Do capacitors sound different?

[Glennjeff]

Duly noted HotBlue.

Well, that was really unexpected.

With 5 way switch I tried 150's, Mojotone Dijon, Orange PS, Industrial 2kv switching cap PP, and a VitaminT. In the coupling between input valve and stage 2. I had several of each so I measured them and chose the ones of closest value they were all 0.021 uF by my meter.

Had to have a lot of listens to actually get my ears around it but;

The 150,  PS and industrial PP had a kind of indistinctness and harshness about them.

The Dijon and Vitamin T were much more transparent, think the Dijon might have had a tiny bit more warmth than the Vitamin T but the Vitamin T had a bit more top end sparkle.

In a way it was very subtle, but in another way it was rather gross and obvious. Three failed, two passed is how I would put it. To be fair one should probably take a larger small sample size than 1 unit.

A person could go insane trying to get every cap in an amp optimised so just in case, it's been nice knowing you folks.

Guess now I will shunt out the fails and try OD 715 and 716 with the winners.

Oh Dear Me !!! It appears I do not have 716's, thought I got them with the big order 8 months ago but nope.

The OD 715's were in the pass category, transparent and full bodied. To be honest the three "passes" all had something to distinguish them from the others but could not really say that any of the three were better than the others. My ears are probably fatigued by now.

[Glennjeff]

One HUGE difference.

Some of the more expensive caps have much better shielding. Doing an "earth end" test on a CRO, a typical cap has 6mV pp hum wrong way round and 2 mV pp right way round, fingers gripping the cap body. Solens  have like 0.1mV pp wrong way around and 0.0mV pp (undetectable with my CRO) the right way round. Vitamin T's are similarly hum free but the outer shell is a metal can and connects directly to the earth end lead internally, no confusion there.

I also observed that different caps of same value from the same "not so expensive" manufacturer can have significantly different hum rejection.

Caps with significant shielding advantages are the obvious choice for any amp stage that has low signal strength such as the input stages.

I have the Vitamin T's so all my guitars are going to get tone control cap replacements. (I think they may be standard in PRS)

[Glennjeff]

Finally, Vitamin T, 715, Dijon and 150's in the two 0.01 uF couplers between PI and PA (via PPIMV). This amp stage has much different operating conditions than input stage, higher DC and a lot more signal strength.

There were some very subtle differences but really - no difference, not even enough to say I subjectively preferred one over the other, although I will go with the 715's here.

I think ones choice of size, shape, color and cost may be more significant factors.

Rolling tubes gives a much greater variability in tone, in the past noticing major differences between 2 different 12AX7's from the same manufacturer, both new, and then choosing a 12AX7 from a different manufacturer.

Transformers also are known to give markedly different sounds, and whilst never having used them, a Classictone is on order for the final build OT.

It was a good exercise in listening vewy vewy carefully when hunting rabbits. In the end they all taste good as long as they don't have  myxomatosis.

[printer2]

Would be interesting to know what amp.

https://www.youtube.com/watch?v=XIaENqCBajY

[Willabe]

Found this last night when looking for something else.   

From Vacuum Tube Valley, issue 11, VTV Listens to Signal Capacitors, page 26;

"Some tips: Remember to wire signal capacitors so that the inner foil points toward the output and the outer foil points toward the input."

Ok, we got that, but then he writes;

"If you wire the outer foil to the output side, the sound will be muffled and rolled off."

[tubenit]

Printer 2,

THANKS for posting that!  The Mallory 150's sounded muffled to me and I thought the Illinois caps had the best definition and articulation.  Of course, that's simply my personal subjective review.   

With respect, Tubenit

[12AX7]

Found this last night when looking for something else.   

From Vacuum Tube Valley, issue 11, VTV Listens to Signal Capacitors, page 26;



"If you wire the outer foil to the output side, the sound will be muffled and rolled off."

Hmmm...interesting to me. The reason being that after dragging a old amp i built years ago out of storage a few months back and tweaking it for the last few obsessively, at one point it started sounding amazing to the point i threw it back into the cab and stopped tweaking it. It had sounded bright and thin and i couldn't seem to get rid of that issue with any of the normal ways. And when i red the quote above it struck me because he only change i made between the point where it sounded bright and thin then great was i changed a tone stack cap. I was trying something and accidentally hit the cap with my iron and melted a rather deep hole in it so i changed it. After i did i started noticing the change and i couldn't imagine thats what did it but i could not think of anything else i did at the change in tone. And it wasn't because they varied in value because i have tried different values in the tone stack and it took a jump to double or 1/2 value to make a difference, and even then it never sounded like this whatever else i tried. Also, i had just installed all new caps of the same brand/value (mallory) because i was renewing the amp since the components were old and reused a zillion times. But after the renew the amp sounded no different. It was after i melted one of those new ones and swapped one of the old malories in that the change happened.

I still have a hard time believing it was that, but again i did nothing else i can remember between the change. My first thought when reading the above quote is that the tone caps in the stack were never oriented correctly (assuming this really CAN change tone)  and when i installed the new caps one was but not the other. Then when i melted that one and swapped a old one in it went in oriented correctly. Then i stated thinking about how i never felt the tone stack was quite right. the way the treble and mid worked just seemed wrong. The treble got very thin and the tone turned to am radio sound when  i turned it past about 2 oclock, and the mids were harsh turned up. And no, it wasn't a bad joint or anything like that because i was always redoing the solder joints as i experimented.

So what i'm wondering is, could there really be something to this orientation thing not just for noise but as far as tone, and could it be ultra important in the tone stack and possibly be one of the things that can make or break an amp? I can't see how or else production amps would be much more inconsistent than they are, but who knows. This makes me want to get the orientation in the rest of the amp right but the problem is, i have tried and b=never been able to hear it, at least not with mallory. I used the cable and cap from hot to ground then reverse it method. I never heard any change in noise when i'd reverse the leads.

[kagliostro]

I've find this link

http://www.mennovanderveen.nl/nl/download/download_4.pdf

Franco

[enorbet2]

Hello.  I realize this thread is over 120 days old and may qualify as "necro" but since my post here is exactly on topic and only creeps slightly into related topics, I figured I'd better try this first.  I'd like to chat about what I think is another factor in how caps can alter sound in guitar amps, or really, any amp, and that factor is ESR, or Efffective or Equivalent Series Resistance.

First in the Power Supply - It is my understanding that ESR not only affects how much ripple is filtered out but also the "flexibility" in regards to the each stage's, and of course especially the power stage's, ease in swinging the power supplied.  There is a term, in fact, for this characteristic, which is escaping my memory for the moment, but I imagine anyone who cares about such things already knows exactly to what I am referring - Power Supplies have an effective "impedance".  Personally while I, like any vintage amp lover, enjoys a bit of "sag" I prefer to achieve it in other ways than poor filtering.  This not only affects tonality and harmonic structure but also that enigmatic dynamic "feel" of guitar amps which are not reproduction devices but musical instruments in their own right.

Secondly, in the Signal Chain -  Wherever possible I prefer non-polarized caps in the signal chain, even in bias resistor bypass caps, because polarized are non-symmetrical and this does have an effect on the waveform.  I can see it to some degree even on an oscilloscope, to a larger degree on a spectrum analyzer or FFT, and I can hear it.  It is somewhat slight especially if only one is swapped out for non-polarized as the effect seems to be somewhat cumulative.  That last bit is actually anecdotal since I haven't been able to measure that other than by ear and that can be subjective.

Lastly, and tangential to ESR, is that it seems guitar amps, especially those that are overdriven and there even more especially those that seek to substantially overdrive the power section, sound sweeter if lower values are used early in gain staging coupling caps and get progressively larger as each step to the power section is considered.  IOW "trebly" at the front and "bassy" further in.

I'd very much enjoy any comments on these concepts so thank you in advance for any hard data or even imaginative musing.

[ewc]

i think the best capacitor is no add any signature to the sound but it never exist,,could you tell me which cap from all caps you had tested that add very little signature to sound?
thank you!

[shooter]

caps by design will have some effect on tone since they interact with frequency. If it allows a large bandwidth to pass, you will hear more frequencies, narrow band will do the opposite.  it's a math n experiment, repeat til happy kinda part 

[enorbet2]

I like HotBluePlates' posts quite a lot because he actually does the work IRL not just imagining or regurgitating others' imaginings.  I fully agree from my own work of many decades that in the majority of cases, cap values have far more tonal impact than cap materials with one possible exception - Cathode Bypass Caps.

A very long time ago I read that electrolytic caps are asymmetrical in ESR and wondered, especially since a bypass cap heavily influences/controls gain (actually amplifying small differences) so I tried replacing bypass caps with true non-polarized versions where possible and reversed back-to-back (series) electrolytic caps where more practical.  I actually expected more difference than I could measure or hear but there IS a difference and it is cumulative depending a lot on how many stages in the signal path have bypassed cathode design. 

This difference  naturally increases with gain influencing whether any overdriven clipping is symmetrical or asymmetrical, with obvious and well documented differences in harmonic generation..  That effect can be reduced if there are subsequent stages that reverse the asymmetry.  Bottom line is there is considerable difference when overdriven and all gain stages are either 100% symmetrical or 100% asymmetrical rather than a mix.

It doesn't take a great deal of effort to try this for yourself and most gain stages will have full frequency response gain with 1uf-2uf  non-polarized caps so it can be tried with no electrolytics at all fairly easily and cheaply.

[jojokeo]

Hello enorbet2, an old thread for sure but findings and/or updates are always appreciated. Since this time I have included a few options to help with what I think is what you're describing with "symmetry of gain" and in the power supply fitration as well. Of course we are talking about small details but these can and do make slight differences in gain, character, sound, possibly limiting/eliminating artifacts and parasitic oscillations. Also, these additional caps on cathode bypassing may be another option yielding similar results to using non-polarized larger caps alone? Attached are a few examples I have incorporated. Tubenit has experimented a lot with capacitor use in many places of the circuits and not only in the manner with which you have commented but if you haven't seen the files called "smoothing caps" you might find it interesting? But, I have not done everything in one amp all at once and don't think there would be this need? But who knows, you might just want to try it all? YMMV as they say. :)
jojo