[[email protected]]

Hi,

we're designing a new "hiend" monophonic poweramp (100W@8ohm -
mosfet). It'll feature a 500W toroidal transformer and we planned to
use ELNA Silmic2 RFS caps as PS filtering for a total capacitance of
20000 to 24000uF.

I'd like to know your opinion about the pros and cons of the following
electrolytic caps configuration :

1] Total of 2x 10000uF + 2x 470uF

2] Total of 4x 4700uF + 2x 1000uF

3] Total of 50x 470uF

Will there be any difference in sound quality? If so, which will sound
better?

Thanks a lot for your help,
Paolo

[Randy Yates]

writes:

Hi,

we're designing a new "hiend" monophonic poweramp (100W@8ohm -
mosfet). It'll feature a 500W toroidal transformer and we planned to
use ELNA Silmic2 RFS caps as PS filtering for a total capacitance of
20000 to 24000uF.

I'd like to know your opinion about the pros and cons of the following
electrolytic caps configuration :

1] Total of 2x 10000uF + 2x 470uF

2] Total of 4x 4700uF + 2x 1000uF

3] Total of 50x 470uF

Will there be any difference in sound quality? If so, which will sound
better?

Thanks a lot for your help,
Paolo

I can't answer your question, but along these lines you may be interested
in Vishay's new amazing wet tantalum caps that boast up to 72 millifarads
(each!). That's *milli*, not *micro*!

At up to $750 a pop, they may not be in your running, depending on how
"high-end" your high-end is.
--
% Randy Yates % "Ticket to the moon, flight leaves here today
%% Fuquay-Varina, NC % from Satellite 2"
%%% 919-577-9882 % 'Ticket To The Moon'
%%%% % *Time*, Electric Light Orchestra
http://www.digitalsignallabs.com

[[email protected]]

I can't answer your question, but along these lines you may be interested
in Vishay's new amazing wet tantalum caps that boast up to 72 millifarads
(each!). That's *milli*, not *micro*!

At up to $750 a pop, they may not be in your running, depending on how
"high-end" your high-end is.

:-) so 1 of these caps is 72000 microfarad if I'm correct!
However I searched the Vishay website but I couldn't find them...

My question about the best sounding power supply caps configuration is
still open :

1] Total of 2x 10000uF + 2x 470uF
2] Total of 4x 4700uF + 2x 1000uF
3] Total of 50x 470uF

Thanks!

Paolo

[Paul[_11_]]

On 22 Sep 2007 02:23:37 GMT, wrote:

Hi,

we're designing a new "hiend" monophonic poweramp (100W@8ohm -
mosfet). It'll feature a 500W toroidal transformer and we planned to
use ELNA Silmic2 RFS caps as PS filtering for a total capacitance of
20000 to 24000uF.

I'd like to know your opinion about the pros and cons of the following
electrolytic caps configuration :

1] Total of 2x 10000uF + 2x 470uF

2] Total of 4x 4700uF + 2x 1000uF

3] Total of 50x 470uF

Will there be any difference in sound quality? If so, which will sound
better?


Check the ESR (Equivalent Series Resistance) for each type. If you
parallel the caps, you are doing more or less the same for the
resistors. You want a low value of ESR.
Layout is important.... Poor layout combined with high output
current can couple unwanted signals into places they don't belong,
causing instability or oscillation. The act of placing caps so that
the AREA bounded by the load current is too great, can cause trouble.
"audio" rated caps are probably not very good. You might find that
switching supply caps may be superior, especially if you intend to
drive high current at high frequencies.
I don't know the topolgy of your "hi end" amp (hi-end usually means
crap), but its more likely that things OTHER than the power supply
caps are going to have more effect on the sound.
In my experience, the types of caps doesn't make a whole lot of
difference. Layout, and stability are much more important. Do you have
test equipment? Decent test equipment can take a lot of guesswork out
of the design.
Amp design (IMHO) has been pretty well beaten to death, unless you
are looking for something special, like a low noise and distortion
switching amp. What is special about the amp you are designing?

-Paul

[Randy Yates]

writes:

I can't answer your question, but along these lines you may be interested
in Vishay's new amazing wet tantalum caps that boast up to 72 millifarads
(each!). That's *milli*, not *micro*!

At up to $750 a pop, they may not be in your running, depending on how
"high-end" your high-end is.

:-) so 1 of these caps is 72000 microfarad if I'm correct!

Yes!

However I searched the Vishay website but I couldn't find them...

I had the same experience today. I saw them advertised in the September
2007 issue of "Electronic Products."

My question about the best sounding power supply caps configuration is
still open :

1] Total of 2x 10000uF + 2x 470uF
2] Total of 4x 4700uF + 2x 1000uF
3] Total of 50x 470uF

I would say the 50x, because the equivalent series resistance is reduced
by 1/50th.
--
% Randy Yates % "And all that I can do
%% Fuquay-Varina, NC % is say I'm sorry,
%%% 919-577-9882 % that's the way it goes..."
%%%% % Getting To The Point', *Balance of Power*, ELO
http://www.digitalsignallabs.com

[jwvm]

On Sep 21, 10:23 pm, wrote:
Hi,

we're designing a new "hiend" monophonic poweramp (100W@8ohm -
mosfet). It'll feature a 500W toroidal transformer and we planned to
use ELNA Silmic2 RFS caps as PS filtering for a total capacitance of
20000 to 24000uF.

I'd like to know your opinion about the pros and cons of the following
electrolytic caps configuration :

1] Total of 2x 10000uF + 2x 470uF

2] Total of 4x 4700uF + 2x 1000uF

3] Total of 50x 470uF

Will there be any difference in sound quality? If so, which will sound
better?

Thanks a lot for your help,
Paolo

What is your objective here? Are you trying to maximize power output,
minimize noise from the AC source or what? Why do you want to use
multiple electrolytic capacitors?

[Arny Krueger]

wrote in message

Hi,

we're designing a new "hiend" monophonic poweramp
(100W@8ohm - mosfet). It'll feature a 500W toroidal
transformer and we planned to use ELNA Silmic2 RFS caps
as PS filtering for a total capacitance of 20000 to
24000uF.

Seems light on capacitance for a high end power amp. I have PA amps with
power supply capacitors of that order. Those amps are effective, and they do
sound good, but from an image standpoint, they'd win no sales contests.

I'd like to know your opinion about the pros and cons of
the following electrolytic caps configuration :

1] Total of 2x 10000uF + 2x 470uF

2] Total of 4x 4700uF + 2x 1000uF

3] Total of 50x 470uF

Will there be any difference in sound quality? If so,
which will sound better?

There's a lot of lore about capactors that is a mixture of ignorance, myth,
and outdated information. One myth relates to putting small value caps
across large value caps. It actually made sense before capacitor
manufacturers started paying more attention to ESL (effective series
capacitance). They have all pretty much found that making low ESL caps is a
very low cost design feature, so they now just do it.

I suspect that a lot of power supply caps are sized these days by figuring
how much total capacitance the designer wants, and then figuring out how to
achieve it with adequate reliability as well as the best ESL and ESR
performance, in available space and for the lowest cost. This is going to
drive you away from collections of very small caps such as your 50 x 470
alternative.

[jwvm]

On Sep 23, 11:12 am, "Arny Krueger" wrote:
wrote in message



Hi,

we're designing a new "hiend" monophonic poweramp
(100W@8ohm - mosfet). It'll feature a 500W toroidal
transformer and we planned to use ELNA Silmic2 RFS caps
as PS filtering for a total capacitance of 20000 to
24000uF.

Seems light on capacitance for a high end power amp. I have PA amps with
power supply capacitors of that order. Those amps are effective, and they do
sound good, but from an image standpoint, they'd win no sales contests.

I'd like to know your opinion about the pros and cons of
the following electrolytic caps configuration :

1] Total of 2x 10000uF + 2x 470uF

2] Total of 4x 4700uF + 2x 1000uF

3] Total of 50x 470uF
Will there be any difference in sound quality? If so,
which will sound better?

There's a lot of lore about capactors that is a mixture of ignorance, myth,
and outdated information. One myth relates to putting small value caps
across large value caps. It actually made sense before capacitor
manufacturers started paying more attention to ESL (effective series
capacitance). They have all pretty much found that making low ESL caps is a
very low cost design feature, so they now just do it.

Agreed! BTW, Arny really meant to say that ESL was effective series
inductance, I am sure. :-)


I suspect that a lot of power supply caps are sized these days by figuring
how much total capacitance the designer wants, and then figuring out how to
achieve it with adequate reliability as well as the best ESL and ESR
performance, in available space and for the lowest cost. This is going to
drive you away from collections of very small caps such as your 50 x 470
alternative.

Ceramic bypass caps distributed over a PC board are used to minimize
high-frequency impedance in digital circuitry and analog circuits with
high-gain elements like op amps to provide for low impedance at high
frequencies. Using distributed electrolytic capacitors in a similar
fashion doesn't really make sense, however.

[Arny Krueger]

"Randy Yates" wrote in message


My question about the best sounding power supply caps
configuration is still open :

1] Total of 2x 10000uF + 2x 470uF
2] Total of 4x 4700uF + 2x 1000uF
3] Total of 50x 470uF

I would say the 50x, because the equivalent series
resistance is reduced by 1/50th.

Trouble is, the ESR of a 470 uF is rarely if ever as low as that of a good
10,000 uF.

For example, in this series of caps, the 470 uF has ESZ of 0.25 ohm, while
the 4,700 as ESZ of 0.060. So instead of getting the 50x benefit you
expect, the actual benefit is only about 2.2 X.

Now lets look at cost.

http://www.mouser.com/catalog/631/634.pdf

a 470/16 costs $0.81 while a 4,700/16 costs $2.75. 3 of the 4,700 uF caps
would give you 3x more capacitance and lower ESR, not to mention a smaller
over-all size.

[[email protected]]

On Sep 22, 7:19 pm, wrote:
I can't answer your question, but along these lines you may be interested
in Vishay's new amazing wet tantalumcapsthat boast up to 72 millifarads
(each!). That's *milli*, not *micro*!

At up to $750 a pop, they may not be in your running, depending on how
"high-end" your high-end is.

:-) so 1 of thesecapsis 72000 microfarad if I'm correct!
However I searched the Vishay website but I couldn't find them...

My question about the best sounding power supplycapsconfiguration is
still open :

1] Total of 2x 10000uF + 2x 470uF
2] Total of 4x 4700uF + 2x 1000uF
3] Total of 50x 470uF

Thanks!

Paolo

Its likely you will hear no difference. Much of the stuff has already
been said about which are superior ways to use caps, and which types.
If I wanted to make a very good cap, I would get decent ESR rated
caps, and using more than one, use a star configuration, meaning there
should not be more than one cap on the way to the connection point. I
would also solder all crimps, if crimps were used anywhere. Solid
foils are pretty good for interconnection many caps, but that gets
away from star configuration unless care is taken.

greg

[Randy Yates]

"Arny Krueger" writes:

"Randy Yates" wrote in message


My question about the best sounding power supply caps
configuration is still open :

1] Total of 2x 10000uF + 2x 470uF
2] Total of 4x 4700uF + 2x 1000uF
3] Total of 50x 470uF

I would say the 50x, because the equivalent series
resistance is reduced by 1/50th.

Trouble is, the ESR of a 470 uF is rarely if ever as low as that of a good
10,000 uF.

For example, in this series of caps, the 470 uF has ESZ of 0.25 ohm, while
the 4,700 as ESZ of 0.060. So instead of getting the 50x benefit you
expect, the actual benefit is only about 2.2 X.

Now lets look at cost.

http://www.mouser.com/catalog/631/634.pdf

a 470/16 costs $0.81 while a 4,700/16 costs $2.75. 3 of the 4,700 uF caps
would give you 3x more capacitance and lower ESR, not to mention a smaller
over-all size.

No arguments here. I guess I presumed "all other things being equal," which
they are definitely not.
--
% Randy Yates % "Maybe one day I'll feel her cold embrace,
%% Fuquay-Varina, NC % and kiss her interface,
%%% 919-577-9882 % til then, I'll leave her alone."
%%%% % 'Yours Truly, 2095', *Time*, ELO
http://www.digitalsignallabs.com