Crossover recommendations for 600hz horn and JBL 2225H

Hi! Got a DIY 600hz horn with a 2" Compression Driver (Paudio DE99) and JBL2225H 15" Woofer bi amplified and active crossover by minidsp.
Listening distance from the speakers is about 3meters away.

Looking for some crossover recommendations that would improve the "centering" of the vocals to improve the sound, and a more solid midrange.

Tried out different points and crossover slopes but I keep coming back to HP at 1200hz 6db/octave butterworth and LP at 600hz 6db/octave butterworth.

The one octave underlap between the crossovers is deliberate since that seems to sound the best for me and sounds more "live".

I may be exhibiting a bias since Im a fan of 1st order crossovers so any recommendation would help and I can immediately try it out since I have a minidsp.

Attaching pictures of the measurements and the speakers.

Priorities are dynamics and a more solid center vocal image.

Thanks in advance!
 

Attachments

  • Spkr1.jpg
    Spkr1.jpg
    484.3 KB · Views: 660
  • Screen Shot 2022-03-23 at 12.41.03 PM.png
    Screen Shot 2022-03-23 at 12.41.03 PM.png
    319.1 KB · Views: 245
  • Screen Shot 2022-03-23 at 12.36.40 PM.png
    Screen Shot 2022-03-23 at 12.36.40 PM.png
    573 KB · Views: 208
  • Screen Shot 2022-03-23 at 12.36.26 PM.png
    Screen Shot 2022-03-23 at 12.36.26 PM.png
    553.7 KB · Views: 229
  • Screen Shot 2022-03-23 at 12.35.22 PM.png
    Screen Shot 2022-03-23 at 12.35.22 PM.png
    449.9 KB · Views: 242
Last edited:
If the Paudio compression driver is PA-D99, then 600Hz should not be a problem. However, if it belongs to the BM-D or other series then it's best to keep things at 800Hz minimum. What horn is that, and will crossing at 1200Hz match directivity to a 15" driver ? And, only 2m listening distance from such a big speaker ?
 
If the Paudio compression driver is PA-D99, then 600Hz should not be a problem. However, if it belongs to the BM-D or other series then it's best to keep things at 800Hz minimum. What horn is that, and will crossing at 1200Hz match directivity to a 15" driver ? And, only 2m listening distance from such a big speaker ?
Hi! Re-edited and adjusted the post, listening distance is 10feet away so around 3 meters. I know, still a bit too close for comfort, but the next option would be smashing in walls to enlarge the room and would make me persona non grata to my wife hahaha. It's a compromise I have to live with hehehe.

The CD is the PA-D99.

The horn is an es600 horn from Joseph Crowe, used a 1.4" to 2" horn adapter for the PAD99.

Tuned this mostly by ear and still a newbie on crossover design and driver integration hence the request for help.

The 1200hz xover was decided on by ear and we can change it up hence the request for suggestions.

The last attempt to try out a 600hz crossover point resulted in horn honk, which I can re-tune with EQ but it affected the naturalness of vocal tracks.
 
Last edited:
You don't have to keep it at 600Hz just because the horn is 600Hz capable, but the convention is to pick a crossover frequency where the directivity of the woofer gradually transitions to the horn, without being noticeable. Now, since this criterion indicates not a single frequency but a band of frequencies (around 800Hz), as you can see in the pictures where the -6dB beamwidths are similar.
 

Attachments

  • Beamwidth.png
    Beamwidth.png
    61.3 KB · Views: 137
  • es600.jpg
    es600.jpg
    33.6 KB · Views: 126
That JBL is a big boy, so beaming will start fairly low. Normaly I would suggest lower xo point, about half an octave above horn cutoff frequency or 850Hz. Unfortunately, there is nasty step in CD response at 800Hz, probably with phase issues as well. So, for HF, I would stay with 1200Hz xo point, but with 2nd order for better attenuation of that problematic area around 800Hz.
For LF, current 1st order at 600Hz dont give enough attenuation above 1200Hz, so despite
"underlap" you actually have JBL overlapping with horn, smearing the midrange.
Also, according to speaker pics, it looks like acoustic center of JBL is in front of acoustic center of the horn. For both reasons I would go with 1200Hz 3rd order for LF.

If you have problems with "centering" of soundstage, one possible reason might be that furniture you have in between speakers.
You should have clean space between speakers, especially at height of HF driver.
 
I'd want to measure to see where the horn flips.
I'd also want to see where the THD starts to rise as that gives a good indication of how well the CD is loaded by the horn.

When I was first using the Arai A290 about 20ya with my then new BMS4590, I was constantly told that they both went to 300Hz, so that's where I should cross. At modest levels 500Hz was much better, and for louder 600. It stayed at 600 until the system as dismantled when I moved.

As the system here is active as mine was, I'd check the THD, use probably LR4 slopes and delay appropriately once I found the nice midpoint between the 2225 and the horn.
I've owned a huge number of 2225/6 drivers and they're great.
 
I'd also want to see where the THD starts to rise as that gives a good indication of how well the CD is loaded by the horn.

When I was first using the Arai A290 about 20ya with my then new BMS4590, I was constantly told that they both went to 300Hz, so that's where I should cross. At modest levels 500Hz was much better, and for louder 600. It stayed at 600 until the system as dismantled when I moved.

As the system here is active as mine was, I'd check the THD, use probably LR4 slopes and delay appropriately once I found the nice midpoint between the 2225 and the horn.
I've owned a huge number of 2225/6 drivers and they're great.
Hello Brett, as you stated you owned 225/6 JBL's how large was the box you built and how low was the bass capable of playing? I do not have the space for open baffle as they are noted to play well in larger rooms. My attempt with Onken failed as the box predicted was 24cu.ft. any box volume suggestions to start at? kingfisher
 
I don't have answers to any of the good questions posed, but I would definitely include a steeper-than-6dB/octave HPF with a lower Fc than your primary crossover frequency for the horns. My old LE-85's were not known for their ruggedness, so maybe I'm biased. But I have a whole shoebox full of shredded diaphragms from trying to crossover too low, or with too shallow of a slope, or built with components that weren't up to the task (leakage, or other shortcomings) that keeps me alert to the issue -- even though it was decades ago now.

An octave and a half, or two, below the original Fc shouldn't cause too much audible trouble. And not too difficult to implement, since you're already in DSP nirvana.

Cheers
 
You don't have to keep it at 600Hz just because the horn is 600Hz capable, but the convention is to pick a crossover frequency where the directivity of the woofer gradually transitions to the horn, without being noticeable. Now, since this criterion indicates not a single frequency but a band of frequencies (around 800Hz), as you can see in the pictures where the -6dB beamwidths are similar.
Wow! Thank you for pointing me in the right direction!

Dont have the foggiest notion of directivity data though.

Sorry if I'm making a mistake but does the chart above suggest the since I have a 15" woofer a 6db/octave 800hz would be a better match directivity wise?

If I'm not making sense (or I'm failing to understand your post) I'd be happy to hunt down suggested links or topics for more knowledge pending some suggestions or guide words to point me in the right direction.

Where can I find that beamwidth table and the accompanying book or website?

Thanks again!
 
That JBL is a big boy, so beaming will start fairly low. Normaly I would suggest lower xo point, about half an octave above horn cutoff frequency or 850Hz. Unfortunately, there is nasty step in CD response at 800Hz, probably with phase issues as well. So, for HF, I would stay with 1200Hz xo point, but with 2nd order for better attenuation of that problematic area around 800Hz.
For LF, current 1st order at 600Hz dont give enough attenuation above 1200Hz, so despite
"underlap" you actually have JBL overlapping with horn, smearing the midrange.
Also, according to speaker pics, it looks like acoustic center of JBL is in front of acoustic center of the horn. For both reasons I would go with 1200Hz 3rd order for LF.

If you have problems with "centering" of soundstage, one possible reason might be that furniture you have in between speakers.
You should have clean space between speakers, especially at height of HF driver.
Now this I understand and can readily implement!

Will try this out when I have the time to hook up my laptop.

Any reasoning as for the different slope usage? Like, will mismatching slopes improve upon phase issues?

I can move the speakers a bit forward for the furniture issue.

Thanks a lot!
 
I'd also want to see where the THD starts to rise as that gives a good indication of how well the CD is loaded by the horn.

When I was first using the Arai A290 about 20ya with my then new BMS4590, I was constantly told that they both went to 300Hz, so that's where I should cross. At modest levels 500Hz was much better, and for louder 600. It stayed at 600 until the system as dismantled when I moved.

As the system here is active as mine was, I'd check the THD, use probably LR4 slopes and delay appropriately once I found the nice midpoint between the 2225 and the horn.
I've owned a huge number of 2225/6 drivers and they're great.
Reposted extra measurements.

Can you elaborate on lr4 slopes and "delaying" appropriately?

Do you mean measuring ms delay with different xover points while using an lr4?

Any recos on xover point to begin with
 
It's a sure bet that modern techniques would do better than my 47-year-old reflex designs. They're 5 cubic foot and served pretty well until 5-string bass guitars became more common. If you can get a 2225 sounding decent to 30 Hz, I'd love to hear about it.

It took tons of parametric EQ to get Nathan East's 5-string to sound [almost] effortless, and that was only at a much lower level than the cabinets were otherwise capable of !

Cheers
 
Last edited: