Karls_speaker_works
Member
- Joined
- Mar 17, 2022
- Messages
- 84
- Likes
- 129
Your original post was excellent, thanks. The only issue that seemed different to me was feed-forward and that was addressed already. The implementation I'm familiar with has the PWM R-C sawtooth ramp derived from the line voltage. Since the slope of the ramp is proportional to the line voltage, the "feed forward" information improves output regulation (DC voltage) when the line voltage changes suddenly.
Funny, that's the thought that most often crosses my mind when looking at simplified Class D schematics. In fact, a swept AC signal is injected into the feedback loop to confirm power supply stability/bandwidth/phase margin. If looking at the output on a 'scope, the injected AC signal shows as an (amplified) AC component on the DC output voltage.
I just like learning so that was why I felt disappointed that the conversation seemed to have ended abruptly. The only other thought I had while reading this is switching at > 1MHz to obtain better performance would likely lower efficiency perhaps into the mid-high 80's and add complication, thus possibly lessening the benefit from Class D. I won't get into the dead time (as I don't fully understand how it might effect distortion), other than to note that it can become a real issue as switching frequencies increase. You're right, this is encroaching on Class D 201 I think. I guess having similar experience makes me an advanced student?
Funny, that's the thought that most often crosses my mind when looking at simplified Class D schematics. In fact, a swept AC signal is injected into the feedback loop to confirm power supply stability/bandwidth/phase margin. If looking at the output on a 'scope, the injected AC signal shows as an (amplified) AC component on the DC output voltage.
I just like learning so that was why I felt disappointed that the conversation seemed to have ended abruptly. The only other thought I had while reading this is switching at > 1MHz to obtain better performance would likely lower efficiency perhaps into the mid-high 80's and add complication, thus possibly lessening the benefit from Class D. I won't get into the dead time (as I don't fully understand how it might effect distortion), other than to note that it can become a real issue as switching frequencies increase. You're right, this is encroaching on Class D 201 I think. I guess having similar experience makes me an advanced student?
