Favorite small value blocking capacitors?

[Shattersignal]

In particular, I'm tinkering with condenser mic capsule coupling caps, which usually call for extremely small values on the order of 100pf-1000pf.

Here's my question to those who care to chime in: in a market rife with excellent foil capacitors, what are our best options for extremely small value caps?  What are your favorites?  There don't seem to be many options out there and I'd love to know which ones I may have missed.

All the other small value caps seem to be just variations on polystyrene, silver mica, or ceramic--all of which I doubt offer anything competitive with the copper and silver foil caps that I love.  I usually don't love the sound of polymer caps. 

Also, and here's a question specifically for the capacitor mic experts, what effect does the capsule coupling cap value have in the sound of the mic?  I'm working from the assumption that a lower value cap will be more transparent, all other things being equal.  This reasoning leads me to wonder whether a smaller valued, lower quality cap, might actually outperform a higher valued, higher quality cap in some situations--this matter I intend to test by comparing the xxx (1000pf) vs. the xxx (100pf), for instance. 

Thanks for sharing your thoughts and experiences!  Cheers. 

 

[JohnRoberts]

COG-NPO caps are considered linear and accurate.

Capacitors should not have a sound.

JR 

 

[CJ]

it the amp does not have these i won't go on.

 

[Shattersignal]

COG-NPO caps are considered linear and accurate.

Capacitors should not have a sound.

JR

Ture, indeed!  The best capacitors come closest to this ideal.  All capacitors have a sound, unfortunately.  The best cap is no cap at all.

 

[Shattersignal]

it the amp does not have these i won't go on.

Dang, CJ--you're hardcore. 

 

[Bo Deadly]

All capacitors have a sound, unfortunately.  The best cap is no cap at all.

You're not going to find a lot of folks hear talking about how capacitors and resistors and op amps "sound". Until someone can produce data that proves that changing a capacitor actually altered the signal in any meaningful way, such discussions are pointless.

 

[Shattersignal]

it the amp does not have these i won't go on.

Just bought some NOS Mallory Ceramics in 100pf.  What the hell, I'll try anything twice. 

 

[Shattersignal]

You're not going to find a lot of folks hear talking about how capacitors and resistors and op amps "sound". Until someone can produce data that proves that changing a capacitor actually altered the signal in any meaningful way, such discussions are pointless.

Hijack alert? ???

Just to be clear, I'm not trying to have that conversation--I'll save us both some dignity and not respond to your comment directly. 

There are tons of folks here who experiment with a variety of caps and who know the sonic differences.  I'd  love to think that this thread hasn't already been killed by hijacking.  I want to hear from you. 

 

[radardoug]

Have you done true double blind tests? Or is it that having paid a LOT of money for fancy caps, they MUST sound better!

 

[CJ]

you can use the scope to see things at upper frequencies,  in particular, you will sometimes see a small triangular peak attached to the crest of a sine wave sine wave in a Fender tone stack,  with some amps with certain ceramic caps this peak is not there,  i need to document  this the next time i see it so i can show you guys what i am talking about,

 

[Newmarket]

  I'm working from the assumption that a lower value cap will be more transparent, all other things being equal.  This reasoning leads me to wonder...

But why would you assume that. If anything the opposite would make more sense (as with electrolytics but I realise we are at the other end of the scale from that). A smaller cap' will have more voltage across it at a given frequency - since its impedance is greater. The actual 'numbers' depend on the whole circuit of course.
Didn't Cyril Bateman nail this some years ago ?

 

[JohnRoberts]

But why would you assume that. If anything the opposite would make more sense (as with electrolytics but I realise we are at the other end of the scale from that). A smaller cap' will have more voltage across it at a given frequency - since its impedance is greater. The actual 'numbers' depend on the whole circuit of course.
Didn't Cyril Bateman nail this some years ago ?

Capacitors are a mature technology while manufacturing technology has continued to evolve. I first wrote about capacitors in my magazine column back in the 80s and they have only gotten better since then.

There are a handful of objective metrics used to characterize capacitors.  For trying to parse out subtle differences between caps, perhaps try null testing. Human listening tests are very labor intensive and time consuming to reach any kind of statistical certainty (double blind et al).

Yes several good studies and articles on capacitors over the years. As I recall our own Samuel Groner did a study of low level distortion in electrolytics caps (but I can't find it right now).

@ CJ I don't recall the Z5F being a particularly clean dielectric but caps in active circuits can interact in different ways, so YMMV.

JR 

 

[Newmarket]

There are a handful of objective metrics used to characterize capacitors.  For trying to parse out subtle differences between caps, perhaps try null testing. Human listening tests are very labor intensive and time consuming to reach any kind of statistical certainty (double blind et al). ...

@ CJ I don't recall the Z5F being a particularly clean dielectric but caps in active circuits can interact in different ways, so YMMV.

Yes - some form of null testing = good idea.  (Some) People will still object to its validity though 

Z5F = volumetrically efficient = greater thermal and voltage coefficients = distortion.
But since a lot of effort goes into designing circuits which introduce "euphonic distortion" via non-linearities it  might not be an issue esp instrument amps etc...

 

[JohnRoberts]

Yes - some form of null testing = good idea.  (Some) People will still object to its validity though 

and some people think the earth is flat.

Z5F = volumetrically efficient = greater thermal and voltage coefficients = distortion.
But since a lot of effort goes into designing circuits which introduce "euphonic distortion" via non-linearities it  might not be an issue esp instrument amps etc...

Small ceramic caps are generally used as LPF poles in feedback networks at high frequency. The ASSumption is that THD (higher harmonics) caused by a pole tuned above 20kHz will not be audible. While this is generally true it ignores intermodulation distortion where two HF signals can cause (audible) lower frequency distortion products that we can hear (as mud and grunge). 

I worked "with" some cost conscious engineers who would measure THD of an audio path to decide which cheap caps they could get away with.  For the reason I have already shared (IMD) this could still miss some distortion mechanisms, but I was not in a position to influence their behavior at that time. 

JR

PS: I won't even venture a guess about CJ's capacitor preference.

 

[Shattersignal]

The big brains have it figured out, so it's time to close the thread: it doesn't matter what cap you pick because they all sound the same. 

In other news, all wine tastes like grapes, so please don't use flowery descriptions like "leather," "gunpowder," "minearality."  We did some research and it tuns out that it's all made from grapes, you dummy!

Will one of the moderators please delete my first post?  Thanks!

 

[JohnRoberts]

The big brains have it figured out, so it's time to close the thread: it doesn't matter what cap you pick because they all sound the same. 

:

In other news, all wine tastes like grapes, so please don't use flowery descriptions like "leather," "gunpowder," "minearality."  We did some research and it tuns out that it's all made from grapes, you dummy!

I just bottled about 5 gallons of wine less than a week ago (Riesling). I also brew my own beer, and roast my own coffee. There is huge variability in grapes (due to soil, vines, and weather) that directly affect the result. Likewise green coffee beans also express huge variation due to elevation, climate, soil, etc. (not to mention how they are roasted).

Your straw man does not even come close to being applicable to  passive electronic components like capacitors, that do not grow on trees. Maybe if they did? 

Will one of the moderators please delete my first post?  Thanks!

You should be able to delete your own content if you wish.... (I'd have to delete the entire thread, including several thoughtful responses).  Apparently this discussion didn't inform you, but maybe others can learn something.  I have written about this for a few decades.

JR

 

[Shattersignal]

:I just bottled about 5 gallons of wine less than a week ago (Riesling). I also brew my own beer, and roast my own coffee. There is huge variability in grapes (due to soil, vines, and weather) that directly affect the result. Likewise green coffee beans also express huge variation due to elevation, climate, soil, etc. (not to mention how they are roasted).

Your straw man does not even come close to being applicable to  passive electronic components like capacitors, that do not grow on trees. Maybe if they did?  You should be able to delete your own content if you wish.... (I'd have to delete the entire thread, including several thoughtful responses).  Apparently this discussion didn't inform you, but maybe others can learn something.  I have written about this for a few decades.

JR

"

Here's the analogy:

Presumably you can taste the difference between extremely similar batches of wine grapes, even when others cannot.  Likewise, there are people who can hear the difference between (*wildly* different) capacitors compositions within the same circuit, even when those capacitors are of the same value and "ought" to perform identically.  Clearly you would agree, since you've spent time advising folks about the parts spec in their commercial designs?  So then why all the trolling from folks when I show up to discuss the differences between caps?  Some of the replies on this thread have been downright rude.  JR, I see you have more to offer here than simple trolling and argumentation by contradiction and I respect your thoughts and experience.

If you have so much experience spec'ing caps, I would love for you to answer my topic questions in earnest.  Only two respondents have done that so far, by actually recommending a cap.  Everyone loves to sling the easy stock refutations, but nobody actually wants to respond to the stated thread topic?  I'm asking specifically about 100pf-1000pf value capsule coupling caps for a LDC mic. 

An additional note to the naysayers at large: you seem to approach this topic with quite a big chip on your shoulder.  Please take a step back, pause, and calm down a little.  I'm not trying to brainwash you with boutique capacitor mumbo-jumbo.  If you don't have an answer for me, then just politely move on.  Is this really how this community intends to comport itself?  Telling me to perform double blind tests simply avoids my questions.  And yes, I have performed double blind tests to guide me in my parts choices so far.  If you wish to create a harangue about testing procedures, please start your own thread.

 

[Matador]

Everybody can be correct here.  I will agree with JR: NP0-COG ceramic caps cannot be beaten in the sub 1nF range, in terms of stability, repeatability, solderability, and 'getting out of the way'.  I haven't found any better cap for use as a capsule-coupling cap, full stop.

But rest is also true: it's just about impossible to define what people Just Know TM that they hear.  I haven't seen any conclusive test in this application that shows anything other than random selection.  Almost every single listening test I have ever seen/heard was plagued with random selection errors, missing simple variable controls like measured capacitance, and even disclosing exactly which caps were used which biased the results.

 

[abbey road d enfer]

Presumably you can taste the difference between extremely similar batches of wine grapes, even when others cannot.  Likewise, there are people who can hear the difference between (*wildly* different) capacitors compositions within the same circuit, even when those capacitors are of the same value and "ought" to perform identically. 

You cannot compare capacitors in a circuit, which involves a limited number of variables, to the composition of wine, where thousands of parameters interact.

advising folks about the parts spec 

Capacitors are usually selected by the designer against a perfectly defined set of constraints: smoothiong caps are electrolytics because they offer the proper value-to-size ratio, NP0/CG0 because they are very stable against temperature and very linear, Z5F because they exhibit the perfect value-to-size ratio and low ESR,.. In fact, compared to 50 years ago, capacitor manufacturing has changed so much that the recommandations have changed, e.g. when some designers specified paper-in-oil for reliability, modern electrolytics have supplanted them.

So then why all the trolling from folks when I show up to discuss the differences between caps?

You're alluding to sonic differences, isn't it? It has been shown and demonstrated that the non-linearities of a capacitor became inaudible when they were in series with a much smaller cap, or when they are in parallels with a much bigger one. Typically in a condenser mic, the coupling cap is in series with the capsule's capacitance, and usually in a ratio of about 15:1.
Indeed using a 100pF in this position is bound to change both the harmonic content and the frequency response, particularly at VLF (<10 Hz); VLF response is a significant factor in how the overall response is perceived.
There is another factor that may explain some sonic difference, when the caps are of significantly different size, the larger one has more stray capacitance that couples the very sensitive Hi-Z connection to other parts of the circuit, which may introduce significant changes in the frequency response.
Also a leaky cap would interact with the capsule bias and the head amp bias, which would likely change the sound.

 

[CJ]

it's the green paint on top of that RMC cap that gives it the funk. 

visual stuff can influence the sound. try listening to a live song by Janis Joplin, and then relisten while watching the video and tell me it don't sound different. 

actually, i think i posted the wrong cap, it is these caps that are the bomb>