Although it saves digging to place the ducts one above the other, the rating will be higher if they are side by side - cables in the upper layers will get heated to a degree by those below. I read the diagram as 150mm dia ducts on 300mm centres so the whole block is 1.2m metres wide ?

According to literature a standard density concrete, at a temperature of 20°C, has a value between 1.33 W/mK and 1.95 W/m.K

ERA 69-30 report assumes a soil thermal resistivity of 1.2 K.m/W, which corresponds with typical soil thermal resistivity for the UK in Annex A (A.22.2) of BS IEC 60287-3-1 . BS7671 uses a figure of about twice the thermal resistance.

So to a good first stab, the concrete is a little bit more thermally resistive than typical UK soil, but even so a better thermal conductor than sandier drier soils.

One way to look at the problems is to calculate the total (all cables) watts per meter dissipation, and then use that figure for the heating at a distance, but then solve the ducts, as if in that higher ambient.

Mike.

If these cables are only to be used in a grid power failure, it MAY be acceptable to make rather optimistic assumptions about cable loading. For long term use, a limiting factor is the gradual heating up of the concrete and surrounding subsoil, not applicable for short term only use. Significant warming can take weeks or even months, how long will the power cut be ? And even if the cables do get hotter than normally allowed, this may be acceptable for short term use. Cable life might be reduced from say 50 years to say 5 years, and that is not 5 calendar years, but 5 years of power cuts.