I've been slowly working on this project for half a year now, and am finally getting close to finalizing the crossover. Things are getting easier now that I have a decent measurement setup.
I used Arta with the following hardware to take measurements:
- Behringer ECM8000 microphone
- Behringer UB602A for phantom power and preamp
- Niles Audio SI-275 amplifier
- M-Audio Delta Audiophile 2496 sound card (in 96khz/16bit mode)
The baffles are 17 inches tall and 5 inches wide. I don't think this is that messy, but the wife is not happy.
A close up. The left and right edges were rounded over with a 1/2" bit. I can remove the baffle by removing those 4 ugly screws.
The back looks worse in this picture than it really is. A flash sure brings out the flaws in my amateur paint job. To save money I used brass screws from Home Depot instead of binding posts.
I put the crossover in a wooden box I bought at Michaels for $3. Once I dial everything in I'll move the crossover into the enclosures.
Here is the nearfield FR and CSD of my Dayton RS125-S. The crossover is a simple coil and cap. No impedance compensation or notch filter. There is an energy problem around 900hz.
The result of my first attempt at a crossover. These are on axis gated measurements (1m, 3.5ms). I matched the levels of the tweeter and woofer pretty well, but the phase could be better. The bottom line is the ambient background noise.
I had initially measured without any compensation for the microphone, and incorrectly thought the tweeter was too bright. I modified the lpad and ended up with this (re-measured with my new setup). The tweeter is padded too low, but the reverse null looks great.
The gated CSD of the speaker as it currently stands. It's pretty easy to see where the Seas 27TBFC/G takes over, because it really does stop on a dime.
My next goal is to adjust the lpad to get the correct level and deep null, since I've read that a deep null is a good indicator of a well designed crossover.
Dan
I used Arta with the following hardware to take measurements:
- Behringer ECM8000 microphone
- Behringer UB602A for phantom power and preamp
- Niles Audio SI-275 amplifier
- M-Audio Delta Audiophile 2496 sound card (in 96khz/16bit mode)
The baffles are 17 inches tall and 5 inches wide. I don't think this is that messy, but the wife is not happy.
An externally hosted image should be here but it was not working when we last tested it.
A close up. The left and right edges were rounded over with a 1/2" bit. I can remove the baffle by removing those 4 ugly screws.
An externally hosted image should be here but it was not working when we last tested it.
The back looks worse in this picture than it really is. A flash sure brings out the flaws in my amateur paint job. To save money I used brass screws from Home Depot instead of binding posts.
An externally hosted image should be here but it was not working when we last tested it.
I put the crossover in a wooden box I bought at Michaels for $3. Once I dial everything in I'll move the crossover into the enclosures.
An externally hosted image should be here but it was not working when we last tested it.
Here is the nearfield FR and CSD of my Dayton RS125-S. The crossover is a simple coil and cap. No impedance compensation or notch filter. There is an energy problem around 900hz.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
The result of my first attempt at a crossover. These are on axis gated measurements (1m, 3.5ms). I matched the levels of the tweeter and woofer pretty well, but the phase could be better. The bottom line is the ambient background noise.
An externally hosted image should be here but it was not working when we last tested it.
I had initially measured without any compensation for the microphone, and incorrectly thought the tweeter was too bright. I modified the lpad and ended up with this (re-measured with my new setup). The tweeter is padded too low, but the reverse null looks great.
An externally hosted image should be here but it was not working when we last tested it.
The gated CSD of the speaker as it currently stands. It's pretty easy to see where the Seas 27TBFC/G takes over, because it really does stop on a dime.
An externally hosted image should be here but it was not working when we last tested it.
My next goal is to adjust the lpad to get the correct level and deep null, since I've read that a deep null is a good indicator of a well designed crossover.
Dan
sreten said:Hi, why do you have a wide driver spacing ? ,/sreten.
Good question. I initially made these baffles for a single HiVi B3S, then chose to try a two way. In order to reuse the baffles, either the tweeter or the woofer would have to be placed in the original cutout. I modeled both, and chose to put the woofer where the original cutout was, because the tweeter was far more senisitve to baffle placement.
Below is the diffraction simulation. The first image is on axis, and the second image is 15 degrees off axis.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Dan
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.