[mlundy57]

If a first order in-line filter is designed to be -3dB at 100Hz, it would be -9dB at 50Hz. How do you determine where it will be -6dB since the drop is not linear?

[jmimac351]

Mike, I've chewed on this in the past and got distracted on something else.  I think I'd figured it was "about halfway in between", but hadn't considered it's not linear - as you point out.  As I think about it more, I guess it's going to end up where it ends up, and the room is going to have a big say in that matter... such that the answer is almost moot, and means I need to measure / tune for wherever it ends up, in my room.  Notwithstanding all of that, the actual answer would be good to know as well.  The answer to this is going to be of great interest to me as well, as I have (yet) another project starting up... that you will probably have some interest in / be able to compare notes on.

This is pretty cool and may be of interest (just because)... these are input boards for my Ayre VX-5 Twenty amp.  They are spec'd for 100hz high pass, are "a thing" for Vandersteen owners as a lot of them use Ayre gear / Vandersteen likes the gear.  The input impedance for the VX-5 Twenty is 1M Ohms, per leg... so the caps can be super small.  The calculations for the value are right on with what's been discussed / shared for "high pass filter calculator" values, e.g., capacitor value needed for a certain frequency given the input impedance of the amp. So, doing it this way is another option for some.

[HAL]

For a single pole high pass filter you can use the equation:

C=1/(2*Pi*R*F-3dB), so for a R=47000 Ohm input impedance, and a F-3dB=100Hz, C= 0.0338uF. 

Linear Technologies has a free version of SPICE for circuit simulation.  I used it to plot the frequency response of that filter.  I put the cursor at the -6.02dB point as close as possible with the system.  The -6dB point is ~58Hz.
 

[jmimac351]

Hal, thanks for sharing that formula...

High Pass Calculator:
http://www.learningaboutelectronics.com/Articles/High-pass-filter-calculator.php

Capacitance Conversion:
https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-capacitance

[HAL]

The 0.0338uF would be 33.8nF or 33800pF.

If you start with the original uF value and multiply x1000, then it is nF, do it again and it is pF.

[jmimac351]

The 0.0338uF would be 33.8nF or 33800pF.

If you start with the original uF value and multiply x1000, then it is nF, do it again and it is pF.

Got it! I see the mistake I made... Should have entered "47" into the calculator instead of 47,000, given that I had the value set to "Kohms".  Thank you. 

[jmimac351]

Here's what I got for a spreadsheet for capacitor value / formula for 100hz high pass filter... or given whatever values are entered into the yellow cells.

[mlundy57]

For a single pole high pass filter you can use the equation:

C=1/(2*Pi*R*F-3dB), so for a R=47000 Ohm input impedance, and a F-3dB=100Hz, C= 0.0338uF. 

Linear Technologies has a free version of SPICE for circuit simulation.  I used it to plot the frequency response of that filter.  I put the cursor at the -6.02dB point as close as possible with the system.  The -6dB point is ~58Hz.
 

Thanks Rich,

I've downloaded the software. Now I have to learn how to use it to get what you are showing. Then I can manipulate the parameters. Based on your example, the -6 point is 84% of the way through the octave, not 50% like it would be if it was linear instead of logarithmic. It will be interesting to see if the -6 point is 84%  of the way through the octave with other -3 points between 40Hz and 200Hz.

[HAL]

Mike,
Glad to help.

Rich