Creating a Passive Crossover for Subwoofer Integration: Seeking Guidance

Hello everyone!

I hope I've chosen the right topic for the issue I'm encountering regarding modifying a line-level signal. I'm reaching out for advice and guidance on how to address a specific challenge with my audio setup.

I recently got intrigued by the latest Class D amplifier on the market, the new Fosi V3. This amplifier has replaced my FX-Audio FX502Spro, which had taken the place of a Breeze Audio TPA3116 in my system. While the Fosi V3 boasts a subwoofer output, it turns out that it's simply an additional port parallel to the line input. Starting from October, Fosi V3 units will supposedly come with a factory-fitted subwoofer output from the preamp stage. However, it's unlikely that this output will include a low-pass crossover filter for the subwoofer.

To overcome this limitation, here's my plan: In my setup, I have a TV with an optical output connected to a Topping D30 DAC. The DAC's analog output is then connected to the Fosi V3, which drives my primary speakers, a pair of Pioneer SP-FS52 towers.

Additionally, I have a budget-friendly Sony SA-WP780 subwoofer. While I don't have high expectations for its contribution to music playback, I find it useful for enhancing movie viewing experiences.

This subwoofer is active and has only a line-level input. Its speaker impedance is 8 ohms, and its lone control is a volume potentiometer with an exaggeratedly large range that's rather low-fi and practically unusable beyond 20% of its travel. When I play frequency sweeps, I can distinctly hear the subwoofer reproducing the range from 50 to 1700 Hz.

I've ruled out connecting the subwoofer to the dedicated parallel output of the Fosi V3, as I want to avoid altering the impedance and potentially sacrificing high frequencies. Therefore, my plan is to use the RCA analog output of the TV, despite its DAC's known characteristics of being opaque and compressed. However, for my intended purpose, I believe it will suffice.

My goal is to construct a passive filter that completely cuts off frequencies above 80 Hz from the television's line-level output.

Can a passive filter accomplish this goal? I've already attempted to construct a basic RC filter, but the level of attenuation was minimal. Given the subpar quality of the subwoofer, my aim is to completely eliminate frequencies above 80 Hz. The primary intention is to generate a tangible sensation of room vibration for movies and some bass support for music.

I've come across dedicated 12V circuits on platforms like AliExpress, specifically designed for this purpose. These circuits feature stereo input and a subwoofer output, complete with two potentiometers – one for adjusting the frequency cutoff and another for volume control. However, if possible, I'd prefer to build a passive filter since my requirements are fairly straightforward.

I would greatly appreciate any insights, advice, or recommendations that you could offer regarding this project. Thank you in advance for your assistance!
 
First I tried building a RC filter with the values that ChatGPT calculated for me, but it missed, because it tried to adapt the RC Filter to the input impedance or something like that.

Then I tried with the values given in the following youtube video (1K + 1uF = -3dB @ 160 hz? ):

The effect wasn't steep enough, but cascading two of these RC filters improved at the point of making the subwoofer usable. The Subwoofer was audible until 1300 hz instead of 1700 hz.

I've used this online calculator, so next time I'll try cascading two or more RC filters with 2K + 1uF.
 
My goal is to construct a passive filter that completely cuts off frequencies above 80 Hz from the television's line-level output.
Okay, a 24 db/octave filter according to David B. Weems, is two 12 db/octave filters in series. A 12 db/octave high cut filter is a capacitor series the load, and an inductor parallel to the load. A 24 db/octave filter is two of those in series.
C=159000/(R*f) where C is microfarads, R is load resistance, and f is frequency in hz.
Lets assume your load is 10000 ohms, which is is lower than the actual load impedance, but you can install a 10000 ohm resistor parallel the amp input to keep the inductor from going to infinity. C=0.199 microfarads. easily purchasable. Digikey doesn't have any polyprophylene, only polyester, https://www.digikey.com/en/products/detail/kemet/RSBEC3200AA00J/13176307 for $.65 . Partsexpress should have some polyprophylene which distorts less.
L is R*1000/(6.28*f) where L is milihenry, R is ohms, F is in hz. So L is 19,904 millihenries . Or 19.9 henries. That is rather expensive. Digikey has some today https://www.digikey.com/en/products/detail/hammond-manufacturing/193C/455280
for $57 each. you need two. Series resistance is 181 ohm. Today digikey has only one in stock. Note the quoted inductance is probably at 60 hz. Inductance at other frequencies probably varies due to interwinding capacitance and non-linear iron core. Read the data sheet.
You should probably put 10000 ohms series each inductor to keep your DAC from shorting out if it tries to drive 20000 hz at 2 vac through your filter.
Have fun.
 
Last edited:
  • Like
Reactions: 1 user
Finally, I've solved my problem using the simplest solution with parts I had lying around the house: 2x 80 Hz RC lowpass filters (20K + 0.1 μF).

I didn't use resistors to combine the L and R channels because there's already a volume drop due to the two filters. I'm not certain if the circuit is perfect, but it's functioning as I needed. Now, the subwoofer is audible up to 600 Hz instead of 1700 Hz, and voices are barely audible. I didn't expect the subwoofer to pair so well for listening to music with the TV's built-in speakers. Not a huge complement for the Pioneer towers, but it's Ok for movies.


IMG_20230830_034109.jpg