COP - diminishing returns on investment?

nxdmsandkaskdjaqd

NedS said:

Absolutely. COP (or SCOP) is a great way to determine your system is running in the right ballpark, and great for identifying when something changes or breaks (e.g, you suddenly go from a COP of 4 to a COP of 2), but ultimately we all just want lower running costs.

I get annoyed when people tell me the cheapest way to run my heat pump is slow and steady, 24/7. That's not a bad way to run a heat pump, but for us it is far cheaper to turn it off for 9-10 hours overnight and reheat the house from cold the following day. We use around 10-12kWh less electricity by turning off overnight. Unless the external temp is below 2C overnight, the house is too warm overnight, even set back to minimum. So for us it's cheaper to turn off overnight, save that 10-12kWh usage, and then run a slightly higher flow temp during the day (which is likely offset by the ambient temp being a few degrees warmer) achieving a similar COP and lower running costs. Having cheap electricity slots on the Cosy tariff helps where we can bump up the flow temps at half price.

Unfortunately we have no way of determining our Cost (of) Heating Annualised Ratio (CHAR) as, although we know our heat pump electrical input usage, we don't know the price per kWh for that usage, only the overall price per kWh for our whole house usage on that day.

I have a similar situation.  If I run the heat pump overnight in 0C conditions I will use 10Kw of electric. So I have now taken to run the system like a conventional gas heating system, on in the morning and off at night. 

The property is very well insulated and will only drop 0.5C in mild conditions overnight and 1.5C in very cold overnight. 

NedS

michaels said:

I think with an external thermostat though you are limited to a flow temp based either on weather comp or a fixed temp?

We just use constant temp at the moment basically because our rads/balancing/rather thin upstairs pipes seems to give us 18 upstairs and 21 downstairs which works for us or be it at the price of excess heat and therefore heat loss downstairs overnight.

Some folks control their heat pumps over a modbus connection, where you can set the flow temps and then automate that either using a python script or in a system such has Homely. This allows setting or adjusting the flow temps on a time-based schedule which is great for TOU tariffs such as Cosy (e.g, +3C when it's cheap, -3C overnight etc).

NedS

nxdmsandkaskdjaqd said:

NedS said:

Absolutely. COP (or SCOP) is a great way to determine your system is running in the right ballpark, and great for identifying when something changes or breaks (e.g, you suddenly go from a COP of 4 to a COP of 2), but ultimately we all just want lower running costs.

I get annoyed when people tell me the cheapest way to run my heat pump is slow and steady, 24/7. That's not a bad way to run a heat pump, but for us it is far cheaper to turn it off for 9-10 hours overnight and reheat the house from cold the following day. We use around 10-12kWh less electricity by turning off overnight. Unless the external temp is below 2C overnight, the house is too warm overnight, even set back to minimum. So for us it's cheaper to turn off overnight, save that 10-12kWh usage, and then run a slightly higher flow temp during the day (which is likely offset by the ambient temp being a few degrees warmer) achieving a similar COP and lower running costs. Having cheap electricity slots on the Cosy tariff helps where we can bump up the flow temps at half price.

Unfortunately we have no way of determining our Cost (of) Heating Annualised Ratio (CHAR) as, although we know our heat pump electrical input usage, we don't know the price per kWh for that usage, only the overall price per kWh for our whole house usage on that day.

I have a similar situation.  If I run the heat pump overnight in 0C conditions I will use 10Kw of electric. So I have now taken to run the system like a conventional gas heating system, on in the morning and off at night. 

The property is very well insulated and will only drop 0.5C in mild conditions overnight and 1.5C in very cold overnight. 

Wow, you are lucky. Our property can drop 3-4C overnight. With an average 1C drop, I'd definitely turn off overnight.

I'm currently experimenting with a half-way house of off for 4 hours from midnight to 4am, and then back on in the cheap Cosy period of 4-7am to reheat at cheap rate before we get up. We do use more electricity that turning off completely, but comfort levels are improved when we first get up as it can otherwise take 2-3 hours to reheat the house. Still uses less than running 24/7, and means the house is not too warm overnight for sleeping.

QrizB

NedS said:

nxdmsandkaskdjaqd said:

The property is very well insulated and will only drop 0.5C in mild conditions overnight and 1.5C in very cold overnight. 

Wow, you are lucky. Our property can drop 3-4C overnight. With an average 1C drop, I'd definitely turn off overnight.

We don't (yet) have a heat pump.

Our house is nothing special; a 1950s council-built semi. Built of brick and fully dense block, cavity wall filled with poly beads a couple of decades ago, less glassfibre loft insulation than is currently recommended.

We did go away for a week over Christmas and left our GCH on but with the room thermostat turned all the way down, nominally to 10C. Outdoor temps were around 5-8C. The house started at 18C and took six days to drop to 12C, initially at about 1.5C/day then levelling off a bit. Once it got down to 12C the heating started cutting in very occasionally; we did have about 2.5kWh/day of electrical load (fridge, freezer, various background stuff) which made up most of the energy input during this period.

When we returned it took a couple of days to heat soak the building structure back to our normal operating temperature.

FreeBear

QrizB said: We did go away for a week over Christmas and left our GCH on but with the room thermostat turned all the way down, nominally to 10C. Outdoor temps were around 5-8C. The house started at 18C and took six days to drop to 12C, initially at about 1.5C/day then levelling off a bit.

That is pretty good going. Way better than the 2-3°C drop I experience overnight.

michaels

QrizB said:

NedS said:

nxdmsandkaskdjaqd said:

The property is very well insulated and will only drop 0.5C in mild conditions overnight and 1.5C in very cold overnight. 

Wow, you are lucky. Our property can drop 3-4C overnight. With an average 1C drop, I'd definitely turn off overnight.

We don't (yet) have a heat pump.

Our house is nothing special; a 1950s council-built semi. Built of brick and fully dense block, cavity wall filled with poly beads a couple of decades ago, less glassfibre loft insulation than is currently recommended.

We did go away for a week over Christmas and left our GCH on but with the room thermostat turned all the way down, nominally to 10C. Outdoor temps were around 5-8C. The house started at 18C and took six days to drop to 12C, initially at about 1.5C/day then levelling off a bit. Once it got down to 12C the heating started cutting in very occasionally; we did have about 2.5kWh/day of electrical load (fridge, freezer, various background stuff) which made up most of the energy input during this period.

When we returned it took a couple of days to heat soak the building structure back to our normal operating temperature.

5-8C - was that day and night?  Heat loss is still going to be considerably more at 0/negative outside temps?

QrizB

michaels said:

5-8C - was that day and night?  Heat loss is still going to be considerably more at 0/negative outside temps?

Sadly my cheap ZigBee temperature sensors (or possibly the almost as cheap Lidl gateway) don't store hourly data for more than a month, so I can't see what the daily lows and highs were over Christmas and New Year. I do get a monthly summary chart and that's where I got the 5-8C range from.

But it was fairly mild and I don't think there was any frost, so yes it stayed above freezing that period. It was about 10C colder outside than in, so if it had been 0-3C instead that would've been 15C difference and 50% more heat loss? Possibly 2-3C a day?

michaels

QrizB said:

michaels said:

5-8C - was that day and night?  Heat loss is still going to be considerably more at 0/negative outside temps?

Sadly my cheap ZigBee temperature sensors (or possibly the almost as cheap Lidl gateway) don't store hourly data for more than a month, so I can't see what the daily lows and highs were over Christmas and New Year. I do get a monthly summary chart and that's where I got the 5-8C range from.

But it was fairly mild and I don't think there was any frost, so yes it stayed above freezing that period. It was about 10C colder outside than in, so if it had been 0-3C instead that would've been 15C difference and 50% more heat loss? Possibly 2-3C a day?

You should be able to get a pretty good idea of local temps from nearby weather underground loggers.

Reed_Richards

You should bear in mind that the rate of heat loss depends on the difference between the inside and outside temperature.  So, for example, if a house starts at 20 C and it's 8 C outside, by the time the inside temperature has fallen to 14 C it should be losing heat at half the rate that it was at 20 C.  This is more than "levelling off a bit" and unless you do the maths carefully, it's difficult to compare what happens overnight with what happens in a week.    

FreeBear

Reed_Richards said: This is more than "levelling off a bit" and unless you do the maths carefully, it's difficult to compare what happens overnight with what happens in a week.    

Throw in solar gain during the days when the sun is shining, and it makes the comparisons next to impossible. As an example. Heating was on between 08:00 and 09:20, from then on, solar gain kicked in (curtains got pulled back at 09:50).