|
If you know the input impedance of the load that will be fed by this filter, then you don't need the shunt resistor.
Use a calculator that has a "1/x" button. That will save you lots of grief.
Your formula is: 1/(20000 * 100 * 2 * π)
Multiply out the denominator first (20k * 100 * 2 * π). Then hit the 1/x button on your calculator.
You should get 0.000000079577 Farads. Multiply that by a million to get microfarads. Approximately 0.080µF. Each capacitor will need to be double this value, or .16µF. The nearest common value is 0.15µF, so I would probably use that. If you want it to be exactly 100Hz, then you could round the capacitor value UP and then bring the R value DOWN by adding a shunt resistor. But this is a gradual filter, so I don't think you'll notice too much if the filter works at 106Hz instead of 100. So sticking with 0.15µF caps will be fine. If you're worried about it, 0.16µF caps can be found, or you can put a 0.01µF cap in parallel with each 0.15µF cap.
In this range of capacitor values, you're going to be looking at film dielectric. Polyester is the most common, but polypropylene performs a little better and should be fairly easy to find in this size. The voltage rating isn't too important, unless you expect to see some significant voltage (like if you were going to use it on a mike preamp input that had phantom power active - which would be pointless because the capacitors would prevent the phantom power from reaching the microphone. Polarity is not an issue for film capacitors, so it doesn't matter which way they face. What does matter is that the two capacitors be well matched. You can pay a little extra to get caps with a 5% or even sometimes 1% tolerance, which will give better noise rejection than caps with say a 10% tolerance. But another option would be to buy a handful of caps with 10% tolerance and measure them with a multimeter that measures capacitance. Pick two that give exactly the same reading on the meter. They don't have to be the two that are closest to the nominal value, they just need to be the two that are closest to one another. If you don't have a meter that measures capacitance, then don't worry about it. Just by the caps with the tightest tolerance you can find.
If you go to digikey.com and search for "capacitor" you'll find a bunch of things. Choose "poly film" and you'll get a parametric table that lets you narrow down the choices to help you pick something. You can choose a few different capacitance values (like 0.15, 0.16, and 0.18µF, for example); choose to only see parts they have in stock (they have thousands listed that are not stocked); choose only through-hole parts if you don't want to screw around with surface-mount caps). Choose parts sold in bulk or on cut tape so you don't have to buy a full reel (which is several thousand pieces).This will get you down to 128 choices and now it's time to make some decisions. If you don't have a capacitance-measuring multimeter, then select tolerances between 1% and 5%. If you know something about the space in which you have to fit these two caps, make some decisions that way (axial-lead caps may be easier to work with, but the selection is limited. Radial-lead caps may require a bit more space). Polypropylene caps will be physically larger and more expensive. If you're trying to fit this into a barrel connector, you may have to use axial-lead metallized polyester caps such as B32232A1154K. (Metallized film caps are generally more compact.) If space is not an issue for you (if you're building it in a little box or on a rack panel) then you could choose a 1% tolerance, polypropylene film capacitor such as ECQ-P1H164FZW.
|