Grounding/Bonding in a subpanel

SETUP: 200 AMP Main Panel connected to a 100 AMP sub-panel via metallic nipple. A 75 AMP breaker in the main panel feeds the sub via #4
conductor. The Neutral is isolated from the ground in the sub. Bare grounding conductors are attached to the subpanel through an equipment bar attached to the metallic case of the sub.
Can someone explain to me why the NEUTRAL and GROUND must be isolated in the sub-panel since they are already bonded in the Main panel. Since the ground in the SUB is derived from the link of the two metallic enclosures, isn't the NEUTRAL in the SUB already bonded to the ground???
What happens if the NEUTRAL AND GROUND are also bonded in the subpanel?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
If N & G are bonded in the subpanel then the grounding conductor(s) become current carrying conductor(s).
This could cause magnetic field problems.
RE

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
O.K., but to repeat:
Since the ground in the SUB is derived from the link of the two metallic enclosures, isn't the NEUTRAL in the SUB already bonded to the ground since they are bonded in the MAIN?
Ryan Evans wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Randell Tarin wrote:

Yes it is bonded to ground back at the service. Neutral and ground are bonded at the service so a failure that connects hot to metal frames, conduits, ... will have a solid path back to the source transformer (via ground - service neutral,)resulting in high current, resulting in rapid breaker trip.
The service neutral-ground connection is also earthed to limit the voltage from earth to frames, conduits, ... and minimize the voltage from earth to the hot and neutral wires.
But bonding downstream puts ground conductors in parallel with the neutral, which puts neutral current on the ground wires as both Ben and Ryan said.
I would view the major reason not to bond donwstream the same as Ben. Another reason is neutral current on the ground system will put electrical 'noise' on the ground system which can cause problems with electronic equipment.
-- bud--

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Bud-- wrote:

Thank you Bud. I get it now. It was just somewhat counter-intuative. For me it was a matter of understanding the actual function of the GROUND.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Randell Tarin wrote:

The grounding chapter in the NEC is probably the most confusing of the commonly used ones. I saw a video from Mike Holt where he talked about "earthing" and "bonding". "Grounding" implies earthing, which largely is not what ground wires do. IMHO "grounding" is almost hoplessly confusing. I have grown to really like "earthing", which is used in the UK, and clearly describes a part of the "grounding" function. The 2008 NEC is supposed to revise the language a little, but probably not as much as I would like to see.
-- bud--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Neutral and ground are only bonded at the service entrance. If they are connected together elsewhere in the system, then the grounding conductors will carry some of the neutral return current. This can cause elevated potential on the housings of equipment, appliances, counduit, etc which creates a possible electric shock hazard. In the event of an open neutral at the supply, those exposed conductive surfaces could reach lethal voltage levels.
Ben Miller
--
Benjamin D. Miller, PE
B. MILLER ENGINEERING
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

You don't want your ground to carry the neutral current in the event of a loss of the neutral connection. Two reasons: 1. The grounds aren't supposed to carry current for obvious reasons, and 2. Your ground is likely a #8 copper in this case and it isn't sized to carry the current that might possibly come from that sub-panel. It might help to imagine a regular duplex receptacle. You wouldn't connect the ground and the neutral at the receptacle, would you? Same thing, larger scale in your sub-panel.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Others have already pointed out the code requires them to be separate. They have also mentioned that bonding them together will cause a portion of the neutral current to return via the ground wire.
As a practical matter, your panels are very close to each other and likely won't cause a problem. But consider what happens if the panels are further apart: The neutral current can develop a significant voltage drop between the panels. If both panels are bonded to ground, there will be a voltage difference between grounds in the house. This can cause current to flow in other paths that were never intended to have a current flow, like the building structure, phone, sat/cable TV, computer network cable, etc.. Worst case is if a fault occurs - the 100's or 1000's of amps of fault current can cause 10's of volts difference with very low source impedance. That could cause a significant current through the stray paths, causing damage to the other conductors or even a fire.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.