Author Topic: Insulating washers & creepage/clearance issue at holes  (Read 1524 times)

0 Members and 1 Guest are viewing this topic.

Offline eevcandiesTopic starter

  • Regular Contributor
  • *
  • Posts: 119
  • Country: us
Insulating washers & creepage/clearance issue at holes
« on: November 04, 2019, 07:59:29 am »
Have an electrically "hot"  semiconductor, let's say 400V or 600V  TO-220 mounted to a chassis/heatsink.
Using the typical shoulder washer & thermal insulator pad with good cut-through prevention.

Or this might be a small "hot" high voltage metal plate mounted using four corner shoulder washers & insulating pad material to a heatsink

However, at the hole itself, the metal heatsink is only thousandths of an inch away from the "hot" metal (via the wall of the hole)...separated only by the thermal insulator pad (which is not insulating at the open hole).  Adding the shoulder washer does not seem to change things, since an arc could travel from the hot metal's hole rim, alongside the shoulder washer, down a few thousandths to the heatsink's metal hole rim.  Seems like the shoulder washer ONLY prevents the screw from causing an arc/short.  A few thousandths gap is certainly not enough distance for 600 volts or hipot tests.

Is this mounting issue discussed anywhere?--if so, I can't seem to find any mention.  I usually see "simply add a pad & shoulder washer to insulate", in shoulder wash catalogs, app notes, etc..but that seems to ignore this path.  Should we ignore?  Do any of the standards address this potential arc path?

 
 

Offline Someone

  • Super Contributor
  • ***
  • Posts: 5155
  • Country: au
    • send complaints here
Re: Insulating washers & creepage/clearance issue at holes
« Reply #1 on: November 04, 2019, 08:57:45 am »
To create more isolation there are many options, spring clamping with a flat pad or sleeve:
http://www.tglobaltechnology.com/wp-content/uploads/2018/03/PTM-THIN-Series.pdf
shaped clamps for use with flat thermal pads:
https://www.essentracomponents.com/en-nl/p/transistor-insulators
Many others...
 

Offline eevcandiesTopic starter

  • Regular Contributor
  • *
  • Posts: 119
  • Country: us
Re: Insulating washers & creepage/clearance issue at holes
« Reply #2 on: November 04, 2019, 09:50:17 am »
Thanks for the tips!!

Have you ever seen this specific condition discussed or actually mentioned anywhere (as well as in any standards)?
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Insulating washers & creepage/clearance issue at holes
« Reply #3 on: November 04, 2019, 10:39:28 am »
You can get shoulder washers with a longer shank to stick down into the mounting hole; that doesn't account for the gap between the outside of the shank, the thermal pad and the hole edge, but it helps the screw anyway.

I'm not sure offhand if regulations would accept a tight-fitting thermal pad as sufficient.  It could possibly work as electrical insulation.  I would guess it's not very reliable.

The preferred solution is a solid pad and a spring clamp, which also improves clamping force -- normally the off-center screw hole causes the device to tilt up, increasing thermal resistance between the body (where heat is actually generated) and the heatsink.  You can use I(2)PAKs and MAX247s in the same way.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline eevcandiesTopic starter

  • Regular Contributor
  • *
  • Posts: 119
  • Country: us
Re: Insulating washers & creepage/clearance issue at holes
« Reply #4 on: November 04, 2019, 03:35:51 pm »
Thanks for the tip.

However, what I'm saying is for years and years you see thermal pads or mica washers & a shoulder washer touted as "the solution"....yet I've never seen this potential issue even mentioned anywhere...so is it an issue?
If it is not, why not?  If it is an issue is there a link anywhere where it has been discussed?  Again, the issue is the distance between the metal surfaces,at the hole, is only the thickness of the sil-pad, or mica washer.  A pic of a mounting:
http://www.diyforums.org/MOSFET-MAX/MOSFET-MAXheatsink.php

They mention the washer insulates the screw...but they neglect to mention the extremely small creepage between the heatsink and metal tab of the transistor through the hole
« Last Edit: November 04, 2019, 03:52:28 pm by eevcandies »
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Insulating washers & creepage/clearance issue at holes
« Reply #5 on: November 04, 2019, 04:49:18 pm »
The short answer is that a single insulating shoulder washer on the screw attaching a TO-220 to a grounded and/or accessible heatsink might very well not meet creepage requirements depending on the peak or RMS working voltage classification that applies. Figuring all of this stuff out is how compliance engineers make their living, and a good living it is, based on all the confusion I see/experience on a daily basis...

 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7678
  • Country: ca
Re: Insulating washers & creepage/clearance issue at holes
« Reply #6 on: November 04, 2019, 06:12:06 pm »
It's a non-issue in almost all of the electrical safety standards. Because a failure of the TO-220 mounting insulator/bushing would not lead to an unsafe condition.

An arc from the tab to the heatsink could make a heatsink hazardous live - but they are not touched by users and usually PE grounded, which would blow the mains fuse. The TO-220 plastics are UL 94-V for flammability. Insulated TO-220 Fullpack helps but the lead spacing is still the limiting factor.

If the TO-220 package and mounting hardware do not meet the creepage/clearance requirements, as treez has queried, then the safety agency certifier will bridge that as a non-countable fault. It's still not a safety hazard in a properly designed product.

I know of one or two safety standards that spell out the issue, I'd have to dig for them but extremely fringe level stuff.
 

Offline dom0

  • Super Contributor
  • ***
  • Posts: 1483
  • Country: 00
Re: Insulating washers & creepage/clearance issue at holes
« Reply #7 on: November 04, 2019, 06:19:09 pm »
Agree with floobydust. Insulating washer and pad are only used for functional insulation, not safety insulation. An interesting case would be a device mounted to the case / external heatsink on a class I device. Not sure if I actually ever saw that. It is of course very common on the secondary, already mains-insulated side of things (amplifiers, power supplies etc.).

Class II devices tend to use fullpak devices in similar situations. E.g. fullpak triac riveted to the case of a dimmer lightswitch. The case isn't grounded (and only covered cosmetically); these are class II.
« Last Edit: November 04, 2019, 06:20:41 pm by dom0 »
,
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16385
  • Country: za
Re: Insulating washers & creepage/clearance issue at holes
« Reply #8 on: November 04, 2019, 06:26:35 pm »
TO220 isolpak is the answer, at least then only the leads are the worry, and you can use additional insulation there to improve thermal performance and heat transfer with silicone thermal pads. Otherwise you have to use ceramic washers, as those ate thick but still transfer heat well, or simply have a second isolation barrier and a small heat spreader that is insulated from the main grounded heatsink.  If you are still not getting clearance you need another package more suited to high voltage use, and this them is where you start using industrial power electronics devices in isolated packages, where they use a thick aluminia substrate internally to provide the isolation, and a thick potting of Parylene to prevent creepage or conductive paths forming. Normally you get 6 IGBT, bipolar or MOSFET transistors, and 12 high power diodes in the package, arranged as 3 half bridges with freewheel diodes, and a separate 3 phase mains bridge rectifier.
 


Offline mzzj

  • Super Contributor
  • ***
  • Posts: 1298
  • Country: fi
Re: Insulating washers & creepage/clearance issue at holes
« Reply #10 on: November 04, 2019, 08:41:23 pm »
Agree with floobydust. Insulating washer and pad are only used for functional insulation, not safety insulation. An interesting case would be a device mounted to the case / external heatsink on a class I device. Not sure if I actually ever saw that. It is of course very common on the secondary, already mains-insulated side of things (amplifiers, power supplies etc.).

Seen more than one power supply where TO-220 mosfets are mounted to smaller aluminium block that is mounted and isolated with rated insulation to the case.

Intermediate piece is usually tied to "live" part of the circuitry, helps to keep the differential mode switching noise away from the ground where its harder to filter. 
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf