[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GSBN:fire test -- your opinions?

Hi Baleheads

in the SB fire tests Barbara and Bee were involved with in the UK, even though the render didn't seal to the metal frame of the test
jig very well it still took over two hours to reach failure with lime rendered straw bales.
I copy the results which Bee sent to me below.

bale on
<a  target="_blank" href="http://www.strawbale-building.co.uk/";>http://www.strawbale-building.co.uk/</a>

The results I've attached are for one of the tests, the one which used the
small regular bales.  I think this will be most useful for you. The other
test used larger bales, and performed very similarly, though it failed a
little later at 164 mins.

The test was what is known as an 'indicative' test, as opposed to one fully
in accordance with the standards - when it says in the report that 'the full
requirements of the standards were not, however, complied with', all that is
meant by that is that it was not conducted in the 3m x 3m test frame... this
is the required test size for a wall, in order to test to the accepted
british standards.  Our test was conducted in 1.5 x 1.5m test frames.  This
was because they were too frightened about the possibility of burning straw
igniting the filters in the flue from a big, burning wall!  They absolutely
refused to perform it in the large test rig because of this fear.  The tests
were delayed for a considerable time because of this, and eventually they
agreed to run them in smaller test frames.  However, it does mean that the
results from these small test frames (1.5m x 1.5m) are not as good as those
that I believe would be achieved in their large frame - because the large
frame, built of steel, has a ram which pushes upwards, so loading the wall -
giving it realistic compression.  In our frames, we simply had to cut some
timber wedges and hammer them in at the top to give the wall stablility/ as
much compression as we could. You'll note from the observations that after
only 50 minutes, the top edge of the wall moved out of  the frame away from
the furnace by 15mm... and continues to move outwards.  This would not of
course happen in loadbearing design.  So these tests go some way to
emulating infill results, not loadbearing, which I am certain would be
significantly better.

They are only allowed to make observations, not comments, on the performance
of the materials.  However, James was able to say to me that the performance
of the wall based on these results, was extremely good.
The tests look at two things:


Integrity failed on the small bales wall after 135 mins (ignited cotton wool
test pad).  This is over 2 hours. Minimum building regs requirements on most
materials is 30 mins.  (Some places will require up to 2 hours - this

Insulation - this shows that the wall provides 144mins of insulation which
is extremely good.  Their tests allow up to 180 degrees on an individual
thermocouple and 140 degrees in domino effect over 5 thermacouples... this
was only 65 degrees after 144 minutes.  Don't know if there is a building
regs requirement here, but this would definitely comply.

We hope to get funding to run the larger test at some point, thereby getting
the official stamp of approval on them/british standards compliance.  But
til then, these results can definitely be quoted as official, indicative BRE
fire test results - and you can see that they easily comply with the stiffer
building reg requirements, and perform far better than minimum building regs

Test 216578

Results of the indicative fire resistance test carried out, on 9 June 2004 for a duration of 146 minutes on a rendered straw bale
wall, 1.5m x 1.5m, using agricultural grade straw.  The furnace heating regime and appropriate procedures and criteria of BS 476:
Part 20/22: 1987 were utilised for the test using a furnace 1.5m x 1.5m x 1.5m.

The straw bales were of rye straw, measuring 460mm wide x 350mm high and 1020mm long.  All the bales except one had to be cut and
twined in order to form blocks of a suitable size for the wall.  The bales were laid flat, with staggered joints, within a steel
frame lined with masonry blocks and lintel.  The bales were packed into the frame as tightly as possible using timber wedges and a
timber plate along the top edge.  A stonewool blanket gasket was fitted between the bales and the test frame.  Each face of the wall
was coated with a traditional lime render comprising 1 part lime putty, 3 parts well-graded sand and 10% metastar burnt china clay
The wall was constructed on 15th March 2004.  The base render layer was completed on 16th March 2004.  The 3rd and final render
layer was applied on 25th March 2004.  A lime wash was applied on 1st April 2004.  Shortly before the test a few fine cracks were
visible on the unexposed face, none were observed on the exposed face.

The unexposed face temperature of the wall was recorded during the test by means of five thermocouples.  One was located at the
centre, and one was located at the centre of each quarter section.

The following observations were made during the test.  Unless otherwise stated they are of the unexposed face:

Table  1 Observations
















     Test started

      Smoke issuing from interface of top edge of wall and lintel

      Top edge of wall moved out of frame away from the furnace by approximately 15mm

      Top edge of wall moved outwards a further 5mm

      Crack, approximately 60mm long x 1mm wide, developed in the central area of the exposed face

      Issue of smoke increased

      Top edge of wall moved outwards a further 5mm

      Top edge of wall moved outwards to a total of approximately 36mm

      Top edge of wall moved outwards to a total of approximately 44mm

      Gap, up to 10mm wide, developed between centre of top of wall and lintel due to the top edge of the wall bowing downwards

      Top edge of wall moved outwards to a total of approximately 65mm

      Sparks and loud popping noises emitted from top edge of wall

      Hot gases issuing from gap at top edge of wall ignited cotton-wool test pad (integrity failure).  Width of gap increased up to
a maximum of approximately 18mm at centre.

      Entire wall fell out of frame.  Test stopped.

Integrity failure first occurred after 135min at the top edge of the wall.  The specimen is shown before and after test in the
attached photos.

The maximum temperature measured on the unexposed face was 65 degrees C recorded after 144min from the start of the test.  The
temperature recorded by the unexposed face thermocouples are shown plotted against time in the attached graph.

These test results relate to an investigation which utilised the test methodology given in BS 476:  Part 20/22. The full
requirements of the standards were not, however complied with.  The information is provided for the test sponsor's information only
and should not be used to demonstrate performance against the standard nor compliance with a regulatory requirement.  The test was
not conducted under the requirements of UKAS accreditation.

This report is made on behalf of BRE.  By receiving the report and acting on it, the client - or any third party relying on it -
accepts that no individual is personally liable in contract, tort or breach of statutory duty (including negligence).

James Knight

Senior Testing Consultant

For &amp; on behalf of BRE

&amp; R A Jones

Associate Director

For &amp; on behalf of BRE