Highpass filter - need help

[Gene]

Hi everyone,

I'm new to the forum and diy electronics, and need some help regarding passive RC high-pass filters. I've been using various calculators online and built several prototype filters for about 2 weeks now, and can't get them to work properly. I am pretty sure that I am completely screwing up when it comes to impedance.

I want to make a variable high-pass filter from approx. 20hz to 400hz. This will be a 12db per octave filter. It will be plugged into the mixing console's input channel inserts.

The console is a Soundcraft GB8-32. The insert send is 75Ω. The insert return is 5kΩ.

What capacitor and resistor values should I be looking for? Should I be looking at the insert send impedance, or the insert return impedance for selecting the appropriate resistor value, or both?

I realize this may be a very noob question, but I learn better when I have step-by-step instructions that I can also reverse engineer. Thank you.

 

[ruffrecords]

Soundcraft insert can, I belive drive a 600 ohm load. We also want to ensure the 5K return input impedacne does not excessively load the filter. With a simple two stage CR 12dB/octave HPF, at high frequencies the total load is the two filter resistors in parallel which eachother and the 5K input impedance. If we use 2K resistors in the filter then the total load is just over 800 ohms which the soundcraft insert output should be able to cope with. The 5K is only 2.5 times the the 2K so it will have an effect but you have to work with what you are given. All you need to do now is calculate the capcitance for the desired turnover frequency fron C = 1/(2*pi*f*R).  So for 50Hz this would be 1.59uF,  and for 100Hz it would be 795nF.

This is a very simple filter and will not be perfect. For starters it assumes each stage is driven by a low impedance. This is true for the first stage but not the second. It also assumes there is negligible load on each stage. This in true or neither stage. The effect of these deviations from the ideal is that there will be two slightly different turnover points, so instead of the single sharp knee you will get two knees close together. I have made lots of simple HPFs like this and I think sonically there is nothing to choose between them.

Be aware that the turnover frequency is the point where each filter drops the level by 3dB. This means attenuation begins at a somewhat higher frequency.

These circuits are very easy to simulate on a PC. I would recommend you try LTspice for this.

Cheers

Ian

 

[Gustav]

Ian gave you the answer from someone smart enough to know, but maybe for inspiration (someone not smart enough to know), I would check with a cap value, measure frequency, work the math backwards to derive the impedance.

Then look at the circuit and see if I could make sense of the measured impedance.

Gustav

 

[Gene]

Thank you both for replying, thats a lot of great info and a nice suggestion from Gustav. I downloaded the LTspice software and now need to figure out how to use it.

A question though: how is the load of 800 ohms derived from 2k resistors? That bit keeps throwing me off.

 

[ruffrecords]

Thank you both for replying, thats a lot of great info and a nice suggestion from Gustav. I downloaded the LTspice software and now need to figure out how to use it.

A question though: how is the load of 800 ohms derived from 2k resistors? That bit keeps throwing me off.

2K in parallel with 2K is 1K. 1K in parallel with 5K (the input impedance) is just over 800ohms.

Cheers

Ian

 

[emrr]

Not sure why variable and passive.  Hardest thing you can do.  I'd do active, easy here:

http://sim.okawa-denshi.jp/en/OPseikiHikeisan.htm

 

[dustbro]

Not sure why variable and passive.  Hardest thing you can do.  I'd do active, easy here:

http://sim.okawa-denshi.jp/en/OPseikiHikeisan.htm

Is there a calculator like this for LPF?

 

[Gustav]

Not sure why variable and passive.  Hardest thing you can do.  I'd do active, easy here:

http://sim.okawa-denshi.jp/en/OPseikiHikeisan.htm

Is there a calculator like this for LPF?

http://sim.okawa-denshi.jp/en/OPseikiLowkeisan.htm

Gustav