diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Digital Source (http://www.diyaudio.com/forums/digital-source/)
-   -   marantz cdp active filter phase (http://www.diyaudio.com/forums/digital-source/222122-marantz-cdp-active-filter-phase.html)

GLENZWORLD 23rd October 2012 10:44 PM

marantz cdp active filter phase
 
4 Attachment(s)
I have been playing with a marantz cdp and have changed the op amps to ad797b x2 on brown dogs also changed and tried many different passive components.

have tried smt components of different types and they sound bright edgy

so I am now redoing the filter using ridiculously priced through hole components hopefully without inductive noise.

I am having trouble trying to set and or simulate a filter and hoping someone can give some real world advice

If I am trying for a cutoff of say 23khz the bode plot for phase looks a mess.

Is why the original circuit datasheet for tda1549 output makes for 74khz fc?

do I want a 23khz (about?) fc on any of the filters from dac out?

where do you set cutoff frequency for best q ?

reading another thread here

Simple, good quality DAC

where Elso Kwak setting his cutoff at 10khz??
presumably to avoid 44khz phase inter modulation?
but if oscillator is on separate board (my cdp) would you get 44khz noise at dac?

I have also heard of others trying to bypass internal op amps on tda1549 on same thread by Rudolf Broertjes

did Lukasz Fikus (lampizator) not try this and it did not work?

I will try to contact Rudolf on that one.

my main question is what to set filter frequency to get phase right or does that matter?
well I will post pictures and multisim bitmaps in the hopes someone that knows op amp behavior can give advice


ps

thanks for all the help from stephen sank and others on this forum and 6 months of googleing I am grasping small amounts of this stuff.

oh well I am a carpenter by trade and would not want a noob of 6 months building my home If you know what I mean;)

thanks
Glen

PlasticIsGood 24th October 2012 09:59 PM

Filters. "Simple harmonic motion" is an oxymoron: how can such simple things be so hard to understand? When Margaret Thatcher said economics is just like housekeeping, I thought "But housekeeping is really hard." Everything dynamic is hard to grasp. Don't worry, hardly anybody understands this stuff. Nearly everyone's bluffing.

I was OK with counting, and arithmetic was easy. Then there was algebra and geometry that caused some pause for thought but I got the hang of it after a while. Trigonometry took a bit longer. Then everything fell apart with calculus, and calculus became everything, in all science, because we can't analyse dynamic systems without it.

We can't handle dynamics directly so we use calculus to reduce it to statics; to the maths we understood before. Consequently it becomes one stage removed from reality. There's an inescapable feeling of having lost the plot. Economists have no choice but to live in ivory towers. This is what we need to do:

Start with descriptions of moving things, because we are familiar with force, distance, velocity and acceleration. This immediately requires an introduction to differential equations, which is where calculus begins.

Move on to apply calculus to things that move in circles, or waves. A weight hanging from a spring, attached to a fixed point, is a common starting place, and is an example of simple harmonic motion. Add some resistance and we have damped SHM, which is fundamental to filters.

Then consider continuous excitement, such as moving the attachment point of the spring up and down in a wave motion. Now we have a low-pass filter. At low frequencies the weight follows the motion of the spring mount exactly. As frequency increases, the weight begins to fall behind in time, and a phase difference emerges. Increase frequency further and, as it approaches the natural frequency of the system, the amplitude of the weight increases. Beyond that frequency, phase difference increases to some maximum, and amplitude falls until, when frequency is very high the weight hardly moves at all. Apply what we know about differential equations and we can analyse and predict this behaviour. Notice how phase and amplitude are inextricably related.

Now consider several damped spring-mass-damper combos of various lengths and weights connected in a daisy chain. That's a multi-pole low-pass filter. The differential equations become too complicated so we use a trick rather similar to when we used logarithms to find square roots: a transform such as Laplace or Fourier. This removes us one more stage from reality and we feel lost again. Now our differential equations become transfer functions, and we've got complex numbers to contend with.

What's worse, the maths of manipulating these transforms eventually becomes too hard, so we use even more transforms, such as Nyquist, or graphical techniques like Bode analysis and pole-zero plots.

The analysis of a low pass filter for a CD player requires all of that. Inductors are like weights, capacitors springs, and resistors are dampers. The equations are the same. Our experiments with springs would show that, if we want a sharp attenuation beyond some particular frequency, then we need lots of sections, and lots of sections means complicated behaviour. It's very hard, if not impossible, to get both amplitude and phase to behave as we wish throughout the pass band. Every section adds wiggles somewhere.

Oversampling reduces the complexity of the filter by requiring a less sharp cut-off. This makes it possible to get close enough to constant amplitude and linear phase throughout the pass band.

NOS forces compromise. Here's an example:

http://www.diyaudio.com/forums/digit...iwa-dxm45.html

There are lots of filter simulators about, many intended for crossover design. Here's one:

SVCfilter download page - Tonne Software

I'm still struggling to grasp all the ramifications of aliasing, amongst all the other things I'm trying to grasp...

One upshot of the Aiwa's filters is that frequencies around 18kHz are modulated by a mysterious audible tone of a few kHz.

martin clark 24th October 2012 10:24 PM

Popping an AD797 on an adaptor and dropping it unbidden into an existing circuit is a recipe for unhappiness if ever there was one. Wrong tool for this task without serious reengineering to suit that (excellent) amp.

There's little or nothing wrong with the circuit or filter design; its the act of using the wrong tool for the job, and poor implementation at that, that makes the result worse.

Read Mooly's excellent write-up on opamp swaps and their sensitivities before attempting anything else.

GLENZWORLD 25th October 2012 05:37 PM

Quote:

Originally Posted by PlasticIsGood (Post 3214221)
Filters. "Simple harmonic motion" is an oxymoron: how can such simple things be so hard to understand? When Margaret Thatcher said economics is just like housekeeping, I thought "But housekeeping is really hard." Everything dynamic is hard to grasp. Don't worry, hardly anybody understands this stuff. Nearly everyone's bluffing.

I was OK with counting, and arithmetic was easy. Then there was algebra and geometry that caused some pause for thought but I got the hang of it after a while. Trigonometry took a bit longer. Then everything fell apart with calculus, and calculus became everything, in all science, because we can't analyse dynamic systems without it.

We can't handle dynamics directly so we use calculus to reduce it to statics; to the maths we understood before. Consequently it becomes one stage removed from reality. There's an inescapable feeling of having lost the plot. Economists have no choice but to live in ivory towers. This is what we need to do:

Start with descriptions of moving things, because we are familiar with force, distance, velocity and acceleration. This immediately requires an introduction to differential equations, which is where calculus begins.

Move on to apply calculus to things that move in circles, or waves. A weight hanging from a spring, attached to a fixed point, is a common starting place, and is an example of simple harmonic motion. Add some resistance and we have damped SHM, which is fundamental to filters.

Then consider continuous excitement, such as moving the attachment point of the spring up and down in a wave motion. Now we have a low-pass filter. At low frequencies the weight follows the motion of the spring mount exactly. As frequency increases, the weight begins to fall behind in time, and a phase difference emerges. Increase frequency further and, as it approaches the natural frequency of the system, the amplitude of the weight increases. Beyond that frequency, phase difference increases to some maximum, and amplitude falls until, when frequency is very high the weight hardly moves at all. Apply what we know about differential equations and we can analyse and predict this behaviour. Notice how phase and amplitude are inextricably related.

Now consider several damped spring-mass-damper combos of various lengths and weights connected in a daisy chain. That's a multi-pole low-pass filter. The differential equations become too complicated so we use a trick rather similar to when we used logarithms to find square roots: a transform such as Laplace or Fourier. This removes us one more stage from reality and we feel lost again. Now our differential equations become transfer functions, and we've got complex numbers to contend with.

What's worse, the maths of manipulating these transforms eventually becomes too hard, so we use even more transforms, such as Nyquist, or graphical techniques like Bode analysis and pole-zero plots.

The analysis of a low pass filter for a CD player requires all of that. Inductors are like weights, capacitors springs, and resistors are dampers. The equations are the same. Our experiments with springs would show that, if we want a sharp attenuation beyond some particular frequency, then we need lots of sections, and lots of sections means complicated behaviour. It's very hard, if not impossible, to get both amplitude and phase to behave as we wish throughout the pass band. Every section adds wiggles somewhere.

Oversampling reduces the complexity of the filter by requiring a less sharp cut-off. This makes it possible to get close enough to constant amplitude and linear phase throughout the pass band.

NOS forces compromise. Here's an example:

http://www.diyaudio.com/forums/digit...iwa-dxm45.html

There are lots of filter simulators about, many intended for crossover design. Here's one:

SVCfilter download page - Tonne Software

I'm still struggling to grasp all the ramifications of aliasing, amongst all the other things I'm trying to grasp...

One upshot of the Aiwa's filters is that frequencies around 18kHz are modulated by a mysterious audible tone of a few kHz.

thanks for the reply

that was a great analogy you should teach

in fact if teachers would show practical applications to the math we would be more interested.
I know I would have paid more attention in math class

thanks for the info and links and letting me know I am not the only one standing there with a blank look on my face

Glen

GLENZWORLD 25th October 2012 05:56 PM

Quote:

Originally Posted by martin clark (Post 3214234)
Popping an AD797 on an adaptor and dropping it unbidden into an existing circuit is a recipe for unhappiness if ever there was one. Wrong tool for this task without serious reengineering to suit that (excellent) amp.

There's little or nothing wrong with the circuit or filter design; its the act of using the wrong tool for the job, and poor implementation at that, that makes the result worse.

Read Mooly's excellent write-up on opamp swaps and their sensitivities before attempting anything else.

thanks for the reply

I am now realizing that it's not like swapping passives

I will read mooly's opamp swapping if I can find and am reading as much as I find from self and others

do you have a link to that thread?

the harder this gets the more I want to at least try this amp, but if I can't make this sound nice I can allways order different amps.

hell it's only money, time and my sanity:(
Glen

GLENZWORLD 25th October 2012 05:57 PM

Quote:

Originally Posted by PlasticIsGood (Post 3214221)
Filters. "Simple harmonic motion" is an oxymoron: how can such simple things be so hard to understand? When Margaret Thatcher said economics is just like housekeeping, I thought "But housekeeping is really hard." Everything dynamic is hard to grasp. Don't worry, hardly anybody understands this stuff. Nearly everyone's bluffing.

I was OK with counting, and arithmetic was easy. Then there was algebra and geometry that caused some pause for thought but I got the hang of it after a while. Trigonometry took a bit longer. Then everything fell apart with calculus, and calculus became everything, in all science, because we can't analyse dynamic systems without it.

We can't handle dynamics directly so we use calculus to reduce it to statics; to the maths we understood before. Consequently it becomes one stage removed from reality. There's an inescapable feeling of having lost the plot. Economists have no choice but to live in ivory towers. This is what we need to do:

Start with descriptions of moving things, because we are familiar with force, distance, velocity and acceleration. This immediately requires an introduction to differential equations, which is where calculus begins.

Move on to apply calculus to things that move in circles, or waves. A weight hanging from a spring, attached to a fixed point, is a common starting place, and is an example of simple harmonic motion. Add some resistance and we have damped SHM, which is fundamental to filters.

Then consider continuous excitement, such as moving the attachment point of the spring up and down in a wave motion. Now we have a low-pass filter. At low frequencies the weight follows the motion of the spring mount exactly. As frequency increases, the weight begins to fall behind in time, and a phase difference emerges. Increase frequency further and, as it approaches the natural frequency of the system, the amplitude of the weight increases. Beyond that frequency, phase difference increases to some maximum, and amplitude falls until, when frequency is very high the weight hardly moves at all. Apply what we know about differential equations and we can analyse and predict this behaviour. Notice how phase and amplitude are inextricably related.

Now consider several damped spring-mass-damper combos of various lengths and weights connected in a daisy chain. That's a multi-pole low-pass filter. The differential equations become too complicated so we use a trick rather similar to when we used logarithms to find square roots: a transform such as Laplace or Fourier. This removes us one more stage from reality and we feel lost again. Now our differential equations become transfer functions, and we've got complex numbers to contend with.

What's worse, the maths of manipulating these transforms eventually becomes too hard, so we use even more transforms, such as Nyquist, or graphical techniques like Bode analysis and pole-zero plots.

The analysis of a low pass filter for a CD player requires all of that. Inductors are like weights, capacitors springs, and resistors are dampers. The equations are the same. Our experiments with springs would show that, if we want a sharp attenuation beyond some particular frequency, then we need lots of sections, and lots of sections means complicated behaviour. It's very hard, if not impossible, to get both amplitude and phase to behave as we wish throughout the pass band. Every section adds wiggles somewhere.

Oversampling reduces the complexity of the filter by requiring a less sharp cut-off. This makes it possible to get close enough to constant amplitude and linear phase throughout the pass band.

NOS forces compromise. Here's an example:

http://www.diyaudio.com/forums/digit...iwa-dxm45.html

There are lots of filter simulators about, many intended for crossover design. Here's one:

SVCfilter download page - Tonne Software

I'm still struggling to grasp all the ramifications of aliasing, amongst all the other things I'm trying to grasp...

One upshot of the Aiwa's filters is that frequencies around 18kHz are modulated by a mysterious audible tone of a few kHz.

thanks for the reply

that was a great analogy you should teach

in fact if teachers would show practical applications to the math we would be more interested.
I know I would have paid more attention in math class

thanks for the info and links and letting me know I am not the only one standing there with a blank look on my face

Glen


All times are GMT. The time now is 11:05 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2