simple u-value calc, not matching reality!

tails_naf

Hi All,

Moved into the new house, and it is getting it's first test with the cold weather.
It's doing well...kind of, except for one thing

The problem is our main bedroom is fierce cold... now before you say 'wrong forum -> marriage advice is over there', it seems to be to do with heat loss - but much more than should be happening!

So the room is big - it has an arched ceiling, approx 4m high, and 4.3x5.3 in dimensions.

My rough calcs are as follows:
roof area = 32m2, u = 0.16 -> loss @ 20C differential is 98w
front wall area = 22m2, u=0.18 -> loss @ 20C is 80w
windows = 5.5m2, u=1.4 -> loss @ 20C is 150w (ouch!)
side-walls (exterior only) = 27m2, 0.18u -> loss at 20C is 100w

So total loss is supposed to be 360w, with 20C difference between inside and out.

Last night we had -6C outside, and inside the room it was 12C, so 18C differential. The room rads are not working yet, so we had a 1.2kwatt fan heater going - so this should have providied 3x the amount that would be required, not to mention some heat seeping in from room below (which is warm).

Note: There are no draughts.

My guess at this stage is some insulation was omitted during building, as it is a TF build, and insulation was provided pre-fitted in the framed wall, but with a spec of 0.18u for the walls.

Is there any other reason that anyone can think of why there would be a 3x difference between energy being supplied and what is needed 'on paper' ?

I'm thinking of getting thermal imaging done, but only if the numbers truly say its likely to find something...

braftery

Are the rads working in the rest of building ?
If not how are the other rooms being heated ?
Are all the other rooms and in particular the other bedrooms warm ?
Are the ceilings as high in the other rooms in the house ?

L driver

hello,
What type of insulation is in the walls?

tails_naf

braftery wrote: »

Are the rads working in the rest of building ?
If not how are the other rooms being heated ?
Are all the other rooms and in particular the other bedrooms warm ?
Are the ceilings as high in the other rooms in the house ?

We have underfloor down stairs, rads upstairs. rads are not on, but downstairs is. other upstairs rooms have no heat source (yet), but actually are not too cold due to airflow. this one bedroom is baltic tho!

ceilings on other rooms are not as high - standard height.

Insulation is rockwool type - 200mm in the timber frame, no other insulation in cavity, etc.

braftery

I am not expert in insulation ... so what i say below is purely my initial thoughts, and given the other bedroom are not cold (what temp are they achieving by the way ?) it might be not be useful ....


When we moved into our house, all underfloor heating, it took about 4 weeks for the whole building to settle down to standard temperatures .. I called the UF guys and we tweaked the therms until all the room temps stablised.

In order to really figure out if the room is losing heat at a higher rate than it should, I think you should first make sure you get the room upto working temperature (18-20deg). Then turn off the heat.

From this point you can see how fast the room is losing temp. The rate of loss will be easier to observe and is closer to real life usage.

After that maybe move to a thermal imaging camera ... and if that doesnt work I can give you the number of good exorcist

Scartbeg

Is your HRV up and running? If so, you will need to make sure the rooms you are extracting from are well heated (wetrooms/kitchen I assume?) - otherwise there will be insufficent energy passed to the incoming air. You would normally expect the incoming air to be ~ 3 deg less than the outgoing, so it could be chilly if one of your bathrooms is at say 15 deg as it's unused. Can you measure temp at one of the incoming vents?

I make your losses to be 430W from the figures given. But you would also have an extra 200W of heat input to the room from the body heat of two occupants (or higher, depending on your level of activity!)

Bear in mind that it takes energy to heat up the cold air in the room in the first place. About 1000J/Kg/C for air. Your room is about 90m3, so that's about 110kg of air. It will take 110,000Joules to heat it up by 1 degC. With a heater of 1200W , that should take 90 secs (assuming even circulation and no heat losses). As you have theoretical heat losses of 430W, the time take to heat the room by 1 degC rises to 2min 20 sec. Hot air will accumulate first at the ceiling so it would take time for this to percolate down - I'd still expect an overall temp rise in a reasonable time frame though.

My best guess is your HRV may be the issue.

Lantus

A few things: -

1/ very high ceilings would effect room as warm air rises and it's much harder to heat the occupied part of the room.
2/1.4 from glass is optomistic unless it's very high spec. If you have good DG then 2.0 is more probable. (taking into account the frame.)
3/You may say no draughts but you need to allow for infiltration. .33.N.V.(deltaT) if you have an air pressure test done then you know it's allright but if not........
4/Extract fans running on en-suite WC? this will pull air from the path of least resistance.

Mellor

Lantus wrote: »

A few things: -

1/ very high ceilings would effect room as warm air rises and it's much harder to heat the occupied part of the room.
2/1.4 from glass is optomistic unless it's very high spec. If you have good DG then 2.0 is more probable. (taking into account the frame.)
3/You may say no draughts but you need to allow for infiltration. .33.N.V.(deltaT) if you have an air pressure test done then you know it's allright but if not........
4/Extract fans running on en-suite WC? this will pull air from the path of least resistance.

it really is prob a lot more simple than people are making out.
The calculated U-values are basically minimums for your build up, in reality you con't achieve these . THe better the build, the closer you'll get.
There are lots of little small things that are unaccounted for, some have been mentioned already.
The heater isn't 100% effcient so its isn't providing 1.2kW, prob not even close.

Scartbeg

Mellor wrote: »

The heater isn't 100% effcient so its isn't providing 1.2kW, prob not even close.

Have to disagree with this - the filament in the heater is close to 100% efficient at converting electricity to heat ( a tiny amount as light if it is glowing). There's a negligible amount (<10W) used to drive the fan, and any frictional losses also result in heating. If the heater is using 1200W in electricity, I'd expect pretty much every drop of it to appear as heat.

WRT the U-values being theoretical, yes, the as-built values are unlikley to match what was planned. However even if you doubled the losses due to the walls etc, you should still be achieveing a net gain of 400W which would result in the room heating up over time.


FWIW, we rented an old farmhouse last winter. One bedroom had 3 external walls. With a large rad and 2kw oil heater going the temp was 4C inside when it was -10C outside! Might as well have slept in the barn!

beyondpassive

Tails,

I'd add more insulation if possible, this would necessitate taking off the slates or internal finishes. Not likely to be an option but heres the rational.

200mm rockwool in a vaulted ceiling isn't very much. The conductivity of rockwool is 0.037 at best. But thats just where the insulation is installed, where you have timber, its lamda is 0.140 lamda, so take 10% from your calc. Next factor in workmanship, take another 40%, cold bridge at eaves and ridge about 5%.

Another factor to consider is the air moving accross or through your insulation. The venting under the slates, cools the top of the insulation reducing its effectivness. Air leakage caused by bouyant pressure in the house caused by rising warm air or dropping cold air.

so in effect your insulation will perform at best half what is calculated. A good spec usually has softboard sarking board to keep the insulation warm and a services cavity inside for airtightness, both of these measures eliminate repeating and occasional thermal bridges and give you some certainty that you get what you pay for. This is why the passive approach is useful it gives that certainty.

Mellor

Scartbeg wrote: »

Have to disagree with this - the filament in the heater is close to 100% efficient at converting electricity to heat ( a tiny amount as light if it is glowing). There's a negligible amount (<10W) used to drive the fan, and any frictional losses also result in heating. If the heater is using 1200W in electricity, I'd expect pretty much every drop of it to appear as heat.

How can you say every drop i appear as heat, and mention two loses in the same paragraph.
The filament is good, but not 100%, 90-95% probably, the fan likely has more than 10W when turn to highest setting.
Plus, the heat may not actually be running at 1200W.

Simply put, I stand by my comments that the heater is putting out 1200w of heat. How much less we don't know, but its less.

tails_naf

Thanks guys, lots of very interesting answers here!

On the insulation in the pitched part - it is 180mm metac between rafters (600 centers) and 50mm insulated plaster-board on the inside.

I am surprised to hear there can be 40% loss in performance due to workmanship - can you elaborate on that?
I was around (and took part in) the installation of the insulation of the roof - left as few air gaps as we could, insulation nice and snug either side of rafters - the only thing I can think of is minor compression of the insulation width-wise, to fit between rafters snugly - do you think that alone can cause insulation's lambda to go up by 40%?

The TF walls are where I'm starting to zero in on as the problem - the laser temp sensor showed that a generic wall section was at 13C, but the section directly beneath one of the windows was 9C - so without thermal imaging (which I may still do), it looks like there is poorly fitted insulation behind there.

There is little chance of fitting more insulation now, as the room is finished inside. Would be a lot of re-work needed if I was say to slab the inside of the room with insulated slab...

I assume I would have some come-back on the TF company, who put on the contract that the wall spec was 0.18u - if I could show them that a given wall section was not achieving that, would they have to put it right?

Finally, maybe a bit pedantic, but on the heater, assuming it does take 1200w from the socket, according to the conservation of energy, 1200w must be used up, and almost everything we do ends up as heat (friction->heat, combustion->heat, electricity usage (in general)-> heat, combustion->heat, etc). the only waste I can see would be the sound, and the bit of light as already pointed out! Of course I could be totally wrong on that!

sydthebeat

tails the simplest way to explain workmanship in terms of u values is:

if thermal bridging detail sare not done correctly the assumption then is that there is a thermal bridging factor of 0.15 (0.08 IF done correctly)

so.... assuming you u values for walls roof etc are all standard minimums, then the average u value for all your house u values is around 0.25

then the thermal briging factor is added to this so 0.25 + 0.15 = an average u value of teh build of about 0.4.

so therefore thermal bridging (workmanship!!) accounts for about 40% of the heat loss in the example above . and i would reckon that about correct for most celtic tiger "slap em up" build.

tails_naf

sydthebeat wrote: »

tails the simplest way to explain workmanship in terms of u values is:

if thermal bridging detail sare not done correctly the assumption then is that there is a thermal bridging factor of 0.15 (0.08 IF done correctly)

so.... assuming you u values for walls roof etc are all standard minimums, then the average u value for all your house u values is around 0.25

then the thermal briging factor is added to this so 0.25 + 0.15 = an average u value of teh build of about 0.4.

so therefore thermal bridging (workmanship!!) accounts for about 40% of the heat loss in the example above . and i would reckon that about correct for most celtic tiger "slap em up" build.

Hi sydthebeat,

- this is a one-off house, self built. So not a slap-em-up, but I can only directly say the workmanship on the ceiling was good quality,the TF walls were insulated in the factory before delivery (plus windows air-tightness, etc).

- also, I was taking into account the 'cold-bridging' of the timber frame at 600 centers in my calculations - so what other sources of cold bridging can you forsee that would take the overall u down by 0.15.

- the 0.15 value is actually huge - it kind of makes the whole energy-efficient goal seem like a waste of time - i.e. I paid for walls that were designed to be 0.18u. If poor workmanship adds another 0.15 to this, it means the wall is about 1/2 as good in reality as it was designed. It Would make it seem worthless to go for 0.14u walls (which was an option), as the % difference is negligible when the 0.15 is added on!

- so would you say that the 1200w usage would not be unreasonable/is expected for the u-values I quoted that the building was supposed to achieve?


Also - a few other things I forgot to reply to other posters:
- the windows were specced with 1.0u glass, and 1.4 overall including the frame - so using a value of 2.0 sounds a lot (even though when that is used, it still is a long way from 1200w)

- the HRV is not running yet - but it's a good point that an intake in an 'unused' room could make the outlet air cold - I'll beware of that when I finally get it going.

- ufh is working well, it is definitely up at temperature, but is only downstairs.

- blower door test was done, it achieved 2.9, and as a result of that test a few places were patched - none were in the bedroom in question.

L driver

Hello,
was the roof made wind tight? How dense was the rockwool? Did any trade do any work on the room after the blower door test? It probably is a combination of things. It is a big volume and big glazing, and in my very humble opinion 1.4 on windows is not enough and 200mm rockwool in vaulted ceiling is probably a bit on the lean side. Crack on the rads, maybe the heater is a bit John Gormallyed.. sorry I mean frigged:rolleyes:

tails_naf

L driver wrote: »

Hello,
was the roof made wind tight? How dense was the rockwool? Did any trade do any work on the room after the blower door test? It probably is a combination of things. It is a big volume and big glazing, and in my very humble opinion 1.4 on windows is not enough and 200mm rockwool in vaulted ceiling is probably a bit on the lean side. Crack on the rads, maybe the heater is a bit John Gormallyed.. sorry I mean frigged:rolleyes:

Wind-tight sounds interesting - I don't think there was anything beyond the standard felt and battens and then slates. inside there is 50mm airgap, and 180mm metac (high density version of rockwool) and then 50mm insulated plasterboard on the inside.

does it sound like something is missing?

davidoco

My observations may explain the problem with planned uvalue and on the ground results from insulation. My comparison is lightweight (timber frame) and solid (block) construction.

I live in a bungalow (main living area) with a converted attic (playroom). I think of the downstairs as solid construction 300 mm cavity walls with 60 Polyiso and 300 mm fibreglass in 225 mm ceilings

I think of attic as timber frame as it is a room in roof 20 x 14 ft and has the insulated floor (ceiling from below) and 125 mm rafters with 100 mm rafter loc underlayed with 1" insulated polyiso board. 100 mm xxx in 1 m high side walls overlayed with same 1" insulated board. Two small north facing velux windows and sealed from rest of house on landing upstairs.

Below are my heating readings from today. It can be clearly seen that the upstairs lightweight construction while is relatively fast to heat up looses 8 degrees over a 6 hour period while the living area lost only 5 degrees over a 6 hour period.

Ext Temp here over the past 24 hours was a max of 0 and a low of –2.5

Attic playroom heating on for two hours at 1 pm today. temp now is 13 degrees.


Living room. Fire was lit last night. Temp was 21 at midnight.

tails_naf

davidoco wrote: »

My observations may explain the problem with planned uvalue and on the ground results from insulation. My comparison is lightweight (timber frame) and solid (block) construction.

I live in a bungalow (main living area) with a converted attic (playroom). I think of the downstairs as solid construction 300 mm cavity walls with 60 Polyiso and 300 mm fibreglass in 225 mm ceilings

I think of attic as timber frame as it is a room in roof 20 x 14 ft and has the insulated floor (ceiling from below) and 125 mm rafters with 100 mm rafter loc underlayed with 1" insulated polyiso board. 100 mm xxx in 1 m high side walls overlayed with same 1" insulated board. Two small north facing velux windows and sealed from rest of house on landing upstairs.

Below are my heating readings from today. It can be clearly seen that the upstairs lightweight construction while is relatively fast to heat up looses 8 degrees over a 6 hour period while the living area lost only 5 degrees over a 6 hour period.

Ext Temp here over the past 24 hours was a max of 0 and a low of –2.5

Attic playroom heating on for two hours at 1 pm today. temp now is 13 degrees.


Living room. Fire was lit last night. Temp was 21 at midnight.

Interesting - what did you use for the measurements of the temps - looks like a nice setup.

How does your energy loss calcs compare to the theoretical values? I suppose your 'solid construction' section looses nothing through the ceiling, which is an advantage.
i.e. do you have numbers for - living space - on paper heat-loss is x, in reality it is y, and do the same for the attic space?

I also came across this:
http://www.viking-house.ie/downloads/Wind%20Chill%20-%20Roof.pdf
it's a very very interesting doc on the effect of wind-tightness - and how it may decrease the effective u of insulation dramatically in roofs.

creedp

@Davidoco

Your set up is very similar to mine except I have a 1 1/2 storey that also has a good sized attic. I want to treat the attic in the same way you are, i.e. seperately from the rest of the house. I want to heat the ground and 1st floor as a unit and seal them from the attic using ceiling insulation and airtight membrane. I then want to insulate and slab the attic for occasional use and storage. One question, do you have an airtight membrane on your ceiling to help stop warm air entering your cooler attic. Also did you put a airtight membrane inside your attic insulation to stop the warm attic air escaping to the relatively cold rafter space and causing condensation on rafers/felt. Would you consider the use of a membrane in the attic to be worthwhile?

Sorry about all the questions, it just that I'd like to hear your views on this issue, particularly as you have a very similar set up to what I'm proposing

Thanks.

davidoco

The pictures are off a Heatmiser touchpad which is the heating controller - see heatmiser.co.uk

creedp wrote: »

One question, do you have an airtight membrane on your ceiling to help stop warm air entering your cooler attic.

No I purchased this house in 2006 and it has trickle vents and no HRV and hence no airtight membrane fitted. In any event it'sthe fibreglass in the ceiling that is stopping the heat reaching the attic. That is borne out by the fact that the temp in the attic does not change during a downstairs heating cycle once the attic has settled down to the 7 or 8 degree differential to the outside.

creedp wrote: »

Also did you put a airtight membrane inside your attic insulation to stop the warm attic air escaping to the relatively cold rafter space and causing condensation on rafers/felt. Would you consider the use of a membrane in the attic to be worthwhile?

as above no airtight membrane and not even a vapour layer as I was worried about putting that behind the 1" insulated backed board but of course the insulated 1" plaster board has a foil backing. I have ensured ventilation across the cold side of the insulation by using 100mm board and I have a void on either side of the room to assist air flow as shown in the image on this page http://www.sabsm.co.uk/sabsm/roofspace.htm

maddragon

Was the fan heater running non-stop or was it cutting in and out on the inbuilt thermostat. That could account for alot of the differential.

tails_naf

maddragon wrote: »

Was the fan heater running non-stop or was it cutting in and out on the inbuilt thermostat. That could account for alot of the differential.

Unfortunately it was running flat out.
I have since used a 1.8kw fan heater, and that runs flat out too.


One thing that is interesting - I was able to use the laser-temp-sensor to go over the whole wall surface slowly and try to do a poor-mans-thermal-imaging, and I did find a very very cold point - the bottom of one corner of the room right at the joint between floor and skirting was 4 degrees!
I'm guessing there is some insulation omitted here - my guess is either the bottom of the wall, or the void between floors (as it is timber frame).

While this is a weak-spot, could one weak spot like this cool down the whole room? I guess we would see some loss, but if I move 4-5 inches away in any direction the temp is back to 1-2 degrees below room temp, so the surface area of this weak spot is quite small...