In my final year at uni i did my thesis on direct audio from power supply amplifiers inspired by a design by Manny Rodrigues
http://www.apsdc.com/DAFPS 598,6498/dafs598,6498.pdf
I made a very simple 10w prototype and got it to work fairly well. However, the biggest problem i found with the distortion found when around the crossover region when the duty cycle was low <2%. You just can't get the mosfets to turn off and on fast enough.
Has anyone seen this circuit before? Any ideas how you could get it to work for full-range audio? Is it beneficial over standard class-d?
http://www.apsdc.com/DAFPS 598,6498/dafs598,6498.pdf
I made a very simple 10w prototype and got it to work fairly well. However, the biggest problem i found with the distortion found when around the crossover region when the duty cycle was low <2%. You just can't get the mosfets to turn off and on fast enough.
Has anyone seen this circuit before? Any ideas how you could get it to work for full-range audio? Is it beneficial over standard class-d?
You should add small offset to polarity comparator to avoid crossover distortion...Or negative bias to rectified input of PWM. Seem to be probleme in that region...
Fred
Fred
Class-BD modulators suffer from crossover distortion, you can reduce it, but it will always be there. It might comfort you to know that the implementation claimed by the inventor himself suffered the same problem; he says his amplifier does not produce noise when idle, any amplifier that does not reproduce its own noise at the output will not respond to any other signal of similar magnitude, and thus will suffer crossover distortion.
- Status
- Not open for further replies.