High current on earth conductor 7

[carlosmelim]

Hi everybody

I am facing a problem in a supermarket electrical installation where I´ve measured 30A in the earth conductor.

I have two diferent earths on the compound ( PE and neutral )since we use TT system in Portugal.

The supermarket has a step down transformer from 6600V down to 400 V. The primary ( 6600 V ) is delta connected and the secondary ( 400 V ) is star connected with the center directly connected to the service earth.

What puzzles me is that I have measured 30 A on the protection earth conductor that goes directly to the protection earth and I have no readings on the conductor that comes from the service earth directly to the center of the star on the 400 V side ( neutral ).

Where is that current closing the path? I don´t think it´s returning through the 6600 V side since it´s delta connected.

By the way I have used a clamp-on current measurer and I don´t find any reasons to doubt it so far.

Has anyone faced a similar situation?

Best regards

Carlos Melim

Portugal

 

[busbar]

Sounds like an interesting discovery, but unless it’s causing other problems, the investment in time tracking it may not be practical.

A few suggestions:

Assure no inadvertent neutral-to-ground/earth connections downstream of the service entrance point.

Isolate various loads while monitoring excess current.

Use an oscilloscope to view if current is sinusoidal or distorted, or a meter that will allow crest-factor calculation.

High neutral-to-ground voltage at a load may point to a problem.

Line-to-ground-connected “filter” capacitors may be contributing to the undesired current.

 

[ausphil]

Don't forget that your earth cable might be carrying current not from your supply transformer, but a customer next door or down the street with a bad neutral connection. Their neutral current will be going into the earth, and may find multiple paths back to their transformer secondary, including some current through your earth.
You may have water pipe connections or other paths that will allow a current to flow in and out of the conductor.
Our company quite often finds that earth currents are not produced by the direct customer who is experiencing the problem, but someone else in the street or around the corner.
To check that the current is yours or not, measure the currents in the phases, neutral and earth conductor. Add the phase currents vectorially and this should equal the magnitude of the neutral current. If this is the case, I hazard a guess that the problem is not generated byb your establishment. If the vectorial sum of phase currents does not equal the neutral magnitude, then start looking internally. If it is deemed to be an external influence, start looking for connections to your earth that outside current may get into the conductor - uninsulated earth conductor at some point, connection to water pipes etc.

Earth current problems can take hours or in some cases days to solve. Sit down and think how to determine what current am I measuring, is it from my plant, where is it coming from, where is it going to, how do I eliminate it? It is a great example of the use of Ohms Law. you will probably need to take readings at a number of locations around the site to get a feel for what is happening, where the current is flowing to/from. Look for earth rod connection outside the building, measure water meter currents etc.

ausphil

 

[carlosmelim]

Thanks a lot for your cooperation.
Let me add some information:

I have used an oscilloscope and the earth current is very distorted and so is the neutral. The phases current is an almost perfect sinusoid. What could this mean?

When I cut all the power inside the installation the earth current disappears. So I assume the current is not an external influence, right?

Could it be a capacitive effect on the cables?

Carlos Melim

 

[lyledunn]

Not sure I agree with Busbar in terms of time wasting although I do agree with his variuos points. I have measured significant current in earth conductors which on investigation turned out to be a combination of diverted neutral current and earth connected filters. Poor insulation values between neutral and earth in the installation are a possibility although will be difficult to trace on a fully operating installation without co-operation of owner / manager. Nonetheless, your installation maybe unsafe. If in the UK it would be subject to the Electricity at Work Regulations and as such it would be incumbent on the owner to ensure that his installation was safe (probably done through you). This means investigation and I think the advice given to you is sound by the other two respondents.
I am often charged with monitoring diverted neutral current for petrol filling stations where such currents can lead to potentially explosive consequences if they flow through metalwork in the hazardous area. Also, for the unsuspecting maintenance personnel significant shock risks may be present even with circuits isolated.
As a minimum,notices providing information on these high protective currents should be placed at the relevant distribution points (required by BS7671 2001 in the UK).
If you do decide to trace, I would appreciate your analysis of outcome.
Regards,

Lyledunn

 

[lyledunn]

Are there a lot of computers or electronic tills? It is suprising the cumulative effect of filter circuits. Such networks will contribute to the distortion you mention.
Regards,

Lyledunn

 

[busbar]

Lyle, I appreciate your professional disagreement. No one has a guaranteed solution for problems posted here. My point is this—30 amperes compared to what? carlos’ reading seems likely to be a bit high and is certainly interesting, but what other problems have been manifested; particularly those thermal in nature?

In this region a 1200-ampere, 480V switchboard is not uncommon for a grocery store, but that is my view and may only remotely compare to the norm in Portugal.

 

[carlosmelim]

Dear fellows,

Following your various tips I think we have discovered the reason for the high current in the earth conductor.

The step down transformer feeds the compound with two 240mm2 copper conductors in parallel for each phase; one 240mm 2 for the neutral and one 240mm2 for the earth.
Our electricians laid the cables side-by-side in a cable way without separating the earth conductor. We could not see this because the cable way is inside a sealed duct for fire prevention.

Today we broke the sealing off and put an extra earthing cable outside the cable way, connected it at both ends and disconnected the old cable. The earth current disappeared.
My conclusion was that the phase cables that carry around 400 A were inducing the 30 A current in the earth cable producing a distorted current waveform.

We can even see in that waveform that when the neutral current rises, the earth current diminishes because the induction effect is opposite to the one produced by the phases.

I really want your opinions on this subject but let me tell you all that this was my first post in E-tips and I am amazed with everybody interest in helping me out.

I am an electrical engineer in Portugal and I am looking forward to hearing from you all.

Best regards

Carlos Melim


Now that we have reached this conclusion we can confirm tha

 

[lyledunn]

Busbar, I apologise, I re-read my post and it is obvious that I did not make the object of my small disagreement with you as clear as it should have been. It was simply that the question of time spent on investigation would be necessary if only to determine that this was not an installation fault. 30 amps flowing in the earth conductor could have been the result of phase earth or neutral earth faults within the installation. In the UK it would have been imperative to rule these out in order to comply with certain statutory provisions. It was the shock risks I was referring to rather than any thermal problems. Just as a matter of interest our national rules for electrical installations (BS7671 2001) demand specific requirements when the protective conductor current is in excess of 10ma.
It looks like Carlos is sorted, although I am trying to get my head round what he is saying.I would have thought that perhaps some N/E problem existed on the supply from tx to service intake rather than suspect the effects of induction.

Regards,

Lyledunn

 

[Ares]

The solution you found is not a good practice.
The equipment grounding conductors in a feeder must always be routed within the same conduit or raceway containing that circuit’s associated power circuit conductors.
the grounding conductor is much more effective that way.

 

[carlosmelim]

Ares,

I agree totally with you but the problem is that with conductors of 240 mm2 it´s usual practice to use single conductor cables instead of multi-conductor ones. Of course in this case, the long extension of the cables ( 150m ) is causing the current induction.

We will separate the ground conductor but it will follow the same path, just a few centimeters apart from the phase conductors.

Could you tell me more in detail why the ground conductor is more effective when it follows the same path?

 

[jghrist]

The inductance is higher for larger spacings.

 

[busbar]

lyle — In your Aug 19 posting, an apology is not necessary, for no offense was taken. I think we both agree that is important to take all incidents of likely hazardous potential difference seriously. Regional differences in how to consistently achieve that through regulations and practices will seldom be the same. That is one reason that eng-tips.com remains so interesting.

 

[Ares]

Carlosmelim,

I often see a current that flows through the grounding conductor in the conduit (usually harmonics + common mode noise) despite the presence of other multiple ground return paths of lower DC resistance. The reson is the magnetic field created by power supply conductors.
(Usually the reason is poorly designed electronics power supplies)
I read a paper (sorry do not remember the source) with empirically collected data showing that a ground installed in conduit is many (?~6) times more effective in short circuit situations than identical conductor installed outside the conduit.

The explanation is - Induction. Surge in power supply conductors will create magnetic field oriented in a way that will enhance the flow in the return path. (lower the
impedance)> Note that short cct is a transient event - loaded with harmonics so the impedance (not resistance) becames a dominating value.
The further away the grounding conductor from the power conductors the less of that benefit you get.

 

[lyledunn]

Ares,
Forgive me, I am trying to understand your exact reason for insisting that the earth is run with its associated phase conductors. Certainly that would perhaps apply to the neutral but I cant see the reason for the protective conductor. In fact the effect of induction on a heavy current earth fault on protective conductors taped to SWA multicores can often render the separate earth conductor superfluous! Regards,

Lyledunn

 

[dpc]

The proximity of the ground conductor to the phase conductor has a large effect on the zero sequence impedance due to the increased inductance, as noted above. This is especially true for circuits run in a steel conduit. A ground conductor run outside the steel conduit is nearly useless at carrying fault current. The steel conduit generally isn't much better after a few years when the resistance of the joints builds up. This is one reason the NEC requires ground conductors be run in each conduit for circuits with multiple conductors per phase.

Quite a bit of testing was done on 480v circuits back in the 1970s that confirms this. I can provide references when I'm back in the office if anyone is interested.

 

[resqcapt19]

dpc,
I'd be interested in the references. Thanks.
Don

 

[dpc]

Don,

Best reference I have is AIEE Paper 60-12, "Determination of Ground-Fault Current on Common A-C Grounded-Neutral Systems in Standard Steel or Aluminum Conduit", by J.A. Geinger, O.C. Davidson and R.W. Brendel. It was published in AIEE Transactions in May 1960!

It was this paper that I believe started the trend toward including separate ground wires in 480V circuit run in conduit. Previously, most people relied on the steel conduit as the grounding conductor. This paper showed the serious limitations of the conduit system (particularly steel) as a grounding conductor. Aluminum conduit is much better.

See also "Let's Be More Specific About Equipment Grounding" by R.H. Kaufmann, published in Proceedings of American Power Conference, 1962, (Vol XX1V).

Hope this is useful to someone.

 

[busbar]

Another AIEE reference is cited in Chapter 11 of IAEI Soares Book on Grounding, 5 ed. It is R.H.Kaufmann, Some Fundamentals of Equipment-Grounding Circuit Design, AIEE paper 54-244. 1954.