Characteristic of same Inductor at 50 Hz, 150 Hz, 250 Hz, 350 Hz, 450 Hz, 550 Hz

[maverick87]

Hi,

I have an Inductor designed to work at 50 Hz.
2.2 mH, 80 Amp, 415 Volt.

However, alongwith 50 Hz signal, also 5th Harmonic (250 Hz) and 7th Harmonic (350 Hz) signal are flowing through the Inductor.

Does the Inductance change or it remains constant for all these frequencies?

 

[betwixt]

The inductance remains constant but the reactance (equivalent resistance to AC) increases with frequency. (reactance XL = 2 * pi * frequency * inductance)

Bear in mind that the reactance alone will not stop the harmonics. Think of your inductor as the top (series) resistor in a potential divider and the bottom (load) resistor being the rest of your circuit. If the bottom resistor is very high in value there will be very little drop in the top one. What you should consider is adding a capacitor across the load side of the inductor to make a low-pass filter circuit.

Brian.

 

[klystron]

Also consisder the BH curves of the inductor core as a function of frequency.

 

[maverick87]

Thank you Brian for your prompt reply.

I have a series LC circuit in parallel to the Transformer grid.
My job is to filter out maximum harmonics before they reach the grid.

What do you think should be my approach?
V = 254 volts (single phase) & 440 V (3 phase)
5th Harmonic Current :- 90 A
7th Harmonic Current :- 40 A

The LC circuit would act as a passive Harmonic Filter.

The inductance remains constant but the reactance (equivalent resistance to AC) increases with frequency. (reactance XL = 2 * pi * frequency * inductance)

Bear in mind that the reactance alone will not stop the harmonics. Think of your inductor as the top (series) resistor in a potential divider and the bottom (load) resistor being the rest of your circuit. If the bottom resistor is very high in value there will be very little drop in the top one. What you should consider is adding a capacitor across the load side of the inductor to make a low-pass filter circuit.

Brian.

 

[chuckey]

The iron losses in the core of the inductor will increase with increasing frequency. You have to place your series circuit in series with your output, as the frequency rises, there is more impedance in series with it. If you place it in parallel with the output, you can resonate the circuit so it looks like a short circuit, but only at one frequency, so it can notch out the 5th but its impedance has risen again at the frequency of the 7th, so it gets through. If its in series with the output then it has to deal with the full fundemental current. Of course if there is a short at the 5th, can your generator/whatever deal with this?
Frank

 

[betwixt]

To honest I know the theory but I have no experience with filtering at these power levels. I would think a good solution would be to use either a pi network (C-L-C) or a two stage LPF (LC-LC) but as chuckey points out, shorting the harmonic may cause excessive current at the higher frequencies and overload your source. I think my solution would be a two stage LC filter, two inductors in series with a capacitor to the other line after each of them. You could possibly make them resonant to give extra suppression but that could cause other problems. What is the power source? It seems to be very rich in harmonic output.

Brian.

 

[FvM]

As mentioned by chuckey, you'll need an LC series circuit ("absorption circuits") for each harmonic to be filtered. For the current in 250 and 350 Hz operation, you have to apply a derating factor to the inductor designed for 80A 50 Hz to consider the core losses. But you would have to know the ratio of core/windings losses and designed magnetization of the choke. The best way is to ask the manufacturer for the derating factors. You should also know the grid impedance at harmonic frequencies to decide if parallel LC circuits are sufficient to pull most of the harmonic current.

P.S.:

shorting the harmonic may cause excessive current at the higher frequencies and overload your source.

You mean the grid power source? No. Harmonic filters as planned here are quite usual. Typically, the grid impedance will be sufficient inductive to decouple the filters. If other plants share the same transformer, there's a certain risk to pull harmonic currents from the outside and overload the filter circuit.

I guess, the harmonic currents are generated by rectifier or SCR power conversion circuits. Using low-pass or other series connected filters between the grid and the local power distribution would cause huge voltage harmonics and possibly failure of sensitive equipment.

 

[Orson Cart]

Industry practice is to use "series traps" (RLC) to provide the harmonic currents drawn by your load, one for each of the major harmonics, obviously the caps and inductors need to be rated for the expected currents at these frequencies - it is common to include a de-tuning resistor (high wattage, high peak power capability) so that the maximum current is limited at the fo or resonant frequency otherwise very high currents will flow if the fo is exactly 250Hz etc.
It is common to isolate the series traps from the supply grid by using small value supply inductors (these inductors carry ALL the load current to the site) if the supply transformer is low impedance or you share a Tx will other people (so that you do not filter their harmonics). Remember that the series traps are capacitive at 50Hz so you are performing power factor correction at the fundamental (50Hz) and may see a voltage rise at your site due to this.

Usually chokes for these applications are specifically designed and built to handle the required rms currents at a given frequency (and the fundamental) and the flux in the iron is well below that for a 50Hz choke.

The most comprehensive solution is an Active Harmonic Compensator which is a big box of power electronics (actually a 3 phase inverter with internal DC bus) that can "inject" harmonic currents in antiphase to those drawn by your loads and are very good at reducing harmonics to low and acceptable levels.

Regards, Orson Cart

 

[FvM]

The most comprehensive solution is an Active Harmonic Compensator which is a big box of power electronics (actually a 3 phase inverter with internal DC bus) that can "inject" harmonic currents in antiphase those drawn by your loads and are very good at reducing harmonics to low and acceptable levels.

It's mostly designated Active Power Filtering (APF). You'll find a huge amount of literature about this topic on the internet. But I guess, it's not an option for the original poster. Conventionel passive harmonic filtering is discussed in many power electronics text books.

it is common to include a de-tuning resistor (high wattage, high peak power capability) so that the maximum current is limited at the fo or resonant frequency

Alternatively, the chokes winding resistance will provide sufficient dampening. Also the nonlinear core characteristic will work as a natural limiter.