Thermal conductivity improving with insulation thickness!!

sas

Noticed on a technical data sheet for a rock wool product that they claim an improved thermal conductivity above a certain thickness.

Can anyone shed any light on this claim?

soldsold

Typo?

sas

Nope. It's all over the lions documentation for their kool products too.

sinnerboy

This is why it is vital to look for certification - from all of them

Rockwool - lambda 0.04 here

Kingspan - variable lambda here

Untill recently thermal conductivities were based on lab test measuring heat flow through a cubic meter of insulation .

That value is then divided by the board or quilt layer thickness to get Resistance - in turn inverted for the U Value

Recent research has found that thickness is a factor ( direction of heat loss is too - insulation is more conductive losing heat upwards )

So for the purposes of design to comply with building control legislation we must follow building regs which in turn lead us to NSAI / BBA certs .

Building Regulation Technical Guidance Documents are a curious hybrid of science and law . The science often moves on leaving the law part behind .

The Rockwool cert is from 2005 the Kingspan one from 2009 .

So maybe expect a revised Rockwool cert soon

GreenTech

I wondered the same thing for ages and I haven't come across anything substantial to help me, although from what i have read on the scientific side of it conductivities and heat transfer the it does make sense

My theory: If you take 25mm of insulation on its own and heat the air to 20degrees on one side and cool to 0degrees on the other side then the temperature difference across 25mm is -20degrees. If you do the same with 80mm the temperature difference is still the same but across a wider cross section. The material itself is affected slightly in these 2 scenarios because every milimeter of insulation is actually insulating the previous millimeter and therefore raising its temperature. Because the temperature difference is more spread out across a wider element it means that the rate of heat loss improves very slightly. Its like the more insulation you add the more it compounds the value of the insulation before it. I hope I havent confused the issue even more.... hard to explain but its one of those things that a digram would explain very quickly...

sinnerboy

I think you are probably spot on there GreenTech .

sas

Glad I asked now. The responses make logical sense to me.

Thanks.

soldsold

So to summarise in laymans terms: if you add more insulation the insulation you add isn't quite as effective as the first bit, but still improves the insulation substantially to lower the heat loss? Ie the law of diminishing returns? (just in case anyone thinks less insulation keeps the heat in better)

sinnerboy

No ! Your understanding it in reverse .

Imagine two insulation boards - one 50mm thick and one 100mm thick .

Now imagine a 25mm strip along the centerline of each board .

The 25mm strip in the middle of the 100mm board is more insulative then the 25mm strip in the middle of the 50mm board .

When you use thicker insulation you get two bites of the cherry i.e. more insulation which in turns is insulating at a better rate .

sydthebeat

sinnerboy wrote: »

No ! Your understanding it in reverse .

Imagine two insulation boards - one 50mm thick and one 100mm thick .

Now imagine a 25mm strip along the centerline of each board .

The 25mm strip in the middle of the 100mm board is more insulative then the 25mm strip in the middle of the 50mm board .

When you use thicker insulation you get two bites of the cherry i.e. more insulation which in turns is insulating at a better rate .

is it not a more direct co-relation between thickness and conductivity??

ie if theres a 20 deg diffeence between the internal and external of a 50mm insulation, then that heat has 50mm of insulation to pass through before dissipating.

if theres 100mm insulation then theres and extra 50mm to pass through, therefore the extra 'lag' and therefore the increased condictivity.... as its measured in w/m2k...

i suppose is it more unusual NOT to see TC values decrease as thickness increases??

sinnerboy

Not quite direct , as the "relationship between the thermal conductivity (k) and optimum thickness (xopt) of insulation material is non-linear which obeys a polynomial function of xopt = a + bk + ck2, where a = 0.0818, b = −2.973, and c = 64.6."

According to this

I suppose that whilst we were only ever using modest thicknesses of insulation , that it was acceptable practice to use a static vale for TC . Now , at the dawn of Passiv House in Ireland to be triggered by this we need to employ better calculation methods


.

Mellor

sydthebeat wrote: »

is it not a more direct co-relation between thickness and conductivity??

ie if theres a 20 deg diffeence between the internal and external of a 50mm insulation, then that heat has 50mm of insulation to pass through before dissipating.

if theres 100mm insulation then theres and extra 50mm to pass through, therefore the extra 'lag' and therefore the increased condictivity.... as its measured in w/m2k...

i suppose is it more unusual NOT to see TC values decrease as thickness increases??

this post, and the previous one is simply the effect of having extra insulation, and would explain decreased conductivity, only increased resistance.

Basically, thermal conductivity is dynamic. Because of the narrrow range of temps in natural environments, an average was ok to use and didn't create much of an issue or error. But as we go bigger and bigger, the error is compounded.

The rate of the improvement differs between materials. I suppose the easiest example would be rockwool or fibre glass.
RW and FG insulations basically use fibres to trap air. Consider the surface of this materials. One would assume that, close to the surface the air, is free to move away and its only a few mm in that it is fully trapped.
Lets say, to keep it simply, that the outer 5mm is only 50% efficient. (made up figures to highlight the issue). so every sheet has 10mm @ 50%

So a 50mm section is 40mm*100% + 10mm*50%, which totals 90% efficiency total.
100mm = 90*100% + 10*50%, which is 95% total
200mm = 190*100% + 10*50%, which is 97.5% total

Similarly, rigid boards aren't of always of uniform density and thicker boards more less dense mass improving TC.
These are jsut two non uniform properties of insulation that might affect TC, also, the thickness for optimum TC may peak, and above that the TC may reduce again. But even though, boards above this are still better insulators

readytostart

In reading your posts I assume a TF construction with 200mm of fibre glass insulation included would be a good construction.Am I wrong in this assumption?
Does anybody know the life span of fibre glass?
Would a TF house with blown in cellulose be a better job.

beyondpassive

@Sinnerboy

I always assumed the incrementally reducing TC was as a direct effect result of the diminished effect of the foil faces as a proportion of the total thickness of the insulant. Can this be ruled out?

sas

beyondpassive wrote: »

@Sinnerboy

I always assumed the incrementally reducing TC was as a direct effect result of the diminished effect of the foil faces as a proportion of the total thickness of the insulant. Can this be ruled out?

How does this explain the rockwool case?

GreenTech

Hi Readytostart,

Your question about celulose vs mineral wool is interesting. This all depends on the conductivity of the insulation and how its installed. Mineral wool has been known to slump over time leaving large spaces uninsulated at the top of your walls after a few years. There are some mineral wool insulation products on the market at the moment which claim conductivities as low as 0.032w/mk which would actually be superior to celulose which is generally around 0.040w/mk. The advantage to using celulose is that it is pumped in at pressure from the inside once the internal OSB boards have been fixed. Pumping it in at pressure means that slumping of insulation is virtually eliminated due to a tighter fit. In my opinion this is the right way to go. Don't forget to make sure the panels are breathable! Avoid OSB or plywood on the outside of the studs as these are not tested as breathable products... wood fibre boards or a similar breathable product is always a good choice.

galwaytt

GreenTech wrote: »

Hi Readytostart,

Your question about celulose vs mineral wool is interesting. This all depends on the conductivity of the insulation and how its installed. Mineral wool has been known to slump over time leaving large spaces uninsulated at the top of your walls after a few years. There are some mineral wool insulation products on the market at the moment which claim conductivities as low as 0.032w/mk which would actually be superior to celulose which is generally around 0.040w/mk. The advantage to using celulose is that it is pumped in at pressure from the inside once the internal OSB boards have been fixed. Pumping it in at pressure means that slumping of insulation is virtually eliminated due to a tighter fit. In my opinion this is the right way to go. Don't forget to make sure the panels are breathable! Avoid OSB or plywood on the outside of the studs as these are not tested as breathable products... wood fibre boards or a similar breathable product is always a good choice.

I'd take issue with the OSB comment - it's there for a structural racking purpose, for a start, in 'conventional' TF, so you won't be leaving that out.

However, in other timber construction, (where I work), there is no interstitial condensation in the panel in any case, and the OSB does good work. It also eliminates the need for VC and internal airtightness membrane's.

So, not all timber construction is the same, and you need to judge each on it's merit's.

On a general note, the XPS we use also has different declared values at different thickness', so the phenomenon isn't confined to loose-fill insulation, either.

GreenTech

Galwaytt,

This wasnt a personal attack on anyone so don't take it that way. In fact I don't think you read my post right at all. I'm aware of the fact that OSB is used for structural racking.... I also worked for a timber frame company for a number of years and now work in the architectural side of things. I have learned more about timber frame since I stopped working for a timber frame company as most companies only concentrate on one single method of construction. OSB is not generally recommended on outside of studs if you require a breathable construction and in my opinion all timber frames should be allowed to breath. The OSB can be fixed on the inside face of the studs to provide structural racking and a breathable board on outside to promote breathability.

Timber frame is generally not known for promoting interstitial condensation and I would agree that it is most likely not going to occur. Interstitial condensation is not the only form of moisture which occurs in construction and the principle of using a breathable board on the outside of the studs is to allow moisture to escape... whether the moisture comes through interstitial condensation or not.

You might explain one thing though.... you say OSB eliminates the need for Vapour control. Why then are you suggesting putting OSB on the outside of the studs??? Shouldn't the vapour control be positioned on the warm side of the insulation?.... unless you are sugessting insulating outside the timber frame?
If OSB is used for vapour control and air tightness it is usually on inside face of studs and a breathable board is used on the outer face of studs. I would be interested in seeing the reports your company carried out on interstitial condensation.