(bold mine). I think (I know, that is) it's the other way 'round. Don't parallel low ESR caps with more than 10:1 capacitance ratio (assuming increased ESL from the bigger packaged ones).
Only way to really know if you are safe or not when paralleling low ESR caps is either an in-situ network analyser plot or a close look at a bell-ringer time response, in-situ as well. Otherwise stick with a high-ESR tant/'lytic plus a single low-ESR film/ceramic.
The caps I normally use are all quite small, generally 0402 and 0603, 0805 at the largest. I've encountered no problems following the 10:1 rule, and yes it has been examined with an oscilloscope and spectrum analyzer, as I do this sort of thing for a living. It has also been simulated using Genesys. YMMV, and all that.
Last edited:
Yep, the trick is that the inductive slope of the large one is swamped out by the ESR (or additional external R) so that it cannot create a problem at the intersect with the capacitive slope of the smaller cap. Reminds me about opamp compensation, more than 20dB/dec slope difference at Aol 1/beta intersect --> potential problem.
No doubt here, as with same ballbark inductance the resonances are only apart by a factor of 3 or 4, problems start at 10x++. My comment was more like when people try to bypass one bigger "audio grade" THT film, notably with axial leads, with something too small (both capacitance and ESL).
Ah well; as I said I would never use leaded parts for bypass with high speed devices. It's just asking for trouble.
So does that mean all the money i have been spending on buying high quality polypropylenes is wasted, that i might as well have stuck to using a high esr al electrolytic all along, potentially better.
If a higher esr cap is better why use ceramics when al electros are more commonly available and in a wider range of values.
No context means no comment.
However, it is certainly true that some purchases of 'high quality' capacitors for audio use are a waste of money. Some may be worse than that; the 'high quality' capacitor might actually be a worse fit for a particular circuit duty than a perfectly ordinary cheap component. Good engineering is the answer.
However, it is certainly true that some purchases of 'high quality' capacitors for audio use are a waste of money. Some may be worse than that; the 'high quality' capacitor might actually be a worse fit for a particular circuit duty than a perfectly ordinary cheap component. Good engineering is the answer.
Ceramics are smaller and they don't wear out, so for some applications where you don't need large capacitance, a ceramic is a better idea than an electrolytic. Because they are physically smaller, they will also have a higher self resonant frequency, making them a better fit for some types of supply bypassing.
High ESR is better when the circuit calls for high ESR; elsewhere high ESR may be a problem or it may not be a problem.
- Status
- Not open for further replies.