Why Is High Feedback Considered Bad In Audio? In Simple Terms

Good!

When Larkin is wrong, he never admits it :)))))

VLV

Good point.

Definition of thud?

Leftist weenies thud, rattle or ring. Engineers talk numbers and formulas :-)

How about this:

HPF = 1 - LPF

If LPF is rings, then HPF rings, too.

VLV

Apparently they didn't teach you FIR filters.

VLV

^^^^^^^^^^

John

Are you writing your own dictionary?

You seem to want to argue definitions, or rather define all HP filters and a ton of simple RC networks as "ringing." Fine, go for it.

So a punching bag is indistinguishable from a brass bell.

John

Well, when it comes to the arguments like that... OK, keep loosing your face :)

VLV

He has his own working definition of the word "ring." Anybody can do that. Even "monotonous."

Which this thread is becoming.

John

Just a language difference amusement...

monotonous =/= monotonic

...Jim Thompson

Abby, The phenomenon you refer to (increased distortion when feedback is applied) is called =93transient intermodulation distortion=94 (TIM). TIM was a hot topic in the early 1970s. You don=92t hear much about it nowadays because it s causes have been identified and, thanks to faster semiconductors, have been eliminated. Here=92s what happens: If the slew rate capability (maximum rate of change of voltage with respect to time) of any stage of an amplifier is lower than the actual slew rate of a signal, then the output can not keep up with the input. The result is a very large error exists between the input and the feedback. This error causes the input stage to saturate. When the output finally =93catches up=94 to the input, it takes awhile for the input stage to come out of saturation. During this period a large intermodulation distortion component appears at the output.

In the 1970s, power transistors were quite slow, preventing the output from keeping up with the input. Today, TIM is easily prevented simply by designing the open loop amplifier so that all stages have adequate slew rate capability. As long as this is the case, negative feedback will always result in lower distortion, and no distortion artifacts will occur that were not present in the open loop amplifier. Regards, Jon

hen

ng&sa...

How about this:

Y =3D A*exp(B*t) + C*exp(D*t) ...etc not ringing

Y =3D A*exp(B*t)*sin(C*t) ... ringing

For C 0 and both series converging

Have you ever seen a real filter act like that? For that filter (after carefully and accurately characterizing the components) does the spice simulation agree with the experimental results?

in be

Not exactly like that but sort of the same but harder to do with ASCII art. The squiggles where much faster. In the intended application it worked ok. I'm sure that spice would agree with the results if the SRF of the inductors was accurately modeled etc.

The first one can have multiple inflexion points in the step response as well.

VLV

Pluto isn't a planet so we can have thins like that and perhaps not call them ringing.

They're not meant to blatently lie. But take for example, Nivea's Oxygen Power products:

"NIVEA VISAGE Oxygen Power Reviving Day Cream with 15% pure oxygen increases surface skin cell renewal. Your skin is revived from deep within and more resistant to combat daily stress."

Sounds good, but what does it mean? Apart from the 15% pure oxygen (as distinct from 15% impure oxygen, presumably) there's nothing there you could pin them down on. Even "increases surface skin cell renewal" is of dubious interpretation given that surface skin cells are dead.

Any strategy is to make a claim of an objectively measurable improvement, but then include a small print caveat showing that the claimed improvement is some trivial amount, and that even that was supported by tests on a statistically insignificant number of people (perhaps being the particular test group who provided the required result, with the other test groups being ignored).

Sylvia.

No, but monotonic waveforms do tend to be monotonous.

John

Yikes, that's serious 1/F noise turf.

What's the signal impedance? What sort of noise levels do you need?

John