Power cable connection on a mobile installation to External equipment using a MIL-DTL-38999 connector.

The regulations  say "Socket-outlet. A device, provided with female contacts, which is intended to be installed with the fixed wiring, and intended to receive a plug. A luminaire track system is not regarded as a socket-outlet system.", is it part of the fixed wiring?

yes it has an onboard generator, no it is not powered up while on the move, just when stationary.

We discovered recently that the designer of the generator unit used a 1/500 winding current transformer when they should have used a 1/1000. This means it was actually tripping at just under 15mA  - no surprise it couldn't take the inrush. (Long and tedious story. Please don't suggest the supplier changes it, been there done that still have a bleeding forehead from repeatedly banging my head against the wall.)

None of the RCD setting adjustment process from the Dev stage was written down (before my time), so I don't know if they gradually turned it up or just went straight to 300mA. Whether they tried a time delay I really couldn't say. The RCD does have detents at 100mA etc, and an adjustable time delay. We are going to have to go through that process now to determine if a lower setting would work. The lower the better in my opinion. 

I think in the mean time I'm going to have to go with 'It's not a general use socket and is a specialised connector supply one aspect of the system only' for now. Therefore not a socket in the sense that the regs mean.

I'd also play the for use by  'skilled and instructed persons only' card as well if you can. It is really not after all anything like a socket for use by ordinary/unskilled persons.
Indeed it is not being plugged and unplugged at all while genset is on I imagine? 

Fully understand the difficulty of changing the plans after something has been built and delivered - concession forms and engineering change costs etc. 

Mike.

Newfutile said:

The regulations  say "Socket-outlet. A device, provided with female contacts, which is intended to be installed with the fixed wiring, and intended to receive a plug. A luminaire track system is not regarded as a socket-outlet system.", is it part of the fixed wiring?

BS 7671 also defines a cable coupler. This pairing has a 'plug' and a 'connector' ... however, this connector doesn't have an IEC, CENELEC or British standard ... so, to use them in an electrical installation for which BS 7671 applies (including those to BS 7909) would require a declaration of some sort (see BS 7671, Regulation 133.1.1 and Section 511).

Samantha Rendell said:

Given that it uses the 38999 connectors and it is solely dedicated to the dismounted control systems that form part of the system is this a valid assessment? 

Are the 'control systems' part of an electrical installation, or Machinery as defined in the Supply of Machinery (Safety) Regulations? If the latter, then BS EN 60204-1 might apply, rather than BS 7671?

gkenyon said:

The regulations  say "Socket-outlet. A device, provided with female contacts, which is intended to be installed with the fixed wiring, and intended to receive a plug. A luminaire track system is not regarded as a socket-outlet system.", is it part of the fixed wiring?

BS 7671 also defines a cable coupler. This pairing has a 'plug' and a 'connector' ... however, this connector doesn't have an IEC, CENELEC or British standard ... so, to use them in an electrical installation for which BS 7671 applies (including those to BS 7909) would require a declaration of some sort (see BS 7671, Regulation 133.1.1 and Section 511).

Samantha Rendell said:

Given that it uses the 38999 connectors and it is solely dedicated to the dismounted control systems that form part of the system is this a valid assessment? 

Are the 'control systems' part of an electrical installation, or Machinery as defined in the Supply of Machinery (Safety) Regulations? If the latter, then BS EN 60204-1 might apply, rather than BS 7671?

Semantics again!

Is it an electrical installation or a machine? Or neither?? My first thought was whether this equipment is in scope or not. 110.1.1 says that the Regulations apply to electrical installations, so if it is not one, BS 7671 does not apply.

Is the device in question a "socket-outlet". Newfutile questioned this. I think that it is difficult to get away from the connexion being a socket and plug, but I am far from convinced that it is a "socket-outlet" as defined. More importantly, 411.3.3 is all about supplying electrical energy (the Grid in a building, PV, generator, WHY) from a socket upstream to an appliance downstream.

Most things are mobile if you have the capacity. I envisage a "machine", for want of a better word, which can be transported (perhaps to a battlefield, or more importantly, an exercise area), but which has to be broken down into components to make it mobile. If you didn't use some form of plug and socket, how else would you put it together? (I discount the notion of terminal blocks in enclosures.)

Be aware that there are significant relaxations of the machinery directive, & it is completely dis-applied for fairground, police or military equipment, or several kinds of vehicle, though generally not otherwise normal equipment that just happens to be mounted on a vehicle.

quoting the 2023 version.

Semantics indeed!. Actually it's both. We have several standards we have to comply by in our requirements, one of which points to compliance with BS7671 for the electrical system.

This discussion has been very helpful. Thank you to all :-)

Chris Pearson said:

Is it an electrical installation or a machine? Or neither?? My first thought was whether this equipment is in scope or not. 110.1.1 says that the Regulations apply to electrical installations, so if it is not one, BS 7671 does not apply.

Agreed.

Chris Pearson said:

Is the device in question a "socket-outlet". Newfutile questioned this. I think that it is difficult to get away from the connexion being a socket and plug, but I am far from convinced that it is a "socket-outlet" as defined. More importantly, 411.3.3 is all about supplying electrical energy (the Grid in a building, PV, generator, WHY) from a socket upstream to an appliance downstream.

I'm also not convinced it's definitely a "socket-outlet".

gkenyon said:

however, this connector doesn't have an IEC, CENELEC or British standard ... so, to use them in an electrical installation for which BS 7671 applies (including those to BS 7909) would require a declaration of some sort (see BS 7671, Regulation 133.1.1 and Section 511).

However, if BS 7671 is the standard to use, then this is still relevant ...

Yes, we have a declaration. BS7671 does not cover military 'equipment' but the clients stipulated compliance with certain Def-Stan's, one of which said we must comply with BS7671 for the electrical installation side of the 'equipment' where reasonably practicable.

The connectors are a must have because of the use of the mobile equipment, and while they don't have an appropriate IEC, CENELEC or BS the design specifications make it very safe for our purposes and will not result in a lesser degree of safety it complies under section 511. 

I agree that while it is a socket I am also not convinced it can be defined as a socket-outlet. Hence the question, as opinion is a little divided here because of the combined use and architecture it is a bit of a grey area that isn't specifically covered  by any of the standards.

As much as I would prefer the belt and braces approach of a 30mA limit, the added complication of the current transformer windings reducing the set trip current to just under half makes it difficult to ascertain if an actual trip current of 30mA would cope with the inrush currents.

The adjustable RCD has detents at 0.03, 0.1, 0.3, 0.5..... so we can test at 100mA which will give a trip current of around 45mA. I don't know if this was done at the Dev stage because there are no notes of any kind. When I spoke to my predecessor about it, it was 7 years after the fact so the details were a bit, shall we say murky.

if an actual trip current of 30mA would cope with the inrush currents.

Can you tell if the inrush current has a significant "leakage" component? There have been quite a few reports of RCDs tripping on switch-on, not because there's a particularly large earth leakage (residual) current during switch-on, but rather the L-N currents get large enough to saturate the sense coils/ferrite , which then don't entirely balance and trips the RCD. Sometimes selecting a RCD with a larger current carrying rating (In) (as distinct from the residual current rating IΔn) can help solve that problem without reducing the level of protection.

Similarly I have heard tell of some nuisance trip resistant 30mA RCDs which as well as some extra filtering incorporate a slight delay although - but still open within the requirements for additional protection. Might be worth a look.

   - Andy.

if you try the 100ma (really 50ma in this case) setting, also set the 0,1 second delay at the same time. - the delay is not long enough to seriously impact the shock survival risk, but the worst of any start-up surge should be over before the device unlatches. Of course the leakage could be too high in the steady state too, but it is a fast way to find out.

Mike.

if you try the 100ma (really 50ma in this case) setting, also set the 0,1 second delay at the same time. - the delay is not long enough to seriously impact the shock survival risk, but the worst of any start-up surge should be over before the device unlatches. Of course the leakage could be too high in the steady state too, but it is a fast way to find out.

Mike.