Heatmiser, Hive and Willis....Help!

[Antonb182]

OK, so i'm adding wet UFH to our outbuilding, slab and pipes are in, manifold is set up and i'm now at the stage of getting it up and running, but I have some questions, which while not insurmountable, I'd definitely benefit from some more sensible/learned opinions on.

 

First one is the Willis. I'm planning on using it to heat two relatively small spaces (approx 25sqm each)  and i'd be super surprised if they ever both needed heating at the same time, so I think i'm going to be OK with a single Willis. However, what i've not figured out, is how people are switching them on/off depending on the system needs. As far as I understand, there's no switched live on the unit, so I assume i'd need a relay/switch of sorts between the (always on) spur and the unit. Is this what people generally do? If not, how are you guys controlling the on off of the Willis?  Not sure if it makes a difference, but the whole thing is going to be controlled via the heatmiser UH4. 

 

Secondly, because I already have them, I was planning on using Hive to control the two rooms in two separate zones. Which in my mind, makes perfect sense, one receiver and one thermostat for each zone. Has anyone else done this already and how did it go? I still have the possibility to switch to something more easily integrated, but i'd definitely rather use something I already have. 

[JohnMo]

To me you don't want Willis on without pump on. So drive a suitably rated multi pin relay from pump switch in wiring centre. Drive and Willis from relay. How you control zones doesn't matter after that. Your flow temperature will be by inbuilt thermostat, but add overheat protection with a pipe stat downstream of Willis interlocked to pump Willis power. So mains to pipe stat, then relay, input. So if pipe too hot Willis is off.

[Antonb182]

I hadn't thought about runaway heat from the Willis, but will definitely consider that too . I had in mind to use something like a Salus 16a relay (Here) (or a more robust Drayton heat switch) and run the call for heat from the wiring centre directly into it, which I guess should then control the on off in line with the rest of the vales/pumps being activated. 

 

If anyone has this setup already in place, I'd be super interested to actually see a pic of how you're already running. 

Edited Friday at 12:49 by Antonb182

[Nickfromwales]

13 minutes ago, Antonb182 said:

I hadn't thought about runaway heat from the Willis, but will definitely consider that too . I had in mind to use something like a Salus 16a relay (Here) and run the call for heat from the wiring centre directly into it, which I guess should then control the on off in line with the rest of the vales/pumps being activated. 

 

If anyone has this setup already in place, I'd be super interested to actually see a pic of how you're already running. 

@TerryE has this and iirc is using SSR’s to power the Willis.

 

Ive installed Willis heaters on a number of jobs, and I chose to create a recirculating loop which I then glean flow and return from (via some close - coupled T pieces) so this acts like a figure of 8.

 

That way the main pump can circulate freely, regardless of demand, which makes the system less erratic and the UFH manifold pump is hydraulically decoupled and can just do its own thing.

[JohnMo]

7 minutes ago, Antonb182 said:

Salus 16a relay

I used something similar on a 2kW heater it has a melt down after about about a month. You need a relay rated more like 20A, to put up with the switching loads.

 

Build or buy something like this, this is what Grant Boilers use on heat pumps to switch immersions. Choose this link because it shows the inside of box.

https://ebay.us/m/iXxCKs

 

[JohnMo]

1 minute ago, Nickfromwales said:

@TerryE has this and iirc is using SSR’s to power the Willis.

 

Ive installed Willis heaters on a number of jobs, and I chose to create a recirculating loop which I then glean flow and return from (via some close - coupled T pieces) so this acts like a figure of 8.

 

That way the main pump can circulate freely, regardless of demand, which makes the system less erratic and the UFH manifold pump is hydraulically decoupled and can just do its own thing.

Do you use a second pump (on the heater circuit), or does the system rely on buoyant flow across the heating element - similar to the original design for cylinder heating?

[Nickfromwales]

2 hours ago, JohnMo said:

Do you use a second pump (on the heater circuit), or does the system rely on buoyant flow across the heating element - similar to the original design for cylinder heating?

Second (primary) pump.

 

Willis heaters do hot water by convection, when in a hot water cylinder, but for a heating arrangement where there are variables I employ the primary pump. This is to ensure flow across the heater is sufficient to avoid early kettling or any kettling at all.

[SimonD]

7 hours ago, JohnMo said:

Build or buy something like this, this is what Grant Boilers use on heat pumps to switch immersions. Choose this link because it shows the inside of box.

 

If you get something like this, get a bigger box! The spacer on this one has to be trimmed down as it won't fit when you have your stuffing glands and cabling installed and the mcb & relay shouldn't be installed next to each other due to heat! It's a very tight fit with all cables installed.

[Mike]

For the second winter in a row, my Wilis UFCH is operating manually, but by next winter it will be more sophisticated.

 

A Shelly Pro 4PM will be capable of scheduling the days and times to switch on and off & providing remote control. Upstream of the Shelly there will eventually be a Raspberry Pi which will calculate the heat required, monitor temperatures and command the Shelly (which will become a fall-back in case the Pi is out of action). Either way, the Shelly will send control signals to a pair of 25A contactors (one for each* Willis heater) and will directly control the pump (since that draws little current).

 

A Hager 60060 load-shedder will temporarily cut the contactor control signals if the mains supply is under heavy load, but in your case, @Antonb182, that would be the place to put a thermostat. However, note that with UFCH isn't as reactive as radiators - there's a significant time lag between heating the floor and the air reaching the set temperature that you'd need to take into account to avoid overheating.

 

Schema + sketch below. All the this (excluding the Pi) will take up one row of my consumer unit. D4.4 & D4.5 in the schema = D21 & S1 in the sketch; not got round to updating that...

 

                                 

 

*Although I could heat my place with 1 Willis heater, I've chosen to install 2. That's partly to have the option to use both to load the floor with heat during off-peak hours, but mainly so that I can switch to the second if the first should fail.

 

Edited Saturday at 00:17 by Mike

[Antonb182]

@JohnMo Thanks for the heads up on the small relay. Saves me the pain of trying to figure out what i'd done wrong at a later date. Have ordered from the link you sent as this looks like an ideal solution for me. 

 

@Mike This is actually a really interesting setup and I think that with the bits I've got in place, I now have something that operates in a very similar fashion. 

 

The issue I now find myself with is that the Willis runs hot and cool, intermittently. I've matched it's output temp to the mix temp on the manifold mixing valve (I'm aware I shouldn't really need one if the input is already the correct temp), but the on off nature of this is preventing the room from getting above about 18-19 degrees. It's a SIPS build and well sealed, So heat loss from the room isn't the issue but even after running it for over 24 hours, the return temp is well down compared to the input, but that damned Willis refuses to stay on. Manifold says i'm running at approx 3L/min. 

[JohnMo]

22 minutes ago, Antonb182 said:

The issue I now find myself with is that the Willis runs hot and cool, intermittently

Your system is reacting the same way as an oversized boiler.

 

The reason it does that, is similar to a kettle, you put a cup of water in it it boils in seconds, fill it it takes an age to heat up. It sounds like you simply don't have enough water volume engaged. Then you have a mixer, this will always mix some of the return water with incoming water, and it doesn't make any different temperatures you set them at they always bypass some return water back to incoming water. 

 

If your loops are doing 3l/min, your flow through the Willis is nearer 2l/min. Which is a trickle.

 

The other thing is target flow temp goes out the window with cycling as the target is based on average flow temp. Example, target of 30 degs, is fine if running heat input for hours at a time, but if you are cycling every couple of mins, it's more like 20 then 30, so now your average is nearer 25. So you have to set the target higher to compensate.

 

I would do this first - keep all loops open or closed so run a single circuit. So one control thermostat connected to all loops.

 

And - Crank up the loop flow rate to get more water moved about. Flow each loop at max flow, if making any noise trim them back a little.

 

Then if it's still not ok

Delete the mixer, add a pipe stat, for floor protection, to the hot pipe going into manifold.

 

Then finally, add a buffer or volumiser to system. 

 

 

 

 

[Mike]

3 hours ago, Antonb182 said:

I've matched it's output temp to the mix temp on the manifold mixing valve

I have a mixing valve too, but only to prevent high temperature water entering the underfloor pipes in case someone mistakenly sets the Willis thermostat too high. It's therefore set to 50°C, not the target flow temperature. But it will still mix the water temperature down a few °C as mentioned by @JohnMo.

 

IR image below. With the Willis thermostat set to about 30°C the water coming out of the Willis at 45°C-ish and entering the floor is about 35°C. Other temperatures are misleading as some of the pipes are insulated.

 

[Nickfromwales]

The stats probably read at the lower cooler section of the heater, so you’re correct in that you see the output temp being much higher than the displayed set temp. 

 

What you need to do is dial these back in live play, to get the output temp that you require, using the stat as a ‘volume control’ vs using the markings denoted on it.

 

I’d still say to add a primary pump and tee off a recirculating loop, maybe via a 25L or 50L volumiser, for the best long term functionality. Should see much fewer on / off switching events and max out longevity of the controls.

 

I’d defo keep the TMV and manifold pump as is, for ultimate floor temp control. You’ve already paid for and installed them anyways. 

[Nickfromwales]

Volumiser on flow, not return btw. 

[Nickfromwales]

Used these previously, from the same supplier who’s based in Poland iirc. 
 

Low loss header for a bit of volume, but primarily for hydraulic separation. 
 

https://culmstoreseltermltd.co.uk/en_GB/p/Low-Loss-Header-40kW%2C-1-insulated/221

 

Another, more compact solution, but the 25L version would give better results methinks. 50L better again if you’ve got room. 
 

For just the other £100 or so I’d go for this 

 

https://culmstoreseltermltd.co.uk/en_GB/p/Buffer-Cylinder-Volumiser-VOLANO-TERMICO-PDC-25l-/357

 

These are mild steel so you need to be well dosed with inhibitor. 

 

[Mike]

1 hour ago, Nickfromwales said:

The stats probably read at the lower cooler section of the heater, so you’re correct in that you see the output temp being much higher than the displayed set temp. 

Yes, from memory I think my Elemex model has a 7" stat too, so temperatures in the top half will have no impact. It will switch off either if the temperature builds up lower down (the hot water isn't being pumped away quickly enough), or when the incoming water reaches the stat temperature.

[Nickfromwales]

1 hour ago, Mike said:

Yes, from memory I think my Elemex model has a 7" stat too, so temperatures in the top half will have no impact. It will switch off either if the temperature builds up lower down (the hot water isn't being pumped away quickly enough), or when the incoming water reaches the stat temperature.

Yup, aka crude! At best. 

[TerryE]

Thanks for poking me on this BTW @Nickfromwales, and I hope that you and the family have a good Christmas.

 

Mike,  Do a deep search on my posts on how I heat my slap etc., as I've covered this in depth elsewhere.  A few comments / adjustments:

 

As I described in my topic Microcontroller based power switching revisited in 2024, this showed my thinking and eventual rework of my UFH control and switching revised design.  This was very much a discussion leading to final solution over the course of this long topic, so read to the end.  But the key points were:

• I moved from SSDs to DIN mounted CU-compatible 20A 240VAC contactors for switching all power circuits, eg Willis, pump,  DHW×2.  These have 24VDC contactor coils.
• I use a generic COTS relay module running Tasmota firmware which allows me to use MQTT to drive all controls.  These have dry contact outputs, which I prefer safety-wise so you should have flyback diode protection across all contactor coils, again see the topic.
• I use ESPhome firmware ESP modules with remote DS18B20 temperature probes on my Willis, loop out and returns, DHW top and bottom and central hall to aid my sequencing and control.
• I have a DIY NodeRED app doing all of my CH sequencing and control, and use Home Assistant (HA) as my GUI linked via MQTT, although the CH app is independent of HA and can keep the control regime running for days without HA input.
• The system implements various safety interlocks (e.g. the DHW Immersions are powered off once the UVC reaches target temp, the Willis heater is powered off if it gets above 40°C
• The Willis do UFH only, as the UVC is heated by Immersion, so no buffer tanks or low loss headers are needed.
• The heating regimes are optimised for use with Octopus Agile tariff.

I have a low-energy, high "thermal mass" house, so in general I (or at least my CH control app can pre-calculate day-ahead heating  requirements  and then time-shift UFH and DHW heating to the cheapest Agile time-slots in the day to buy the electricity at lowest prices, so last night for example we put about 22 kWh into our slab and UVC at about 3p / kWh so under £1 (and another £2 to charge our EV).

 

As I describe in our blog the Willis OP gets dumped into ~17ton of insulated slab so this heats up slowly over the, say, 6 hr heating period by 5-10°C and the return goes directly into the Willis, so it never goes above 30°C in normal operation.

 

Yes I do have a fairly complex NodeRED app controlling everything, but the electronics and hardware set up is all stack items and cheap as chips.