A friend has a condition that can cause seizures when listening to frequencies over 1500Hz. To help her out I made a passive first order (RC) low pass filter with a cut-off frequency at ~1300Hz but when testing this I noticed that frequencies of up to 13kHz were still audible.

Obviously what I tried isn't the solution, but with no proper electrical engineering training, I'm struggling with what I should be looking up online regarding keywords and names of circuits.

TL;DR I'm looking for a circuit that will completely cut out frequencies over around 1450Hz with 0 wiggle room for anything over that frequency. Preferably as a DIY solution, but I'm open to all suggestions.

Any circuit names or links that can point me in the right direction is greatly appreciated.

EDIT: Also, would stacking these passive low pass filters eventually achieve what I'm after?

Update (thanks transistor): The general use case is with headphones and listening to music or audio output from a TV etc. She wears noise cancelling headphones when out in public with no music only to dampen outside noise and helps little. This is just a side project to hopefully let her be able to plug into any device with headphones and know that she wont be harmed.

Update 2: Thank you everyone for your help, the general consensus that I received is that it is harder than a simple passive filter. I found this site which lets you specify requirements on a filter and suggests (with some tweaking) a "10th order Chebyshev 0.20 dB (5 stages)". So I'll see how I go. Thanks again for the input, I really appreciate it.

  • \$\begingroup\$ @pipe. I agree. The OP has admitted to not having a working background in electronics, much less knowing that a 1K ohm resistor and a .01uF capacitor are off-the-shelf values the roll off at about 1.5KHZ. With a single IC a drop of 24dB octave is possible. But this seems personal. I will say no more. \$\endgroup\$
    – user105652
    May 4 '16 at 5:19
  • \$\begingroup\$ If you find that it doesn't help with just the filter, it might help adding some white or pink noise after filtering, to mask whatever notes that remain. If you want that, it'll have to be a new question though.. \$\endgroup\$
    – pipe
    May 4 '16 at 5:22
  • \$\begingroup\$ My Bad. R should be 12K ohm and C = .01uF. for a roll-off of about 1.4KHZ. Without trim pots an exact cutoff value is difficult to set. \$\endgroup\$
    – user105652
    May 4 '16 at 5:33
  • \$\begingroup\$ @Adam: "... when listening to frequencies over 1500Hz." In general? On loudspeakers? On earphones? Music only? Speech? How will your low-pass filter be used? Just wondering: do regular ear-protection plugs or ear-muffs help? Pop the info into your original question. \$\endgroup\$
    – Transistor
    May 4 '16 at 6:26
  • \$\begingroup\$ It sounds like you need a brick wall filter, lets everything in upto 1.4kHz and then brick wall straight to zero. Problem is, that's really hard to do physically, most solutions are done in software where there's not as much of a limit of filter complexity. I feel for your friend, there's not that much worth listening to below 1.4kHz compared with the higher frequencies. But a really good pair of earmuffs might do the trick. \$\endgroup\$
    – Sam
    May 4 '16 at 9:22

You will need a higher-order LPF (low-pass filter) in order to achieve a sharper cutoff. This article describes using two second-order LPF in series to achieve a 4th-order filter.

Ref: http://www.circuitstoday.com/higher-order-filters

enter image description here

  • 2
    \$\begingroup\$ This answer would be better if you transfer the important equations from the article. As it stands, it just gives the hints. You can't really trust that the article is still there at the same URI tomorrow. \$\endgroup\$
    – pipe
    May 4 '16 at 5:18
  • \$\begingroup\$ The formulae for calculating filters have been around longer than the interweb. The design methodology is very well described in dozens of places across the internet. I used this article as just one example of cascading higher-order filter stages to achieve sharper cutoff. \$\endgroup\$ May 4 '16 at 5:20
  • \$\begingroup\$ This is what I was thinking of, except using a quad op-amp like a TL074. Either way the OP has some learning to do. \$\endgroup\$
    – user105652
    May 4 '16 at 5:22
  • \$\begingroup\$ If R = 12K and C = .01uF, that should give a 1.2KHZ cutoff frequency, or did you split the middle to create a peaking effect? \$\endgroup\$
    – user105652
    May 4 '16 at 5:26
  • \$\begingroup\$ @Sparky256 Confusion like this is exactly why I wanted the equations in the answer. \$\endgroup\$
    – pipe
    May 4 '16 at 7:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.