Help me with this active filter modification

Hi, my active crossover, a/d/s/ 642ix, can only be adjusted filter frequencies by changing filter modules. It has instruction for modifying these modules. They also had provided table of modified values. Please find attached pictures. I’d like to have the low-pass filter frequency to be at 320 Hz. Please help me to calculate component values to achieve 320 Hz. Thanks in advance
 

Attachments

  • B06B1A36-2E18-4787-952D-77B1224BDEC2.jpeg
    B06B1A36-2E18-4787-952D-77B1224BDEC2.jpeg
    52.3 KB · Views: 123
  • 0AD89F07-E03C-4D6E-B766-B899C80474DD.jpeg
    0AD89F07-E03C-4D6E-B766-B899C80474DD.jpeg
    128.6 KB · Views: 127
  • 2BD4AF79-22A2-439B-B6FF-12CFFC2E1949.jpeg
    2BD4AF79-22A2-439B-B6FF-12CFFC2E1949.jpeg
    159 KB · Views: 114
R1 Value for 320Hz is 6.81* 340/320 = 7.24k which isn't an E24 value. You could put 7k5 in parallel with 200k which will be close enough given the cap tolerances.

Sorry, but is it as easy as simply using the rule of three?

I tried applying the rule of three on the other values in the table, and found they weren’t linear relationships, constant slope when plotted graphs. However, I’m not sure about my math.
 
If you do some quick math on the table you've shown, you'll see the product of the frequency (first column) and the R1 value (right column) is a constant, around 2320 give or take a few dozen. I used that relationship to calculate the 320Hz value. You can double check because there are 150 and 170Hz values in the table which if you interpolate between them gives 160Hz. And that's half the frequency you want, hence has double the R1 value you're looking for.
 
  • Like
Reactions: presscot
If you do some quick math on the table you've shown, you'll see the product of the frequency (first column) and the R1 value (right column) is a constant, around 2320 give or take a few dozen. I used that relationship to calculate the 320Hz value. You can double check because there are 150 and 170Hz values in the table which if you interpolate between them gives 160Hz. And that's half the frequency you want, hence has double the R1 value you're looking for.

That’s clear. Big thanks to you.

Let me ask another final question. This is slightly more difficult. Please find attached again. This is the high-pass filter. I would like 540 Hz on the high-pass filter. But it has various C values. I’m not sure if I should use 0.1 or 0.022 uF. Please assist me one more time.
 

Attachments

  • EF2FC4DB-1B79-49E9-AFA9-D4F46E6C7A8B.jpeg
    EF2FC4DB-1B79-49E9-AFA9-D4F46E6C7A8B.jpeg
    75.2 KB · Views: 67
  • 17754D81-8691-44B3-89E5-A80400A17B39.jpeg
    17754D81-8691-44B3-89E5-A80400A17B39.jpeg
    138.8 KB · Views: 67
  • F4E84EF3-B960-4433-BE5B-5D0AE11E2982.jpeg
    F4E84EF3-B960-4433-BE5B-5D0AE11E2982.jpeg
    168 KB · Views: 68
Last edited:
Given that 540Hz is between the two tables and the cap value jumps down by a factor almost 5X, I'd go for a cap value of 0.047uF to create the 540Hz turnover frequency. Lower resistor values are less noisy but too low is going to load the driving stage so something in the region of 10k would be fine.

With the cap at 47nF, extrapolating from the left table gives 8.9k for R1,R3 at 540Hz and 4.46k for R2,R4. So 9k1//470k and 4k7//91k.
 
  • Like
Reactions: presscot