Testing the Active Crossovers for my 10 Channel Amp

These crossover boards are a key component in my 10 channel amp build, and I have to make sure the board layout works properly before I order the professionally made PCBs.
Each board contains an input buffer, a two stage high pass filter, a two stage low pass filter and an output buffer that has some gain and also has a trimpot to adjust the output level.
The filter stages combine for a 4th order slope on the high pass and low pass, letting a limited range of frequencies through. In this example, the board is set to drive the woofer of my 4 way speakers from 80Hz to 350Hz. Another board will drive the midwoofer from 350Hz to 900Hz, and so on.
There will be 10 of these, one for each individual amplifier board.
Testing was done first with my scope just to check if it actually works. While the circuit design is proven, the wildcard is the board layout and whether I made any mistakes doing that.
I then set up my computer to run REW to measure the frequency response. This is done in a similar way to measuring a speaker (or room), except instead of using a microphone, the signal is fed directly through the crossover board.
As shown in the video, the measured response very closely matches the simulated.
The next step is to finalize the board layout and order the boards, and then I can start putting the amp together.
Active vs passive: is a topic for another video.
You can watch part 1 in this series here:

You can help support the work I do in making these videos:
Support this channel on Patreon:

#diyspeakers
#johnheisz
#audio

My "Scrap bin" channel:

My main channel:

[IBuildIt]

These crossover boards are a key component in my 10 channel amp build, and I have to make sure the board layout works properly before I order the professionally made PCBs.
Each board contains an input buffer, a two stage high pass filter, a two stage low pass filter and an output buffer that has some gain and also has a trimpot to adjust the output level.
The filter stages combine for a 4th order slope on the high pass and low pass, letting a limited range of frequencies through. In this example, the board is set to drive the woofer of my 4 way speakers from 80Hz to 350Hz. Another board will drive the midwoofer from 350Hz to 900Hz, and so on.
There will be 10 of these, one for each individual amplifier board.
Testing was done first with my scope just to check if it actually works. While the circuit design is proven, the wildcard is the board layout and whether I made any mistakes doing that.
I then set up my computer to run REW to measure the frequency response. This is done in a similar way to measuring a speaker (or room), except instead of using a microphone, the signal is fed directly through the crossover board.
As shown in the video, the measured response very closely matches the simulated.
The next step is to finalize the board layout and order the boards, and then I can start putting the amp together.
Active vs passive: is a topic for another video.
You can watch part 1 in this series here:

[thomasduby]

I laughed out loud! “As you know, it’s how it looks”. So many things in the world definitely fit that model.

[mfkhometheater]

That definitely brings back memories! I use to do a lot of that but I'm just too lazy now days so I just use DSP. At one point I was thinking of doing analog boards for my whole house audio just to use up a bunch of left over old OPA2134 opamps etc but I ended up just getting a bunch of 2x4HD's for the job.

[peterwu]

The high pass section has equal value capacitors and resistors with unity gain. The cutoff frequency is calculated to be 106Hz instead of 80Hz. And the design yields +3dB overshoot at the cutoff frequency - could be desirable. If replacing R18 with 20k, the cutoff would be at 75Hz, with -3dB “normal” roll-off.

[garyrobinson]

This was a very straightforward explanation, thank you.

[rickharms]

I have a question. Please understand, this is curiosity, not judgmental. Why cut off at 80Hz? I like going down to at least 20Hz. I am fully capable of admitting that I am full of crap. I greatly admire what you do.

[peterwu]

Why not just use an eight channel DSP, or two of four-channel Mini DSP for this build, as you have used previously for your 4-way open baffle project? Do you like analog crossover better than the digital?

[FixthisCD]

Is there any pros/ cons to putting the gain at the end versus the beginning of the circuit?

[jeffbeck]

Why did you not include pots so that you adjust the frequency cut-offs, like a commercial active crossover? I want to build what you are building, and that is an active crossover with 3 pairs of stereo amps for a tri-amp'd speaker system. But I want to be able to control the frequency of the 2 cut off points, between low and mid, and between mid and high, and I want a volume for each of those outputs.

[AquaPeet]

No sponsorship from PCBWay? ;) So awesome that you were able to do the whole process of prototyping.
Does the length of copper tracks on your board matter?

[KipdoesStuff]

I like learning new things, even if I will never apply that knowledge. Knowing is half the battle, YO JOE!

[hazembata]

Looking to build my own speakers soon. I am no expert. DSP is new to me. Can one design a speaker with no passive crossover and just use DSP. If so, can you explain that process? Thanks.

[analoghardwaretops]

..so these active xovers are slightly over 2 octaves wide....hence 5 per stereo channel??
How difficult would it be to have a one octave wide... design ....
(a 10+10 active)
in terms of acoustic performance only...??

[sudd]

where in the chain does the Active Crossovers fit in?
maybe between the dac and a amplification is my guess.
and my second question: is it easy to make a Active Crossovers that fits on before the dac?

[bbfoto]

@IBuildIt
Maybe it's been asked, but what will your Final Total Cost be for all prototyping + the finished versions of these 10 active analog crossover boards?

[valdimirputin]

At what stage Active Crossovers are connected? Between pre-amp and power-amp or between power-amp and subwoofer?

[TriAmpMyFi]

Why not variable x-over
You'll be stuck if new drivers are chosen, like a horn mid-range. 🔈🔉🔊

[lohikarhu]

I was wondering if you have any "funny things" in phases among the crossovers...24dB makes a lot of phase changes... You could probably, or did, simulate the crossovers, but, then, speaker phases, et al...

[bradstone]

Rpi, paxoverrack, 5 dacs, likely cheaper too...

[johncnorris]

"It's not how you sound, it's how you look. And you look..."