25 Jan 2007

Multifoils: last throw of the dice

I went to London this morning to attend the first ever press conference of the Confederation of Multifoil Manufacturers (CMM), held at the Building Centre. The CMM consists of four members being Actis, YBS, Euroform and IPP who, combined, account for 90% of UK multifoil market.

They put on a 90-minute presentation consisting of four presenters and a short Q&A session in front of an audience of around 30, made up of people working in the multifoil industry and a number of trade journalists.

• Paul Newman, the secretary of the CMM, opened the proceedings and talked a little bit about the challenges facing them; in particular, the August 2006 release of a technical guidance notice from the LABC (Local Authority Building Control) advising against the use of multifoils as standalone products. Newman insisted that multifoils were being held back by old (redundant?) testing methods. He used a graph to show how multifoils had by then captured 6% of the total UK insulation market, up from just 1% in 2002, but this was now expected to fall; the graph suggested 5% in 2007.

• Paul Mitton, Technical Manager of Euroform, then spoke generally about the installation advantages of multifoils, highlighting speed of installation and minimal depth.

• Matthew King, Technical Manager Actis UK followed. He spoke at length about the failings of the guarded hot box (GHB) test used to evaluate conventional insulation and claimed that GHBs only tested for conduction. He claimed that 93% of heat transfer through building fabric was by radiation (a claim later challenged by Roger Bisby) and that somehow GHB testing was not measuring this. This is a controversial view and is disputed by GHB experts such as Ray Williams (not present) of National Physical Laboratories, who has told me that the GHB test doesn’t distinguish between conduction, radiation and convection. King also contested that the GHB measures a temperature difference in a steady state between 10°C and 20°C and that, in reality, temperature differences in the built environment are much larger than this. This was questioned by a lady from the AJ who thought that 20°C would cover all but the most extreme weather events. King further claimed that GHB lab tests do not take into account
- temperature variations which disturb the thermal system
- rain and humidity which influence thermal conductivity
- solar radiation
- wind strength and direction (convection)
- difference in pressure (interior, exterior)
- water and air tightness
- quality of the installation
- ageing of materials.

However, whilst emphasizing that site tests would be more realistic than lab tests, as has been show in the world of acoustics, he did not explain why site tests would bring about improvements over lab tests. In acoustics, elements on site invariably perform worse than they do in lab tests: here we are expected to believe that multifoils perform better on site than they do in GHB lab tests.

King then gave a lengthy exposition of how the in situ tests are carried out and how close attention is paid to ensuring that the controls are fair:
- two roof assemblies are constructed and tested without insulation to see if they are within ±3% of each other, as regards energy used to heat to a pre arranged internal temperature.
- then both assemblies are tested with 200/250mm mineral wool to see if they remain within the ±3% range.
- only then is multifoil placed in one of the roofs to carry out the comparison testing.

Results from the Actis tests have shown that whilst 200mm of mineral wool results in higher than predicted energy use, the Actis multifoil performs far better than the simple R-values derived from GHB testing would suggest. In fact the multifoil matches the results from the 200mm of mineral wool. This will come as no surprise to those who have followed this debate: it is the essential claim made by all multifoil manufacturers and the one disputed by virtually every other insulation manufacturer.

• The last speaker was Prof Nico Hendriks, professor of Building Materials at Eindhoven University of Technology and the Principal of the BDA Test Institution, a third part approval agency. Hendriks has been hired by the CMM to take on the independent testing of multifoils. Basically, he is charged with replicating the famous in-situ tests carried out by Actis and to use the (hopefully) positive results to persuade EOTA (European Organisation of Technical Approval) to grant multifoils an ETA (European Technical Approval), which would enable multifoils to be awarded an equivalent R value and thus to be used in all European territories.

He ran through the methodology that would be employed in carrying out the tests, but he was later questioned in depth on this by Roger Bisby of Professional Builder who noted that they were only planning to test against mineral wool and that in a configuration that would no longer pass UK building regs (basically wool stuffed between rafters). “Why not test against Pu foam instead,” asked Roger. Hendriks appeared to be unaware of the current requirements of Part L, saying that no one built like that in Holland anyway as they mostly used panelised roof sections. He also pointed out that there were both cost and time implications for further comparitive tests. “Why not three roofs, why just two?” went the questioning. “I am sure Kingspan would be willing to help with costs.” This observation met with a good deal of laughter from the audience, rather less from the panel.

I asked what the timescale was for their application to EOTA for an ETA. EOTA is like an umbrella organisation for national bodies, such as our own BBA, which have a remit to test innovative building products. It moves slowly and is still considering whether to even look at this testing programme, let alone pronounce on it. One of the drawbacks of in situ testing is that it takes three months to carry out a winter test and three months to do a summer one as well, and the CMM have already missed this winter as a test time, so the very earliest they could have results before EOTA would be mid-2008. EOTA has to reach a consensus on the matter so it is far from clear what the outcome will be, assuming that Henrdriks’ BDA test outcome is positive for multifoils.

Interestingly, there is a second independent testing station being used by the CMM for their in situ tests to place before EOTA. This was mentioned in passing by Prof Hendriks, without reference to who this might be. Later I discovered that this other test station was none other than our own BM Trada, who of course were the original independent verifiers of the Actis in situ tests back in 1997. Why Hendriks was flown in from Holland when they could have called on BM Trada from up the road remains a mystery. Indeed BM Trada didn’t appear to be present at the event at all. Maybe they are seen as being too close to Actis for comfort.

In the question and answer session that followed, the lady from the Architects Journal asked what the middle layers of radiant foil actually did. Paul Mitton gave a reply that was both convoluted and unintelligible. I report that I am none the wiser!

Another questioner asked about Web Dynamics, a multifoil manufacturer who are conspicuously not members of the CMM, having made a decision that multifoils were best used as a supplementary insulation layer. Web Dynamics have achieved BBA third party approval (a UK-only version of ETA) after testing their multifoil, Thinsulex, in conjunction with conventional insulation materials in the BBA’s guarded hotbox. The CMM speakers were dismissive of the Web Dynamics approach and claimed that their multifoil was an inferior product that couldn’t match the performance of the ones produced by CMM members. Mitton claimed that: “Their wadding is black and as you know black material absorbs heat, not what you want in thermal insulation.”

The last question was about the BRE tests that took place at a site near Aberdeen. These were essentially in situ tests of multifoils against conventional insulation and the results appear to confirm the GHB lab tests – i.e. multifoils do not match the claimed for 200mm of mineral wool insulation. Mathew King’s view was that the test was not as rigorous as the ones that they had carried out, nor as the ones that they want BM Trada and BDA in Eindhoven to undertake.

Over lunch I was able to question Andrew Whittle from IPP and Claire who had flown over from Actis in France. I asked Claire if it was true that Lafarge were in the process of buying Actis: this was news to her, so it appears to be just gossip. I also asked her how Actis sold throughout Europe. She said it was a difficult time for them, as authorities seemed to be closing ranks against multifoils for similar reasons as we are seeing in the UK. Interestingly, they have never been able to sell into Germany (why does that not surprise me?) because the Germans have always insisted on hard R-value figures, which means positive GHB test results. But most other countries have been receptive to multifoils.

In summary, the multifoil industry feels very aggrieved that it has been unable to establish that its products do actually work as they claim. They say that their UK market has been badly affected by the LABC advice to stop using multifoils on their own, but that many metropolitan local authorities are still happy to accept multifoils, especially in loft conversions where it is often virtually impossible to build in the correct depth of insulation using conventional materials. They seem to be pinning their hopes on a) independent testing verifying their own test results and b) consequent EOTA approval, which would give multifoils access to all European markets via an equivalent R value method which even the Germans would have to accept.

My own feeling is that they are being extremely optimistic about their chances. What I didn’t hear today was any convincing explanation as to why multifoils perform so poorly in the guarded hotbox test and yet appear to equal large depths of mineral wool in comparison tests. As happens so often with the multifoil debate, I came away with more questions than answers. It’s all very well carrying out in situ tests, but it’s not altogether clear what exactly is being measured. It may be that 200mm of mineral wool as an insulator is very much worse than the lab tests indicate, in which case the multifoil comparison tests are simply illuminating this point. And why always test against mineral wool? It is rarely used in roof applications these days because it’s hard to meet the requirements of Part L without an absurdly deep rafter. Why not test against more realistic options for today’s housebuilders? And why can’t they come up with a scientifically plausible explanation of the workings of multifoils? Whilst I don’t doubt the sincerity of the multifoil manufacturers, their arguments are still some way short of compelling and on the basis of what they showed us today I really think they have their work cut out if they are to win the approval of EOTA, certainly first time around. It is far more likely that EOTA will ask the same question as Roger Bisby. “Why just test against a roof stuffed with mineral wool? If you want our approval for this unorthodox test procedure, let’s see more tests against a wide range of insulation products.” It could be 2010 before EOTA reaches a conclusion by which time the multifoil industry may well be history.


  1. Thanks for being there and giving us your report.

    Perhaps you've put your finger on what continues to be so baffling:
    "It may be that 200mm of mineral wool as an insulator is very much worse than the lab tests indicate, in which case the multifoil comparison tests are simply illuminating this point"

    Maybe it's not about how multifoils inexplicably manage to perform so much better in the field, than their hotbox results. Instead, maybe it's a revelation that conventional insulations performs *very* much worse in the field, than their hotbox results. In the field, a multifoil may indeed equal 200mm of mineral wool - but at a R-value about half what the conventional insulation industry and the testing houses have been telling us.

  2. As Tom says, thanks again for reporting on this.

    His comments echo a thought I had at the Green Build conference you organised in Cambridge. There, we heard about the success of hempcrete as an insulating building material. However, the in situ tests seemed to imply as much the failure of traditional materials to perform at ideal levels as the high quality of hempcrete.

    Could it be that sagging, air gaps, cold bridging and moisture retention may all contribute to the failure of many traditional methods that otherwise do well in laboratory tests?

  3. My hope is that aerogels manufacturers eventually solve the ridgidity problem, and come up with a blanket that if the conductivity values reported are correct will simpy wipe out both multifoils and the ridged foams such as Celotex and Kingspan.
    But perhaps I may not live long enought to see this happen...

  4. Very interesting post thank you.

    I read a good story about the problems of installed versus theoretical R/U values, the essay is American in origina and titled "R Fairy Tale The Myth of Insulation Values" http://www.monolithic.com/plan_design/rfairy/

  5. multifoil works! with an infrared meter and a reflective foil placed in the opening of the fridge in combination with a polyprop honeycomb I measure a R-value of 5! The ouside temperature stays 1 celcius warmer then the outside panel of the fridge itself. In combination with the outer temperature and the temp in the fridge you can simply calculate te U and the R values.
    Its all about lobbying, money and politics. But I'm sure multifoil works!

  6. The reason the glasswool performs so different in comparason to its lab results is the diffent conditions it is under.

    The traditional lab test for wool is stick it between 2 plates at 20C and 0C, and measure the heat flow across them. As the wool is trapped there is little effect caused by the wool being subject to air movement.

    This result with minimal air movement through the wool would be something you would expect to see from wool installed between rafters that have a breather membrane on top and a VCL and plasterboard on the underside. (the effect of air movement by having an air flow adjacent the wool IS something that can be accounted for in a u-value calculation).

    As pointed out by ASA when they looked into the BM Trada test for triso, there was no plasterboard to the underside of the rafters (because all loft conversions with glasswool just leave it exposed to the rooms!!) and there was no felt under the tiles. So the wool IS subject to air movement and so will not perform to the certified value.

    This is where the comparative testing is manipulated to give the desired result. Or Actis simply forgot half the stuff you would install with the wool (if you used wool between rafters).

    Question, why can't they follow the ISO standard for in situ testing? Where the R-value of the roof is measured instead of being compared to something else?

    On another note, who is Mitton person? And what is the blabbing on about black wadding for? The only heat the black would absorb is radiant heat, aparently his foils have just reflected up to 95% of radiant heat so what difference is the colour of the wadding going to make when it only has 5% left to absorb? So surely black would perform just as well as white in a multifoil? Contradiction of his own claims or what?