Author Topic: Class D amp  (Read 1036 times)

0 Members and 1 Guest are viewing this topic.

Offline Shinobot2

  • Newbie
  • Posts: 4
  • Country: pt
Class D amp
« on: October 15, 2015, 10:28:12 pm »
Hi, I tried to build a class D amplifier but I´m having some problems.

I see a lot of distortion in the output specially at frequencies above 3kHz.
I already changed the filter to see if it was the problem, but not much improvement. The triangular wave is nice and clean, but the frequency is not as I predicted, 420kHz instead of 500kHz.
I removed the feedback and it got a lot worst.

The pictures bellow are related to the 3 points in the schematic, Vfeed, Verror and Vo, as you can see Vfeed looks good, but Verror doesn´t and it´s  similar to the output.

I didn´t measured the signal across the load, because of grouding problems, don´t know if it makes a big diference.
I connected to my speaker and sounds ok.


Offline NoItAint

  • Contributor
  • Posts: 33
  • Country: us
Re: Class D amp
« Reply #1 on: October 16, 2015, 01:57:09 am »
Looks good to me  :)
You should measure with the speaker or load connected.
The LC filter is calculated with the load.  A lot of the switching noise will filter out when a load is applied.
The frequency of the switching can change, just keep in mind it's much higher than the signal your reproducing.

Offline Shinobot2

  • Newbie
  • Posts: 4
  • Country: pt
Re: Class D amp
« Reply #2 on: October 16, 2015, 10:28:26 am »
The load is connected, it´s a 4ohm, but not purely resistive.
The amp is a full brige with fixed switching frequency around the 500kHz.

I´m not measuring the signal across the load because I dont have an isolated power supply.

What I dont understand is why the output looks realy weird  and the filtered feedback signal looks like a sinewave (with switching ripple as expected). Is it because I´m not measuring signal across the load?

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo