Author Topic: Applience metal chassis ground protective earth connection rules/regulations?  (Read 2949 times)

0 Members and 1 Guest are viewing this topic.

Offline ELS122Topic starter

  • Frequent Contributor
  • **
  • Posts: 937
  • Country: 00
I can't find any documentation on the requirements for a protective earth mains ground connected to a metal chassis in an electronic device. only info on fuse box grounding.
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9940
  • Country: us
US National Electric Code - NEC Article 250-110 (for fixed equipment)

https://www.electricallicenserenewal.com/Electrical-Continuing-Education-Courses/NEC-Content.php?sectionID=567.0

There are other standards but, in the end, they boil down to 'all exposed non-current carrying metallic surfaces must be grounded'.  That's why BNC connectors on oscilloscopes are grounded via the power cord.
« Last Edit: October 26, 2022, 04:05:26 pm by rstofer »
 
The following users thanked this post: ELS122

Offline helius

  • Super Contributor
  • ***
  • Posts: 3681
  • Country: us
You may get better results searching for "IEC Class I II"
When a protective earth is required (Class I, when the "double insulated" requirement cannot be met) it should be bonded from the receptacle to the chassis and from that point to any other panel. Typically this is achieved by brazing a threaded rod into the chassis and using nuts and lock washers to attach several ground wires to the rod using ring lugs. There are other methods which may be better for specific application requirements.
Class II devices are double insulated, which means that a single insulation failure of any current-carrying conductor will not place a hazardous voltage on any exposed surface. The IEC does not require protective earth for those.
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8593
  • Country: us
  • Retired, now restoring antique test equipment
One detail I remember from the US requirements on appliance wiring is that a ground lug attached to the metal frame with a screw fastener should not share that screw with another mechanical purpose.
Therefore, you should not use a mounting screw for a transformer or other object to hold the ground lug that attaches to the green/yellow wire.
The ground lug needs its own screw.
 

Online themadhippy

  • Super Contributor
  • ***
  • Posts: 3038
  • Country: gb
Quote
https://www.electricallicenserenewal.com/Electrical-Continuing-Education-Courses/NEC-Content.php?sectionID=567.0

american standards are a bit irrelevant in latvia
Quote
There are other standards but, in the end, they boil down to 'all exposed non-current carrying metallic surfaces must be grounded'.

even on  class II  equipment ?
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7527
  • Country: ca
What safety standard is OP following? For which countries/regions? Without knowing, this thread is going to be a collection of rules.

Another critical aspect is the appliance ground must be able to withstand fault current of the feeding branch circuit breaker. You have to be able to trip the breaker.
I design so that even if you have an IEC entry fused at 3A, in the event of reversed line/neutral, a ground fault for the feed 15A breaker can be withstood.
I've had certifiers inject into the grounding 15Arms for an hour and then check that connection/wire temperatures are not a problem.

As well, the usual green+yellow coloured wire, double screw/nut and non-corroding connections. It's largely common sense.
 

Offline CaptDon

  • Super Contributor
  • ***
  • Posts: 2067
  • Country: is
On non-medical equipment the ground path resistance from receptacle to chassis must not exceed 500 milliohms. On medical grade equipment it must not exceed 300 milliohms. Our testers at the hospital had a test lead to attach to the chassis and had a built-in receptacle. I also ran into a specification that said the ground path had to be able to support a current of 30 amps (I forget the time period, but it was short, perhaps 500 milliseconds) without failure, failure being defined as an increase in resistance which exceeded the specification (500/300) or burning / burned to open circuit. One of the old brutal testers did a 30 amp D.C. current regulated time controlled slam then switched to 15 amps regulated and measured voltage drop to calculate resistance. Crappy cords and connections couldn't sneak past 'Brutus". The chassis connection probe on Brutus has a thick and a thin wire as does the receptacle and Brutus (that was our nick-name for the unit) did a real 4-wire measurement. Brutus was known to blow the ground wire of the device under test to an open circuit and the failure usually occurred at the flex point right where the cord joins the plug at either end. Fun times in BioMed!!
Collector and repairer of vintage and not so vintage electronic gadgets and test equipment. What's the difference between a pizza and a musician? A pizza can feed a family of four!! Classically trained guitarist. Sound engineer.
 

Offline jkostb

  • Regular Contributor
  • *
  • Posts: 69
  • Country: np
Within EU you need to know to which EU directive your product must comply. The electrical safety is covered by the Low voltage directive or the Medical device directive (for medical devices). Your product is compliant with a directive if it is compliant with a standard as specified in the directives. Usually this means one of the following standards:
IEC60335-1 for consumer devices
IEC61010 for lab and industrial devices
IEC60601-1 for medical devices
IEC60079 for devices to be used in explosive atmospheres.
These standards are usually adopted worldwide. So you need to first determine which safety standard is applicable and then follow the rules as specifieid in the standard.

 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8593
  • Country: us
  • Retired, now restoring antique test equipment
Although national standards vary in detail and quantitatively, and depend on the application (medical, etc.), the basic requirement is that any metal surface that a human can encounter during operation should be connected appropriately to the appropriate PE wire (usually green or green/yellow), which in turn plugs into a good PE "ground" connection at the socket.
Thus, should anything fail inside to apply substantial voltage to the metal surface (such as a capacitor from line to chassis shorting), sufficient current will flow to blow the fuse or circuit breaker, and there shall be only low voltage from the metal surface to "ground" (e.g., water pipes) to avoid a shock to the human.
Double insulation requires that two levels of insulation must fail before such a connection to the metal occurs.
 

Online themadhippy

  • Super Contributor
  • ***
  • Posts: 3038
  • Country: gb
Quote
Although national standards vary in detail and quantitatively, and depend on the application (medical, etc.), the basic requirement is that any metal surface that a human can encounter during operation should be connected appropriately to the appropriate PE wire (usually green or green/yellow), which in turn plugs into a good PE "ground" connection at the socket.
Again maybe in the usa thats the case,but here in the uk it most certainly isn't, that only applies to class 1 equipment,if the internals are class 2 there no need to earth the  external metal bits,im sitting looking at a cd player,full metal case but no electrical earth in sight,same for the dab radio or  my sds drill full metal case,but no earth.
 
The following users thanked this post: Someone

Offline ELS122Topic starter

  • Frequent Contributor
  • **
  • Posts: 937
  • Country: 00
it seems the IEC documents are a lot less available (for free) than the NFPA ones, and im more interested in the US standard anyway.
I had a discussion on another forum of reducing hum in a guitar amp, and I brought up moving the mains earth connection next to the input jacks instead of next to the power transformer, this makes more sense anyway apart for the extra wite length which I thought might not comply with regulations, but it seems there is no "max wire length", and the resistance of the ground connection would probably be lower than when going trough a steel chassis.", or is that measured from the connection TO the chassis?
 

Offline Someone

  • Super Contributor
  • ***
  • Posts: 5016
  • Country: au
    • send complaints here
On non-medical equipment the ground path resistance from receptacle to chassis must not exceed 500 milliohms. On medical grade equipment it must not exceed 300 milliohms. Our testers at the hospital had a test lead to attach to the chassis and had a built-in receptacle. I also ran into a specification that said the ground path had to be able to support a current of 30 amps (I forget the time period, but it was short, perhaps 500 milliseconds) without failure, failure being defined as an increase in resistance which exceeded the specification (500/300) or burning / burned to open circuit. One of the old brutal testers did a 30 amp D.C. current regulated time controlled slam then switched to 15 amps regulated and measured voltage drop to calculate resistance. Crappy cords and connections couldn't sneak past 'Brutus". The chassis connection probe on Brutus has a thick and a thin wire as does the receptacle and Brutus (that was our nick-name for the unit) did a real 4-wire measurement. Brutus was known to blow the ground wire of the device under test to an open circuit and the failure usually occurred at the flex point right where the cord joins the plug at either end. Fun times in BioMed!!
Different countries, different standards, but that thump+buzz or "clunk" noise when engaging the large test current around a physically big loop is quite the experience.
The old regulations for Australia:
"equal to 1.5 times rated current of the equipment or 25 A, whichever is the greater" (now 2.0 times)
"resistance shall not exceed 100mOhm" (4 wire measurement, excluding all cabling)
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
american standards are a bit irrelevant in latvia

Electricity behaves the same everywhere in the world. The standards from one country may not be ideal for somewhere else but the basic principals should apply. Protective earth grounding serves the same purpose worldwide, only some of the minor details will vary.
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 10872
  • Country: us
  • $
What safety standard is OP following? For which countries/regions? Without knowing, this thread is going to be a collection of rules.

Another critical aspect is the appliance ground must be able to withstand fault current of the feeding branch circuit breaker. You have to be able to trip the breaker.
I design so that even if you have an IEC entry fused at 3A, in the event of reversed line/neutral, a ground fault for the feed 15A breaker can be withstood.
I've had certifiers inject into the grounding 15Arms for an hour and then check that connection/wire temperatures are not a problem.

As well, the usual green+yellow coloured wire, double screw/nut and non-corroding connections. It's largely common sense.

I always kind of figured this, on old equipment I try to replace the ground wire (usually same gauge as the line wire) with a higher gauge wire.

It kind of makes sense because if you interconnect equipment something with a bigger fuse could energize the ground and blow a thin ground wire before it fuses. At least I replace the 26? gauge ground wires used in some old equipment with 16 gauge lol. Like lets say you have a stack of equipment and a frayed cable that energizes the chassis of the equipment you are working on. Not a internal fault. Not the most likely thing to happen but unless the unit is externally insulated, it makes sense and its cheap enough IMO. Tektronix puts either plastic or power coat on the outside of some their units, but alot of stuff ends up just being shoddy deteriorating paint on sheet metal.. (also a good reason to re-do the rivets on metal chassis connectors with bolts, serrated nuts and star washers+grease, to make sure that is a solid earth point).

Good reason not to make equipment stacks anyway, I try to avoid tower more then 2 units high. Kind of makes me want to invest in a powder coating setup... just be careful reaching behind equipment. And it never hurts to make a rounded edge on the chassis, particularly if its banged up on the corners.
« Last Edit: October 26, 2022, 10:06:40 pm by coppercone2 »
 

Online themadhippy

  • Super Contributor
  • ***
  • Posts: 3038
  • Country: gb
Quote
The standards from one country may not be ideal for somewhere else but the basic principals should apply.
Following a set of standards in a different country isnt only not ideal,but  could lead to a dangerous situation. For example what are   black and  white wires used for according to the electrical standards?
usa answer neutral and live
uk answer phase and functional earth
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8593
  • Country: us
  • Retired, now restoring antique test equipment
Before I retired, we were sometimes told that our equipment had to meet both US and European standards.
I asked "Is neutral blue or white?"  US code says white is neutral (not live).
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Before I retired, we were sometimes told that our equipment had to meet both US and European standards.
I asked "Is neutral blue or white?"  US code says white is neutral (not live).

I see the European blue/brown colors fairly often in US equipment, I don't think anyone really cares when it's wiring inside equipment. UK/Europe adds a yellow stripe to ground IIRC but those or the plain green we typically use are easily recognized as ground. I've never seen anything other than some variant of green or bare metal used as protective earth.
 

Online themadhippy

  • Super Contributor
  • ***
  • Posts: 3038
  • Country: gb
Quote
I asked "Is neutral blue or white?"
or  black pre 2006 in the uk
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8593
  • Country: us
  • Retired, now restoring antique test equipment
We might as well bring back red for "PE".
For many years, in the US, white or grey are reserved for neutral, green (and the new-fangled green/yellow) is reserved for ground/PE, and any other color can be used for "hot", whether line, switched, or specific colors for phases.
Black is the default color for line.
These code assignments need not be rational, but they must be consistent to avoid problems.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
The colors are nearly irrelevant anyway, it's trivial to look up whatever color is required in the location in question. The basic principals of protective earth are the same everywhere, only minor details will change. Obviously there is no substitute for looking up the exact regulations where you are, but to understand the concepts any standard will do. These days there is a lot of overlap since products are often marketed worldwide.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf