JC: I know the DA effect well from the eighties, when I was designing HV impulse voltage dividers - a lot of resistor damped capacitors in series.
Reinventing the wheel.
Sure, don't read books, let alone learning from someone else's experience. A waste of time.
Philip the GH said:
Sure, don't read books, let alone learning from someone else's experience. A waste of time.
john curl said:
Before anybody jumps in, would you do everybody a favor and define what exactly "zero linear distortion" means?
Re: Calling Ernst Mach!
Yes, but my question of which come first, the current or the field, is to figure what cause the electrons to move. The electron on the other side of the cable seems to pop out at the same speed as field propagation (which are influenced by dielectric and conductor resistance?), that make me think that the field control the electron flow, not necessarily the electrons 'pushing' each other. But then again why does the conductor resistance influence propagation speed?
Perhaps it is easier just to listen than to try and understand as well. 😀
André
SY said:
Yes, but my question of which come first, the current or the field, is to figure what cause the electrons to move. The electron on the other side of the cable seems to pop out at the same speed as field propagation (which are influenced by dielectric and conductor resistance?), that make me think that the field control the electron flow, not necessarily the electrons 'pushing' each other. But then again why does the conductor resistance influence propagation speed?
Perhaps it is easier just to listen than to try and understand as well. 😀
André
As you know, the signal propagates ALONG the conductor, and energy is carried by elmag field. There would be no field and no propagation inside the ideal conductor, and there is a very very low propagation speed inside the real conductor. It is not about "electron move".
Re: Reinventing the wheel.
That's not what I was saying.
Edmond Stuart said:Sure, don't read books, let alone learning from someone else's experience. A waste of time.
That's not what I was saying.
SY said:
I'm not necessarily talking about audio, I'm trying to understand what can influence the signal.
Do you agree with my thinking that the field control the electron flow?
Good question, someone might have imagined traffic lights.
http://www.pcmag.com/encyclopedia_term/0,2542,t=signal&i=51326,00.asp
http://www.pcmag.com/encyclopedia_term/0,2542,t=signal&i=51326,00.asp
SY said:
I'm thinking of an audio signal although I guess we are going to differ on what is audible in the end. 😀
His Audio power amplifier design book is certainly one of the best stuff you can read on amps, also for people for which low-THD is not important. Would be a shame to miss it.
Have fun, Hannes
EDIT: By the way, Drude's theory was rather soon slightly reworked (by Sommerfeld) and as such very successfull even today.
Re: Ad797
OK back at my schematic;
1) Q61 is superfluous or is there for some other reason not shown. The correct bias for Q5/Q6 is 0 Vce.
2) No, if you check the collectors of Q16 and Q17 are at the same potential, two Vbe's up from the top of I3. Q50 and Q61 are b to b diodes clamping the current mirror during slew. They are drawn as transistors because they are, the excess current during clamping action gets dumped out the supply which is often more benign.
3) No, Q15's and Q18's Cob are cancelled by symmetry the same way this single stage gets Aol of 5,000,000.
I guess the schematic was pretty close afterall 🙂 Also to clarify again the term "neutralization" was probably inappropriate. The cap I call Cn is actually bootstraped EXCEPT for the voltage signal across the output stage. So it amounts to a 1-A ~ 0 bootstraping or positive feedback of the error.
Edmond Stuart said:
OK back at my schematic;
1) Q61 is superfluous or is there for some other reason not shown. The correct bias for Q5/Q6 is 0 Vce.
2) No, if you check the collectors of Q16 and Q17 are at the same potential, two Vbe's up from the top of I3. Q50 and Q61 are b to b diodes clamping the current mirror during slew. They are drawn as transistors because they are, the excess current during clamping action gets dumped out the supply which is often more benign.
3) No, Q15's and Q18's Cob are cancelled by symmetry the same way this single stage gets Aol of 5,000,000.
I guess the schematic was pretty close afterall 🙂 Also to clarify again the term "neutralization" was probably inappropriate. The cap I call Cn is actually bootstraped EXCEPT for the voltage signal across the output stage. So it amounts to a 1-A ~ 0 bootstraping or positive feedback of the error.
SY said:
Yes, I knew you would say that 😀 but it's not answering my question, do you agree with my thinking that the field control the electron flow? Just trying to understand something.
We all know that cables are not perfect, just use a longer length and measure, the question is only when the differences become audible. Surely that point will differ depending on equipment used and also the listeners ability to discern them.
Wavebourn said:
This does not seem correct.
Keep in mind that I am a fan of line sources and planars (especially ESLs).
Planar speakers do not automatically sound any better than a so-called point source. Most real world speakers are not point sources in the first place, just worse and better approximations - therein lies a big issue.
Planar speakers approximate the expansion of a wave coming from a point source at some distance from the point source. Albeit usually only in one axis...
Witness the Quad 63 design - a planar that emulates the expansion of a wavefront from a single point source at some distance from the surface of the planar diaphragm.
The audible differences between excellent speakers of the sort you allude to are due to distortions that relate back to the specific mechanical means that is used to produce the sound and the SPL vs. distortion (especially non-linear) that results from these differences.
Not to mention that deadly "7th" harmonic that just might be produced as the result of some of these distortion and IM producing mechanisms.
So other than the perception that the ear has of the sound source's apparent size and location the only other difference that is actually audible is going to be the spectra and level of distortion vs. frequency and level.
That is going to be different for speakers that are physically different, no matter what. Physically different is going to include everything right down to the cone/diaphragm material and past that point.
Planars and line sources are not magic solutions nor do they have automatically beneficial characteristics.
_-_-bear
PMA said:
I did in 1970'th when participated in design of some industrial test equipment. We used S/H there, so a DA played significant role.
FrankWW said:
I did not comment his mistake, I commented your statement that you instead of arguments try to enforce by link on some authority.
bear said:
From your explanation I don't see what does not seem correct.
I am not, because they have own drawbacks.
They are not, but cylindrical waves, relatively light diaphragms and heavy non-resonant in the band frames, tiny excursions, all factors help to reduce non-linear distortions caused by air pressure and deformations of surfaces. On high SPLs mechanical resonances are more audible not because of uneven frequency response (that presents on any levels), but because of non-linearities caused by deformations. This is my point. Some manufacturers are cutting costs on speaker constructions and materials trying to compensate resulting mechanical resonances by DSP equalizers, but that does not work as intended because making frequency response flatter they can't "equalize" distortions produced by speakers and their enclosures on resonant frequencies. Working in this direction I had to make very heavy speakers with complex shapes. It is costly and inconvenient, but so far it is the only way to go I see.
Edit: let me explain something. Long time ago working on musical synthesis I tried to simulate acoustical instruments using analog electronics. The idea seems to be simple: to generate certain waveforms that change in dynamics, then apply formants using analog filters. I had to give up because found that filtering is not enough: filters had to add own harmonics, depending on levels of signals. It was too complex and I gave up because digital sampling technology got over real synthesis. What is the difference between speakers and musical instruments in terms of physics? Nothing. The same laws work, the same principles sound. But what is desirable for musical instruments is bad for speakers.
- Status
- Not open for further replies.