Designing and capacitors

[Dylfish]

Hi guys, 

Do the different types of capacitors make a difference with their applications or will they all work but might result in a slightly different sound?  Eg can ceramic be substituted with electrolytic of the same value with minimal effect?

[GibsonGM]

Electrolytic are mainly for power supply use.  Yes, they're found here and there, but not usually in a place that will affect signal quality (say, as an emitter bypass). 

Ceramic have a reputation of being a bit "gritty" sounding, and their value changes with frequency, so they're not 'consistent'.   But they are 'the real deal' for vintage tones, for those who really want to replicate the 'days of old'. That's what they had, and it's fine by me.

Poly types are said to be a little 'sterile' sounding, or 'too perfect'.   

IMO, you can sub in most anything, as long as it's a fairly well-made part, in most places. I have yet to be *shocked* by a big change in tone by doing so.   I use poly caps in the signal path (input, output, coupling....).  Electrolytic for power stuff, places you need uF's instead of nF.    Ceramics when you need pF.

Just that simple.    For me it's about value ranges...you would need a LOT of typical ceramics to equal what you'll find with most electros...

YMMV, some really feel there are HUGE differences...I deal with amplifiers and effects of limited bandwidths (umm, guitar...it's not hi fi) so don't see any need for utter boutique components.   The only caveat I'll make is to say that if I were making a super-wonderful AMP, I might get spragues or mallory's, because they have a rep as being WELL-MADE.   Quality = mojo, to me, vs. type.

[Dylfish]

Thanks Mike.  I've been wanting to breadboard a up to test the gain and such so I'll use what I have around at the moment then go from there later.

Cheers

[Gus]

You need to be careful when posting about ceramics
There are different types of ceramics
Ceramics have different classes
One is a really good capacitor

With capacitors like carbon comp resistors the difference might be heard at greater voltage differences from lead to lead.  PI to grids of the output tubes in a tube amp is one place to do a test.
If you search the web look for the Cyril Bateman capacitor PDFs  IMO the one of the best web sources of info on caps


Now at 9VDC caps can make a difference
Microphonics at loud volume/ feedback levels
DA can matter with switch popping

[nate77]

I have to agree with mike. He's been doing this much longer than I have ( and has solved a good handful of problems of mine for me), but in my experience as well, it's about the values. I typically get the best caps I can, but value is the most important thing fr me when building a circuit. Almost all commercially and semi-commercially made PCBs have standardized spacing and silk screening based on value not material, as well as most layouts for perf and vero. I know there are some builders out there that obsess over every possible avenue of tone enhancement, but for me at least, being that exacting is most useful for 1:1 clones of classic or out-of-production pedals. There are a few good books out there that explain the suble differences in cap types and there relationship to placement in the circuit (I think alologman and Brian wampler have some good info in their books). I guess that's not an answer to your question in retrospect, but after my own aggressive search and experiments with component placement , I came to the conclusion that the vast magority of standard build practices are the result of other people's search and experiments as well. Good luck man, I hope that you discover something awesome!

[Jdansti]

To boil down what's already been said, in general, use electrolytic if the schematic or layout calls for a polarized cap and use ceramic or poly if it calls for a nonpolar cap. You normally don't have to worry about this because of the values that polar and nonpolars are made; however, there is some overlap in values between 0.22μF (220nF) and 1μF.

As for what various types of caps sound like, if you were to take a poll, my guess would be that for caps in the signal path, most people say they like the sound of poly caps over ceramic for nonpolar, and tantalum over aluminum electrolytics for polarized. This is not a scientific statement!  Sound is very subjective. I'm sure that someone's done a controlled study using an o-scope, but in the end, the question is what can the ear perceive? 

[Gus]

I have been experimenting with capacitors since the late 70's After the TAA and audio articles.

As with most stuff on the web things get exaggerated sometimes you can hear things sometimes you can't, it matters where the cap is in circuit

Search for and read the Batman PDFs you will learn some things I did a google search and you can still find the PDFs
Sometimes it is not about the cap value.  I posted two things that matter at 9VDC
Take different caps connect them to a scope input and tap them.

I left a really microphonic cap in A friends guitar I worked on because they said it sounded better when played live on stage at a loud volume

Read about DA and think about the input and output coupling caps and DPDT switching sometimes used in effects

Here is a good example of cap type mattering not 9VDC

When I build a plate out triode tube microphone with an output transformer the coupling cap TYPE as well as VALUE matter.  
This is a RLC network sometimes with 60VDC to 80VDC at the plate that connects to the cap the other side often goes from cap to transformer to ground.  
Now at lower voltages like you might find in a phantom power transformer out microphone the tone change is less evident if at all noticeable.

I have changed the power supply caps in a tube amp to polypro 630VDC and the cathode caps to PET film(you need to think about what changes and you need to add a part).  This changed the amp a lot, the output plate and screen power supply caps do matter

Try different treble caps in a Fender or Marshall tube amp tone circuit.  Measure and match the value between different types to remove the value change issue

EDIT you can hear things in isolation but you might not hear any difference when a song is mixed or when people are playing live on stage.  
Think about this give a good player a solid state amp with a good speaker and cab (instead of a good tube amp and speaker and cab)how many people would notice anything when they are at a bar or club listening?

[merlinb]

Even the worst ceramic in the worst place seldom introduces more than 1% harmonic distortion. It's questionable whether you could hear that in a hifi situation, but for electric guitar you're certainly never going to know it's there. Heck, the loudspeaker will introduce maybe an order of magnitude more than that, even if you DON'T add deliberate distortion elsewhere! For guitar, capacitance value is everything; dielectric is neither here nor there.

[nate77]

I've read a few articles about the difference in cap types when looking at an O-scope but as Merlin said, if you are pushing tubes aggressively or using any kind of dist/overdrive, any difference heard in a controlled situation will be negligible at best. This is just my experience and opinion, but once I have my signal running through my pedal board or even a single dist type circuit, I can't perceive any difference. I do have to agree with Gus about tantalums in the signal path. A few of the first mods I did (a few boss an Ibenez pedals), replacing electrolytics with tantalums made a noticeable difference.

[Jdansti]

I knew I was taking a risk talking about differences in sound between caps. I'm not an expert on that subject.

I was mainly trying to answer his last question: can ceramic be substituted with electrolytic of the same value with minimal effect?

I believe that in most cases, the answer is no. Perhaps there are situations where this would be ok, but for a beginner, I recommend following the schematic or layout.

[GibsonGM]

Debate is good and fun; nobody is mad, JD     Yeah, tantalums are a little better than electros, which are really 'the junk pile of capacitors'.    Tantalums might also take up less space - bonus!   One should use the schematic parts list, WHEN available...sometimes they're not.

This is like "NOS tubes vs. new"....very subjective.  But hey - will the circuit work, and be *fine* if you intermix cap types? Yes, as long as the cap is well-made and you observe polarity rules etc.    Will it be "the most awesomeness that you ever heard?".  Maybe, maybe not.  

The answer to the original question is, "they might make a slightly different sound" but will work.   The "slightly different sound" part is where the debate starts!  I didn't think he meant replace electro with ceramic....guess you could, but wouldn't that require a BANK of them to work out??  I saw it as more of a ceramic vs. poly, LOL!   Makes me think of making up 10K with 100R resistors or something - the next installment should be "what type of cap goes where, and what values you can get using them"!   

Overall, everything everyone said above is true.  I've SEEN the deviations that ceramic caps make on a scope....but can't hear them!  As a GENERAL rule; but there's always going to be some exception. Let your ears do the deciding.

[merlinb]

Debate is good and fun; nobody is mad, JD     Yeah, tantalums are a little better than electros, which are really 'the junk pile of capacitors'.  

Tantalums have lower leakage and ESR than electros, but their distortion is about an order of magnitude worse- practically as bad as cheap ceramics. Conversely, distortion in non-polar electro's is about an order of magnitude better than ordinary electros. Just sayin!

[GibsonGM]

Debate is good and fun; nobody is mad, JD     Yeah, tantalums are a little better than electros, which are really 'the junk pile of capacitors'.   

Tantalums have lower leakage and ESR than electros, but their distortion is about an order of magnitude worse- practically as bad as cheap ceramics. Conversely, distortion in non-polar electro's is about an order of magnitude better than ordinary electros. Just sayin!

LOL! SEE!?  :laughing over here:    And I prefer tantalum to electro's, for signal-path work. And even tho I PREFER them, I almost always end up using the darn electro's!  That is what's more available in my area.

I bet if you could parallel enough ceramics to make up, say, .047uF, and compare THAT with a .047 poly, THEN you'd probably have something you might really *hear*....provided you found a place in the circuit that would 'show off' the differences.    Maybe one day, someone will come up with a cool circuit that DOES let you make a definitive sound comparison.

[LucifersTrip]

Has anybody discussed multi-layer monolythic ceramics?  In 9V fuzzes, I've been using the crap out of them in lieu of 1uF - 10uF electros, and many times
use 2 or more in parallel instead of 20+uF electros.

The OP should consider the endurance & tolerance....electros dry out over time and generally are less accurate (tolerance) than other types.