Funniest snake oil theories

Status
Not open for further replies.
Jump to Latest
#3,742
AuroraB said:
maybe not so - I'm beginng to think that some here are dead serious with this loudspeaker transmission line stuff.....
Click to expand...

Some might like to fuel the non-argument. I posted a sim a couple of weeks age using the traditional piecewise approximation to a 100 Ohm TL driving an 8 Ohm load, at line lengths much less than a wavelength the TL reflection steps fall exactly on the lumped RLC rise time. They have to, think of it as an energy time problem the time it takes to fill the line has to be the same no matter how you solve the problem.

Part of the confusion arises because in the end the waveform abberation maps as a simple phase delay not as a pure delay in the TL sense. An amp/cable/speaker system with sufficient phase margin should/might be able to be Kelvined at the load making this all moot.

And yes, matching the line to load moves the R's, L's, and C's so that the line is "charged" in one transmission length in both senses.
 
Last edited:
#3,743
You could add a common mode ferrite at the amp end 🙂

Joking apart, lots of circuitry these days almost have a ferrite bead or a cap ferrite cap pi filter on the input, the 0Vs is also often isolated with a single ferrite the split between the connector 0V and board 0V being down the centre of the ferrites (careful placement is the key). Not popular in audio circles but maybe as we live in an rf rich world that is only going to get richer maybe the only way to go. I did a month on site last year to re layout some boards because of rf ingress, interestingly they had a (very expensive) rf engineer instead of the usual 'EMC' expert to solve the problem ( +18GHz conducted immunity, a months paid education). I learnt more working with that gentleman than ever from most industry EMC courses.
 
#3,744
tvrgeek said:
jn,
So what audible issue is your settling time manifesting itself as within the AF spectrum?
Click to expand...

Why do you ask the question if you've bandied about about how it doesn't exist??

First, you need to learn the topic, you've only learned part of the subject. If it is possible, go find a prof in a university, and ask them about the settling time I mentioned. You may indeed find out that you've taken the simplified approximations too literally at the expense of overall understanding.

Humans localize (soundstage) using two parameters, inter-aural delays and interaural intensity difference.

We can discern 1.5 uSec interaural delays. Delay a portion of the audio band in the 10 to 20 uSec range, what do you expect to happen to the soundstage? Model the settling time for a speaker that vary's it's impedance from 2 to 50 ohms, with a line that's 150 ohms. Look at the settling time variance.

Using t-line settling time is only one way to skin the cat, as scott said, you can use a good rlc modelling, but you need lots of elements to get the resolution down to the microsecond domain in the audio spectrum.

jn
 
#3,745
I tried to keep it simple, my responses in red within.

tvrgeek said:
The nominal impedance is for an infinite long line. (No, that's silly.)
Nominal impedance is only correct at one frequency. (again, that's silly. Imagine a 50 ohm coax that's only good at one frequency)
Transmission line effects are primary relevant at one quarter wave and more.(never used a smith chart, eh?)
The wavelengths at audio frequencies are in the thousands of meters.
Therefore:
Transmission line characteristic impedance is totally irrelevant for AF.(you've some research to do, I suggest a univ prof at the minimum) I really don't give a flying hoot if I could get a reflection back at 100Mhz.

Now, cable inductance, resistance, and capacitance are relevant.
Well, finally, you said something correct. Now, what do you think the characteristic impedance of a cable is equation wise? Didja ever wonder why they use that sqr(L/C) part?

And your belden site? didja ever look at their chart showing characteristic impedance down into the audio realm?
Click to expand...

jn
 
Last edited:
#3,750
According to my calculations, & your assertions, the human being can localize a point-source of sound...........At 32.04 feet distant, the difference of one particular point of sound at 1.77 feet apart? I seriously doubt it.



__________________________________________________Rick.............
 

Attachments

  • DISTANT.JPG
    DISTANT.JPG
    13.7 KB · Views: 191
#3,751
tvrgeek said:
Sorry jn, you are incorrect. Maybe physics works differently in "away"
Click to expand...

I learned this back in 1974. I apply this now. Perhaps you should put on an engineering hat and work it out. I will be very happy to teach you, but you must ask questions.

Not only is settling time an electronic function, it is mechanical as well.
When I power up motion control systems I build, I have to consider the mechanical settling time in the PID setup.

You obviously do not understand control system theory. I can direct you to a book or two if you like. If you want to continue your illiterate tact, I cannot help you.

Speedskater said:
A 10 Volt step function?
Just what kind of music do you listen to that has a 10 Volt step function?
What kind of audio amplifier do you have that can deal with a 10 Volt step function?
Click to expand...

Characterization of an engineered system requires engineering understanding. How do I explain color to a blind person?

I take it you listen to sine waves in lieu of music, since all the equipment out there is characterized using sines?

Richard Ellis said:
According to my calculations, & your assertions, the human being can localize a point-source of sound...........At 32.04 feet distant, the difference of one particular point of sound at 1.77 feet apart? I seriously doubt it.



__________________________________________________Rick.............
Click to expand...

It doesn't matter what you "seriously doubt ". But keep in mind, those are NOT my numbers, they are from the audio researchers. Nordmark demonstrated human ITD capability back in 1972, and others have reviewed, duplicated, and repeated the results.

I'm just the messenger here.

Edit: Richard, I admire you for working out the ramifications if an ITD number. You at least are considering the science, thank you.
jn
 
Last edited:
#3,752
just a messenger with a very strange interpretation of psychoacoustics and audio reproduction realities to posit the cable impedance mismatch variable group delay/settling time effects as audible effects

you don't consider that the effect is largely balanced in each channel unless you have 20' of cable on one side and 2' on the other

you don't compare with the magnitudes of several phase/group delay response modifying effects in speakers: frequency varying radiation impedance, XO (both sum response and physical diver separation) - even Doppler from cone motion is bigger than single digit microsecond at reasonable SPL, typical multi way dynamic driver loudspeakers
 
Last edited:
#3,754
jcx said:
just a messenger with a very strange interpretation of psychoacoustics and audio reproduction realities to posit the cable impedance mismatch variable group delay/settling time effects as audible effects

you don't consider that the effect is largely balanced in each channel unless you have 20' of cable on one side and 2' on the other
Click to expand...

Yes, I do consider.

And frequency dependent delays will NOT be evident when the signal is mono, and when the affected information is centrally located within the image.

When the image is off axis, the game changes. Frequency dependent delays will shift off axis image components with respect to other images within the field. With respect is the key here, as it is a differential sensitivity that we are speaking about, not absolute positioning (which would require a head vice).

jcx said:
you don't compare with the magnitudes of several phase/group delay response modifying effects in speakers: frequency varying radiation impedance, XO (both sum response and physical diver separation) - even Doppler from cone motion is bigger than single digit microsecond at reasonable SPL, typical multi way dynamic driver loudspeakers
Click to expand...
Of course I do.

The very first thing I recommend when testing audibility is to use one amplifier to drive the two speakers, and evaluate the stability in position of all of the spectral content, using the speakers and identical wiring.

If the system cannot perform this properly, it cannot be used to properly evaluate anything else.

You do understand that the Doppler/dithering actually increases human sensitivity to ITD right? Without it, our sensitivity above about a kHz drops quickly, with it we retain about 1.5 uSec discernment out to about 12 KHz.

jn
 
#3,757
Logic??? At all? Today??? Anywhere???? A current thread paraphrased for clarity: "Help me design a speaker my wife might allow me to place in my own house" Apparently a man is still allowed to have male parts in his house, kept properly perfumed of course, as these could conceivably be of some use to a limited portion of the ruling class, but a male bringing a speaker in to the house he is allowed to stay in is quite controversial regardless of cables used.
 
Last edited:
Status
Not open for further replies.
Top Bottom