Assessment of the burning behavior of protected and unprotected cables and cable trays in nuclear installations using small- and large-scale experiments

Electric installations and cables are a main fire risk source in industrial buildings and power plants. In general, cables and cable systems are associated with flash-over phenomena due to pyrolysis of fuel gases induced by the heat of an adjacent fire, fire spread along cable trays affecting additional areas besides the fire origin, being an ignition source due to malfunction. If burning, cables can emit large amounts of smoke and toxic products affecting occupants as well as the long-term functionality of structure and installations. Paying attention to these risks has led to the development of fire retardant non-corrosive (non-halogenated) cables which are qualified to reduce the individual or all of the risks mentioned. For existing installations in industrial buildings and power plants with halogenated cables, different protection measures are available and widely applied retroactively. Important protective measures are intumescent or ablative coatings, cable casings and bindings. For qualification of the effects of the protection measures, small-scale tests investigating a single cable specimen as well as large-scale cable tray test setups have been developed and carried out in the last 20 years at iBMB. In this paper, these test results are analysed regarding their effects on the heat release, ignition time and fire spread over cable trays. Furthermore, national and international research projects have investigated the burning behaviour of different cable types, tray installations, tray loading and spacing and ventilation conditions. As a conclusion, the main outcomes of past researches are summarized. Influence factors (e.g. pre-heating due to high power utilization, influence of cable aging) which have not been accounted for in detail are emphasized. The modelling of unprotected cables has been internationally studied in recent years. For future applications, the question of applicability of recently developed sub-models on the fire behaviour of protected cables has to be answered. The results presented in this paper may provide the basis for the planning of further validation experiments to the fire performance of protected cables.