• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Taking Advantage of a Horn's Beamy Top End

Tim Link

Addicted to Fun and Learning
Joined
Apr 10, 2020
Messages
978
Likes
813
Location
Eugene, OR
I've been messing around with my Selenium HL14-25 exponential horns lately, mating them to some JBL2426-H compression drivers and finding out what I can get them to do.

My initial thought was to cross them over high like I used to, at 2000Hz. I've since learned that my midrange horn has some pretty bad behavior up that high. So, dispersion be darned, I decided to push the HL14-25 down to 1000Hz. It actually sounds pretty good when EQ'd flat on axis. I didn't even check the off-axis response. But tonight I decided to do that because I was listening to them somewhat off-axis.

To get a baseline I turned off all the digital EQ, but still have the 1 uF caps that I was using for the constant directivity horns I had on there earlier. What this does, besides protect the tweeters, is give a rising treble response on axis. But off-axis, it's pretty close to flat. As a matter of fact, the upward tilted slope on axis had the same little bumps the off axis flat curve had, so a few EQ points and both responses were straight lines, just tilted up on-axis, and flat at an appropriate angle off axis. It starts rolling off too quickly above about 12 kHz off axis, but the energy is pretty good on-axis. So, I turned the horns further inboard so that I would be off axis appropriately for a mostly flat direct response and had a listen.

This sounds really, really nice! There's lots of air! It doesn't sound hard and bright. It sounds sweet and airy.Voices sound natural, not thin, not overly sibilant. Another interesting thing is that on-axis, REW shows continually rising 2nd order distortion as the frequency goes up. Off axis, this isn't the case. I don't understand why that's happening. I'm trying to grasp what all might be causing this enjoyable sound. Some thoughts:

1. I'm getting excellent loading on the driver down to 1000 Hz.
2. The early reflections are similarly off axis to my listening position, so they should be tonally similar to the direct sound I'm getting - at least down to 1000Hz.
3. The higher frequencies are perhaps getting overdamped by my room. By allowing the upward tilting response on-axis, and then pointing it away from me, a better room balance is achieved from the later reflections, with more high frequency energy, which is needed.

I remember once looking at specs on some high end dome tweeters that had a significant rising response at their top ends on axis, but leveled out off axis. Somebody on a forum wrote that it's correct for the tweeter to do that, and you should listen off axis. So I'm not the first to hear something pleasing with this kind of response. I suppose it may depend on the room.
 
Another interesting thing is that on-axis, REW shows continually rising 2nd order distortion as the frequency goes up. Off axis, this isn't the case. I don't understand why that's happening.
I think I understand this. The distortion overtones are all getting beamed in one direction. When I'm measuring off axis the distortion tones are reduced relative to the fundamental.
 
I have a JBL S3100 that has the same compression driver.

I filter them through a DBX 360 VENU, but taking REW measurements with its asymmetrical horn is a real pain.
 
I have a JBL S3100 that has the same compression driver.

I filter them through a DBX 360 VENU, but taking REW measurements with its asymmetrical horn is a real pain.
That is one strange looking horn! It's hard to even look at the pictures without thinking I'm seeing some weird perspective effect.
 
I have a JBL S3100 that has the same compression driver.

I filter them through a DBX 360 VENU, but taking REW measurements with its asymmetrical horn is a real pain.
I have a pair as well and yeah, measuring the high end output is tricky. I also observe a rising response around 10khz most of the time with that horn.

I might take my pair outside one of these days (they are on castors) and try making some quasi-anechoic measurements of them.
 
Back
Top Bottom