Another R-C Filter Question

[Nick C.]

Let's say I want a high pass filter with a cut off of about 720Hz. Is there any difference between using a 100k-.0022uF pair vs 47R-4.7uF pair? I've been wondering if there is any difference in volume or slope after the cut off point since the resistor to ground is so different? Sure would help in part selection.

Thanks in advance,

Nick

[R.G.]

If you consider the frequency response of the resistor/capacitor ONLY there is no difference. The cutoff point, slope, etc. are the same.

There is obviously a big difference to whatever drives the filter and whatever takes the filtered signal out of the filter.

Driving the filter is an issue. The source impedance of the signal source appears in series with the source. The cutoff frequency and volume ARE affected unless the source impedance is less than about 1/10 of the impedance of the filter - in this case less than 1/10 of 100K or 10K. That's easy enough. But driving a 47R/4.7uF pair requires a source impedance below 4.7 ohms to be neglegible. That's a lot harder.

Obviously, a lot of signal current goes through a 47R/4.7uF pair, even if there is no load.

Then there's the issue of the load. To avoid affecting the cutoff and slope, the load must be greater than 10x the impedance of the filter pair; effectively this is 1M for the 100K/0.0022 and 470 ohms for the 47/4.7. Obviously, the low impedance filter can supply more current and drive a lower impedance load with no apparent degradation.

Since there are no zero-impedance sources, and no infinite-impedance loads, the source and loading impedance of every network have to be considered, at least to the point that you can decide they can safely be neglected or not.

[Nick C.]

Thanks RG for the quick reply.

Any approximations on what a "typical" source impedance would be coming from the op amp output of a Dist+ or Shaka?