inverter voltage output 1

[bigheadted]

i have started to develop my knowledge a bit at work about pwm inverter drives, i think i have a good basic knowledge of how they work but when i put my scope on the output supplies to the motor i was suprised to find that when measuring volts the trace i recorded was a sinusoidal envolope of 50hz made up of the 4khz pulses

the things that supprised me were as follows
1)the frequency of the envelope remained the same no matter what the actual output frequency of the drive was
(i know that this was changing correctly as i could see motor speed)
2)i expected the output to be a block of increasing and decreasing width pulses all at the same voltage to create the current sine wasve but the voltage levels were changig sinusoidally too?
can anyone explain am i missing a fundemental point
the drive was very small single phase to three phase output and the motor connected in delta?
the current waveform frequcency changed as expected

 

[itsmoked]

Yes that is what first impressions would be...but! Notice somewhere that the drive has a modulation frequency. Something like 10kHz. This means the drive makes up the 50/60 Hz from 10kHz pulses that are all little PWM blocks.

Simply put these little blocks get larger more towards 100% the faster the motor is commanded. So you should see lots of little pulses spattered all over each line cycle of synthetic 50/60Hz.

 

[Skogsgurra]

I wonder where you were measuring?

The constant 50 Hz indicates that you have parts of the mains voltage in the signal. This is exactly what you get when you measure between ground and one of the motor terminals.

You need to take the measurements across the motor to read the correct motor voltage. To do this, you need either a differential probe (isolating) or a transformer that isolates the scope from the mains.

The waveform seen shall be a PWM square signal with varying duty cycle. When measuring across two motor terminals, you will see the sum of two PWM signals which means that you will have a three-level PWM signal with the extra level being zero volts.

Gunnar Englund

http://www.gke.org

 

[Marke]

A red star to you skogsgurra!

bigheadted, if you measure between one of the DC bus terminals and ground, you will see an AC waveform. This is because the DC is derived by rectifying the AC input voltage. Measuring one output phase to earth will give you this same AC waveform modulated by the high frequency switching waveform.

If you wish to get a good indication of the effective voltage wavform applied to the motor, you need to measure across two phases as Gunnar has correctly recommended. This will show you the switching pattern. If you add a low pass filter and measure after that, you will see a better representation of the applied waveform.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[itsmoked]

bigheaded; be very careful with a scope.

These particular signals have no problem frying your scope or you!

If you hook a standard scope gound to one of the output signals to the motor it will fry something in the scope. The scope case may then be LIVE. As you will be until you touch the scope.

 

[bigheadted]

bigheadted, if you measure between one of the DC bus terminals and ground, you will see an AC waveform. This is because the DC is derived by rectifying the AC input voltage. Measuring one output phase to earth will give you this same AC waveform modulated by the high frequency switching waveform.

can you explain this a bit further for me is this because the dc bus is floating and not grounded.