Grounding and bonding NEC 250 questions. 2

[cmelguet]

Hi,

I am from Chile where we have our own standard for low voltage installations, but since they are not updated since 2003,we normally also follow the NEC as a good practice guide. Article 250 of the NEC states:

250.24(C) For 1000 V or less AC Systems, neutral conductor shall be routed with the ungrounded conductor to each service disconnecting mean and shall be connected to each disconnecting mean grounded conductor terminal or bus. Whether or not it is used to supply a load.

In the project I am working we do not route the neutral of a transformer to an MCC since we only connect 3 phase loads on the MCC. If we need a neutral we use an isolation transformer between the MCC and the single phase load.

For the return of a fault current to the source we use the ground grid wich is common for the whole plant.

Please comment if attached drawing will be ok under NEC standards, do you always have neutrals on the MCC`s?. Isnt good enough for the earth fault current to return using the ground grid ?

Regards
cmelguet

 

[cmelguet]

Sorry forgot to attached the drawing.

 

[cmelguet]

i think now is really attached, sorry for the mess. Regards,

 

[Kiribanda]

cmelqut,
Per your sketch I donot think you are providing an effective ground path (NEC 250.2) for the ground fault current back to the source which is the "system ground".It looks like you have mixed the "safety ground" with the "system ground" serving two different purposes.

 

[cmelguet]

Hi Kiribanda,

From my point of view, if there is a ground fault at the motor, the current will flow through the motor housing to the ground grid and then through the ground grid to the neutral point of the transformer. Isnt an effective ground fault path the 4/0 AWG conductor used as ground grid conductor. Thanks again,

Regards.

 

[jghrist]

The inductance of a circuit is proportional to the log of the distance between current paths. If there is a grouding conductor run with the phase conductors, as required by NEC, then the impedance of the ground path is low. If the ground current is forced to flow through the ground grid which is distant from the phase conductors, then the impedance will be large (and difficult to calculate). There may not be enough ground fault current to trip the overcurrent protection.

 

[7anoter4]

In my opinion you cannot use the Grounding Grid as an
Effective Ground-Fault Current Path [see 250.2] since the short-circuit impedance cannot be controlled [the reactance between the faulted phase conductor and the Grounding Grid is too large].See Art.250.118 for permitted means.

 

[Kiribanda]

cmelguest,
While I was planning to write an answer to your question, jgrist & 7anoter4 had done it.Please see their replies.
Thanks gentlemen.

 

[acog]

It is interesting to note that in many distribution and medium voltage installations (even in cases where only overcurrent or distance schemes are employed) a neutral conductor is not always run and this is accepted practice.

Why should low voltage installations be treated differently?

It is possible to measure the loop impedance at power frequency of the zero sequence return path cmelguet has sketched.

If this impedance is low enough such that the current drawn in a fault scenario is large enough to operate protection for all credible faults, then can anyone suggest a technical reason why this arrangement is not acceptable?

 

[cmelguet]

Thanks for all your answers. it is clear to me now, i hope i can make the consultant company understand. Regards

 

[VTer]

You do not have to run the neutral but there has to be some sort of equipment grounding conductor. I believe the NEC mandates that this conductor has to be in the same raceway as the circuit conductors. The equipment grounding conductor size is determined by the OCPD upstream of the circuit in question. The drawing you provided would not be NEC compliant as others noted.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic — and this we know it is, for certain — then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". – Nikola Tesla

 

[rcw retired EE]

ACOG - You asked "Why should low voltage installations be treated differently?" -

Let's look at typical MV & LV systems. Assume we have 1.0 ohms to remote earth through our ground grid, a very typical number for large grids. Let's neglect the inductive impedance for now and just look at a medium voltage and low voltage faults.

Assume a 13.8kV MV system with 1000 MVA of short circuit capability. The bolted fault level is around 40,000A. 13.8 kV/(1.732 x 40,000A) = 0.2 ohms total system impedance.

Now add in the 1.0 ohm return path. Fault level = 13.8 kV/(1.732 x 1.2 ohms) = 6,640 Amps. That is still enough fault current to trip most MV protection systems.

A typical 480/277 V system fault level is 50 kA. 277V/50,000A= 0.00554 equivalent system impedance. Add in the 1.0 ohm ground fault return path and the fault drops to 277V/1.00554 ohms = 275 Amps. If that fault is on a 200 A breaker it will take a few hours to trip. A 400A or larger breaker will never trip until the arcing melts it into a phase-to-phase fault.

At 208/120V the effect of the return path impedance is even more pronounced.

Maybe I'm overstating the fault return path impedance. What if it is only 0.2 ohms? The 480V ground fault current goes up to 1350 Amps. That's enough to trip smaller breakers but it will take the 400 Amp many seconds to trip.

For safety reasons, the NEC, which is aimed at low voltage systems, tries to make the fault current return path impedance as low as possible to maximize the fault current and minimize the overcurrent trip time. A proven method of doing that is to include a fault return path through a ground (earthing) conductor run with the power conductors or through the metallic raceway encircling the conductors.

You can't depend on the earth to provide a low impedance return. At higher voltages, the added impedance has a smaller effect.