Polypropylene vs polyester vs polystyrene caps? What's better?

[Lew-Dawg]

Gents,

Just made it back to shore after a chopy night on the bay.

I don't understand what the audible difference would be from capacitors made from different materials.

Most of the work I do is simple cap replacements for me and my buddies.

Should I aim for a certain material? Polypropylene? Mylar even? I'm talking tiny 50-70v caps. Is it more of a longevity thing?

Lewis

[BetterOffShred]

Do a search for this topic.  Every 6 months or so there is a 5 page thread on it.     I use whatever I have that's the right size and rated to handle the voltage seen by the component, and finally a package that will fit where it needs to go.

[EBK]

Some people are cursed with ears that make different types of capacitors sound different in audio circuits.  Some of these cursed individuals even appear to experience excruciating pain when their ears are exposed to sound from mediocre audio circuits.  I'm blessed with ears that enjoy whatever is cheapest.

That said, if I'm looking at a build doc that recommends a specific type of cap, I blindly follow that advice.

[Mark Hammer]

If it's a clean pedal (e.g., preamp for sampling mic, or an EQ), cap type may make a difference.  But the majority of builders here preoccupy themselves with things that produce distortion.  And in that context, cap type likely doesn't matter very much.  I mean, cripes, it's fuzz, and it's rock and roll.  All the distortion and fuzz pedals you just wish you had an original of used big ugly ceramic caps.

[ElectricDruid]

The "accepted wisdom" is that polyester caps are ok, polypropylene are better, and polystyrene are the best.

Now let's talk about a few of the reasons why that's not likely to actually be important and why that "wisdom" isn't as accepted as all that.

a) We're talking about small differences. As Mark said, in a fuzz, who cares? No-one will *know*, let alone hear it.
b) A lot of the time it's more to do with tolerances than specific type. Polystyrene caps used to be available in closer tolerances than some other parts (+/-2% even) and *that* can make a difference for things where it matters, like multi-pole filters. In that situation, better matching between the filter poles can *sometimes* be audible. Consequently polystyrene got used, but as much for the tolerance as the material.
c) Caps aren't the only variable by a long shot. If you use fancy caps and then put 5% carbon resistors in the circuit for "mojo" you can forget any benefit you might think you're hearing. Same deal if you decide that a vintage 741 op-amp or CA3080 OTA is the way to go. Why fuss about caps if you've got some other part in the circuit that'll mess with with the sound way more than the caps ever would?
d) What's "better" anyway? Film caps of all types have a sound I'd describe as clean and transparent and perhaps even "modern". That might not be what you're after.
e) It depends on the application. For decoupling radio-frequency noise across power supply rails for example, film caps are not appropriate. Ceramic is used for this because it's the best thing for the job.
f) That classic vintage pedal you desire that costs an arm and a leg used the cheapest bin-end parts they could get, and they slung any old thing in there from one week to the next. And now everyone goes wild about it. So why should we worry so much now?

I'm sure there are others, but you get the general idea. There might be some rule of thumb, but unless it really applies to your particular situation, it's not worth a great deal.


Ok, now a question: Choppy night on which bay?!

[amptramp]

Polystyrene caps were originally specified when you needed a precise capacitance that stayed within 5% of rated value over the temperature range, but the temperature range is low, going to 85C max, so you have to be careful about soldering them.  They have a temperature coefficient near zero so they were commonly used for oscillators and filters but have become less common now that other film capacitors have been used with higher temperature tolerance.

Polypropylene is common and it has a temperature limit of 105°C.  They have the lowest dielectric absorption which is a phenomenon where you can apply a voltage to a capacitor, short it out, remove the short and some of the voltage reappears.  This makes it suitable for sample-and-hold circuits.  They have a high dielectric strength meaning that high voltage values are available.

Polyester capacitors are made of polyethylene terephthalate, the same material used in soft drink bottles.  It absorbs little moisture so it is often seen without a coating, particularly in European radios.  It is suitable for operation at 125°C but there is a frequency dependence where the capacitance measured at 100 Hz may be different from the capacitance measured at 100 KHz.

Polyethylene naphthalate is another polyester with higher voltage limits and a temperature rating of up to 175°C.

There are a lot of other film capacitor dielectrics but for our use, any one of them will do.

[vigilante397]

Multi-layer ceramic. All the way

Especially if you're into that SMD scene.

[EBK]

One of these days I'll make a pedal with a homemade capacitor in it.  Thinking either a cut up mylar party balloon or a chewing gum wrapper.   

[Mark Hammer]

In junior college, we learned that, in the absence of voltage-meters, Benjamin Franklin would "measure" capacitance in his home-made Leyden jars by discharging them on chickens.

Fortunately, for Mr. Franklin, the "tongue test" did not emerge as a viable crude measurement strategy until many decades later, else I fear his scientific career would have been severely shortened.

[EBK]

In junior college, we learned that, in the absence of voltage-meters, Benjamin Franklin would "measure" capacitance in his home-made Leyden jars by discharging them on chickens.

Fortunately, for Mr. Franklin, the "tongue test" did not emerge as a viable crude measurement strategy until many decades later, else I fear his scientific career would have been severely shortened.

It's a good thing kite-flying electrical testing had advanced to a very safe level during his scientific career.   

[Tony Forestiere]

Fortunately, for Mr. Franklin, the "tongue test" did not emerge as a viable crude measurement strategy until many decades later, else I fear his scientific career would have been severely shortened.

As if that old "Kite and Key" trick didn't do him in. Sometimes, you've got to run with scissors.

*edit* Eric beat me to it...

[EBK]

This talk about Leyden jars and homemade caps sparked a memory.  In one of my college Physics classes, a group of students demonstrated a homemade Tesla coil that had a capacitor made out of a champagne bottle filled with oil, wrapped in foil, with a large nail through a cork.  It worked amazingly well.  Another group of students made a one-string electric guitar (Smoke on the Water was their chosen demo riff).  There was also a rudimentary rail gun (it made a staple twitch about an eighth of an inch) from yet another group.  And, one student suspended a bubble in the middle of a flask of water using ultrasonic transducers.  My group, um...  made nothing*, but we basked in the genius glow of the others.

*We must have presented something, but I can't remember what it was.

[Rob Strand]

What about rare poly-purple-ene caps.  The purple film blocks any harmful signals passing between the plates. 

[thermionix]

The secret to Hendrix' tone!

[diffeq]

One of these days I'll make a pedal with a homemade capacitor in it.  Thinking either a cut up mylar party balloon or a chewing gum wrapper.   

Add the homebrew diode to that and you get a true vintage sound.  Bonus points for pencil resistors. 

[amptramp]

One of these days I'll make a pedal with a homemade capacitor in it.  Thinking either a cut up mylar party balloon or a chewing gum wrapper.   

Add the homebrew diode to that and you get a true vintage sound.  Bonus points for pencil resistors.

One of the radios in my collection has a coherer as a detector diode.  This predates the crystal set.

It is rumoured that there was a FET as early as 1925 made using a carbon semiconductor.  It might go along with a pencil resistor.

[duck_arse]

I'm guessing the "bay" in question began with the letter "E".

colour is important, especially now that box-caps seem to be defaulting to grey. greencaps are good and green, sometimes brownish-red, and the outer epoxy will withstand soldering iron pokes. but the leg spacing is inconsistent and the fatness varies greatly. and don't bend the legs too close to the body.

box caps can be had in colours like yellow and red and green and blue/silverish, but the shell is prone to melt if soldering ironed. but the pin spacings are predictable, as are the thicknesses.

polystyrenes, at one stage, were silver/green, tubular Philips. melty inside and out. stacked film caps always had me baffled - why not encase them? I was less baffled once the legs fell off the endplates.

ceramics used to be much more interesting than they seem now, in that the diameter could be nearly anything, totally disproprtionate to the value. nowdays, I see all the same, tiny, no matter the value (tayda, I'm looking your way). and don't bend the legs too close to the body.

[2 cents well spent on my part, I think.]

[Mark Hammer]

What about rare poly-purple-ene caps.  The purple film blocks any harmful signals passing between the plates. 

I didn't know they made Thrills gum with component leads.

[EBK]

"It still tastes like soap!"
WTF? 

Never heard of Thrills gum, but I'm looking it up now to make my life more complete.   

[vigilante397]

Never heard of Thrills gum

Me neither, judging by the French translation I'm assuming it's a Canadian thing