Picking up from earlier, what we’re looking at here is an audio DAC reconstruction filter built around a prototype discrete opamp-like differential gain cell I’ve had in the works for quite a while. I finally chased out the last engineering details and have been listening to the final setup for about a month. I am still astonished at how good it sounds.
I designed the gain cell from the ground-up as a dedicated high-performance audio device. It uses some novel topological and other features that I’ll probably go into in a future post. For now all I want to say that the thing is wicked fast for a discrete device and has been rock-solid stable.
But why bother? Aren’t there already tons of reasonably decent, some even cheap, audio IC opamps out there? Yes, there are. But I’ve never been totally happy with any of them. Some have too much LF bloat, some are too strident—none to my ears do everything right (which is to say, do as little as possible apart from making the signal bigger and stronger).
Designing a discrete device let me optimize the gain structure specifically for audio, minimize and more effectively manage the number of parasitic interactions throughout, thermally couple and, more importantly, decouple elements as necessary, and a few other things. It started as a “Gee, let’s see…” exercise, and I have been rather shocked by the results.
Now I’m contemplating where to take things next. I’ve designed a couple small-footprint packages for the gain cell. I’m implementing a few other ideas with it too. I suspect this surprising little circuit will see some commercial application soon.
Right then. Back to listening. 😀