The 'Sound' of Capacitors..

culturejam · Post by **culturejam** » 16 Feb 2010, 23:20

bumblebee wrote:The last thread I posted in like this was about Dialectical Materialism not Dielectric Materials.

Ah, I miss college.

Greg · Post by **Greg** » 16 Feb 2010, 23:21

soulsonic wrote:Man, this thread is off tha hook!

I love you guys.

Cool huh ??

Some good knowledge and links.. differing opinions.. and a dash of emotion.

It's like discussions about Mojo components... anytime you start telling people they're imagining things (which they may be)...

IvIark · Post by **IvIark** » 16 Feb 2010, 23:23

And United beat MIlan so who gives a shit about capacitors? [smilie=wine.gif]

mictester · Post by **mictester** » 16 Feb 2010, 23:39

bumblebee wrote:
Its not the tolerances,its the cap, I know this. For example I've tried various caps in the two clipping stages of the big muff from 100nF to 2.2uF. The value's between 470nF and 2.2uF don't make much of a difference but when replaced with an electro from 470nF to 2.2uF it does make a difference. It cant be tolerances because I have already established that the big muff sounds fine with clipping caps from 470nF to 2.2uF with poly type. When an electro of between 470nF to 2.2uF is placed in both stages the sound changed for the worse and considering the exact pedal sounds fine with caps from 470nF to 2.2uF it must be the cap type,not the value as 470nF to 2.2uF is a pretty wide variation which sounds fine with poly caps but not electro's, this rules out the possibility that is the tolerances unless of course you would like to argue that the tolerances of a 1uF electro can measure from 1nF to 1000uF.....

But there's a flip side to this too, I have other big muffs (American) that have all electro's in them and they sound fine. Its ONLY in the old SBM's I have noticed such a difference between electro and box type which makes me think its got to do with the other components and their tolerances used in the SBM's used in conjunction with electro's as the parts values used between American BM's and SBM's are quite different.

All you're hearing is the change in leakage!

mictester · Post by **mictester** » 16 Feb 2010, 23:43

KMG wrote:Real capacitor consists not only of C but also from R (dielectric losses) L (stray inductance).
Also capacitor loaded for example on diode limiter has large enough amount of bipolar repolarization.

We're not talking about high power here - we're talking a volt or two and little current, so R isn't significant. Similarly, the self inductance of capacitors at audio frequencies is entirely negligible. Next you'll be warning me about the coupling between tracks....

Greg · Post by **Greg** » 16 Feb 2010, 23:47

mictester wrote: All you're hearing is the change in leakage!

http://www.tadiranbatteries.de/eng/down ... th_Eng.pdf

If leakage affects sound, and different types have varying leakage properties, is that not a characteristic of different cap types that can change the sound ?

IvIark · Post by **IvIark** » 16 Feb 2010, 23:48

mictester wrote:All you're hearing is the change in leakage!

So they can sound different then?

MoonWatcher · Post by **MoonWatcher** » 17 Feb 2010, 00:15

So then who's going to pony up and do a hot swap test?

So far we've got a few who've done it, others that are sticking with what the tech papers say, and a quasi group that have impressions of certain types in certain pedals, but with no real constants.

Pepsi challenge, anyone? Isn't curiosity killing anyone? I mean, we've already got 5 pages of this stuff...

Yeah, I miss college too.

oldgravity · Post by **oldgravity** » 17 Feb 2010, 00:28

MoonWatcher wrote:So then who's going to pony up and do a hot swap test?

So far we've got a few who've done it, others that are sticking with what the tech papers say, and a quasi group that have impressions of certain types in certain pedals, but with no real constants.

Pepsi challenge, anyone? Isn't curiosity killing anyone? I mean, we've already got 5 pages of this stuff...

Yeah, I miss college too.

Yeah, I'd love to see this. I will put up one, count 'em, 1 American dollar, via Paypal, to the person who puts this test together, for their trouble. If everybody else here will do the same, we could get like 8 bucks, which is an alright wage for a half-hour's work, eh?

I don't really care that much about capacitors one way or another, but I do really like to take part in experiments like this. Awhile ago somebody on Talkbass did a great audio comparison of a Sansamp VT versus a real SVT. There were of course those who argued with the test procedure, but it was fun and the majority of voters guessed wrong about which was which. Something like 70% thought the Sansamp was the SVT.

IvIark · Post by **IvIark** » 17 Feb 2010, 00:39

I'm on, but to be honest I was just going to do it myself for my own interest because unless we do one build and ship the pedal round to every individual I can't see how we could offer an opinion. It's not like we can do a vid and post it on youtube. What are the suggestions for how it's done?

Greg · Post by **Greg** » 17 Feb 2010, 00:49

IvIark wrote:I'm on, but to be honest I was just going to do it myself for my own interest because unless we do one build and ship the pedal round to every individual I can't see how we could offer an opinion. It's not like we can do a vid and post it on youtube. What are the suggestions for how it's done?

I think Moonwatcher's suggestions on the last page are good.. a single circuit with switched caps..
That takes out all variables from other components, and allows an immediate comparison.

Then I think you need a switcher and some good ears.. make the switching hidden and see if any detected differences can be repeated.

What types to test ?
For reference, a good Poly* Film type.
Ceramics
Electros

IvIark · Post by **IvIark** » 17 Feb 2010, 00:54

But who's going to actually test themselves with it or are we talking about each doing something individually?

oldgravity · Post by **oldgravity** » 17 Feb 2010, 00:55

I just assumed there would be some audio clips made. It's imperfect, surely there will be things said about how things are different when you're playing through something versus just listening, "feel", imperfection of lossy audio CODECs, etc. It isn't ever going to be double-blind unless you can set up a switching arrangement in which you don't know what you're switching to, and ideally you need to allow for a third switch position which could be either A or B, and so on. We're not going to get a really solid testing procedure in any case, but it's interesting to just try.

Greg · Post by **Greg** » 17 Feb 2010, 00:57

If you do the test and can get a few friends with good ears to join in, I think the results would be very worthwhile.
If you then wanted to send it to someone else to do the same, even better.

I'm thinking of doing one also (might sound different in the Southern Hemisphere

), but I can't do it straight away.

Personally I suspect any differences will be too subtle to hear in clips.

oldgravity · Post by **oldgravity** » 17 Feb 2010, 01:06

But if we're going to take your friends' words for it, it would definitely need to be a double-blind ABX test where the audience and the guitarists and the person switching would not be able to see what was being switched to. That gets pretty complicated, which is why I think sound clips and letting us each make our own observations is better.

Greg · Post by **Greg** » 17 Feb 2010, 01:15

oldgravity wrote:But if we're going to take your friends' words for it, it would definitely need to be a double-blind ABX test where the audience and the guitarists and the person switching would not be able to see what was being switched to. That gets pretty complicated, which is why I think sound clips and letting us each make our own observations is better.

Greg_G wrote:Then I think you need a switcher and some good ears.. make the switching hidden and see if any detected differences can be repeated.

If the listeners don't know what they're hearing, and are able to pick one type from another consistently, I think you've (reasonably) proven there's a difference.

IMO clips are not an effective way to hear subtle, or often not-so-subtle differences.. too many variables with loss of quality, recording techniques, playback equipment, etc... but they might be fun.

oldgravity · Post by **oldgravity** » 17 Feb 2010, 01:24

Totally, but the ABX part is really important. If you tell a bunch of people, "we're going to play you 2 different things and you tell us if you can hear a difference", people will say they hear a difference. You have to do cap A, then cap B, then cap X, which could be cap A or cap B, and see what people say. Really, what should be done is to play some guitar through cap A, then B, telling people which is which, then switch randomly back and forth between the 2 about 10 times. If there's really a meaningful difference, they should be able to identify which one it is each time. But still, this is so much harder to orchestrate than simply recording 2 sound clips and letting us all decide for ourselves.

Edit, response to your edit: the audio test will never be perfect for the reasons I listed above, but personally I think they're better than rounding up your friends and having us believe what your friends say. No offense to your friends, of course.

But I certainly wouldn't trust my friends' ears for this sort of thing. My girlfriend listens to music on her iBook speakers and I don't think she could tell the difference between 2 completely different amps, let alone just swapped caps.

IvIark · Post by **IvIark** » 17 Feb 2010, 01:25

Greg_G wrote: What types to test ?
For reference, a good Poly* Film type.
Ceramics
Electros

They're probably the three I'd choose.

Greg_G wrote:Personally I suspect any differences will be too subtle to hear in clips.

I agree that could well be the case, and it would also be possible for someone to make their clips fit in with their previously expressed opinion if the person doing the recording was that way inclined, so there can still be the same amount of doubt as before. And I can't see us all getting together, sitting in a room and doing a Visual Sound hands-up test.

IvIark · Post by **IvIark** » 17 Feb 2010, 01:28

oldgravity wrote:Totally, but the ABX part is really important. If you tell a bunch of people, "we're going to play you 2 different things and you tell us if you can hear a difference", people will say they hear a difference. You have to do cap A, then cap B, then cap X, which could be cap A or cap B, and see what people say. Really, what should be done is to play some guitar through cap A, then B, telling people which is which, then switch randomly back and forth between the 2 about 10 times. If there's really a meaningful difference, they should be able to identify which one it is each time. But still, this is so much harder to orchestrate than simply recording 2 sound clips and letting us all decide for ourselves.

But at what settings? The Visual Sound opamp comparisons were basically pointless to me because they were able to set all the controls in the way they wanted beforehand, which there can be no doubt were the ones that made their pedal sound most like the Klon. So it would have to be more involved than that, playing and swapping through the gain range. I just don't think it could be done reasonably in a way that is going to alter anyones opinion unless you're there, can see what's going on, and can play more than a low gain blues riff through it.

Fun maybe but not really answering any questions.

RnFR · Post by **RnFR** » 17 Feb 2010, 01:33

yeah, i think it would be pointless to do this with mp3's. also, you would have to do a few "fake switches". just to see if they are really hearing shit that isn't there. i'm sure some will.