Transformer configuration and harmonics 4

[bozzah]

I am modelling a system with 5 medium voltage transformers feed to the common bus. On seconday side of each transformer are PV solar panels connected to DC/AC inverter. The inverter is connected to the secondary side of MV transformer.
The inverter generates voltage harmonics (at the moment I do not have right harmonic data, so I placed arbitrary "close" values for 3rd, 5th, 7th, 9th and 11th harmonics for each inverter). When run the model in ETAP (v 7.1.0) it is clear that transformer (Delta-Y[grounded]) "captures" 3rd/9th harmonics which is according the theory behind it. However, when I change configuration of the transformers (TO any configuration) I still got the same results on the common bus, that is 3rd/9th is "eliminated" while other harmonics are idenitical on the commnon bus as it is on Delta-Y[grounded]. THE ONLY exception is Y[grounded]-Y[grounded] in which case I get all of the harmonic frequencies on the primary side of the bus. I am trying to undersand why every configuration gives out similar results (with exception go Y[grounded]-Y[grounded]) or there is a issue with the software.
Do harmonics pass though common ground at Y[gr]-Y[gr]?

 

[jghrist]

Triplen harmonics (multiples of three) are zero sequence and will not pass through a transformer with an ungrounded winding.

 

[electricpete]

That sums it up very well. To elaborate the reason why triplen harmonics are zero sequence: assume that we have a 60hz system and that the 3phase current waveforms are identical except shifted in time by and interval corrresponding to 120 degrees of the fundamental. That interval is 0.0167 sec / 3 = 0.00556 sec. But 0.00556 sec is one full period (360 degrees) of the 3rd harmonic (180hz). So the 3rd harmonic currents are all shifted from each other by 360 degrees... in other words they are in-phase with each other (zero sequence). They cannot flow into a winding unless there is some path other than the three phases for them to flow out (like ground).

=====================================
(2B)+(2B)' ?

 

[bozzah]

thank you

 

[electricuwe]

Unless your PV-inverters are using some ancient thyristor design, you will not have any concern with harmonics. At least in Europe using IGBT voltage source inverters is common practice for PV inverters of any size since many years.