Mooly
I actually quite like the quasi configuration, as you can match the output transistors usually very much more closely than for a complementary pair, and in a relatively simple circuit like this (a classic design) matching of the drivers and output is quite important.
But another potential issue (as others have pointed out) is its tendency to latch up on overdrive. That can easily be cured by limiting the current in the pre-driver. Ideally to twice its quiescent current.
And with global phase lead compensation (as you showed) I have, in other circuits of similar concept, obtained more or less the same low (excellent) crossover performance whether quasi or complementary. Although the phase lead I had in mind was the one Bailey used, with the capacitor connected between the pre-driver collector and emitter of the input transistor. But your configuration gives the lowest crossover distortion.
I actually quite like the quasi configuration, as you can match the output transistors usually very much more closely than for a complementary pair, and in a relatively simple circuit like this (a classic design) matching of the drivers and output is quite important.
But another potential issue (as others have pointed out) is its tendency to latch up on overdrive. That can easily be cured by limiting the current in the pre-driver. Ideally to twice its quiescent current.
And with global phase lead compensation (as you showed) I have, in other circuits of similar concept, obtained more or less the same low (excellent) crossover performance whether quasi or complementary. Although the phase lead I had in mind was the one Bailey used, with the capacitor connected between the pre-driver collector and emitter of the input transistor. But your configuration gives the lowest crossover distortion.
Ah ! I would say its pretty much ideal then for your intended use. These circuits can always be tweaked and everyone has their own ideas.
John, I did try the Bailey type compensation on the simulation for this amp and tbh it just didn't seem as good a fit. Maybe the very low VAS current of only 1ma or so played a part although it could probably be optimised
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.