[Shaun]

I found this while I was poking around on the internet; see link below

http://gilmore2.chem.northwestern.ed...meier4_prj.htm

It filters frequencies around 15Khz and up for CD players and is suppose to
get rid of the harshness sound of CDs.

Does this theory hold any water, if it does, I might build one.

thanks in advance,

Shaun

[Serge Auckland[_2_]]

"Shaun" wrote in message
...
I found this while I was poking around on the internet; see link below

http://gilmore2.chem.northwestern.ed...meier4_prj.htm

It filters frequencies around 15Khz and up for CD players and is suppose
to
get rid of the harshness sound of CDs.

Does this theory hold any water, if it does, I might build one.

thanks in advance,

Shaun

I'm not impressed. Firstly it assumes that digital is harsh, and I for one
don't find it at all harsh, just rather more accurate than the alternative,
presumably vinyl.

Secondly, all it does is to take off the top, and you can do that just by
using the tone controls, which is what they're for. It's one of the
mysteries of HiFi that people got rid of their tone controls, then try and
use all sorts of alternatives, the silliest of which is cable choice, to try
and reinstate some control of tone.

Thirdly, if you want an external box as a tone control, then get one of the
many decent equalisers which cost very little, like the Behringer DEQ2496,
and use the parametric and/or 1/3rd Octave graphic EQ to achieve exactly
what's required.

This filter project could have been a lot more useful if it had variable
frequency and slope controls so could also be used for filtering LP
playback, as older pre-amps like the Quad and Leaks used to have.

S.

[Shaun]

"Serge Auckland" wrote in message
...
"Shaun" wrote in message
...
I found this while I was poking around on the internet; see link below

http://gilmore2.chem.northwestern.ed...meier4_prj.htm

It filters frequencies around 15Khz and up for CD players and is suppose
to
get rid of the harshness sound of CDs.

Does this theory hold any water, if it does, I might build one.

thanks in advance,

Shaun

I'm not impressed. Firstly it assumes that digital is harsh, and I for one
don't find it at all harsh, just rather more accurate than the
alternative,
presumably vinyl.

Secondly, all it does is to take off the top, and you can do that just by
using the tone controls, which is what they're for. It's one of the
mysteries of HiFi that people got rid of their tone controls, then try and
use all sorts of alternatives, the silliest of which is cable choice, to
try
and reinstate some control of tone.

Thirdly, if you want an external box as a tone control, then get one of
the
many decent equalisers which cost very little, like the Behringer DEQ2496,
and use the parametric and/or 1/3rd Octave graphic EQ to achieve exactly
what's required.

This filter project could have been a lot more useful if it had variable
frequency and slope controls so could also be used for filtering LP
playback, as older pre-amps like the Quad and Leaks used to have.

S.



It isn't exactly a filter. What it does is add a time delay to the original
signal, then sums the original signal with the time delayed signal. At low
frequencies it doesn't really change the signal (audio), at higher
frequencies, the time delayed signal partially cancels the original signal;
thus it works like a filter but does not change the phase of it.
So a tone control or an EQ would not be an exact replacement for it.

Shaun

[Andrew Haley]

Shaun wrote:
I found this while I was poking around on the internet; see link below

http://gilmore2.chem.northwestern.ed...meier4_prj.htm

It filters frequencies around 15Khz and up for CD players and is suppose to
get rid of the harshness sound of CDs.

Does this theory hold any water, if it does, I might build one.

Those diagrams are horribly misleading. A real reconstruction filter
in a CD player doesn't do that: it reconstructs the original waveform,
almost up to the Nyquist frequency. The author could easily have
checked this with an oscilloscope; I presume he has one. He says
"Although this sampling frequency allows us to record signals up to 22
kHz the upper frequencies are not very well presented. Figure 1
compares the sampling of a 2.5 kHz and of a 21 kHz waveform. After
sampling of the original signals I connected the consecutive
measurement-points by straight lines." But a CD player does not
"connect the measurement-points by straight lines." The mathematics
behind what it actually does are he

http://en.wikipedia.org/wiki/Whittak...lation_formula

Andrew.

[Doug McDonald[_6_]]

On 3/15/2011 8:57 AM, Dick Pierce wrote:
Shaun wrote:
It isn't exactly a filter.

Yes, it is.

What it does is add a time delay to the original
signal, then sums the original signal with the time delayed signal. At low
frequencies it doesn't really change the signal (audio), at higher
frequencies, the time delayed signal partially cancels the original signal;

So it's a classic comb filter.

thus it works like a filter but does not change the phase of it.

Yes it does. The cacellation occurs because you have
two paths, each with different delays, that are summed.

f(w) = g1 sin (wT) + g2 sin(wT + t)

The delay adds an additional frequency-dependent phase
term, and the net output signal is the sum of both
the amplitude (determined by g1 and g2) AND phase
differences (determined by w and t).

So a tone control or an EQ would not be an exact
replacement for it.

No, what you have implemented is a non-minimum phase comb
filter, as compared to a minimum-phase EQ or tone control.

But it IS EXACTLY a filter, nonetheless.

So how is comb filtering done by this better (it is
different, to be sure) than comb filtering doe by, oh,
delayed acoustic reflections and partial cancellation?



However, if my calculations are correct this sum is a phase
shift linear in frequency. So it represents a simple time delay
superimposed on a comb filter frequency response.

Doug McDonald

[Doug McDonald[_6_]]

On 3/15/2011 8:57 AM, Dick Pierce wrote:

Yes it does. The cacellation occurs because you have
two paths, each with different delays, that are summed.

f(w) = g1 sin (wT) + g2 sin(wT + t)



That formula is wrong .... I had the exact same wrong one when
I started.

The correct one is

f(w) = g1 sin (wT) + g2 sin(w(T + t))

The second term is delayed by a frequency independent time,
so the w (omega) is outside the sum. Your original formula one is
a constant phase shift (in radians) independent of frequency.

Doug