Do I need to fuse power supply output?

Merry Christmasto you also!

You should post your question in the IEEE EMC-PSES group [don't have to be a member of IEEE] Their expertise includes those pesky RFI/EMC compliance issues, too.

Yes, fuse both ends..

Jamie

thank you. I will post over there.

On Tuesday, December 23, 2014 11:56:14 AM UTC-5, Maynard A. Philbrook Jr. w rote:

plies. These are near 100 watt each at about 4amps out. I was told we wil l design to NFPA 79.

or fusing the output of the supply. NEC wants a fuse or breaker on lines t o protect. The vendors I have asked so far have been absolutely useless. I would think an application engineer would know right off the top of his h ead what particular agency ruling would govern this aspect of applying thei r product. SELV, UL508, IEC60950-1, etc....

electronic motor overload function in software and got it by UL508. I am l ooking for a similar function in a small power supply.

look to for guidance.

CHRISTMAS to all of you great engineers here on S.E.D.

Thanks for reply. Do you have the standard that calls out fuse both ends?

If it's a current limiting power supply, I don't think you have to (and it wouldn't matter, because it won't clear the fuse anyway).

If you have lower amp secondary loads, it might be worthwhile?

Tim

-- Seven Transistor Labs Electrical Engineering Consultation Website:

There doesn't seem to be a lot of information concerning requirements for fusing the output of the supply. NEC wants a fuse or breaker on lines to protect. The vendors I have asked so far have been absolutely useless. I would think an application engineer would know right off the top of his head what particular agency ruling would govern this aspect of applying their product. SELV, UL508, IEC60950-1, etc....

I worked for a large motor drive company in the 80s and we designed an electronic motor overload function in software and got it by UL508. I am looking for a similar function in a small power supply.

To end this tome/rant I am asking if you may know which agency(s) I may look to for guidance.

thanks for all the help I have received here over the years and MERRY CHRISTMAS to all of you great engineers here on S.E.D.

Bob N9NEO

What is the short-circuit current of the power supply?

Merry Christmas to you as well.

It all depends on how much you like lawsuits. Passing an agency is only part of the game. Even if the supply sits inside a chassis and never sees the outside world, there is always that dork with a DVM or scope probe shorting things out.

Not exactly appropos, but once I went to use an Audio Precision distortion analyzer using the digital ports, only to find them dead. Somebody fried the ports and just put the unit back. I was very thankful the designer used a socket for the bus driver chips.

You should really engineer for bad shit to happen because it will.

If you have a component failure on the boards fed by the power supply, there should be a fuse to prevent elevated temperatures. You can also use a PTC if you can tolerate a device like that. I'm not familar with NFPA 79, but 100W will get some attention by the reviewer. If you can prove the supply will turn off on overload or fold back with out generating high temperatures, then it may be good enough.

Cheers

supplies. These are near 100 watt each at about 4amps out. I was told we will design to NFPA 79.

s for fusing the output of the supply. NEC wants a fuse or breaker on line s to protect. The vendors I have asked so far have been absolutely useless . I would think an application engineer would know right off the top of hi s head what particular agency ruling would govern this aspect of applying t heir product. SELV, UL508, IEC60950-1, etc....

an electronic motor overload function in software and got it by UL508. I a m looking for a similar function in a small power supply.

may look to for guidance.

RY CHRISTMAS to all of you great engineers here on S.E.D.

s?

With reference to this thread, what safety schemes are put into potted power bricks. I have used a few, and all I have seen are 5(2 for AC input, 3 for +/- DC and DC ground) sticking out of the shiny black block. Where are the fuses ? For that matter common PC SMPS's do not have any input/output fuses, although a decent PC SMPS is supposed to deliver 450 -

Small wall wart transformers often have such high primary and secondary coil resistance and such magnetic core characteristics that the transformer does not overheat even with indefinite short circuit conditions.

Others may have some operate once fuse embedded into the transform and hence sense the heat from the transformers.

If a SMPS has output short circuit current limiting or even fold back so that the PS does not overheat even during short circuit, so why would it need output fuses ?

The lack of input fuses is a bit strange, if the SMPS switcher transistor fails short circuit. Depending on topology,this might be a problem or not.

Anyway the mains cord to the SMPS must be quite thick, so that the loop resistance is sufficiently low to blow the in house fuse if something went wrong in the power cord or device. Thus a less than 100 W device might have a power cord capable to carry 2 or 3 kW (at least in the 230 V world) to ensure that the domestic mains fuse operate quickly.