Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

"Jack"

i.e. are bi-polar caps generally not used in the audio signal path?


** Bi-polar electro caps are INTENDED for use in audio signal paths were
there is no DC bias.

And they do it very well.

Any posturing lunatic here who says otherwise is a POS asshole.




............. Phil

Vishay Roederstein EKSU-Series.

Samuel

Jack wrote:
Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

Hi John,
The panasonic SU series are bi-polars, available at digikey.

Brian

Brian Allen wrote:

Jack wrote:
Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

Hi John,
The panasonic SU series are bi-polars, available at digikey.

Can't tell your location since you're posting through google.

If you're in the UK ( or anywhere else they operate ) , Farnell has a range
too. Also known as 'non-polarised'.

http://uk.farnell.com/jsp/endecaSear...gensearch&y=11



Graham

(Jack) wrote in message om...
Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

You can deal with this a couple of ways. If you use polarized
electrolytics, first use a VOM to determine the side of the cap which
has the greater voltage. Rotate pots and push switches to determine if
the voltage reverses. If not, place the polarized caps with the + pin
to the + voltage. If the dc shifts excessivly you can "create" a
polarized cap by tying the two + input pins of the caps together to
make a non polarized electrolytic. You can further experiment by
adding a polarization voltage to the juction of the + pins.
I would expect using a pair of Panasonic FM series or Rubycon Z
series caps would make a superior non polar cap than any of the pre
made ones out there (Black Gates being an exception). Avoid the
Roedersteins, I've removed thousands of them over the years and they
sound like there's cotton stuffed up your ears. (Plus they are very
large).

Jim Williams
Audio Upgrades

For sound quality & (especially) reliability, NOTHING can touch the Nichicon Muse bipolars,
which are available cheaply from Handmade Electronics in Allentown, Penna. Digikey finally
started carrying Nichicon caps recently, but not the top grade Muse series. Vastly better than
the Panasonics of any grade or era.

--
Stephen Sank, Owner & Ribbon Mic Restorer
Talking Dog Transducer Company
http://stephensank.com
5517 Carmelita Drive N.E.
Albuquerque, New Mexico [87111]
505-332-0336
Auth. Nakamichi & McIntosh servicer
Payments preferred through Paypal.com
"Jack" wrote in message
m...
Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

Jack wrote:

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

What is the device?

Bipolar caps are almost always found in the audio path when you see them,
but you'll need the schematic to be sure. You don't need them in supply
decoupling applications.

Panasonic does make some okay NP types, which you will find a couple pages
later in the Digi-Key book.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Stephen Sank wrote:

For sound quality & (especially) reliability, NOTHING can touch the Nichicon Muse bipolars,
which are available cheaply from Handmade Electronics in Allentown, Penna. Digikey finally
started carrying Nichicon caps recently, but not the top grade Muse series. Vastly better than
the Panasonics of any grade or era.

What characteristics affect the sound quality in your opinion ?

In the audio band, I'd expect a cap to be pretty much a cap. Excepting medium and Hi-Z ceramics
with their voltage dependent dielectric characteristics.

Graham

"Stephen Sank" wrote in message ...
For sound quality & (especially) reliability, NOTHING can touch the Nichicon Muse bipolars,
which are available cheaply from Handmade Electronics in Allentown, Penna. Digikey finally
started carrying Nichicon caps recently, but not the top grade Muse series. Vastly better than
the Panasonics of any grade or era.
Have you evaluated the new Panasonic FM series? I would submit that the Rubycon Black Gates are by far the finest electrolytic caps. I also believe no electrolytic cap is good enough for high quality music without another high Q bypass cap around it. Electrolytic caps smash pulses and therefore are transient killers.

Jim Williams
Audio Upgrades

--
Stephen Sank, Owner & Ribbon Mic Restorer
Talking Dog Transducer Company
http://stephensank.com
5517 Carmelita Drive N.E.
Albuquerque, New Mexico [87111]
505-332-0336
Auth. Nakamichi & McIntosh servicer
Payments preferred through Paypal.com
"Jack" wrote in message
m...
Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

Rubycon Black Gates have quite good performance(though still not as good as Nichicon Muse), as
do Elna Cerafine/Silmic caps, but both makers have a horrible reliability track record, at
least in my 30 years of experience. In all of that time, I can count on one hand, with fingers
left over, the number of Nichicon caps of ANY age or grade that I have had to replace, compared
to many hundreds(thousands, perhaps) of Elna & Rubycon caps. I would say that Rubycon in
particular is the single most UNreliable lytic maker in the world. Just ask any VCR tech.
Panasonics are very reliable & good performers, but a distant second on both counts at any
grade versus Nichicon.
And I do very much agree that even the best lytics need good film byassing for audio apps,
whether in the signal path or power supply. Considering the chemistry/physics of lytics, it's
a miracle they work at all.

--
Stephen Sank, Owner & Ribbon Mic Restorer
Talking Dog Transducer Company
http://stephensank.com
5517 Carmelita Drive N.E.
Albuquerque, New Mexico [87111]
505-332-0336
Auth. Nakamichi & McIntosh servicer
Payments preferred through Paypal.com
"Jim Williams" wrote in message
m...
"Stephen Sank" wrote in message ...
For sound quality & (especially) reliability, NOTHING can touch the Nichicon Muse bipolars,
which are available cheaply from Handmade Electronics in Allentown, Penna. Digikey finally
started carrying Nichicon caps recently, but not the top grade Muse series. Vastly better
than
the Panasonics of any grade or era.
Have you evaluated the new Panasonic FM series? I would submit that the Rubycon Black Gates
are by far the finest electrolytic caps. I also believe no electrolytic cap is good enough for
high quality music without another high Q bypass cap around it. Electrolytic caps smash pulses
and therefore are transient killers.

Jim Williams
Audio Upgrades

--
Stephen Sank, Owner & Ribbon Mic Restorer
Talking Dog Transducer Company
http://stephensank.com
5517 Carmelita Drive N.E.
Albuquerque, New Mexico [87111]
505-332-0336
Auth. Nakamichi & McIntosh servicer
Payments preferred through Paypal.com
"Jack" wrote in message
m...
Hey,

I'm taking on my first recap project, and I've noticed several
bi-polar caps in the circuit. I was planning on using Panasonic FCs
for the whole thing, but it appears as though there aren't any in the
FC series, any suggestions on types I could use, or are they
generally less important (i.e. are bi-polar caps generally not used in
the audio signal path?) Thanks

/John\

Stephen Sank wrote:

Rubycon Black Gates have quite good performance(though still not as good as Nichicon Muse), as
do Elna Cerafine/Silmic caps, but both makers have a horrible reliability track record, at
least in my 30 years of experience. In all of that time, I can count on one hand, with fingers
left over, the number of Nichicon caps of ANY age or grade that I have had to replace, compared
to many hundreds(thousands, perhaps) of Elna & Rubycon caps. I would say that Rubycon in
particular is the single most UNreliable lytic maker in the world. Just ask any VCR tech.
Panasonics are very reliable & good performers, but a distant second on both counts at any
grade versus Nichicon.
And I do very much agree that even the best lytics need good film byassing for audio apps,
whether in the signal path or power supply. Considering the chemistry/physics of lytics, it's
a miracle they work at all.

Reliability is one issue for sure. If you want long term ultra reliability, don't use caps with an
electrolyte that can dry out.

I sometimes wonder about the condition of those NOS electros being sold on ebay to gullible fools
who think old parts are better than new.

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.


Graham

Stephen Sank wrote:

Rubycon Black Gates have quite good performance(though still not as good as Nichicon Muse), as
do Elna Cerafine/Silmic caps, but both makers have a horrible reliability track record, at
least in my 30 years of experience. In all of that time, I can count on one hand, with fingers
left over, the number of Nichicon caps of ANY age or grade that I have had to replace, compared
to many hundreds(thousands, perhaps) of Elna & Rubycon caps. I would say that Rubycon in
particular is the single most UNreliable lytic maker in the world. Just ask any VCR tech.
Panasonics are very reliable & good performers, but a distant second on both counts at any
grade versus Nichicon.
And I do very much agree that even the best lytics need good film byassing for audio apps,
whether in the signal path or power supply. Considering the chemistry/physics of lytics, it's
a miracle they work at all.

Reliability is one issue for sure. If you want long term ultra reliability, don't use caps with an
electrolyte that can dry out.

I sometimes wonder about the condition of those NOS electros being sold on ebay to gullible fools
who think old parts are better than new.

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.


Graham

"Pooh Bear"

I note an apparent reluctacnce to discuss a scientific reason for the
different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?


** .......... :-0 !!!!


I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.



** There is one, the original one that the practice has likely derived
om - tube radios have such bypassed electros in their PSUs. The armchair
experts there have assumed the use of 0.1uF paper caps in parallel with 8
uF, 350 volt electros was to counter the electros alleged "inductance".
But it was not.

It was there to counter RF instability due to the electros ESR at AM radio
frequencies.

The ESR of an old style, low cost 10 uF, 350 volt electro is 15 - 30 ohms -
paralleling a 0.1uF paper cap brought this down to 1 or 2 ohms.





.............. Phil

"Pooh Bear"

I note an apparent reluctacnce to discuss a scientific reason for the
different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?


** .......... :-0 !!!!


I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.



** There is one, the original one that the practice has likely derived
om - tube radios have such bypassed electros in their PSUs. The armchair
experts there have assumed the use of 0.1uF paper caps in parallel with 8
uF, 350 volt electros was to counter the electros alleged "inductance".
But it was not.

It was there to counter RF instability due to the electros ESR at AM radio
frequencies.

The ESR of an old style, low cost 10 uF, 350 volt electro is 15 - 30 ohms -
paralleling a 0.1uF paper cap brought this down to 1 or 2 ohms.





.............. Phil

I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.


Phil wrote:

** There is one, the original one that the practice has likely derived
om - tube radios have such bypassed electros in their PSUs. The armchair
experts there have assumed the use of 0.1uF paper caps in parallel with 8
uF, 350 volt electros was to counter the electros alleged "inductance".
But it was not.

It was there to counter RF instability due to the electros ESR at AM radio
frequencies.

The ESR of an old style, low cost 10 uF, 350 volt electro is 15 - 30 ohms -
paralleling a 0.1uF paper cap brought this down to 1 or 2 ohms.



This was what my dad told me as well, but even more applicable to the
broadcast transmitters and radar systems he was responsible for. In the
early '70's the electrolytics were still no where near as nice as they
are now, and the film bypass of larger 470uF or so was an audible
improvement.

Nowadays, thanks to the advances made to accomodate switching supplies
the caps have all improved rather dramatically, and I really don't hear
any significant difference with bypassing or different brands.

I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.


Phil wrote:

** There is one, the original one that the practice has likely derived
om - tube radios have such bypassed electros in their PSUs. The armchair
experts there have assumed the use of 0.1uF paper caps in parallel with 8
uF, 350 volt electros was to counter the electros alleged "inductance".
But it was not.

It was there to counter RF instability due to the electros ESR at AM radio
frequencies.

The ESR of an old style, low cost 10 uF, 350 volt electro is 15 - 30 ohms -
paralleling a 0.1uF paper cap brought this down to 1 or 2 ohms.



This was what my dad told me as well, but even more applicable to the
broadcast transmitters and radar systems he was responsible for. In the
early '70's the electrolytics were still no where near as nice as they
are now, and the film bypass of larger 470uF or so was an audible
improvement.

Nowadays, thanks to the advances made to accomodate switching supplies
the caps have all improved rather dramatically, and I really don't hear
any significant difference with bypassing or different brands.

Pooh Bear wrote:

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'm not willing to say it's rubbish. And I might be willing to believe it
has something to do with rectification effects. I have not seen any good
measurements on this, in either direction.

I _do_ know that if there is too low a DC bias on an electrolytic capacitor,
there are extreme low-level nonlinearities. These are very audible, and
the normal solution, of course, is to make sure the voltage across the cap
is never even close to the zero-crossing. The question is whether any of
these nonlinearities exist at higher levels, and I can't answer that.

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.

If the issue is high order harmonics being generated by small rectification
effects, a bypass capacitor will clean that up.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Pooh Bear wrote:

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'm not willing to say it's rubbish. And I might be willing to believe it
has something to do with rectification effects. I have not seen any good
measurements on this, in either direction.

I _do_ know that if there is too low a DC bias on an electrolytic capacitor,
there are extreme low-level nonlinearities. These are very audible, and
the normal solution, of course, is to make sure the voltage across the cap
is never even close to the zero-crossing. The question is whether any of
these nonlinearities exist at higher levels, and I can't answer that.

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.

If the issue is high order harmonics being generated by small rectification
effects, a bypass capacitor will clean that up.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

"Scott Dorsey"
Pooh Bear

I note an apparent reluctacnce to discuss a scientific reason for the
different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'm not willing to say it's rubbish. And I might be willing to believe it
has something to do with rectification effects. I have not seen any good
measurements on this, in either direction.


** Dorsey is a life long proponent of anti-science - the notion that
ignorance is knowledge.


I _do_ know that if there is too low a DC bias on an electrolytic
capacitor,
there are extreme low-level nonlinearities.


** Gobbledegook.

This is just like Dorsey falsely saying he *KNEW * that the LM301 has
crossover distortion when it dos not.

Where DOES this colossal fool drag all this misinformation up from ??



I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.

If the issue is high order harmonics being generated by small
rectification
effects, a bypass capacitor will clean that up.


** What a load of absolute bull crap - the bypass cap would need to be
larger in value than the one with the non-linearity to have any benefit.

Back to your rubber bands tape recorder mechanic.




.............. Phil

"Scott Dorsey"
Pooh Bear

I note an apparent reluctacnce to discuss a scientific reason for the
different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'm not willing to say it's rubbish. And I might be willing to believe it
has something to do with rectification effects. I have not seen any good
measurements on this, in either direction.


** Dorsey is a life long proponent of anti-science - the notion that
ignorance is knowledge.


I _do_ know that if there is too low a DC bias on an electrolytic
capacitor,
there are extreme low-level nonlinearities.


** Gobbledegook.

This is just like Dorsey falsely saying he *KNEW * that the LM301 has
crossover distortion when it dos not.

Where DOES this colossal fool drag all this misinformation up from ??



I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.

If the issue is high order harmonics being generated by small
rectification
effects, a bypass capacitor will clean that up.


** What a load of absolute bull crap - the bypass cap would need to be
larger in value than the one with the non-linearity to have any benefit.

Back to your rubber bands tape recorder mechanic.




.............. Phil

"Dan Kennedy"


I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.


Phil wrote:

** There is one, the original one that the practice has likely derived
om - tube radios have such bypassed electros in their PSUs. The
armchair experts there have assumed the use of 0.1uF paper caps in
parallel with 8 uF, 350 volt electros was to counter the electros
alleged "inductance". But it was not.

It was there to counter RF instability due to the electros ESR at AM
radio frequencies.

The ESR of an old style, low cost 10 uF, 350 volt electro is 15 - 30
hms - paralleling a 0.1uF paper cap brought this down to 1 or 2 ohms.



This was what my dad told me as well, but even more applicable to the
broadcast transmitters and radar systems he was responsible for. In the
early '70's the electrolytics were still no where near as nice as they are
now, and the film bypass of larger 470uF or so was an audible
improvement.



** I have seen many electros that were made in the 1960s of between 20 and
100uF @ 500volts ( in old Fender amps etc ) and their ESR values, even now,
are not more than 1 or 2 ohms. Most modern equivalents are much the same.


Nowadays, thanks to the advances made to accomodate switching supplies
the caps have all improved rather dramatically,


** The electrolyte formulation has improved to give far higher
onductivity - so permitting lower ESR values and smaller packages.


and I really don't hear
any significant difference with bypassing or different brands.



** Correct.



............... Phil

"Dan Kennedy"


I'd like to see a rational reason posted for bypassing electros with film
caps in coupling
applications though.


Phil wrote:

** There is one, the original one that the practice has likely derived
om - tube radios have such bypassed electros in their PSUs. The
armchair experts there have assumed the use of 0.1uF paper caps in
parallel with 8 uF, 350 volt electros was to counter the electros
alleged "inductance". But it was not.

It was there to counter RF instability due to the electros ESR at AM
radio frequencies.

The ESR of an old style, low cost 10 uF, 350 volt electro is 15 - 30
hms - paralleling a 0.1uF paper cap brought this down to 1 or 2 ohms.



This was what my dad told me as well, but even more applicable to the
broadcast transmitters and radar systems he was responsible for. In the
early '70's the electrolytics were still no where near as nice as they are
now, and the film bypass of larger 470uF or so was an audible
improvement.



** I have seen many electros that were made in the 1960s of between 20 and
100uF @ 500volts ( in old Fender amps etc ) and their ESR values, even now,
are not more than 1 or 2 ohms. Most modern equivalents are much the same.


Nowadays, thanks to the advances made to accomodate switching supplies
the caps have all improved rather dramatically,


** The electrolyte formulation has improved to give far higher
onductivity - so permitting lower ESR values and smaller packages.


and I really don't hear
any significant difference with bypassing or different brands.



** Correct.



............... Phil

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors. snip

Hobbyist-grade mag "Audio" did a fairly good article on this many
moons ago, but concentrated more on the TYPE of dielectric, rather
than descend into grading various brands. Electrolytics are crap in
the audio path...period. Micas are an improvement over 'lytics, but
have their own distortion problems. The winners? Mylar and
polypropylene, hands down.

The transfer characteristics of a capacitor are just as important in
an audio chain as those of an inductive device. To put it more
simply, certain types of dielectrics add more distortion to an AC
signal in its passband than do others. Aluminum and tantalum
electrolytics are notorious in this regard, as are old ceramic disks.
The "audiophool" community also seems to have glommed onto supposed
"benefits" of archaic paper caps, where none exists.

I was intrigued enough by this article to start doing some studies
with what bench gear I had at the time, and sure enough...my results
turned out identical to those in the article, with polyprops
distorting less than any other dielectric type tested. Just for
funsies, I took a new Mexican electrolytic and tested it vis à vis an
ancient Mallory I had in a junk box. The newere cap was far better,
but not just on distortion. ESR played a key role here, too, as the
aged cap was obviously suffering from effects of age, and high ESR's
one of the first parameters to go awry on old 'lytics.

Might that be because it's simply rubbish ? snip

I goofed around with mylar caps from three different manufacturers,
and got identical results. So much for "branding!" Same results were
gotten from three different polyprops and three different aluminum
electrolytics.

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though. snip

See above. However, you don't always (almost never?) have the luxury
of unimpeded real estate when changing out a 'lytic for a film cap,
especially when dealing with B+ voltages. Those 'lytics are there for
a reason far simpler than distortion characteristics...they're the
only ones that'll fit!

dB

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors. snip

Hobbyist-grade mag "Audio" did a fairly good article on this many
moons ago, but concentrated more on the TYPE of dielectric, rather
than descend into grading various brands. Electrolytics are crap in
the audio path...period. Micas are an improvement over 'lytics, but
have their own distortion problems. The winners? Mylar and
polypropylene, hands down.

The transfer characteristics of a capacitor are just as important in
an audio chain as those of an inductive device. To put it more
simply, certain types of dielectrics add more distortion to an AC
signal in its passband than do others. Aluminum and tantalum
electrolytics are notorious in this regard, as are old ceramic disks.
The "audiophool" community also seems to have glommed onto supposed
"benefits" of archaic paper caps, where none exists.

I was intrigued enough by this article to start doing some studies
with what bench gear I had at the time, and sure enough...my results
turned out identical to those in the article, with polyprops
distorting less than any other dielectric type tested. Just for
funsies, I took a new Mexican electrolytic and tested it vis à vis an
ancient Mallory I had in a junk box. The newere cap was far better,
but not just on distortion. ESR played a key role here, too, as the
aged cap was obviously suffering from effects of age, and high ESR's
one of the first parameters to go awry on old 'lytics.

Might that be because it's simply rubbish ? snip

I goofed around with mylar caps from three different manufacturers,
and got identical results. So much for "branding!" Same results were
gotten from three different polyprops and three different aluminum
electrolytics.

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though. snip

See above. However, you don't always (almost never?) have the luxury
of unimpeded real estate when changing out a 'lytic for a film cap,
especially when dealing with B+ voltages. Those 'lytics are there for
a reason far simpler than distortion characteristics...they're the
only ones that'll fit!

dB

"DeserTBoB" = Bob the Tosser


I note an apparent reluctacnce to discuss a scientific reason for the
different alleged *sound* of
various brands of capacitors. snip

Hobbyist-grade mag "Audio" did a fairly good article on this many
moons ago,


** A massive load of pseudo-scientific garbage.



but concentrated more on the TYPE of dielectric, rather
than descend into grading various brands. Electrolytics are crap in
the audio path...period.

** A massive load of pseudo-scientific garbage.


Micas are an improvement over 'lytics, but
have their own distortion problems. The winners? Mylar and
polypropylene, hands down.


** A massive load of pseudo-scientific garbage.



The transfer characteristics of a capacitor are just as important in
an audio chain as those of an inductive device.


** A massive load of pseudo-scientific garbage.


To put it more
simply, certain types of dielectrics add more distortion to an AC
signal in its passband than do others.


** A massive load of pseudo-scientific garbage.



I was intrigued enough by this article to start doing some studies
with what bench gear I had at the time, and sure enough...my results
turned out identical to those in the article, with polyprops
distorting less than any other dielectric type tested.


** There is no distortion in normal usage.

Your tests were obviously stupid and false - as is proved by your NOT
detailing them.





................. Phil

"DeserTBoB" = Bob the Tosser


I note an apparent reluctacnce to discuss a scientific reason for the
different alleged *sound* of
various brands of capacitors. snip

Hobbyist-grade mag "Audio" did a fairly good article on this many
moons ago,


** A massive load of pseudo-scientific garbage.



but concentrated more on the TYPE of dielectric, rather
than descend into grading various brands. Electrolytics are crap in
the audio path...period.

** A massive load of pseudo-scientific garbage.


Micas are an improvement over 'lytics, but
have their own distortion problems. The winners? Mylar and
polypropylene, hands down.


** A massive load of pseudo-scientific garbage.



The transfer characteristics of a capacitor are just as important in
an audio chain as those of an inductive device.


** A massive load of pseudo-scientific garbage.


To put it more
simply, certain types of dielectrics add more distortion to an AC
signal in its passband than do others.


** A massive load of pseudo-scientific garbage.



I was intrigued enough by this article to start doing some studies
with what bench gear I had at the time, and sure enough...my results
turned out identical to those in the article, with polyprops
distorting less than any other dielectric type tested.


** There is no distortion in normal usage.

Your tests were obviously stupid and false - as is proved by your NOT
detailing them.





................. Phil

DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Strange how those nice new reservoir caps *don't* go bang when you switch on a power supply for the
first time too !


I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors. snip

Hobbyist-grade mag "Audio" did a fairly good article on this many
moons ago, but concentrated more on the TYPE of dielectric, rather
than descend into grading various brands. Electrolytics are crap in
the audio path...period. Micas are an improvement over 'lytics, but
have their own distortion problems. The winners? Mylar and
polypropylene, hands down.

Strange that almost every single high end recording console uses electrolytics for coupling then ?

I'll elaborate on this later.

Mica would be daft choice for a coupling cap. I've only ever seen it used for picofarad type values.


The transfer characteristics of a capacitor are just as important in
an audio chain as those of an inductive device. To put it more
simply, certain types of dielectrics add more distortion to an AC
signal in its passband than do others. Aluminum and tantalum
electrolytics are notorious in this regard, as are old ceramic disks.
The "audiophool" community also seems to have glommed onto supposed
"benefits" of archaic paper caps, where none exists.

Agreed that paper caps are rubbish except in line filtering applications where the dielectric has some
spinoff advantages ( better self-healing properties ).

The audiophools simply appear to want to replicate the sound of ancient valve amps by including ancient
rubbish caps too it seems. :-)

All ceramics aren't bad btw. The ones with low-K dielectrics like NPO don't suffer the dielectric
non-linearity You'd only use them for feedback or small value caps in EQ sections though. Not good for
any more than 330-470 pF in practice..

The 'rectification effect' in electrolytics only appears AFAIK when a reverse voltage of around 0.5 V
is applied. Keep the reverse volts due to an a.c. signal below that and it doesn't happen. That
*doesn't* mean restricting the a.c. signal value to below 0.5 V though. Think 'potential divider' and
use large value caps. I commonly use 100uF for outputs and 10uF for inputs where Zin is = 10k ohm.


I was intrigued enough by this article to start doing some studies
with what bench gear I had at the time, and sure enough...my results
turned out identical to those in the article, with polyprops
distorting less than any other dielectric type tested.

How were you measuring distortion ?

How can you explain ployester/mylar 'distorting' ? I find this puzzling for any plastic film. I never
saw a polyester cap distort when measured by an AP test set.


Just for
funsies, I took a new Mexican electrolytic and tested it vis à vis an
ancient Mallory I had in a junk box. The newere cap was far better,
but not just on distortion. ESR played a key role here, too, as the
aged cap was obviously suffering from effects of age, and high ESR's
one of the first parameters to go awry on old 'lytics.

Might that be because it's simply rubbish ? snip

I goofed around with mylar caps from three different manufacturers,
and got identical results. So much for "branding!" Same results were
gotten from three different polyprops and three different aluminum
electrolytics.

That'll be because the dielectric is what it says it is. Branding can't change the properties of
plastic films.


I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though. snip

See above. However, you don't always (almost never?) have the luxury
of unimpeded real estate when changing out a 'lytic for a film cap,
especially when dealing with B+ voltages. Those 'lytics are there for
a reason far simpler than distortion characteristics...they're the
only ones that'll fit!

That's because low noise ( low impedance ) circuitry and low cut-off frequencies require large caps.
It's not possible to use 100uF film types either practically ( size wise ) or remotely economically.

B+ isn't an issue for designing with dual supply op-amp circuitry though.

What value electros did you test btw ? I'll guess *not* 100uF into a 10k load.


Graham

DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Strange how those nice new reservoir caps *don't* go bang when you switch on a power supply for the
first time too !


I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors. snip

Hobbyist-grade mag "Audio" did a fairly good article on this many
moons ago, but concentrated more on the TYPE of dielectric, rather
than descend into grading various brands. Electrolytics are crap in
the audio path...period. Micas are an improvement over 'lytics, but
have their own distortion problems. The winners? Mylar and
polypropylene, hands down.

Strange that almost every single high end recording console uses electrolytics for coupling then ?

I'll elaborate on this later.

Mica would be daft choice for a coupling cap. I've only ever seen it used for picofarad type values.


The transfer characteristics of a capacitor are just as important in
an audio chain as those of an inductive device. To put it more
simply, certain types of dielectrics add more distortion to an AC
signal in its passband than do others. Aluminum and tantalum
electrolytics are notorious in this regard, as are old ceramic disks.
The "audiophool" community also seems to have glommed onto supposed
"benefits" of archaic paper caps, where none exists.

Agreed that paper caps are rubbish except in line filtering applications where the dielectric has some
spinoff advantages ( better self-healing properties ).

The audiophools simply appear to want to replicate the sound of ancient valve amps by including ancient
rubbish caps too it seems. :-)

All ceramics aren't bad btw. The ones with low-K dielectrics like NPO don't suffer the dielectric
non-linearity You'd only use them for feedback or small value caps in EQ sections though. Not good for
any more than 330-470 pF in practice..

The 'rectification effect' in electrolytics only appears AFAIK when a reverse voltage of around 0.5 V
is applied. Keep the reverse volts due to an a.c. signal below that and it doesn't happen. That
*doesn't* mean restricting the a.c. signal value to below 0.5 V though. Think 'potential divider' and
use large value caps. I commonly use 100uF for outputs and 10uF for inputs where Zin is = 10k ohm.


I was intrigued enough by this article to start doing some studies
with what bench gear I had at the time, and sure enough...my results
turned out identical to those in the article, with polyprops
distorting less than any other dielectric type tested.

How were you measuring distortion ?

How can you explain ployester/mylar 'distorting' ? I find this puzzling for any plastic film. I never
saw a polyester cap distort when measured by an AP test set.


Just for
funsies, I took a new Mexican electrolytic and tested it vis à vis an
ancient Mallory I had in a junk box. The newere cap was far better,
but not just on distortion. ESR played a key role here, too, as the
aged cap was obviously suffering from effects of age, and high ESR's
one of the first parameters to go awry on old 'lytics.

Might that be because it's simply rubbish ? snip

I goofed around with mylar caps from three different manufacturers,
and got identical results. So much for "branding!" Same results were
gotten from three different polyprops and three different aluminum
electrolytics.

That'll be because the dielectric is what it says it is. Branding can't change the properties of
plastic films.


I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though. snip

See above. However, you don't always (almost never?) have the luxury
of unimpeded real estate when changing out a 'lytic for a film cap,
especially when dealing with B+ voltages. Those 'lytics are there for
a reason far simpler than distortion characteristics...they're the
only ones that'll fit!

That's because low noise ( low impedance ) circuitry and low cut-off frequencies require large caps.
It's not possible to use 100uF film types either practically ( size wise ) or remotely economically.

B+ isn't an issue for designing with dual supply op-amp circuitry though.

What value electros did you test btw ? I'll guess *not* 100uF into a 10k load.


Graham

Scott Dorsey wrote:

Pooh Bear wrote:

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'm not willing to say it's rubbish. And I might be willing to believe it
has something to do with rectification effects. I have not seen any good
measurements on this, in either direction.

Me neither. I'm tempted to have a look though !


I _do_ know that if there is too low a DC bias on an electrolytic capacitor,
there are extreme low-level nonlinearities. These are very audible, and
the normal solution, of course, is to make sure the voltage across the cap
is never even close to the zero-crossing. The question is whether any of
these nonlinearities exist at higher levels, and I can't answer that.


My understanding is that any rectification effects take place when an electrolytic has 0.5 reverse V
approx applied. It's easy to ensure this never happens in a coupling cap application - not least by
the use of large values since the component of the signal appearing across the cap will be in the
millivolt area.

Seems to work for the likes of Neve and SSL to mention a couple of high end names using electros with
zero bias. Not to mention almost all pro-audio gear made.


I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.

If the issue is high order harmonics being generated by small rectification
effects, a bypass capacitor will clean that up.

Suggested value to bypass 100uF ?


Graham

Scott Dorsey wrote:

Pooh Bear wrote:

I note an apparent reluctacnce to discuss a scientific reason for the different alleged *sound* of
various brands of capacitors.

Might that be because it's simply rubbish ?

I'm not willing to say it's rubbish. And I might be willing to believe it
has something to do with rectification effects. I have not seen any good
measurements on this, in either direction.

Me neither. I'm tempted to have a look though !


I _do_ know that if there is too low a DC bias on an electrolytic capacitor,
there are extreme low-level nonlinearities. These are very audible, and
the normal solution, of course, is to make sure the voltage across the cap
is never even close to the zero-crossing. The question is whether any of
these nonlinearities exist at higher levels, and I can't answer that.


My understanding is that any rectification effects take place when an electrolytic has 0.5 reverse V
approx applied. It's easy to ensure this never happens in a coupling cap application - not least by
the use of large values since the component of the signal appearing across the cap will be in the
millivolt area.

Seems to work for the likes of Neve and SSL to mention a couple of high end names using electros with
zero bias. Not to mention almost all pro-audio gear made.


I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.

If the issue is high order harmonics being generated by small rectification
effects, a bypass capacitor will clean that up.

Suggested value to bypass 100uF ?


Graham

"Pooh Bear" wrote in message
...
DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold
on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Plenty, if the coupling caps are used on a single-supply circuit.

Peace,
Paul

"Pooh Bear" wrote in message
...
DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold
on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Plenty, if the coupling caps are used on a single-supply circuit.

Peace,
Paul

Paul Stamler wrote:

"Pooh Bear" wrote in message
...
DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold
on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Plenty, if the coupling caps are used on a single-supply circuit.

I think the reference was to using them as supply 'bypass caps'.


Graham

Paul Stamler wrote:

"Pooh Bear" wrote in message
...
DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold
on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Plenty, if the coupling caps are used on a single-supply circuit.

I think the reference was to using them as supply 'bypass caps'.


Graham

Pooh Bear wrote:

Strange that almost every single high end recording console uses electrolytics for coupling then ?

Because in the solid-state world there isn't much choice. And one of the
big problems with the early solid-state world is that people adopted the
usual single-ended capacitively-coupled designs that were common with tubes.
Since input impedances were so low, massive caps were required, and so of
course electrolytics turn up.

If somebody made a film cap that was almost as small as an electrolytic and
in the same price range, I don't think you'd see any high end recording
consoles still using electrolytics.

All ceramics aren't bad btw. The ones with low-K dielectrics like NPO don't suffer the dielectric
non-linearity You'd only use them for feedback or small value caps in EQ sections though. Not good for
any more than 330-470 pF in practice..

I really want to know more about ceramic caps... I used to have a strong
anti-ceramic bias until I tried some of the newer COG types which are less
microphonic by a long shot. I want to thank John Hardy for turning me on
to some of the better quality ceramics today. I have used them as coupling
caps in very high-Z circuits and have actually been pleased with the
performance compared with most of the film caps.

The 'rectification effect' in electrolytics only appears AFAIK when a reverse voltage of around 0.5 V
is applied. Keep the reverse volts due to an a.c. signal below that and it doesn't happen. That
*doesn't* mean restricting the a.c. signal value to below 0.5 V though. Think 'potential divider' and
use large value caps. I commonly use 100uF for outputs and 10uF for inputs where Zin is = 10k ohm.

Right. Our question is whether there might be some other nonlinear effects
in addition to this one.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Pooh Bear wrote:

Strange that almost every single high end recording console uses electrolytics for coupling then ?

Because in the solid-state world there isn't much choice. And one of the
big problems with the early solid-state world is that people adopted the
usual single-ended capacitively-coupled designs that were common with tubes.
Since input impedances were so low, massive caps were required, and so of
course electrolytics turn up.

If somebody made a film cap that was almost as small as an electrolytic and
in the same price range, I don't think you'd see any high end recording
consoles still using electrolytics.

All ceramics aren't bad btw. The ones with low-K dielectrics like NPO don't suffer the dielectric
non-linearity You'd only use them for feedback or small value caps in EQ sections though. Not good for
any more than 330-470 pF in practice..

I really want to know more about ceramic caps... I used to have a strong
anti-ceramic bias until I tried some of the newer COG types which are less
microphonic by a long shot. I want to thank John Hardy for turning me on
to some of the better quality ceramics today. I have used them as coupling
caps in very high-Z circuits and have actually been pleased with the
performance compared with most of the film caps.

The 'rectification effect' in electrolytics only appears AFAIK when a reverse voltage of around 0.5 V
is applied. Keep the reverse volts due to an a.c. signal below that and it doesn't happen. That
*doesn't* mean restricting the a.c. signal value to below 0.5 V though. Think 'potential divider' and
use large value caps. I commonly use 100uF for outputs and 10uF for inputs where Zin is = 10k ohm.

Right. Our question is whether there might be some other nonlinear effects
in addition to this one.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Pooh Bear wrote:
Scott Dorsey wrote:

My understanding is that any rectification effects take place when an electrolytic has 0.5 reverse V
approx applied. It's easy to ensure this never happens in a coupling cap application - not least by
the use of large values since the component of the signal appearing across the cap will be in the
millivolt area.

I will buy that.

Seems to work for the likes of Neve and SSL to mention a couple of high end names using electros with
zero bias. Not to mention almost all pro-audio gear made.

Yes, agreed. But, I once changed the tantalums in a Neve channel strip out
for film caps, with large enough values to get good low end. Sounded very
clean and very nice to me. Everybody else in the studio also heard a change
in sound, and they all hated it and I was almost fired.

Clearly there was _something_ changing the sound about the tantalums, even
though they were all carefully biased.

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.

If the issue is high order harmonics being generated by small rectification
effects, a bypass capacitor will clean that up.

Suggested value to bypass 100uF ?

I think the tradition is to use a bypass cap that is about 1/100th the value
of the electrolytic as a rule of thumb. I am not sure where that came from
or how it was derived.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Pooh Bear wrote:
Scott Dorsey wrote:

My understanding is that any rectification effects take place when an electrolytic has 0.5 reverse V
approx applied. It's easy to ensure this never happens in a coupling cap application - not least by
the use of large values since the component of the signal appearing across the cap will be in the
millivolt area.

I will buy that.

Seems to work for the likes of Neve and SSL to mention a couple of high end names using electros with
zero bias. Not to mention almost all pro-audio gear made.

Yes, agreed. But, I once changed the tantalums in a Neve channel strip out
for film caps, with large enough values to get good low end. Sounded very
clean and very nice to me. Everybody else in the studio also heard a change
in sound, and they all hated it and I was almost fired.

Clearly there was _something_ changing the sound about the tantalums, even
though they were all carefully biased.

I'd like to see a rational reason posted for bypassing electros with film caps in coupling
applications though.

If the issue is high order harmonics being generated by small rectification
effects, a bypass capacitor will clean that up.

Suggested value to bypass 100uF ?

I think the tradition is to use a bypass cap that is about 1/100th the value
of the electrolytic as a rule of thumb. I am not sure where that came from
or how it was derived.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Pooh Bear wrote:
Paul Stamler wrote:

"Pooh Bear" wrote in message
...
DeserTBoB wrote:

On Thu, 02 Dec 2004 04:53:12 +0000, Pooh Bear
wrote:

I sometimes wonder about the condition of those NOS electros being sold
on ebay to gullible fools
who think old parts are better than new. snip

No need to wonder...they're "unformed" and instantly turn into a bomb
when B+ is applied.

What has that to do with *coupling* caps ?

Plenty, if the coupling caps are used on a single-supply circuit.

I think the reference was to using them as supply 'bypass caps'.

As Paul points out, a supply bypass cap is directly in the audio path in
a single-ended circuit, and could arguably be considered a an audio coupling
cap if you wanted to get pedantic about it.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."