High pass filter in an op-amp

[microbug]

I know the title is vague but I couldn't think of how better to phrase it. I'm designing a small headphone amplifier for myself, using the op-amp circuit on this page. I understand it except for how the value of C2 is calculated, and the importance of the value (what is the disadvantage to increasing it, which would better suppress power supply ripple?). As suggested on that page, I'm using R1=100k, R2=100k, R3=1M. Could someone explain?

[uncle_bob]

Hi

C2 is a bypass capacitor. It is a "bigger is better" sort of thing up to a point. If you make it big enough (relative to R1 parallel R2) the time constant could get nasty compared to your expected turn on time ( = why does it sound like crap for the first 20 seconds?).  It is desirable to keep the R3 and C2 corner below 20 Hz. In any practical circuit that is not going to be a problem you run into.

One other hint:

It is common to try to pick one cap to be the one that sets the high pass for the amp and to keep the others "out of the way". That's not the only way to do it. In some cases it was done that way to make the math easy. In the days of LTSpice, there's no need for math

Bob

[microbug]

OK, thanks