are multilayer mono ceramic capacitor good for audio? for pedals?

[Yoshi]

Hello, are multilayer mono ceramic capacitor good for audio? for pedals?

I visited some pages, and many of them say that ceramic capacitores are bad for audio. Im makin a tubescreame clone and it has a 51pf capacitor, but here, were I live, there are only ceremic capacitors.

Thanks

[soggybag]

There is a lot of debate. Most products claim to use higher quality caps for better quality audio. All of the measurement and empirical examinations I’ve seen say you can’t hear a difference between different types of caps.

Pretty much any type of cap will work if it’s the correct value. The material doesn’t matter.

Some caps are polarized in which case you need to respect the polarity.

I like the multilayer ceramic caps, they are cheap, come in a wide variety of values, and reasonable sizes.

[idy]

Almost all pedals use ceramic for values under 1nf. For "authentic period correct" the old style ceramic were everywhere.

Most of the time experimenters hear a difference between types its because of tolerance: one is 10% high, the other is 10% low.

[antonis]

I have nothing to add but Douglas Self "Small Signal Audio Design" Chapter 2, Components..

[Mark Hammer]

It's rock and roll.  You're making a distortion pedal.  Where does "quality" fit into all of this?

All the classic pedals we pay huge sums for used the cheapest garbage around, and we think it sounds fantastic.  So what exactly is there to "improve"?

NOW...if you were making a mixer, or a mic preamp for sampling, or a high-end home stereo system, THEN it would be relevant to talk about quality, especially since many of those devices are for processing multiple sources and an overall wide-bandwidth signal.  If you were building an analog synth and you REALLY needed the unleakiest caps possible for your sample & hold module, or to assure your VCOs tracked each other flawlessly, you'd want precision and immunity to any heat effects.  But when it comes to adding crunch to a guitar that will be played through an amp with speakers that roll off above 6khz, and when your final tone will be determined by many more factors than one pedal, I doubt there is really anything useful to be added by being picky about caps.  If they measure the value it says on the cap, it fits in the space on the PCB, and is rated at a suitable voltage, exactly what more do you need from it?

[antonis]

If they measure the value it says on the cap, it fits in the space on the PCB, and is rated at a suitable voltage, exactly what more do you need from it?

Fancy facies (e.g tropical fish or orange drop) and/or unjustified extraordinary cost (e.g. Nichikon "premium audio"), perhaps..??

[Mark Hammer]

Again, I will emphasize that sometimes quality really DOES matter, when the goal is consistent (over time and across manufacturing) high fidelity processing of wide-bandwidth multi-source signals for music reproduction and other types of  clean processing.  That's simply not what guitar players do or need, however.

If a person was a quality-control food taster at some expensive high-end food manufacturer, I could see where they would be picky about the composition and mouth-feel of the spoon they would dip into each vat of product, wash and reuse, all day long, since it could very easily influence how the product tasted to them, and potentially result in unwanted variation across batches.  But would such spoon aspects matter if you were quickly eating a microwaved bowl of canned spaghetti in tomato sauce before rushing off to a soccer practice?

[jonny.reckless]

I use standard ceramic caps for 10pF -> 820pF, box polyester for 1nF -> 470nF, MLCCs for 1uF -> 10uF and electrolytics above that. The secondary characteristics of capacitors: tolerance, temperature coefficient, voltage coefficient, frequency coefficient, dielectric absorption, loss angle, ESR might be important for super high precision applications and in special use scenarios, but I wouldn't worry about it too much for anything related to this site. Decoupling of high frequency (100MHz+) digital circuits is a different matter entirely as you care about resonance, complex impedance and Q there.

I used to work in the high end HiFi industry where people claimed they could hear the difference between different dielectrics. Nobody ever differentiated consistently in a blind test, to my knowledge. Just remember that NPO > X5R / X7R > Z5U in terms of relative 'perfectness'.

Oh, and you'll probably not hear the difference between 51pF and 47pF in a screamer 

[PRR]

.....it has a 51pf capacitor, but here, were I live, there are only ceremic capacitors....

Up to 1,000 pFd, ceramic IS fine, nearly perfect.

To go much over 1,000 they have to 'salt' the ceramic with funny polar chemicals, which makes an imperfect capacitor.

Which backs-up Jonny's "standard ceramic caps for 10pF -> 820pF".

If you need a "boutique" part in <1000pFd, Silver Mica used to be good. A wee bit less good than mature (1960+) ceramics, but what a great name! However the world is running out of easy-pickings mica, also there's a lot of child-labor in the mica mines (not necessarily in capacitor mica, but nobody knows the whole truth). The coating on silver-mica is often prettier than the coating on low-price ceramics.

[Phend]

But they look cool

[Rob Strand]

OK for small values but for audio you should also be using COG (or NPO) types because of the low voltage coefficient.   For some positions in the circuit you might be able to stretch it to X7R types.   Small values automatically force you to choose COG anyway.  The not so small values is where people slip-up.

You should be able to find many similar recommendations on the web.
https://www.diyaudio.com/community/threads/ceramic-capacitors-in-audio-signal-paths.317685/

[anotherjim]

COG MLCC can exceed 100nF. Certainly, many low-cost pro-audio products now don't contain any film caps. MLCC for small values and electro for large.

[antonis]

But they look cool

Definately, yes...

[amptramp]

But they look cool

Definately, yes...

I think you got two of them switched.  Third row down, second from the left, labeled 220 pF is the 39 pF (orange, white, black) and the one third from the left, labeled 39 pF is 91 pF (white, brown, black).

[Phend]

Don't know if these come in 9v.
But they do compliment tropical fish.
Tough to get these slimy things in the effect box.
Wear rubber gloves and keep one hand in your pocket and keep your pants zipped.

[duck_arse]

.... and keep your pants zipped.

what if you are fly fishing?

[Processaurus]

As has been touched upon the problem with the common X7R ceramics is the capacitance will change with the applied voltage, not a problem for DC circuits but  for AC (audio) it ends up contributing distortion. This is the worst when used as a coupling capacitor with large voltage swing.

The C0G type ceramics are better for audio. You see those in modern mass produced equipment like audio interfaces and mixers.

Tantalum caps are reportedly not great for audio coupling, compared to electrolytic, and cost more. 

Film caps are cheap and perform very well for audio, the only complaint is they aren't commonly used in reflow oven ready surface mount (expensive, moisture sensitive), so mass manufacturing looks for other types.

For my bootweaking production I do surface mount boards with SMT electrolytics for power supply decoupling and large coupling caps (>1uF) for the audio path, SMT ceramic X7R in power supply decoupling and RF shunt positions, but thru hole film caps for the audio path. Lately I've had good results with the Kemet box type film caps from Mouser. After reading about the distinction between types of ceramics though, I'll use C0G ceramic caps for the audio if the film cap won't fit or isn't available in a needed oddball value.

More reading:
https://www.ti.com/lit/an/slyt796/slyt796.pdf?ts=1656220907124

Will ceramics ruin your fuzz pedal? Debatable at best, but I had good luck doing the distortion intentionally and keeping the rest of the circuit, like filters for toneshaping hi-fi.

[Rob Strand]

Definately, yes...

The fishlips capacitor wall chart.

For those too young to remember, Philips use to produce these, (look closely at the caps)

https://i.pinimg.com/originals/ab/8c/b7/ab8cb70015d86b2d62534eef98ac76c4.jpg
https://www.radios-tv.co.uk/wp-content/uploads/2015/09/rescap.png

[m4268588]

I don't think there's wrong with using as a coupling cap. (not HPF)

[Rob Strand]

I don't think there's wrong with using as a coupling cap. (not HPF)

What the TI article is showing is when you operate the coupling cap below the cut-off frequency you get distortion.

Distortion is caused by the capacitor voltage dependency. It can get quite bad, especially for small sized parts,

https://www.diystompboxes.com/smfforum/index.php?topic=122880.msg1160267#msg1160267

For the HPF case, above cut-off frequency the cap is more or less a short circuit.   That means the ac voltage across the cap is small and the voltage dependency doesn't have much effect, so the distortion is low.

Below the cut-off frequency the ac voltage across the cap gets larger.  That makes the capacitor voltage dependency have more effect on the output signal of the HPF and as a result you get more distortion.

That's what the TI article shows.    It is also why in some circuit positions you can get away with X7R caps.

What the TI article doesn't show but is shown in my old post is the DC bias can cause the value of the X7R (and other) caps to be much reduced.    It's possible in some circuits to move resistors after the coupling cap from gnd to vcc/2 in order to reduce the DC bias across the cap.