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

Standards for a PSU to be considered as "SELV"

Hi - quick standards question, for hobby purposes only:


Is a declaration of being "Class II" and within the appropriate voltage band, sufficient for a PSU to be considered as SELV for usage scenarios in BS7671?


I'm doing a Raspberry Pi project with my son to PWM drive some 30V LED outdoor fairly lights. So I'm choosing a PSU that's nice and safe.


The candidate PSU I have is this one: 30V PSU (RS) or possibly this one 30V PSU (-10C) for it's lower operating temperature (it's going to be in the shed, occasional drops below 0C possible)


First one is defined as Class II, it's 30V. It also claims medical approval. A whole bunch of quoted standards on the datasheet above. Apart from "Class II", not sure if I should be looking at any particular standard that defines it as meeting SELV PSU requirements. Isolation voltage is given as 4kV?


Second one claims very little but implies Class II by lack of protective conductor pin.


Pretty sure either is a good choice (noting operating temperature limits of the first), but academically I have been curious what defines a SELV PSU if it doesn't actually have the word SELV printed on it?


Many thanks as always for letting me tap the font of knowledge that this forum is :)


Tim


  • Tim


    BS 7671 recognises various sources for SELV. It has to be "separated" and Extra Low Voltage i.e 50V or less AC.


    Safety isolation transformers to BS EN 61558-2-6 and BS EN 61558-2-8 are in my view the safest as the electronic type SELV sources being electronic will probably fail short circuit. Below is the actual wording from BS 7671.


    The following sources may be used for SELV and PELV systems:

    (i) A safety isolating transformer in accordance with BS EN 61558-2-6 or BS EN 61558-2-8

    (ii) A source of current providing a degree of safety equivalent to that of the safety isolating transformer specified

    in (i) (e.g. motor-generator with windings providing equivalent isolation)

    (iii) An electrochemical source (e.g. a battery) or another source independent of a higher voltage circuit (e.g. a

    diesel-driven generator)

    (iv) Certain electronic devices complying with appropriate standards, where provisions have been taken such

    that, even in the case of an internal fault, the voltage at the outgoing terminals cannot exceed the values

    specified in Regulation 414.1.1. Higher voltages at the outgoing terminals are, however, permitted where, in

    case of contact with a live part or in the event of a fault between a live part and an exposed-conductive-part,

    the voltage at the output terminals is immediately reduced to the value specified in Regulation 414.1.1 or

    less.

    NOTE 1: Examples of such devices include insulation testing equipment and monitoring devices.

    NOTE 2: Where higher voltages exist at the outgoing terminals, compliance with this regulation may be assumed if the

    voltage at the outgoing terminals is within the limits specified in Regulation 414.1.1 when measured with a

    voltmeter having an internal resistance of at least 3 000 ohms.
  • My understanding is that the PSU must be so designed that the ELV output can not become connected to the mains input under fault or failure conditions.

    In practice this means double or reinforced insulation between mains and ELV. For example the mains and ELV transformer windings must be on separated insulated formers, such that BOTH would have to fail to become dangerous.


    AFAIK there is also a requirement that it must EITHER be protected against over current and overvoltage, OR that it must fail safely.

    For example output overload blows a fuse=fine.

    Output overload destroys the PSU without exposing the user to mains=fine

    Output overload melts plastic internal insulation and allows the ELV output to connect to the mains=NOT FINE.


    If the PSU is of reputable manufacture and is marked SELV and is CE marked, then that in my view is acceptable, you are not required to do your own tests regarding the veracity of the markings.

    Just marked "ELV" is not in my view sufficient, for say lightly insulated equipment that should be connected to a SAFETY ELV source.
  • Hi John & Broadgage,


    Thank you both for your kind replies. Everything seems to have gone down the switched mode route these days - for green/efficiency reasons I guess...


    However, it turns out that searching for SELV 30V is a terrible search term. Better to search for LED driver on RS and work down to the right parameters.

    30V SELV Datasheet RS Product page


    30V DC, IP67, SELV, Class II, block diagram on datasheet shows full magnetic + opto isolation between input and output, over temperature/current/voltage/short circuit protections.

    Operating temperatures suitable for Siberia!


    £20


    Whilst LED drivers are not necessarily pure clean DC supplies this one is not dimmable and the ripple and output specs suggest it is proper DC out.


    The rest as always is down to trust in the manufacturer to have designed as claimed, but sourcing from RS is usually an assurance.


    Thank you for making me doubt my first choice - this one looks a lot better for the application :)


    Kind regards,


    Tim

  • Just to add that the standards that BS 7671 specifically lists are primarily for conventional wound transformers - BS EN 61558-2-6 (safety isolating transformers and power supply units incorporating safety isolating transformers) and BS EN 61558-2-8 (transformers and power supply units for bells and chimes) - but there are other product standards which might give equivalent protecton - the first device you referred to mentioned EN60950-1 which will be known in the UK as BS EN 60950-1 (Information technology equipment. Safety. General requirements) (although that standard has now apparently been superceded by BS EN 62368-1 (Audio/video, information and communication technology equipment. Safety requirements)). You'd have to read the standard proper to confirm the details (which aren't freely available on the web) - but perhaps your local library could help with that.


    I wouldn't be too worried about the 'electronic' designs per se. The traditional switch-mode PSU normally had a transformer between the "mains" and ELV sides (if a much higher frequency, smaller and more efficienet one that would traditionally been the case) - which together with opto-coupling in the feedback loop which can provide isolation from the mains every bit as good as a traditional safety isolating transformer. Not all designs will provide safety isolation of course, but it is possible.


      - Andy.
  • This issue is set to become more "nebulous".


    BS EN 60950-1 is sometimes used for switch-mode power supplies. It is being replaced by BS EN 62368-1 (which is also replacing BS EN 60065), and the new standard will be in full force in 2020 I understand.


    Now, BS EN 62368-1 does NOT recognise SELV and PELV, and it is not at all clear whether, under all fault conditions conditions, the output of a supply to BS EN 62368-1 might be guaranteed to be "just as safe" as a safety isolating transformer.


    Now, of course a manufacturer that made an SELV (or PELV) supply under BS EN 60950-1 could continue to do so, probably without design change, and certify it to BS EN 62368-1, BUT the important point for those specifying sources for SELV and PELV is that, just by virtue of the product complying with BS EN 62368-1 does not mean you're getting an SELV (or PELV) supply that we used to assume we were getting.


    Would a product only conforming to BS EN 62368-1 (but not BS EN 60950-1) comply with BS 7671 as a source for SELV/PELV?


    Well, possibly it could be argued to be a source according to 414.3 (iv), and is likely to be fine in dry conditions. But I'm not, at present, 100 % sure  (without further investigation) that, under all fault conditions in situations it would protect a person we are trying to protect against shock would be substantially wet / submerged, and unclothed, in the same manner as an SELV/PELV source to 414.3 (i), (ii) or (iii).