|
Home > Archive > Electrical code Compliance > February 2006 > ESD Grounding with separate ground rod
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
ESD Grounding with separate ground rod
|
|
| John Hill 2006-02-19, 11:21 am |
| Hello everyone,
We have a Safety Committee at work that has declared it safe to ground
ESD mats and wristbands to a separate ground rod that is not connected
to the building green ground. To my knowledge this is unsafe due to the
possibility of dropping a phase and having the local green ground rise
above earth ground.
My understanding is that during some AC power system faults, the local
building green ground can rise above the earth ground. If a separate
ground rod is connected to the ESD equipment during one of these
faults, a voltage would appear between the building green ground and
the separate ESD ground rod.
What I am looking for is a definitive document reference to give to the
Safety Committee to help them understand the issues. They are very
convinced that what they are doing is safe.
John
| |
|
|
"John Hill" <jhill28590@comcast.net> wrote in message
news:1140361843.168954.257430@g44g2000cwa.googlegroups.com...
> Hello everyone,
>
> We have a Safety Committee at work that has declared it safe to ground
> ESD mats and wristbands to a separate ground rod that is not connected
> to the building green ground. To my knowledge this is unsafe due to the
> possibility of dropping a phase and having the local green ground rise
> above earth ground.
>
> My understanding is that during some AC power system faults, the local
> building green ground can rise above the earth ground. If a separate
> ground rod is connected to the ESD equipment during one of these
> faults, a voltage would appear between the building green ground and
> the separate ESD ground rod.
>
> What I am looking for is a definitive document reference to give to the
> Safety Committee to help them understand the issues. They are very
> convinced that what they are doing is safe.
>
> John
John your right, and they are DEAD wrong.
Best explanation of NEC 250 is the Soars Book on Grounding. Puts grounding
into words that MOST people can understand. Lots of illustrations, for the
complete and simple.............
Connecting the ground rods to the building ground system is covered in the
NEC as supplemental grounding. Problem, the conductors can be huge to the
new ground rod. It is after all an extension of the service ground.
I worked at a university that had all sorts of "intelligent degreed" folks
that would and did drive "clean" ground rods to ground equipment. One in
particular was a 400 amp 480v medical machine. They even used smurf tube to
prevent the disconnect on the wall from being involved. Only problem this
was on the 6th floor of a research building. Ya they drilled a hole in the
concrete and drove a ground rod in to it. So stupid it was sad. First time
they fired it up, they fried all of the their computers, ooops forgot that
Ethernet can create a ground loop. That is why I got called to fix the
problem.
I went through risk management folks, safety folks, and finally the project
managers. I borrowed slides from my local Independent Electrical Inspectors
Association and gave a 1.5 hour class on grounding. They have associations
in most cities across the country.
I do not know your situation but you usually can consider the building steel
as "effectively grounded" attach your ESD to building steel. I helped write
a new ground spec for the U. It contained a 1/2 thick by 12 inch high by 18
inches long steel plate installed in every equipment room. That way we could
use building steel. Architectural committee would have had kittens if you
ran a 4/0 down the outside of one of the buildings.
Other thing to make damn sure of is if there is more than one transformer
feeding the building. The transformers must be grounded in common. Other
wise you get a machine on one service with Z potential and the office with
the computers on another service with M potential. POOF the magic can fly
away. (equal potential is a good thing now days)
Good luck,
| |
| WeathermanBill 2006-02-20, 10:21 pm |
| On Sun, 19 Feb 2006 12:08:47 -0700, "SQLit" <sqlit@qwest.net> wrote:
>
>John your right, and they are DEAD wrong.
>
>Best explanation of NEC 250 is the Soars Book on Grounding. Puts grounding
>into words that MOST people can understand. Lots of illustrations, for the
>complete and simple.............
>
>Connecting the ground rods to the building ground system is covered in the
>NEC as supplemental grounding. Problem, the conductors can be huge to the
>new ground rod. It is after all an extension of the service ground.
>
>
>I worked at a university that had all sorts of "intelligent degreed" folks
>that would and did drive "clean" ground rods to ground equipment. One in
>particular was a 400 amp 480v medical machine. They even used smurf tube to
>prevent the disconnect on the wall from being involved. Only problem this
>was on the 6th floor of a research building. Ya they drilled a hole in the
>concrete and drove a ground rod in to it. So stupid it was sad. First time
>they fired it up, they fried all of the their computers, ooops forgot that
>Ethernet can create a ground loop. That is why I got called to fix the
>problem.
>
>I went through risk management folks, safety folks, and finally the project
>managers. I borrowed slides from my local Independent Electrical Inspectors
>Association and gave a 1.5 hour class on grounding. They have associations
>in most cities across the country.
>
>I do not know your situation but you usually can consider the building steel
>as "effectively grounded" attach your ESD to building steel. I helped write
>a new ground spec for the U. It contained a 1/2 thick by 12 inch high by 18
>inches long steel plate installed in every equipment room. That way we could
>use building steel. Architectural committee would have had kittens if you
>ran a 4/0 down the outside of one of the buildings.
>
>Other thing to make damn sure of is if there is more than one transformer
>feeding the building. The transformers must be grounded in common. Other
>wise you get a machine on one service with Z potential and the office with
>the computers on another service with M potential. POOF the magic can fly
>away. (equal potential is a good thing now days)
>
>Good luck
Have a lot of RF chambers. Seems the preferred method for grounding
is a new rod connected to large copper plates to 4/0 to ground post on
chamber. My beef was the receps, disconnects whatever attached to the
chamber. All had dialectic brakes but the problem I saw was the green
wire ground from the house power. Any ideas
Weathermanbill
| |
|
|
"WeathermanBill" <no@spam.com> wrote in message
news:Y2b6Q=EhDhXEcEeKrbA=SLWM+nKY@4ax.com...
> On Sun, 19 Feb 2006 12:08:47 -0700, "SQLit" <sqlit@qwest.net> wrote:
>
grounding[color=darkred]
the[color=darkred]
the[color=darkred]
folks[color=darkred]
to[color=darkred]
the[color=darkred]
time[color=darkred]
that[color=darkred]
project[color=darkred]
Inspectors[color=darkred]
associations[color=darkred]
steel[color=darkred]
write[color=darkred]
18[color=darkred]
could[color=darkred]
you[color=darkred]
with[color=darkred]
fly[color=darkred]
> Have a lot of RF chambers. Seems the preferred method for grounding
> is a new rod connected to large copper plates to 4/0 to ground post on
> chamber. My beef was the receps, disconnects whatever attached to the
> chamber. All had dialectic brakes but the problem I saw was the green
> wire ground from the house power. Any ideas
>
> Weathermanbill\\\
RF as in radio frequency?
Wire the recpts in plastic boxes and plastic conduit. Radio can be brutal.
Only problem I can think of is if they use the recpts for powering the
equipment making the radio signals.
An medical isolation panel might help. the ones used in explosive gas
operating rooms. Best to sit down and take a deep breath before asking a
price.
I will think on this and see if I can come up with some thing else.
| |
|
| SQLit wrote:
>
>
> RF as in radio frequency?
>
> Wire the recpts in plastic boxes and plastic conduit. Radio can be brutal.
> Only problem I can think of is if they use the recpts for powering the
> equipment making the radio signals.
> An medical isolation panel might help. the ones used in explosive gas
> operating rooms. Best to sit down and take a deep breath before asking a
> price.
> I will think on this and see if I can come up with some thing else.
>
It is not obvious to me what the "large copper plates" are. Specifically
if they are burried or otherwise are in contact with earth. An 8' ground
rod is not a particularly good ground
Also not obvious what "RF chambers" are - screen rooms that are occupied
or test cells that equipment only is placed in for RF test.
Or if the chamber green grounds connect only to the local ground
rod/system or also connect to the electric supply.
If not connected: the National Electrical Code REQUIRES these grounds to
connect to the service ground; not connecting breaks the path of ground
fault currents back to the power source to trip breakers - can be a
major safety issue.
Of interest might be an Army publication dated 2002: Grounding and
Bonding in Command, Control, Communications, Computer, Intelligence,
Surveillance, and Reconnaissance (C4ISR) Facilities
http://www.usace.army.mil/usace-docs/armytm/tm5-690/
This covers screen rooms containing electronic equipent and issues of:
power/safety grounds
shielding for EMI interferrence coming in
shielding for EMP (electromagnetic pulse) coming in
shielding (TEMPEST) for any signals getting out that could
compromise security
It is 3.8M, 183 page PDF; I started reading it - looks interesting.
bud--
| |
|
|
"Bud--" <remove.BudNews@isp.com> wrote in message
news:34dc7$43fb4070$4213eb89$14126@DIALUPUSA.NET...
> SQLit wrote:
brutal.[color=darkred]
>
> It is not obvious to me what the "large copper plates" are. Specifically
> if they are burried or otherwise are in contact with earth. An 8' ground
> rod is not a particularly good ground
Some engineers use copper/steel plates cause they think the surface area is
larger than a ground rod.
I did some work in the Prescott AZ area and had to make 20 1 foot square by
1/4 inch thick plates for a job I did. Brazed on some #4 bare copper and
installed them per the plans. Testing was about what was expected in
decomposed granite, barely below 25 ohms.
My personal favorite is the chemical ground rods that are available. I do a
ground resistance test on a map and send it to the manuafacture. They then
tell me how big the chem rod needs to be.
I have installed a pair of 40 foot long by 2 inch in diameter for one
electrical system. Not much earth in an old river bed. Not cheap but they
can do the job.
-------snipped==================
| |
| WeathermanBill 2006-02-21, 9:21 pm |
| On Mon, 20 Feb 2006 21:51:11 -0700, "SQLit" <sqlit@qwest.net> wrote:
>
>RF as in radio frequency?
>
>Wire the recpts in plastic boxes and plastic conduit. Radio can be brutal.
>Only problem I can think of is if they use the recpts for powering the
>equipment making the radio signals.
>An medical isolation panel might help. the ones used in explosive gas
>operating rooms. Best to sit down and take a deep breath before asking a
>price.
>I will think on this and see if I can come up with some thing else.
That's correct - radio freq chambers. Like the idea of plastic boxes
and PVC - that solves the ground strap problem on the commercial grade
receps. The equipment inside uses power that is supplied thru factory
installed filters. We usually install a loadcenter to distribute this
filtered power based on customer needs. To go one step further -
might be a good idea to use PVC pipe inside to cut down on the metal.
Weathermanbill
| |
|
| SQLit wrote:
> "Bud--" <remove.BudNews@isp.com> wrote in message
> news:34dc7$43fb4070$4213eb89$14126@DIALUPUSA.NET...
>
>
> brutal.
>
>
>
> Some engineers use copper/steel plates cause they think the surface area is
> larger than a ground rod.
>
My question was whether the plate was an electrode or a ground bus but
the OP referrs to plates - multiple.
> I did some work in the Prescott AZ area and had to make 20 1 foot square by
> 1/4 inch thick plates for a job I did. Brazed on some #4 bare copper and
> installed them per the plans. Testing was about what was expected in
> decomposed granite, barely below 25 ohms.
>
> My personal favorite is the chemical ground rods that are available. I do a
> ground resistance test on a map and send it to the manuafacture. They then
> tell me how big the chem rod needs to be.
> I have installed a pair of 40 foot long by 2 inch in diameter for one
> electrical system. Not much earth in an old river bed. Not cheap but they
> can do the job.
>
I would think Ufer grounds would be useful. IIRC they were tested in the
Arizona desert and had resistances below 5 ohms. I think alkaline
materials leach out of the concrete making it also 'chemical' source.
bud--
| |
| John Hill 2006-02-25, 1:21 pm |
| Hello again,
The story gets better and I am going to need more help to deal with
this issue. I have been told that a local Building Inspector wrote a
citation and forced the maintenance group to disconnect a ground rod
from the green ground of the building power system. This ground rod is
currently being used to ground a flammables storage cabinet containing
spray paint. Hence, the cabinet and separate ground rod is not
connected to the power system green ground.
I expect the wire between the supplemental ground rod and the building
green ground was a small gage wire. During a ground fault this wire
would have been a fusible link. I expect the inspector said fix it (put
in a large gage wire) or disconnect it and the maintenance group
disconnected it.
Now every time the building maintenance group needs a ground for
electrostatic dissipation, they drop a new ground rod through the
concrete floor and do not connect it to the building green ground.
These ground rods are not connected to anything other than the thing
they are trying to protect. There are now separate ground rods for ESD
(Electrostatic Discharge) protected benches, flammables storage
cabinets, and Radio Frequency Shield Rooms.
By the way, a Radio Frequency Shield Room is a large metal box with a
metal door and special filters for the AC power going into the room.
These rooms allow testing of electrical products without the radio
signals that we use everyday like TV, cell phones and the like. It
creates a quite radio frequency environment to test electrical
products. The metal shields of the room are typically connected to the
building green ground in the filters. It is the one place in the
building where I cannot be reached by cell phone. The radio frequencies
use by the cell phone cannot get through the shielding of the room.
Friday my copy of the NEC and the Soars book came in. I am reading them
now, but have not found anything that is definitive. This is really
getting out of control and turning into a big project.
I am going to need to find a specific section of the NEC or other
nationally recognized document that specifies that all supplemental
Ground Rods must be connected to the power system ground. Does anyone
know the specific section of the NEC that covers this?
Best regards,
John
| |
|
| John Hill wrote:
> Hello again,
>
> The story gets better and I am going to need more help to deal with
> this issue. I have been told that a local Building Inspector wrote a
> citation and forced the maintenance group to disconnect a ground rod
> from the green ground of the building power system. This ground rod is
> currently being used to ground a flammables storage cabinet containing
> spray paint. Hence, the cabinet and separate ground rod is not
> connected to the power system green ground.
>
> I expect the wire between the supplemental ground rod and the building
> green ground was a small gage wire. During a ground fault this wire
> would have been a fusible link. I expect the inspector said fix it (put
> in a large gage wire) or disconnect it and the maintenance group
> disconnected it.
>
> Now every time the building maintenance group needs a ground for
> electrostatic dissipation, they drop a new ground rod through the
> concrete floor and do not connect it to the building green ground.
> These ground rods are not connected to anything other than the thing
> they are trying to protect. There are now separate ground rods for ESD
> (Electrostatic Discharge) protected benches, flammables storage
> cabinets, and Radio Frequency Shield Rooms.
>
> By the way, a Radio Frequency Shield Room is a large metal box with a
> metal door and special filters for the AC power going into the room.
> These rooms allow testing of electrical products without the radio
> signals that we use everyday like TV, cell phones and the like. It
> creates a quite radio frequency environment to test electrical
> products. The metal shields of the room are typically connected to the
> building green ground in the filters. It is the one place in the
> building where I cannot be reached by cell phone. The radio frequencies
> use by the cell phone cannot get through the shielding of the room.
>
> Friday my copy of the NEC and the Soars book came in. I am reading them
> now, but have not found anything that is definitive. This is really
> getting out of control and turning into a big project.
>
> I am going to need to find a specific section of the NEC or other
> nationally recognized document that specifies that all supplemental
> Ground Rods must be connected to the power system ground. Does anyone
> know the specific section of the NEC that covers this?
>
> Best regards,
>
> John
>
Interesting story.
There are 2 main purposes of power system grounding. First is to bond
equipment or provide a low resistance path back to the neutral (which is
connected to the metal enclosures at the service in almost all systems)
to rapidly operate a breaker/fuse if there is an accidental fault from
hot to ground (enclosures, raceways, etc). Second is to hold the neutral
voltage at 'earth' potential to limit the system voltages and provide a
sink for lightning and similar events.
For grounds on line powered equipment 2005NEC 250.4-A-5 prohibits the
use of the earth as a ground fault path. That is because a path through
the earth will have a high resistance and won't reliably provide the
bonding function to trip a breaker. Say you have an electrical device
grounded only through a ground rod, and the device develops a
hot-to-ground fault. In this case the return path is through the ground
rod, through the earth, through the system grounding electrode, to the
neutral. The path via the earth, especially the ground rod-to-earth
resistance will be high. If the ground rod-to-earth resistance is 10
ohms, much better than the 25 ohms NEC requirement for a single ground
rod, and there is a fault to 120V, the current will be 120/10 = 12 Amps.
This likely won't trip the supply breaker, and if it does there will be
a long time delay. In the mean time, the "ground" will be at 120 Volts
with respect to properly grounded power wiring. For power wiring, a
solid metal return path is required for equipment ground/bond. If the
power wiring ground on an ESD bench is attached only to a rod it is a
major hazard. (If it si the ESD surface it is probably resistance
grounded an doesn't make much difference.) If the screen rooms are
connected only to a ground rod, not the system bond/ground that is a
major NEC violation. OSHA incidentally would take a very dim view of the
situation which could cost a lot of money, an effective motivtor.
Grounding of paint cabinets is covered by NFPA standards other than the
NEC and I am not familiar with them. However assume a paint cabinet is
grounded to a rod only. And assume the service is grounded only through
a rod - of whatever ground resitance. If there is an event that causes a
ground current, like a utility surge that lifts the neutral voltage 100
volts, there is a rapid fall-off of voltage away from the ground rod.
From the earth 5' from the ground rod to the rod will be about 100
Volts. The paint cabinet rod far away will be at the same potential as
the earth at 5' while the grounded/bonded electrical aparatus will be
100 Volts higher. An event like this would be short lived. And the power
system grounding electrode would hopefull be larger area than a rod, but
still the remote rod can wind up at a very different voltage than the
power system grounded/bonded surfaces. Stray earth currents, which are
often present, could also cause isolated ground rods to be at a
different voltage than the grounded/bonded electrical system surfaces. I
do not understand why they don't just use the the power ground with no
rods, which in most cases I would think would be easier. In all cases I
would think the goal would be for everything to be at the same
potential. OSHA might be interested in some of these non-power practices
also.
bud--
| |
|
| Article 250.54: Supplementary grounding electrodes shall be permitted
to be connected to the equipment grounding conductors specified in
250.118 ... but the earth shall not be used as the sole equipment
grounding conductor.
You are confusing numerous purposes of earthing. Earthing for human
safety is defined by code which is why all earthing connections must be
interconnected. Grounding for static electricity need not be earthed.
For static electricity, charges are stored on both sides of a shoe.
Static protection is about connecting top of shoe (ie via arm) to
bottom of shoe to discharge static charges. Otherwise static would
create electricity down arm, through electronics, through table, floor,
to bottom of shoes. Note no electricity without a complete circuit.
Trace the circuit. Note that earth is not in this circuit. Concrete
and linoleum tile can be two examples of conductors when discussing
static electricity. Also remember what is and is not conductive per
each type of electricity.
Same applies to an RF anechoic chamber. Earthing serves multiple
masters - including safety ground. Your earthing must meet NEC
requirements AND address a solution to your problem. Earthing a
chamber means that a radio wave does not have separate circuits through
air and through ground that would create one kind of signal leakage.
That means every wire leaving the chamber must first connect to the
single point ground that also includes earthing electrode. Remember,
wire is not just a conductor. Wire is an electronic component which is
why electricity is different at both ends of a wire.
So why is a paint storage locker earthed? Before defining an
earthing system, first define what circuit is being protected or what
the discharge path is. And remember, many things you normally call
non-conductive, instead, become conductive depending on electrical
parameters such as voltage, current, transient frequencies, etc. They
may be conductive to your problem and yet just also meet other NEC
requirements for NEC's problem - human safety.
Those who recommended earthing of anti-static mats for static
electric protection don't understand the problem they are trying to
solve. Earthing for static protection should be obviously nonsense.
But many just use a concept called ground - and then assume that is
earth. Ground is a subjective concept unique to each circuit.
Earthing is only another type of ground not to be confused with the
safety ground in a break box, the motherboard ground inside
electronics, and virtual grounds used in op-amps. Ground does not
necessary mean earthing - in part because wire is an electrical
component - not just a conductor when discussing grounds.
John Hill wrote:
> Hello again,
>
> The story gets better and I am going to need more help to deal with
> this issue. I have been told that a local Building Inspector wrote a
> citation and forced the maintenance group to disconnect a ground rod
> from the green ground of the building power system. This ground rod is
> currently being used to ground a flammables storage cabinet containing
> spray paint. Hence, the cabinet and separate ground rod is not
> connected to the power system green ground.
>
> I expect the wire between the supplemental ground rod and the building
> green ground was a small gage wire. During a ground fault this wire
> would have been a fusible link. I expect the inspector said fix it (put
> in a large gage wire) or disconnect it and the maintenance group
> disconnected it.
>
> Now every time the building maintenance group needs a ground for
> electrostatic dissipation, they drop a new ground rod through the
> concrete floor and do not connect it to the building green ground.
> These ground rods are not connected to anything other than the thing
> they are trying to protect. There are now separate ground rods for ESD
> (Electrostatic Discharge) protected benches, flammables storage
> cabinets, and Radio Frequency Shield Rooms.
> ...
| |
| Joerg 2006-02-28, 10:21 pm |
| Hello John,
>
> Now every time the building maintenance group needs a ground for
> electrostatic dissipation, they drop a new ground rod through the
> concrete floor and do not connect it to the building green ground. ...
Do the owners of the building know about this? They will not be
enthused, to say the least. Also, this scenario can one day lead to a
punctured sewer line or something like that.
>
> I am going to need to find a specific section of the NEC or other
> nationally recognized document that specifies that all supplemental
> Ground Rods must be connected to the power system ground. Does anyone
> know the specific section of the NEC that covers this?
>
I don't believe that the NEC deals with ESD much. But this document has
several links in it:
http://www.desco.com/info/esdhigh.doc
Regards, Joerg
http://www.analogconsultants.com
|
|
|
|
|