Earth Resistance Testing in a large industrial facility

[pittengineer]

I have been requested to perfrom an earth resistance test here at our industrial facility. This tested is mandated by MSHA requirements that govern our site.

Our distribution voltage is 4.16kV and our utilization voltage is 480V.

I have been researching this test and plan on performeing a 3-point fall of potential test. I plan on following the IEEE 81 standard as the guidelines for this test and the acceptance results will be a result of less than 5ohms as stated by NETA.

Our site is laid out as such that we have a utility substation located on-site at our plant, and then have several electrical rooms throughout the plant which have various unit subs, MCC's, swgr etc... Each of these electrical rooms has a ground ring around it with various parts of the ring being connected to the equipment inside each room. The various grids around each room are connected together through copper in ductbanks etc.. to create the ground grid for the plant. A couple of questions regarding this test.

1) Since all of the electrical room ground grids are tied together do I need to perform this test at each individual room or just perform in one location? How is this typically handled when multiple grids are tied together such as in a large facility?

2) Do I even need to pefrom this test at the electrical room ground grids, or can we just do one fall of potential test at the utility substation and check that everything is connected back to it? Would this sill ensure the LV side of the grid for each of the unit subs in the electrical rooms?

3) Once a ground reference is establised I plan on having a 2-point test performed between the ground reference and all the associated equipment such as transformers, mcc's, vfd's, etc... I am going to follow the NETA standards of less thatn .5ohms for this test. Other than this 2-point test, is there anything else that should be included with this testing procedure?

4) We perform ground continutiy checks on motors and equipment routinely. Is there a standard for guidance on methods for performeing ground continuity checks, and equipment bonding checks?

I was hoping that maybe someone here had experience with performing these test in a large enviornment such as mine.

 

[Zogzog]

To answer some of those I would need to see the site or at least your 1 lines and the MSHA spec you are following. But you have a sound plan and are doing everything right. You would save a lot of time and expense having a NETA testing company come out and do it for you, they do these everyday.

 

[zaza123]

In reply to your point No:1
We used to do fall in potential 3 point Test by removing the connections of Earth Electrodes & measuring earth resistance. After that again the connections were made & same test repeated. This used to be done for all the Earth Electrodes in the plant. For any electrode where the Resistance was very high used to be treated with salt water & charcoal.
But you are doing better Than the old ways !

 

[pittengineer]

Zogzog

I am indeed having a NETA certified testing company perform this testing. I am confident that they indeed know how to perform a 3-point fall of potential test, however there seems to be some confusion even with them as to may many points we need to test, and at what locatins since grid is tied together. Thats what spurred the questions above.

I am not able to post a one-line right now but I'll do the best I can to describe it. We have a utility transformer with a 4.16kV wye secondary which is solidly grounded. This 4.16kV is then distributed to a number of electrical rooms at wich point it energized a switchgear in each room before being fed to the high side of 4.16kVDelta to 480V wye solidy grounded secondary. These transformer then feed MCC's and other LV equipment. We also have 4.16kV motors which have starters located in the switchgear circuit in each room I mentioned.

As I mentioned each electrical room has has a ring around it, with all of the rooms tied together to create a grid. It would be very difficult to isolate each one of these rings due to the fact that there are may conductive paths between rooms such as conduit etc...

Myself and the NETA company are trying to determine how to approach this.

 

[dpc]

I've never found the fall-of-potential tests at large complex facilities to be very meaningful. There is no way to get to a "remote" point and there is so much metal in the ground that results can vary wildly as you move in different directions.

The clamp-on ground resistance tester at various locations may give you a better insight into what is actually going on with the grounding system.

Maybe it's just me....

"Theory is when you know all and nothing works. Practice is when all works and nobody knows why. In this case we have put together theory and practice: nothing works... and nobody knows why! (Albert Einstein)

 

[jghrist]

Ground potentials for a fault on the secondary side of the utility transformer do not depend on the ground resistance because you have metallic ground paths to the ground source (the utility transformer). You need to perform the test only to meet MSHA requirements, not for safety reasons. Perform whatever test will be accepted by the AHJ (Authority Having Jurisdiction). I see no reason to test at more than one location unless the AHJ requires it.

Because of the extent of the grounding system, you will probably have to use the Slope Method to get a resistance value along with the Fall of Potential method. You may not be able to get the current probe far enough away to be outside the zone of influence of the ground system.

 

[tem1234]

There's something that i don't really understand about this.

What's a good value for a ground grid resistance? In IEEE 80, there's example of ground grid with resistance of 2 ohm, 20 ohm and 40 ohm and all these results are good since they respect the step and touch voltage.

As i see it, the fall of potential method (or slope method for large system) is only meaningfull just after construction, before interconnecting with other grid, to validate that the calculation was good. For an existing facility, you often don't have access to the initial calculation result, so it's difficult to say if the resistance is good or not!

On an other subject, be carefull before disconnecting any ground. This should be made in shutdown and with precaution.

I think the continuity test is good. Some would say that you most inject at least 20-30 ampere because if the cable is damaged and only few strand remained, the result will still be good if you inject only 100 mA.

 

[dpmac]

According to Mine Safety and Health Administration Mandatory Standards #55, 56.57.12-28 (there may be newer standards available)you are required to maintain certain records concerning continuity and resistance tests of grounding systems immediately after installation. repair, modification and annually thereafter.

Several MSHA Regulations have been written that directly deal with grounding of electrical equipment at plants and quarries. These regulations, as written, are of a general nature. Since proper grounding of electrical equipment is vital to maintain a safe work environment, regulations concerning grounding must be strictly adhered to.

I would imagine MSHA is well aware that the "earth" is being moved or re-arranged every day at a mine or quarry as well as equipment and its wiring may be relocated on a regular basis. Comparing results from initial plant construction may be a moot point. In any case they do require that these annual tests reports be kept on file.

 

[pittengineer]

dpmac

I have seen the same MSHA standards that you are referring to, and although they do state that these systems need to be tested anually, they are very vauge in what they are looking for.

Its hard to decipher weather or not they are looking for a test to be performed at every transformer or just one location. They are also not very clear in what values they are expecting so I am therefore following those set forth by NETA.

 

[jghrist]

I agree with TEM1234. The value of the resistance test is just after construction to validate ground resistance calculations. Disconnection of the grid from the source neutral/shield wire connection is necessary before making the fall-of-potential measurement. This connection is probably under the control of the utility and most likely you will not be able to disconnect it while the plant is energized.

The most important measurement is continuity. This ensures that you have a good metallic ground fault return path to the source.

 

[resqcapt19]

We did some work in a MSHA covered facility and I was talking with their electrician, and he told me they have to do a continuity check for every EGC once a year. He didn't say any thing about testing the grounding electrode system's resistance. However I did not specifically ask about that.

 

[eamonkerrigan]

Whenever I do a ground resistance test, I always do a soil resistivity measurement. It's probably the same meter, just use four pins instead of three.

This will help evaluate whether or not you were able to escape the zone of influence. That is if you really want to build an accurate computer model and replicate your test. Even if you don't ever want to do that, you might as well do your soil resistivity measurements.

This will also answer your question about what is a good measurement? 2 Ohms, 5 Ohms, 40 Ohms....It all depends on your soil.

 

[Thedroid]

Your original post is exactly how we do our ground testing at our facility. We start at the utilities substation and perform a fall of potential. I believe that you furthest stake is supposed to be some multiple of the total length of the grid. If your facility is built on one big interconnected grid, then you'll be walking for awhile.

We do perform a fall of potential at each corner of the plant, and we usally see the fall at around 80 ft. After that we start from the utility and work our way out. Every transformer, mcc, panel, motor, and even 120v outlet is referenced back, along with all machines and appliances.

We are under msha also, so this has to be done yearly, and they check them also. It's definitely the most miserable time of the year. It takes us about 2 weeks to complete with four men.

We inquired about an outside company doing the testing and they quoted over 20 grand, and said they could do it in 2 days with 2 men. I don't thinks considering we're familiar with the plant, and the test move quick, once you've established the references.