Typical capacitor values for audio

[Philbo King]

So - I'm learning a lot here as per usual. I might still search for starter kits, but bothering to read up and try and pick parts to suit my needs is... pretty handy!

Shelf Life... got me switched on. Here's a good quick read:

https://www.google.com/url?sa=t&sou...8QFnoECBkQAQ&usg=AOvVaw1HqCk6fmC8rh5fX2O8awSv
(That link is massive sorry!).

The "de-aging" section bottom right in particular was very new to me!

While tantalum have the enviable shelf, I'm not too keen to blow too many thing up at first haha... So - I'm thinking about mainly C0G for audio path stuff. Any thoughts? I'm happy to buy more specific stuff later, but for tinkering and bread boarding, it would be good to have a longish shelf life stock of audio related values.

Electrolytics?... not sure just yet. They're pretty short on the shelf it seems. I'm expecting to get a variety, but not really "stock up".

C0G & NP0 caps are generally less than 100 pF, so they have little applicability for audio.

 

[RCAguy]

For a quick calculation (even in your head) of a common 1st order RC combination, I multiply µF times Ω. A result of 100,000 implies a corner frequency (–3dB) of 1.59Hz that translates to a small fraction of a dB at 20Hz for audio DC blocking (signal-passing) or PS decoupling capacitors (perhaps larger for LF peak signal stiffness). The small voltage drop at 20Hz and above also means little added distortion. E.g. 1) For blocking\signal-passing, 10µF looking at an impedance of 10k "cuts off" at 1.59Hz. 2) Similarly 100µF seeing 1kΩ. 3) But 1µF seeing 1kΩ corners at 159Hz. 4) A 'lytic main filter in a linear PS supplying 500mA at 15vdc (+ or – side if bipolar) feeds a 30Ω impedance, so it needs to be at least 3300µF (3300*30=100,000) to reduce 120Hz full-wave ripple –35dB.

 

[JohnRoberts]

C0G & NP0 caps are generally less than 100 pF, so they have little applicability for audio.

In a quick search I found 0.22uF and 0.15 uF cog capacitors at digikey... most are SMD, but there were a few radial thru hole components from TDK.

JR

 

[JohnRoberts]

For a quick calculation even in your head of any 1st order RC combination, I multiply µF times Ω. If the result is 100,000 then the corner frequency (-3dB) is 1.59Hz, which works well for audio DC blocking (signal carrying) and PS decoupling caps (main PS filters may be larger for sufficient LF peak signal stiffness). As examples: 1) For blocking\signal-passing, 10µF looking at an impedance of 10k "cuts off" at 1.59Hz. 2) Similarly 100µF seeing 1kΩ. 3) But 1µF seeing 1kΩ corners at 159Hz. 4) A 'lytic main filter in a linear PS supplying 500mA at 15vdc (+ or - side if bipolar) appears to feed a 30Ω impedance, so needs to be at least 3300µF (3300*30=100,000) for 1.61Hz ripple - scale accordingly.

+1... back when we designed circuits using slide rules, we had to do the rough math filter calculations in our heads, and then let the slip stick (slide rule) provide the significant digits.

JR

 

[Cqwet Dbdfte]

Hi gang - another basics question from me ...

After years of digital tinkering in spice I'm deciding to go all in this coming year and want to start doing what most of you do... lol... actually building stuff !! Sure, I'm good with an iron, comfy building kits etc... but these silly spice projects just seem cool and I want to start making them real.

Silly as it sounds (to me), I'm only really just getting head around resistors (not basics, but more what all the crap in data sheets actually mean). Capacitors... I know what they do etc, know the types... but I'm not all over it in terms of tinkering.

I'm about to load up on bits and pieces to start breadboarding things, mainly compressor designs. But wouldn't mind just getting a random foray of capacitors for audio related fun - eq / filter related.

Is there a kind-of good "list" of cap values (and perhaps type / value combo) to have on hand? Like, do you all have a pile of "go-to" values and types for general use, or is it more project specific??

Almost changed the title to "general component values" ... but I'll keep it capacitor related. I might come back with diode and transistor questions!

No worries if it's more project specific - happy to trawl through schems more to get a feel (which I've been doing, but super keen to just finish this mouser cart and thought I'd reach out in case you all just stock up on some standard values).

Many thanks for any guidance!!

Typical value would depend on what function the cap has in the circuit, like coupling, decoupling, tuning etc, and expected frequency range. Caps are also "inductors" and "resistors" so parasitic properties play in. Coupling caps are high pass filters and have to be dimensioned for low frequency rolloff with the loading resistance considered. Your spicy sim would show that. Good spice models include parasitic properties.
Film caps should have outside foil end marked.
NPO COG caps have great properties, temp stability etc, but are seldom used in audio, unnecessary and expensive, when PP or teflon would do.

 

[RCAguy]

+1... back when we designed circuits using slide rules, we had to do the rough math filter calculations in our heads, and then let the slip stick (slide rule) provide the significant digits.

JR

Yep!

 

[Gold]

In a quick search I found 0.22uF and 0.15 uF cog capacitors at digikey... most are SMD, but there were a few radial thru hole components from TDK.

The TDK 0.22 was the largest value C0G I found. I'm using them in the SVF sections of the Barry Porter EQ. I've been using it for over a year and I am very happy with the sound.

 

[JohnRoberts]

The TDK 0.22 was the largest value C0G I found. I'm using them in the SVF sections of the Barry Porter EQ. I've been using it for over a year and I am very happy with the sound.

they should not contribute any sound character of their own... a very linear dielectric.

These puppies did not exist back when I was actively designing audio circuits.

I struggled finding film SMD that would survive reflow about 10-15 years ago.

JR

 

[Whoops]

But to my opinion the recent class of (budget) converters sound great, big improvement in sound quality compared to the older generation of yesteryear.

I agree, present day converters, even budget ones sound great, and all of them will be SMD.

I can see that I'm not coherent in my perception or opinions on this, as I have a Focusrite Red4Pre soundcard, and I was amazed with it's sound the first time I listened to it, it sounds incredible and I love it. Love the A/D, D/A, Headphone Amp and Mic Preamps, it's just that every section sounds Great.
Well I never opened the unit, but I'm pretty sure it's completely SMT.
Maybe the fact that I never saw the circuit board or that I never though about it didn't affect my perception and judgment of the sound.
But I have to honest to myself, so SMD can actually sound pretty good and my judgement is cluttered when I visually see the circuits, I guess that's the reality

 

[Gold]

they should not contribute any sound character of their own... a very linear dielectric.

It is a very clean, clear and neutral sound. Not what most want out of an analog EQ these days but it suits my need well.

 

[Cqwet Dbdfte]

Polystyrene is a great dielectric for audio, but how well does it age?

 

[JohnRoberts]

Polystyrene is a great dielectric for audio, but how well does it age?

Polystyrene dielectric does not have aging issues AFAIK, but I have encountered issues with poor temperature handling (mostly the mechanical package). I also experienced an issue back in the 70s/80s where my business partner, while cleaning off PCBs with a high pressure air hose blew water moisture inside the caps, ruining them. I am not aware of them absorbing moisture without a high pressure air hose assist.

I would have loved using polystyrene caps while at Peavey (because of the relatively low prices and good dielectric) but they were not robust enough to survive wave soldering.

JR

 

[Mr. Moscode]

Polystyrene dielectric does not have aging issues AFAIK, but I have encountered issues with poor temperature handling (mostly the mechanical package). I also experienced an issue back in the 70s/80s where my business partner, while cleaning off PCBs with a high pressure air hose blew water moisture inside the caps, ruining them. I am not aware of them absorbing moisture without a high pressure air hose assist.

I would have loved using polystyrene caps while at Peavey (because of the relatively low prices and good dielectric) but they were not robust enough to survive wave soldering.

JR

What JR said about aging. I would guess water infiltration would depend on who made it. We use small value 100pf and 47 pf 500 v caps in the Moscode amps and washed the stuffed boards in a dishwasher and never had a problem. What they were used for was HF rolloff in the nfb loop and also between stages plate to ground for stability. We also used .01mfd 500 V in a preamp as bypass caps across coupling caps and a passive RIAA net with no problems in the 80s.

 

[JohnRoberts]

What JR said about aging. I would guess water infiltration would depend on who made it. We use small value 100pf and 47 pf 500 v caps in the Moscode amps and washed the stuffed boards in a dishwasher and never had a problem. What they were used for was HF rolloff in the nfb loop and also between stages plate to ground for stability. We also used .01mfd 500 V in a preamp as bypass caps across coupling caps and a passive RIAA net with no problems in the 80s.

Somewhere I picked up a few 0.1 uF polystyrene capacitors. These were about the diameter of my thumb. I used lots of <0.01uF polystyrene values in RIAA stages and EQs back in the 70s/80s.

JR

PS: I suspect the high pressure air hose was at fault for the moisture contamination. No more problems like that after we stopped using high pressure air to clean PCBs.

 

[Whoops]

PS: I suspect the high pressure air hose was at fault for the moisture contamination. No more problems like that after we stopped using high pressure air to clean PCBs.

You didn't have a Moisture Filter installed in the high pressure machine?
Or the Moisture Filter was not enough to reduce the moisture?

 

[Whoops]

C0G & NP0 caps are generally less than 100 pF, so they have little applicability for audio.

Sorry mate but that's not correct.

C0G & NP0 capacitance can range from 0,5pf to 0,1uf. So they can go to 100000pf

They see common use in Audio electronics

 

[JohnRoberts]

You didn't have a Moisture Filter installed in the high pressure machine?
Or the Moisture Filter was not enough to reduce the moisture?

IIRC the shop air had oil filters (but this was 40+ years ago).

The water that was blown inside the polystyrene caps was water residue on the PCBs after washing off water soluble organic flux. (Don't get me started on organic flux problems). The polystyrene packaging was not very secure. Great sounding dielectric, but not compatible with mass production.

JR

 

[Whoops]

I see,
polystyrene caps seem to be quite sensitive in a lot of different areas.
They can't seem to withstand much heat from the soldering iron also, my main Iron is 30watts that I use for everything, but when I need to solder a Styrene cap I have to use the 11watts iron otherwise 1 second more with the iron can make them fail.

 

[JohnRoberts]

I see,
polystyrene caps seem to be quite sensitive in a lot of different areas.
They can't seem to withstand much heat from the soldering iron also, my main Iron is 30watts that I use for everything, but when I need to solder a Styrene cap I have to use the 11watts iron otherwise 1 second more with the iron can make them fail.

yup the really thin wire leads on polystyrene caps are not an accident, thicker leads would conduct more heat to the low melting temperature styrene.

But they do sound clean.

JR

 

[Cqwet Dbdfte]

Dielectric absorption is very favoable in Styrene caps. Have not used them audio, nor any ceramics, some SMT caps are microphonic, pressure sensitive. Leaded ceramic may be OK.
I have an old stash of 7uF 300V (mylar?) with very low ESR, like 0.008 Ohm, I use for decoupling.