Even with a balanced load connected phase-to-phase, the non-linear load component will generate harmonics that will distort the sinusoidal wave.
A portion of the harmonic spectrum, the triplen harmonic current, will conduct current from the neutral point of the transformer because there are in phase with each other as indicated in the illustration below.
The degree of neutral current conduction is limited exclusively to the triplen harmonics that are the only components contributing to the neutral current similar to the zero-sequence.
Grounded-wye windings transformer on both sides allows balanced triplen harmonic flow from the LV to the HV side with the same proportion to the voltage ratio.
Consider cuky's diagram. Zero-sequence current flowing in the non-linear Ø-Ø load will circulate in the delta and produce no zero-sequence line current. ØA line current is the B-A load current minus the A-C load current. For zero-sequence, all three phases are equal in magnitude and phase: current B-A load current equals A-C load current. The difference is zero.
Compared to a single phase non linear loads:
With one line to neutral load there will be a neutral current comprised of the fundamental frequency current and the harmonic currents.
With three line to neutral loads the fundamental currents will cancel on the neutral and the triplen currents will add on the neutral.
Three phase:
With one phase to phase load there will be a neutral current composed of the fundamental current and the harmonic currents.
With three line to line non linear loads, the fundamental frequency currents will cancel on the neutral and the triplen currents will add on the neutral.
In a wye/wye bank the secondary neutral current will be reflected on the primary neutral.
Bill
--------------------
"Why not the best?"
Jimmy Carter
The sum of a balance fundamental phase current is zero as expected. However, the sum of the triplen harmonics (green arrow)is ≠ 0 (See the illustration of 3rd harmonic all in phase). Therefore, ground current in the LV side flows into the ground and back into neutral of primary transformer winding.
Cuky, this is very had for me to grasp. Excluding the neutral point ground, the system is completely sealed off from earth. I can not imagine current flowing in what appears to be an open circuit.
Simple Kirchoff problem. Consider the region comprising the load, and the three phases sum to zero. Now consider the region of the secondary winding. The three phases, which already sum to zero by themselves, sum to zero with the ground current. This is the basis for restricted earth fault protection, which is secure in the presence of linear and non-linear loading. The triplen harmonic currents circulates around the delta connected load.
If present and depending on the actual physical arrangement of the circuits, line-neutral (linear) loads can provide an alternate flow path for the third harmonic currents, capacitors being of prime concern since they could offer a low impedance path relative to the path back through the wire to where the currents from "load 1" and "load 2" meet up to cancel. These loads would also be exposed to the harmonic voltage generated by the flow of the harmonic current and thus conduct their own harmonic current.
I understand the challenge of the flow of neutral conductor current with phase-to-phase load. Rather than the conductor we suggest to focus instead on the neutral point at the grounded transformer in which we believe that there is current flow caused by the triplen harmonic not only in the transformer neutral-to-ground but also in each phase in both sides of the wye-grounded/wye-grounded transformer.
I am taking the liberty to enclose an excerpt of the paper 100 Years of Symmetrical Components by G. Chicco & A.Mazza that I hope found interesting and useful to bring light to this discussion.
After reading this excerpt, let us know if there further doubts that the terms of the triplen harmonics cancel in the neutral to ground at the transformer with a balance non-linear load.
