Question about Zs and PSCC needs some clarification thanks

[pc1966]

It's the exception of 434.5.1 if I understand it correctly. I asked some questions on it a while back and it sounds like it's very unlikely to be an issue due to the resistances between the supply and the fault quickly adding up, lowering it to something more acceptable.

What you are looking at with 434.5.1 is the "cascading value" for protective devices.

Basically in the domestic case the upstream supply 60-100A fuse limits the peak fault current to a level that downstream domestic 6kA MCBs are able to break even when the measured PSCC is above 6kA.

Also in practice you rapidly lose PSCC as you move away from the CU on any final circuit. Even if you had an infinite PSCC at the supply, to bring it to 6kA needs 38 mOhm and you get that from around 1m of 1mm T&E or around 4m of 6mm T&E.

 

[pc1966]

See this guide:

https://www.beama.org.uk/static/uploaded/992ea5f3-2044-46b2-9db965480b99d7c8.pdf

 

[pc1966]

on TNC-S , as Terre and Neutral are combine hence PSCC and PEFC isnt the same as they are combined at one point?
and on TNS as they are separate so PEFC as earth is smaller in size hence during faulty scenario when Earth and L comes in contact shouldnt it generate less fault current than PSCC when Neutral would be same size as Line conductor ?

Mostly.
In the TN-C-S case you would expect the PSCC and the PFC to be the same as they are both using the same L and PEN impedances, with the tail N and main earth connection being short and usually of negligible overall effect.

The OSG has a max TN-C-S Ze for this as 0.35 ohm which for a nominal 230V would be 660A and that would give you about a 9% drop on 60A. That is, of course, more than the regs permit for circuit VD but the DNO have different rules to follow. But as a sanity check if you measure Ze above this on a TN-C-S setup then something is wrong. Same for seeing a significant difference in PSCC and PFC.

In the TN-S case the usual expectation is the supply CPC is smaller than the N and so PFC will be less than PSCC. But that is not always the case as you can have the earth-bond going to shared metal service pipes that are in turn bonded to other supply earths and so the PFC measured in-use (for Zs at DB) is higher than the PSCC since the parallel earths have lowered the CPC impedance to below that of the supply N cable.

However, you should always base the design on the supply Ze and not the in-use Zs at DB as in the future those metal service pipes may be replaced by plastic, etc. Usually you want to be able to disconnect on the OCPD (switch-fuse, MCB, etc) as it is simple and reliable, but if the Ze is too high to permit it than you have to look at using RCDs and then the selectivity of cascading them, etc.

 

[pc1966]

Just to add to the above discussion: the DNO fuse does not have to meed 5s disconnection on earth faults as the regs for a sub-main would dictate. Again, they operate under different rules and that fuse is to protect their network from overload-style faults, not to provide shock protection ADS for downstream use as the BS7671 wiring regs wiring dictate.

Just in case anyone asks about the often quoted TN-S max Ze of 0.8 ohms being too high for even a 60A fuse!

 

[Aaron b]

Just to add to the above discussion: the DNO fuse does not have to meed 5s disconnection on earth faults as the regs for a sub-main would dictate. Again, they operate under different rules and that fuse is to protect their network from overload-style faults, not to provide shock protection ADS for downstream use as the BS7671 wiring regs wiring dictate.

Just in case anyone asks about the often quoted TN-S max Ze of 0.8 ohms being too high for even a 60A fuse!

but would this then be treated like a TT with all conductors (but not busbars) needing to be double insulated until the first protective device, assuming the consumer unit was metal and an RCD was required for fault protection for outgoing circuits?

 

[pc1966]

but would this then be treated like a TT with all conductors (but not busbars) needing to be double insulated until the first protective device, assuming the consumer unit was metal and an RCD was required for fault protection for outgoing circuits?

Having a L-E fault on the incomer is always a bad day, TT or TN!

These days you would be using double-sheathed tails anyway, and (hopefully) a proper gland for them, so you would meet that sort of "no single point of failure" anyway even with a metal CU.

Where it gets more complicated is a SWA sub-main, it is not double insulated in any real sense, though it has the metal armour for protection to a degree. I would be considering adequate protection to be placed before any such run (even if < 3m) so if you can't disconnect on any supply fuse safely I would be considering an up-front delay RCD.

 

[Aaron b]

Having a L-E fault on the incomer is always a bad day, TT or TN!

These days you would be using double-sheathed tails anyway, and (hopefully) a proper gland for them, so you would meet that sort of "no single point of failure" anyway even with a metal CU.

Where it gets more complicated is a SWA sub-main, it is not double insulated in any real sense, though it has the metal armour for protection to a degree. I would be considering adequate protection to be placed before any such run (even if < 3m) so if you can't disconnect on any supply fuse safely I would be considering an up-front delay RCD.

Yeah I doubt it's normally an issue. Split load boards and SWA incomers are likely the only trouble.