Protecting runs between buildings from short

[dobby1]

You do not really want an RCD on the supply side of the cable.

It will be inconvenient in the workshop if a small fault causes the disconnection of everything.

I did think about that, and am still conflicted about it. The plan was to use 30mA RCDs in the workshop, and a 100mA at the supply. Since each machine will be on it's own circuit then I should be able to get away with trips on the supply, but yes, any trips on supply will be a huge PITA. No harm in trying it out I guess, can always failover to an MCB if things go pear shaped.

 

[ban-all-sheds]

Just for clarity's sake, I was always intent on using an RCBO type of arrangement to feed the outbuilding.

Why?

I was merely curious as to whether there was a reason why some sparks were omitting MCBs, which to me seemed a bit silly and dangerous.

They are not, and it is. And it is no much that anybody who dose it is lying if they claim to be an electrician.


Given all that you have written here, I wonder who is going to sign this (with particular attention to the highlighted parts):

I being the person responsible for the design, construction, inspection & testing of the electrical installation (as indicated by my signature below), particulars of which are described above, having exercised reasonable skill and care when carrying out the design, construction, inspection & testing hereby CERTIFY that the said work for which I have been responsible is to the best of my knowledge and belief in accordance with BS 7671:2008, amended to 2015 except for the departures, if any, detailed as follows:​

 

[ban-all-sheds]

That americans drive on the wrong side.

Where?

 

[ban-all-sheds]

The plan was to use 30mA RCDs in the workshop, and a 100mA at the supply.

Why?

 

[ban-all-sheds]

Whereabouts in the UK do you live?

Do you need Building Regulations approval for this work?

 

[SimonH2]

Those people were blunderers who had no idea what they were doing.

Indeed. Many will have seen where it says "63A" on the front of the RCD and (incorrectly) assumed that it was an overcurrent trip rating. It's one of the first things I need to fix when (if !) we move as the vendor (or rather, his father in law) made exactly this assumption.

The plan was to use 30mA RCDs in the workshop, and a 100mA at the supply.

And guess what, when one trips, you'll almost certainly find that the other does as well
The reason for that is because when you have a fault, it is very unlikely that it will only be between 30mA and 100mA. In practice, it is likely to be significantly higher than 100mA and will trip both RCDs. It is a false assumption that RCDs somehow limit the actual fault current.

Where it is necessary to have multiple RCDs in series, then they are normally configured with time discrimination. For example, there might be a 30mA/30ms RCD on the final circuit, and a 100ms RCD upstream - that way, if the fault is on the final circuit it will trip the downstream RCD and remove the fault before the upstream RCD trips.
Whether an RCD with 100ms trip time is appropriate to a circuit is a matter of designing correctly while armed with detailed knowledge.

In general you would be well adviced to try and design (cable type and installation methods) the circuit such that RCD protection of the submain cable is not required - that way there's no RCD in the house to have to go and reset when the garage lights go out. Put the RCD protection in the garage where it's easy to get to when it needs resetting.

As to what rating of MCB (or RCBO)/Fuse you need - that can only be determined after you have designed the circuit. That design will include (not exclusively) : ensuring that the protection is appropriate for the size/type/installation method of the cable; and ensuring that the prospective fault current will be sufficient to ensure that required disconnect times are met.

 

[JohnW2]

Where it is necessary to have multiple RCDs in series, then they are normally configured with time discrimination. For example, there might be a 30mA/30ms RCD on the final circuit, and a 100ms RCD upstream - that way, if the fault is on the final circuit it will trip the downstream RCD and remove the fault before the upstream RCD trips.

We tend to think that is always going to be true, but time/speed discrimination alone does not guarantee that - one really needs both the trip current and speed to be different.

If the trip currents were both the same (say 100mA), then the maximum permitted operating time of a standard RCD is appreciably greater than the minimum permitted operating time of a time-delayed ('Type S') RCD at both In and 2In, and even at 5In the two figures are the same (40ms) - so, at least at currents up to 5In, it would be possible for the non-delayed one to trip first.

If, as is the usual situation, the downstream RCD is a standard (non-delayed) 30mA one, and the upstream one is a 100mA Type S, then what you say will usually be the case (particularly if the residual current is well above the trip threshold of both devices), but not necessarily so. You may remember my recent test of a 'high current' L-E loop impedance measurement with my MFT, which resulted in the upstream 100mA Type S operating but not the downstream 30mA non-delayed one (and both RCDs were 'within spec' when subsequently tested).

Kind Regards, John

 

[ban-all-sheds]

In general you would be well adviced to try and design (cable type and installation methods) the circuit such that RCD protection of the submain cable is not required

In general he would be well advised to have someone competent do the design.

 

[stillp]

Where?

This reminded me of working at an American air base in Suffolk in the 80s, where they had a sort of cross-over arrangement for the entry lane - entered on the left, then crossed over to the right-hand side of the road while within the base!

 

[JohnW2]

This reminded me of working at an American air base in Suffolk in the 80s, where they had a sort of cross-over arrangement for the entry lane - entered on the left, then crossed over to the right-hand side of the road while within the base!

If I recall correctly, when, a long time ago, there was talk of a possible road Channel Tunnel, people were considering that sort of arrangement.

Kind Regards, John

 

[EFLImpudence]

Well, they would have had to, wouldn't they?

I'm sure Stillp means - and the base had - just white lines and possibly barriers.

Had the tunnel been built for cars, I am sure they would have had something more structurally unavoidable - a flyover.

 

[JohnW2]

I'm sure Stillp means - and the base had - just white lines and possibly barriers.

Yes, maybe.

Had the tunnel been built for cars, I am sure they would have had something more structurally unavoidable - a flyover.

Indeed - that was what was being talked about

Although obviously less dramatic, I suppose we have something along those lines at the M40/M42 junction. If one is travelling north on the M40 and wants to move onto the East-/North-bound M42, one leaves the M0 in the LH ('slow') lane and then soon finds oneself on the RH ('fast') lane of the M42. If there were barriers down the middle of those two (one-way) carriageways, one would have unavoidably moved from the 'left half' to the 'right half', courtesy of a bridge/flyover (which one might not even realise one is btravelling over).

Kind Regards, John