Sidebands of fundamental frequency during PWM generation

[hwgeek_1221]

Unidentified low frequency components around 50Hz in a dc-ac inverter

Hello,

Im designing microcontroller based dc to ac inverter based on unipolar PWM switching. The switching frequency used is 20khz and the sinusoidal frequency that I want to produce is 50hz with an amplitude of about 10V. I have an LC low pass filter at the output of my full bridge transistor circuit. The dc bus voltage is stiff and is around 40V. The sine wave reference given in the micontroller is based on a table lookup and contains only a sine wave of 50hz [made of 400 points].

When observing the output voltage im not able to get an exact 50hz signal. The signal produced contains additional 2 low frequency components at about 45 and 55 Hz however with smaller amplitudes (other than the 50Hz).
Im unable to trace the reason of why im getting such a signal, though the amplitude that im targeting (10V) can be observed. However this 10V is not directly seen due to the superposition of the 45 and 55hz components.
When removing the LC filter and observing the frequency spectrum of the voltage signal, more frequency components around the 50hz were observed. All the components were symmetrical to the 50hz and had smaller amplitudes as moving to the left or to the right of the 50hz component.

Has anyone experienced something similar before?

 

[andre_luis]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Without seeing the code is hard to give you some advice.

 

[hwgeek_1221]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

I can't post the code neither the schematics because I don't have the soft version of the schematic and the console that im working on is not connected to the internet so i cant upload any code.

I have an infinite while loop running along with a timer that overflows/underflows every 50us. The unipolar PWM switching functionality is ran every 50us and is handled within the timer's function by updating 2 compare registers. Complementary outputs along with the deadtime feature are used.

 

[andre_luis]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

I can't post the code neither the schematics

I reafirm what was said just above

 

[treez]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

it sounds like aliasing but im not sure.
What you want, a 50hz sine from pwm, can be done with a comparator with a sawtooth into one input, and the sine reference into the other...but what is the lc resonant freq of the LC......?..that I believe will have effect....shouldn't it be well below 50hz (the res freq)?

 

[andre_luis]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

When observing the output voltage im not able to get an exact 50hz signal. The signal produced contains additional 2 low frequency components at about 45 and 55 Hz however with smaller amplitudes

I'm curious to know how the values of these marginal frequencies were measured.
Anyway, a picture of the waveform would help.

 

[FvM]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

The +/- 5 Hz sidebands indicate a 5 Hz amplitude (or possibly frequency) modulation. The question is however, how you or detecting the signal spectrum. Under circumstances, it might be a just a measurement artifact like aliasing by undersampling of the pwm carrier.

 

[chuckey]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

what are the levels of the spurious += 5HZ side bands, are they too high? Moving coil multimeters will clearly show such a low frequency, measure your output with a moving coil voltmeter. So measure your output with a moving coil voltmeter, if the pointer vibrates its a real effect.
As all the timings are determined by the micro's crystal, are the supplies to it decoupled well?
Without access to the software and different compilers, it is not possible to get any further.
Frank

 

[hwgeek_1221]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

I used the FFT feature on the oscilloscope to see them. I'm attaching a snapshot of the oscilloscope screen where the time waveform in white and the FFT in purple (in the back).

 

[KlausST]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Hi,

Maybe it is a measurement problem:
* some kind of FFT window mismatch --> Try to use an additional FFT windowing function like hanning...
* Or it is an undersampling effect --> choose an other time base

Or it is a real signal problem.
* maybe caused by DC rail fluctuations --> check DC rail voltage with a scope

Klaus

 

[hwgeek_1221]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Hi,

Maybe it is a measurement problem:
* some kind of FFT window mismatch --> Try to use an additional FFT windowing function like hanning...
* Or it is an undersampling effect --> choose an other time base

Or it is a real signal problem.
* maybe caused by DC rail fluctuations --> check DC rail voltage with a scope

Klaus

I don't think its a measurement problem. I checked the DC rail and i didn't observe any frequency component that I had observed on the output of the fullbridge.

 

[FvM]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

The time waveform shows an about 4.1 Hz amplitude modulation, you don't need FFT to identify it clearly. You should be able to find out if it's real or a measurement artifact, I guess it's real. Can be better seen with faster time base setting. In any case, that's the origin of the side bands.

 

[dick_freebird]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Do you see significant jitter on either edge of the
controller output waveform(s)? Is there significant
supply ringing (ringing tail) present at the point
where current mode compare (presuming this is a
current mode controlled design) is done? If you
find jitter, is it deterministically banded and does
the excursion just happen to look like about 10%
of cycle time?

 

[KlausST]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Hi,

Check your pwm,

Either all rising edges, or all falling edges should have equal dustance.
(As long as you don't use symmetric pwm generation)

Best is to use a digital scope.

Klaus

 

[andre_luis]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Is this inverter performing some kind of loopback control ?
I'm guessing this could be happening due to a self oscillation.

 

[hwgeek_1221]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Do you see significant jitter on either edge of the
controller output waveform(s)?

No I don't.

...presuming this is a
current mode controlled design?

No I'm not using current mode control. It's open loop.

- - - Updated - - -

Hi,

Check your pwm,

Either all rising edges, or all falling edges should have equal dustance.
(As long as you don't use symmetric pwm generation)

Best is to use a digital scope.

Klaus

Yes I am using symmetric PWM since my timer running is an up-down counter. Counts from 0->PERIOD and then back from PERIOD->0 and outputs are set to HIGH when CMPR > Timer_counter

I didn't understand the part of the rising and falling edges. Can you clarify please?

- - - Updated - - -

Is this inverter performing some kind of loopback control ?
I'm guessing this could be happening due to a self oscillation.

No I don't have feedback at all!

 

[KlausST]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

Hi,

I didn't understand the part of the rising and falling edges. Can you clarify please?

&

As long as you don't use symmetric pwm generation

--> but you use symmetric PWM. So the thing with rising and falling edges does not relate to your application.

Klaus

 

[andre_luis]

Re: Unidentified low frequency components around 50Hz in a dc-ac inverter

While hwgeek_1221 do not provide more relevant information, such as a piece of code or part of the schematic, we will stay restricted at the field of the assumptions.

 

[hwgeek_1221]

Hello,

I am trying to generate a sin wave by using PWM. My MCU is an stm32f405 running at 168 MHz. The timer that I'm using ( the one generating the PWM signal) also runs at 168 MHz in up-down count mode (symmetric PWM). The timer has a frequency of 10KHz (switching frequency) and the frequency of my sin wave is 50Hz. I have a sin lookup table of 200 points that cover a full 50Hz cycle generated from matlab and imported to my project. On every overflow of the timer a new point is taken from the sin table and placed in the compare register. The effect of the change in the compare value takes place on the next cycle of the timer (Due to the existence of preload and shadows registers). I also have a lowpass RC filter on Channel 1 of my timer.

The output is taken on the capacitor of the RC filter. I am able to get a 50Hz sin wave however I have sidebands added to it, one on about 46Hz and the other on 54Hz. I'm using the FFT on the oscilloscope to find the frequencies of the sideband. Looking at the oscilloscope in the time domain, one can see that the output voltage on the capacitor of the RC is "wiggling". The sidebands have a smaller amplitude than the fundamental.

Has anyone encountered something like this before? What can be the cause of this?

Thank you!

 

[c_mitra]

This is natural and cannot be avoided but I am wondering the origin of the 4Hz modulation. It is part of the synthesis process and if you change the 200 points (sample you are taking for the sine table) this 4Hz will also change.

Only with an infinite number of points this 4Hz sidebands will finally disappear. Better for you will be take lesser number of points (so that sidebands will be further) and then you can filter it off. 50+/-4 will be a bit tricky...