BTU Calcs - 2 methods, 2 different results - any thoughts?

[jsstevenson]

Hello...

I have been working out the heat loss for rooms in a new build (140mm timber frame with 140mm insulation, cavity, 100mm block and render) in order to specify correctly sized radiators but I am getting very different results depending on the method that I use.

Method 1 is where I calculate the heat loss for all surfaces based on the U-Values for the surfaces in question.

Method 2 is where I simply calculate the volume of the room, and make some basic corrections (e.g. north facing, french doors, number of external walls, etc).

The problem is, if I take Bedroom 1 (which is 3.8m x 4.8m with a ceiling height of 2.7m, 2 external walls, and set of french doors) and apply Method 1 I arrive at a BTU requirement of 2700 BTU's (791 Watts), and if I use Method 2 I arrive at 6405 BTU's. The results differ by a factor of over 2!

Given that Method 1 involves explicitly calculating heat loss of all surfaces and temperature differences between the inside and outside of these surfaces I would think method 1 is the most accurate method and Method 2 is just a crude estimate, but the difference seems substantial. But maybe my thinking is incorrect or I am missing something!

Any thoughts?

P.s. I have posted a spreadsheet online at the following url (just click at the top of the page to move between the two methods(sheets):
http://spreadsheets.google.com/pub?key=pmaiDN932ryiMM3UyaJfKVg

P.p.s....

www.trademate.co.uk heat loss calculator arrives at 3275 BTU's
www.plembcentre.co.uk heat loss calculator arrives at 1356.8 BTU's

 

[D_Hailsham]

I have been working out the heat loss for rooms in a new build (140mm timber frame with 140mm insulation, cavity, 100mm block and render)

If it is a new build, your U-values for the external envelope have to be no higher than given in current building regulations; i.e walls 0.35 W/m².K, floor 0.25Wm².K, roof 0.25Wm².K and doors/windows 2.2Wm².K. Some of your U-values are better than required but your floor and ceiling values are worse.

The U value of your window is very good; is it triple glazed, argon filled with K glass or something?

Looking at the bedroom in Method 1, I assume you are working with an outside temperature of -3°C and room temperature of 16°C, giving a difference of 19°C. But, assuming the bedroom is on the first floor, There would be a heat gain through the floor for the rooms downstairs (presumably at 19°C as you show a "ceiling difference" of -3°C) and a heat loss through the ceiling into the attic. But you have shown no heat loss through the ceiling. Have you got floor and ceiling transposed?

Method 2 is where I simply calculate the volume of the room, and make some basic corrections (e.g. north facing, French doors, number of external walls, etc).

How did you calculate the uncorrected BTU's (I wish you would work in watts ) and where did the correction factors come from?

... the difference seems substantial. But maybe my thinking is incorrect or I am missing something!

Or you assumptions in method 2 are way off the mark!

If you want good calculators, there are the Stelrad Stars and the Kermi

 

[gas4you]

Taking the 4.8m length as the external walls, using the Mears calculator for energy efficient dwellings, it gives 900 watts for a bedroom @18C.

This is slightly OTT as this mears calculator is designed for walls with a factor of 0.45, whereas yours, as said should be 0.35.

(I haven't got round to buying the most up to date one yet, I find the 3 different ones I have suffice well enough for me, as I rarely do new build)

 

[jsstevenson]

Thanks for your reply D_Hailsham...

You are right to pick up on the apparantly poor u-values of the floor and ceiling. I should have said, the house is three floors with the room that I have used, Bedroom 1, being on the 1st floor and as such it has a heated room above it and below it, thus the ceiling of the room is an intermediate ceiling and the floor is an intermediate floor hence the apparantly poor uvalues. Simply put the room in question (Bedroom 1) has a heated room above and a heated room below. The u-value for the Ground Floor is 0.2 and the u-value for the roof is 0.2. Thus bedroom 1 has a heated bedroom above at the same temp thus for calculation purposes I have assumed no heat loss through the ceiling, however the heated room below (Sitting Room) is designed to be 3 degrees C higher than Bedroom 1 thus the 'negative' BTU value as heat is actually gained through the floor rather than lost.

By memory the windows are argon filled K glass though I seem to think they are only double glazed! Now that you have mentioned it I will need to check!!!

I have actually been basing my design temperatures on -1C outside and 18C inside for bedrooms and 21C for Sitting Room thus temp difference of 19C as you said yourself for bedrooms in relation to outside and 3C difference between Sitting Room and Bedrooms.

I calculated the "uncorrected BTU's" by multiply the volume of the room by 153!!!!! Now where did I get that value (153) from? To be honest I cannot remember! I suppose that doesn't help me be 'sure' of the btu value arrived at. The correction factors were from the same online source that I can't remember and equally cannot verify as accurate!

The more I look at things the more I see Method 2 as simply filled with assumptions that I cannot qualify whereas Method 1 seems more and more precise the more I work with it and cross check and cross reference it.

Working in Watts... I worked in Watts and BTU's for Method 1 but for some reason not Method 2!

I will have a look at the calculators you mentioned.

Are you aware of any correction factors that might be worthwhile applying to Method 1?

Again, thanks for your reply.

Jamie.

 

[jsstevenson]

Thanks gas4you...

The 4.8m wall and one of the 3.8m walls are external with the external 3.8m wall having a french door in it. The door is 1.8m wide by 2.3m high thus an area of 4.14m2.

I think one of the things that has been making me question the results of Method 1 is that I am used to living in old tenament buildings with 3m high ceilings and am used to notably larger radiators to heat these rooms, as such it is a bit of a shock to see how small the radiators can be for similiar sized rooms in a new build due to the improved insulation and resistance to heat loss. Or so it seems, to me at least as I would have expected much larger radiators than being suggested by Method 1. Then again I am learning this as I go so I shouldn't be surprised when my assumptions are wrong!

Based on your calculation of 900 watts and the fact that it might be over the top it is looking more and more likely that Method 1 might be acceptable to specify my radiator sizes from.

Thanks for your reply.

Jamie.

 

[snb]

using the latest heating and hot water council program it gave about 735watts corrected to 835 for fast heat recovery.

However this is for flow at 82 degrees.

if you wanted the flow temp to be lower e.g 65 degrees then the adjustment would be to around 1300 watts

 

[jsstevenson]

Thanks snb...

Does that mean that if I install larger radiators that I can then run the overall system and boiler at a lower temperature thus optimising as far as possible the efficiency of the boiler? As I understand it a combi boiler is more efficent (or its efficiency is more utilised) with Underfloor heating which can be run circa 50-65C instead of the higher 85C that I might run radiators at.

Equally does this mean that if you have a radiator of a given size with a specified output that the output is based on the assumption that the flow temperature is 85 degrees C? Therfore if I knew that I was going to reduce the flow temperature to the 65 degrees C that you mentioned that I would then manually recalulate the wattage/BTU of a higher rated radiator to confirm that it outputs my required wattage/BTU at the lower temperature?

Assuming this is correct, would this also produce a more 'gentle' heat?

Thanks for your reply.

Jamie.

p.s. I take it that "fast heat recovery" is simply the increasing of the radiator size/wattage requirement in order to have the ability to quickly boost the temperature of the room from cold?

 

[D_Hailsham]

I should have said, the house is three floors with the room that I have used, Bedroom 1, being on the 1st floor and as such it has a heated room above it and below it, thus the ceiling of the room is an intermediate ceiling and the floor is an intermediate floor hence the apparantly poor uvalues.

That explains the anomalies!

By memory the windows are argon filled K glass though I seem to think they are only double glazed! Now that you have mentioned it I will need to check!!!

Pilkington K glass has a U-value of 1.5, so that explains your figures.

I have actually been basing my design temperatures on -1C outside and 18C inside for bedrooms and 21C for Sitting Room.

I had assumed that -3C meant the outside temperature!

Now where did I get that value (153) from? To be honest I cannot remember! I suppose that doesn't help me be 'sure' of the btu value arrived at. The correction factors were from the same online source that I can't remember and equally cannot verify as accurate!

Reminds me of the old rule of thumb that you multiply the room volume by 3 or 4 or 5, depending on the type of room to bet the required rad size - heat loss was much too complicated for the "heating engineer"!

Are you aware of any correction factors that might be worthwhile applying to Method 1?

None of the calculators available take Solar gain into account - for that you need a SAP calculator; but they cost £2-300!

 

[snb]

whilst few of the calculators take into account solar gain this of little importance during winter with 16 hours of darkness.

I have used my current version heat loss program with new builds and the results for heat loss have all been within 5% of the sap results.

 

[jsstevenson]

Thanks D_Hailsham, gas4you, and snb...

Writting my questions/thoughts and reading your comments and questions has pretty much sorted my concerns about the results I was arriving at and I now feel quite trusting of the results I now have!

D_Hailsham... I found where the magic "multiply by 153" to get the uncorrected BTU's value came from... it was from the wickes website, specifically the information brochure (http://www.wickes.co.uk/pcat/49centralheating) helping to choose radiator sizes. This is also where the correction values came from! I'm glad I have spent time learning about heat loss as based on the Wickes material I might have ended up being baked!

Again, thanks.
Jamie.

 

[doitall]

I'm glad I have spent time learning about heat loss as based on the Wickes material I might have ended up being baked!

Again, thanks.
Jamie.

You may end up freezing as well.

TRV's etc should control the final temperature.

If you calculate at -1 I would add 10% to allow for -5 plus days.

Just a personal thing, this year for example we had 6 weeks continuous with temperatures below your calculations.

 

[D_Hailsham]

D_Hailsham... I found where the magic "multiply by 153" to get the uncorrected BTU's value came from... it was from the wickes website

I gave up on Wickes, apart from things like sand etc as everything they sell is branded Wickes, so you don't know whether it is any good or not.

I wonder how many other unsuspecting customers have bought rads which are way over-sized.

One final comment. The calculators all increase rad sizes by 10-15%, to allow for fast heating up from cold. All rads outputs are quoted for a flow temp of 75C and return of 65C - a 10C difference, but boilers work best with a temperature difference of about 20C. This means that rad out puts are about 15% lower than specified. so you really need to have rads which are 15 % larger than the calculator says.

 

[David937]

I wonder how many other unsuspecting customers have bought rads which are way over-sized.

Aren't over-sized rads supposed to be better?

 

[D_Hailsham]

Aren't over-sized rads supposed to be better?

1. Only if the boiler works with a 20C differential.
2. The oversize has to be calculated not guessed.

Just taking Bedroom 1 as an example, the detailed calculation gave a heat loss of 2700 BTU, the Wickes ready reckoner gave 6400 BTU. That's not 10-15% oversize but 137% oversize.

If all rooms are oversized by that amount, not only will you have wasted money on larger radiators than necessary but you will have bought a boiler which is more than double the size you really need - I am thinking of heat-only boilers, not combis.

 

[David937]

Now it makes sense, thanks.