Bonding RSJ question.

[mainline]

Sorry, I forgot to mention is a TN-S earthing system.

As long as you can guarantee that it is TN-S and isn’t taken from a TN-C network.

 

[fatboyslim]

As long as you can guarantee that it is TN-S and isn’t taken from a TN-C network.

Yeah definitely TN-S. We have two separate private LV transformers, as the site is spilt into two substations.
Recently had taps changed on both as voltage was around 250-260v now sitting comfortably at 230-240v. Some extra savings in the company's pocket!! Shame I don't benefit from seeing them in my payslip

 

[davesparks]

Could I bond the nearest RSJ of the storage building to the metal container which is 40 metres away, as the RSJ structure is bonded near the sub main? This would save 70 metres of cable.

Yes, subject to the steel framework being suitably continuous, not likely to be removed etc, see reg 543.2.1(vii) and reg 543.2.6

 

[pc1966]

Correct.I always bond the metalwork to the CPC .I understood from the OP that he wondered about running a separate bonding cable to the metallic structure.

My apologies, I was not really thinking of what the OP actually asked about.

 

[pc1966]

You could consider this as supplementary bonding so minimum would be 4mm copper.

If you are sure it is only ever going to be TN-S (so no PME conditions to satisfy) and only ever going to be fed via 1.5mm or so cable (so 4mm well below adiabatic for any supply OCPD) that should be sufficient.

There might be other considerations like lightning protection, mechanical strength, or potential increases in supply capacity that mean 10mm or more might be appropriate, but I guess not just now.

 

[fatboyslim]

Thanks for all the advice guys. Much appreciated

 

[narkedatninety]

Is the metal container itself not an extraneous conductive part ?

Very unlikey to introduce an outside potential into the system.

 

[narkedatninety]

But if the metalwork is not bonded to the CPC system they risk being at differing potentials, say should the CPC be on a faulted-PEN segment of the supply network.

So while all structural metal work might be at the same potential, any class I equipment may be at another. So bonding is still needed.

Not neccessarily. Bonding it may mean you are actually introducing a fault-path. We test to ascertain whether it is extraneous and whether it is likely to introduce an outside potential into the system. If the test proves otherwise, we do not bond.