BROWNIE POINTS.
Christer,
Have you ever talked to more than one person at the time?
It ain't simple.
I did see your points,only It really gets confusing when things are taken out of context.
IMHO,the amount of voltage applied to the cap will influence the degree of absorption of the signal though.
I would consider that a non-linear behaviour.
After all a phase shift does not equal the input hence the use of the term non-linear behaviour.
Can we agree that phase shifts,time lags in the signal are actually a kind of distortion?
The test used real world capacitors but concentrated on dielectrical behaviour under a set of conditions.
I think it was interesting,revealing and well thought out.
Naturally a number of other factors will enter the equation at some point.
No arguing there.
What I would like to ask you:
If in case of a high resolution audio system you were to replace say a polyester cap for a polypropylene one (equal values,same voltage rating) would you then be able to hear an increase in resolution?
Now,if so,to what element would you then contribute this?
Dielectrical materials have a "memory",in my book the less the better.
And I restate it clearly:we are talking applied coupling film caps here.
I agree that this would remain constant with voltage unless one would surpass the cap's ratings.
In that case some other deviations may occur.
Difficult stuff,won't you say?
And we're only talking DA,you can add tangens delta factors etc,etc....
It would,IMO be a more fruitful discussion if we were to ask ourselves what would make a good coupling cap wouldn't you agree?
Cheers,
Christer,
Have you ever talked to more than one person at the time?
It ain't simple.
I did see your points,only It really gets confusing when things are taken out of context.
IMHO,the amount of voltage applied to the cap will influence the degree of absorption of the signal though.
I would consider that a non-linear behaviour.
After all a phase shift does not equal the input hence the use of the term non-linear behaviour.
Can we agree that phase shifts,time lags in the signal are actually a kind of distortion?
The test used real world capacitors but concentrated on dielectrical behaviour under a set of conditions.
I think it was interesting,revealing and well thought out.
Naturally a number of other factors will enter the equation at some point.
No arguing there.
What I would like to ask you:
If in case of a high resolution audio system you were to replace say a polyester cap for a polypropylene one (equal values,same voltage rating) would you then be able to hear an increase in resolution?
Now,if so,to what element would you then contribute this?
Dielectrical materials have a "memory",in my book the less the better.
And I restate it clearly:we are talking applied coupling film caps here.
I agree that this would remain constant with voltage unless one would surpass the cap's ratings.
In that case some other deviations may occur.
Difficult stuff,won't you say?
And we're only talking DA,you can add tangens delta factors etc,etc....
It would,IMO be a more fruitful discussion if we were to ask ourselves what would make a good coupling cap wouldn't you agree?
Cheers,
Re: BROWNIE POINTS.
I know, that is one of the problems with forums.
I did see your points,only It really gets confusing when things are taken out of context.
I was pressing you a bit, I admit, but I don't think it was really
out of context.
IMHO,the amount of voltage applied to the cap will influence the degree of absorption of the signal though.
Really? Do you have anything to back up that claim with, assuming
we are operating within the voltage rating and not using voltages
high enough to cause significant heating of the cap?
I would consider that a non-linear behaviour.
After all a phase shift does not equal the input hence the use of the term non-linear behaviour.
Can we agree that phase shifts,time lags in the signal are actually a kind of distortion?
Phase shifts are per definition linear phenomena, that is
elementary circuit theory. Time lags which are not frequency
dependent are a different animal, but that is not the case here.
With your definition of non-linearity any amplifier would be
non-linear since the output does not equal the input - the
signal levels differ.
The test used real world capacitors but concentrated on dielectrical behaviour under a set of conditions.
That is something we onviously have not agreed upon yet and
SB certainly does not say anything about that in the article.
I think it was interesting,revealing and well thought out.
Absolutely, but we still do not quite know exactly what it
revealed.
If in case of a high resolution audio system you were to replace say a polyester cap for a polypropylene one (equal values,same voltage rating) would you then be able to hear an increase in resolution?
Now,if so,to what element would you then contribute this?
Given what I know right now, I would not have any opnion on
why there is a difference. SB or the DA models has not shown
me anything that seems to explain the difference for such types of caps. For ceramics and lytics, I would probably suspect the
behaviour that show up in SBs test, whatever it is.
Dielectrical materials have a "memory",in my book the less the better.
Yes, it is a type of memory, but it does not introduce any
non-linearities, since the DA behaves as RC networks.
Difficult stuff,won't you say?
Difficult and intriguing, yes.
It would,IMO be a more fruitful discussion if we were to ask ourselves what would make a good coupling cap wouldn't you agree?
But I think we are. The best coupling cap would be an ideal
capacitor. Since we can't make those, we must try to understand
what the problems are with those we have.
fdegrove said:
I know, that is one of the problems with forums.
I did see your points,only It really gets confusing when things are taken out of context.
I was pressing you a bit, I admit, but I don't think it was really
out of context.
IMHO,the amount of voltage applied to the cap will influence the degree of absorption of the signal though.
Really? Do you have anything to back up that claim with, assuming
we are operating within the voltage rating and not using voltages
high enough to cause significant heating of the cap?
I would consider that a non-linear behaviour.
After all a phase shift does not equal the input hence the use of the term non-linear behaviour.
Can we agree that phase shifts,time lags in the signal are actually a kind of distortion?
Phase shifts are per definition linear phenomena, that is
elementary circuit theory. Time lags which are not frequency
dependent are a different animal, but that is not the case here.
With your definition of non-linearity any amplifier would be
non-linear since the output does not equal the input - the
signal levels differ.
The test used real world capacitors but concentrated on dielectrical behaviour under a set of conditions.
That is something we onviously have not agreed upon yet and
SB certainly does not say anything about that in the article.
I think it was interesting,revealing and well thought out.
Absolutely, but we still do not quite know exactly what it
revealed.
If in case of a high resolution audio system you were to replace say a polyester cap for a polypropylene one (equal values,same voltage rating) would you then be able to hear an increase in resolution?
Now,if so,to what element would you then contribute this?
Given what I know right now, I would not have any opnion on
why there is a difference. SB or the DA models has not shown
me anything that seems to explain the difference for such types of caps. For ceramics and lytics, I would probably suspect the
behaviour that show up in SBs test, whatever it is.
Dielectrical materials have a "memory",in my book the less the better.
Yes, it is a type of memory, but it does not introduce any
non-linearities, since the DA behaves as RC networks.
Difficult stuff,won't you say?
Difficult and intriguing, yes.
It would,IMO be a more fruitful discussion if we were to ask ourselves what would make a good coupling cap wouldn't you agree?
But I think we are. The best coupling cap would be an ideal
capacitor. Since we can't make those, we must try to understand
what the problems are with those we have.
JUST ONE THING.
Hi Christer,
They wouldn't be if they occured over the entire frequency band in the same way.
When these differ and change with varying frequency it would be valid IMO to classify these as non-linear.
And they are not linear but that is not the issues here.
Here you really surprise me.
If that cap behaves like a sample and hold circuit where it shouldn't you say it does not happen?
Now,even if the hold gets released it will smear the next signal,won't it?
Hope so too.
Cheers,
Hi Christer,
They wouldn't be if they occured over the entire frequency band in the same way.
When these differ and change with varying frequency it would be valid IMO to classify these as non-linear.
And they are not linear but that is not the issues here.
Here you really surprise me.
If that cap behaves like a sample and hold circuit where it shouldn't you say it does not happen?
Now,even if the hold gets released it will smear the next signal,won't it?
Hope so too.
Cheers,
Re: JUST ONE THING.
But then any filter is non-linear according to you. This is not
in line with accepted terminology.
And they are not linear but that is not the issues here.
Real-world amplifiers no, but I meant that even an ideal
amplifier would be non-linear with your definition.
Here you really surprise me.
If that cap behaves like a sample and hold circuit where it shouldn't you say it does not happen?
Now,even if the hold gets released it will smear the next signal,won't it?
But it doesn't behave that way. I must ask you, do you
accept the standard models for DA? The Dow model, which
is the simpler one, has been used in audio articles on modelling
DA. This model introduces a number of links in parallel with
the capacitor, each link consisting of an R and a C in series.
The more accurate Cole&Cole model also introduces only Rs
and Cs, but in a slightly different way. Hence, the cap with
DA is just a passive circuit and cannot be non-linear, and it
does not behave as a sample and hold circuit. It can be used
in such a circuit, but so can the ideal cap, which would be the
best choice there.
Your view of how DA works is what I thought too when I first
read about it, but it turned out not to be so bad when I read
more about it.
I have seen claims that Richard Marsh says that DA causes
distorsion in his article on DA. Does anybody know it that
article is available online? I am puzzled by that claim, since
it is hard to see how any form of distorsion could result, unless
he uses some different model for DA.
fdegrove said:
But then any filter is non-linear according to you. This is not
in line with accepted terminology.
And they are not linear but that is not the issues here.
Real-world amplifiers no, but I meant that even an ideal
amplifier would be non-linear with your definition.
Here you really surprise me.
If that cap behaves like a sample and hold circuit where it shouldn't you say it does not happen?
Now,even if the hold gets released it will smear the next signal,won't it?
But it doesn't behave that way. I must ask you, do you
accept the standard models for DA? The Dow model, which
is the simpler one, has been used in audio articles on modelling
DA. This model introduces a number of links in parallel with
the capacitor, each link consisting of an R and a C in series.
The more accurate Cole&Cole model also introduces only Rs
and Cs, but in a slightly different way. Hence, the cap with
DA is just a passive circuit and cannot be non-linear, and it
does not behave as a sample and hold circuit. It can be used
in such a circuit, but so can the ideal cap, which would be the
best choice there.
Your view of how DA works is what I thought too when I first
read about it, but it turned out not to be so bad when I read
more about it.
I have seen claims that Richard Marsh says that DA causes
distorsion in his article on DA. Does anybody know it that
article is available online? I am puzzled by that claim, since
it is hard to see how any form of distorsion could result, unless
he uses some different model for DA.
MODELS ON THE CATWALK
Christer,
That is all very well in theory.
I understand where you're coming from.
Trouble is real components do not respond to models the way theory would have them to.
The whole topic is then: why?
And how does this affect the audible differences?
Shall we start all over again?
Shall we give it a little rest too?
Cheers to you,
Christer,
That is all very well in theory.
I understand where you're coming from.
Trouble is real components do not respond to models the way theory would have them to.
The whole topic is then: why?
And how does this affect the audible differences?
Shall we start all over again?
Shall we give it a little rest too?
Cheers to you,
Re: MODELS ON THE CATWALK
Yes, there is always a difference between theory and reality,
but that is the reason science tries to refine the models.
The ideal capacitor wasn't a good model, so the model was
refined with series resistance and inductance, parasitic
capacitance etc. Then this model was refined with models
for DA. I have asked myself too if the models for DA are
reasonable, but they are derived from careful measurements
of real capacitors. The point is, then, that we don't know if
DA is captured by these models with sufficient accuracy or not,
but since we don't know that we cannot claim DA responsible
for phenomenae not captured by the models.
The whole topic is then: why?
And how does this affect the audible differences?
I think this is about where we started
My point is simply, DA as it is known in the literature seems
not able to explain the differences and some of the phenomenae
SB captured cannot be explained with DA.
Why, then, is there a difference (I am all the time reasoning
under the assumption that there is a difference, you see)?
Well, either there is some other effect than DA and which we
do not know or DA has some nasty effects in practice which
are not captured by the models. I cannot see how we could
decide on which of these is the case with what we know right
now. I would like to know more about the claims of Richard
Marsch though, since they do not fit with what others say. If
he has managed to model DA in another way and confirmed
that model with measurements, then it would be a different
case.
Shall we start all over again?
Shall we give it a little rest too?
Yes, let's give it a rest. It seems we cannot agree, so i don't
think it will lead any further.
Good night,
fdegrove said:
Yes, there is always a difference between theory and reality,
but that is the reason science tries to refine the models.
The ideal capacitor wasn't a good model, so the model was
refined with series resistance and inductance, parasitic
capacitance etc. Then this model was refined with models
for DA. I have asked myself too if the models for DA are
reasonable, but they are derived from careful measurements
of real capacitors. The point is, then, that we don't know if
DA is captured by these models with sufficient accuracy or not,
but since we don't know that we cannot claim DA responsible
for phenomenae not captured by the models.
The whole topic is then: why?
And how does this affect the audible differences?
I think this is about where we started
My point is simply, DA as it is known in the literature seems
not able to explain the differences and some of the phenomenae
SB captured cannot be explained with DA.
Why, then, is there a difference (I am all the time reasoning
under the assumption that there is a difference, you see)?
Well, either there is some other effect than DA and which we
do not know or DA has some nasty effects in practice which
are not captured by the models. I cannot see how we could
decide on which of these is the case with what we know right
now. I would like to know more about the claims of Richard
Marsch though, since they do not fit with what others say. If
he has managed to model DA in another way and confirmed
that model with measurements, then it would be a different
case.
Shall we start all over again?
Shall we give it a little rest too?
Yes, let's give it a rest. It seems we cannot agree, so i don't
think it will lead any further.
Good night,
NOT SO DIFFERENT AS YOU THINK.
Christer,
Before you hit the sack:
I'm pretty much convinced we understand what we're talking about.
The Marsh and Jung articles took an interesting approach and remain as interesting as ever.
Moreover I am convinced by you're latest post that we do think along the same lines.
I do hope you understand that caps are particularly difficult since they do present rather complex models?
Ain't easy,good night to you too.
Christer,
Before you hit the sack:
I'm pretty much convinced we understand what we're talking about.
The Marsh and Jung articles took an interesting approach and remain as interesting as ever.
Moreover I am convinced by you're latest post that we do think along the same lines.
I do hope you understand that caps are particularly difficult since they do present rather complex models?
Ain't easy,good night to you too.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.