Telecommunication Equipment Factory Grounding

Status
Not open for further replies.
J

jpower

Member
Hi friends,
There is a factory manufacturing telecommunication equipment such as TV TX/RX or mobile cells. Do you think that the following grounding scheme would be resonable for such a premise (double-floor, more than 1000 square meter area, and fed by a 20kV/380v delta/star)?

- grounding electrode system:
#4 earth well connected in a star manner to the service panel grounding bar, which is bonded to the neutral bar and, also, the structural steelwork and piping as well (overal resistance less than 4 ohm). Distance between transformer frame ground electrode and building grounding electrode system is less than 20 meters, Doesn't it poses any danger? Isn't a ground ring around the building necessary?

- equipment grounding:
The equipment signal reference (on printed board) is bonded to the equipment chasis/enclosure and to the safety equipment grounding conductor (EGC), i.e. a common safety/signal EGC. All the insulated EGCs are routed in a star manner down to the service panel ground bar (single point grounding). Anyway, when testing equipment, there is a ground loop between equipment and instrument. To resolve that, is it reasonable to feed both items from a common AC receptacle and swap the feeding cords up to them?
Equipment work at MHz Range.

Any advise or comment is highly appreciated ...
 
I will take a shot at this as I seen this on TV once and slept at a Holiday Inn last night.

In the telephone business we have power and ground specifications that equipment manufactures must meet in order for their equipment to qualify for use in our offices, and it directly addresses your questions.

The equipment must be able to meet either an ?Integrated Ground Plane? or ?Isolated Ground Plane? environment via option straps. All this means is the battery return, logic/signal ground, equipment frame all have to have a means to either be kept isolated or bonded all together.

The means to do this is just like I stated via option bonding jumpers and terminals. By doing this it allows the operating companies to either install the equipment in an Integrated Ground Plane or Isolated Ground Plane.

In the Integrated Ground Plane we just run either battery or AC power, and connect a frame ground to the equipment rack, then install the optional bonding straps that bond all the internal equipment ground together.

For the Isolated Ground Plane we again either run battery or AC power, leave the optional bonding straps out, run a frame ground to bond the equipment frame, and an additional ground to the Logic/Signal ground terminal. This separates (isolates) the safety (EGC), Frame Ground (supplemental), and signal grounds.

In your case, sounds like you are using AC power. The AC power is then converted to DC via some sort or rectifier which constitutes a separately derived system (SDS). The common return on the DC side is your signal ground. You have a choice what you reference it too. You can float it because it is internal to the equipment and is not under NEC rules (UL), bond it to the equipment chassis, or run it to a terminal called SG for an external reference. This brings us back to where I started. Make it an option

To address your ground electrode question, its not that important for a single piece of equipment powered from AC as it contains a SDS which forms a new reference point. The impedance of a 20 meter single wire is an open circuit to high frequencies, so who cares what size it is, if it is radial, parallel, or looped. The more loops, the lower the impedance. That is another subject all together.
 
Dereck,
Thanks for your comments,
According to your explanations, I think an integrated ground plane (IGP) serves our purpose. As a matter of fact, the signal ground of all of these high frequency equipments is internally bonded to the chassis and that is in turn bonded to the AC incoming EGC (i.e. a black box with just one grounding conductor). This is all according to the manufacturer practice.
Now, my question is what kind of structure (topology) is necessary for this IGP considering there is just one grounding point is accessible for each equipment's signal/safety ground through its plug grounding pin and, also, considering equipment are manufactured and tested in all the corners of this double story building with more than 1000 square meter? If it's advisable to wire all the building's grounding type receptacles down to the service panel ground bar in the basement (with or without bonding to steel structure in the halfway). But as you stated it would be open circuit for high frequencies. Also, it's not possible to put in a grid-type reference plane for each story, because the factory's civil work is ended up and it's in operation. So, what do you think we should do in this respect?
By the way, there are some other concerns regarding this factory as follows,
- the distance between the main transformer (20kV/380v) frame's grounding electrode and factory's grounding electrode system, GES, to which the transformer's neutral is connected, is less than 20 meters violating standards such as VDE. What's your advice? This transformer is next to (outward) the factory's perimeter wall.
- equipment test set-up is such that a ground loop is established between equipment and instrument. I think we should feed both of them from the same receptacle to eliminate the potential difference and, of course, twist their cords to minimize induced currents in the ground loop. What's your advice?
- there is a microwave tower behind the building acting as an air-terminal against lightning in the surrounding area. Regardless of the interconnection of its earth termination (an earth well) to the factory's GES, is there any other requirements, say in terms of nature of this interconnection or anything else?
Thanks a lot in advance ?
 
Sorry for my delay in responding. Not sure I understand all your questions. so lets start with you answering one question:

Are you experiencing any problems?
 
Excuse me, I think that's my fault not to clear the problem from scratch. The fact is that this factory is in operation for some 6-7 years, but there's not been a true grounding system in place unless bonding building receptacles' EGCs to structural steelwork at one point. Of course, #4 isolated ground well had been previously put in on the opposite sides of the building (2/2), but no particular application was made of them and they were actually useless. Also, there was no main bonding jumper in place in the service panel.
The other day, a friend of mine, who works in this factory, asked me to devise a grounding system for this factory. That was interesting to me why after this long time!?
Anyway, I noticed that the most dominant problem was intermittent shocks due to potential difference between adjacent/interconnected equipment, especially between equipment and interconnected instrument. Simply, the equipment bonding wasn't observed or one item was without ground connection. In one case, I measured more than 150V difference between osciloscope ground and equipment enclosure under test.
Now, I'm going to come up with the problem basically; not only resolving the abovementioned issue but also avoiding future likely problems that somehow may be related to grounding. So, for this factory with all the details I've described earlier, I'm going to devise a grounding system. In my previous messages, I described my proposed scheme. Now, I'll be grateful to know your or any of the other kind friends' viewpoints in this respect.
the best,
 
Jpower, my advice is fix the problem rather than re-invent the wheel. If the grounding and bonding were done per code, there is no way you would see 150 volt potential differences on a grounding system. One or more grounding systems is not properly bonded. Find it and the voltage difference goes away.
 
One more comment.

Sounds to me like one or more of your Ground Electrode Systems is not bonded together to form a common Ground Electrode System (GES). Some manufactures are real bone heads and make people install a dedicated GES isolated from the power GES. This is plain ignorance, stupidity, dangerous, and quite likely to render the equipment inoperable.

Make sure all GES are properly bonded together including the microwave tower and all electrode used in the facility. No if's, and's, or butt's.

Once you do that all voltage differences go away for good.
 
Thanks Dereck,
But, after all, is it reasonable to connect all the EGCs (which are also the equipment signal ground) through the building to the service panel ground bar, regardless of their length?
 
jpower said:
Thanks Dereck,
But, after all, is it reasonable to connect all the EGCs (which are also the equipment signal ground) through the building to the service panel ground bar, regardless of their length?
Click to expand...

Not only is it responsible, but the law (aka NEC) Failure to have all grounds bonded together properly creates an extreme danger to personnel and property, and in some cases leaves the equipment inoperable.

If someone were hurt or property damaged as a result, the persons responsible for intentionally or knowingly knew there were code violations would be liable for all damage and casualties.
 
Status
Not open for further replies.
Top