Author Topic: Creepage between live and neutral in offline SMPS?  (Read 2176 times)

0 Members and 1 Guest are viewing this topic.

Online FaringdonTopic starter

  • Super Contributor
  • ***
  • Posts: 1947
  • Country: gb
Creepage between live and neutral in offline SMPS?
« on: June 26, 2021, 01:06:25 pm »
Hi
Supposing the DS5022P-334 (330uH) SMD inductor is used in the live line of the AC filter of an offline 240vac SMPS (ie just downstream of the mains input connector) . Then suppose that the neutral trace is tracked right under the ferrite body of this inductor, with just the usual  solder resist covering the neutral trace . Would you agree this is a bad move? (in terms of creepage distance between live and neutral)

DS5022P-334
https://www.coilcraft.com/getmedia/5408414a-17b6-4981-89ae-94de22e4c8d9/ds5022p.pdf
'Perfection' is the enemy of 'perfectly satisfactory'
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #1 on: June 27, 2021, 08:35:40 pm »
Oh FFS!
Treez, just read the bloody standards already!
 
That part (Far more importantly then the standards themselves in a practical sense) seems to lack any sort of NRTL approval, so by using it you are at least into major cost when your customer starts asking about UL/CSA/VDE/SEMCO/PSE/<whatever> approvals. Far better to make sure all your mains side components have file numbers for at least one of the big test houses, makes for a far quieter and less annoying life.

It almost certainly has a thermoset plastic base, but solder resist does not count as insulation and there is no specification for that use, so I would not even consider it.

Seriously, you been doing SMPSU work for HOW LONG?
 
 
The following users thanked this post: Faringdon

Online FaringdonTopic starter

  • Super Contributor
  • ***
  • Posts: 1947
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #2 on: June 28, 2021, 05:11:17 am »
Thanks, this is interesting....do you know what is the actual issue with this inductor, ie, why it isnt good for UL approval?
I take it its the winding insulation?
Having said that, ive seen plenty of SMPS that use filter inductors upstream of the bridge that dont have any reference to passing NRTL etc  in their datasheet.
I did one offline 150W LED driver, of an unusual design, and it had to use only SMD everywhere....and i used SMD filter inductors, admittedly downstream of the bridge, and it passed 10kV transient test, thermals, EMC, everything. Those inductors made no ref to NRTL in their datasheet.

Ive always wondered just what it is about being "downstream of the diode bridge", that makes such a big difference to a components rating in an offline SMPS...(other than its DC there.).........what is this great protective cloak that the diode bridge throws over for us?

I admit the diode bridge could   fuse and blow open and so provide safety.......but so could other components upstream of the diode bridge. If anything, i'm sure you agree, mains diode bridges are amazingly tough, and would be one of the last things to fuse open circuit.
« Last Edit: June 28, 2021, 05:13:02 am by Faringdon »
'Perfection' is the enemy of 'perfectly satisfactory'
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #3 on: June 28, 2021, 10:55:41 am »
The thing about having a NRTL approval is not that it makes the part different in any way from one that has not got such, but that it means the NRTL can simply use the fact that they have examined the part before to avoid the need to do a while pile of tests again (Flammability, spread of flame, insulation, lead content, mercury content, cadmium content, whatever the fuck else).
 
A part with a UL file number will be presumptively assumed to be compliant, where a random inductor will not have that assumption applied.

This means that certifying your product is MUCH simpler and thus much cheaper, you do NOT want to have an NRTL come back with questions about say the flammability of the plastic used in a fuse holder, that shit is a nightmare. 

You can get something using parts not known to the test lab thru the NRTL, but the process will be longer and MUCH more expensive.
 
The following users thanked this post: Faringdon

Online FaringdonTopic starter

  • Super Contributor
  • ***
  • Posts: 1947
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #4 on: June 28, 2021, 07:46:19 pm »
Thanks, if my internet connection was better, id do a search for NRTL registered inductors. Will manage soon.
'Perfection' is the enemy of 'perfectly satisfactory'
 

Offline Fiona

  • Newbie
  • Posts: 1
  • Country: it
Re: Creepage between live and neutral in offline SMPS?
« Reply #5 on: June 29, 2021, 02:27:15 am »
 :) ;) ;D
 
The following users thanked this post: Faringdon

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3287
  • Country: fr
Re: Creepage between live and neutral in offline SMPS?
« Reply #6 on: June 29, 2021, 04:54:31 pm »
Hello: This is a safety issue and mistakes can resut in fire, shock or death.

Suggest to take compliance issues seriously.

Simple compliance is to use an approved IEC or other line filter/connector with CM and DM inductors built in and tested.

But first determine the applicable standards of what you are doing.

Just a thoght


Jon
Jean-Paul  the Internet Dinosaur
 
The following users thanked this post: Faringdon

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9393
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #7 on: June 29, 2021, 07:14:39 pm »
A very good thought Jon.

Unfortunately one that has been recommended to the OP many times over the years.
« Last Edit: June 29, 2021, 07:16:51 pm by Gyro »
Best Regards, Chris

"Victor Meldrew, the Crimson Avenger!"
 
The following users thanked this post: Faringdon

Online FaringdonTopic starter

  • Super Contributor
  • ***
  • Posts: 1947
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #8 on: June 29, 2021, 07:30:05 pm »
Thanks, for these replies. Im very grateful for your thoughts, and you will no doubt  completely agree that , as you know, a huge number of offline SMPS's in EU and UK & USA  use  diff mode filter inductors which are not UL approved. And huge numbers dont use approved filter modules.
I'm sure you would agree that these are not dangerous because of this.

 The issue here is whether there would be a creepage issue in the routing of the (solder resist covered) neutral trace under the body of the SMD inductor. The SMD inductor being in the live line.
'Perfection' is the enemy of 'perfectly satisfactory'
 

Offline penfold

  • Frequent Contributor
  • **
  • Posts: 675
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #9 on: June 29, 2021, 08:34:28 pm »
The issue here is whether there would be a creepage issue in the routing of the (solder resist covered) neutral trace under the body of the SMD inductor. The SMD inductor being in the live line.

Have you read the standard to which you are aiming to comply? Which standard is it? How thick is the neutral trace? Is it on the same side of the board as the inductor? What colour is the board? Is the "240vac SMPS" going to be powered or is it just there to raise the BOM count? What's the over-voltage category? Pollution degree? Altitude?
 
The following users thanked this post: Faringdon

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9393
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #10 on: June 29, 2021, 08:52:38 pm »
The issue here is whether there would be a creepage issue in the routing of the (solder resist covered) neutral trace under the body of the SMD inductor. The SMD inductor being in the live line.

Ok then, to be blunt, don't do it. Solder resist cannot be relied upon to provide reliable insulation at mains voltage. In addition to dielectric breakdown you have potential issues with it having pinholes, being scratched during processing or assembly, and being impossible to inspect once an smd component has been placed on top of it.

The inductor does look as if it has a thermoplastic base beneath the ferrite (thickness unspecified!), but this is very thin and, unless the manufacturer has used a grossly oversized bobbin, the winding will be far too close to the edge to provide sufficient creepage distance.

Without specific written confirmation from Coilcraft that the inductor would meet the required creepage to a conductive surface that meets the standard you are working to (which you apparently don't possess) then only a complete cowboy would consider doing this. There is nothing in the linked Coilcraft data to support what you propose. Best case, zero polution degree etc, you have to be looking at a bare minimum of 3mm for 240V mains, and I can't see how that is achievable.
« Last Edit: June 29, 2021, 09:02:21 pm by Gyro »
Best Regards, Chris

"Victor Meldrew, the Crimson Avenger!"
 
The following users thanked this post: Faringdon

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3287
  • Country: fr
Re: Creepage between live and neutral in offline SMPS?
« Reply #11 on: June 30, 2021, 05:39:50 am »
Rebonjour cher Farringdon:
The  schematic, BOM and PCB layout will get you better advise.

1/ PCB silkscreen ink is never to be used as an insulator, it is easily damaged and of unspecified material and thickness.
Act as if it did not exist and any top layer conductor was bare.

Just  place a piece of 2.2 mil UL rated electrical tape over the trace and under the part, e.g.  3M 56 Mylar tape.

2/components  of  mains filter ARE safety rated, such as  VDE X and Y capacitors, and  CM toroidal chokes.

3/Very often a line filter has TH and not SMD parts. One reason is most SMD parts are NOT designed for mains use.

4/ The 6mm creep and strike standards are from protection against HV transients, ~ 6 KV.
https://www.tempoautomation.com/blog/understanding-pcb-creepage-and-clearance-standards/

5/ Nominal mains ~ 100-260V  Transients line...ground are several kV.

Your question involves a tradeoff,  between safety and cost/convenience/size, etc.
Just decide what your priorities are.

Bon Chance,

Jon

(designed SMPS and electronic ballasts since 1970s)


Jean-Paul  the Internet Dinosaur
 
The following users thanked this post: Faringdon

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9393
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #12 on: June 30, 2021, 08:13:58 am »
...
1/ PCB silkscreen ink is never to be used as an insulator, it is easily damaged and of unspecified material and thickness.
Act as if it did not exist and any top layer conductor was bare.

Just  place a piece of 2.2 mil UL rated electrical tape over the trace and under the part, e.g.  3M 56 Mylar tape.

I have a couple of concern with this one:

- I don't think Mylar tape would survive a full-on smd Lead-free reflow process [Edit: Melting point 254'C]. You should maybe consider the 130'C max operating as a temperature that is not to be exceeded under any circumstances (including reflow) if its insulating and physical properties are to be maintained.

- Tape application is a tedious manual operation which could easily be 'forgotten' by the manufacturer over time, returning the situation to solder resist alone. Application accuracy would need to be carefully controlled too.

Quote
2/components  of  mains filter ARE safety rated, such as  VDE X and Y capacitors, and  CM toroidal chokes.

Agreed. Just because the OP doesn't see them covered in logos (impossible on an irregular shaped varnished CM choke) doesn't mean that they do not have agency approval. It is necessary to actively seek this information from the manufacturer. It is impossible to achieve certification of the SMPS (or the overall product) without obtaining this data anyway.  It would be pretty silly for a company to manufacture CM chokes that didn't meet agency approval, they wouldn't be able to sell them.
« Last Edit: June 30, 2021, 08:45:37 am by Gyro »
Best Regards, Chris

"Victor Meldrew, the Crimson Avenger!"
 
The following users thanked this post: Faringdon

Online FaringdonTopic starter

  • Super Contributor
  • ***
  • Posts: 1947
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #13 on: July 01, 2021, 05:04:58 am »
Thanks, i appreciate  you are entirely correct about the standards situation, but its interesing about a series diff mode filter inductor in series with the live...because if it fuses, we're then safe, and if it goes short...well..its just like its not there any more....again...safe.
If it catches fire..then thats bad...but there's nothing in that inductor to catch fire.
« Last Edit: July 01, 2021, 07:37:54 pm by Faringdon »
'Perfection' is the enemy of 'perfectly satisfactory'
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9393
  • Country: gb
Re: Creepage between live and neutral in offline SMPS?
« Reply #14 on: July 01, 2021, 10:49:03 am »
Thanks, i appreciate  you are entirely correct about the standards situation, but its interesing about a series diff mode filter inductor in series with the live...because if it fuses, we're then safe, and if it goes short...well..its just like its not there any more....again...safe.
If it catches fire..then thats bad...but there nothing in that inductor to catch fire.

Well that's one way to look at it. Not my way, or maybe even that of a self proclaimed champion of re-introducing quality SMPS design to the UK, but it is certainly a viewpoint.


Edit: Sorry, maybe you were talking about using agency approved CM choke rather than your actual question on the live inductor on top of a neutral trace. I'm not sure. You wouldn't expect an approved CM choke to break down in a hazardous manner anyway.
« Last Edit: July 01, 2021, 06:11:02 pm by Gyro »
Best Regards, Chris

"Victor Meldrew, the Crimson Avenger!"
 
The following users thanked this post: Faringdon

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3287
  • Country: fr
Re: Creepage between live and neutral in offline SMPS?
« Reply #15 on: July 03, 2021, 04:58:47 am »
Did not think of SMD oven temp, use Kapton or fiberglass insulationg tape, not Mylar/3M56

Jon
Jean-Paul  the Internet Dinosaur
 
The following users thanked this post: Faringdon

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 8951
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Creepage between live and neutral in offline SMPS?
« Reply #16 on: July 03, 2021, 07:54:30 pm »
Assuming you can't just reroute the neutral trace to not be under the inductor, what about using a through hole jumper or zero ohm resistor on the other side of the board? Might be cheaper than adding a strip of tape under the inductor since it would not require custom assembly procedures.

Or move the inductor to be in series with the neutral and now you have a good place to put a through hole fuse for the live.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 
The following users thanked this post: Faringdon


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf