Ground Cable Integrity Testing

[VLFit]

What is used for testing substation, and other, ground cables for their resistance/integrity? I don't mean the earth resistance grid testing but testing individual ground cables into the ground and up the structures? Is this test routinely performed? Thanks

 

[Skogsgurra]

A very similar question was asked only one or two days ago: thread238-384792 Go there (click link) for some answers.


Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[VLFit]

Thanks Skogsgurra. However, that thread was for ground grid testing, not testing the cables that ground the structures to the earth, both into the ground and from the ground level up into the structure overhead. I am not close enough to it to have all the nomenclature correct but the testing I ma asking about is performed with AC voltage not DC and many amps are pumped into the cable, perhaps 50 or more, with current measurements taken and resistance and power factor measurements calculated. A 5 or 10 kVA test set is used. It is not a test that can be performed by some handheld instrument that puts out a DC voltage and measures high resistances. Megger and others used to have them but maybe they are old school and a different method is now used.

 

[Skogsgurra]

The text is just about the inadequacy of the hand-held instruments. And the problem with high-current injection. It describes a method that, although not hand-held, is portable and doesn't need high-powered outlets. It also avoids destructive, dangerous and lethal voltages in the plant and, finally, avoids the false measurements that unreflected use of high test frequencies cause.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[marks1080]

When you consider the size and amapcity of the conductors used in ground grids, or bonding structures in the switchyard I don't see how running 50A through the cable is any better than a simple resistance check with an ohm meter.

Are you simply wanting to confirm the ampacity of the guage of cable you have? I've never heard of this being done before. If you buy a 4/0 stranded conductor, trust the datasheet. If it doesn't look physically damaged, then it isn't. Test the bonding points (at the structures and the ground grid) with a resistance check. If the bonding points were for circuits that normally carried current you could use thermal imageing at the bonding points to see that they are good (not too hot). Ground grids shouldn't normally carry any significant amount of current. When a fault does happen and they have to marshal many kA's of current we rely on the protections to stop the current flow. If the protection fails to do this than you have much more than just a damaged ground grid to worry about. Testing conductors/bond points with 50A does not come close to simulating a fault condition. Also, make sure the proper connecting hardware is being used for the size and type of ground conductor you choose.

 

[Skogsgurra]

Thanks a lot for the badly needed lesson!

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[VLFit]

There are several methods that can be used, some better than others depending on what is being tested and what results one is after. I was asking mostly about the higher current testing to verify the ground cable integrity that may be compromised by corrosion at some locations. The IEEE Standard 81-2012 is the defining document.

 

[Skogsgurra]

Right. But that practice is sometimes destructive and quite often not allowed. What I point out is that there are several clueless measurements being made without thinking about resolution, accuracy or time constant of the meshes.

We prefer to do a measurement that takes all these circumstances into consideration and produce real results, like 18.7 milliohms instead of a fictive and false value like 567 milliohms or thereabout.

Sorry that you haven't heard of the method before. Now you have.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Jebb]

http://www.hubbellpowersystems.com/...pdfs4lib/operating-instructions/p403-3222.pdf

 

[7anoter4]

I agree with Skogsgurra, of course, about the measurement accuracy. However I agree with marks1080 also about ineffectiveness of cable resistance measurement in order to test the cable temperature.
First of all it is not important how precisely will be it is actually an average and it is not any indication of the local actual temperature.
In my experience with tens of thousand power cables in duct banks I never came across with cable overheating .Of course, in a power station it is permissible to oversize a bit the conductor cross section.
For instance the DIN/VDE 276 standard takes into consideration a 0.7 load factor usually, calculated from the 24-hour load cycle, also.
Even if it is a constant load considered , in my opinion ,the earth temperature changes in a day time and the transient constant of cable temperature variation in a duct bank is high [high inertia with respect the changes].
However the cable laid directly in the ground was overheated and damaged often. I think here is the problem of earth uncontrolled drying around the cable.

 

[waross]

I thought that I understood this thread until temperature and load factor were brought in.
The OP asked:
"What is used for testing substation, and other, ground cables for their resistance/integrity?"
In a ground grid the resistance of a 2/0, 4/0 or larger cable which has been properly sized for the calculated fault current is immaterial compared to an anomalously high resistance caused by a corroded joint or cable.

We prefer to do a measurement that takes all these circumstances into consideration and produce real results, like 18.7 milliohms instead of a fictive and false value like 567 milliohms or thereabout.

A high resistance point in a ground grid has at least two implications:
1. It may not limit step and touch to the calculated safe levels.
2. It may fuse open under fault conditions with the result that step and touch voltages are even higher.
VLFit; I suggest that for the best answer to your question, you read over thread
thread238-384792: Grounding Grid
and follow the link to Gunnars presentation at:

http://www.gke.org/presentationer/files/Low ohms...

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[7anoter4]

I am sorry I did not understand the question. I meant all the time insulated power cables in underground laying.
There are two-basically-ways to build a grounding grid: bolted connection or cadweld[or brazing] connection.
The advantage of the cadweld mode is the elevated permissible temperature[for copper conductor up to 1060 oC, instead of 250 for bolted connection].The vertical electrode connection does not need a handhole [visitable].
The bolt connection grid is provided with handholes in order to connect the electrode.If you will open the connection between two adjacent handholes you may test the ground cable connected with.
However,in my opinion, it is not required if the touch and step voltage will be in the permissible limits[see IEEE 80/2000].

 

[Supertester]

An often overlooked maintenance procedure is a simple visual inspection of these ground cable tails assemblies. The mechanical connectors should be electrically bypassed and then disassembled, cleaned and brushed, and reassembled with new hardware, or at least bolts, and actually torqued to a prescribed value. Similar to repairing Hotspots in overhead, just tightening old hardware and connections is not enough, clean it - update it and then you have assurance of low Z bonding under fault conditions (at the connection point anyway).

I did this at a remote cable terminal station and found the grounding cable hardware loose enough to move when I gave it a boot with my foot. Lots of loose ones. Turns out the protection hardware was failing for years during transients on the DC transmission system. Construction went back to the site and after repairs (probably just tightening) no more P&C worries.