Spwm filter design (lc filter)

[FvM]

You already posted this filter design in a parallel PWM filter thread https://www.edaboard.com/threads/290376/

Unfortunately the design is useless for the discussed purpose
- A PWM inverter doesn't implement 50 ohm source impedance and impedance matching
- A filter topology with "1st element shunt" will propably destroy the inverter
- The filter must implement a compromise for a load impedance range

A stop band attenuation of 100 dB is rarely needed for this application, but that's a minor point.

 

[sabouras]

You already posted this filter design in a parallel PWM filter thread https://www.edaboard.com/threads/290376/

Unfortunately the design is useless for the discussed purpose
- A PWM inverter doesn't implement 50 ohm source impedance and impedance matching
- A filter topology with "1st element shunt" will propably destroy the inverter
- The filter must implement a compromise for a load impedance range

A stop band attenuation of 100 dB is rarely needed for this application, but that's a minor point.

This is an example of filter as the attachment title refers to. Of course and the load impedence is very different.
As for topology yes it need to be coil as first element, and for better design it needs broadband transformer in input and output if that is possible for matching purposes. The stopband is theoretical only. Could not achieved easy.
The huge stopband is only to understand the attenuation which may have with a simple filter.

 

[HARSHID1993]

take a look on the sample filter. At 50HZ the attenuation is about 0.2db and at 4KHZ 100db.

thanxx for your help from this design i got sine wave..
so can you pleASE give the design funda behind it?

 

[sabouras]

When you have to signal , different frequency you could distinguish by a low pass filter.
In order to work properly you should achieve measure the input and output impedance (the max power transfer theory).
As for calculations and topologies you must use a program to do those. You could use this **broken link removed**
which is free.

 

[FvM]

In order to work properly you should achieve measure the input and output impedance (the max power transfer theory).

I'm curious how you'll apply this method for an inverter output filter, because:

- The design isn't based on impedance matching ("maximum power transfer")
- The inverter output impedance is low (about zero in terms of filter design)
- The load impedance is varying between no load (infinity) and a value for rated output power

 

[sabouras]

I think at the inverter output could use a transformer. The transformer it gives an stable impedance so the filter could be re-design at that impedance for example 1ohm.
So you could do the same for the output.
Another option is to avoid the transformer with very poor performance on the filter section. The complex impedance of the load could be inductive or capacitive. In the new attachment (changed so we have inductors at the input and the output) the filter is most tolerant to complex impedance.
Another option is the usage of https://en.wikipedia.org/wiki/Line_Impedance_Stabilization_Network

 

[HARSHID1993]

i didn't get proper sine wave from all above discussion .
please guyz help me out..
i used LC low pass filter... if any body got pure sine wave plzz gave that circuit...

 

[Audioguru]

Why does your inverter produce a 4kHz or 5kHz squarewave output that is not PWM modulated at 50Hz? A pure sinewave inverter has a high frequency carrier that is PWM modulated at 50Hz then a simple LC filter eliminates the carrier.
You cannot filter a high frequency squarewave to produce a lower frequency.

 

[BradtheRad]

i didn't get proper sine wave from all above discussion .
please guyz help me out..
i used LC low pass filter... if any body got pure sine wave plzz gave that circuit...

There is a vital factor that is not evident in schematics...

Between pulses your load network must not see high impedance. Current must be permitted to continue to flow through it.

This is slightly different from typical H-bridge operation.
Current flows upward through your switching devices at certain times.
The body diodes may permit it, or external diodes may be needed.

My simulation in post #9 shows how this happens (greatly simplified). The op amps are ideal op amps, in the way their outputs source and sink current. One could say this is cheating, of course. Because it requires more work to achieve this in a real circuit.

 

[HARSHID1993]

SPWM Inverter and Filter

i had designed Spwm Inverter now i want to convert my SPWM Voltage Wave to pure Sine wave...

by using LC filter i got pure sine wave Current but my Voltage waveform is still SPWM please help me to resolve this Problem guys...

 

[D.A.(Tony)Stewart]

Re: SPWM Inverter and Filter

Do you have any specs? Load impedance? input impedance? Source impedance? Power? frequency? Budget ?