Rules and Guidelines
Want to participate in the discussions? Please read our Community Rules and Guidelines. Need some help? Visit Support and Answers.

  • Replies 4 replies
  • Subscribers 40 subscribers
  • Views 113 views
  • Users 0 members are here
Options
You may also be interested in

BS 7671 Thermal Insulation Derating Factors for Cables Passing Through Fire-Stopped Penetrations

Kai Tai Fok

I'm interested in hearing how others apply BS 7671 thermal insulation derating factors in practice when cables only pass through a relatively short section of thermal insulation or fire stopping.

BS 7671 Appendix 4 provides derating factors for cables surrounded by thermal insulation, and in some cases the resulting reduction in current-carrying capacity can be significant. However, many building installations only have cables passing through localised fire-stopped penetrations, fire collars, fire pillows, or fire batt systems, typically over a short distance compared with the total cable run.

My question is:

Do you apply the thermal insulation derating factor to the entire cable length and its current-carrying capacity, or do you consider the effect to be more localised?

From a heat transfer perspective, the highest cable temperature will occur within the insulated/fire-stopped section, but heat can also be dissipated longitudinally along the conductor and sheath into adjacent cooler sections of cable. This suggests the thermal behaviour may not be equivalent to a cable being continuously surrounded by thermal insulation over its entire length.

Has anyone come across:

  • Guidance from BS 7671, IEC 60364, ERA reports, manufacturers, or DNOs on this specific situation?
  • Any accepted industry practice for short fire-stopped penetrations?
  • Thermal modelling, testing, or engineering assessments that justify a less conservative approach than applying the full derating factor to the complete circuit?

I'd be interested in both the theoretical basis and how designers, contractors, and reviewers typically approach this in practice.

  • You definitely only need to consider the section of the cable that is surrounded by thermal insulation, rather then the entire cable length. Heat loss along the length of the cable to cooler sections is explicitly acknowledged in BS 7671 - see the table in section 2.6 of appendix 4.

    That said, it might be that the fire stopping materials are less thermally resistive than purpose designed thermal insulation ... so depending on the actual material the situation might be even less onerous again. As an extreme, if cement mortar is used to make good the holes, the thermal effect would be similar to direct embedding (method C) which probably means no re-rating at all. Your more plastic intumescent materials might need closer examination.

       - Andy.

  • in reply to AJJewsbury
    Thank you very much for your helpful explanation, that aligns with my understanding that the thermal effect should be local to the insulated section rather than the entire circuit, particularly given the heat dissipation available through adjacent cable lengths. I also agree that the fire-stopping material itself is likely to be the key factor. My main uncertainty is how to justify this in a design calculation, as BS 7671 provides derating factors for thermal insulation but does not give a clear method for localised fire-stopped penetrations. 
  • in reply to Kai Tai Fok

    There is also the interaction with grouping factors - if cables that are all in a duct are then rammed very tight through a single hole in the firestopping, that is thermally much worse than if the cables are spaced so that in effect each passes though a different bit of the fire stop and there is  a spacing of a few cable diameters.
    With the best will in the world it is not reasonable to expect the authors of the regs to foresee every possible combination of installation conditions, and fire stopping is very variable, so some situation specific thinking is often needed, though this may be as simple as selecting the nearest standard case, and assuming its a bit better or worse than that.

    Intumescent materials (that swell in the heat of the fire) do not even need to touch the cables they protect,  so there is ventilation while other methods may be more like sand bags or putty and close the aperture even when there is  no fire.

    In general a short section in the middle of the cable running a bit warm is not dangerous - after all PVC does not suddenly explode above 90C, or even above 100C - drop an offcut in your tea or coffee and it does not suddenly melt like butter... (might be better to try that in someone else's drink) Mildly overheated cables tend  to age rather faster than normal and become less flexible, but this is quite a slow effect.
    This and the ability of heat to flow along the cable once a hotter region is established tends to protect cables that pass through joists or thin sheets of insulation and a similar logic applies. (Its probably safe to assume any common firestop is less good as a thermal insulator than real insulation, or it would be sold as insulation as well. )

    However if your chosen design pushes right up against the cable ratings before adding the firestopping, and there is a significant (many cable diameters) tightly embedded length then it might be wise to revisit the design.
    Mike

  • In my opinion, a cable passing through a short fire-stopped penetration should not automatically be considered thermally equivalent to a cable continuously surrounded by insulation.
    From a heat transfer perspective, the fire-stopped section creates a local thermal bottleneck and will likely be the hottest point on the circuit. However, conductor heat is not confined to that point alone. Longitudinal heat flow through the conductor, insulation and sheath allows dissipation into the adjacent cooler cable sections. Therefore, a cable traversing a 100 mm fire batt will not exhibit the same thermal profile as a cable embedded in insulation for several metres.

    In practice, I would assess:

    • Length of cable within the fire-stopping material.
    • Circuit loading relative to cable capacity.
    • Number of grouped cables within the penetration.
    • Thermal properties of the fire-stopping product.
    • Available design margin.

    For typical fire batt, fire pillow or sleeve penetrations of short length, I would generally regard the effect as localised rather than circuit-governing, provided the circuit is not operating close to its thermal limits. Where the enclosed length approaches or exceeds the 0.5 m threshold, or where the circuit is heavily loaded, I would be more inclined to apply the Appendix 4 derating factors or undertake a detailed assessment.

    I would be interested to know whether anyone has access to IEC 60287 calculations, ERA test reports, manufacturer data or fire-stop testing that quantifies conductor temperature rise through short fire-stopped penetrations, as this would help bridge the gap between the simplified assumptions of BS 7671 and actual installation conditions.

Community Rules & Guidelines