Problem using Ecodial finding single phase short circuit currents.

[healyx]

Hi All,

I'm using Schneiders free Ecodial program to try and calculate some short circuit currents. I was doing a sanity check on the maximum single phase current (Ik1max) and am a long way off their calculated answer. Can anyone confirm I have this wrong?

The system is:

400V Phase/Phase TN-S or TN-C (doesn't matter)

500 MVA upstream network (Rq = 0.035, Xq = 0.351 - standard ratios).
160kVA transformer (4% impedance, Rt = 11.714, Xt = 42.516)
1x95mm2 Cu PVC cable per phase.
1x95mm2 Cu PVC neutral.
1x35mm2 Cu Earth.
So the system is just a transformer feeding a main switchboard via the 95mm2 cables - run length is 50m.

Ecodial apparently uses the UTE NFC-15-500 guidelines to calculate currents.

For maximum 3 phase short circuit current, according to NFC-15-500, the current (Ik3max) is:

Ik3max = (Cmax x m x Uo)/(sqrt(((Rq+Rt+Rph)^2)+((Xq+Xt+Xph)^2))

Where:

Cmax = 1.05
m = 1.05
Uo = 231 (Ph-N voltage)
Rph = Phase cable resistance = 18.51*50/95 = 9.74
Xph = Phase cable reactance = 0.09*50 = 4.5

So I get Ik3max = 4.895 kA - This is the same as Ecodial - so I must be correctly calculating network, transformer and phase impedances.

I can't work out how Ecodial calculated Ik1max, which should be for a Phase to Neutral fault with 20oC conductors.

For maximum 1 phase short circuit current, according to NFC-15-500, the current (Ik1max) is:

Ik1max = (Cmax x m x Uo)/(sqrt(((Rq+Rt+Rph+Rn)^2)+((Xq+Xt+Xph+Xn)^2))

Where all the parameters are the same with new ones:

Rn = Neutral cable resistance = 18.51*50/95 = 9.74 (same as Rph)
Xn = Phase cable reactance = 0.09*50 = 4.5 (same as Xph). It is basically the same as Ik3max, but with the additional impedance of the neutral return path.

I get an answer of Ik1max = 4.21 kA with Rtot = 31.23 and Xtot = 51.87.

Ecodial gives and answer of Ik1max = 3.57 kA with Rtot = 34.16 and Xtot = 62.50.

I just can't work out where they got the higher impedance from. I have found examples of how to calculate Ik1max on the web and they all concur with my calculation. Can anyone help?

 

[jghrist]

I'm not familiar with NFC-15-500, but the difference in calculation could be that a symmetrical component calculation is being used by the program. If so, and the transformer is delta-wye, the upstream network impedance would be included in the positive- and negative-sequence networks only. The cable impedance would be separated into positive-, negative- (same as positive-), and zero-sequence components instead of adding neutral cable impedances. This would be a more accurate approach.

 

[healyx]

Thanks for the suggestion jghrist - but I have reversed engineered the Ecodial calculation and have worked out how to reproduce the issue.

So for any single phase current (Ik1max, Ik1min and If) the transformer component of the circuit impedance is increased by 25%. So for 3 phase entities (Ik3max), they say the transformer impedance is 11.714 + j42.516, but for 1 phase entities they say it is 14.64 + j53.15 (e.g same R/X ratio just 25% larger magnitude).
Can anyone shed any light onto why they would randomly introduce 25% more impedance to single phase entities? I could understand if it was for safety but it applies to both min and max quantites, so by lowering ground fault current to be safe and ensure device disconnect times, they are also lowering Ik1max, which would be used to spec the breaking capacity of the breakers. I couldn't find any reference to this 25% factor in UTE NFC-15-500 either - which is what this program is supposed to comply to.

Another weird thing I found, was with the installation set to TNC (Shared neutral and earth conductor - called PEN conductor) - I'm pretty sure Ik1min (Phase to Neutral Fault) should equal If (Phase - Earth Fault). The programs works out Ik1min fine (if the 25% xfmr Z increase is OK), but it doesn't include the impedance of the PEN conductor at all in the If calculation - almost as if it was a 3 phase balanced fault.