For ripple rejection split the value of the current reference resistor into two and add a polarised capacitor from the positive rail to the middle of the two new current reference resistors. This may reduce the current from the current sources.
I think the two bridge rectifiers in parallel won't work as you intend it to. Diodes are not like resistors, as in they won't share the load. If two parallel diodes aren't matched to the same forward voltages, whichever turns on first will dictate the voltage across both of them. ie the other one will never turn on. This is the reason LEDs in parallel need a resistor for EACH LED, not one for all of them. It would be best to just get a single bridge rated for the current you need to run the amplifier.
@@andrewshepherd383 No worries, thanks for putting in the effort to make a video! I don't know much about amps (which is how I ended up here), but I learned a few things from this. Maybe a cap on the output as well would help with the DC, you won't need a large value since the transistors aren't biased too much. A follow-up video with improvements would be awesome!
Yeah I know, I threw this thing together with parts I had on hand. One rectifier was getting a little too hot so I added another even though I knew one would probably current hog the other. In fact, not long after I made this video I realized how bad this design is in general. It has really bad input offset current so when you turn the volume knob it shifts the operating point of the output stage. This puts DC across the speaker which is not good. I ended up taking out the pot and added input caps instead. It has almost zero PSRR because the hum is awful unless you use a regulated supply. It also drifts with temperature worse than I thought. The only good thing about this amp is the sound. These are things I should have seen in advance when I saw the schematic. Instead, I didn't read into it too closely and just built it.
Even order distortion is what you expect with asymmetrical rails. Yours is low relative to third, so great!
Nicely done! Distortion numbers not as bad as I would have thought.
I really like how you mounted the switch and potentiometer, pretty clever. :)
Nice video. Could you please post the reddit link of the ckt ?
Also it would be nicer if you'd shown the actual demo of your amplifier.
Wow. Appreciate it. Thanks for sharing your experience.
For ripple rejection split the value of the current reference resistor into two and add a polarised capacitor from the positive rail to the middle of the two new current reference resistors. This may reduce the current from the current sources.
So doubling voltage in the mono mode quadrupled power as load R is fixed? Nice. You need to be able to handle the higher dissipation though.
It does get hot. Those tiny clip-on heatsinks are too small and I can't run it full blast for too long.
Bro is there anywhere that explains these parts more detailed.
I think the two bridge rectifiers in parallel won't work as you intend it to. Diodes are not like resistors, as in they won't share the load. If two parallel diodes aren't matched to the same forward voltages, whichever turns on first will dictate the voltage across both of them. ie the other one will never turn on. This is the reason LEDs in parallel need a resistor for EACH LED, not one for all of them. It would be best to just get a single bridge rated for the current you need to run the amplifier.
@@andrewshepherd383 No worries, thanks for putting in the effort to make a video! I don't know much about amps (which is how I ended up here), but I learned a few things from this. Maybe a cap on the output as well would help with the DC, you won't need a large value since the transistors aren't biased too much. A follow-up video with improvements would be awesome!
Yeah I know, I threw this thing together with parts I had on hand. One rectifier was getting a little too hot so I added another even though I knew one would probably current hog the other. In fact, not long after I made this video I realized how bad this design is in general. It has really bad input offset current so when you turn the volume knob it shifts the operating point of the output stage. This puts DC across the speaker which is not good. I ended up taking out the pot and added input caps instead. It has almost zero PSRR because the hum is awful unless you use a regulated supply. It also drifts with temperature worse than I thought. The only good thing about this amp is the sound. These are things I should have seen in advance when I saw the schematic. Instead, I didn't read into it too closely and just built it.
Can you expound on that and how can we use a volume pot to make it better?
Mind blown … 😳🤓🤘