Abject failure of mineral wool and questionable BBA test rigs

I have just read the BBA/ NHBC foundation's  Full Fill Cavity Wall Insulation in Areas of Very Severe Exposure to Wind-driven Rain 2016

When BBA issue a certificate for cavity wall insulation it must have passed their standard test which essentially involves building a section of brand new cavity wall in a lab with standard wall ties and either constructing it with built in insulation batts or applying a blown in product after it has been constructed.
The test rig has no features that are representative of a real house. Most people would say It's reasonable to assume that most houses will have doors, window and utilities passing through the cavity. But the document indicates that BBA test have not included these for the past 40 years.

Most certificates have been issued with the condition that the cavity must be >=50mm and for that the product was deemed effective are limiting water penetration to the inner skin.
There are some restrictions on fair faced masonry and exposure zones which meant full fill cannot be used in those conditions.

The 2016 study attempted to determine if a fully filled cavity of more than double that size (125mm) would reduce the risk of rain penetration and so allow new builds to utilise a fully filled cavity in areas of high exposure.
The main deviation from the standard BBA tests was that a PVC-u window was built into the wall. This was considered novel.

Summary of findings

In general, the test showed good resistance to penetration with built in materials (such as batts), some minor problems with polystyrene balls and complete failure with blown mineral wool.
As it relates closely to my situation I will concentrate on the results related to blown mineral wool.


65mm standard cavity as a baseline with a window typical of a 1995 building :
  • Mineral wool performed significantly worse than an empty cavity leading to saturation of two big areas in the internal skin where as there was only minor penetration of the unfilled cavity.
  • Even under laboratory conditions in a clean cavity, the fill was uneven with small voids and areas of much higher density than specified.
Full fill mineral wool in a standard 65mm cavity

125mm Cavity with modern window and full fill blown mineral wool
An empty cavity of this width was not tested for comparison but with the mineral wool, the following was found:

  • Water was able to penetrate and cause damp to the inner leaf. On one of the walls this could not be blamed on the window (see photo below).
  • Although there were no voids, the report notes that there were plugs of very high density insulation surrounded by less dense insulate which could be acting as moisture bridges to the inner cavity.


Full fill mineral wool in a standard 125mm cavity - penetrating damp not related to the window.

Full fill mineral wool in a standard 125mm cavity - penetrating damp around the window

Discussion

I hope the above is a fair representation of the tests. It's important to emphasize that these tests are designed to test NEW BUILD walls and the insulation is injected from the inside of the building and not from the outside as with retrofit cavity wall insulation (which is more risky in terms of penetration!).
The conclusions are pretty obviously applicable to retrofit:
  • If the mineral wool fill failed in a laboratory environment in a cement constructed 125mm CLEAR cavity then what hope does a sub 50mm 1930s cavity with 90 year old lime mortar,  cavities filled with rubble, snots and old cavity trays have at resisting penetrating water of any kind.
  • The test supports the notion that an empty 65mm cavity is far more protective against penetrating rain than blown mineral wool even when that wool is installed under laboratory conditions into a clean cavity.

There are three things that really bother me about the closing statement

The investigation was undertaken using a completely novel adaptation of the standard BBA masonry cavity wall Water Resistance test rig.

Really? Having a window in a test rig is considered novel. It's clear that the standard BBA test rig bears no resemblance to an actual new build let alone an old 1930s property.

The BBA and the Cavity Wall Insulation industry are also aware of the fact that features such as windows, doors, flue pipes etc incorporated into walls can accentuate any weaknesses in the construction and lead to water penetration. It is therefore a requirement in all blown-in Cavity Wall Insulation Agrément Certificates that a detailed survey of the property ...
This makes no sense at all. If the test rig failed and there were no defects found in construction when it was dismantled then how is a survey going to help!

Since investigations have not been made before, using the adapted rig, there is no experience of the behaviour of the insulation systems in this condition. The findings of this investigation have no bearing to existing certification for cavity wall insulation systems.

This is the most concerning statement in the report. This sort of back covering  looks as if it was added by the lawyers.  The study in my view directly calls into question the validity of all previous testing and should trigger some serious soul searching at the BBA.
It's fine to do this basic first pass test as it takes out lot of variables (such as windows and pipes) from the equation but it is not representative of a whole house. To draw an analogue, It would be the equivalent of releasing a drug after a phase II trial in medicine!

I'm am genuinely interested in hearing the industry/BBA side of this story so please get in touch if you disagree or have something to add about BBA test that are not clear from public documents.

Comments

  1. An interesting read, would you be happy if we shared the content ?

    ReplyDelete
  2. Yes. Please do but link back so comments may be made here.

    ReplyDelete
  3. This comment has been removed by a blog administrator.

    ReplyDelete

Post a Comment

Popular posts from this blog

The controversy over Cavity Wall Insulation 'topups'

Lintels, weep holes and render

Discrepancies between Internal and external DPC (damp proof course)