The amplifier output impedance (with feedback) should be near enough zero and it’s current output capability many amps.
Anything the speaker tries to push back simply gets rejected (swamped, sunk) by the action of the feedback, so in fact the exact opposite of what you say.
It is because someone who do not understand how to measure, he explain about measurement 😀 😀 😀
In my experience, many solid state amplifier which people think sound bad is not enough slew rate. Average people do not sensitive about distortion compare than timing (reproducing of envelope wave). And then, they blame the global negative feedback.
Just for the record, I would say that analog_sa is one of the few on this site who holds sound opinions, has good taste and the ability to accurately judge sound quality without calculating. You can`t simply brush aside his views as irresponsible speculations.
I sense an incomplete understanding of feedback. It is true that speaker currents, whether they are distorted or not, flow through the amplifier output. And yes, if those currents threaten to make the amplifier different from the wanted signal, the feedback helps to prevent that by compensating the output so it stay only the wanted signal.
But saying that therefor the feedback amplifies the speakers distortion is misleading at best.
Jan
Again, quite misleading. The amp does not amplify the signal generated by striking the cone. It generates a compensating signal so that the amp output remains zero, despite the currents the speaker generates in this example. It is exactly the same as when you would short the speaker terminals and strike the cone. The cone movement is heavily damped and the cone quickly comes to a standstill. Exactly the same with shorted driver - you could say the feedback amp act as an 'active short'.
Now think about an amp with low or no feedback, say it has an output impedance of 3 ohms. This is exactly equivalent to shorting the speaker with 3 ohms. Now strike the cone, and it will take much longer for the movement to die out because the 3 ohms is an imperfect damping. You get more resonance and ringing now.
Jan
What views? The only contribution I can find of his to this thread is the very dubious assertion that there have been few or no successful amplifiers based on the high-feedback principle. I suggest that this is largely an exercise in question-begging, as it rests entirely on his definition of 'successful'.
Please supply examples to support this view. Effects described would be easily visible in tone burst tests.
What the difference does it make between 0.1% and 0.0001% amplifier distortion if speaker distortion is 5%? I think I explained good enough in my yesterday's post how feedback loop amplifies speaker distortion.
Nobody does tests with speaker, all tests are with dummy load. I dont understand what "examples" you want.
No you did not. The whole point of feedback, seen from a systems view, is that it very effectively isolates amplifier performance from speaker performance by delivering an exceedingly accurate facsimile at its output terminals of what is presented to its input. Without feedback, the ability of the speaker to effect the amplifier performance is much, much higher (hundreds or thousands of times).
You’re clearly a very smart guy, so do a bit of research and I’m sure all will be clear.
😉
You’re clearly a very smart guy, so do a bit of research and I’m sure all will be clear.
😉
Only true for low frequencies. Mechanical latencies in the speaker do not allow real cancellation.
Do another gedanken experiment. If your theory of magic effects of feedback is true, when you rap the cone, you should hear nothing at all because feedback should cancel it.
No you didn't explain anything, you just gave your opinion. I explained why you were wrong.
And you bragged about the less than 1%, and hopefully you do understand the difference between speaker and amp distortion. Because if the 5% from the speaker swamps the amp 1% or 0.0001%, how in the world can you ever hear a difference between amps??
Jan
I think you have answered your own question. As no one in your view tests with speakers, then provide test examples with speakers that illustrates what has been stated.
Paul Miller devised tests to measure the performance of amplifiers under dynamic conditions. I will need to review his load method, but I suspect issues with feedback would have shown up.
Paul Miller devised tests to measure the performance of amplifiers under dynamic conditions. I will need to review his load method, but I suspect issues with feedback would have shown up.
Don't talk about 'magic effects', that is too cheap a trick if you run out of factual arguments.
Please reread my post. If you short the speaker, or drive it with a (feedback) amp with very low Zout, the rap is damped quickly. Physics 101. Of course it is not inaudible, feedback doesn't 'cancel' things in the way you mentioned, don't make up things nobody said.
If you short the speaker with some resistance (I gave 3 ohms in my example) or a (non-global feedback) amp with some Zout, you hear the rap longer as it takes longer to come to the rest position.
I know you won't change your view because you can't reason someone out of a view that he didn't arrive at by reasoning, but others might understand how things are.
Jan
Last edited:
Not sure who you were addressing, but this might help. Ian Hegglun in Linear Audio Vol 1, showing deteriorating speaker response with higher amp Zout.
It also explains why there can be an audible difference depending on feedback factor. And it should be clear that while it is perfect OK to prefer the open loop response, it definitely is not transparent reproduction.
Jan
Attachments
Last edited: