The 'Sound' of Capacitors..

Frequently asked questions on capacitor types, ratings, brands, use and abuse.
User avatar
andregarcia57
Cap Cooler
Information
Posts: 572
Joined: 16 Nov 2008, 15:42
Has thanked: 73 times
Been thanked: 109 times

Post by andregarcia57 »

Nocentelli wrote:I don't think anyone could answer that, but I read online (from Bill Finnegan himself) that the reason he wanted to eliminate electrolytic caps from the Klon KTR was because the tolerance of such caps was too high to give consistency between pedals, not because they sound inferior.
thanks....

KTR it uses film capacitor film in place of the electrolytic?
I came to think of tantalun because of constancy too.

User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

andregarcia57 wrote:what is the difference
1st of all, where in the Klon? There are many caps in that circuit that do different things.
2nd, two of those are polarized so you can't just swap one type for another.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
andregarcia57
Cap Cooler
Information
Posts: 572
Joined: 16 Nov 2008, 15:42
Has thanked: 73 times
Been thanked: 109 times

Post by andregarcia57 »

FiveseveN wrote:
andregarcia57 wrote:what is the difference
1st of all, where in the Klon? There are many caps in that circuit that do different things.
2nd, two of those are polarized so you can't just swap one type for another.

precisely by not fully understand the functions, I put all polarized, more will use tantalum

User avatar
Thecomedian
Information
Posts: 3
Joined: 17 Jul 2013, 04:08

Post by Thecomedian »


User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

Well, who could argue against such precise measurements as tone realism, space clarity and voice position?
"Listening session" of reservoir caps before a voltage regulator. What will they think of next?!
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
astrobass
Cap Cooler
Information
Posts: 548
Joined: 02 Mar 2013, 01:30
Has thanked: 42 times
Been thanked: 74 times

Post by astrobass »

Nocentelli wrote:I don't think anyone could answer that, but I read online (from Bill Finnegan himself) that the reason he wanted to eliminate electrolytic caps from the Klon KTR was because the tolerance of such caps was too high to give consistency between pedals, not because they sound inferior.
They also take up way too much space. I doubt that had anything to do with HIS choice in specific, but it's been a factor for me. Ceramic monolithic for anything 10 uF or lower.

User avatar
fuzzlabrador
Information
Posts: 6
Joined: 10 Dec 2013, 13:39
Been thanked: 2 times

Post by fuzzlabrador »

just sharing my tests



there are 2 different 10nf caps in the bunch I've tested .... I think they did sound different but not by much.

User avatar
astrobass
Cap Cooler
Information
Posts: 548
Joined: 02 Mar 2013, 01:30
Has thanked: 42 times
Been thanked: 74 times

Post by astrobass »

fuzzlabrador wrote:just sharing my tests



there are 2 different 10nf caps in the bunch I've tested .... I think they did sound different but not by much.
The problem is that you haven't isolated the source of the difference: Is it due to the material of the cap (it isn't) or due to slight variations in the true capacitance provided by the components in question (this is what it is)

10% tolerance means that two 100 nF caps could be 90 nF and 110 nF, and both would be within tolerance and "good". Depending on where those caps are in the circuit though, you're quite likely to hear a noticeable difference in sound, especially if these are being used in tone controls.

That's the whole problem. It's actually a huge pain in the ass to isolate for component variation, without which you cannot say that any difference in sound is owed to the cap's material or construction. Especially since a difference in component value is actually known to drive real differences in performance. Especially since the old caps that allegedly have the "better" sound typically come with a 20% tolerance. If the schematic calls for 100 nF, and you use a modern 10% cap that gives you 105 nF, but the original unit that somehow "just sounds better" has a 20% cap whose actual value is 85 nF, there's your mojo.

Especially since you have to consider that commercial units aren't designed on a one-off basis, they're designed to accommodate tolerance for an entire production run. So maybe you have a hypothetical circuit where 85 nF is the ideal value, but anything below 70 nF is unusable, though values up to 120 nF are acceptable if not ideal. So you use a 100 nF cap and sometimes it sounds great and other times it sounds okay.

Now look at how many caps fall into the signal path on an effect and consider that every single one of them is subject to this issue. Same with every resistor and transistor, too.

User avatar
fuzzlabrador
Information
Posts: 6
Joined: 10 Dec 2013, 13:39
Been thanked: 2 times

Post by fuzzlabrador »

not the best experiment but still it;s one. :)

User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

I wouldn't want you to lose your interest and enthusiasm but I'm inclined to say that such "experiments" that lack control do more harm than good. The hazards of hasty conclusions are well illustrated within this thread.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
fuzzlabrador
Information
Posts: 6
Joined: 10 Dec 2013, 13:39
Been thanked: 2 times

Post by fuzzlabrador »

Oh, I've done harm... what a pity. I will definitely make it my new year's resolution to stop being curious and experiment,and start saving for a laboratory.

User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

There were many things you could have done without extra equipment. And understanding experimental methodology would benefit you in many ways, including developing and debugging effects and general problem solving.
Like I said, there's nothing wrong with curiosity, but "a little knowledge is a dangerous thing".
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
fuzzlabrador
Information
Posts: 6
Joined: 10 Dec 2013, 13:39
Been thanked: 2 times

Post by fuzzlabrador »

well I would go into a private parts measuring contest if I was 20 years younger for sure... but instead I will wish you and everyone else a Happy New Year. :thumbsup

User avatar
KMG
Solder Soldier
Information
Posts: 209
Joined: 02 Feb 2010, 20:26
Location: Russia
Has thanked: 5 times
Been thanked: 80 times
Contact:

Post by KMG »

Interesting test of the different types of capacitors in tube amplifier (unfortunately in russian, use google translate)
http://forum.guitarplayer.ru/index.php?topic=301773.0
Comments:
- Used reamp to eliminate difference of playing
- Same controls & mic position while recording
- Measured capacitors difference not more than 5%
Image

User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

KMG wrote:unfortunately in russian, use google translate
Unfortunately Google isn't helping much. If you or another Russian speaker can find the time, maybe you could give us a summary.
Here's what I did to compare the samples:
1. Used only one (the right) channel, to make things a bit easier and clearer
2. Aligned the samples as best I could. I found that they were not perfectly aligned to begin with, so be aware if you're doing this kind of test.
3. Compared the three by subtracting one from the other (A-B, A-C and B-C), by reversing the phase.

It's just a quick test, I didn't use long averaging for smoother and more precise results, but I did use the same part of the song in each case.
Here's the baseline, peaking around -1 dBFS with -11 dB RMS: http://xonedesign.ro/publik/captest/baseline.png This is the right channel of the SBE715 sample, and the other ones look just about the same.
Here are the results of the subtractions (I'm hoping the file names are self-explanatory):
http://xonedesign.ro/publik/captest/SBE715vsSOZO.png
http://xonedesign.ro/publik/captest/SBE715vsWIMA.png
http://xonedesign.ro/publik/captest/SOZOvsWIMA.png

What one can conclude:
1. The difference is lower than -23 dB RMS down. In other words, less than one part in 14 if the flu hasn't affected my math.
2. The difference is mostly broadband, that is mostly a matter of loudness (signal level) and not of frequency respose.

I have to say, this resource is a lot better than most silly "experiments" on YouTube. Wish I had more time to look into it, but the points I've made above provide me with plenty of evidence to infer that any difference in cap type is less significant than systematic errors, even with these fairly controlled conditions.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
KMG
Solder Soldier
Information
Posts: 209
Joined: 02 Feb 2010, 20:26
Location: Russia
Has thanked: 5 times
Been thanked: 80 times
Contact:

Post by KMG »

Subtracting tracks is not good way.
Better way is to subtract cumulative spectrums.
Human ear does not "feel" exactly small difference in total volume, but skewness in spectrum products at the same level it "feels" distinctly.
Image

User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

Maybe there's a language barrier but that's precisely the point of my tentative findings: there is no "skewness in spectrum products". The spectra are pretty much the same.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
KMG
Solder Soldier
Information
Posts: 209
Joined: 02 Feb 2010, 20:26
Location: Russia
Has thanked: 5 times
Been thanked: 80 times
Contact:

Post by KMG »

The main difference in spectrum is between Wima & other types of capacitors.
I think this is due to the difference in construction - Wima has "comb" construction, other types rolled construction.
SBE715_WIMA.jpg
But the difference between rolled types of capacitors is also "hearable".
Image

User avatar
FiveseveN
Cap Cooler
Information
Posts: 591
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 215 times
Been thanked: 116 times
Contact:

Post by FiveseveN »

Fine, we'll do it your way. But first I normalized the samples' volume to the loudest RMS value (-11.2 dB).
overlap.png
Almost perfect overlap. Ain't that some shit?!
I even zoomed in on the biggest difference I could find: 1.9 dB @ 2827 Hz.
difference.png
KMG wrote:The main difference in spectrum is between Wima & other types of capacitors.
Yes, but it's not in spectrum (frequency response) but in crest factor: the peak value is 0.2 dB louder. And there are exactly seven samples that have this amazing anomaly. That's seven out if 9910481 (ten million). I think we can safely blame that on quantization error.
peak.png
But the difference between rolled types of capacitors is also "hearable".
And yet to be proven.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
KMG
Solder Soldier
Information
Posts: 209
Joined: 02 Feb 2010, 20:26
Location: Russia
Has thanked: 5 times
Been thanked: 80 times
Contact:

Post by KMG »

My tests show difference about 2 dB in area 2-3 kHz at level about -25 dB from highest peak at 150 Hz.
Under spectrum difference I don't mean low signal frequency response, but different amount of produced harmonics.
In overdriven tube stages interstage capacitors works differently than in "linear" circuits due to several side effects.
Such as assymetry of the stage input impedance due to "grid clipping". In this case capacitor parasitic parameters gives more difference than in "linear" circuits.
In amplifier clean channel, I think difference will be neglible.

PS "I don't know why" doesn't mean "it can't be".
Image

Post Reply