Jump to content


ebuild is sad to announce its closure - it has become too time and resource intensive to develop, manage and maintain.

However, ebuild will remain on-line in archive mode (ie no posting facilties) for several weeks so that users can use it as an information resource.

Barn Conversion, Stone Wall Uvalues, Thermal Detail Questions+


  • Please log in to reply
10 replies to this topic

#1 dc08

dc08

    Member

  • Members
  • PipPip
  • 40 posts

Posted 07 February 2014 - 08:36 AM

Hi it has been a while( 2008) since I built a barn conversion but I now have three to do this year, two for myself and one for a client. We are structural carpentry contractors and developers predominantly new build timber frame/oak frame. Although we have done many barn conversions in the 90s I feel a bit out of touch in this area. I have instructed my architect to produce BC drawings but want to make sure for myself that we are using the latest technology when it come to retaining as much internal space as possible. I was wondering if anyone on this site may have had some thermal heat loss or U value calcs done on existing stone walls (idealy blue lias rubble fill 450 wide) . obviously I want to keep my loss of internal space to a minimum and will maximise insulation in the roof and floors to off set against a thinner profile on the walls. I want a minimum of 25mm cavity and flexibility to follow wall contours ( to a degree just to retain at least some character) I know the steel systems offered by knauf etc but they look expensive. I am guessing that even the steel will need over boarding with celotex to prevent cold bridging at the studs. In out timber frames we always use a reflective VLC but this does little to effect U values. Any recommendations gratefully received. I am aiming for a 100mm finish including a small cavity for ventilation. I know the mortgage companies will insist I inject the walls but I still like to retain a cavity to ventilate between the wall and the internal studwork.

#2 temp

temp

    Advanced Member

  • Moderators
  • 10,200 posts

Posted 07 February 2014 - 09:33 AM

Hi.

So the wall will be something like..

Stone/rubble
25m cavity ventilated to the outside
?? Insulation between studs
Vapour barrier
Plasterboard

The two issues I see at the available thickness of insulation and the risk of the ventilation penetrating gaps around the insulation.



#3 Nickfromwales

Nickfromwales

    Short cuts take three times longer.....Fact

  • Moderators
  • 8,182 posts
  • LocationSouth Wales

Posted 07 February 2014 - 10:36 AM

Hi,
I think I'd rather sacrifice the extra 50mm and get more insulation in there. What's your chosen method of providing heating?
Regards, nick.

#4 jsharris

jsharris

    Please ignore all posts by me, some are erroneous

  • Member Blogger
  • PipPipPipPip
  • 11,461 posts
  • LocationWiltshire/Dorset border

Posted 07 February 2014 - 11:04 AM

The U value of a 450mm rubble stone wall will probably be around 2 to 3 W/m².K ( λ value is around 1.2 W/m.K for sandstone, and varies from maybe 0.6 W/m.K for packed chalk to perhaps as high as 1.6 W/m.K or more for granite).

To calculate the total U value of the build up, find out the λ value for each material, calculate the R value (where R value = thickness (m) / λ value) and then add up all the R values. Take the reciprocal of the answer (1 / R value) and you have the basic U value for the wall (ignoring the surface effects).

Here are some example λ values:

Stone: varies from about 0.6 W/m.K to maybe as high as 1.6 W/m.K
Concrete: typically about 1.2 W/mK
Plasterboard: 0.22 W/m.K
PIR foam insulation ("Kingspan"): 0.022 W/m.K
Rockwool insulation: 0.044 W/m.K
Low density EPS foam: 0.038 W/m.K
High density EPS foam: 0.034 W/m.K
PUR foam: 0.025 W/m.K
Dry timber: 0.15 W/m.K
OSB/3: 0.13 W/m.K
Warmcell insulation: 0.04 W/m.K
Steico wood fibre: 0.038 W/m.K
Silica aerogel board: 0.017 W/m.K
Silica carbon aerogel: 0.0135 W/m.K
Air in a 25mm unventilated cavity: 0.138 W/m.K

With the data above and knowing the thickness of each layer in the build up of the wall (or roof/floor) you can get a pretty good estimate of the total U value.

Edited by jsharris, 07 February 2014 - 12:44 PM.
typo


#5 joiner

joiner

    Advanced Member

  • Moderators
  • 10,718 posts
  • LocationWest Midlands

Posted 07 February 2014 - 12:14 PM

:) Just for completeness, Jeremy, what (if any - and why) "contingency factor" would you apply to the final figure?

#6 jsharris

jsharris

    Please ignore all posts by me, some are erroneous

  • Member Blogger
  • PipPipPipPip
  • 11,461 posts
  • LocationWiltshire/Dorset border

Posted 07 February 2014 - 12:42 PM

I wouldn't bother, because the true U value will always be slightly better than this calculation gives, because it takes no account of the surface resistance at each side of the wall/roof/floor. There are standard surface resistance figures that are used, but if you have a non-standard wall, for example, or are in a particularly sheltered, or exposed, location, then these may be well off. Just using the material build up means that you can be sure that the U value will always be a little bit pessimistic, which is no bad thing.

#7 joiner

joiner

    Advanced Member

  • Moderators
  • 10,718 posts
  • LocationWest Midlands

Posted 07 February 2014 - 02:38 PM

:) I think that's good enough. Cheers. ;)

#8 jsharris

jsharris

    Please ignore all posts by me, some are erroneous

  • Member Blogger
  • PipPipPipPip
  • 11,461 posts
  • LocationWiltshire/Dorset border

Posted 07 February 2014 - 03:15 PM

If you want to get a better approximation (but it will only be an approximation) you can use the following rough and ready surface resistance (R values) for specific conditions:

For a 5 deg C external wall with no external air movement (i.e. a windless day) then the surface resistance will vary from around 0.1 to 0.25 m².K/W depending on the surface emissivity (which is a function of the surface finish).

For a 5 deg C external wall with a 5m/S wind blowing (about 11 mph) then the surface resistance will vary from around 0.03 to around 0.04 m².K/W, depending on surface emissivity.

For a 20 deg C internal wall with no draughts, then the surface resistance to horizontal heat flow will be around 0.13 m².K/W to 0.4 m².K/W, depending on surface emissivity. Dark matt paint will tend to have a higher emissivity (lower R value) than shiny white paint, for example.

Change any of these parameters and the surface resistance changes quite markedly, particularly the "wind chill" effect of increased external wall heat loss resulting from accelerated convection with increasing wind speed, as the example given for just the effect of a 5m/S wind speed versus zero wind speed shows. The surface emissivity of the external walls and roof becomes far less significant as wind speed increases.

As a real-world worked example, here's the effect of surface emissivity on the overall U value for a wall that's made up from a 12.5mm layer of plasterboard, a 45mm closed service void, a layer of 12mm board, a 300mm layer of Warmcell, a 10mm outer OSB layer, a 25mm air gap and an outer skin of 19mm timber:

For just the wall build up with no correction for surface resistance, the total U value would be 0.122 W/m².K

On a still, no wind, day with the internal wall surface at 20 deg C and the external wall surface at 5 deg C, with the internal wall and external wall both having an average emissivity of 0.9, the total U value will be 0.1184 W/m².K

On a day with the wind blowing at 5m/S, same temperature and surface emissivity conditions as above, the total U value will be 0.1195 W/m².K

So, using just the material λ values and not bothering to calculate the surface emissivity or take account of surface temperatures or wind speed, results in U value that is pessimistic by around 3%, which is probably less than the variation that will result from manufacturing and fitting tolerances.

Edited by jsharris, 07 February 2014 - 03:20 PM.


#9 dc08

dc08

    Member

  • Members
  • PipPip
  • 40 posts

Posted 07 February 2014 - 03:42 PM

Thanks for the help guys. Some light reading to digest later tonight. I was hoping to use say a 25 mm cavity 40mm steel stud 40 celotex interstud with 20mm celotex over. VCL 12,5mm board. But I will need to put the numbers in and calculate from there. I your opinion do expensive reflective VCLs actually make a difference to uvalues. As often on the SAPS calcs they appear not to despite the cost. I have an interesting article from mortise and tennon mag (carpenters Fellowship) actually testing walls in situ ranging from straw/clay light earth @ 0.29W/m2K to brick infill @ 2.48W/m2K

#10 jsharris

jsharris

    Please ignore all posts by me, some are erroneous

  • Member Blogger
  • PipPipPipPip
  • 11,461 posts
  • LocationWiltshire/Dorset border

Posted 07 February 2014 - 04:05 PM

The reflective layers don't do anything useful internally, but can be useful at reducing solar gain if used under roof coverings, for example. This is why SAP ignores them, I think. The majority of heat transfer through walls is conductive, so adding a radiative heat barrier doesn't have any significant effect. It's the same reason that multifoils are snake oil, as they try to pretend that there is benefit in having multiple radiative heat barriers inside a structure.

There are some really good solid wall materials around, things like thick, lightweight cob can be pretty good, as can thick, dry, rammed chalk. We're keen to suggest using high tech foams and other insulation, but there are alternatives that, if made thick enough, are surprisingly effective.

A 12.5mm layer of plasterboard, with closed 25mm service void, 60mm of PIR foam and 450mm of stone will give you a U value of around 0.3 W/m².K (ignoring surface resistance on either side).

#11 temp

temp

    Advanced Member

  • Moderators
  • 10,200 posts

Posted 07 February 2014 - 04:34 PM

I was in a rush earlier.

I have a roof with >100mm celotex between the rafters. I'm disappointed with the way this is performing and believe this is possibly due to air getting through gaps between insulation and the rafters from the cold side. Part of the problem is that the sheets of Celotex had to be cut to fit between the rafters and with the best will in the world it's hard to cut Celotex accurately to fit between rafters/studs.

If you want a ventilated void I think it might be worth looking at..

Stone wall
50mm x 25mm treated battens to form 25mm void
Vapour permeable Membrane?
Combined insulation and plasterboard (something like Celotex PL4000)

This would allow the sheets of insulation/plasterboard to be butted together. The down side is no void for services. They would have to be cut into the plasterboard/insulation. There may also be a problem if the walls aren't very flat.