Electronic/Instrument Grounding Methods 3

[nightfox1925]

What type of grounding configuration is acceptably employed in electronic or instrument grounding? Is this necessarily be isolated from the Elctrical Safety Grounding? Are there useful links and reference for me to expand more on this subject?

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata

 

[waross]

Control room equipment grounding is per NEC in the US. The ampacity of the grounding conductor is dependant on the ampacity of the feeders.
The field runs are grounded or floating depending on the individual instruments but are often floating. The shields or sheaths of the field signals are grounded at the control room end, or the source end and then insulated from ground.
There should never be more than one ground point on a instrument loop and never more than one ground connection on the shield of an instrumentation loop cable.
yours

 

[dpc]

As waross mentions, in the US you must comply with NEC grounding requirements for safety reasons. Instrument vendors will likely have their own recommendations that should be followed, provided that these recommendations do not violate the NEC requirements.

In years past, some major control system suppliers advocated something called a "quiet" ground, or "isolated" ground. This involved a magical triangle of ground rods that were used only for instrumentation grounding and not tied to the normal facility grounding system. Aside from questions regarding the merit of this approach, it is a blatant violation of the NEC, which requires that all grounding system be bonded together.

You can run "isolated" insulated grounding conductors and grounding systems for instrumentation, but ultimately this must be bonded to the power system grounding at a minimum of one point.

 

[waross]

Hello dpc
I agree with you.
I always shake my head when I see those "special" grounds, like in the special olympics. When you ponder the grounding of the main system that may have hundreds of feet of buried cable ringing the facility, multiple ufer grounds, and the capacity to safely ground thousands of amps of ground current those "Dedicated instrument grounds" look pretty Mickey Mouse. I never worried about them though because when we installed the 4/0 jumper from the instrument ground bus to the main ground bus as per code, the instrument system had a good ground.
It is regretable, but over the years, I have seen electronics engineers, comunications engineers and computer engineers show an unprofessional contempt for power engineering. In regards to grounding in particular, they don't understand grounding systems and the normally expected characteristics of a ground system under fault conditions. And they don't ask an expert because they don't know they don't know. There seems to be an assumption that if you know electronics and data systems that 60 hz power are so simple that you can just put a ground symbol on a drawing and magically you have connected your circuit to an infinite common point. (end of rant)
I am glad to see that 02101972 is asking. This is an intelligent and professional approach.
Respectfully

 

[dpc]

waross,

Yep, I feel your pain. I've seen instrument guys want to put a piece of plastic in to insulate "their" conduit system from the power system's. Did they think that a lightning pulse that just traveled 30,000 feet *through the air* is going to care about the little piece of plastic??

In fairness, most of major control system suppliers now at least acknowledge the need to meet NEC and bond all grounds. But I'm sure there are still a lot of facilities with "quiet" grounds. Just trouble waiting to happen.

 

[Senselessticker]

I've seen instrument ground systems which are bonded to the the main ground system in one location. Then the the bare copper instr. ground embedded in conductive concrete trench, while the main ground grid is kept seperated by a couple of feet. From there....all instr. ground extends out like branches, not loops, to various instr. ground bars (via insulated copper).

Its my understand that for such a system...nothing changes concerning equipment grounding, but a signal ground has been established.

If your talking to the instr./controls type folks about such grounding....good luck! Be prepared to confuse each other!

 

[ScottyUK]

Our site has some of the magic 'quiet grounds' - exactly as described above: a triangle of driven rods bonded together. The fact that the triangle is surrounded by the main earthing grid for site, in soil that has a very low [Ω]-cm value and with a water table approx. 30" below grade is obviously lost on the genius who designed it.

I can wholly accept the need for the instrument ground to be separated from the power grounds right up to the connection into the earth grid to avoid stray currents from the power system, or fault currents, causing interaction with the control system through a shared conductor to the earth grid. That is common sense, and shouldn't be much of a surprise to analogue electronic designers who would recognise the star point earthing strategy from their own discipline.

The problems arise when controls guys who have neither a sound electronics background nor a sound electrical background start expounding their witch-doctor theories on noise supression to others who have inadequate knowledge to know better. From this is born unsafe practice and urban myth, which eventually makes it into the specifications.

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[nightfox1925]

Thanks folks, I am presently checking things out with IEEE Std. 1100-1999, IEEE Emerald Book: IEEE Recommended Practice For Powering and Grounding Electronic Equipment.

Well, this is a 424 pages book mostly discussing grounding stuff for Electronic Equipment. I am in the process of looking into it and hope it won't confuse me further when dealing with Instrumentation or Communication grounding practices.

Your further views on the specific approach is well taken.

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata