[sdiy] Wanted: non mathematical description of the function of RC-filters

[rsdio at sounds.wa.com]

I like the second example, but the first is problematic. The  
implication when an analogy is made between a sieve and an audio  
filter is that there is a binary relationship. With a sieve, either a  
particle fits through the holes or it doesn't. With an audio filter,  
it's not so black and white. Filters have a slope, and this allows  
frequencies from input to appear on the output at reduced amplitudes.

The misconception that filters completely stop all frequencies beyond  
their cutoff is such an issue that I wrote an article about it a  
while back. The article does nothing to explain the electronics of RC  
filters, but it might be interesting reading now that I've brought it  
up:

http://www.plasmodium.net/music/eq-filters-and-frequency-responses- 
are-not-perfect/

Brian Willoughby
Sound Consulting


On Aug 8, 2013, at 10:29, Byron G. Jacquot wrote:
> A couple of intuitive models from the physical world that might be  
> worthwhile to explore:
>
> The sieve:  Consider that you have a pile of mixed gravel, with  
> particles from maybe 2mm up to 50mm.  With a screen, you can  
> separate particles of a size the fit through the holes in the  
> screen.  Depending on whether you take the particles that pass  
> through the screen, or the ones that won't pass, you've effectively  
> got a "large pass" or a "small pass" filter.  With a pair of  
> screens of slightly different hole diameter (say 9mm and 11mm  
> holes), you could achieve a size filtering that either passed only  
> ~10mm particles, or everything except ~10mm particles (roughly  
> equivalent to bandpass or notch filtering).
>
> The shocks and springs in a car:  The mass of the car pushes the  
> tires to the ground through the flexible medium of the spring.   
> When terrain varies gradually (say, climbing and descending a  
> hill), the spring translates the tire position to the car body more  
> or less literally.  But when the terrain varies quite rapidly (such  
> as a heavily rutted "washboard" dirt road), the tire will track the  
> surface details, but spring absorbs the motion, and the mass of the  
> car body doesn't track those small movements.  It's a mechanical  
> lowpass filter.
>
> Do either of these help?