Hipot and approach boundaries?

[StaySafe]

Looking for guidance on hipot (dielectric withstand) testing and approach boundaries.

We hipot test completed assemblies in the open (i.e., no insulated enclosure) with 2000 VDC at 5 mA. I can’t be specific about the type of assembly, but they have metal frames so the whole thing could be potentially energized if it fails.

For the testing, the completed assembly power plug is inserted into an adapter connected to the hipotter and the hipot test is performed for about a minute. The hipot test operator does not interact with the device under test during testing.

Is there any exception to the approach boundaries in 70E (2018) since the current is so low? Or is 2000 VDC, 2000 VDC no matter the current so table 130.1(D)(b) applies regardless? My past experience with hipot had a higher current, so boundaries were a no-brainer.

The minimal current appears to make this a low risk test – please correct me if you feel differently. I would really like your rationale one way or another since this is currently a debate in my company – basically, “5 mA won’t hurt you so the boundaries aren’t needed.”

If it makes a difference, we use the Associated Research 3765. The hipot tester is capable of 7.5 mA maximum with DC.

Thanks.

 

[zbang]

The minimal current appears to make this a low risk test – please correct me if you feel differently. I would really like your rationale one way or another since this is currently a debate in my company – basically, “5 mA won’t hurt you so the boundaries aren’t needed.”

Well, 5ma will kill you, and it's a lot easier to get 5ma through the body from a 2.0kv source than from a 0.120kv source.

Also, look to OSHA for workplace safety rules/regs, I don't think 70E is really relevant here.

I've always seen HiPot product tests set up in a grounded cage with a door interlock.

 

[zog]

Many years ago I sent that same question to OSHA for interpretation and was told it should be treated the same as an energized conductor at the same voltage. While I disagree with it, that is what OSHAs response was. I do not have the letter anymore (Was a different company) but perhaps it can be found in a search of OSHA website?

 

[Jraef]

In every place I have worked where hi-pot testing was performed, the tester was hooked up and an area was cordened off with caution Tape or physical barriers that kept everyone away from it during the test, including the tester (unless he was in the proper PPE).

Not the best example because the guy is posing for the picture in his work area and may not have the right PPE for a hi-pot of 480V gear...

 

[StaySafe]

Please nobody take anything I am going to post as challenging your expertise, I will be challenged with the same questions when I propose changes, so I need to get my ducks in a row. My hipot experience has always had a designated space; there is currently no extra space for hipot at my workplace, so requiring barriers/creating a space will be a battle. Challenging "the way we have always done it" is really fun.

I agree, at 2.0kv, you will essentially get full current available. However, 5 mA DC is lethal? Most sources I have seen puts that at >15 mA (typically for 60 Hz, though, not DC). The "safe" 5 mA has been the defense given for not having a designated space.

If 70E wouldn't set the approach boundary (i.e., barricades in this case) what should apply? As a company, 2018 70E has been adopted, so the more protective requirements will apply.

I scoured the OSHA interpretations and couldn't find one applicable. However, at least it suggests OSHA's approach. Thanks.

 

[zbang]

https://www.osha.gov/Publications/3075.html
(I didn't give this more than a cursory read, but it looks like there are some useful things)

5 milliamperes Slight shock felt; not painful but disturbing. Average individual can let go. Strong involuntary reactions can lead to other injuries.

Most GFCI's are suppose to trip at 5ma.

What is guarding and what protection does it offer?

Guarding involves locating or enclosing electric equipment to make sure people don't accidentally come into contact with its live parts. Effective guarding requires equipment with exposed parts operating at 50 volts or more to be placed where it is accessible only to authorized people qualified to work with it. Recommended locations are a room, vault, or similar enclosure; a balcony, gallery, or elevated platform; or a site elevated 8 feet (2.44 meters) or more above the floor. Sturdy, permanent screens also can serve as effective guards.

 

[GoldDigger]

For what it is worth, at those voltages, even with a 5ma AC current limit on the HiPot tester source, there may be a high momentary capacitive current when the arc to the "operator under test" is struck.

 

[StaySafe]

Thanks to all for your replies and help. It sounds like pretty much every other company uses proper distances and designated areas for hipotting. Sometimes I wonder if mine has been in a black hole for 20+ years.

Thanks, zbang, for the OSHA link. Pretty clear language to keep unqualified people away.

I think the correct and most defensible approach is to put up the boundaries to keep unqualified persons away. If nothing else, any shock requires medical evaluation so it is taken seriously by the company (thankfully there has only been one shock reported in the last several years - not in hipot).

I am going to use 70E approach boundaries, they may not be fully applicable, but I think it provides a proper measure of safety and is not just an arbitrary distance out of my head. Looking at the Associated Research hipot manual, they refer to 70E, so we are then also following manufacturer recommendations.

 

[ATSman]

In every place I have worked where hi-pot testing was performed, the tester was hooked up and an area was cordened off with caution Tape or physical barriers that kept everyone away from it during the test, including the tester (unless he was in the proper PPE).

Not the best example because the guy is posing for the picture in his work area and may not have the right PPE for a hi-pot of 480V gear...
View attachment 21112

Looks to me like he has the hipot test set lead connected to phase A of a GE PowerVac MV vacuum breaker.
480V gear is not normally hipot tested. Megger tested? yes

 

[Sahib]

480V gear is not normally hipot tested. Megger tested? yes

It is better both hipot and subsequent insulation resistance testing done.