CH - most efficient way to run in cold weather?

[Velvet]

With the current chilly weather, I'm mulling over the best (most efficient) way to keep my place warm. I work (or rather did) from home, so weekdays normally see me here pretty much all the time. Weekends I may be away for the weekend, or again pretty much here all the time.

I have gas-fired CH, via combi boiler, no HW tank. I live in an old (1880's) property (small hall off which toilet, bathroom, kitchen, livingroom off which 2 bedrooms), the upper of two maisonettes. Insulation in the roof, though not to all that high a standard. No insulation in floor, so I get the benefit of heat from downstairs when it's running. Double glazing throughout, one external door not d/g. I have a CH temp control on the boiler (and separate one for HW). It's a non-modulating boiler. Property is quite large for room sizes but sloping ceilings help reduce actual volume, and the rads are well sized for the rooms. No TRV's but programmable room stat located in the hall.

The wall construction is somewhat unknown. I suspect brick for some walls, either brick or brick/timber/lathe&plaster construction for others - it's possible they're just timber/lathe&plaster throughout a lot of the walls but not drilled into them to find out. Internally everything apart from kitchen broom cupboard is drylined/plasterboarded due to condensation/damp being a problem previous to my occupancy.

I'm not about to start adding insulation/upgrading doubleglazing/replacing front door etc, so suggesting I stop more of the heat from being lost in the first place *isn't* a useful OR valid suggestion no matter how well-founded you think it might be.

Now, after all that background, the question. Since I'm not out all day at work, I'm not sure what the most efficient way of heating the place is.

Should I:

Turn the boiler temp down, and keep the room stat at a fairly constant setting and just a couple of degrees cooler at night, so that the system is circulating fairly warm water around on a more frequent basis (on the basis that eventually the heat lost from the house will equal the heat the boiler is putting in near enough, with the CH running for long periods with less frequent firing of boiler).

This would presumably avoid having to shove heat back into the walls etc during the heat-up phase in the mornings.

Or:

Turn the boiler temp up, and set the room stat to drop significantly overnight, with a blast of heat in the morning, then lower level during the day, then another blast of heat in the evening. This means the place will have cooled down significantly overnight, and the boiler will be working hard to get the place back up to temp in the morning, it'll cool slightly again during the day (I just wear jumpers cos it's easier to regulate my heat with clothes depending on what I'm doing than it is to turn the heating up and down all the time!), heat it back up during the evening again before effectively switching it off over night.

This means the boiler's heating up the fabric of the house twice a day, and then keeping the heat constant.

Now, I originally thought that it was easy to work out, and if the place spent longer not needing to be heated (ie, overnight/during the day if I'm wearing warm clothes) then it worked out more efficient to do the latter, but I'm starting to wonder now.

Heat loss would be the same through the fabric, presumably it'll lose heat faster when the difference between inside and outside is greater, in which case keeping a lower overall temp would be more efficient (and this is actually what I've done already, reduced the room stat by a couple of degrees, which is fine for me).

So do I match the heat loss on a fairly constant level, thus keeping the place heated to a fairly constant level, or do I let the outside suck the heat out of it overnight, then heat the place back up and start maintaining heat in to counter heat loss?

As you can tell if you've stuck with this to the end, I'm a bit confused as to what the best thing to do would be. I've googled and googled both in web and groups to see if this has been discussed and any evidence found on the different ways to keep a place heated, and can find nothing that really explains it, just lots of home energy efficiency tips. I guess I'm looking for an article that explains the physics of heating a home, and looks at the differences between the two ways of doing it.

Any replies will be much appreciated, and thanks for sticking with it if you got this far.

Velvet

[Christian McArdle]

It will probably be more efficient to let it drop overnight, particularly as your maisonette probably has low thermal mass. However, the difference might not be all that great.

You really should install TRVs on all the radiators not in the thermostated room. They could save you a lot of cash AND make you more comfortable as it will prevent rooms overheating. It will also considerably reduce the overshoot when heating from cold.

Christian.

[Velvet]

For the same reason as the insulation, I'm not prepared to installed TRV's. I don't own this place, long term rent (though that's not as certain to continue as it used to be due to changed landlords).

Don't find it overshoots much, one of the reasons I have the stat set a couple of degrees lower than normal - the overshoot brings it up to a nice temp rather than being overly hot for a bit.

Will have to go look up thermal mass for an idea on what my place would be catagorised as, I think. The age, building materials and style mean there often seems to be quite a lot of difference between how it behaves and how it should, window placements and the sloping ceilings etc.

Ta for the reply though, given me something else to go investigate. Much appreciated.

Velvet

[Alan Shilling]

----- Original Message ----- From: "Velvet" Newsgroups: uk.d-i-y Sent: 28 January 2004 10:01 Subject: CH - most efficient way to run in cold weather?

[Velvet]

Sorry, maybe I wasn't that clear. I know running the overall temp lower leads to less heat loss, but my question was whether keeping it at (for example) a min of 17C overnight (which would mean the boiler coming on and off) would be more cost-effective than letting it cool down overnight from the 19C in the evening to perhaps 14C before stat turns it up to 18C for the morning blast of heat.

Bearing in mind I'd likely get no heat overnight from downstairs, since I don't think they run theirs overnight, I lose a lot of heat quite quickly, and I'm not sure if it's better to maintain the temperature, albiet a bit lower, overnight, or just let it drop then have the boiler firing for a long period in the morning to bring it back up to temp?

Added to this, I've got the option to changing the boiler temp (of the water circulating in teh rads). I normally have it set to about a third, not quite sure what that is in degree's C of water but I have thermometers on the pipes at teh boiler exit/return so can find that out if it would help. I turned it up to about 2/3 because it does take quite a long time to warm up in the morning. Not sure if that will make much difference as I only did that after I got up this morning.

Having already lowered the stat temp and got used to (and indeed finding I prefer a cooler rather than warmer house most of the time, as long as I don't have cold feet), I don't really want to drop it any further. Handy bit about the programmable room stat is it gives me an easy indication of what the current temp is, just by flicking the override up/down and seeing when it stops/starts firing. This is how I know how cold the house drops to overnight (sometimes I'm up before the next programme on the stat kicks in, though that's rarer these days).

Ta for the reply though.

Velvet

[Bob Eager]

I look at this in terms of simple input and output.

To get it to a given temperature in the morning, you either leave it on all night or you don't.

Heat loss is greater if you leave it on (as the differential is greater), so you lose more heat overnight if it's on than if it's off. In both case you need to replace the heat lost. You have more heat loss to replace if you leave it on.

QED. Turn if off!

[Ed Sirett]

The least heat lost from the house wil be the regime that runs the lowest average temperature. So you need to find what the lowest temperature you're content with for the night-time (suggest 15C until you know better). Like-wise anytime you are not there you need you need to find the lowest temperature you would be happy with considering:

1) The house will take a while to heat up from the low temp. 2) You don't really want the house getting much below 5-10C becasue you may get temporary condensation on objects with a high thermal mass until the house is warmed up again.

[Velvet]

I keep the house to a fairly normal cycle of heat even when I'm not there for a weekend (tend not to be gone any longer than that) because I have pets who, though they'd no doubt survive, would take a while to acclimatise to lower temps, and then have to acclimatise to higher ones again - and other pets who will simply end up with their own heaters running constantly to keep them up to temp. With that in mind, the stat never sets itself below about 12C at any time day or night, and while daytimes are cooler than if I'm around, they're still no lower than 16, with a burst of a couple of degrees of an evening.

My previous experiences with letting a house cool to around 10C is that it took hours, literally hours and hours, to warm back up again. Possibly even longer than that - a day, maybe. It wasn't this place so I'm not sure how much of that was the heating (different system, not going to start an argument so won't say what it was).

I'm perfectly happy with the stat dropping to 11 or 12 overnight, because at the moment (well, before this recent cold spell) it's never got down beyond 14 or so. Me personally I don't mind if it drops below that, cos I have a thick duvet and prefer a cold rather than hot room to sleep in. But getting up in the morning I'm conscious of how long the heating runs for to bring it back up to temperature, vs how often and long it runs for if it's just maintaining the steady temp.

That's how I got to wondering which one actually does use more gas, and therefore, costs more... and searched the web, but to no avail, no concrete examples and theories that demonstrate the differences between the two methods seem to exist. And whilst I'd give it a go as a home experiment, I know the weather's too variable to make the results worthwhile.

Oh well. Ta for the replies. Anyone else?

Velvet

[Andy Hall]

Well...... you've introduced additional information that you have animals who require an environment influenced by the room temperature.

If there are a significant number of them or more to the point their enclosures and the energy requirements, then the cost difference of energy for the house may well become irrelevant. If the heating for the enclosures is electric, then that is costing 3-4 times the gas cost anyway.

If economy is a real issue for you, then insulating the animal enclosures makes sense.

I've done this for an enclosure housing two creatures who like to have an environment of about 29 degrees and 80% humidity. It was a challenging project, but I made a significant difference on electricity use with insulation to avoid heatloss to the room.

.andy

To email, substitute .nospam with .gl

[Doctor D.]

On a purely practical and non-scientific test over the last two winters I've left our heating on constant and turned the thermostat down to 20 at night. (I don't think the scale is very accurate as we run it at about 25 during the day - and the house is seldom too hot.)

It's a 1988 4 bed detached and we have no TRV's. The kids used to wake at night and it was far more pleasant to deal with in a warm house, I also appear at home at all hours of the day after site visits to do my paperwork - nice to return to a warm house.

The result has been a very marginal increase in our gas bills, though not enough to increase our monthly direct debit. The heating seems to be off for the vast majority of the time during the night, and when I'm sitting at home during the day. The seemingly endless hot water is also blissful!

[Alan Shilling]

Hi it's me again, maybe I didn't provide a simple enough explanation - if you run continuously you'll use more gas. This is because the average temperature inside the house will be higher and hence the heat loss to the outside will be greater (as I'm sure someone else has also pointed out).

You're probably thinking that there is some complex stuff going on here - but there isn't. Think of your house as a box with gas going into it and heat coming out. The hotter it is inside, the more heat comes out and the more gas has to go in. You can reduce the gas going in by lowering the temperature (overall or for a few hours at a time) or by improving the insulation.

[Velvet]

That's an interesting point. In my situation, it's only one that'd actually cost me more financially (potentially) - tropical fish tank. I found somethign that gives an estimate of cost/month based on wattage of thing, and hours run, which was quite handy. Need to establish what the heater wattage is of course, then it'll be a stab in the dark for how long it actually spends running.

The majority of the other critters wouldn't mind, but a couple of them would be prone to getting far more chilled than they should, and lack of 'freedom to roam' if I'm not there means they can't use exercise as a means of keeping body temp up. They don't have any form of heating, cos there's absolutely no way to stop it escaping straight into the room.

I'm still really looking for a fairly simple explanation of the energy used to heat a place from a colder temp, vs energy used to maintain a temp. Someone somewhere must have done the research for this, I just can't find it :-) I thought maybe there were lurking people here who had the physics etc knowledge to calculate this sort of thing, I was thinking it was something that people might know from sizing CH systems since you have to generate enough heat to actually get the place warmed up in the morning, but I think I'm realising that the efficiency isn't looked at in an overall sense like this but more as an 'insulate as much as possible then fit the most efficient boiler' - just seems there's a bit bit of the puzzle being missed out, which is how to run the CH most efficiently.

Off to google again for more physics based stuff I think!

Velvet

[Velvet]

Ok, but what about once the house (fabric of) has cooled. Surely you end up pumping more energy/heat into it, some of which leaks out, some of which is absorbed by the house fabric, and only then do you start actually raising the temperature of the air within (once the house has stopped sucking heat out of the air).

That's the crux of the question for me. The longer the heating is off, the further it cools, and the more energy you need to put in to raise the temperature again. It's the comparison between energy used to raise it back to temp and energy used to maintain a temp that I wanted to find out about.

Velvet

[Christian McArdle]

The efficiency will be higher letting it cool. The only effect that will reduce efficiency compared to the alternative as that things get overheated in an attempt to reheat the house to a comfortable level too quickly. This is only likely to happen if the house is high thermal mass (i.e. thick stone walls) and the cycling frequent. It would actually require a pretty pathological set of parameters to make it less efficient, too.

Higher insulation would tend to reduce the heat losses, keeping the fabric warmer for longer. This will reduce the differences between the two methods.

Christian.

[The Natural Philosopher]

There is tho.

In order to warm up the house fully - structure and all - you have to actually make it hotter inside for a time. This *partially* offsets the benefits of timed running.

With UF heating, I have found it simpler to run it 24 hours for the last month. A complex predictive timeclock and stat is probably better, but I haven't yet installed one.

The fuel burn rate seems very little different from running it timed.

Thsi takes no account of thermal interia. If teh time constant of the building is anywhere over a few hours, it will stay warm and decay agfter heating is off - a net gain for you, but rewquire a huge burn first thing on switch on to restore. And take several hours to get fully warm. By understanding this you can adjust your timeclock accprdingly, but the option of switching off, and immmediately ceasing to lose heat as the inside temperature plummets to zero when you leave for work, followed by instant restoration to 22C or whatever the moment the boilre fires up is not possible in practice.

The better the insulation and the more thermal mass there is the less there is to gain from timed usage.

[The Natural Philosopher]

Heat required is the integral of temperature difference times the calculated U values over the surface areas of the house PLUS the amount of heat needed to raise the temperature of teh contained air mass and structure to the same level.

However this is a 'once and only' heat 'loss' since once warm. it takes no extra heat to keep it warm - its all about the insulation and air change losses. And if you like it gives it back to you by staying warm after the heating goes off.

The pertinent issue is whether you need to e.g. raise air temperature significanbtly MORE to feel confortable in a room with a stone cold concrete floor and icy thick exposed brickwork etc. etc.

The 'tmed' soulktion is most effective where heat loss is high and thermal mass low. There there is very little retention of heat, and teh internal temperature responds rapidly to CH heat input.

If insulation and mass is high, the system takes longer to respond if driven by the rated boler outpurt, since this will be lower anyway due to teh better insulation. In effect, it will take time to get up to temperarture, teh boiler will run at peak output for a lomngish initial period, and the rooms may need to be set hotter in order not to feel cold until it does.

There are so many variables that its very hard to predict, but having moved fom only moderately well nsulated house to one with big mases and good insulation, the oil burn rate is very little extra on 24 hour heating on the UF sections.

Upstairs where its studwork walls only, its timed, and these warm up very quickly anyway.

[The Natural Philosopher]

In theory iots always less to time, because the energy you put in to warm it up, keeps it warm after you have switched off teh heating. This of course is a net waste and leads you to advance the timing so teh house is warm when you get in and when you leave. In teh case of high enough thermal inertia:heat loss ratio this results in the system being on nearly 24 hours anyway.

So you might as well leave it on.

[The Natural Philosopher]

Yes. Experience shows that it is marginally less efficient to do 24 hrs, but not nearly as much as 'simple theory' predicts.

And teh iusses are as you say, thermal mass to insulation level ratios.

There is a lower limit on thermal mass - and that is the enclosed air volume of the house.

As insulation levels rise to the point where heat input is drastically reduced, the rate of cooling of that air mass makes it almost silly to stop heating it. Air change factors start to dominate the loss equations anyway.

There are side issues to do with whether boilers are more or less efficient dong long burns to heat up big storage elemnents, or lots of short blasts to keep a fast responding house 'up to temperature'

So it is complex.

Whilst timed CH in older less well insulated houses of lightweight constrctions is definitely a huge saveing, its by no means clear ij teh fcvase of modern well insulkated stuff.

My sister, in Sweden, has no concept of time clocks. The CH is switched on in the autumn, and off in the late spring, and thermostats control the house temperature. Thats all triple glazed 8"" rockwool no cold bridges stuff.

[IMM]

If not a condensing boiler have the boiler stat full on, or near to. Don't go for night setback temps. If a condensing boiler have boiler stat as low as possible. Trial and error if not auto modulation controls.