They do, CWaters. Isn't that what ws described from B&Q? There are others with closed cell bubblewrap sandwiched between two foils.

No, Tuna, I don't see that the foil can conduct heat away - except for the few microns of its thickness. Not imortant. And the heat won't flow sideways as there will not be a heat gradient (except maybe right at the edge).

>Most tables tell us that an air gap 25mm+ gives a 0.2 R-value.
That presuposes that the air is not moving. It will be moving so it would be silly to use that figure. Convection is a good way of moving (losing) heat.

Thanks for Mylar link, David.

Biff: you asked why the 25mm gap each side of multifoil was required by the manufacturers and I gave my opinion. Whether the air space actually gives 0.2R is probably irrelevant. The fact it can be included to boost the claimed U-value is probably the important thing!

deleted.

I have read the whole of the first 148pp of the MF forum and also the 217 posts here - it is all extremely interesting.
HOWEVER I have still not found the answer to my own question(s) from the real world.
I have a 3-room loft in which will include a bathroom. I have stripped it right back to the rafters, which vary in depth from 3 inches to 2 and a half inches (i.e. not much depth for insulation). Much of the area to be insulated is the sloping underside of the roof, with only a small area of ceiling in the centre below the roof ridge.
The roof consists of slates with a bituminised felt layer directly beneath them. This layer seems to me to be a problem because it is not permeable except where pieces of the felt overlap allowing draughts between the rafters.
These overlaps are random, so that some of the spaces between the rafters are probably quite well ventilated, others less so, if at all.
What problems might I have with condensation? Would it be wise to leave the rafter voids untouched and attach 9mm plasterboard with 50mm PIR foam incorporating a foil vapour/radiation barrier under the rafters? - as suggested by funcrusher on 14.02.07?
The problem with this is the degree to which it would lower the head-height below the rafters. Can I save space by using Aerogel in some combination? What is the minimum depth of insulation that would be effective in preventing heat loss and cold downdraughts to the lower floors??
In short, what combination of foil, plasterboard, rigid or non-rigid insulation material, vapour control layer etc etc
does the team think would best deal with this situation, where the roof is lined with a non-permeable material and the rafter depth and the head-height is restricted? In addition the bituminised felt sags down a little into the between-rafter spaces here and there.
I would be extremely grateful for some advice.
Please let me know if I should put this post in some other part of the site. I happened to come to it via the Multifoils thread because I started out believing in these. (No longer the case).

Thank you Biff. That says it all. I am almost ready to give up and let my builders do as they will - but reluctant to do this without understanding. Loadsamoney at stake. Incidentally (worryingly) two of the rafters have rotted at the bottom of the slopes. This could perhaps be due to condensation running down behind the original lath and plaster lining (now removed), or to leaking roof before waterproof lining added. God knows.........

Incidentally, Building Control has said that they don't mind what we do on the slopes, since the roof structure has not been altered (and we are in a Conservation area). In the centre, where there is some ceiling, they stipulate 300mm Rockwool or similar. My builders have suggested "Acoustic wool"? or some such (it's cheaper). I don't mind spending money, as long as it is worthwhile. Feeling weak this a.m....

Kevin

Aerogel is worth it if your headroom is very constrained and you can afford the premium (it has also been tested in Biff's lab - which is good enough for me!). It is unfortunatley not yet mainstream enough for the price to have come down to competetive levels for normal humans. I have not persued a brief conversation with Proctors about it as a result of knowing pretty quickly that I could not afford to use it.

Today I am mostly wiring!

Happy Kiting

S.

I'm surprised they don't want specific insulation on the sloping parts. Perhaps they believe these rooms were already habitable rooms built ages ago when the regs were more relaxed.

How much head room can you afford to loose? For example could you put 1" rigid foam insulation between the rafters (leaving a 50mm air gap) and another 3-4" below the rafters? The important thing is to stop vapour from the rooms getting to the cold rafters so whatever insulation you use you must put a vapour barrier up before the plasterboard. Make sure that barrier forms a total seal, no holes, tape the joints etc.

Kevin

Not sure if this is too late for you or not. I suggest a bit of pragmatism based on my own practical experience. Like you, I have old oak rafters of differing depths and straightness and a sloping roof and with a non-breathable sarking felt (roof membrane). Like you, I wanted to minimise encroachment into the room space. And, again, like you I thought that multifoil was the answer. After a discussion with my local BCO it soon became apparent that multifoil by itself would not be acceptable. They would accept 60mm in between the rafters and multifoil then laid on top.

That was what I proposed to do until I realised that to apply multifoil you have to batten first to allow the 25mm expansion area 'behind' it, then batten again with another 25mm to allow for the expansion area 'above' it and also for somewhere to fix your plasterboard. OR I could fit another layer of 60mm Kingspan horizontally to cover the thermal bridging that would have been there from the existing rafters. That' what I did.

Next was whether or not there was sufficient depth of rafter to let me slide in 60mm Kingspan in between the rafters. The regs say 50mm air gap. Come on...get real... we're talking about an old house here..50mm is big enough for the Blackwall tunnel. There's loads of places where air is going to get into the roof void to ventilate it. So if some air gaps are a bit thinner than 50mm, frankly I don't see it as a big deal and any reasonable BCO will agree.

Condensation..if you're going to plasterboard and otherwise seal the bathroom space up then any water vapour ingress is going to be minimal. I really think that one can go overboard in slavishly following the regs to the nth degree.

My roof void is also relatively small especially as I raised the ceiling joists to expose the lovely old oak purlins. I over-engineered and used 4x2 (no access left in the roof void) and laid 120mm Kingspan over the top. That means that I have a sufficient airgap between the halogen downlighters and the insulation that I don't need to worry about the fire hazard.

Hope this helps.

Hi Martian. Good to have you back.

Thanks Keith and Skywalker (I hope) I am pleased to see you all again too.

My own reading of the NPL report leads me to a number of conclusions:
1. That for a number of reasons, in situ comparative testing is impracticable.
2. That when all factors are correctly accounted for the actual performance of multifoils corresponds almost exactly with both theoretical calculation and with hotbox measurement.
3. That because of points 1 and 2, in situ comparative testing is wasteful and not necessary.
4. That the thermal performance of Actis Tri-iso Super 10 varies to a much greater degree than conventional products depending on the direction of heat flow. With heat flow DOWN Tri-iso marginally outperforms 100mm of low density fibreglass quilt but in all other orientations is in fact inferior to 100mm fibreglass. (Many people including myself have already predicted and allowed that multifoils would score quite well in resisting solar gain through roofs in this way.)
5. That the actual U-Value of Actis Tri-Iso Super 10 in their recommended roof construction is approx 0.51W/m²k - very close to the LABC guidance value. Considering the 0.2W/m²k U-Value allowed under the building regs for a vaulted roof construction, we can see that the Actis roof would let more than 2½ times more heat escape than it should.
6. If you consider whole year performance and aggregate heat flow UP (winter) and DOWN (summer) through a roof, an 85mm construction of Tri-iso will be roughly equivalent to 100mm of fibreglass - about 2½ times WORSE than the 250mm equivalence Actis and TRADA claim!
7. Dividing the rafter depth into four 25mm deep air gaps using light foil faced card (or heavy paper) outperforms the multifoil as an insulator by a factor of roughly 1½ times which is not a practical way to insulate a roof but helps to demonstrate how effective trapped air gaps bounded by low emissivity surfaces are as insulation. By inference we can see that a large part of the effectiveness of expensive multifoils is actually achieved by the very cheap air on either side!

An interesting observation of my own is that if both roofs were constructed of 100mm deep rafters with a breathable felt above, the tri-iso roof would actually be a thicker construction because of the counterbattens below (ie - less headroom in loft conversions) and would be about five-six times more expensive per m² just for the insulation (counterbattens extra again).

If anyone can draw alternative conclusions I would be interested to know how and what they are .... ?

Martian

I always wonder what these Martians get up to when they just disappear for a while - hope it was fun anyway!

Posted By: biffvernonAn excellent summary of the previous zillion posts on this thread

Not quite - no account taken of http://www.greenbuildingforum.co.uk/forum/index2.php?DATEIN=tpc_wlpssdlpg_1142805843&showpage=120 (my post of 31.1.07) Martian didn't understand it then (I later wrote a version in plainer terms, but can't find it now), and I've seen no similar explanation published. However Paul Mitton eventually confirmed to me that this is indeed how multifoils work.

Their concentration on resisting radiant transfers on a micro scale within the thickness of the insulation is unique to multifoils (so far - something like Cellotex with chrome-plated interior bubble surfaces would work even better). All other insulants (whose solid matrix is composed of high-emisivity material) positively encourage such micro-scale radiant transfer within their thickness. The latter doesn't matter under steady-state conditions, but becomes overwhelmingly important under real-life fluctuating temperature conditions, which cause step-change temperature fronts to criss-cross through the thickness of the insulant.

Posted By: MartianThe multifoil manufacturers case rests primarily on the claim hotbox testing does not measure heat transfer by radiation (I think I even saw this claimed in their legal submissions). This is simply not true as you will see from a comprehensive report dated Dec 2007 from the National Physical Laboratory:
http://publications.npl.co.uk/dbtw-wpd/exec/dbtwpub.dll?&QB0=AND&QF0=ID&QI0=%204835%20&TN=NPLPUBS&RF=WFullRecordDetails&DL=0&RL=0&NP=4&AC=QBE_QUERY" >

I read that abstract, admittedly not the whole report, and see no mention there of the radiant issue. There's FEA modelling of predicted results, which should imply inclusion of radiant; but when it comes to testing, no disabling of the usual hotbox methodology, which systematically eliminates radiation as a source experimental error. Anyway it's not radiation to/from the external faces of the insulant; it's internal radiation - did the NPL's FEA analysis model that? The multifoil guys have.

Until Martian gets his head round the vital link between a) internal radiant transfer and b) fluctuating temperature, so that he can't mention one without the other in the same breath, his "excellent summary" will be so incomplete as to miss the point entirely. Though ignored by Martian, the fluctuating temperature issue is in fact prominent in the first sentence of the NPL Report abstract. But the NPL's "Temperature cycling" suggests such weedy fluctuations - one can imagine the carefully-stabilised hotbox temperatures being permitted to change slowly and steadily over a period of hours or minutes. A far cry from the sharp step changes that occur in real life - the sun comes out, a gust of cold wind, someone opens a window. all of which create a seething complex of non-steady-state temperature inequalities, advancing temperature wavefronts, possible anihilation and reinforcement of same where they intersect, all of which get evened out predominantly by internal micro-scale radiant transfer - readily in the case of conventional insulants, against great resistance in the case of multifoils.

Without having read the full NPL report, it smells just like more of the same to me. What we need is a better cafetiere, which is exactly where I came into this forum so long ago!

Hi Martian, welcome back.

I have read the full report . It took me about three days to understand it all and even now I'm not sure if I have the measure of it. Anyway, I would be interested to hear about how it is open to peer review. Do you mean by forum such as this or is the NPL requesting feedback? Is it published in a peer reviewed journal?

My main criticism is that the Authors have reviewed the possible methods for assessing thermal performance. [Note this includes 2D Finite Element Analysis]They have not however considered Dynamic Thermal Simulation [nor Biff's Cafetiere]. It is not therefore the comprehensive document you proclaim. This is something which I have investigated in my research; and which I previously posted initial findings here. These seem to have been removed though I can send you some details if you wish.

There are also limitations with the testing the NPL have carried out, some of which are acknowledged, some not.

Hi Mike, thanks for the welcome it is nice to be here.
What I mean about it being open to peer review is that it has been made freely available to anyone to read and comment upon. In an institution like the NPL I think that all reports are already internally peer reviewed before publication but this by no means prevents you or anyone else having a go. If you have any criticism of their scientific method I am quite sure that Messrs Williams and Ballard will be happy to respond to it. There are many ways to do this. You could ask for their contact details from the NPL and write/email them or you could counter-publish a critique of their report and send them a link so that they could comment in an open blog or forum.
You could possibly publish your critique in the Green Building magazine and hopefully slug it out with them here!
(Before taking any of these routes you would be wise to first need to seek permission to use extracts from the report from the Managing Director of the National Physical Laboratory - if consent is given the source must be acknowledged and the extracts may not be used out of context. Such permission is usually given quite freely.)

From what I remember of your previous posts you were trying to obtain the exact composition of the mulitifoil insulation so that you could input all of the varying layers and materials to see what your sofware would predict the thermal properties to be. Is this correct?
I think you should be more specific about the limitations with the testing which you think the researchers have not acknowledged and let them know. I am sure if you are right they would welcome your input.

Posted By: MartianWhat I mean about it being open to peer review is that it has been made freely available to anyone to read and comment upon.

Not quite ‘freely’ due to the strict quotation policy you highlight.
In my experience, making queries and /or comments to Authors does not often lead to a response - the word never springs to mind but I will give it a go.

Posted By: MartianFrom what I remember of your previous posts you were trying to obtain the exact composition of the mulitifoil insulation so that you could input all of the varying layers and materials to see what your sofware would predict the thermal properties to be. Is this correct?

Yes it is correct.

Posted By: MartianI think you should be more specific about the limitations with the testing which you think the researchers have not acknowledged and let them know. I am sure if you are right they would welcome your input.

We will see