A question about analog PWM generation

Hi, Yini,

When you say you "tried to build the circuit on our experiment board" does this mean some kind of prototyping board? I expect that stray capacitance caused issues. You can download the gerbers and build the actual board that we built and tested here at TI. You can find those files here.

I personally haven't worked on this design, but I have mentioned this to the person who designed and tested this board. I imagine that he is out for the holidays now, so you'll need to wait until next week to hear back from him, most likely (he is in a different group from me).

-d2

Hello...

The paper itself only talks about the PWM generator circuit and not the about the low pass filter. Page 15 will show you how the simulated low pass filter is designed, which isn't using RC only (there're some L included).

How did you design your actual low pass filter? Have you calculated the bandwidth?

Don's suggestion to have the PCB fabricated from the included gerber files is a very good one. PWM circuits include extremely fast edge rates and are difficult to build on basic prototyping boards (breadboards, perforated circuit board, etc.). 

If the low-pass filter you are using presents a very low impedance to the output comparator then the circuit will distort. Could you include the schematic for the low-pass filter that you are using. It would also be helpful to see an oscilloscope screenshot of the triangle wave and PWM output (before the filter). 

Don Dapkus said:

Hi, Yini,

When you say you "tried to build the circuit on our experiment board" does this mean some kind of prototyping board? I expect that stray capacitance caused issues. You can download the gerbers and build the actual board that we built and tested here at TI. You can find those files here.

I personally haven't worked on this design, but I have mentioned this to the person who designed and tested this board. I imagine that he is out for the holidays now, so you'll need to wait until next week to hear back from him, most likely (he is in a different group from me).

-d2

Hi d2:

The experiment board is like a 4 layer PCB board which is used for other products, there are suitable packages for the chips of this board.

And I will read the paper and the gerbers carefully. 

Thank you.

Yini

Innocentio Loe said:

Hello...

The paper itself only talks about the PWM generator circuit and not the about the low pass filter. Page 15 will show you how the simulated low pass filter is designed, which isn't using RC only (there're some L included).

How did you design your actual low pass filter? Have you calculated the bandwidth?

Hi Innocentio:

My low pass filter is a 3rd order RC filter which the R is 4 ohm and the C is 0.5 uF. I've tried to simulate it in LTSpice, and the bandwidth is satisfied for the system. 

But maybe there's other problem that I didn't concern about.

Thank you.

Yini

John Caldwell said:

Don's suggestion to have the PCB fabricated from the included gerber files is a very good one. PWM circuits include extremely fast edge rates and are difficult to build on basic prototyping boards (breadboards, perforated circuit board, etc.). 

If the low-pass filter you are using presents a very low impedance to the output comparator then the circuit will distort. Could you include the schematic for the low-pass filter that you are using. It would also be helpful to see an oscilloscope screenshot of the triangle wave and PWM output (before the filter). 

Hi John:

The pic of the filter is as the attachment,  it's a 3rd RC filter.

Since that the class D amplifier is supposed to drive a capacitive load,  so we plan to use RC filter. 

I've checked the triangle wave and the PWM output, and the output of triangle wave is as the attachment too. 

And for the waves with the same frequency, does the amplitude also have influence to the filter result?

Thank you.

Yini

Don Dapkus said:

Hi, Yini,

When you say you "tried to build the circuit on our experiment board" does this mean some kind of prototyping board? I expect that stray capacitance caused issues. You can download the gerbers and build the actual board that we built and tested here at TI. You can find those files here.

I personally haven't worked on this design, but I have mentioned this to the person who designed and tested this board. I imagine that he is out for the holidays now, so you'll need to wait until next week to hear back from him, most likely (he is in a different group from me).

-d2

Hi Don:

I've tried to download the design files of this board from the link. And when I tried to open the PCB project in protel, An error happened and the notice reads some file couldn't be found and is removed from the project. 

But actually I can find them in the fold. I don't know if  there's some problem with my protel, So I want to make sure that I have the authorization to open and modify the files right?

And on the information page of this board, a note reads that Fully assembled board is not available, so does it mean the fully assembled board is out of stock now, or we just can't purchase it from the website?

Thank you.

Yini

Hi Innocentio:

Thank you for the respond and I think the swing limitation is exactly where the problem is.

Yini

I see... Well, glad that helped you

Also thanks for the friend request :)