Set boiler on high or low for heating?

[kasser]

I've got a mid-terrace property at around 18 deg in the morning at the moment, without heating overnight. It's probably not an accurate thermometer as there's another showing 15 deg. But we're only interested in the temp rise and this property has wall and loft insulation. I didn't do a heat loss calculation but you can see already it's going to be very low.

Is it cheaper to warm this house with a short burst of heating or on low heating for a few hours?

There are massive, oversized rads in the house. It's a 2 bed btw. With a short burst of high heat, it takes about 10 min to warm up and can be turned off 30 min after switching on. Something like 18 to 21 deg in 30 min. Stays warm for a few hours.

For low heating, I've turned down the boiler output to 20% (from 24kW although min heating is supposed to be 9kW) and flow temp to 45 deg. Return temp is 40 deg. It takes 1 hr+ to get to 21 deg. The boiler can run for much longer as the thermostat doesn't switch it off immediately but the temperature also takes much longer to keep rising past 21 deg, probably due to the low flow temp.

At 80 deg flow temp and boiler at full power, the 21 deg often gets overshot and reaches 22.

Good practice is to run the boiler for a long time at low heat but I don't think this is possible here given min output is 9 kW.
I once saw the calculations and modelling that were used to prove that it's cheaper to keep a house warm than to heat it from cold. I don't remember the exact numbers but the temperatures used were quite low.

My point is that this is just modelling - wouldn't the conclusions only apply to the model used (temperatures, boiler, type of house, insulation, etc)?

In the example I've given, I've provided a lot of vague numbers. I suppose the only way to find out accurately would be to heat the house in short bursts for a day or two and record gas usage, then repeat with heating on low and compare?

 

[Brambles]

The only true way is to try both options and to measure your daily gas consumption at the meter. You obviously need to do it on days with the same outside temperature and similar house occupancy hours.

In more general terms you don’t want the flow temperature to exceed a level where the return temperature is above 55degrees C or the boiler won’t condense - so is working at a lower efficiency.

In essence for what you are trying to achieve try to keep your flow temperature to 70 degrees or below.

From my experiments, I have never achieved a situation on a gas fired panel rad system where a constant low heat was more economical than a well thought out programmed control - but others may have had different experiences.

If is this is relevant, it is certainly not the case for occasional use swimming pools - they are far more efficient if left to cool then boosted for 8 to 12 or so hours before use

 

[gmartine]

Even though you're concerned about the boilers lowest output you don't suggest the boiler is short cycling so I would say if that is the case the low temp approach should be more efficient for a couple of reasons. The boiler is obviously in condensing mode at 40 deg C and the other key here is you have oversized radiators, had they been "properly" sized you'd struggle to get the rooms up to temp using those low temperature flow rates. As mentioned the true measure of course is to measure the gas consumed using both methods on similar days.

 

[kasser]

Even though you're concerned about the boilers lowest output you don't suggest the boiler is short cycling so I would say if that is the case the low temp approach should be more efficient for a couple of reasons. The boiler is obviously in condensing mode at 40 deg C and the other key here is you have oversized radiators, had they been "properly" sized you'd struggle to get the rooms up to temp using those low temperature flow rates. As mentioned the true measure of course is to measure the gas consumed using both methods on similar days.

I meant to imply that the boiler does short cycle even in low output.
I can still have it in condensing mode in high output by ensuring the return temp is no more than 55 deg.
Agree, it would be hard to heat a house with flow temp of 45 deg. I even went down to as low as 30 but the rads only got lukewarm.