sizing of cables for adjustable short-circuit threshold

Perhaps take a look at 433.1.1 noting the words "rated current or current setting" and the sentence regarding the current setting selected is the effective In of the device.


However, that's for overloads


typically under short circuit, reducing the effective In also reduces the threshold at which the short circuit current operates the device and in what time (ie the inverse characteristic) - so you still need to design for the SC condition at the actual fault level using the effective disconnection time. It's then up to you if you decide that upstream devices don't provide any cut off characteristic or are likely to maloperate to an extended disconnection time (and further upstream doesn't act.


As an example, just to maintain temperature rise characteristics sensibly within the switchboard, I might well have a 3200A incoming ACB and put a 1600A ACB device dialed down to 1250A as an outgoing device and then use say 4 x 185mm2 conductors to feed the downstream switchboard - and then use say 630A devices dialled down to 400A in that downstream switchboard with the outgoing circuit assessed accordingly. Under short circuit, I might ignore the cut off characteristics of the 1600A/1250A device and assume the full prospective current is applied to the 630A/400A device. But I would probably ascertain the cut off of the 3200A device and use that for the 1250A outgoer.


Essentially, you only need to design for what you consider credible rather than what might be possible - how far you go in each direction is up to your appetite for risk - BS 7671 is only minimum compliance and would allow what you describe without any "whatiffery" involved


Regards


OMS

Perhaps take a look at 433.1.1 noting the words "rated current or current setting" and the sentence regarding the current setting selected is the effective In of the device.


However, that's for overloads


typically under short circuit, reducing the effective In also reduces the threshold at which the short circuit current operates the device and in what time (ie the inverse characteristic) - so you still need to design for the SC condition at the actual fault level using the effective disconnection time. It's then up to you if you decide that upstream devices don't provide any cut off characteristic or are likely to maloperate to an extended disconnection time (and further upstream doesn't act.


As an example, just to maintain temperature rise characteristics sensibly within the switchboard, I might well have a 3200A incoming ACB and put a 1600A ACB device dialed down to 1250A as an outgoing device and then use say 4 x 185mm2 conductors to feed the downstream switchboard - and then use say 630A devices dialled down to 400A in that downstream switchboard with the outgoing circuit assessed accordingly. Under short circuit, I might ignore the cut off characteristics of the 1600A/1250A device and assume the full prospective current is applied to the 630A/400A device. But I would probably ascertain the cut off of the 3200A device and use that for the 1250A outgoer.


Essentially, you only need to design for what you consider credible rather than what might be possible - how far you go in each direction is up to your appetite for risk - BS 7671 is only minimum compliance and would allow what you describe without any "whatiffery" involved


Regards


OMS