Building in provision for a later ASHP.

[epsilonGreedy]

The more I consider ASHP I realize the basic installation is very simple. Am I correct in thinking the complexities are further downstream with buffer tanks and other gadgets needed to raise the luke warm ASHP output to that required for DHW?

 

If I wanted to build in support for later adoption of ASHP and coupling to UFH what is needed? Here is my guess:

1. High capacity electrical circuit of say 5 to 8 amps terminated on an outside wall socket.
2. A pair of water pipes routed from the external wall through to a point close to the UFH manifold.
3. A control wire conduit with either a mouse line or multi-core general purpose control cable. [Edit as per @ProDave]
4. Concrete base slab for ASHP [Edit me]
5. Planning permission in a conservation area [Edit me]

 

Is it really that simple? Where would a timer and the thermostatic control go?

Edited April 16, 2019 by epsilonGreedy

[joe90]

2 minutes ago, epsilonGreedy said:

needed to raise the luke warm ASHP output to that required for DHW?

 

Not the case, I heat my DHW to 47’ (ish) with my ASHP and I have not seen it defrost once since installing it 6 months ago. I have immersions in the tank more as a backup than anything. I think you will find others on here that do the same. I have an insulated direct cylinder and if the DHW is kept to 47’ there is very little heat loss from the tank.

[ProDave]

You are correct, BUT you also need a load of control wires.  This is where it gets complicated. different heat pumps work in different ways.  My first (dud) one just needed a CAN bus pair to connect to an internal control wiring box.  But it's (different make) replacement required all pumps, valves, programmers, thermostats, actuators etc connected direct to the heat pump.  To simplify pulling loads of separate cables, I just used one bit of 10 core control cable and made my own distribution junction box in the plant room.

 

So unless you know exactly what make of heat pump you might fit, for cabling the best you can do is install a decent sized bit of trunking that is accessible, or at least has a draw string included to pull whatever control cables through when you find out what you actually need.

 

In my case I brought all my cables to the plant room.  The controller that came with the HP is in the wall there.  But it is a very non intuitive thing and I wanted the "user interface" to be simpler, so I fitted a conventional 3 channel boiler tine clock which is in the utility room so you set heating and hot water times just like any other heating system that everyone is familliar with, and the HP's own programmer is there just for looking at or adjusting parameters, I don't use any of it's timer functions.

 

As @joe90 says an ASHP can heat HW okay, I have mine set to 48 degrees and that has proved okay.  The immersion heater heats it further with surplus PV on a sunny day and it will easily reach 60 degrees or more from that.,

 

[epsilonGreedy]

2 minutes ago, joe90 said:

Not the case, I heat my DHW to 47’ (ish) with my ASHP and I have not seen it defrost once since installing it 6 months ago

 

 

Ok his is interesting, would the ASHP efficiency (COP) be improved if it was set to run at the ideal temperature for UFH alone?

 

I ask because I have read that in the super insulated @jsharris house his UFH slab runs a smidgen above ambient house temp, whereas in my less than passive new build I assume my slab would need to run a few degrees higher. Which leads to the question what would be the ideal circulation heat from the ASHP to whack 20kw hours in to the slab during the overnight economy 7 period between say 2am and 7am?

[Jeremy Harris]

1 minute ago, epsilonGreedy said:

 

Ok his is interesting, would the ASHP efficiency (COP) be improved if it was set to run at the ideal temperature for UFH alone?

 

I ask because I have read that in the super insulated @jsharris house his UFH slab runs a smidgen above ambient house temp, whereas in my less than passive new build I assume my slab would need to run a few degrees higher. Which leads to the question what would be the ideal circulation heat from the ASHP to whack 20kw hours in to the slab during the overnight economy 7 period between say 2am and 7am?

 

 

COP improves if the ASHP flow temperature is reduced, but there seems to be little improvement below about 35 to 40°C, from what I can tell.  I run ours at 40°C flow temperature.

 

You can calculate the required UFH floor surface temperature for a given heat output to a room from this spreadsheet, together with the heat loss from UFH to the ground beneath the floor: Floor heat loss and UFH calculator.xls

[epsilonGreedy]

48 minutes ago, JSHarris said:

You can calculate the required UFH floor surface temperature for a given heat output to a room from this spreadsheet, together with the heat loss from UFH to the ground beneath the floor: Floor heat loss and UFH calculator.xls

 

 

Hey that is excellent, just ran the numbers for a theoretical version of my house. I nearly doubled the whole house heat loss to 3kw, reduced the slab to 80mm and insulation to 150mm and get a required floor temp of 25.2.

 

The spreadsheet seems to assume a ground bearing slab, would a beam & block suspended floor introduce a large error?

 

48 minutes ago, JSHarris said:

COP improves if the ASHP flow temperature is reduced, but there seems to be little improvement below about 35 to 40°C, from what I can tell.  I run ours at 40°C flow temperature.

 

 

Ok and I imagine that at some lower temperature point there is a cross over where pumping costs exceed the improved COP savings. Just to confirm, heating water enters your UFH slab at 40 degrees?

Edited April 16, 2019 by epsilonGreedy

[Barney12]

Interesting thread as I'm still minded to install an ASHP to bypass my reliance on the Sunamps for UFH. I would then use my Sunamps purely for DHW. 

I got Nick to put two 32mm flow and return pipes and a duct for cabling to an outside point which current sit unused in the plant room. 240v is also provisioned too with 2.5mm SWA. 

 

As others have done my intention is to use the slab as a buffer to keep things simple. The other benefit is the ability to cool the slab too. 

 

The spreadsheet above is interesting which leads to a further question; Is there a way to calculate the lag? I.e. If say the slab was at 18 degrees and you needed to get it to 23 degrees to hit your target room temp how long would it take for the room to warm up? 

 

Ta! 

[epsilonGreedy]

38 minutes ago, ProDave said:

 made my own distribution junction box in the plant room.

 

 

The absence of a plant room in my design is beginning to feel like a problem. I was hoping to box in the basics of the UFH manifold and pump under the stairs of the open plan sitting room as this is a good central point in the house and would avoid a UFH Clapham Junction at the main internal ground floor door threshold.

[Dreadnaught]

55 minutes ago, epsilonGreedy said:

The more I consider ASHP I realize the basic installation is very simple.

 

True but only for a mono bloc ASHP. I believe that a split ASHP would require a heating engineer for gasing. A whole new kettle of fish.

 

(Non-expert here, just throwing in my pedantic tuppence.)

Edited April 16, 2019 by Dreadnaught

[Jeremy Harris]

17 minutes ago, epsilonGreedy said:

The spreadsheet seems to assume a ground baring slab, would a beam & block suspended floor introduce a large error?

 

For a ventilated space under the floor, then the "under slab soil temperature" needs to be changed to be close to the outside air temperature, as it will vary pretty much in line with that, so will be colder in cold weather and warmer in warm weather.

 

17 minutes ago, epsilonGreedy said:

 

Ok and I imagine that at some lower temperature point there is a cross over where pumping costs exceed the improved COP savings. Just to confirm, heating water enters your UFH slab at 40 degrees?

 

 

The losses in the heat pump from either the fan power or the circulating pump are pretty small.  The biggest impact seems to be from defrost cycling, as in cool, damp, weather, with the ASHP being asked to run at a high heat output, the external evaporator tends to ice up, and ASHPs have a de-icing system that often just runs the heat pump in reverse for a short time, to transfer heat from the house back out to the evaporator to melt any ice.  I've found by experiment that if the ASHP flow temperature is kept to 40°C or less, the need for de-icing is dramatically reduced, which improves performance a fair bit.  Icing isn't a problem in cold weather, as the air tends to be quite dry then, it's at it's worst when it's cold and wet.

 

I run the UFH with a lower flow temperature, using a thermostatic valve.  The ASHP charges a 70 litre buffer tank and the UFH can draw from that as needed.  Typically our UFH flow temperature will be around 25 to 26°C.

 

[epsilonGreedy]

On 16/04/2019 at 10:16, Barney12 said:

The spreadsheet above is interesting which leads to a further question; Is there a way to calculate the lag? I.e. If say the slab was at 18 degrees and you needed to get it to 23 degrees to hit your target room temp how long would it take for the room to warm up? 

 

 

@TerryEwrote a blog entry for his house but a with 1kw heating requirement his house sounds like a Passiv++ example.

 

Also worth noting that from other posts @TerryEcalculated that the heat capacity of his plasterboard is about the same as the heat capacity of his floor slab, this might be due to a 3-story design??

 

https://blog.ellisons.org.uk/newbuild/heating-the-slab-an-overview/

Edited April 28, 2019 by epsilonGreedy

[Jeremy Harris]

23 minutes ago, Barney12 said:

The spreadsheet above is interesting which leads to a further question; Is there a way to calculate the lag? I.e. If say the slab was at 18 degrees and you needed to get it to 23 degrees to hit your target room temp how long would it take for the room to warm up? 

 

I think you could have a go at this, but it wouldn't be that easy.  It depends on the heat capacity of the heated floor and the thermal conductivity from the pipes through that.  Concrete is a reasonably good conductor of heat, so if you know the volume (or mass) of the floor, the heat output the room with varying temperatures and the heat losses, it should be possible to work out how long it takes for the floor to warm up.

 

I know that our slab takes several hours to warm up from cold, but as we've been running it at a near constant temperature for a long time now I can't say for sure exactly how long it takes to go from, say, 20°C to 22°C for any given heat input and room temperature.

[Ed Davies]

1 hour ago, epsilonGreedy said:

Ok his is interesting, would the ASHP efficiency (COP) be improved if it was set to run at the ideal temperature for UFH alone?

 

Bear in mind that you can (usually, depending on the heat pump/controller model?) run the ASHP at different temperatures depending on what it's heating. AIUI, the usual thing is to run it in either of two modes: UFH or DHW. When you want to heat the DHW you stop heating the UFH for a while. You can then set the flow temperatures differently for the two modes.

[jack]

56 minutes ago, JSHarris said:

I know that our slab takes several hours to warm up from cold, but as we've been running it at a near constant temperature for a long time now I can't say for sure exactly how long it takes to go from, say, 20°C to 22°C for any given heat input and room temperature.

 

From memory, it took several days to warm up our slab from around 14°C to 20-ish when we returned from several weeks' holiday having left all heating off over a cold, dull and windy Christmas. That was only allowing a 27°C maximum flow temp. It might have worked faster with a high temperature flow, but I didn't want to heat things up too quickly.

[epsilonGreedy]

5 minutes ago, jack said:

That was only allowing a 27°C maximum flow temp. It might have worked faster with a high temperature flow, but I didn't want to heat things up too quickly.

 

 

A concern over thermal expansion stress being induced into your floor slab? 

[jack]

Yes, but also because we already had a couple of cracks in our polished concrete layer and I didn't want to make those worse or add any more.

[Declan52]

2 hours ago, ProDave said:

You are correct, BUT you also need a load of control wires.  This is where it gets complicated. different heat pumps work in different ways.  My first (dud) one just needed a CAN bus pair to connect to an internal control wiring box.  But it's (different make) replacement required all pumps, valves, programmers, thermostats, actuators etc connected direct to the heat pump.  To simplify pulling loads of separate cables, I just used one bit of 10 core control cable and made my own distribution junction box in the plant room.

 

So unless you know exactly what make of heat pump you might fit, for cabling the best you can do is install a decent sized bit of trunking that is accessible, or at least has a draw string included to pull whatever control cables through when you find out what you actually need.

 

In my case I brought all my cables to the plant room.  The controller that came with the HP is in the wall there.  But it is a very non intuitive thing and I wanted the "user interface" to be simpler, so I fitted a conventional 3 channel boiler tine clock which is in the utility room so you set heating and hot water times just like any other heating system that everyone is familliar with, and the HP's own programmer is there just for looking at or adjusting parameters, I don't use any of it's timer functions.

 

As @joe90 says an ASHP can heat HW okay, I have mine set to 48 degrees and that has proved okay.  The immersion heater heats it further with surplus PV on a sunny day and it will easily reach 60 degrees or more from that.,

 

This is the kind of setup I would be looking if I install an ashp in the future. I basically use my pellet boiler in this fashion but use its onboard time programs to just heat my thermal store. There is nothing in my system that can cause the boiler to fire up other than me hitting the button. 

Can you just use the simple controller that comes with them to provide a set time that the ashp works. As in it comes on at 4am to 10 am then comes back on at 4pm to 8pm. These Time periods would be adjusted once I figured out how long it takes to get my tank to 45-48. 

[PeterW]

37 minutes ago, Declan52 said:

These Time periods would be adjusted once I figured out how long it takes to get my tank to 45-48. 

 

You can calculate this pretty easily but from observation currently, I’ve got a 300 litre tank going from 15-45c (bottom stat is about 25% up the tank) and it’s doing that in a little over an hour. 

[Declan52]

18 minutes ago, PeterW said:

 

You can calculate this pretty easily but from observation currently, I’ve got a 300 litre tank going from 15-45c (bottom stat is about 25% up the tank) and it’s doing that in a little over an hour. 

I didn't think it would be as quick as that, pretty impressive.

[joe90]

4 hours ago, epsilonGreedy said:

The absence of a plant room in my design is beginning to feel like a problem. I

 

I don’t have a plant room per se, but a large cupboard (think double wardrobe) in the cloakroom with fuse board, timers, buffer tank, pumps and also washing machine ironing board and all that crap you don’t want to see round the house ?

[ProDave]

3 hours ago, Declan52 said:

This is the kind of setup I would be looking if I install an ashp in the future. I basically use my pellet boiler in this fashion but use its onboard time programs to just heat my thermal store. There is nothing in my system that can cause the boiler to fire up other than me hitting the button. 

Can you just use the simple controller that comes with them to provide a set time that the ashp works. As in it comes on at 4am to 10 am then comes back on at 4pm to 8pm. These Time periods would be adjusted once I figured out how long it takes to get my tank to 45-48. 

The programmer that came with the ASHP will do all the timing, but it is a fiendishly complicated thing to program.  I wanted instead a simple boiler programmer  type interface that anyone can operate without having to pour over a poorly written manual to figure it out.  So the programmer we now use has all the usual functions, advance, on / off, 1 or 2 hour boost etc, all easy to understand.

 

Other makes of ASHP (mine is LG) many have a programmer that is easier to understand and operate.

 

It is worth noting that an ASHP will normally do space heating OR DHW heating, never both at the same time.  And it has a supply water temperature setting for each mode.  Exactly how they schedule heating and hot water depends a bit on the make, buy my LG one does the DHW heating in half hour bursts with a half hour gap before the next half hour burst if needed.  There is no explanation in the manual why it is set like this. My best guess is partly to only run it hard for short bursts to minimise risk of icing and defrosting, and partly so as not to stop heating the rooms for too long (though in our well insulated house it could turn the space heating off for several hours before you would even notice).  You can change the timings, and you can also select whether heating or hot water has priority.

[ProDave]

5 hours ago, epsilonGreedy said:

 

The absence of a plant room in my design is beginning to feel like a problem. I was hoping to box in the basics of the UFH manifold and pump under the stairs of the open plan sitting room as this is a good central point in the house and would avoid a UFH Clapham Junction at the main internal ground floor door threshold.

Not really.  My "plant room"  houses the mvhr and a circulating pump and some controls for the ASHP.  These could easily have gone in any convenient cupboard space anywhere in the house.

[PeterW]

16 hours ago, ProDave said:

Other makes of ASHP (mine is LG) many have a programmer that is easier to understand and operate.

 

It is worth noting that an ASHP will normally do space heating OR DHW heating, never both at the same time.  And it has a supply water temperature setting for each mode.  

 

That depends on the units and the sophistication of the programmer, not the ASHP itself. 

 

For example, @readiescards and I have both installed the IVT units which are rebadged Mitsubishi heat pumps. They come with no controls at all and are pretty binary in operation in that they just dump heat like a boiler does into the circuit. There are connections on the board for the  Mitsubishi controller but I’m not sure anyone has ever tried as the correct controller is around £400. 

 

Most heatpumps can be put into either an S Plan, W Plan or Y Plan config without too much hassle. They need a call for heat and a set temperature. I have configured this by setting the bottom stat on the DHW to 45c, and the stat on the UFH buffer to 35c. Hit either of these and the ASHP stops. This means in “heating” mode it may well cycle but it’s not that bad tbh. 

 

19 hours ago, Declan52 said:

I didn't think it would be as quick as that, pretty impressive.

 

It’s a straight factor of power output tbh - that rise needs around 10.5kw of heat and it’s a 9kw ASHP so given it was 8c outside I’d expect it to be able to go flat out with no de-icing and 70 mins or so would be the target. 

 

I do need to find some cheap clip on pipe stats though - anyone got a decent source..?

[bissoejosh]

Interesting thread. Are they are maximum distances that one should adhere to when siting ASHP / UVC or is this not an issue in reality? At the moment my ASHP will be around 8-10m away from the cylinder. 

Edited April 26, 2019 by bissoejosh

[ProDave]

4 minutes ago, bissoejosh said:

Interesting thread. Are they are maximum distances that one should adhere to when siting ASHP / UVC or is this not an issue in reality? At the moment my ASHP will be around 8-10m away for the cylinder. 

That won't be a problem. Mine is further. Just lag the pipework between well to minimise heat loss.