Type B rcd availability

Hang on, this “tripping of type A or AC” is NOT why you fit a type B! A type B should trip just as a type A or AC does, it is only that it is not STOPPED from NOT tripping by DC faults. You have a problem here that is probably inrush rather than DC faults. It is possible that type Bs are more resistant to this unspecified parameter, but there is no reason not to use a type AC or A with a time delay (S type). A type B should see a single half-cycle surge as a DC fault, but from your comments does not, perhaps the inrush lasts several cycles, although I don't quite see why it should unless there is some internal series resistance. 

Andy, whilst it is theoretically possible that there could be an internal DC short to Earth, surely reducing or eliminating this possibility would be a DESIGN requirement? Are we simply making the wiring regulations correct improper designs (again - EVs)? I have an invertor where this is not possible by any reasonable mechanism except poking a bare wire into the equipment. Have we got completely to the “idiot-proof” point for everything? Surely if I poke wires into things I should expect some kind of risk? Note that such a short would pop the CPD due to excess current, or perhaps the short would be in the range where the CPC is raised to a dangerous voltage but not quite blow the CPD? You may like to calculate that range for a 32A supply and a 2kW inverter. Is it credible?

Hang on, this “tripping of type A or AC” is NOT why you fit a type B! A type B should trip just as a type A or AC does, it is only that it is not STOPPED from NOT tripping by DC faults. You have a problem here that is probably inrush rather than DC faults. It is possible that type Bs are more resistant to this unspecified parameter, but there is no reason not to use a type AC or A with a time delay (S type). A type B should see a single half-cycle surge as a DC fault, but from your comments does not, perhaps the inrush lasts several cycles, although I don't quite see why it should unless there is some internal series resistance. 

Andy, whilst it is theoretically possible that there could be an internal DC short to Earth, surely reducing or eliminating this possibility would be a DESIGN requirement? Are we simply making the wiring regulations correct improper designs (again - EVs)? I have an invertor where this is not possible by any reasonable mechanism except poking a bare wire into the equipment. Have we got completely to the “idiot-proof” point for everything? Surely if I poke wires into things I should expect some kind of risk? Note that such a short would pop the CPD due to excess current, or perhaps the short would be in the range where the CPC is raised to a dangerous voltage but not quite blow the CPD? You may like to calculate that range for a 32A supply and a 2kW inverter. Is it credible?