SFPE Discussions

Variations in fire wall types

Hi All
Looking for some feedback. NZBC requires a received level of radiation at the boundary of 30 kW/m2, and 16 kW/m2 at 1.0m over the boundary.
I normally use 175mm thick concrete for my fire walls, but recently I have been using a 78mm thick steel faced aerated concrete instead. The methodology is to determine the exposure to the boundary (and 1.0m over the boundary) from the unprotected portion of the exposing face, and then to determine the exposure to the boundary from the protected portion of the exposing face - in the relevant product’s case at 350C (at 240 minutes). Add them up (ignoring any variation in target points) and ensure that the NZBC limits are not exceeded. This generally increases the height of the fire wall by ~ 6%. The big advantage is the large reduction in the mass of the wall, plus ease of erection, plus application in high seismic zones. Note that the application is associated with cold stores, so the fire wall does not form part of the weather envelope.
I have been heavily criticised by a recognised fire engineer (to one of my clients, not to me directly) for applying this methodology. The Branz fire wall testing exists, there is at least one reviewer for accepts the methodology as being in compliance with NZBC, plus it is deemed to be acceptable to multiple Councils.
What am I missing?

The methodology of the combined sources of radiation is ok, as long as you have the data and correctly applied. For example, you would need to have the fire test data for 240 minutes integrity with the cold side temperatures for the full duration of the test. You can’t extrapolate out from a shorter duration test.
I have always found it hard to find this information from the manufacturers as they often treat it as commercially confidential.

If the steel faced aerated concrete product is Korok (speedwall), I note the published results for a single thickness wall is -/120/120 for an external wall and there is nothing published with a higher integrity rating without additional layers or plasterboard. I would be surprised if a single thickness would achieve a 240min integrity, as I assume if they went to the expense of a test and achieved that result, it would be published as a good selling point where integrity only is required.

Hi Geoff
The test was taken out to 240 minutes for integrity and insulation, without any extrapolation. You are correct in that the information is proprietary, but it does exist.
My query was in regard to the methodology, as the relevant fire engineer seems to consider that all fire walls cannot exceed + 140C average above ambient, or + 180C above ambient for point temperature. Reality is that it does not matter what the cold side temperature of the fire wall is, so long as the NZBC maximum levels of received radiation at the boundary are not exceeded.