This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

Data cables in electrical control panels

Hi, 

Almost all industrial control panels and others like lighting / automaton, have devices inside of them which have Ethernet ports, connected generally to network switches within the panels. 

BS7671 is quite strict when it comes to routing data and power cabling; in control panels, data cabling (so Ethernet over Cat5 or Cat6 or indeed RS232 etc) is generally in the same trunking as AC (single and three phase). The cables land into the panel separated from AC, but are then right on top of AC inside of the panel itself.

Obviously panel design standards are different. Does anyone know the exclusions / best practice for this?

There are also other subjects like Hybrid cables, for EV chargers, that I think come into question. 

Are there any special distances/creepages that need to be maintained in control panels between power and data? 

Thanks. 

  • Doesn’t BS7671 just cover the installation of the cable?

    The cable itself is manufactured and CE certified. If the cable manufacturer complies with guidance set out in whatever standard applies to the cable itself (the power and the data) and they can prove the method bundling as they are, does not cause issues with data transmission, surely when installing this, it complies with BS7671 and BS6701, which makes firm reference to following manufactures instructions? 

  • I do not agree.

     Then that standard is either incomplete or technically unjustifiable, I do not have it to hand to see their safety explanation for 50mm.  It should not matter which standards we look at, if they are not based on the real physics that applies to the situation, they are unhelpful.

    We know that wires from the various 'bands' have to meet somewhere, not least where mains is converted to ELV of course, and the real amount of insulation to hold off mains may be little more than a thin -but well specified to be CE marked- coat of varnish by the time you get deeper into the guts of the power supply. (so called triple insulated magnet wire allows you to ditch the inter-winding papers )

    If you open an ethenet mag jack, designed to hold off 500V or more, the wires are a lot closer together but 'globbed' to prevent arc strike.

    I suggest the actual problems in the field that come with network cables in power boxes and vice versa are usually related to EMC, not voltage breakdown of the insulation of the cables.  To jump an air gap of 50mm needs in excess of 30kV !!  Once an arc is struck of course a lot less. This is why we are happy to work with bare hands on open systems isolated from the mains by a 3mm or greater air gap at the mains switch or MCB.

    The exception needing real separation is probably the few remaining bare copper phone lines sharing poles or crossing under bare mains overheads, and there I think we want a lot more than 50mm!

    I'm happy to say the 50mm rule is a bit arbitrary.

  • BS 6701 doesn't refer to manufacturer's instructions in this particular respect.

    Also, are you claiming that if a manufacturer recommends something which is not compliant with BS 7671, that is OK?

    For example, I manufacture a 13 A socket-outlet, and put in the installation instructions "This device must not be protected by a residual current device" ... Does that mean you install it, and not supply from an RCD - or do you (as I would hope) decide that this socket-outlet ought not to be selected for an installation to BS 7671 due to Regulation 411.3.3?

  • Hi, 

    I’m not suggesting that at all, however there are tones of products out there that combine data and power into hybrid cables, with many of these featuring connectors that have specialist pin-outs that cater for power and data. An example is here:

    These are fitted by lighting manufactures to products which combine power and DMX into the same cable. This then lands into a junction box, where you then terminate your power and data. They also provide pre-made jumper cables to link products together. 

    The cables and the luminaires and equally and jointly as a finished product, CE certified and sold for fixed installations. 

    An installer would follow manufactures instructions to install this equipment. 

  • I added a quick discussion last week about the cost of running separate power and data cables to EVSE versus using a combined cable with an electrician working for a specialist EVSE installation company, cost wise separate cables win hands down if you are installing EVSE on a daily basis.

  • I understand that, but BS 6701 clearly states this safety requirement, so you couldn't necessarily claim that installation was compliant.

    DMX is a separate network ... but Ethernet has another issue, as it often connects to equipment that's connected directly to a copper-wired telecomms network.

    Until BS 6701 changes, it's difficult to see Ethernet or RS 485 connecting via EV Ultra cable being compliant.

  • I understand what you’re saying but there are lots of examples where the regulations seem to contradict themselves.

    These hybrid cables are also available for temporary installs, where the DMX would end up in the back of a lighting console. 

    A theatrical lantern can also have hybrid cables which land into special junction boxes for star type distribution. The DMX end if you like, would end up being a hand-held component in this instance for plugging into a console. 

    If you take a standard multi-core power cable, which could host a combination of ELV (band 1) and LV (band 2), BS7671 states this is OK as long as the insulation rating of the cores are to the highest voltage present.

    You can also run an SELV circuit (ELV) inside of a multicore with low voltage (LV). If this 50mm rule is for safety, why doesn’t BS7671 make reference to this (or some sort of spacing) for SELV circuits? Ethernet is also defined as SELV in BS7671 (telecommunications).

    If you have a 24V SELV garden light, powered from a multicore cable that’s also carrying mains voltage, the risks associated with it being fixed or handheld are almost the same in my opinion. You could step onto a ground light with bare feet, which in my view would be more risky than holding onto a Cat5 cable with a plastic connector on the end. 

  • If you have a 24V SELV garden light, powered from a multicore cable that’s also carrying mains voltage, the risks associated with it being fixed or handheld are almost the same in my opinion. You could step onto a ground light with bare feet, which in my view would be more risky than holding onto a Cat5 cable with a plastic connector on the end. 

    I agree - but we are not simply talking about BS 7671, this is an other standard.

    Strangely, BS 6701 contradicts itself, because it says that 50 mm separation is required for cables indoors, but underground cables refers to BS EN 50174-3, which appears not to require separation of 50 mm if all cables are insulated for the highest voltage present.

    The problem is, how do you bring the single cable out of the ground to connect it to separate cables that are kept 50 mm apart?

    I'm happy to say the 50mm rule is a bit arbitrary.

    And therefore, I agree.

    Are you saying that EV Ultra cable is therefore not allowed? The product itself is CE certified and it’s being installed everywhere. 

    Well, perhaps not allowed for cabling installations to BS EN 50174-2 (indoor and within structures cabling), especially where it comes out of the ground?

  • Given the confusion brought about in BS 6701 itself, I am wondering whether there is an error in the wording of the relevant Clause (5.4.4.2) of the standard?