Power supply filter cap size

[Locrian99]

Anyone have a link to a decent article on the effect of power supply filter size. Ive always just kind of assumed in general you would want as small as possible for the job at hand. But ive never really seen the understood why some use 220 vs 100. I guess id think bigger is always going ti be better for filtering out unwanted ripple, but what is the trade off?

What if i used a 22 for c103 here in the caesar? Did they only go that small since at that point its pretty well filtered? Etc etc

 

[manfesto]

I guess id think bigger is always going ti be better for filtering out unwanted ripple, but what is the trade off?

Physical size.

 

[Locrian99]

Found thread in test kitchen…

 

[PunchySunshine]

Anyone have a link to a decent article on the effect of power supply filter size. Ive always just kind of assumed in general you would want as small as possible for the job at hand. But ive never really seen the understood why some use 220 vs 100. I guess id think bigger is always going ti be better for filtering out unwanted ripple, but what is the trade off?

What if i used a 22 for c103 here in the caesar? View attachment 73858Did they only go that small since at that point its pretty well filtered? Etc etc

 

[JTEX]

You can use a 22u for C103 if you want. It just filters Vref_B. A bit more subtly, it also shorts out the hiss generated by R100 & R101, but in this case it's not much, coming from 10k resistors. You can make an informed choice about the value you need by calculating the cutoff frequency of the filter that the cap forms with the resistor feeding it. 10k and 10u will give you 1.6Hz. Plenty low enough that il will remove all hum from Vref_b. No need for a bigger cap (unless the circuit uses Vref_B as a low impedance return path for audio, which is another possible subtlety).

R102/C104 is another RC filter. I haven't looked at the whole circuit, but if it draws little current, say <10mA, I would consider increasing R102 for better filtering (and lower liklihood of it or the Zener burning up from supply overvoltage).