I'd narrow it down by capacitance, voltage, dielectric, and physical size/pin pitch as you've already done. Then it usually comes down to tolerance, price, and availability for me. Occasionally I'll look at the temperature coefficient and voltage coefficient (if specified, which it rarely is). Brand/manufacturer factors in as well. I would much rather buy a WIMA cap than a Lelon or some other no-name brand.
Note that you'll find two kinds of film capacitors: Metalized film and metal foil.
I wouldn't worry about the material of the pins. I've yet to see any hard evidence that one material is better than the other as long as the material is conductive and that solder will stick to it.
Tom
Note that you'll find two kinds of film capacitors: Metalized film and metal foil.
- Metalized film capacitors use dielectric film that has had metal deposited on it. Often methods like sputter deposition and chemical vapour deposition are used. The result is a very, very thin (a few molecules thick, probably) film of metal on the dielectric.
- Metal foil capacitors use a think metal foil sandwiched between layers of dielectric. The foil is much, much thicker than the metal on metalized film caps so foil caps tend to be physically larger. They also tend to have lower ESR than metalized film caps. ESR is usually not a huge concern in audio, but it can be in some applications.
I wouldn't worry about the material of the pins. I've yet to see any hard evidence that one material is better than the other as long as the material is conductive and that solder will stick to it.
Tom
I am sold on COG - and especially in SMD form. Prices have come down the last 10 yrs or so and they are available in tight tolerances. I never understand why people insist on using film when COG is available.
You're right in general, but in this particular case it won't matter at all.
After listening to the Bruce Hofer presentation in its entirety, also going to 1206 resistors.
Thanks for posting the video Tom!
Generally the larger parts perform better. The same goes for 1206 vs, say, 0402 capacitors.
I also find it interesting that the C0G parts show much less sample-to-sample variation on H2 than the film caps in Hofer's video.
Tom
I also find it interesting that the C0G parts show much less sample-to-sample variation on H2 than the film caps in Hofer's video.
Tom
Another thing not mentioned so far: PCB's HATE multiple desoldering/soldering. They usually survive a couple of times, then the pads may come off. Not fatal, but ugly.
Your amp may be less than excellent design needing a shrill sounding cap to increase treble. No muffled sound or lack of treble in my tube amp with K40Y-9. There is no reason a paper/oil cap of any brand would be muffled. K40Y-9 are mil-spec with glass sealing ends and the metal case can be grounded wrapping wire around the body and applying heat shrink that is valuable in phono sections. Compact size will fit in any amp.That`s your opinion, i think they muffled and lacking treble. Polypropylens are way to go. CDE 940 series are good caps.
BTW, COG/NP0 SMD are generally available up to 250 VDC and some types 500V. You can get 5mm leaded devices at even higher voltages. Once values go above 0.1uF, the field thins a bit, but Mouser have a great selection. They are perfect for RIAA networks, amp compensation, and general filter work. For PSU local decoupling, still stick Roth X7R or similar.
Self did some thin film 1206 resistor tests on his AP and found they were as as good as high quality leaded devices. Go figure. 0402 starts to lose the plot a little bit but this only arises if there are large signal excursions across the resistor which in line level is not normally the case. I would struggle to place 0402 so would never use them anyway.
🙂
Self did some thin film 1206 resistor tests on his AP and found they were as as good as high quality leaded devices. Go figure. 0402 starts to lose the plot a little bit but this only arises if there are large signal excursions across the resistor which in line level is not normally the case. I would struggle to place 0402 so would never use them anyway.
🙂
You are so funny. I have read so many your post that i know what is your knowledge. Now i put you to my ignore list.
by "polyethylene" do you mean polyethylene terephthalate, or polyethylene naphthalene?
Ken Stevens from Convergent Audio has made and designed his own caps for decades, and has made similar comments about teflon/polypropylene, and the tendency for oil caps to muddy HF.
these caps are not compliant with security due the use of a mineral oil that can produce dioxine with heat ( and explosion) and there are no so good for sound
Much better on Wima or Vishay cap
On Digikey a lot of choice
Walter
Scott Wurcer (RIP) and Stuart Yaniger - both seriously experienced engineers - explained at length on the forum why DA does NOT cause distortion and could not materially affect the sound. What will affect the sound and introduce distortion are capacitance non-linearities with applied voltage, something that happens with X7R (primarily a voltage vs capacitance mechanism) and with electrolytic capacitors at LF as voltage develops across the terminals. Cyril Bateman wrote the definitive set of technical articles on capacitor distortion IMV.
According to Horowitz & Hill, PIO capacitors have high voltage coefficient. They speculate that this is due to the paper being slightly squishy. The electrostatic field set up by the applied voltage causes the two foils in the capacitor to attract, so the distance between them is shortened, which increases the capacitance.
In an audio circuit, capacitors with high voltage coefficient will have a signal-dependent error voltage in addition to the signal -> higher distortion.
Tom
In an audio circuit, capacitors with high voltage coefficient will have a signal-dependent error voltage in addition to the signal -> higher distortion.
Tom
What do you think of these Epcos TDK?
I looked into wima and Vishay but it seems buying these when you live in Australia is a problem. You can only buy in bulk (750 or over) from all the websites I’ve seen. These Epcos TDK can be purchased in single units. So it seems like a good option.
Attachments
For one example, in some cases ESR can be too low to avoid ringing on transients. A series damping resistor can be added but then inductance increases. In certain cases SMD film can be a good choice. Depends.