Optimising a heat pump

shinytop

70sbudgie said:

Reed_Richards said:

I don't have UFH and radiators, just radiators.  But the way it could be done is for the output water temperature to be set by weather compensation when the radiator zone is demanding heat but to be a fixed temperature when the radiator zone is not demanding heat.  This assumes that the WC temperature will be higher or equal to the UFH zone temperature so the UFH always gets water that is at least as hot as it needs.   

If I am understanding this correctly, it means the room thermostat for the radiator zone, would need to be the one that works with the weather compensation? Hence it is that one that is suggested as being replaced by the Mitsubishi stat? 

How would the radiator zone call for heat - I thought that the way to make the weather compensation more efficient (and avoid short cycling) was to not use a room stat? Or, by having additional zones, will the short cycling be avoided anyway, because there is always demand?

When the Mitsubishi room stat is working with auto-adapt, it does a pretty good job of keeping cycling down.  This is different to a 3rd party stat because it works with the ASHP and isn't just an on-off switch. Your problem is going to be when both zones are running because the ufh zone will be using the radiator zone flow temps, which will probably be too high and too variable.  I don't know much about ufh but can you not have some sort of a thermostatic mixer valve set up that controls the flow temp, a bit like a shower?     

70sbudgie

shinytop said:

70sbudgie said:

Reed_Richards said:

I don't have UFH and radiators, just radiators.  But the way it could be done is for the output water temperature to be set by weather compensation when the radiator zone is demanding heat but to be a fixed temperature when the radiator zone is not demanding heat.  This assumes that the WC temperature will be higher or equal to the UFH zone temperature so the UFH always gets water that is at least as hot as it needs.   

If I am understanding this correctly, it means the room thermostat for the radiator zone, would need to be the one that works with the weather compensation? Hence it is that one that is suggested as being replaced by the Mitsubishi stat? 

How would the radiator zone call for heat - I thought that the way to make the weather compensation more efficient (and avoid short cycling) was to not use a room stat? Or, by having additional zones, will the short cycling be avoided anyway, because there is always demand?

When the Mitsubishi room stat is working with auto-adapt, it does a pretty good job of keeping cycling down.  This is different to a 3rd party stat because it works with the ASHP and isn't just an on-off switch. Your problem is going to be when both zones are running because the ufh zone will be using the radiator zone flow temps, which will probably be too high and too variable.  I don't know much about ufh but can you now have some sort of a thermostatic mixer valve set up that controls the flow temp, a bit like a shower?     

I believe that the UFH already has something like a thermostatic mixer, because it has a max temp of 40°C (I think), whereas the radiators run at a higher temp. As I learn more about this, I am inclined to reduce the UFH max temp and run it for longer. It has such large thermal mass that I only run it for about 5 hours in the morning which is then good for the rest of the day. 

Reed_Richards

If you were going to run your radiators at 40 C then you don't need to worry about keeping the UFH going if and when the radiators are off.  

QrizB

For @Spoonie_Turtle as I didn't want to clutter the "Octopus Agile" thread with this!

The context, for anyone else who's interested, is whether you could run a heat pump solely during the 8hrs/day of cheap leccy you get with Cosy Octopus.

Spoonie_Turtle said:

I guess we could try bumping the temperature up a lot if we were to try that, but then what we'd be saving from the lower rate would be eaten into - even possibly wiped out - by the MUCH extra energy it would be using to get up into the 20s℃.  And I'm not convinced it'd work anyway, to get the house up to temperature within a 3hr slot enough to maintain comfort the rest of the time.

This would, in effect, be using the structure of your house as a thermal store to bridge the "off" periods. It's unlikely to be economic, but you could do something similar by greatly increasing the volume of your heating circuit.

Consider a hypothetical house with a standing heat loss of 3kW (72kWh/day heat demand). The longest gap between Cosy cheap periods is six hours, from 0700-1300. During that period, the house will lose 18kWh of heat.

Adding a 1500-litre volumiser to the system means that if your normal system temperature was 45C then you'd be able to extract almost 18kWh by recirculating the water through the system and allowing it to cool to 35C. In principle this would allow you to keep the property warm during the gap, with the caveat that the output of your emitters will fall as the system temperature drops.

You'd need a big enough heat pump to then return the system to it's normal operating temperature during the subsequent three cheap hours, so at least 9kW (3kW standing plus 6kW for 3 hours to reheat 1500 litres of water).

Is this a realistic solution? I'm fairly sure it would work, but the cost of 1500 litres of volumiser tank (purchase, installation, filling with glycol, and the problem of finding room for that much tank in a typical house) probably outweighs the benefits.

TroubledTarts

Spoonie_Turtle said:

I guess we could try bumping the temperature up a lot if we were to try that, but then what we'd be saving from the lower rate would be eaten into - even possibly wiped out - by the MUCH extra energy it would be using to get up into the 20s℃. And I'm not convinced it'd work anyway, to get the house up to temperature within a 3hr slot enough to maintain comfort the rest of the time

@Spoonie_Turtle

Also to move this away the idea was not to turn off the heat pump out of the cosy periods just set it back 7am-1pm as an example. 

Depending on your house and heat loss maybe 40-45oC during the Cosy period and 30-35oC during the 7am-1pm period. Then back up for the 1pm-4pm period etc etc

We found we could average around the 17p kWh mark with this change on Cosy with no batteries to assist.

Spoonie_Turtle

TroubledTarts said:

Also to move this away the idea was not to turn off the heat pump out of the cosy periods just set it back 7am-1pm as an example. 

Depending on your house and heat loss maybe 40-45oC during the Cosy period and 30-35oC during the 7am-1pm period. Then back up for the 1pm-4pm period etc etc

Oh that makes more sense as a suggestion.  They're nice ideas but we don't have a setback temperature (or rather, if we do I haven't been able to find it), any changes have to be manual.  Also not a clue how to change the flow(?) temperatures.

QrizB I did 'Thank' your post but forgot to respond, it's way over my head but I appreciate the effort that went into thinking about it and maybe it will be useful for anyone else thinking about this kind of thing.

TroubledTarts

Spoonie_Turtle said:

TroubledTarts said:

Also to move this away the idea was not to turn off the heat pump out of the cosy periods just set it back 7am-1pm as an example. 

Depending on your house and heat loss maybe 40-45oC during the Cosy period and 30-35oC during the 7am-1pm period. Then back up for the 1pm-4pm period etc etc

Oh that makes more sense as a suggestion.  They're nice ideas but we don't have a setback temperature (or rather, if we do I haven't been able to find it), any changes have to be manual.  Also not a clue how to change the flow(?) temperatures.

QrizB I did 'Thank' your post but forgot to respond, it's way over my head but I appreciate the effort that went into thinking about it and maybe it will be useful for anyone else thinking about this kind of thing.

I think you have an Ecodan so weather compensation with a temperature set higher on cosy periods and lower on non cosy could work unless you want to set your Ecodan to a flow temp(which should be possible) and manually adjust it during the day.

70sbudgie

On the topic of sizing a HP for optimum operation. My installer has (verbally) told me that my heat loss is 8.4kW, but they still recommend the 11.2kW unit rather than the 8.5kW unit. 

I thought it wasn't advisable to over size the HP as it becomes less efficient. At what point would the 11.2kW unit be considered as over sized?

Reed_Richards

I think you must have blocked me, @70sbudgie, because you're not paying attention to the comments I make.  To repeat, an 8.5 kW unit is insufficient for an 8.4 kW heat loss.  

70sbudgie

Reed_Richards said:

I think you must have blocked me, @70sbudgie, because you're not paying attention to the comments I make.  To repeat, an 8.5 kW unit is insufficient for an 8.4 kW heat loss.  

Apologies if I have blocked you,  it isn't something I have consciously done. On the other hand, I can't see where you have previously said an 8.5kW unit is insufficient, so I must have done/ missed something.

I have read NedS 's comment about not undersizing, but also got the impression that you disagreed

6) Conventional wisdom is don't oversize your heat pump.  It's rarely - 5 C outside but a lot of the year it could be relatively mild out so your heat pump only needs to operate at low power.  Heat pumps don't have the modulation range of a gas boiler and the modulation range can vary a lot from one model to another.  If your heat pump can't modulate down to the low output required for milder weather if will be forced to make the water hotter and cycle and that will reduce your efficiency.  It could be cycling for months for the sake of not working so hard on a few really cold days so I think Ned is wrong on this point

I am trying to understand where the threshold is that would be over / undersizing.