Rule of Thumb for Calculating Capacitive Coupling Charging Current

[rkallies]

I have a project that is in design phase that includes paralleled 12.47kV diesel generators feeding a medium voltage distribution system. I am planning to specify a high resistance ground (HRG) for the generators. I understand that the let through current of the HRG must be equal to or greater than the system charging current. How do I estimate the system charging current for the system? Are there any proven rules of thumb for calculating charging current for a 12.47kV distribution system?

The calculated demand load is approximately 15 MVA. The generators are feeding (16) 2500 kVA liquid insulated substation transformers.

 

[Live2learn]

This reference is 20 years older than me.
From the industrial power systems data book by GE Dec 21, 1964.

Table one on page 1 .32 has a sable
460 v charging current 3Ico - amps/1000kVA of system capacity to be around .1-2.0. that is without andy contribution from surge protective capacitors for rotating machines.

2400 v is the same.

the charging current must be tested to obtain a perfect number and distances of conductors and rotating machines.

Transformers are usually negligible becuase of the spacing between the transformer core and the windings, and the shielding effect of the xfmr winding layers adjacent to the core prevents the other winding layers from materially increasing the winding to ground.

overhead vs underground is a huge factor.

even if you do your best and get every component - it is still a close approximation at best.

 

[thisisnotatest]

i dealt with the same exact problem a little while ago and after scratching my head and doing countless calculations I went with about 1A/1000kVA

 

[rkallies]

I understand charging current is proportional to the distance between conductors thus the response about the significance of overhead vs underground. In this case all phase conductors are in conduits so I don't suspect it matters whether or not the conduits are underground.

The only publicized rule of thumb I could find was from I-Gard (google igard charging current). Here the rule of thumb is 1A/2000kVA for 600V systems and 1A/1000kVA for 4160V systems. Since this system is 12.47kV I'm thinking of estimating 1A/500kVA. With 15MVA of calculated demand load then there would be 30A of charging current. Does this approximation make sense?

 

[rockman7892]

I believe from informaiton I've seen in the past, the best way to find the charging current is to actually meausre it by putting an inttentional varibale shunt resistance between the system and ground and then vary this resistance towards zero to meausre the charging current.

 

[rkallies]

Thanks Rockman. Agreed. The issue is that the distribution system does not yet exist so measurement is not possible. The neutral resistance of the generator plant needs to be specified equal or higher than the anticipated charging current.

 

[rkallies]

FYI. I was able to locate a useful tool. An older white paper that I am not sure is available via the web. Anyway, I thought I would note the source in case others find it useful.

Westinghouse Industrial and Commercial Power System Application Series White Paper (PRSC-4A dated January 1978). Title = System Neutral Grounding and Ground Fault Protection.

 

[niallmolloy]

rkallies,

The charging current can be dependent on the voltage level and the conducting medium (OHL vs Cable), also take into account the whole network at this voltage level as anything else that is connected to the busbar will contribute to the charging current.

If you are installing a very high impedance ground i suggest you treat it as if you were isolating your star point and consider a neutral displacement relay to pick up a single phase to earth fault.

My attachment is a siemens document outlining same.




Niall Molloy
Electrical Engineer
ESB