AV Equipment Rack Power Strip Bonding Jumper: Required? Not Required?

Status
Not open for further replies.
We have an AV equipment rack manufactured by Chief Manufacturing. It's a wall-mounted rack model (SWR-24-17). Inside, we keep AV equipment mounted to the rack rails. Power for the AV equipment is being supplied by two, rack-mounted power strips manufactured by Furman. The power strips are model M-8x2. These power strips are plugged into wall outlets below the wall-mounted AV equipment rack. This rack is used in a classroom with the usual students, faculty, and staff coming into contact with it.


Chief supplies two green bonding jumper cables. One of the green bonding jumper cables connects the front door of the rack to one of the rack's interior bonding studs. A second green bonding jumper cable connects another bonding stud within the rack to one of the rack rails.


The Furman power strips do not have bonding studs.


However, I'm reading a document produced by rack manufacturer "Great Atlantic" that suggests power supplies within a rack have metal chassis that should be connected to one of the bonding studs within the rack with a similar green bonding jumper cable.


In addition, InfoComm International, the professional association representing the commercial audiovisual industry, has just published a new textbook for AV installers who seek to become certified by InfoComm International (CTS-I). This is an ANSI-accredited certification.


That textbook strongly suggests likewise: Power strips within an audiovisual equipment rack have metal chassis that should be bonded to the rack's bonding stud using the green bonding jumper cable.


I've asked the authority having jurisdiction, a State of Idaho Division of Building Safety, if the green jumper cable is required for power supplies within the rack to be connected to the rack's bonding stud.


Idaho's current "interpretation" of NEC code is, apparently, that such a bonding jumper cable is *not* required as long as the power strip "is properly grounded." I assume they are talking about the standard three-prong power plug that the power strips use to plug into the wall outlet.


I'm confused. On the one hand, InfoComm and a major rack manufacturer strongly advise the bonding jumper to connect the power strip chassis to the rack stud for safety reasons. On the other hand, the local authority having jurisdiction says it's not required. As you may already surmise, when something like this is not required, it won't get done.


What does the NEC code really require, in this circumstance?


I wish to point out that InfoComm International makes the point that the rack ears of an audiovisual rack-mounted power supply should not be expected to bond, necessarily, due to paint on the rack rails. That's why the bonding jumper solution is recommended, apparently. The author, Dr. Shonan Noronha, Ed.D., writes: "Racks cannot be effectively bonded through PDU (power distribution unit) rack ears or through power strip mounting clips. Rack power should be bonded to the grounding stud on the rack."

I've found a U.S. Department of Energy document that also seems to support this type of bonding.

Are the Idaho folks mistaken?

P.S. Every time I touch this rack, I get the familiar static electricity "pop."
 

cadpoint

Senior Member
Location
Durham, NC
The NEC doesn't care about self contained electrical equipment.

From what I've seen, low voltage equipment and their installers have always seemed to be more sensitive to a complete ground system!

The A/V equipment is covered by the UL and anything else they submitted their
product to.

Putting the various electrical equipment on a steel platform is all covered by each
piece having a ground prong for the power strip.

Any company can state and describe preferred additional grounding technic's for their equipment.

With use of or installing a rack or using a portable structure. The A/V Group is
trying to just eliminate the small potential difference that might build up from the type of the structure itself, not generally from the equipment that's on it.

Walking the equipment across tile to carpet is enough to build up some static.
It can also be as simple as the soles on your shoes!

You could go seek out NEC 70 B and E. Where they will state that additional bonding
and grounding is in Article 250 if the NEC.

I don't know what to say about some of the other statements.
 
Here is the Middle Atlantic advisory. See page 51. The bonding jumper is there.

https://www.middleatlantic.com/~/media/MiddleAtlantic/Documents/WhitePapers/PowerPaper.ashx

On page 50, Middle Atlantic states:

"Since power strip mounting hardware cannot be relied upon to conduct fault current, a bonding jumper must be installed between the power strip chassis and the rack."​

Meanwhile, the document from the U.S. Department of Energy is here.

"All metal parts of electrical equipment enclosures and chassis shall be bonded and grounded as per the NEC. The methods chosen to avoid ground loops and reduce noise shall meet the requirements of the NEC 250.6."​
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
Here is the Middle Atlantic advisory. See page 51. The bonding jumper is there.

https://www.middleatlantic.com/~/media/MiddleAtlantic/Documents/WhitePapers/PowerPaper.ashx

On page 50, Middle Atlantic states:

"Since power strip mounting hardware cannot be relied upon to conduct fault current, a bonding jumper must be installed between the power strip chassis and the rack."​

Meanwhile, the document from the U.S. Department of Energy is here.

"All metal parts of electrical equipment enclosures and chassis shall be bonded and grounded as per the NEC. The methods chosen to avoid ground loops and reduce noise shall meet the requirements of the NEC 250.6."​

Sounds like boiler plate to me written by someone who does not have a really good idea what he is talking about. My guess is a tech writer.

The NEC does not care a whit about whether the equipment works or not. Only that the electrical system is safe. Bolting and screwing together the metal hardware has always been an acceptable means of bonding. Now you might be able to do that fastening in an inappropriate way that results in a system that is not effectively bonded, but that is poor workmanship, and not an issue with how it was done. besides that, the ground pin of the power strip is where the ground fault current probably goes, not through the mounting hardware.

In any case, look at the title of 250.6.

250.6 Objectionable Current.

It appears to have nothing whatsoever to do with bonding, but rather with grounding, and nothing in the instructions is telling you to make a direct connection to a grounding electrode.

In any case, there is nothing in the NEC that would require this rack or any other similar rack to be grounded, only bonded.
 
Last edited:
I think a lot of the bonding on audio visual equipment is above and beyond the NEC. It is more like an equipotential plane to keep everything in the area at the same potential. This is not for personal protection but for protecting the sensitive electronics. Similar to a computor room raised floor where all the metallic objects are bonded together. At a tv station we had to run two inch copper pipe which was silver soldered at all joints under all of the racks. This is bonded to all of the racks and also to all metallic objects in the area including steel structure and metallic conduit. My understanding of this is that they want any potential difference to dissipate, not in a linear fashion as using the egc, but spread out and use every available path. At a 911 center we even had to install dragging strips on the rolling office chairs, so the static would not build up when rolling accross the carpeted raised floor panels. The static shocks can do damage to the electronics.
 
Petersonra wrote:

"Bolting and screwing together the metal hardware has always been an acceptable means of bonding."

Not according to InfoComm International. From their new CTS-I textbook:

"Racks cannot be effectively bonded through PDU rack ears or through power strip mounting clips."
(Page 145 of my Barnes & Noble version of the textbook)

I have decided to double check this with InfoComm International. They aren't backing down. Their director of education programs development graciously responded to a recent email that I sent requesting clarification.

Here is an excerpt of the prompt and direct response from InfoComm International:
NEC250.6 (A) (2), ?Grounding of Electrical Equipment. Normally non-current-carrying conductive materials enclosing electrical conductors or equipment, or forming part of such equipment, shall be connected to earth so as to limit the voltage to ground on these materials.​

"The statement above means that racks must be grounded. For example, a PDU is mounted in a rack and plugged into the electrical receptacle. The EGC on the receptacle is grounded, and the PDU has an EGC in its power cord. That EGC in the power cord is bonded to the chassis of the PDU inside the PDU. This creates a continuous conductive EGC pathway from the receptacle, through the EGC of the PDU?s power cord and to the chassis of the mounted PDU.​
"However, this does not necessarily create sufficient electrical continuity between the painted/powdercoated surfaces of the PDU chassis and the rack rails, especially if insulated washers were used underneath the rack screws to avoid scratching the PDU?s faceplate. You therefore cannot be assured in this case that the rack has been sufficiently grounded according to NEC requirements.​
"If you have a relatively modern equipment rack, this problem is not difficult to solve. Many equipment racks include a grounding stud in the bottom of the rack. To ground the rack, you simply need to run an equipment grounding conductor from the grounding stud to the PDU. In fact, many PDUs have an external grounding stud to accomplish this connection. To size this EGC properly, look at the size of the EGC used in the PDU?s power cord. If the PDU?s power cord uses 12 gauge conductors, use a 12 gauge or larger EGC between the rack grounding stud and the PDU?s grounding stud.​
"This procedure would be the same regardless of whether you are using an Isolated Ground system.​
"In short, if the PDU is properly connected to a grounded receptacle and the rack is grounded to the PDU, the rack should be sufficiently grounded. NEC requirements are met."

So, I've decided to check with the major rack manufacturers. Most recommend that the rack be grounded in either the way InfoComm has recommended here or by connection to the building ground.

Tripp Lite, for example, states:

"Connect your facility's earth ground connection to the grounding stud not used by door connections, using an 8 AWG (3.264mm) wire."

Petersonra wrote:

"Now you might be able to do that fastening in an inappropriate way that results in a system that is not effectively bonded, but that is poor workmanship, and not an issue with how it was done. Besides that, the ground pin of the power strip is where the ground fault current probably goes, not through the mounting hardware."

But therein is the problem, right? Paint and powder coating might get in the way; the professional association representing the worldwide audiovisual association seems to be correctly pointing to the very real possibility that simply screwing bolts into the rack rails won't necessarily pierce paint/coatings (possibly washers) that might impede electrical continuity between the rack ears and the rack rails when mounting power sources. So, the conclusion I'm drawing is that could lead to a situation where fault current might take the path of least resistance, perhaps a human being. This is why I believe the local authority having jurisdiction might have a degree of expertise in electrical safety, but also may not be aware of the issues that pertain to racks, specifically, what the rack manufacturers are recommending, and how the industry is interpreting the code requirements stated in NEC250.6 (A) (2).

In summary, this code seems to be subject to a wide variety of interpretations, depending on whom one asks.
 
One more observation:

I have noted that the networking department, in my organization, grounds all metal server racks, as recommended by Tripp Lite. The director of that department strives to meet BICSI standards.

But the AV department, in this particular organization, is a CONTRAST.

I am the first individual in my organization to have ever achieved InfoComm International certification and, while achieving my certification was mildly praised by a few folks, the overall response of most AV managers/coworkers was to simply yawn. Only one other individual in my organization, after learning of my certification, decided to go forward with earning his own certification, which made me proud, because it meant my achievement had persuaded a colleague to achieve something I feel is significant. But, believe it or not, he and I are the only two in our entire audiovisual division to hold these certifications and I sense many of our fellow AV colleagues -- AND EVEN MANAGERS -- sneer and snicker (and get paid a whole lot more money for AV work that you would have to see to believe).

NONE OF THE AV DEPARTMENT RACKS WITHIN MY ORGANIZATION ARE GROUNDED IN THE SAME WAY THAT THE NETWORKING RACKS WITHIN MY ORGANIZATION ARE GROUNDED.

Since both BICSI and InfoComm International often partner on various initiatives, I feel relatively safe following their recommendations.

In my state, there is no limited energy licensing requirement. So you would not believe these AV racks that I deal with in my organization. Not only do they fail to comply with InfoComm International standards for proper signal separation and basic rack dressing, cable labeling, rack documentation recommendations, they also don't appear to even be grounded in the way that major industry players recommend, which is a bit frightening.

I've lived in Idaho all my life. But I've also seen -- firsthand -- the dysfunction that often manifests itself here due to a stubborn streak that often appears to resist any constructive criticism, valid concern for safety, or recommendation from anything or anybody outside of the "Idaho Royalty."

Tripp Lite's warning about grounding is explicit: "DO NOT USE THE ENCLOSURE WITHOUT AN EARTH GROUND CONNECTION."
 
Last edited:

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
My guess is that a lot of AV people are not very good electricians, nor do they have a good understanding of how to make an electrical system safe. To protect themselves from amateurs working on stuff that can kill them, they add requirements like this, hoping to kill fewer of them.

The reality is that if the frame is just screwed together without making sure there is actual good contact between all the metal pieces, the bonding stud on the frame does not do all that much for you, since you can still have energized parts of the frame during a fault event with no way to clear it.

The fault clearing path starts at the ground pin on the plug that supplies power to the rack. Every piece of metal inside the rack needs to be effectively bonded to that pin in some way. If screwing the case of that power strip to the frame does nor result in an effective fault clearing path, you have a potentially dangerous condition. I have never seen a power strip that has a bonding stud on it to connect to the bonding stud of the frame of the cabinet. That would be the only bonding that I could see making a difference at all, and even then it would be dependent on effective bonding of all the metal pieces of the cabinet.
 

NetfloorUSA

Member
Location
U.S.A.
Don't rule out carpet on a raised floor just yet!

Don't rule out carpet on a raised floor just yet!

I think a lot of the bonding on audio visual equipment is above and beyond the NEC. It is more like an equipotential plane to keep everything in the area at the same potential. This is not for personal protection but for protecting the sensitive electronics. Similar to a computor room raised floor where all the metallic objects are bonded together. At a tv station we had to run two inch copper pipe which was silver soldered at all joints under all of the racks. This is bonded to all of the racks and also to all metallic objects in the area including steel structure and metallic conduit. My understanding of this is that they want any potential difference to dissipate, not in a linear fashion as using the egc, but spread out and use every available path. At a 911 center we even had to install dragging strips on the rolling office chairs, so the static would not build up when rolling across the carpeted raised floor panels. The static shocks can do damage to the electronics.


Hi there, I just wanted to add to your very informative post that they do make static-resistant carpet tiles for just this situation. It is somewhat more expensive than standard carpet tiles, but it does make a significant difference.

Also, carpet is not necessarily the best choice for covering a raised floor. Some people like it, but it can harbor germs, dirt and like you said, the whole static electricity issue, which is compounded with the more electronics you add. You can get raised access flooring that does not have carpet on it and has built in grounding strips (on the panel edges, and the support stringers). You can get finishes like tile, HPL (high pressure laminate), terrazzo, etc.

We see a lot of 911 centers moving away from carpet because A) sick building syndrome and B) 911 directors love the look of a floor finish other than carpet.

Have a wonderful day!
 
Status
Not open for further replies.
Top