Sorry if my post offended you. I am fully aware that you have earned highest reputation here and elsewhere, many times over. There was no offense intended. What I meant was the virtual impossibility of creating a sound reproduction chain that "disappears". There will be always something that betrays the artifice.
Thanks for the explanation about 4P10S. I once had a chance of getting some 30 6CD6 tubes, for free, at store closure, but decided against it due to low g2 rating. But I lucked with a cheap batch of Mullard PL38s, another sweep tube, which turned fantastic both as audio pentode or triode. The bonus was 30 V series heaters allowing 120 V transformerless heater connection for 4 tubes.
I am afraid I am hijacking the thread, so I stop here.
Hi,
I suggest to cleary state what kind of feedback is talked about as the parameters influenced by the feedback may be influenced oppositely, for example the impedance values of a circuit.
One can divide at least into four types:
- voltage derived voltage- or current-feedback
- current derived voltage- or current-feedback
Further one can divide by the number of encompassed stages between local feedback and global feedback.
It seems that the global methods are in the mids of the critique ... probabely since they are easier to ´recognize´ in most schematics.
Local loops are harder to spot and often fly fully below the radar ...
... and hey, putting a resistor somewhere in a circuit almost always introduces some kind of feedback anyway .... so if You can´t avoid FB, can it be bad?
I agree with BP that if You use FB, use as much as possible.
Interesting I find his claim, that the OL bandwidth be large also, as the typical very low OL-Bandwidth, resp. the falling-with-frequency feedback factor leads to a raising distortion amplitude response ... and a associated frequency dependent dynamic behaviour of the circuit.
jauu
Calvin
You've got the conundrum. A typical recommendation is to have as broad open loop bandwidth as possible and as low open loop distortion as possible before applying global NFB. But who needs NFB if things are good without it?
Although not tube-related, some 15 years ago there was an interesting Motorola application note regarding the use of their newly developed low distortion complementary bipolar power transistors MJL1302A/MJL3281A. It was recommended to reduce global NFB to -10 dB from the commonly used -60 dB. I think Motorola engineers knew something about adverse effects of excessive NFB in audio.
Many years ago I vehemently resisted that view, probably because I had taken out a patent on it not being the case. 😱
But at the end I had to agree that indeed that is the case. An important learning occasion, although somewhat traumatic for my tender ego 🙂
Learning can be painful.
Jan
For best sound a speaker needs the right amount of damping of the bass resonance. This usually comes from a combination of electrical and mechanical damping. If you remove the electrical damping (e.g. by using a pentode output with no feedback) then the mechanical damping needs to be increased. Such a speaker will then have too much damping when used with a conventional amplifier.45 said:
Surely the opposite is true? If the active devices are nearly linear then they would need a bigger control voltage change to change their gain then if they were more nonlinear. Of course, they will need smaller adjustments so the control voltage needed will be the same either way: a certain control voltage will produce a big change in a nonlinear device which needs a big change, and a small change in a linear device which only needs a small change. Anyway, it seems a daft idea to me as it introduces a servo loop which will have its own problems. There is no free lunch.smoking-amp said:
In most cases NFB would be needed to reduce output impedance even for a distortionless amplifier.sser2 said:
15 years ago is sufficiently recent that the 'zero feedback' meme had already got started. Application notes are not just 'helpful info for our customers'; they are marketing too, and in marketing you need to hit whatever are the current hot buttons. Also, 15 years ago is sufficiently recent that an app note could be written by an 'audiophile' rather than a real engineer. There was a famous app note (written for a chip amp IIRC?) which had a PSU design which followed so many 'audiophile' myths that it could almost have been a spoof note.
Error correction schemes usually consist in making the gain of voltage followers output stages to be as most as possible equal to 1, as they are naturally a bit less than that. The EC circuit extracts the missing voltage and injects ahead of the output stage. It's positive feedback. If for some reason, the output voltage is higher than intended, the excessive voltage will be reinjected in the same manner, it's then negative feedback. There is a notion of balance between the inputs of the error correction circuit, Cordell sometimes used the expression "feedback on demand" about this process.
Last edited:
Is this feedback method known ?
Also, is there an op-amp that outputs such a low distortion at 200khz ?
This is not an optimised version. Maybe it will get even better.
Also, is there an op-amp that outputs such a low distortion at 200khz ?
This is not an optimised version. Maybe it will get even better.
Attachments
Last edited:
Surely control is needed. That's one aspect of speaker design.
The conventional amplifier that has very low Zout unfortunately produces distortion in trying to damp the driver. It's not better at all. It's just a compromise like others. The only reason why is common is because it's more convenient and cheaper for mass production. As this is one advantage I would suspect that it is weak in other areas as nothing comes for free.
The reason why it distors the signal is that the feedback current in the coil following excitation is proportional to moving mass velocity. Whereas the suspensions force is proportional to the displacement. The contemporary presence of two different forces gives non-linearity and therefore distortion. The transint response is poorer. It's not intuitive but that is.
Two linear forces do not create non-linearity and distortion. They may create phase shift.45 said:
No. You said "proportional". That means linear. That means no distortion. A linear filter does not create distortion, no matter how much people frighten themselves by looking at waveforms.
No means that is not simply a phase shift. Transient distortion is more general as it involes both amplitude and phase. As the forces in play depend on different variables both are affected. Musical signals are NEVER made of single tones and so if you can't reproduce the initial burst information is lost as you loose the amplitudes relationships with harmonics.
Last edited:
Musical signals are ALWAYS made of single tones. Mr Fourier says so, and he is right. Phase shift is not distortion.
However, if you are now backtracking and saying that some of the forces in a loudspeaker are not "proportional" then I might agree with you.
However, if you are now backtracking and saying that some of the forces in a loudspeaker are not "proportional" then I might agree with you.
I don't know any acoustical intrument that emits a pure single tone when I play a note. A note is made of a fundamental tone + harmonics. Nothing to do with Fourier.
I never wrote they are proportional. So it's not be backtracking, sorry!
I wrote than one is proportional to velocity and another is proportional to displacement.
Yes, not "a single tone" but lots of single tones. Each tone is single, but there are lots of them.45 said:
In your post 70 you wrote45 said:
Sorry if I misunderstood your use of the term "proportional" to mean 'proportional' (i.e. linear) when you actually mean something else.45 said:
So you did write proportional. You just didn't mean 'proportional'? For the avoidance of doubt or confusion, "proportional to velocity" and "proportional to displacement" are both proportional and so both linear. Think of a resistor and a capacitor: both linear components yet one has a current proportional to voltage while the other has a current proportional to rate of change of voltage (analogous to velocity). Combine a resistor and capacitor and you have a linear filter, which changes waveforms and frightens people yet introduces no distortion.45 said:
Nonsense.
Are you kidding? Your example of RC filter is not representative for a speaker driven by an amplifier.
guys, please cut the tedious/tendentious tit for tat
45, if you want to argue speaker driver linear vs nonlinear modeling just state that up front
with your history you have no excuse for mangling "nonlinear" vs "linear" "distortion"
recognizing that most here assume the unqualified word "distortion" is by default referring to nonlinear distortion
linear filtering can change phase and amplitude as a function of frequency - but doesn't create harmonics or IMD nonlinear distortion products
45, if you want to argue speaker driver linear vs nonlinear modeling just state that up front
with your history you have no excuse for mangling "nonlinear" vs "linear" "distortion"
recognizing that most here assume the unqualified word "distortion" is by default referring to nonlinear distortion
linear filtering can change phase and amplitude as a function of frequency - but doesn't create harmonics or IMD nonlinear distortion products
- Status
- Not open for further replies.