Cross sectional area of a protective conductors

What is the voltage and current of the system, and what is the electrode resistance to terra- firma.?


Unless this is part of a lightning system or perhaps an indoor substation,  I'd be surprised if any electrode on an LV system, could really sink enough current to need a 300mm bond during the time needed to clear a fault, and 'd have no issue with doing that  with a small no. of more conveneint sized cables.

What is the cross section of the re-bar that you are connecting to ? After all there is no point in connecting 300mm sq of copper to 500mm2 of steel ?

The adiabatic expression wouldn't size the bonding conductor - only the main earthing conductor, from which you can derive the bonding conductor size depending on the earthing arrangements.


300mm2 seems a fantastically large conductor to be using for bonding. It might be credible for earthing but would be unusual. As an example, it's perfectly feasible to operate a pair of 1.5MVA transformers with a combination of a single 95mm2 conductor and the transformer tails armouring. Assuming this is a fairly large installation, and if you really do need a 300mm2 conductor, then personally speaking, I would use two conductors if you have the opportunity to split up the electrode for testing in future - other than that, I would favour a single conductor  - although there is no reason not to use more, smaller conductors as long as you are assured they all stay in place


Regards


OMS

Well 1.39 ohms, (by the way, a surprisingly precise number,  is that a test result or calculation ?) will pass less than 200A  into earth, assuming we are talking about the bit of cable that only goes to the electrode. As such, even if the fault was never cleared, in practice a piece of 50 or 70 mmsq would probably do. The stuff that does carry the full fault current is going from the supply neutral to the CPC that goes to the enclosures for the live parts. I presume the 90kA fault current does not go very far before the impedance of the cables themselves reduce the prospective fault current quite a bit.

Can you describe, in a bit more detail, or show a sketch of, the conductor you are talking about because there is something adrift with the numbers being quoted.


If it's the connection from the switchboard back to the transformer star point then this is usually the "neutral" conductor - everything else is G/Y and is probably either equipotential bonding, or the connection to the neutral to create the reference - so even in a 5MVA installation probably isn't bigger than a range between (say) 50mm2 and 95mm2


Regards


OMS

MS01 - I'm sure you are confusing potential fault current PFC (Live to Live or Live to Neutral) with potential earth fault current PEFC (Live to earth). 


I'm no engineer so I hesitate to comment (Let alone with an installation of this size) but the I understand that the impedance of the Live to Neutral fault path  give you an approximation of the PFC - and if you multiply this by root 3 you'll get an approximation of the live to live fault current.


The live to earth (PEFC) size depends on the live to earth impedance. As shown above by AJ - this'll only be 168 Amps......And that'll be your main earth fault path I believe. 


When you say you have an earth electrode system made up of structural rebar - is this what needs bonding?


I must be missing something.........


If you look at Table 54.8 in the regs book - page 203 - this would indicate that your main bonding size should be circa 50mm if its PME or if you look at 544.1.1 - 25mm if TNS. It doesn't seem to have any limitations on this size.....

I think you need to draw the line diagram, and then look for those parts of the installation that carry earth fault current.


Typically, both your transformers are closed into the switchboard and would have to be in parallel to achieve the fault level you quote. Presuming you can identify the N-E bond as either at the transformer, or more credibly at the switchboard, you should see that by far the largest proportion of fault current will return along that conductor in an earth fault.


That conductor is referenced to earth via connections to the electrode - the resistance of which sets how much fault current can flow from bonding to rebar and back to the source point.


Assuming all of your cabling design is "normal" and has adequately sized CPC's, then the bonding cannot see a fault current of 100/1.73 kA


Regards


OMS

Still feels a bit odd - if the 1.7kA is the HV side to earth fault current, we are looking at  6.4kV to ground (for 11kV phase to phase) so perhaps the HV electrode impedance is  something like 3.8 ohms

Well perhaps not really, that is HV electrode plus the HV phase loop impedance, but  at  5MW  you are looking at perhaps 150 amps per line  on the primary at full load I'd assume the HV feed is not drooping by 10%  at that point -or there will be horrible problems of secondary side voltage drop)

Presumably then most of the HV fault loop is indeed electrode related and it is at least a couple of ohms. and may be more like 3. Part of me is then a bit surprised that HV and LV earths are combined - it does not sound a very 'cold' site (due to LV side bounce during HV faults).


In any case, you will not get thousands of amps flowing into your rebar, which is as well, as the rebar would probably need to be beefed up to take it, so the cable to the rebar can be reduced accordingly.