This is an active crossover.
Suitable for biamping.
It is a 2nd order filter 12dB slope Linkwitz-Riley.
XO frequency is 1591 Hz, which could be for 1 inch tweeter + 8 inch woofer.
Example of power amplifiers are LM1875 chip amps.
Suitable for biamping.
It is a 2nd order filter 12dB slope Linkwitz-Riley.
XO frequency is 1591 Hz, which could be for 1 inch tweeter + 8 inch woofer.
Example of power amplifiers are LM1875 chip amps.
Other frequencies can be had by changing the resistors.
Resistors - Frequency
11k - 1446
10k7 - 1487
10k5 - 1515
10k2 - 1560
10k - 1591
9k76 - 1630
9k53 - 1670
9k31 - 1709
9k1 - 1748
9k - 1768
Resistors - Frequency
11k - 1446
10k7 - 1487
10k5 - 1515
10k2 - 1560
10k - 1591
9k76 - 1630
9k53 - 1670
9k31 - 1709
9k1 - 1748
9k - 1768
When the four resistors are 6k34
the crossover frequency is 2510 Hz. See image.
By changing any frequency can be achieved.
the crossover frequency is 2510 Hz. See image.
By changing any frequency can be achieved.
You should add an all pass filter to add a delay on the tweeter : All pass filter wiki
Delay ~ 176us to be evaluated.
You will have better transient 🙂
Delay ~ 176us to be evaluated.
You will have better transient 🙂
I added a delay allpass to the highpass.
This is the circuit I used: Would this work correctly?
Something like this would definitely come in handy, and not just for me! I'm currently building a 4 channel amplifier SA2015 for bi-amping, so I hope you will finish the project.
I subscribe 😍
I subscribe 😍
Nice project. I would think before adding an all pass filter to the tweeter circuit that you would want to measure the time alignment between your tweeter and woofer in order to get a better sense of how much delay you need to add on the tweeter channel. (A simple impulse measurement in REW would tell you what you need to know.)
As as aside some years ago I designed a 24dB/oct 3 way LR crossover and came to profoundly regret that I did not make provisions for time delays. This was made even worse because I was using a long mid-range horn and physical time alignment wasn't possible. I actually needed to delay the woofers in this case. (Ultimately went to DSP and learned that having the ability to adjust time alignment electrically was critical to achieving the performance I was looking for.)
As as aside some years ago I designed a 24dB/oct 3 way LR crossover and came to profoundly regret that I did not make provisions for time delays. This was made even worse because I was using a long mid-range horn and physical time alignment wasn't possible. I actually needed to delay the woofers in this case. (Ultimately went to DSP and learned that having the ability to adjust time alignment electrically was critical to achieving the performance I was looking for.)
I tried to get the phase assuming a given delay in vituixcad so we could compare them but got lost. Need chime-in from analog circuit experts. This is how vitiuixcad draws an opa all-pass block (at 2510hz with different default resistance):
I must admit I find delay circuits pretty murky. Here an example from Goran of a passive implementation: https://www.audioexcite.com/?page_id=1988
I guess the proof would be some kind of test against a known control. The difficulty vs dsp I suppose is how does the user calculate/fiddle/set the delay depending on the drivers and baffle layout/cabinet used, per design,
I guess the proof would be some kind of test against a known control. The difficulty vs dsp I suppose is how does the user calculate/fiddle/set the delay depending on the drivers and baffle layout/cabinet used, per design,
Also, this jogged my memory of a TI engineer who posted a pcb with delay here: https://www.diyaudio.com/community/threads/analog-active-crossover-pcb.249409/
Me too, and I have a lot of analog design experience. Have a look at the schematic and manual (attached) of the EV XEQ-3 for some workable all pass filter design as a guide. A single stage all pass is not a great solution as the delay is not constant over frequency range to be covered IIRC.
You don't need to calculate anything, measure it with an inexpensive calibrated USB mic like the MiniDSP Umik-1, REW and a cheap USB stereo dac. The end result is you know almost exactly how much delay you need to add.
Last edited: