Anybody got any figures for these (and ideally pointers to sources of such figures)? I can't find them in Part L or BS5449(1990). The BS has values for floors with Bare boards and "Parquet, lino, rubber" (rubber?!): I wouldn't have expected them to have laminate in 1990, but I'm sure we had carpets even back in those days!
I know I could calculate the U of the flat roof from its component materials - I'm just being lazy! I am surprised they don't include it though.
There are tables in the help files of the Myson and the Barlo heatloss calculating software. If you want the Myson one (their web site is under construction) I can mail it to you. There is also a good table and sets of numbers the HVCA Central Heating Design Guide.
For the roof, the main factor is going to be the amount of insulation.
Figures I have are 2.04 for uninsulated, 0.80 for 25mm and 0.54 for
50mm. Probably after that, the figure is going to tend towards the U value of the insulation and the rest becomes almost irrelevant.
For the floor, you have to take into account the perimeter/area factor if it is a ground floor so there are tables of numbers rather than single figures.
In any case teh tables are not hugely accurate. In my case they are probabvly accurate for a floor that is unvented underneath, but when the east wind blows and the stupid an unncessary underfloor vents (its a suspended block and beam floor) let oit all in, it drops in temperature a LOT even WITH insulation.
Anyone know what a rotless suspebed concrete floor has to have underfloor vents? The Building inspector muttered something about letting flood water out...Oh well. At least it allows me to pick up air for open fires by punching down a foot into the void...
I suppose you could then treat it as though the entire floor was exposed to outside temperatures (i.e. like a wall) and ignore the periphery effect. For the purposes of a heating requirement calculation, and assuming no draught through the floor adding to air changes, this would give a conservative figure (i.e. worse heat loss than it probably is).
Do they have four toes? I know that there's strange tales from those parts..... :-)
You should have built the floor stronger....... ;-)
It has a bunch of other useful data as well such as pipe sizing, radiator derating, pressure vessel sizing. I found it useful to have everything in one place.
Certainly at 100mm and up figures get quite close to the material's U value. In reality the loss will be a shade less, but that goes in the right direction for a heating design.
For a boarded floor (non T&G), I suppose one should consider the draught proofing effect of the carpet as well, but that is pretty hard to assess.....
---8 >> For the floor, you have to take into account the perimeter/area factor
Well it's all pretty approximate (unless you get into the sort of detail in Part L1) so I'm surprised they haven't at least included some average-ish sort of figures e.g. cheap'n'nasty rubber-backed | deluxe thick pile on cushioned underlay.
I see Barlo's help file also refers to the BS and shows the same table (as also in the Barlo) but then goes on to describe the P/A method. However their calcs seem to just use the BS table - no P/A.
I suppose for uninsulated suspended floors the heat loss through the floor vastly exceeds the loss round the peripheries so you can in practice pretty much ignore P/A - ?
This is basically what TNP said. For a suspended block floor, you could effectively treat it as a wall on the basis that the outside temperature is across the entire underside. This would give a worse heat loss calculation than using the P/A factor, but for a heating design will mean that you err on the side of more heat provision.
For timber floors, possibly the U values published assume no P/A anyway.
Obviously it's a different situation if you are trying to demonstrate that a new house construction falls within energy consumption guidelines as opposed to a heating design for an existing property where you want to make sure that there is adequate heat provisioning. Generally on a property without cavity insulation, the largest heat losses tend to be through the walls anyway, and the floors about third on the pecking order.
Another factor is how does one treat party walls? Should it be to assume that next door has no heating and that therefore the party wall is effectively external, or that next door has heating and that therefore there is a net zero transfer - or something in between? If you consider a typical terraced property, they are often three times as deep as wide, so the party wall areas are substantial in comparison to the external.
Yes, that is more or less my subjective experience.
However raised woden floors with underfloor ventilation - good for rot - can be extremely draughty. Even with carpets. I acheved considerable gains by hardboarding such a floor and actually laying vinyl and a non-fixed carpet.
If airflow beneath the floor is restricted, is hugely warmer. The air is locked, and very little transfer to the soil takes place, and, over a period, the soil under the house becomes warm anyway. 2-3m of wet soil is still an effective insulator.
Ther are two aspects to heat calculations that need to be done.
Worst case - how much power do you need running flat out to heat a house on the three really cold days a year. That determines boiler and radiator sizes.
Average losses - which dictate the overall annual heating cost.
I think my house seems to be rated as needing 12Kw to fully heat it but the average yearly bill would reflect about 1.2kw averaged over the year. hether a 10:1 peak to mean ratio is typical I do not know, but its a benchmark anyway.
Finally, when calaculating heat reqiremenst, practical limitations often outweigh theory. My UF heating for example could be laid on one of two grids only - I picked the looser grid, which was supposed to give about
50w/sq m, and this has proved entirely adequate. Corresponding to about
1.5Kw per large room (6x5 meters). It seems small but it works. Likewise in wet radiator territory, how much psacehave you got?
Over engineering the heating system gives you rapid wram up, and, provided you have proper temperature control, I do not think a larger boiler operating on 20:1 duty cycle is MUCH less efficient than a tiddler working on 5:1.
Additionally, if CH pipes are routed as much as possible inside the insulation barrers, the losses from them are not overall losses, but serve to het the house anyway. One is left with overall heat leakage and the hot flue gases as the major heat losses from the house.
Returning to the original posters question, if the floor has underfloor draughts, I would seal it with hardboard, and mastic around the skirting, and assume a U value of about 5 for the wood floor, and then calculate the carpet as somehwat similar to rockwool and get the U value from rockwool conductivity over the thickness of both carpet an underlay.
The HVCA guide reckons you should treat adjacent property as unheated, but then half the U value for the wall because the adjacent property probably will be heated.
I think this rather bizarre alternative to using the real U value and changing the temperature drop is just because that's the way their example worksheet does things. (It does a single temperature drop multiplcation at the end of everything.)
Having just done a complete heat-loss for my house, and realised that the fabric loss was about 55% of the total, with the vent loss being 45%, it does seem that there's a slightly misplaced obsession with exact 'U' values when vent rates are basically picked from a list of 1,1.5,2 or 3.
For example, disregarding all the windows (i.e. treating them as wall) in this house makes about 5% difference to the total loss. As you're probably going to be adding 10-20% of hand-wavy 'distribution loss' and 'intermittent heating factors' at the end of the whole process, I did finish up wondering why I'd bothered measuring everything so carefully.
JOOI, what do other people who've done detailed heat-loss calcs come up with for vent vs. fabric loss?
When I moved in to a 1900 house 18 years ago, you could see where the drafts came through the floor as it had left a dirty mark on the carpet. In this case, it was all round the edge of the room where there was a 1/4" gap between the bottom of the skirting and the floorboards. After that carpet had been ripped out, I repainted the skirting, and deliberately spilled over 2" onto the floorboards too. When the paint was dry, I bought a very wide roll of good quality sellotape, and stuck it carefully along the gap between the skirting and the floor, rising to about 20mm above the floor so it would be hidden by the new carpet which was subsequently installed.
18 years on, that's still doing fine and has to this day stopped any draft coming up through the edge of the carpet, at least to the extent that there is no discolouration caused by the carpet filtering dust out as the previous carpet had done.
I'm sure it is. I think that you have to look at the circumstances and come to a decision based on what you want to achieve.
For example, in one house that we had, an old lady lived next door and we knew that she only heated one room. I therefore treated that party wall as being external.
In a more rigorous calculation, one should also account for internal heat transfers if the rooms are maintained at different temperatures. In the same house, I had a large bathroom which had a wall adjoining a bedroom. THe bathroom design temperature was 23, while the bedroom was
In that scenario, the transfer was significant. The radiator manufacturer programs take heatlosses via this route into account but not usually heat gains - presumably because you'll buy a larger radiator.
Like any other engineering exercise, it is important to look at what are the significant factors and what are not.
To some extent, you have control over the rate of air change in that you can go round and hermetically seal the place if you're stupid.
Fabric heat losses may be less easily controlled.
For example, in older properties with solid walls, the heat loss through them is substantial, yet people fuss about whether there is
100mm or 200mm of roof insulation.
By doing the sums, you gain an appreciation of what the factors actually are, what is important and what is not and how you are going to deal with them - i.e. with insulation or more heat.
Depending on the property, anything from 25%/75% to 75%/25%.
Yes, it was interesting - having done it in Excel rather than one of the radiator mfg's programs, it's particularly easy to look at what the effect of double glazing or cavity wall insulation is.
Sure. Along that lines, are you aware of any standard methods of converting heat loss rates into annual heating costs? (Or at least, total annual energy consumption.)
There must be some rules of thumb used to calculate pay-back times on additional insulation, for example.
Obviously, there's a regional variation, but I have got rather stuck at 'degree day' figures, which don't seem to mesh well with 'U x area' figures.
Interesting. Maybe I should go and measure a brand-new house, to see what that works out at...
I did it once using tables of daily temperatures sent to me by the Met Office. I assumed two types of day - weekend and weekday, and then made actual measurements of gas use by my boiler. I then factored for the other days based on the temperature. In the end I was about
5% high on my estimate,
You could try making contact with the people who do the SEDBUK work. They will likely have a better model to use. Whether that is based on measurements, mathematical modelling or both I am not sure.
You can do that reasonably easily based on looking at the element involved. One approach is to take monthly average temperatures and compute the heat loss through the surface for the month and then the energy used based on your heating patterns. Then you just add the numbers.
Last time I did this, I remember that cavity insulation looks pretty good on an ROI basis, double glazing rather less so.
In a multi-zone system, don't forget that you might choose to have an adjacent zone switched off, which could radically change the temperature differential.
Normally, a good amount of the upstairs heat actually comes through the ceiling/floor from downstairs. However, this takes time from a standing start (i.e. at least the time for the downstairs to warm up), so if you want to be able to get the upstairs up to temperature without having to wait for the downstairs to first heat up, you should probably assume the floor of the upstairs is rather more like a suspended downstairs floor.
These points are probably most relevant where you have a zoned heating system controlled by occupancy where rapid response is required, rather than just a heating system blindly switched by a timeswitch with no idea if or when or which parts of a house are occupied.
This can be easily sorted with heat recovery ventilation, though.
Also, at least you gain something through ventilation losses, as you get nice fresh air. Losing heat through the walls or roof gains you nothing other than damp insulation if you don't have a vapour barrier.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.