Log in to the online community

Want to post a reply? You'll need to log in

RCD Protection at Data Cabinets

34 Replies

  • New Question

RCD Protection at Data Cabinets

Posted by Bob C on Mar 14, 2019 2:21 pm

I have a Client that will not provide a Risk Assessment to omit an RCD Commando Socket for a Data Cabinet. He wishes to employ a local isolator for the Cabinet, however he is concerned that the PDU Strip that arrived on site from the factory and pre-fitted within the data cabinet would still need an RCD. It is my opinion that the Factory built Data Rack is classed as a piece of equipment and as such does not form part of the Contractor's installation. If the Rack had arrived on site without the PDU strip fitted and it was installed by the Contractor then this would form part of the installation and therefore would require an RCD. Any advice/assistance would be appreciated.

Re: RCD Protection at Data Cabinets

Posted by AJJewsbury on Apr 2, 2019 1:59 pm

Bob C:
I have a Client that will not provide a Risk Assessment to omit an RCD Commando Socket for a Data Cabinet. He wishes to employ a local isolator for the Cabinet, however he is concerned that the PDU Strip that arrived on site from the factory and pre-fitted within the data cabinet would still need an RCD. It is my opinion that the Factory built Data Rack is classed as a piece of equipment and as such does not form part of the Contractor's installation. If the Rack had arrived on site without the PDU strip fitted and it was installed by the Contractor then this would form part of the installation and therefore would require an RCD. Any advice/assistance would be appreciated.

Apologies for the apparent tangents - we're still debating the finer points.

So far we seem to agree that if the PDUs contain BS 1363 socket outlets then the 30mA RCD requirements would apply - so somehow or other that would need to be dealt with - either by a 30mA RCD in the supply circuit, or one or more 30mA RCDs within the PDU arrangement (or better still by risk assessment to eliminate the need for RCDs altogether - probably what most PDU manufacturers assume). Ditto if there are BS EN 60309 sockets (or even BS 546 sockets) rated 32A or less within the rack.

We seem to have mixed opinions still if other types of 'connectors' are used - e.g. the IEC 60320s chassis mount outlets often found on PDUs.

To me, trying to supply an IT rack without an risk assessment to omit 30mA RCDs is rather like trying to run a taxi service without a driving licence - while it's not impossible, the contortions you have to go through (hard wiring things that should be pluggable, perhaps needing 30mA RCDs when they're not really needed) make it a very dubious choice in the long run.

  - Andy.

Re: RCD Protection at Data Cabinets

Posted by gkenyon on Apr 2, 2019 2:05 pm

AJJewsbury:

Bob C:
I have a Client that will not provide a Risk Assessment to omit an RCD Commando Socket for a Data Cabinet. He wishes to employ a local isolator for the Cabinet, however he is concerned that the PDU Strip that arrived on site from the factory and pre-fitted within the data cabinet would still need an RCD. It is my opinion that the Factory built Data Rack is classed as a piece of equipment and as such does not form part of the Contractor's installation. If the Rack had arrived on site without the PDU strip fitted and it was installed by the Contractor then this would form part of the installation and therefore would require an RCD. Any advice/assistance would be appreciated.

Apologies for the apparent tangents - we're still debating the finer points.

So far we seem to agree that if the PDUs contain BS 1363 socket outlets then the 30mA RCD requirements would apply - so somehow or other that would need to be dealt with - either by a 30mA RCD in the supply circuit, or one or more 30mA RCDs within the PDU arrangement (or better still by risk assessment to eliminate the need for RCDs altogether - probably what most PDU manufacturers assume). Ditto if there are BS EN 60309 sockets (or even BS 546 sockets) rated 32A or less within the rack.

We seem to have mixed opinions still if other types of 'connectors' are used - e.g. the IEC 60320s chassis mount outlets often found on PDUs.

To me, trying to supply an IT rack without an risk assessment to omit 30mA RCDs is rather like trying to run a taxi service without a driving licence - while it's not impossible, the contortions you have to go through (hard wiring things that should be pluggable, perhaps needing 30mA RCDs when they're not really needed) make it a very dubious choice in the long run.

  - Andy.

 

I agree with Andy for the most part. It's probably get the risk assessment, or fit the RCDs to be honest ... the appliance-outlet is a technicality (and only that). If a client's not happy with the risk assessment, they won't be happy getting out of the RCD on such a technicality anyway!
Graham Kenyon, Managing Director, G Kenyon Technology Ltd www.gkenyontech.com G Kenyon Technology Ltd Logo © G Kenyon Technology Ltd 2015-2019

Re: RCD Protection at Data Cabinets

Posted by AJJewsbury on Apr 2, 2019 2:17 pm

 

FCU -> Flex -> BS 1363 outlet (as in the pendant socket mentioned before, or a autorewind reel with a trailing socket on the end of the flex) - apparently a connector rather than a socket, so no RCD requirement?

I think this is a socket-outlet, or an accessory with socket-outlets, so I think 411.3.3 (i) applies.
But BS 7671's definition of a socket outlet specifies that it's 'installed with the fixed wiring' (which I take to mean connected to the fixed wiring rather than at the same time as the fixed wiring) - sockets one the end a bit of free flex doesn't seem to match that at all. They do seem to match BS 7671's definition of a Connector though.

Life would be so much simpler if BS 7671 definitions just said that equipment was a 'xyz' if it complied with a standard that had the word 'xyz' in its title (not least in the old "is an FCU a distribution board?" debate) - but whether through historical accident or otherwise, JPEL/64 seem to have its own opinions on such matters. We surely can't just ignore Part 2 when reading BS 7671?

  - Andy.

Re: RCD Protection at Data Cabinets

Posted by Bob C on Apr 2, 2019 2:44 pm

Guys, thanks for the responses it is much appreciated. Our Client appears to be going down the route of an Isolator to supply the Cabinet with an IEC Connector in the rack. I still have concerns on this arrangement because the connectors in my opinion are effectively still sockets (albeit non standard) that can be plugged/unplugged to suit the rack's internal configurations. I would be much happier with this set-up if the Client provided a Risk Assessment so I could insert a condition that any internal power re-configuration is carried out by a Qualified Electrician and not an IT Technician, not perfect but gives a bit of comfort.
I have a situation where some of the IT cabinets are 3 Phase and I dread the thought of an IT Technician interconnecting the power to dual fed Servers for obvious reasons.
Bob C

Re: RCD Protection at Data Cabinets

Posted by gkenyon on Apr 2, 2019 2:52 pm

Bob C:

I have a situation where some of the IT cabinets are 3 Phase and I dread the thought of an IT Technician interconnecting the power to dual fed Servers for obvious reasons.
 

I wouldn't worry about that, provided the outlets are single-phase. In servers or network with multiple PSUs, they are usually single-phase only, and the supplies are not "bussed" on the AC side.
Such equipment is made to use in situations where some of the supplies are connected to UPS. Even if the two separate supplies were on the same phase (albeit potentially from different transformers in redundant supply situations), the UPS is not "phase locked" with the live supply (or another UPS), and therefore you might get times where there's 480 V between Line conductors (of the separate supplies), but each enclosed PSU only sees 230 V to earth (and hopefully Neutral).
Graham Kenyon, Managing Director, G Kenyon Technology Ltd www.gkenyontech.com G Kenyon Technology Ltd Logo © G Kenyon Technology Ltd 2015-2019

Re: RCD Protection at Data Cabinets

Posted by mapj1 on Apr 2, 2019 3:07 pm

The 'obvious' fear of 3 phases is probably over stated, so long as the neutral is not shared and then accidentally lost.
any one load will see only a single phase supply,and redundant supplies don't merge outputs until they are a ground referred DC.
It takes a certain incompetence to get the phase to phase voltage anywhere that only 230V is expected, and when it happens ADS normally sorts it safely although some kit may be lost.
regards Mike

Re: RCD Protection at Data Cabinets

Posted by AJJewsbury on Apr 2, 2019 3:17 pm

But this is precisely why I asked the specific question whether you were advocating 30 mA RCD protection for couplers as well!
In fact, I agree it's a difficult situation to grasp, but at the end of the day the Reg exists for "couplers" separate to "socket-outlets", and I can't see the difference between "appliance outlet" and "free outlet" in the context of a cable coupler.
So, the question - should couplers be included in 411.3.3 ? I sort of think you're saying they should be.

From the point of view that if an ordinary person has a lead with a plug on the end of it (be it BS 1363, BS EN 60309, BS EN 60320 or anything else), they're just going to plug it into anything that fits - they're not going to be bothered if BS 7671 considers it a socket-outlet or a connector. All else being equal (i.e. unless a risk assessment says otherwise) the risks will be similar - so, yes, it would seem logical to me to treat them similarly.

It would seem more that a little absurd that where a 32A BS EN 60309 wall mounted socket needs 30mA RCD protection, a wall mounted isolator with a stuffing gland supplying 6" of flex with a 32A BS EN 60309 female coupler on the end of it, in the same situation, would be exempted. (I note that the title of BS EN 60309 is "Plugs, socket-outlets and couplers..." so that's no help.)

I do take your point however, that compliance with BS 7671 isn't always the same thing as complying with my idea of logical. But I still think we need to deal with BS 7671's definitions when reading BS 7671 regulations - rather than definitions (or titles) from other standards - and to me the chassis mounted C13s look to better fit to BS 7671's definition of a socket outlet than to its definition of a connector. Whether it's a socket outlet of itself, or other equipment incorporating a socket outlet (which may or may not have to comply with table 55.1 etc.), is to me a side-issue.

Maybe we're just going to have to agree to disagree on this one...

  - Andy.

Re: RCD Protection at Data Cabinets

Posted by AJJewsbury on Apr 2, 2019 3:24 pm

IT cabinets are 3 Phase and I dread the thought of an IT Technician interconnecting the power to dual fed Servers for obvious reasons.

I totally agree with Mike and Graham - there's absolutely no problem mixing phases on modern equipment (we did it frequently with dual-PSU servers - typically one from mains and the other from a UPS so there might have been more than 400V between them!) - there's not even any need for a 'Danger 415V' sticker any more - since Uo is still only 230V.

  - Andy.

Re: RCD Protection at Data Cabinets

Posted by gkenyon on Apr 2, 2019 3:33 pm

It would seem more that a little absurd that where a 32A BS EN 60309 wall mounted socket needs 30mA RCD protection, a wall mounted isolator with a stuffing gland supplying 6" of flex with a 32A BS EN 60309 female coupler on the end of it, in the same situation, would be exempted. (I note that the title of BS EN 60309 is "Plugs, socket-outlets and couplers..." so that's no help.)

Precisely the same thing could be said about lighting couplers, and to an extent installation couplers, but see Note 3 to Reg 411.3.3 ... and again 553.2.1 (assume not domestic/household premises so 411.3.4 doesn't apply).

I think however, that precisely because all of these different uses of effectively what is a "socket-outlet" exists, and because we have a requirement for "general user-accessible socket-outlets" to have some basic safety standards for the electrical installation (e.g. shuttered socket-outlets) is why it's difficult to put together a coherent set of requirements.


One last thought on the "coupler" issue.

There's a requirement for household and similar use socket-outlets to be shuttered. Many IEC 60320 outlets are not shuttered. You've said "What's the difference"?

Perhaps BS 7671 should make the statement that if a coupler is accessible in a household or similar use, then it should be shuttered as well as RCD protected?

But to be brutally honest, what's the point if we've all got appliances with supply cables having unshuttered free IEC 60320 outlets on them?
Graham Kenyon, Managing Director, G Kenyon Technology Ltd www.gkenyontech.com G Kenyon Technology Ltd Logo © G Kenyon Technology Ltd 2015-2019

Share:

Log in

Want to post a reply? You'll need to log in