable for my ASHP install?

[Lofty718]

You shouldn't need a blending valve with UFH and ASHP

 

if you do have a blending valve it should be an electronic blending valve suitable for low temp UFH

 

 

[JohnMo]

6 hours ago, ectoplasmosis said:

Reliance HEAT970366

It's just an adjustable blending valve, you are only altering the amount of return water that gets mixed into supply water. They aren't the best mixer. They generally will always drop the temperature of the supply by around 10 degs or more. So if your max supply is 45 degs you will get somewhere around 35 degs in the floor.

 

Really depends on the difference in design temp of the floor compared to radiators and also the floor buildup.

 

If you need to blend electronic (ESBE) controlled by your Vaillant controller would be best. But at full retail prices - ouch not cheap. A mechanical solution would be an Ivar mixer that will control to exact temperatures and is very adjustable.