The IET is carrying out some important updates between 17-30 April and all of our websites will be view only. For more information, read this Announcement

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

SELV Fusing

Hi All,

I recently had a discussion with regard to fusing SELV lines, traditionally I always fused the +ve line feeding equipment.

A colleague recently mentioned that he had read an article from Sage? Arguing that both lines should be fused, from memory this ties up with 60204:2019.

A bit of thought suggests this would be wise as obviously with SELV there is no ref to earth for ADS. 

I cannot find the article, can anyone point me in the direction of some relevant info.   

Thanks

Martyn

  • I'm not sure you are looking at the right problem Marty. The only problem you can get with SELV is surely overload, nothing to do with Earth faults. A single fuse will protect against excess current, if you want anything else you must consider some mechanism of Earth fault detection. SELV is completely isolated from Earth, 414.4.1. Do you mean PELV?

  • I think I can argue that fusing of only one pole of 2-wire SELV is required.

    Note before I start the argument: This is NOT necessarily the case for multi-pole SELV (where at least two poles are required to be protected against overcurrent) or three-phase SELV, where in delta, the lines should be protected against overcurrent, and in star, the lines other than star point lines should have OCPDs. Nor is it the case for FELV or LV where electrical separation is used as there may be risk of electric shock.

    First, does SELV require fusing for ADS ? No ... SELV requires protective separation from Earth, as well as limitation of voltage (and hence touch-voltage after one or more faults). Even after first fault to Earth, provided an appropriate voltage is applied for the conditions (e.g. in BS 7671 dry condition 50 V AC/120 V DC, wet condition 12 V AC/30 V DC, although note these voltages vary between standards), there should be no shock risk.

    Which leaves us with overcurrent protection (faults and if necessary overload current). Assuming no other faults, overload current is provided by a single OCPD (in 2-wire SELV). There are two scenarios which chiefly address fault current:

    • Transformer-type source, where, either after secondary line-to-line fault, or one line to Earth then second line to Earth, a single OCPD in the primary (and/or a OCPD in the secondary) could provide overcurrent protection
    • Electronic (switching-type) supply, either after secondary line-to-line fault, or one line to Earth then second line to Earth, the current-limit in the supply device, or a single OCPD in the secondary of the supply device, will provide protection.

    It's a little more tricky to see the fault scenarios if you are distributing a single source for a number of circuits, and require especially if choosing one SELV line to be common, and providing OCPDs in the other, but I'm sure if you draw out the two fault scenarios (e.g. common first, then line in a final circuit, or line in a final circuit then common), fusing just the non-common lines should still work.

    In PELV, where necessary (i.e. protection against overcurrents not provided in primary, or converter, or the supply feeds a number of final circuits) it's normal to fuse the lines which are not neutral or mid-point conductors, and the neutral/mid-points ought not to be fused.

  • Hi David,

    I was not really thinking of any particular fault type, more the principal of fusing both legs of the ELV.

    Just found the source of the original debate…

    “One problem with a PELV system is that faults elsewhere in the installation may induce voltages on the entire system via its protective earth conductor.

    The advantage of an SELV circuit is that there is no earth return path which might occur via the human body. A disadvantage is that an over-current protection device must be installed in each leg of each ELV circuit and sub-circuit.”

    source David W Otterson

    journals.sagepub.com/.../0020294017701880

  • One problem with a PELV system is that faults elsewhere in the installation may induce voltages on the entire system via its protective earth conductor.

    Not sure how an OCPD will protect against, that, but the advantage of PELV for a control system, is that faults operate protective devices, and, if the controls are arranged so that "ground" or "off" is a default safe state, the fault is both detected, and leaves the system in a safe state.

    Another advantage of PELV is that static charges are eliminated, and , provided the current-carrying conductors in the system are separated from PE (a TN-S arrangement), PELV can also be an effective way of dealing with certain electromagnetic compatibility and overvoltage phenomena.

    Strictly, the voltages impressed on a PELV system are controlled in the same way as main protective bonding, and that is why standards such as BS EN 60204-1 have such strict requirements for the protective bonding of the machinery covered by the standard.

    However, SELV also has its advantages under other conditions, and equally  it may be advantageous to separate some signals from PE.

    The advantage of an SELV circuit is that there is no earth return path which might occur via the human body. A disadvantage is that an over-current protection device must be installed in each leg of each ELV circuit and sub-circuit.

    PELV has the same voltage source considerations as SELV, so I'm not quire sure about the "earth return path" point there. Yes, if there is a voltage rise on PE resulting from a fault in another circuit, that voltage will be impressed on the PELV system, as on the whole PE system, but the relevant systems or equipment should be designed so that's not an issue?

    As per my earlier reply, I don't think it's necessary to provide an OCPD in every pole on an SELV circuit according to BS 7671, certainly not in a 2-wire SELV circuit, for the reasons provided, so unless the previous discussion expanded on why it was thought this was necessary. I'm not saying other standards don't have the requirement.

  • I detect a "what if"  kind of question, whilst at the same time assuming two or more faults. I agree with Graham, should you have a multi-phase SELV system or PELV then more fusing may be required, but you specifically said SELV. I do not see that this is "wise" as you say, but then if you want you can fuse as much as you like on these systems. Of course you cannot fuse the neutral in any system depending on disconnection of Earth faults where the neutral is Earthed even if you fit RCDs, but I am sure you know all that. I wonder who wrote the article, I have no direct access to me?

  • Arguing that both lines should be fused

    I've a recollection of something similar - I think in some (German?) instructions for 12V caravans electrics. From memory I think the position was that if you were using the chassis for one pole (typically -ve) then single pole fusing in the other pole was fine, but if you were using an insulated 2-wire system then they recommended fusing both poles. I didn't get to the bottom of it at the time, but my suspicion was that the thinking was similar to the old wiring regs that required 2-pole fusing - i.e. with no chassis/earth to catch faults the only faults likely were between insulated wires - and if those wires happened to be from different rated circuits (say the +ve wire from a 50A circuit and a -ve wire from a 5A circuit) then the -ve wire of the lower rated circuit wouldn't be adequately protected. With a chassis return system the problem didn't exist and the -ve conductors would always have a massive c.s.a. so +ve pole only fusing is fine.

    I'll see if I can dig out the original instructions...

       - Andy.

  • As soon as one side is chassis, or ground, it is not SELV, but PELV.  Generally vehicle and trailer electrics do use a chassis return, and there is very clear sense of the other one being the side you fuse or switch.

    Also in some systems it is safer to not fuse and to allow it to run to failure - both high and low power systems.

    As an example of low power, most ELV bathroom fans are not fused for example on the ELV side, rather there is a one time thermal cut-out in the transformer primary.

    As a common  example of unfused higher power consider the car or lorry starter motor - there is no sensible fuse size that will give much  better protection than the motor windings melting, and that is safely contained,  so that may as well be the limiting factor.

    Mike.

  • Yes indeed, correct circuit cable protection is essential.

    Superyacht worth £6m with 8,000 litres of fuel onboard goes up in flames in Devon harbour | Daily Mail Online

    Z.

  • Generally vehicle and trailer electrics do use a chassis return

    In (touring) caravans it seems that they usually have two distinct systems - one for the 'road' - mostly the road lights (side lights/indicators/brake etc) but also the 12V feed to the fridge for use when moving, which is often chassis return (in common with the tow vehicle), and then there are the 'leisure' circuits fed from a 12V leisure battery (or mains fed PSU/Zig unit) which supplies the caravan's internal lighting, pumps and so on - which are invariably have an all insulated 2-wire system to keep it deliberately separate. (Although there are times where the internal leisure system is connected to the tow vehicle's 12V system - e.g. to recharge the leisure battery when on the road - just to make conventional definitions tricky to apply!)

       - Andy.

  • https://www.dailymail.co.uk/news/article-10866651/Torquay-6M-superyacht-owners-flashing-lit-electric-nameplate-sparked-inferno.html

    Z.