Frequency Converter With 3 Winding Transformer

I never knew a system with those parameter´s! normally a pulsed source have freq´s on the order of the megahertz! with such low freq´s it seams hard for me to find those transformers transform!

Are sure that all the motors will be loaded equally all the times? If not, harmonics problem would not get solved with a three winding trafo even.

Thank you for your information. The load is balanced always under the trafo, i.e both LV windings are loaded equally.

This project is in Russia and we have to define the arrangement of big frequency converters and motors under the provided local supply network. My query is as follows:-

1) Three winding transformer + frequency converter without harmonic filter (7.5 - 10% of harmonics in the primary side of the transformer and 35 - 43% in the secondary side). There is one 12 pulse frequency converter (1500kW) and rest are 6 pulses.

2) Three winding transformer + FC with harmonic filter (both primary and secondary side the harmonics percentage almost be same, 5.29%). There is one 12 pulse frequency converter (1500kW) and rest are 6 pulses.

3) Two winding transformer + FC (all 6 pulse) with harmonic filter (both primary and secondary side the harmonics percentage almost same, between 6 - 7%)

The price of a harmonic filter is around 20% of the price of the frequency converter. Similarly the price of 3 winding trafo is 20% more than the 2 winding.

So first of all I can filter out the arrangement 2 because it is the most expensive method out of these three arrangements.

From 1 & 3 arrangements it is clear that 3 is much expensive than 1 because of the filter cost nevertheless the arrangement 1 do more energy consumption to produce the same output because of this much % of harmonics (this is my thought), then I think the additional amount required (cost of filter) to implement the arrangement 3 on site that amount can be taken back within an year from our electricity consumption/bill. According to the arrangement 1, I see, throughout running, the plant will have harmonics around 35 - 40% and this makes an additional amount of power requirements and it has to be delivered by the transformer (in this case we have oversize the transformer) and we have to make the payment for this additional load to the utility provider for nothing.

So anybody can confirm whether the energy consumption of the arrangement 1 is higher/lower or equal with the arrangement 3 for running the same load?

The load is as follows:- 1500kW, 400kW (2 Nos.) and 200kW (2 Nos.) are fed by a 20kV/690VAC supply network with a 3.5MVA trafo ( I have two option for the trafo either 2 winding or 3 winding ). I have gone through some document what I have here to get the answer of my question but unfortunately it was totally helpless on this topic. I look forward to hearing from you guys.

Thanks in advance for your help and support.

This is not something that can be answered so flippantly, it requires a detailed Fourier analysis that takes into account the existing available fault current of the system, the existing harmonics, the percent loading, all kinds of other details too numerous to discuss here. You need to hire someone with expertise to do this.

But IN GENERAL, 12 pulse inverters are a legitimate means of reducing the harmonics to a level close to IEEE519 acceptable levels at the PCC. 18 pulse is better and IN GENERAL will meet IEEE519, 6 pulse is the worst and the usual suspect when it comes to excessive THD. But everything also really depends on your total DEMAND distortion, ie how much is the distortion as it relates to the total power being used. So for example at a facility using 5MW, a 5kW 6 pulse inverter all by itself is insignificant.

As to the energy issue, you do not mitigate harmonics to change energy consumption, and you do not forego harmonics mitigation to save energy. They are mutually exclusive concerns, unless you are comparing multi-pulse inverters against some other form of harmonic mitigation.

Thank you very much for your valuable information. Could please go through with my previous comment which I made in the forum at 12:43pm and advise me about which option will be feasible for this project. Last week I submitted these three options to Siemens in Austria and Germany. Siemens Austria selected the option 1 and Germany selected option 3 as best. I am totally confused at the moment.

The triplen harmonics will short themselves out around the first delta winding they find and will be lost as heat.

Change from 6 pulse to 12 pulse shift the area where the harmonics form. A 6 pulse machine will create havoc around the 5th, 7th and 11th harmonic whereas the 12 pulse machine will affect the 11th 13th etc.

Try not to use the drive without the filter. You will seriously reduce the life of the VSD, massively increase the current/power used and have issues with the cable length between the VSD and motor.

Help me here... for pulses you are referring to?

suggest looking at manufacturers details. At high level it is to do with the front end of the vsd drive. From your point of view its whether you supply it from a 3 phase transformer or get an additional phase shift transformer as well. The 12 pulse is more complicated but might be necessery if you have harmonic issues lower down. Suggest an analysis first to determine whether you can stick with 6 pulse.

I can't go-ahead with AFE/regenerative type FC's because it is too expensive and takes more place. I do not have that much space in my electrical building.

Here, I need to make a comparison report and inform my team, use, either 3 winding trafo with 30 degree phase shift in the secondary (the THDI in the primary will be around 10%) or a normal 2 winding trafo with FC and harmonic filter. If I use LHF, the efficiency of the FC will go down around another 1%. I contacted the following companies like WEG, ABB and Siemens etc. and each of them have different opinion.

Dear,

One can't assume anny thing in this complicated sequences...But,

If some thing can be og help to you, pl. find the coments here..

1. Using 3 transformer windings with winding configurations of Dyn11 , Dyn21 , Dyn31 you can cut off the harmonics in HV side. As, secondary harmonics will nullify each other. Of course, that depend upon the balancing of the loads in each 3 transformers windings.

2. Many a times, Harmonics need be calculated at the operating level. Not the designed maximum power level. With frequency converters - 6 pulse , inherent harmonics are shall be in the tune of 18% - 38% depend upon the operating lload

3. Filter size should be finalized after the commissioning of the equipment, & then, it will be economical solution.

Hope, you get some idea from above...

Thank you,

Regarding point 1, I think the harmonics in the secondary side of the transformer (all even order will cancel each other) never alleviate even if the load is balanced until use a LHF in the line side of the variable speed drive. For e.g. if I use two 6 pulse VSDs in parallel for each motor feed and each of them get supply independently from the LV windings (this arrangement makes a 12 pulse configuration), the harmonics in the secondary terminals of the trafo will be 35 to 43%. The 3 winding trafo only eliminate the 5^th, 7^th, 13^th, etc., but the 11th will still be there around 10%. But I need to know the energy consumption will be same or not in 3 winding with FC + no harmonic filter and 2 winding with FC + harmonic filter (10%). I am sure that in a non-linear load under normal operating condition it will have some additional load due the harmonics. i.e. S = SQR (P^2+Q^2+D^2. If the power consumption in both cases is same (or a small variation) then I do not have to think about the 2 winding with FC + LHF because if we compare the price it is expensive.