Instrument Grounding 3

[ksps100]

Dear ALL,
Kindly advise the import of the following:

1)Instrument Reference Ground
2)Instrument Safety Ground

We have Electrical Main Grounding grid of 70mm2 Copper conductor,but the entire Instrumentation system(DCS,etc) is connected to a Ground grid developed using separate tripods and separate Ground grid.

What are the applicable codes and standards for the above development of separate ground for Instrumentation system.

I noted that finally all things are connected to the Electrical Ground grid.
Thanx

 

[mohanlal]

Experience shows that separate grounding system proved better. The instument system is not subjected to any surge voltage when there is lightning.

still grounding resistance could be kept minimum and separate. Also ensure that the voltage between neutral and ground is bare minimum if neutral is grounded.

subra

 

[Skogsgurra]

What you have is a kind of "tree grounding" or "star grounding". The idea is that no heavy currents shall flow in the ground system used for instrumentation. No heavy currents -> quiet ground potential -> minimum interference coupled back to instrumentation through shielding and filters.

But it needs to be connected to the main ground terminal (usually the heavy copper bar close to the incoming transformer(s)). Do not connect anywhere else and make sure that no unintentional connections between the two systems exists or is being created. It will ruin the "instrument ground".

A fairly easy way of checking the quality of the instrument ground is to drive a rod somewhere in an open field (if possible, you may need to be 50 or 100 yards away from the building) and measure HF and LF voltage between instrument ground and the rod. There should be very little voltage. You should expect less than 100 mV RMS combined LF and HF.


Gunnar Englund

http://www.gke.org

 

[ScottyUK]

Experience shows that separate grounding system proved better. The instument system is not subjected to any surge voltage when there is lightning.

Google on 'ground potential rise'. Any major current flow to earth will cause voltage rise local to the point at which the current enters earth. A lightning strike near the isolated triad ground will cause a voltage increase at the instrument ground conductor, and will almost certainly introduce an additional hazard from a touch voltage that would not occur if the instrument ground was tied to power ground via a dedicated connection.

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

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

 

[ksps100]

Dear ALL,

I checked out NEC for any such specs which specifies any minimum clearances required between Instrument Ground and Electrical Ground but could not find much.
NEC clearly specifies that ALL grounding should be bonded together (whether Electrical or Instrument).
But my Control Systems Engineer insists that the Instrumentation ground be segregated and clearance of 10 Meters(!!!) be kept between the Instrument Ground and the Electrical Ground.
Anyway-I am trying the IEEE Emerald book-hope something is there.

Thanks all of you folks for the replies-I will try convincing my Control Systems Engineer

 

[waross]

Hello ksps100
With all respect, some of us have been watching this scenario for decades. It is common for electronic people to not understand grounding systems. An instrument circuit may be connected to ground rods, but if it is not properly connected to the main system ground, lethal voltages may be developed between the instrument inputs and the structure supporting the instruments.
By the same token, the same voltages that may be lethal may be developed between the instrument grounds and the grounded circuit conductor.
The reason that neither the NEC nor the CEC give a minimum spacing for "Dedicated instrument grounds" is that it doesn't matter.
The codes state that all the grounds must be connected together.
ALL THE GROUNDS.
THAT'S THE LAW.
The inspector will not sign off the permit until the electrical contractor connects the two systems together with the connection mentioned by skogskura.
I have always understood that professionalism meant consulting with experts in any field in which you are not trained. For more than 40 years I have been astonished and embarrassed by the lack of understanding that many communications experts have of grounding principles, and even more by the indications that they don't know that they don't know.
I would also point out that in the event that the connection between the dedicated ground and the system ground is less than the code minimum ampacity for the system ground, (When the engineer says "It's only an instrument ground, it doesn't need to be that big.") a serious ground fault has the potential to destroy the undersized connection jumper and allow dangerous voltages to develop.
Any station engineer calculating voltage gradients can show you locations for a "Dedicated" ground rod cluster that will be guaranteed to result in dangerous voltages between ground systems in the event of a serious ground fault.
We have all seen many instrument ground systems, and the reason that there is not problems later is that the contractor has to make it safe before he leaves the site or the inspector will not allow the utility current to be connected.
I guess it's partly the fault of the power group, because we know that it's not worth talking to the instrument people when it's so quick any easy to inatall the appropriate jumper to make the system safe. I am sure that most comunication engineers never learn that the reason that they have excellent grounds is because the instrument ground has been connected to the much more extensive and effective main ground after they left the site.
If you consult the NEC for the minimum ampacity grounding conductor for the main grounding system and use that ampacity for the connection between the main grounding systen and the instrument ground bus, you will be safe and almost legal and your engineer will be happy.
end of rant
respectfully

 

[waross]

Hello ksps100
I acknowledge your professionalism in seeking advice when you recognise that an assignment may be outside your area of expertise.
Earlier this year there was a thread in this forum concerning safety grids or mats for the protection of operators operating high voltage disconnect switches. The subject of equipotential grounds was discussed. The same issues of equipotential zones for safety that apply to switch operation also apply to instrument safety. A separate ground bus to establish an equipotential for instrument field circuits is good engineering. Choosing the point on the main bus to connect the instrument ground bus with consideration for the flow of fault currents may reduce the potential difference between instrument field circuits and instrument grounded circuit conductors to negligible levels.
It is not always safe or desirable to connect grounds of any kind to the ground mass of planet earth. It is safer and more effective to design ground systems with equipotential zones of protection.
Do some googling on "Touch potentials" and "Step potentials". These are potentially lethal voltages that develop during fault events between a grounded structure and a point on the ground about 1 meter away in the case of touch potentials and between two points on the earth about 1 meter apart in the case of step potentials. The potential that can be developed between points 10 meters apart can be much higher and more dangerous to both men and instruments.
Conductors that are assumed to have zero impedance often have significant impedance and unexpectedly high voltage drops with the passage of fault currents of thousands of amps. The high rise times of lighning induced currents further increase the effective impedance and the resulting voltages under fault conditions. Instrument systems are affected by events occuring on adjacent power circuits and the resulting high voltages on improperly applied instrument grounds can be a hazard to workers and may damage equipment.
The solution.
1> Use a dedicated instrument ground bus.
2> Connect the instrument ground bus to the main system ground bus with a substantial jumper.
In normal service you have now established an equipotential zone for your instruments.
In normal operation there will be no current flowing in the ground system and there will be no voltages on the ground system. In the event of a fault on the power system the instrument grounds will track with the power ground voltage and maintain the equipotential protection. It can be seen by the explanations in the thread on equipotential grids that the addition of the instrument safety ground rods will create an additional path to ground for fault currents and render the equipotential protection less effective.
The addition of ground rods that can allow current to flow in the instrument ground system under fault conditions, will cause voltage drops in the grounding system which will result in voltage differences between the instrument circuits and the instrument supports and power circuits. The effectiveness of your equipotential protection will be reduced.
Control system designers for substations must consider the high voltages (hundreds and possibly thousands of volts) that will be developed under fault conditions between different locations in the plant even though the locations are tied together by a ground grid that is, I assure you, much more massive than anything imagined for instrument grounding. These professionals may be challenged by the high voltages that grounded instrument transformer circuits may develop between the ground potential at a field mounted current transformer and the circuit ground point in the control room.
Many years ago I was appointed safety person on a project installing capacitors in series with a 500,000 volt transmission line. The utility provided me with extensive information and case studies on safety grounding issues. Some of the case studies detailed worker injuries and deaths resulting from a failure to properly interconnect safety grounds. A worker was rendered unconcious by the shock he received when his body completed the connection between two different ground systems that were at different potentials.
Of course these voltages and currents are not present in instrument grounding situations. Think again. One nearby lightning strike will supply adequate voltage and current to destroy improperly grounded instruments and pose a hazard to nearby workers.
respectfully

 

[dpc]

Your control engineer is wrong. In the US, the NEC requires the instrument grounding system to be bonded to the power system ground. There are no exceptions that I am aware of. It is serious safety hazard, as others have mentioned.

The "isolated" triad ground rods was one of those magic solutions that was promoted by control system suppliers twenty-five years ago. They disavowed it about 20 years ago.

 

[ksps100]

Dear Waross,dpc,

Thanx for the detailed replies.

I checked out the designs in adjacent Plants.
The design is simple and I understand it is quite effective until now.

Basically a Tripod consisting of 3 Mtr Copper rods are buried below ground at a depth of 1.5 Mtrs.
The tripod is then connected to Copper Flat bar and from this Flat bar,long copper cables are taken to cover the entire Plant area and a interconnecting Grid of Copper cables is formed.
All the Instruments are connected to this Grid of Copper cables using Exothermic or Compression connections.

Similarly the Main Electrical Ground Grid is designed to have a overall resistance less than 1 Ohms.
This Electrical Ground grid is then connected to one of the Instrument Tripod Ground Rod by using a 120mm2 Copper cable(same size as the Main Electrical Ground Grid.

Thats the skeleton of the design.

Thanx everybody for the informative responses.

 

[waross]

Hello ksps100
That sounds like the scheme that skogsgurra suggested.
That will work and probably meets code.
You may find the following excercise interesting and it will help your understanding of ground protection.
If you care to sketch the main ground grid and the instrument ground grid and do a quick calculation the results may interest you. The sketches don't have to be accurate to illustrate the point.
You will find that in the event of a fault there will be a voltage difference between grounds that is inversely proportional to the impedance of the jumper connecting the two ground systems and directly proportional to the current flowing in the jumper.
You will see that if you eliminate the tripod ground rods, the current in the jumper will be zero and the two systems will be at the same ground potential.
Do your sketch with the one tie cable between the main bus and the instrument ground bus.
Now assign resistances to the grounding cables and to the ground resistance of the ground rods. Accuracy is not important to illustrate the point.
Now consider that there is a ground fault in the plant and ground currents are flowing. Some of this current will flow in your instrument grounding system in the inverse ratio of the impedances of the two systems.
When current flows in an impedance there is a voltage drop. You will have a voltage drop in the cable that joins the two systems dependant on the magnitude of the current flowing. Now suppose that you improve the ground contact of your tripod ground by some method. As a result, more current will flow in the connecting jumper and the voltage drop will be greater. You have the paradox that if the tripod ground contact is improved, the ability to maintain equipotential protection is degraded.
As I said, the exact values are not important because the current levels are variable.
Bear in mind that the ground currents may be thousands or tens of thousands of amps.
You mentioned a resistance to ground of 1 ohm. That would effectively limit the current to ground of a 277/480 volt system to 277 amps (of the thousands of amps available fault current) and the voltage on your ground bus will be 277 volts above earth ground.
In practice, the voltage difference between the systems will be very low. If the connecting cable is sized as per the main grounding cables the voltage drop will be very small but it can be calculated.
If you don't install the "Magic tripod" the current through the connecting jumper will be zero and the voltage difference between the main ground bus and the instrument bus will be zero. No current equals no voltage.
Of course, if the current is the result of a lightning strike on the primary neutral, that is going to ground through the plant system ground, the voltages and currents will be much higher including the voltage rise on the instrument system.
One reason that the tripod systems work so well is that they are relatively ineffective. By far the greatest proportion of the ground currents flow through the system ground that may quite easily have an impedance value that is one or two orders of magnitude less than the impedance of the "Magic tripod"
respectfully