NP0 ceramic caps in signal path, are they good enough?
sean broderick
feedback paths). Could I expect much improvement by replacing them with
polystyrene/polypropylene types, or are they good? The values are very small (100-400
pF). Thanks.
Sean B
Mark McQuilken
Although ceramics exhibit significant distortion to audio signals, if
they're not used as interstage coupling (or lead/lag xfmr terminations),
there shouldn't be an issue. If the values are as small as you say, then
they're probably being used as feedback compensation at fairly high
frequencies. Since the distortion products are multiples of the
fundamental (as the capacitor becomes the dominent element) and
high-frequency roll-offs are typically fairly high (>50kHz), the
distortion product will also be fairly high-frequency and beyond the
usual 20-20kHz passband of interest...
McQ
__
Mark McQuilken
FMR Audio
www.fmraudio.com
(800)343-9976 - US Only
(512)280-6557 Voice
(512)280-8627 Fax
Scott Dorsey
sean broderick <seanbr...@uswest.net> wrote:
>I bought an 8 channel converter, and it has these type caps in the analog signal path (and
>feedback paths). Could I expect much improvement by replacing them with
>polystyrene/polypropylene types, or are they good? The values are very small (100-400
>pF). Thanks.
Ceramic caps are bad, because they can introduce all kinds of interesting
distortions. They are basically piezoelectric transducers.
However, if they are such small values, they really aren't going to do much
in the audio band. They sound like they are decoupling caps intended to keep
circuits stable at high frequencies.
If you put film caps in there, you may find you have a wild, oscillating mess,
since film caps don't tend to work well at high frequencies. They turn into
inductors and cease being capacitors.
If you want to replace them with something with higher performance, you can
try silver micas, but I suspect you won't hear a difference.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
John W. Hardy
There is a common misconception regarding ceramic
capacitors, which is: ALL ceramic capacitors are bad
for audio.
NOT TRUE!
There are three basic types of ceramic capacitors,
COG/NP0, X7R and Z5U (there are other types, but these
are the main ones). Each type uses a different
formulation of ceramic material for the dielectric (the
insulator between the plates of the capacitor). The
COG/NP0 caps use a pure ceramic material, and they are
excellent for audio circuits. They are NOT
piezoelectric.
The other types of ceramic capacitors add various
ferroelectric materials to the ceramic in order to
achieve higher volumetric efficiency (higher
capacitance per cubic inch), change the temperature
coefficient to temperature compensate a critical
circuit, or lower the cost. It is these added materials
that cause the X7R, Z5U and others to be ferroelectric.
It also causes them to actually change value as the
voltage that is applied to the capacitor changes. In
other words, the actual audio signal can modulate the
capacitance, which can increase the distortion when
such capacitors are used in EQ circuits, crossover
circuits, or any circuit where the audio signal
frequency is near the "-3dB" point.
Specifically, a JOHANSON DIELECTRICS brochure states:
"Class I dielectrics display electrostiction, but, are
not piezoelectric, as the crystal lattice of the
ceramic in these dielectrics possesses a center of
symmetry." They are referring to the COG/NP0
dielectrics. A Z5U capacitor would be fine for the
bypassing of power supplies in some types of equipment,
but would be terrible in the RC network of an
equalizer, RIAA network, etc. The COG/NP0 capacitor is
an excellent choice for audio, particularly when very
small values (<1000pF) are required, or a physically
small package is required. Surface mount versions are
now available in .062"x.031" sizes, with .040"x.020"
and smaller coming soon.
I have been using COG/NP0 caps in the 990 op-amps that
I have been manufacturing since 1979. I also use these
caps in all of my mic preamps (M-1, M-2, Jensen Twin
Servo 990 Mic Preamp and others) in the feedback loops
of the 990 op-amps, as part of the termination network
on the secondary winding of the JT-16-B mic-input
transformer, etc.
I wrote a paper entitled "Ceramic Capacitors" in 1980
to explain the advantages of the COG/NP0 caps. If
anyone wants a copy, please e-mail me privately. Thank
you.
John Hardy
David Josephson
ones such as the X and Y prefixed types are awful for audio (including
bypass), because they change value with applied voltage. NPO types
however, are stable, have very low dielectric absorption, and are
generally quite OK up to about 500 pF. So in the range you specify,
leave them alone. Dorsey's point about inductance is absolutely important
too. They made the circuit work with those caps, it might not be stable
with units having higher inductance.
--
David Josephson / Josephson Engineering / San Jose CA / da...@josephson.com
Dan Kennedy
I don't think you'll hear any difference with polywhatevers. The
size differences may make more problems than the potential quality
improvement is worth. If they were coupling caps and about two orders
of magnitude bigger I'd feel differently.
Dan Kennedy
Great River Electronics Inc.
sean broderick wrote:
>
> I bought an 8 channel converter, and it has these type caps in the analog signal path (and
> feedback paths). Could I expect much improvement by replacing them with
> polystyrene/polypropylene types, or are they good? The values are very small (100-400
> pF). Thanks.
>
> Sean B
Anonymous
Thank you, Mr Hardy! I did not know that, and worse, didn't know that
I didn't know it, but it is something I really needed to know,
--
the artist formerly known as Bill
--------== Posted Anonymously via Newsfeeds.Com ==-------
Featuring the worlds only Anonymous Usenet Server
-----------== http://www.newsfeeds.com ==----------
John W. Hardy
> Thank you, Mr Hardy! I did not know that, and worse, didn't know that
> I didn't know it, but it is something I really needed to know,
You are welcome. I would like to bury that
misconception, once and for all!
BTW, let me correct myself. I said "It is these added
materials that cause the X7R, Z5U and others to be
ferroelectric." I MEANT to say "piezoelectric", not
"ferroelectric". Thank you.
John Hardy
Scott Dorsey
>Anonymous;
>
>> Thank you, Mr Hardy! I did not know that, and worse, didn't know that
>> I didn't know it, but it is something I really needed to know,
>
>You are welcome. I would like to bury that
>misconception, once and for all!
Well, I will reply that the NP0 dielectric, although it's a whole lot
better than most of the ceramics out there, still has some residual
effect. Put an NP0 cap into a high-Z input and tap it, and you will
hear microphonics.
I didn't believe this was a problem until I started replacing NP0
coupling caps on microphones with the AVX glass capacitors and heard
an audible difference.
(Admittedly if you do this with a lot of film caps you can also hear
microphonics, but that's a different mechanism. Just as bad, though.)
sean broderick
which is nice to see.
Sean B
David Josephson
>> Thank you, Mr Hardy! I did not know that, and worse, didn't know that
>> I didn't know it, but it is something I really needed to know,
>You are welcome. I would like to bury that
>misconception, once and for all!
Me too, there are enough already in this "business"
>BTW, let me correct myself. I said "It is these added
>materials that cause the X7R, Z5U and others to be
>ferroelectric." I MEANT to say "piezoelectric", not
>"ferroelectric". Thank you.
With that correction, I agree completely with what John said. And the
reason piezoelectric capacitors are bad, is that they change value with
applied voltage -- built-in distortion generator.
David Josephson
>Well, I will reply that the NP0 dielectric, although it's a whole lot
>better than most of the ceramics out there, still has some residual
>effect. Put an NP0 cap into a high-Z input and tap it, and you will
>hear microphonics.
True, but so will a polypropylene, and it won't have any such
microphonics unless it has a voltage across it. What you are hearing
is the capacitor acting as a condenser microphone. Many mic makers embed
their NP0 chip caps in a very hard resin to (a) keep the moisture out
and (b) keep the microphonics down. I still use film caps but have to
anchor them too.
>I didn't believe this was a problem until I started replacing NP0
>coupling caps on microphones with the AVX glass capacitors and heard
>an audible difference.
Much less ability to transfer acoustic energy to the dielectric.
>(Admittedly if you do this with a lot of film caps you can also hear
>microphonics, but that's a different mechanism. Just as bad, though.)
I think it's the same mechanism. But I might be wrong. Try it without
bias and see if it still does it.
--
David Josephson
Scott Dorsey
>>(Admittedly if you do this with a lot of film caps you can also hear
>>microphonics, but that's a different mechanism. Just as bad, though.)
>
>I think it's the same mechanism. But I might be wrong. Try it without
>bias and see if it still does it.
You can still hear film caps being microphonic without bias. I think
that is as much triboelectric noise as actual capacitance change.
David Josephson
>David Josephson <dav...@rahul.net> wrote:
>>>(Admittedly if you do this with a lot of film caps you can also hear
>>>microphonics, but that's a different mechanism. Just as bad, though.)
>>
>>I think it's the same mechanism. But I might be wrong. Try it without
>>bias and see if it still does it.
>You can still hear film caps being microphonic without bias. I think
>that is as much triboelectric noise as actual capacitance change.
That might be, if they are teflon, but I don't think any of the other
resins used for film caps are triboelectric.
David L. Rick
klu...@netcom.com (Scott Dorsey) wrote:
> Well, I will reply that the NP0 dielectric, although it's a whole lot
> better than most of the ceramics out there, still has some residual
> effect. Put an NP0 cap into a high-Z input and tap it, and you will
> hear microphonics.
>
> I didn't believe this was a problem until I started replacing NP0
> coupling caps on microphones with the AVX glass capacitors and heard
> an audible difference.
Which AVX cap is this, Scott? I have the AVX catalog here, and I
don't see anything that looks likely.
David L. Rick
Seventh String Recording
dr...@hach.com
Sent via Deja.com http://www.deja.com/
Before you buy.
Scott Dorsey
>
>That might be, if they are teflon, but I don't think any of the other
>resins used for film caps are triboelectric.
Well, have you got a better explanation for the effect? Try it with
a cap attached across a high-Z preamp input and see. I can hear
things! I'm grasping for explanations here, so if you can do better
I'd greatly appreciate it.
Scott Dorsey
David L. Rick <dr...@hach.com> wrote:
>In article <89rapm$gpc$1...@nntp8.atl.mindspring.net>,
> klu...@netcom.com (Scott Dorsey) wrote:
>
>> Well, I will reply that the NP0 dielectric, although it's a whole lot
>> better than most of the ceramics out there, still has some residual
>> effect. Put an NP0 cap into a high-Z input and tap it, and you will
>> hear microphonics.
>>
>> I didn't believe this was a problem until I started replacing NP0
>> coupling caps on microphones with the AVX glass capacitors and heard
>> an audible difference.
>
>Which AVX cap is this, Scott? I have the AVX catalog here, and I
>don't see anything that looks likely.
The old Corning glass capacitor line got bought out by AVX a few years
ago. Call AVX and ask about them, because they may not be in the full
line catalogue. We use them in avionics and they are very stable and
quiet.
I have replaced the coupling caps on a lot of mikes and always noticed
an improvement. On the other hand, I replaced all the NPOs in my Citation II
amplifier and couldn't hear any difference at all.
David L. Rick
DR>>Which AVX cap is this, Scott? I have the AVX catalog here, and I
DR>>don't see anything that looks likely.
>
> The old Corning glass capacitor line got bought out by AVX a few years
> ago. Call AVX and ask about them, because they may not be in the full
> line catalogue. We use them in avionics and they are very stable and
> quiet.
Hmm. Now if someone would just put those Corning glass power
resistors back in production...
jay....@oceansonics.com
> since film caps don't tend to work well at high frequencies. They turn into
> inductors and cease being capacitors.
Scott Dorsey
>> If you put film caps in there, you may find you have a wild, oscillating =
>mess,=20
>> since film caps don't tend to work well at high frequencies. They turn i=
>nto=20
>ges with very low series inductance, though at the time of this original th=
>read they were only available in radial leaded packages. If using SMT pack=
>ages, beware -- these capacitors will melt when hand soldered without suffi=
>cient skill, and cannot be desoldered and re-used without destroying them. =
you see at 1 MHz on those things? Paktron keeps trying to get me to try them
but I haven't really yet.
JackA
> I bought an 8 channel converter, and it has these type caps in the analog signal path (and
> feedback paths). Could I expect much improvement by replacing them with
> polystyrene/polypropylene types, or are they good? The values are very small (100-400
> pF). Thanks.
>
> Sean B
Jack