I read lately about how speakers aren't used enough in circuit calculations. Merely being "just the load" instead of "part of the circuit." With that said, that got me thinking about other basic principles of amplification. Feedback. I would imagine though, after being amplified, that signal would be difficult to bring back down and put back into the loop. What are some methods to do just that? (sticking mostly to board level but open to any and all)
its a moot point because the negative feedback its taken off the speaker's voltage source. And any changes in the speaker is only a small variables in the negative feedback.
Yes, unclear what you mean.
If you refer to the feedback loop with "that signal would be difficult to bring back down and put back into the loop" that's a no-brainer: just a two-resistor divider, with the output of that inserted at a convenient point near the input side.
Or is that not what you mean?
Jan
If you refer to the feedback loop with "that signal would be difficult to bring back down and put back into the loop" that's a no-brainer: just a two-resistor divider, with the output of that inserted at a convenient point near the input side.
Or is that not what you mean?
Jan
I meant electrically. Sorry for the confusion. Like a simple feedback loop for an op amp for example.
Wouldn't that still cause a good amount of distortion? At that point I would imagine you would have issues like heat noise from bringing down power in the signal like that. Or is it negligible enough that it can be solved with just simple feedback?
No. More in terms of using feedback to pull the signal closer to what the input would look like.
In good old times of tube amps, there used to be modest feedback taken from the secondary of output trafo. In some cases, even separate secondary just for feedback.
You could use it separately to preamp, but what would that achieve?
Not sure what is the point of this thread. Unless you are trying to create nested feedback. Pre and amp tied in one global negative feedback. Again, this has been done. Search chip amp with opa nested feedback.
You could use it separately to preamp, but what would that achieve?
Not sure what is the point of this thread. Unless you are trying to create nested feedback. Pre and amp tied in one global negative feedback. Again, this has been done. Search chip amp with opa nested feedback.
Most modern amps are some kind of "big" Op Amp.
Even Tube ones (as long as they use NFB), only difference being you can't use too much because transformers complicate it by adding unwanted phase shifts.
That's the very definition of (negative) feedback.
I think there is confusion here. Feedback lowers distortion, it does not increase it.
I don't know what heat noise is, unless you mean thermal noise in resistors for instance?
That's no issue is a competent design.
Also, as said, feedback lowers distortion thus 'pulling the signal closer to the input' except for gain, but gain is wanted normally.
I would suggest find a good text on feedback design and put in a few hours. That'll save you (and us ;-) a lot of time and confusion.
Jan
From a technical point of view, there is nothing wrong with using loads of overall negative feedback, even though it has had some bad press since 1966 (see http://linearaudio.net/sites/linearaudio.net/files/volume1bp.pdf and http://linearaudio.net/sites/linearaudio.net/files/volume1ltemvdg.pdf ).
What is a bit silly is that we usually use feedback configurations fixing the amplifier's output voltage when driving loudspeakers that distort less when driven with current. Anyway, there are many threads about current drive on this forum.
What is a bit silly is that we usually use feedback configurations fixing the amplifier's output voltage when driving loudspeakers that distort less when driven with current. Anyway, there are many threads about current drive on this forum.
Good links Marcel, although I feel the OP should start at a more basic text ;-)
See Guest Editorial here: https://myemail.constantcontact.com...back-.html?soid=1104292817535&aid=JMR8abv02-k
Jan
See Guest Editorial here: https://myemail.constantcontact.com...back-.html?soid=1104292817535&aid=JMR8abv02-k
Jan
That's a good idea. Definitely going to look into that.
Agreed. I understand that speakers perform a linear function, but that doesn't mean you can't design something that isn't completely linear. Composite amps are a terrific example of this.
One thing I've also been looking at are optocouplers and digital isolators. That may be a better idea than just outright feedback because it doesn't cause as much loading and heat noise. It should allow me to keep the rails separate as well.
It makes absolutely no sense to take feedback from power amp output and bring it to the preamp input. It would be a crazy idea, resulting only in stability issues. SOTA preamps and power amps have distortions of 60dB lower than any actual audibility threshold, especially with music. Ear masking is well investigated. Most of trials to "improve" distortion of nowadays amplifiers are made only for satisfaction of technical ambitions. Nothing in common with any kind of audible sound improvement. People should rather concentrate on unconditional stability to any kind of load and on reliability.
One problem with bringing feedback to the preamp input rather than the power amp input is that there are many different combinations of commercial pre- and power amps available, and any such feedback loop would have to be tailored for each combination. Then there's the whole problem with the volume control. You could be listening at one volume level, and then decide to raise the volume, just to be horrified by a loud tweeter end-of-life howl. Tone controls, if used, would be a nightmare to implement. Best to just use separates.