This is interesting @sunandair. The basic weather compensation, relying solely on instantaneous outdoor dry bulb temperature, is crude. I've wondered if offsets or time lag delays could be added to the basic 'curve' (we say curve, but most are linear) to consider solar gain, which is largely instant via glazing, and transmission gain, which through the fabric of a typical house, has a lag of several hours. My ASHP controller doesn't have any offset or time lag option.

It's more an issue with wet rad conversions where there's little thermal lag in the heat delivery. The house tends to overheat a little in the later day, as the thermal mass transmission absorbed during the day slowly releases from the fabric to inside, combined with gradual solar gain. Meanwhile, the outdoor temperature is falling, so the ASHP starts to raise the LWT because that is it's only control reference point for climate compensation. My crude way of getting round this is smart TRVs set to limit output from certain radiators at certain times e.g.: south facing room radiators in the evenings. it's not very elegant, but it works as a basic 'overheat protection' method.

Homely has some much clever ASHP control considering smart tariffs and predicted weather.

This is interesting @sunandair. The basic weather compensation, relying solely on instantaneous outdoor dry bulb temperature, is crude. I've wondered if offsets or time lag delays could be added to the basic 'curve' (we say curve, but most are linear) to consider solar gain, which is largely instant via glazing, and transmission gain, which through the fabric of a typical house, has a lag of several hours. My ASHP controller doesn't have any offset or time lag option.

It's more an issue with wet rad conversions where there's little thermal lag in the heat delivery. The house tends to overheat a little in the later day, as the thermal mass transmission absorbed during the day slowly releases from the fabric to inside, combined with gradual solar gain. Meanwhile, the outdoor temperature is falling, so the ASHP starts to raise the LWT because that is it's only control reference point for climate compensation. My crude way of getting round this is smart TRVs set to limit output from certain radiators at certain times e.g.: south facing room radiators in the evenings. it's not very elegant, but it works as a basic 'overheat protection' method.

Homely has some much cleverer ASHP control considering smart tariffs and predicted weather.

Posted by: @sunandair

↑

I have been experimenting with ways to “Influence” our Weather Compensation Curve to modify its target settings based on the amount of solar gain we may get on a particular day.

If something like this existed (or could be set up) within the existing thermistor controls this might greatly improve the accuracy and efficiency of the heat delivery when using both WC and Auto Adaptation. Which may avoid overheating at times when solar gain might be high.

So, we have in our house orientation an East-West aspect with high solar gain in early morning and less so in early evening. This is from 4 largish windows heating lounge and study and two bedrooms.

so I have been using the east elevation to position the Ambient thermistor ensuring it is shielded from direct sun but is able to get a reasonable and adjusted amount of reflected sun from the house frontage. 

On otherwise cold mornings with high sun output this has resulted in a modification of the “perceived” ambient that the HP uses to set the target room temperature. 

So we have seen the weather compensation curve act as though the morning temperature is possibly 5c warmer than the real frosty morning. This has successful altered the target LWT which has still maintained a comfortable 21c room temperature… but it still will be monitored next autumn.

It seems to be working and it’s the early morning solar gain which we are particularly happy to affect since that generally is the colder part of the active day.

I would be interested to hear any views or if anyone has been doing anything along these lines. 

You are using what in an industrial control system would be described as a 'feed forward' input. In practice a feed forward signal is used to try to predict what will happen in the near future, so that the controller can start taking appropriate action.

When correctly adjusted, feed forward optimisation can be highly useful.

Well done.

Posted by: @allyfish

↑

This is interesting @sunandair. The basic weather compensation, relying solely on instantaneous outdoor dry bulb temperature, is crude. I've wondered if offsets or time lag delays could be added to the basic 'curve' (we say curve, but most are linear) to consider solar gain, which is largely instant via glazing, and transmission gain, which through the fabric of a typical house, has a lag of several hours. My ASHP controller doesn't have any offset or time lag option.

It's more an issue with wet rad conversions where there's little thermal lag in the heat delivery. The house tends to overheat a little in the later day, as the thermal mass transmission absorbed during the day slowly releases from the fabric to inside, combined with gradual solar gain. Meanwhile, the outdoor temperature is falling, so the ASHP starts to raise the LWT because that is it's only control reference point for climate compensation. My crude way of getting round this is smart TRVs set to limit output from certain radiators at certain times e.g.: south facing room radiators in the evenings. it's not very elegant, but it works as a basic 'overheat protection' method.

Homely has some much cleverer ASHP control considering smart tariffs and predicted weather.

 

At the end of the day I suppose it comes down to what one is trying to achieve, is it constant indoor temperature or reduced energy consumption, or a combination of both?

Solar gain may cause an increase in indoor temperature, but is this always a bad thing, provided the increase is kept within certain limits?

As I have mentioned before, we have a hybrid heating system consisting of a gas boiler and an A2A ASHP. We also have a 4kWp solar PV system. At the moment the gas boiler is shutdown completely, and we run the heat pump during the day when it is powered almost exclusively by the solar PV. By increasing the indoor temperature from the desired 21C, to the 23C to 24C region, we then do not required any form of heating overnight. At present we draw approximately 3kWh of electrical energy each day from the grid, and use no gas.

In the shoulder months we use any available solar PV to raise the indoor temperature during the day, which reduces how much work the gas boiler is required to do at night.

In the Winter months the gas boiler provides the bulk of our heating, but is controlled by an industrial type controller, which not only senses the indoor temperature, but also the temperature of the radiator in my study. Just as a heat pump controller varies the LWT to control the indoor temperature, my controller varies the radiator temperature to match the heat demand. Even without any WC, the controller can keep the indoor temperature at 21C +/- 0.2C. It is interesting to see how a 10C variation in outside temperature can cause a 10C change in radiator temperature, whilst the indoor temperature remains an almost constant 21C.

So it proves that fancy predictive controls are not always necessary, at least in our situation, provided one has the correct type of control system that has been optimised for best performance.

My next project will probably be battery storage, to try to reduce our total energy consumption during the Summer months to zero, and to further reduce our electricity and gas consumption during the remaining months of the year.

Posted by: @allyfish

↑

My crude way of getting round this is smart TRVs set to limit output from certain radiators at certain times e.g.: south facing room radiators in the evenings.

Hi @allyfish

that sounds a useful way to get heat in the right places, using smart TRVs… Is it possible to link one TRV with another? What I mean is- Can TRVs be linked so if one valve closes- another valve opens to increase the heat in another room?

This, crucially, will maintain the overall flow volume in the HP. I think that would help avoid short cycling which seems to be a popular topic in these threads at the moment.

what I’ve been trying to do is lower the target LWT by making the outdoor ambient thermistor respond to localised solar gain which should place an immediate adjustment to the Weather Compensation setting. In our case this targets the early morning sun when the WC curve would otherwise show a possibly frosty cold outdoor temperature and try to raise the flow temperature only to overshoot later in the sunny morning. Obviously if there is no sun then the thermistor wouldn’t alter the WC curve. We are attempting to target an average 4c rise when the sun shines and see how it changes things..