that the capacitive “bond” if you like between N and E does not meet the requirements of BS7671 to claim to be a TN-S setup

A capacitive "bond" is definitley not a TN-S arrangement - where the N is directly connected to Earth. It could be IT (which might be either isolated from Earth or connected to Earth via an impedance).

It might be good to establish the values of these "capacitors" - often they're just incidental from the proximity of conductors or small values intended for noise suppression and small values won't be capable of passing significant currents.

IT systems from supplies (such as inverters) that can only provide very modest fault currents can be "interesting" from a shock protection point of view. RCDs won't trip on first faults unless the capacitance is reasonably large. Tripping on second faults (with the second fault from a different live conductor from the first) may not be detectable by RCD either, so normally relies on overcurrent devices for disconnection - but if your source can't provide enough ommph to flow fuses or open circuit breakers in a short time, then that isn't going to happen either. The inverter itself might spot the overload and shutdown, or just produce a reduced voltage, but then you're relying on details of the inverter's behaviour to be sure of safety from shock (and ensure that that behaviour is reliable).

   - Andy.

that the capacitive “bond” if you like between N and E does not meet the requirements of BS7671 to claim to be a TN-S setup

A capacitive "bond" is definitley not a TN-S arrangement - where the N is directly connected to Earth. It could be IT (which might be either isolated from Earth or connected to Earth via an impedance).

It might be good to establish the values of these "capacitors" - often they're just incidental from the proximity of conductors or small values intended for noise suppression and small values won't be capable of passing significant currents.

IT systems from supplies (such as inverters) that can only provide very modest fault currents can be "interesting" from a shock protection point of view. RCDs won't trip on first faults unless the capacitance is reasonably large. Tripping on second faults (with the second fault from a different live conductor from the first) may not be detectable by RCD either, so normally relies on overcurrent devices for disconnection - but if your source can't provide enough ommph to flow fuses or open circuit breakers in a short time, then that isn't going to happen either. The inverter itself might spot the overload and shutdown, or just produce a reduced voltage, but then you're relying on details of the inverter's behaviour to be sure of safety from shock (and ensure that that behaviour is reliable).

   - Andy.

Thanks, one would like to think this has been detailed and thought through by the manufacturer.

This is a large brand, and the products are sold globally and in the UK, they are approved to G98, G100, etc. 

I am thinking maybe I am over thinking this. The manufacturer has confirmed this is normal behaviour. 

I have asked for other specifics and await their response. 

Thanks.