Discussion in 'Tube Audio' started by drtool, Jan 2, 2018.
Alan, The value drift with temperature is a good reason to use polysty. as they are very stable with temperature.
I don't use them for coupling caps because I can't get high enough values in polysty. They generally stop at 10,000pF. Plus the small, clear ones are rated 50v. Russia does sell 450v versions though , I believe. I use polypropylene for couplers. Or Russian paper in oil caps.
Unless I'm badly mistaken, C0G/NP0 caps are used in RF circuits specifically because they don't drift with temperature. Other class-1 ceramic formulations are available with controlled drift characteristics, to compensate thermal drift in precision tuned circuits.
Ha!!! Tell me more, I love to be wrong on this.
NP0 means negative-positive zero in terms of capacitance change.
There's a table of the various available temperature coefficients for class-1 caps in this article on Wikipedia.
Guys, I am happy to be wrong. I assume all ceramics drift!!!
So the next question is why are we still using film caps for low values values like below 0.0022uF?
Have seen many caps in the FB path and they are usually Silver/Mica. The question is 'why'? Do film caps have more ESR or ESL that can do things that are less than useful? Consider just exactly what these have to do, and what makes the Mica caps useful. Then plug it in and have some measurements and a good listen. There are more SS amps out there with vanishingly low distortion figures that are quite un-listenable than I care to comtemplate...so don't count on a single measurement to tell the tale, yes?
Yes Douglas. Having La Scallas new in 82. Having a number of SS amps, none sounded anywhere as good with them as my DG SE Maggie.
C0G/NP0 caps are second to none for most parameters. Oddly, though they can make them have close to zero TC, very few are dead on zero. They actually seem to shoot for slightly off zero, but it's still very low. Now you can get them in MLCC form, so higher values are available than used to be. Silver-mica is a strange beast. They have very low losses (dissipation factor), yet have higher than expected dielectric absorption. I don't detect any sonic problem with them, but one has to be a bit suspicious. They're getting quite expensive considering they only compete with C0G/NP0 if you need higher voltage or high values- though those tend to be huge and crazy expensive. I think the largest I've got is 65000 pF, 0.5%.
There's a trick used for precision measurement bridges like some old General Radio. The TC of silver-mica goes one way. The TC of polystyrene goes the other. If you proportion a parallel combination of the two correctly, the resulting TC will be near zero. That's how they got their reference caps, usually about 0.15 uF, to be temperature stable.
It's always horses for courses. Not every circuit needs a low loss capacitor, and some will work very poorly if you use one. Not every circuit needs great TC, though it's rarely a bad thing.
Surplus mica from Russia can get pretty big, and at useful voltage. ebay used to have lots and lots...though like the .1 uF/500V Teflon foil caps, they have gotten slightly rare.
0,1 uF/500V silver mica from alexr
Separate names with a comma.