UFH + heat pump Advanced control in PH

[LukasV]

Hi,

 

I am new to this forum but not to the topic and as many of you I am thinking too "how to control the UFH". 

Here were some systems described that looked that they reached the comfort-efficiency, but, unfortunately, I would say that they lack energy-/cost-efficiency.

 

Probably the best way how to easily reach a stable comfort and solid cost-effectivity is to use the pre-builded weather compensation curve and run the heating 24/7. After proper setting this curve deliver the exact necessary amount of heat and the inflow would have the lowest possible temperature (thus highest COP). Ofc the pump must be connected directly without any blender/buffer or so. ... so this is the way I would probably go.

 

I do not think that it is a good idea to rely on weather forecast so I would use a temperature sensor to provide an accurate temperature instead of forecasted.

As our passive house should be in modern style - cube with massive south facing glazing - I am a bit afraid of solar gains... to prevent overheating I would use a south sun-facing weather station. My idea is to use a weighted average from the sensor on the north and the weather station in the sun on south. I guess the temperature from the weather station should have weight about 1/3 - 1/2 (but I must run the numbers first to find out how many solar gains compared to heat loss expect). The weighted average would be the final temperature for the heating curve. Hopefully, this could be enough to keep the temperature stable.

 

This is also a quite effective in terms of energy/cost, but not fully effective. It is much better to run 16 hours inflow at 28 degree instead of 24 hours of 25 degree (the COP will change minimally) but with the price for MWh 60 EUR instead of 150 EUR. With the electricity from spot market I need to let the pump run just something like 12-16 hours a day. And here is the problem... I do not know how to compensate the curve/inflow temperature. Everyday could be different. For example the pump can run 18 hours on Monday, 12hours on Tuesday and 16hours on Wednesday to heat only with a good price. But which inflow temperature do I need?? Let's say I can properly set the curve for 24/7 after which the inflow would be at 25 degree... but then I need to change it to higher to compensate (preheat and warm up) the time when the pump was stopped. 

.... I know the exact outside temperatures during the day, I know outside temperatures during shutdown and a day-before I know when and how long will the shutdown be... but I must calculate the inflow temperature that will be able to preheat the house before shutdown and warm up the house after the shutdown. How to do that?

 

I do not think that PHPP can calculate the heat loss exactly, if it could, no one would need to titrate the curve... there are also another variables like wind speed, humidity and so. And I am definitively not able to exactly calculate solar and internal gains. Any gains or inaccurate calculation in 24/7 mode would stop the pump immediately as the reverse flow temperature rises so it would be self-controlling, but these gains during the shutdown cause only that the slab will remain with higher heat inside at the end of shutdown period and the subsequent warm up with "hotter" water will overheat the house.

 

I could measure the heat (temperature) in the slab but I would need to do it in many rooms to get a good overview about the amount of heat there and yet it would be extremely difficult to calculate the correct need of heat to be added and according to it necessary inflow temperature. So this is probably not the way.

 

Also except for the weather station in every room there will be a thermostat (the house will be driven by Loxone) so I could do that in the way that when the average temperature in few rooms hits some point I could stop the pump... maybe is this the way how to find out that the warm up is done?

Unfortunately, the whole family in the living room during the shutdown would heat the slab there enough that in warm up period (until other rooms would be warmed up too) there would be overheated living room. So this way with thermostat feedback looks problematic too.

 

It is quite problematic although I am now dealing only with weather control and shutdown period problems,.... to deal with internal gains all over the day would be total crazy - although I somehow believe that internal gains cannot be solved be heating anyway (I see the only way how to solve internal gains through MVHR/windows).

 

I am on the way to abandon the shutdown and just buy a bigger battery. With the 24/7 mode I would need something like 2x4 hours "shutdown". With 4,5kW heat loss of the house and ASHP/GSHP it needs only something like 1,3-1,8 kWh per hour under the worst weather conditions ie something like 2x 6kWh electricity from battery (which can be restored immediately when price drops) to cover the most expensive periods of the day. That would be better than loose the comfort ie having cold or overheated house just to save a money. 

 

Does anyone have any idea how to solve it and how to find out correct inflow temperature please?

Thanks

 

[JohnMo]

Really not even sure you need to use weather compensation on a passivhaus build, ours isn't and it just added no value running wise.

 

Also found that even though the heat pump can deliver 25 degs, you actually have to run 28 to 29 as the floor doesn't cool enough to allow the hp to restart after a heat cycle.

 

Two options

1 Batch charge floor on time of use tariff, set flow target about 30 to 35.

2 Run with a low hysterisis thermostat (0.1 Deg) such as Computherm Q20RF as a single zone flow at 28, the low hysterisis will stop overshoot and undershoot.

 

Both I would do on a fixed flow temp, or a slight WC starting at +3 Deg add 3 degs to -3 to compensate for defrosting.

 

Simple  is best.

 

I started with 7 zones, now have 1

Had a buffer - now I don't 

Had a mixer and pump - now I don't 

[LukasV]

9 hours ago, JohnMo said:

Really not even sure you need to use weather compensation on a passivhaus build, ours isn't and it just added no value running wise.

 

Also found that even though the heat pump can deliver 25 degs, you actually have to run 28 to 29 as the floor doesn't cool enough to allow the hp to restart after a heat cycle.

 

Two options

1 Batch charge floor on time of use tariff, set flow target about 30 to 35.

2 Run with a low hysterisis thermostat (0.1 Deg) such as Computherm Q20RF as a single zone flow at 28, the low hysterisis will stop overshoot and undershoot.

 

Both I would do on a fixed flow temp, or a slight WC starting at +3 Deg add 3 degs to -3 to compensate for defrosting.

 

Simple  is best.

 

I started with 7 zones, now have 1

Had a buffer - now I don't 

Had a mixer and pump - now I don't 

Hi,

 

thank you for opinion,

 

it is possible to keep things simple when they are simple or ... when you can not do the math. I know that there is a thought to run the UFH with thermostat with low hysteresis. Unfortunately, this is total nonsense. I am aware of that that @Jeremy Harris reached good control with it, but this way is nonsense and should be banned to advise to do this to anyone. Do the math.

Heat pump lifespan is limited by "running hours" or "starts". In passive house the running hours are unimportant, we will be dead by the time it reach the maximum.

But using a thermostat with low hysteresis (ie 0,1 deg) the only you will get is a cycling heat pump. It will start and stop every few minutes or seconds. No way to reasonably prevent this. And now the math ... regular passive house needs something around 8-10 MWh heating energy in a year, with heat pump it means something like 2,25-4 MWh of electricity per year... it is circa 225-400 Euro per year (contra 900 Euro with heating directly with electricity). But the pump cost is circa 10 000 Euro or more. So you need something like 15-20 years even to payoff the investment to the heat pump!! With the heat pump constantly cycling you will not get even half of that. This way of use makes no sense, it is better to heat the house with electricity directly and heat the water with heat pump boiler. But I need to use the UFH also for cooling in summer, otherwise I would need to use an A/C (cca 5000 Euro) so the investment cost for electrical (dry) UFH + heat pump boiler + A/C would be quit similar like the investment costs for wet UFH with heat pump which can do all of that in one device... but I cannot destroy it too soon. Beside that where should be the thermometer placed? In a hallway? I do not care about the temperature in hallway, I want to prevent overheating in living room meanwhile I heat the kids' rooms enough.

 

As to the option 1),... how do you know that I need a temperature 30 to 35? Which one should I use when I heat 12 hours? When 14 hours? When 16 hours? Should I use the 35 also when I am roasting a goose for 12 hours? Any difference when we are out of house vs. when we are at home and kids are running in living room with dogs and TV playing whole the time or not? How long should I charge the floor this way?

 

How should the fixed flow temp help? Why to heat with 30 both time when outside is -10 or +7? Yes, it is better to use a lower COP and heat the house with cheap price faster than maximize COP and heat with price going from 20 to 150 Euro, but how to use it? I can run the UFH on 42 (or so maximum) few hours a day (when the price is minimal) to charge the slab but that will cause an overheating. I can stop the charging the slab when the temperature in a few rooms reach the max. to prevent overheat. It could be a cost-efficient (not top but good) but again I will get in troubles... during the shutdown the whole family will be in living room, duck is roasting, TV is running, same like kids and dogs... the slab there is not discharging there but in other rooms is. Then the pump/UFH starts again with 35-42 inflow... The living room is reaching the top temp (despite the fact that I am the only one there now) but I can't stop heating because other rooms are still not heated enough. I am waiting for another representative room to call to stop heating. This takes another few hours and meanwhile the living room is overheated because the slab there had a lot of remaining energy and with inflow temperature 35-42 there is no way ho to stop discharging the heat into air there or even to revers a heat flow to other rooms. The comfort would be totally unstable and low.

 

So how to deal with it?

 

[JohnMo]

8 hours ago, LukasV said:

using a thermostat with low hysteresis (ie 0,1 deg) the only you will get is a cycling heat pump. It will start and stop every few minutes or seconds. No way to reasonably prevent this

Thermal inertia prevents this. A passivhaus or one close to it doesn't change temperature quickly, also we are talking specifically about UFH not radiators. UFH systems are ultimately a huge buffer of heat store. The only way to prevent temperature overshoot by a simple anyone can use, low hysterisis thermostat. A normal thermostat will give an undershoot (in our house) of 3 degs and the same overshot. So a 6 Deg swing in room temperature. The 0.1 hysterisis is closer to 0.3 swing in 24 hrs, with heat pump running cycles and long off cycle.

 

Through practical experiments I and many other have proven this works.

 

Cooling I use the same thermostat set to cooling mode, mine is in the hall, but every house is different.  You get temperatures in different rooms through balancing the UFH.

 

UFH cooling isn't the same as Aircon, which cools the air. Underfloor cooling,room air temperature stays warm, but the room feels cool because the floor is sucking the heat from you.

 

8 hours ago, LukasV said:

But the pump cost is circa 10 000 Euro or more. So you need something like 15-20 years even to payoff the investment to the heat pump!

It doesn't have to, I paid £1300 for mine.

 

The rest you are talking extremes, that makes no sense. The fact is you can set the target flow to any temperature you want, the heat pump would struggle to get there due to the floors thermal inertia. The heat pump sets it's actual flow temp based on return temperature first, to manage dT. Once dT has started to reduce it adds to the flow temperature... 

 

I have done the maths and the experiments.

 

The facts are 

Anyway you heat the floor, the floor settles to approx 1 to 3 degs warmer than the room air, depending on the house heat loss (delta between inside and outside). Otherwise you overheat the house. The closer the room temperature is to floor temp, the less heat that is given off to the room. So if the floor is at 23 the room cannot get any hotter without an additional heat source. You set flow temp via calculation and experiment (combination of flow rate and mean flow temp).