[marklivia]

I have an Ack!dac 2.0 w/ the Auricap upgrade on both the output and bypass caps. What do you guys feel are the best caps in this application for best transparency and solid bass. My friend has the DynamiCaps in his Ack! and it's a step up in the transparency dept.   Any other suggestions?  Thanks, Mark

[tanchiro58]

I have an Ack!dac 2.0 w/ the Auricap upgrade on both the output and bypass caps. What do you guys feel are the best caps in this application for best transparency and solid bass. My friend has the DynamiCaps in his Ack! and it's a step up in the transparency dept.   Any other suggestions?  Thanks, Mark

Mark,

I think 0.1 or 0.15uF Mundorf Silver/Oil is a good bypass cap for BG in my Apollo preamp. For coupling caps I heard the Obbligatos are also good one. Personally I 'd like to put the Auricaps for power supply. Good luck.

[Daryl]

The high load impedance in line level systems swamps any contribution that a capacitor might have so long as you use a sufficiently large value to keep the capacitors reactance insignifigant compared to the load impedance.

Film capacitors by nature have extremely smooth impedance transfer functions which make any tonal or transient contributions impossible.

Film capacitors (especially poly and teflon) also have extremely low non linearity which precludes intermodulation and distortion.

Then when you use the capacitor as a coupler/blocker the load impedance takes control of the circuit and the contribution of the capacitor is divided by the ratio of load impedance and the impedance of the capacitor (and the capacitor is damn impressive to start with).

If you set the corner frequency of the coupling capacitor at 1hz (8uf@20kohm) then it's contribution is swamped 1000 to 1 by 1000hz if you set the corner frequency at 0.1hz (80uf@20kohm) then the capacitors contribution is swamped 10,000 to 1 by 1000hz.

Use something affordable and decent size and don't allow yourself to be misled.

[rajacat]

Buy some relatively cheap caps from a variety of manufactures, install them in your system to determine for yourself if you can hear any difference in sound texture. Trust your ears. If you can hear a difference, go with your favorite that fits your budget. I too like Obbligatos and I can hear the difference between them and the stock caps in my Trends 10.1 and Constantine dac. They're cheap too!

-Roy

[Daryl]

The job of power supply and local supply rail bypass capacitors is simply to short out any signal which might be present.

The supply rails ideally would be the equivilant of ground with a DC offset.

The only 'special' quality a bypass capacitor can have is low impedance over as wide a band as possible (microdynamics and inner-detail are pure foolishness).

The impedance transfer function of capacitors is shaped like the letter 'V' (for Vanquishing snake oil salesmen and idoliters).

Capacitance causes the impedance to have a downward slope at the low end of the spectrum and inductance causes the impedance to have an upward slope at the high end of the spectrum.

Inductance is mostly due to the lead length so 'special' capacitors aren't likely to be much improved in terms of inductance unless they feature a low inductance lead design (Litz, ribbon, tubular) which you rarely see though countless manufacterers claim to have gone to the ends of the earth to make every possible improvement to their product.

The bottom of the 'V' is the self resonant frequency of the capacitor where it achieves it's lowest impdance.

Film/Foil type capacitors which have lower ESR have a higher 'Q' at their self resonant frequency and consequently a lower impedance in the vicinity of the self resonance frequency but at lower and higher frequencies their impedance is dictated by their capacitance and inductance so they aren't necessarily going to net you much improvement over deposited metal types.

The physical size and price of film and foil types might in the end limit the level of performance you can achieve in your design.

The best way to counter lead inductance is to use multiple smaller capacitors which divides lead inductance by the number of units employed, raises the self resonant frequency by the square root of the number of units employed and results in lower ESR for a given capacitor type.

Lower impedance at the lower end of the spectrum is a simple function of the total amount of capacitance.

Capacitors with more capacitance in a smaller space will allow you to achieve a lower impedance path to ground and locate it closer to where it needs to be which means you would likely use electrolytics bypassed by say ceramics to extend the bandwidth into the RF spectrum.

When combining different capacitors you must consider their impedance functions, design the circuit so there is no possible parallel resonance between the capacitors employed.

Again don't allow yourself to be misled.

 

[art]

I'm so glad that we have experts such as yourself to tell the hobbyists all that they need to know.

(Paul.........go ahead and slap me now. I confess!)

Pat

[rajacat]

Maybe Daryl's hearing is shot.

[Occam]

Well, to paraphrase some very clever, experienced folks over at diyhifi.org -

Sometimes ya gots to stop messing with your calculations and sims, and just build the damn thang, and if it doesn't oscillate, listen to it.



Last time I checked, we enjoy musical reproduction via listening, not by connecting the amps electrical output directly to various bodily orifices.

Don't allow yourself to be misled, indeed.

[Dracule1]

The high load impedance in line level systems swamps any contribution that a capacitor might have so long as you use a sufficiently large value to keep the capacitors reactance insignifigant compared to the load impedance.

Film capacitors by nature have extremely smooth impedance transfer functions which make any tonal or transient contributions impossible.

Film capacitors (especially poly and teflon) also have extremely low non linearity which precludes intermodulation and distortion.

Then when you use the capacitor as a coupler/blocker the load impedance takes control of the circuit and the contribution of the capacitor is divided by the ratio of load impedance and the impedance of the capacitor (and the capacitor is damn impressive to start with).

If you set the corner frequency of the coupling capacitor at 1hz (8uf@20kohm) then it's contribution is swamped 1000 to 1 by 1000hz if you set the corner frequency at 0.1hz (80uf@20kohm) then the capacitors contribution is swamped 10,000 to 1 by 1000hz.

Use something affordable and decent size and don't allow yourself to be misled.

Huh? 

[Dracule1]

The job of power supply and local supply rail bypass capacitors is simply to short out any signal which might be present.

The supply rails ideally would be the equivilant of ground with a DC offset.

The only 'special' quality a bypass capacitor can have is low impedance over as wide a band as possible (microdynamics and inner-detail are pure foolishness).

The impedance transfer function of capacitors is shaped like the letter 'V' (for Vanquishing snake oil salesmen and idoliters).

Capacitance causes the impedance to have a downward slope at the low end of the spectrum and inductance causes the impedance to have an upward slope at the high end of the spectrum.

Inductance is mostly due to the lead length so 'special' capacitors aren't likely to be much improved in terms of inductance unless they feature a low inductance lead design (Litz, ribbon, tubular) which you rarely see though countless manufacterers claim to have gone to the ends of the earth to make every possible improvement to their product.

The bottom of the 'V' is the self resonant frequency of the capacitor where it achieves it's lowest impdance.

Film/Foil type capacitors which have lower ESR have a higher 'Q' at their self resonant frequency and consequently a lower impedance in the vicinity of the self resonance frequency but at lower and higher frequencies their impedance is dictated by their capacitance and inductance so they aren't necessarily going to net you much improvement over deposited metal types.

The physical size and price of film and foil types might in the end limit the level of performance you can achieve in your design.

The best way to counter lead inductance is to use multiple smaller capacitors which divides lead inductance by the number of units employed, raises the self resonant frequency by the square root of the number of units employed and results in lower ESR for a given capacitor type.

Lower impedance at the lower end of the spectrum is a simple function of the total amount of capacitance.

Capacitors with more capacitance in a smaller space will allow you to achieve a lower impedance path to ground and locate it closer to where it needs to be which means you would likely use electrolytics bypassed by say ceramics to extend the bandwidth into the RF spectrum.

When combining different capacitors you must consider their impedance functions, design the circuit so there is no possible parallel resonance between the capacitors employed.

Again don't allow yourself to be misled.

 

Double Huh?

[BobM]

Last time I checked, we enjoy musical reproduction via listening, not by connecting the amps electrical output directly to various bodily orifices.

The visuals here have me quaking in my boots 

Bob

[markC]

Any DIYers out there hear a difference between a Solen Polyprop. and a Sonicap or Auricap? I'm willing to bet that I know the answer.

[tanchiro58]

Any DIYers out there hear a difference between a Solen Polyprop. and a Sonicap or Auricap? I'm willing to bet that I know the answer.

Markc,

That would be nice of you if you tell me (or us) the differences between these caps. Thanks.

Tan

[art]

Quoting a bunch of technical terms has done nothing to answer the o/p's question.

Terms such as ESR, ESL, lead length and inductance are not points of concern when choosing an output coupling cap. Which is the title of the thread.

Those items would be of interest in a bypass cap, which he has also expressed interest in.

While low ESR should be preferential in that case, in some applications it can cause problems. (Some 3-terminal regs will oscillate at worst, and at best cause noise peaking in the output.)

If you guys insist on spouting techno-babble, you should also be prepared to be ready to give more insight.

But as for coupling caps, in specific:

A lot will depend on personal tastes. Even different types from any manufacturer will sound different from ones using the same basic construction. Oval vs round axial caps is one case that comes to mind. Stacked foil films that are square vs ones that are rectangular is another.

On paper, none of that should matter. If you feel that it doesn't, then you can chose your coupling caps from a catalogue. (And don't try to start your own high-end audio company.) Otherwise, you are going to have to listen to them. And, yes, they will sound different.

Pat

[markC]

Any DIYers out there hear a difference between a Solen Polyprop. and a Sonicap or Auricap? I'm willing to bet that I know the answer.

Markc,

That would be nice of you if you tell me (or us) the differences between these caps. Thanks.

Tan

I could give you my opinion.
I've noticed significant improvements in sound by changing coupling caps in my pre, cd players and dac. I've had similar experiences in speaker x-overs.
In x-overs with Solens the high end is not extended. I hear the fundamental of, say, a triangle being struck, but the sound ends quickly, with very little sustain. With Sonicaps, the sound continues on much longer, giving more of a view into the recording venue.
The same thing applies to the coupling position; more of the subtle reverb and higher frequency information is presented. Mind you, I haven't gone from Solen to Auri or Sonic here. These cases were either polyester, (yuck!), or some other cheap metalized polypropylene.

[Daryl]

I clearly explained why any reasonably chosen coupling capacitor cannot contribute to the sound of an componet.

I also gave a perspective (or insight) which others don't give which makes it clearer.

If you don't understand why not just ask and learn rather than posture?

I separated my explaination into two different posts covering coupling and bypass separately yet somehow you still got lost.

Not only that but you need the same qualities for both so why would you try to differentiate them?

That is the lowest impedance across the widest possible bandwidth.

Only capacitors that will be employed in filters have more critical requirements because you cannot diminish their control over the circuit by locating their corner frequency far outside the band.

No the shape of the capacitor is not a concern and I can't believe you would suggest such nonsense.

The original poster posted in the LAB forum about upgrading output and bypass capacitors which is a simple issue and I gave accurate and insightfull information.


 

Quoting a bunch of technical terms has done nothing to answer the o/p's question.

Terms such as ESR, ESL, lead length and inductance are not points of concern when choosing an output coupling cap. Which is the title of the thread.

Those items would be of interest in a bypass cap, which he has also expressed interest in.

While low ESR should be preferential in that case, in some applications it can cause problems. (Some 3-terminal regs will oscillate at worst, and at best cause noise peaking in the output.)

If you guys insist on spouting techno-babble, you should also be prepared to be ready to give more insight.

But as for coupling caps, in specific:

A lot will depend on personal tastes. Even different types from any manufacturer will sound different from ones using the same basic construction. Oval vs round axial caps is one case that comes to mind. Stacked foil films that are square vs ones that are rectangular is another.

On paper, none of that should matter. If you feel that it doesn't, then you can chose your coupling caps from a catalogue. (And don't try to start your own high-end audio company.) Otherwise, you are going to have to listen to them. And, yes, they will sound different.

Pat

[Occam]

I clearly explained why any reasonably chosen coupling capacitor cannot contribute to the sound of an componet.

I also gave a perspective (or insight) which others don't give which makes it clearer.

If you don't understand why not just ask and learn rather than posture?
......

Circular arguments allways work best when they're not clouded by actual empirical verification. 

Coupling caps meeting certain minimal characteristics all sound the same. Have you empirically verified this? No, I don't need too, because they can't....

[Daryl]

While low ESR should be preferential in that case, in some applications it can cause problems. (Some 3-terminal regs will oscillate at worst, and at best cause noise peaking in the output.)

In LDO regulators the output capacitor helps to shape the bode plot of the amplifier.

The impedance function of the capacitor must no longer be in decline before the frequency where the gain of the amplifier starts to decline is reached because too much phase shift will be realized if they are both in decline at the same time.

It's not that low ESR is causing anything, simply that both a capacitor and resistor are required in the circuit and the design of the regulator is such that you can use a typical capacitor as both so long as it's impedance is no longer declining above a given frequency (c x ESR must be greater than a given value specific to each model).

Low ESR isn't a problem here it's just that they are trying to use one componet to fill two slots because they know ESR exists and a capacitor can function as both C and R so long as you don't use too perfect a componet.

[Daygloworange]

Daryl,

Can you share what your 2 channel system is made up of?

Is your gear of your own design? Are you using stuff you modded?

Cheers

[Daryl]

Hi Paul,

If you have much skill with electronics you would have to admit my arguments are compelling and insightfull (me talking).

Listening tests are only of very limited use.

People's suceptability to placebo's and proclivity toward occultism are well known especially with regards to hifi.

Even if what your hearing is to be believed (which it's not) you come to the roadblock that the better sounding choice could be the less accurate one because you might prefer the error of the less accurate choice.

The psychology of the situation is obvious when you look at obvious facts like the hundereds of componets in the system and the fact that folks are replacing the least critical of all of them and lavishing in the difference they imagined.

I pointed out already how manufacturers pedal capacitors they claim are 'better' but most don't' even have the sense to incorporate a litz or ribbon type lead which would likely allow them to cut ESL to 1/4 or so and extend usable bandwidth and octave higher for audio filter applications and two octaves for supply bypass and coupling applications.

These facts should alert you that there is more at work here.

I clearly explained why any reasonably chosen coupling capacitor cannot contribute to the sound of an componet.

I also gave a perspective (or insight) which others don't give which makes it clearer.

If you don't understand why not just ask and learn rather than posture?
......

Circular arguments allways work best when they're not clouded by actual empirical verification. 

Coupling caps meeting certain minimal characteristics all sound the same. Have you empirically verified this? No, I don't need too, because they can't....