Can i connect any circuitry that convert the PWM output of a drive into pure sinewave

[dineshdeshmuk]

Can i connect any circuitry that convert the PWM output of a drive into pure sine wave output (U, V & W ) , so that i may measure the exact AC voltage and its exact Voltage level ??

 

[chuckey]

I seem to remember the the amplitude of a square wave is equal to the amplitude of its fundemental component (sine wave) + 1/3 the amplitude of the third harmonic + 1/5 of the fifth harmonic. Check with WIKIPEDIA for FOURIER series. So just measuring the amplitude of the waveform* will give the same amplitude as the sine wave recovered from it. Failing that what you need to do is to pass it through a low pass filter, so that only the fundemental gets through. Remember to calibrate the filters loss at the fundemental frequency. One way of doing this for a mains like voltage is to use a 1M ohm series resistor and a 100K load resistor with a 1 Mf capacitor across it. This will reduce the mains vol;tage to a safe level. So calibrate it by connecting the network across a low voltage AC supply, 24 V? measure what goes in and out of it with a DVM, then connect the 1M to your mains and the 1K to neutral/earth and measuure the voltage across the 1K. Sit down and calculate the mains input voltage.
Frank
* BEWARE if you are measuring with a DVM, check how they sense the voltage and how they are calibrated, you need a peak reading device.

 

[FvM]

I seem to remember the the amplitude of a square wave is equal to the amplitude of its fundemental component (sine wave) + 1/3 the amplitude of the third harmonic + 1/5 of the fifth harmonic.

Preceeded by a factor 4/pi in addition. But PWM drive putput isn't a square wave of fundamental frequency rather than sine shaped PWM waveform. It's comprised of the fundamental frequency and PWM components. Usual VFD drives have 4 to 20 kHz PWM frequency, thus the fundamental is clearly separated from PWM frequency.

Most modern VFD inverters have a quite accurate built-in output voltage measurement, so you usually don't need an external measurement.

 

[siongboon]

Use a low pass filter on the PWM, it will become sine wave.
A simple circuit can be a capacitor and a resistor.

 

[dineshdeshmuk]

of what value & of what power capacitor & resistor ?? Cap & Res in series , huh ?

 

[siongboon]

of what value & of what power capacitor & resistor ?? Cap & Res in series , huh ?

A simple LPF could be something presented on the following webpage. Resistor in series, capacitor across the signal.
**broken link removed**

The value depends on the frequency and duty cycle.
If you design the filter base on the lowest frequency, the higher freq sine wave will look smaller in amplitude.
Different freq, will give you different attenuation, unless you do a dynamic filter.

Hope it helps.

 

[FvM]

You would want to specify maximum motor and PWM frequency, measurement instrument impedance, acceptable voltage error and PWM residual and calculate filter requirements. For "exact AC voltage" and "pure sinewave" as claimed in the original post, a simple RC filter is most likely not sufficient.

 

[stacksmith]

Regular old PWM will have some bad harmonics. I would suggest reading this:
Don Lancaster's hardware hacker http://www.tinaja.com/glib/hack86.pdf issue about Magic Sine Waves. Basically, by shaping your pulses you can approximate the sine wave much better, with pretty amazing results.

 

[dineshdeshmuk]

by shaping up , can the dvm be easily read out the exact volts ?