Thanks Paul

A lot of "what ifs" there Tom. Conventional insulation, especially any kind of fibre based, is very much affected by air leakage. One of the advantages of the continuous foil type insulation types is that they give better air tightness. Since UK construction practice seems to be unable to reliably produce airtight buildings, this is the first place I'd look for any discrepancy in measured versus expected performance. Once that is done, then other models of the differences, if there are any at that point, can be invoked. As far as I know the test houses in Canada, which are built air tight (and with measured air leakage paramters), the models are very close to the measured performance for conventional insulations. I don't know if they've tested multifoils as there are none on the market over here, excepting the "double bubble" type (which is not multifoil in the sense you've been describing).

Don't forget Occam's Razor: one should not increase, beyond what is necessary, the number of entities required to explain anything. Also known as the Principle Of Parsimony. (see http://pespmc1.vub.ac.be/OCCAMRAZ.html et seq)

Paul in Montreal.

Further to Paul's comments on real life results matching up with software predictions, there are further examples from the German Passivhaus programme. Their software PHPP is an Excel speadsheet based on steady state assumptions and gives excellent correlation with measured results. See http://www.passivhaustagung.de/Passive_House_E/Passivehouse_measured_consumption.html for details. And not a hint of micro chaos or new world theories so far as I can see - just good air tightness and build quality.

Unventilated airspaces - 1980s info source but shd still be current? (tabulation may not work completely but hopefully you can interpret):

Emissivity of surface / Heat flow horiz or up / Heat flow down

5mm gap High / 0.11 / 0.11 R (m2K/W)
Low / 0.18 / 0.18

20mm gap High / 0.18 / 0.21
Low / 0.35 / 1.06

Your calc doesn't take account of thermal bridging e.g. where there's a timber stud instead of insulation/air - this will knock your U down significantly.
Foil backed plasterboard is hard to get, expensive and isn't formulated to stay bright and shiny.
'3 inch' Cellotex is far cheaper than the other thicknesses.
This macro-scale interpretation of how multifoil works isn't accurate - it's the multiple layering within its thickness that exploits the effects of real-life dynamically varying conditions. What you're proposing hardly does at all, so you might as well save your money rather than trying to mix relectives in with a traditional 'steady-state-thinking' construction.

I'd suggest:
Min 120x50 studs with '3 inch' Cellotex carefully fitted between (studs @ 450 or 650c/cs, cutting Cellotex from 1200 wide without offcuts), alumin taped joints, foam sealed to studs. Outboard, 9 OSB, Tyvek,19x50 battens, cedar boarding. Inboard of the studs, multifoil - the imperforate, spotwelded variety, not the through-stitched variety - that means Euroform Xfoil 25 at present), 35x50 horizontal battens, plasterboard. Similar for rafters - allows the inner layers to run from wall to roof seamlessly.

Or:
Wet plaster on 140 blockwork. Outboard, '3 inch' Cellotex, 35x50 vertical battens, multifoil (preferably imperforate; through-stitched debateable - interstitial condensation question), 35x50 vertical battens, cedar cladding.

The block rather than stud solution is superior; a bit (albeit not enough) of thermal mass inboard of the insulation; cheap and simple as long as you've got solid wall to build it off. !40 won't pass building regs without a structural engineer's calc; 190 (what's that?!) is the min deemed-to-satisfy. If it's a conversion/extension, bldg regs might allow you to omit the Cellotex, relying entirely on the multifoil, but you may be too sceptical for that.

For overcladding existing masonry, '3 inch' Cellotex, 35x50 vertical battens, multifoil, 35x50 vertical battens, cedar cladding (or render on rendermesh, ventilated behind) is great - again possibly omitting the Cellotex. In the latter case, within 94mm additional wall thickness outboard, you get t'he equivalent of 250 min fibre' insulation - if you're a multifoil believer!

Note 35dp battens, not 25 as usually recommended. As multifoil is usually 30thk, it needs at least 35 batten space, inboard and outboard, if (as is suggested) a cross-battening arrangement is used. In the above studwork solution, the difference between the 120 stud depth and the '3 inch' Cellotex provides this 'batten space'.

ASA Adjudications
ACTIS Insulation Ltd
Unit 1
Cornbrash Park
Bumpers Farm Industrial Estate
Chippenham
Wiltshire
SN14 6RA
Number of complaints: 1
Date: 23 April 2008
Media: Brochure
Sector: Household
Agency: Outside Agency Not Stated


COMPLAINT:
Objection to a brochure for roof insulation distributed in 2005. The brochure stated "TRI-ISO SUPER 9 Insulation for roofs ... Thermally equivalent to 200 mm of mineral wool when installed in a roof situation, as certified by the European certifying body, BM TRADA CERTIFICATION (following real building trials, certification n°0101) ... THERMAL EFFICIENCY equivalent to 200 mm of mineral wool RT = 5* ... *in situ measured values." The complainant challenged:

1. the claim "Thermally equivalent to 200 mm of mineral wool" and

2. the quoted thermal resistance "RT = 5".



ADJUDICATION:
THIS ADJUDICATION REPLACES THAT FIRST PUBLISHED ON 31 MAY 2006 AND SUBSEQUENTLY WITHDRAWN ON 26 JULY 2006. THE COMPLAINTS REMAIN UPHELD BUT THE WORDING HAS CHANGED.

Actis Insulation Ltd (Actis) said they had stopped advertising TRI-ISO SUPER 9 because it had been replaced with their new product TRI-ISO SUPER 10. They said the efficiency of their products was demonstrated by their track record in the market. Actis said they had commissioned TRADA Technology Ltd to assess and report on the TRI-ISO SUPER 9 product. Its sister body, BM TRADA Ltd had certified TRI-ISO SUPER 9 on the basis of those results. They provided us with a copy of the BM TRADA Certificate dated February 2005, and Report dated August 1997, which they said substantiated their claims. They said that BM TRADA was a leading multi-sector certification body accredited by the United Kingdom Accreditation Service.

Actis explained that TRI-ISO SUPER 9 was different from traditional bulk insulation because it was a multi-foil product that used layers of reflective foils spaced with synthetic wadding and foams. They said the product required less space than traditional bulk insulation and, therefore, internal insulation cavities could be made smaller, and internal useable spaces could be enlarged without compromising efficiency of insulation.

Actis argued that traditional methods of testing were not appropriate for their product because such methods measured thermal efficiency mainly by measuring heat transfer by conduction. They said their product inhibited various methods of energy transfer including radiation and convection as well as conduction, and worked principally by reflective radiative heat. They said the effective functioning of their insulation system required installation of the sheets of multi-foil as a "sealed envelope" with taped joints between sheets, so testing of small pieces of product in a laboratory was inadequate. Actis also argued that traditional methods of testing did not allow representation of the real behaviour of building materials once installed. The test supervised by TRADA Technology had used 'in situ' testing involving a real external environment with variations in temperature, humidity, etc. The tests were carried out in Limoux in France over two months in early 1997.

They explained that the 1997 testing used two identical and adjacent chalets, one of which was lined completely with TRI-ISO SUPER 9 and the other was lined with 200 mm of mineral wool. Because the manufacturers quoted a thermal resistance of 5 m²K/W for the mineral wool and the test results showed an equivalent thermal performance in the two chalets, the report concluded that TRI-ISO SUPER 9 was "thermally equivalent to 200 mm of mineral wool" and had a thermal resistance of 5 m²K/W. Actis therefore maintained that the BM TRADA Certification demonstrated the thermal efficiency of their product and provided proof of their claims.

1. Complaint upheld
The ASA obtained expert advice. We understood that the 1997 testing had not tried to measure the thermal resistance of TRI-ISO SUPER 9 directly, but had compared the measured energy consumption and internal temperatures in a chalet lined with TRI-ISO SUPER 9 with those in a similar chalet lined with 200 mm of mineral wool. The TRADA report stated that the construction method of both test chalets was representative of real buildings. We noted that both insulation systems were installed in the test chalets without a tiling underlay between the insulation and the ventilated space under the roof tiles, and with no internal plasterboard lining to the interior. Actis argued that the use of underlay and plasterboard was not common in France in 1997, but we understood that a tiling underlay and a plasterboard lining were used in most real buildings and noted that both the underlay and the plasterboard lining featured in the brochure for TRI-ISO SUPER 9 and the BM TRADA Certificate for TRI-ISO SUPER 9. Because there was no tiling underlay and no plasterboard lining, air would penetrate the mineral wool from the outside, and to a lesser extent, from the inside degrading its thermal performance. In addition we noted that care was taken to seal the joints between the pieces of TRI-ISO SUPER 9, which in itself was completely airtight, so as to eliminate air infiltration in the chalet that contained TRI-ISO SUPER 9. According to the TRADA report and the photographs annexed to it, the joints between the pieces of mineral wool in the second chalet were left open, which would lead to some ventilation in that chalet. Actis produced recent statements suggesting that TRADA were aware of the problem at the time and had directed Actis employees to seal the joints, and that this had been done, but there was no evidence in the contemporaneous documentation that that had been done. We concluded Actis had not substantiated the claim. We noted the ad was no longer appearing, but told Actis not to make claims in future advertising unless they held sufficient substantiation at the time of publication.

2. Complaint upheld
We understood that RT was a symbol of total thermal resistance and typically had the standard unit of measurement of m²K/W. We noted that the claim "RT=5" was not qualified by any recognised units of measurement e.g. m²K/W and a small footnote stated only 'in situ' measured values" without further explanation. While a professional trade reader would probably not be misled by this omission, Actis agreed to include a unit of measurement in future. We noted that the TRADA Technology report did specify an overall resistance (RT) of 5.0 m²K/W derived from the "in situ" testing. Because heat loss in the chalet with TRI-ISO SUPER 9 was by transfer through the fabric, whereas heat loss in the chalet with mineral wool was from a combination of fabric transfer and air penetration (see 1 above), we considered that it was not possible to compare accurately the thermal resistance in the two test chalets. On the other hand, the ASA's expert considered that it was possible to estimate the thermal resistance for the well sealed TRI-ISO SUPER 9 chalet, from the data recorded there, as being between 1.6 and 1.8 m²K/W, not 5 m²K/W. Even if, as Actis argued it was not possible to estimate an actual RT value from the data for that chalet alone, it was clear to us that the report did not substantiate the claim "RT=5". We noted the ad was no longer appearing, but told Actis not to make claims in future advertising unless they held sufficient substantiation at the time of publication.

The brochure breached CAP Code clauses 3.1 (Substantiation) and 7.1 (Truthfulness).


Adjudication of the ASA Council (Non-broadcast)

Here is the real link. Well done to John Willoughby who told me recently that he stands by the belief that this product is the equivalent of just 64mm of normal insulation not the 200mm that Actis quoted.

http://www.asa.org.uk/asa/adjudications/Public/TF_ADJ_44335.htm

Posted By: (GBP) KeithJohn Willoughby who told me recently that he stands by the belief that this product is the equivalent of just 64mm of normal insulation

Keith, are you satisfied that John's taking into account the reality of rapid dynamic thermal change (not the sluggish change in the recent NPL Report), and that performance in those conditions is very different from in steady-state?

Posted By: Paul in MontrealTom, you can't have it both ways. You're a high thermal mass evangelist which implies that rapid dynamic change is not possible due to the, well, thermal mass! Real world changes are relatively slow - just look at an hour-by-hour temperature chart. Your argument might apply for lightweight construction if the multifoil is essentially adjacent to a low-mass outer sheathing. Otherwise there can be no rapid dynamic changes can there?!

Ho yuss there can. Your 'real world changes' recorded in a 'hour-by-hour temperature chart' are the

Posted By: Paul in MontrealMacro models (which) already account for micro-scale chaos which, despite what people often think, does not always have macro effects

but in this case certainly do have macro effects. I think you're conceding that on a micro scale, there's seething dynamic temperature inequalities being constantly started and evened-out, in lightweight and heavyweight materials alike? If so, then that's why on a micro scale heat transfer across any voids within any material, heavy or light, will be overwhelmingly begun and completed (delta-t anihillated) by instantaneous radiant means, before time-lagged conductive and convective means can reach any destination. That effect of micro-scale 'chaos' makes a huge difference at the macro level.

LABC Guidance Notes April 2008 available from here:

http://www.kontrol-insulation.com/news

Thought I'd post this here as it has some relevance. http://www.limetechnology.co.uk/upload/documents/1201787723_the_thermal_performance_of_tradical_hemcrete.pdf

They seem to be arguing that the hot box method is not suitable for measuring the thermal conductivity of hemcrete. Interesting that Building Control accept this argument for Hemcrete but not for multifoil.

Well, I can add some real world experience to this discussion.
I just had my loft done - well a complete new roof really, and went for Tri-Actis 10 foil insulation. Its been complete since Christmas. We do not need ANY heating in there - and this is a room which is 4.5m x 5.5m.
I would add that the Architect & I spent an inordinate amount of time ensuring the builders got the insulation correct, as they clearly did not want to do it properly.

I could have dispensed with the heating in there and saved myself quite a few bob!

So even if the insulation is not as theoretically good as it should be, if it were any better we'd have to lose the heat by opening the windows - which kind of defeats the point really.

It's late, and I'll try to do something this weekend - if I can work out how to post photos.

First, to explain, I am an engineer - chartered chemical engineer & I work with architects. When I was looking into my loft project - mid 2006, one of them had a sample of the multifoil on his desk - it wasn't Tri-Actis, but a similar product. It looked cool, I did some digging and was persuaded, particularly the difficulty of getting 250mm of solid foam on the roof slope.

I'm an engineer so I spent a weekend reading ALL the installation advice, specs, guidance etc, which the builders clearly had not! My architect did care as well, because he's a pedantic so & so as well. He regarded it as his job to keep the builders honest - they really did not like it!

The main issues were a) the stapling b) the overlaps/taping and the noggins required c) the detail at the soffits.

You need a really meaty stapler, fortunately I had a beefy 14mm stapler which does the minimum needed. Lots and lots of stapling needed.
The overlaps need stapling to noggins - so you need noggins
At the edges you need to do lots of stapling and taping, and then stuff the eaves with rockwool insulation to close off any residual air leakage.

Tim

Biffvernon. It is indeed a VERY big thread. I tried reading it all, but it got late, and I'm tired. Could you summarise the main points? Are the majority for the thin, tin foily stuff, or agin it?

I'd like to see this thread reach 300 posts. Only three more needed now ....

Biff, you did your spreadsheet to prove diminishing returns; I pointed out the conceptual error therein and with martian's help did calcs to prove that every foil does its bit; you avoided acknowledging sight of that; and have maintained your position regardless! I posted: "you haven't answered my question about ..." and you posted back "it's me that asks the questions around here!".

There - no one post now - leave Biff his rightful priviledge, of the 300th. Respect.