I'm afraid to ask what CFA-B is, if it's what I think it is it's no surprise the rest of the EE world sometimes does not take audio engineering seriously. If the feedback network does not change the OL forward transfer function it is not a CFA, PERIOD.
I've never understood the "importance" of CFA vs. VFA. Aren't we talking topologies with their respective compromises. Ergo, means to an end? And, tbh, I see a lot more CFA topologies used for "stupid-high" frequency analog.
So proof is in the eating of the pudding. May the best amp win (I hope this is appreciated as a joke), but ultimately we're (largely) hamstrung by the speed of our output stage.
So proof is in the eating of the pudding. May the best amp win (I hope this is appreciated as a joke), but ultimately we're (largely) hamstrung by the speed of our output stage.
Really? Current suffers all the time in countless designs, for many reasons. Few of us have or could measure it well during music playback, but we can hear when it's being corrected to work better. For example applying an RC in certain areas may really change the sound of the amplifier. And I don't mean just for huge oscillations. It's a drawn out process, and CFA is going to have a similar effect.
I'm not sure, but maybe he meant CFA-B as CFA-Balanced, based on the post.
Current probes are not that exotic a technology. It would be interesting to set up a scope X-Y with current on the X and voltage on the Y and really see what the maximums and relationships are. Arm waving speculation is not really necessary any more. You can always specify limits on the load you want to drive.
I may be able to do this but I'm swamped with other work for the next week or so.
I'm not sure what you mean by : "but we can hear when it's being corrected to work better". How do you correct current without affecting the voltage? Ohm's law has withstood the test of time.
If your referring to the fact that magnetic drivers are actually current driven devices whose acoustic output follows the voltage input I may be able to follow your argument.
Ultimately still limited by the output stage. You can buy yourself only so much.
6 of one, 1/2 dozen of the other. Especially when we're worried about 20 Hz to 20 kHz primarily. (Yes, loop gain at 20k is needed...)
Sure, I don't care much whether it's VFA or CFA, but I believe CFA has easily attainable benefits, and VFA takes more work to get similar results.
Still no one ever checks current, and compares it to the input signal. It just isn't part of the process in audio, currently. You aren't even doing it (yet).
But it's easy to tell there's a dysfunction often when you say, dampen an inductor. That or use a larger one because despite it being well within tolerances, musical playback may tend to cause very small saturation that actually comes across as pretty audible. There's many instances of improvements to be made, and the audible payback is easy to register when say drums just flat-out sound more real and visceral instead of wimpy.
Because it sounds cool, a little kinky, and has the potential to fit in a fashion audio niche.
I will need to chew on this for at least the duration of this weekend to grasp its deeper meaning, but most likely a bit longer.
In the meanwhile, in my urge for oversimplication, how wrong am I to regard topologies in which I can recognize a symmetrical LTP as VFA, and those where something resembling a complementary pair that is being driven at its common emitter point as CFA?
- Status
- Not open for further replies.