Interesting new horn by JBL. Looks like it using similar concept to LC-1A
JBL Professional :: Recording & Broadcast :: M2 Master Reference Monitor
JBL Professional :: Recording & Broadcast :: M2 Master Reference Monitor
So the compression driver is a coaxial, similar to the BMS units?
There appears to be some type of diffraction employed at the throat.
Someone with more knowledge on the subject care to elaborate on the waveguide?
They do seem nice though. Expensive, without a doubt..
Yeah, $12,000/pair.
Edit: Ah, I just got the RCA LC-1A concept relation. Yep. I notice a similar horn with a dome in the little 3 Series.
There appears to be some type of diffraction employed at the throat.
Someone with more knowledge on the subject care to elaborate on the waveguide?
They do seem nice though. Expensive, without a doubt..
Yeah, $12,000/pair.
Edit: Ah, I just got the RCA LC-1A concept relation. Yep. I notice a similar horn with a dome in the little 3 Series.
Last edited:
Not very similar to the BMS units despite the superficial resemblance.
There is a lot of discussion of these over on the AudioHeritage forum.
Expensive driver, around $1400 but apparently available.
Best wishes
David
In horn designs, it has always appeared to me that the more outrageous one makes the device look while maintaining some reasonable performance it can then be claimed to be "new" or "revolutionary" and even patented. But there is simply not much new in this realm and none of these "new" devices works any better than the older ones.
This maybe true, but if a horn sounds and performs well. I can't help it when the looks are great. Nothing beats(in my opinion) a horn that sounds and looks great, you can say I have a thing for big ugly horns(as long as they sound good).
They're reputed to sound superb, but they do appear to be designed around diffraction in the throat, which I'm not a fan of historically. The Array horns have a slot and are also well regarded, but I see no good reason to use diffraction slots, given that polars can be maintained without them.
No diffraction based horn will ever measure as smooth as one without. And no sharp edged horn will either. Maybe they "sound better" to some, but I prefer to enjoy the one that "measures best" rather than one that "looks cool".
I would expect the JBL horn to act very much like the SEOS, neither of which would measure nearly as well as one of my waveguides.
Its all about what you get used to and I prefer to set my standard to the one that is the smoothest. Some people will always like coloration, but after listening without it, I prefer to not have it.
I would expect the JBL horn to act very much like the SEOS, neither of which would measure nearly as well as one of my waveguides.
Its all about what you get used to and I prefer to set my standard to the one that is the smoothest. Some people will always like coloration, but after listening without it, I prefer to not have it.
If the waveguide went down to 400hz, it would need to use a 34" woofer!
The waveguide is crossed over at 800hz, which is just a bit lower than you'd expect for a 15" woofer and a 15" waveguide. (The Summas that I'm listening to right now use an xover of 900hz iirc, along with a 15" waveguide and a 15" woofer.)
Here's my 'take' on how the horn in the JBL M2 works:
In a conventional waveguide the beamwidth is constant from the top of the devices output, to the bottom:
The waveguide in the Econowave is a good example of this. It's a clone of the JBL PT waveguide. See how the beamwidth is quite constant from top to bottom? (the beamwidth above 13,500 hertz will always narrow, because 13,500hz is one inch long, and that means that above 13,500hz the beamwidth is basically dictated by the compression driver itself. IE, the beamwidth of a waveguide and horn combo will depend on the exit angle of the compression driver that's used.)
But sometimes you want a narrower beamwidth for part of the devices bandwidth. The 18Sound XT1086 is an example of this:
^^^ In the measurements, see how the beamwidth varies? It broadens at 13,500hz, as the wavefront exits the compression driver. And then it narrows again at 5000hz (2.7inches). IMHO, the narrowing is due to the throat being 'pinched' in the X axis. (Google some pictures of the 18Sound XT1086; you'll see what I mean. It doesn't have the smooth 'progressive transition' of the econowave waveguide. I believe the primary reason that 18Sound would 'pinch' the horn is that it raises the output level on axis. Everything else being equal, a narrower horn is louder on axis than a wider horn.
OK, still with me?
I know this gets confusing, particularly if you don't have the waveguides sitting here! (I must have two dozen waveguides in my garage
)An externally hosted image should be here but it was not working when we last tested it.
^^^ Here's the money shot. The JBL M2 throat, and the 18Sound XT1086. In the pic, we see that both companies are using a diffraction slot. The difference is that the JBL diffraction slot is mirrored vertically *and* horizontally. The 18Sound isn't - the shape varies depending on the axis. (Wide vertically, narrow horizontally.)
Another 'trick' with the JBL diffraction slot is that it's narrow on the measurement axis (up, down, left, right) but wide on the orthogonal axis. I'm not sure if this was done for looks, or for something more sinister. (Nobody measures loudspeakers on the orthogonal axis, so perhaps shaping it this way improves the polar response on one axis, at the expense of another.)
In engineering you rarely improve one aspect of a product without hurting another. Put a bigger motor in a car and you make it faster, but you also make it handle worse. Use a better interior in a car and it may sell better, but it will also add to the cost of the vehicle.
So I *do* wonder if the oddly shaped diffraction slot hurts measurements on the orthogonal axis. Another possibility is that it simply allows JBL to raise the output on axis, but the 'slits' in the diffraction slot 'bleed off' some of the energy towards the orthogonal axis.
It's definitely, uh, "interesting."
I still agree with Geddes that a big part of this was likely marketing. The cheap PT waveguides measure pretty darn good.
Last edited:
The way JBLpro speaker measurements are shown, is special
At first look it is excellent. When we read it more carefully we notice that (perhaps)
Listenig window response is averaged response 0 to +/-10 or 15 degrees horizontally
First reflection response is 30¤ or averaged 30-45¤ horizontal response
Power response is sum/average of 0 to 60¤ horizontals
This way of presenting anechoid measurement data is not standard but very good! Ít tells a lot of a speaker's performance
I didn't find the answer, but I believe it is somewhere in the excellent webpages of JBL
At first look it is excellent. When we read it more carefully we notice that (perhaps)
Listenig window response is averaged response 0 to +/-10 or 15 degrees horizontally
First reflection response is 30¤ or averaged 30-45¤ horizontal response
Power response is sum/average of 0 to 60¤ horizontals
This way of presenting anechoid measurement data is not standard but very good! Ít tells a lot of a speaker's performance
I didn't find the answer, but I believe it is somewhere in the excellent webpages of JBL
Attachments
We were discussing a lot of these same topics in the "Uniform Directivity" thread a few days ago:
Note that sharp edges aren't the only problem that can cause resonance or diffraction. Even a smoothly shaped flare can cause these problems if its profile is wrong. The wavefront propogation transitions from planar to a spherical section when progressing through some waveguide/horns. That's the ideal. But some horns will cause wavefront propogaton to remain planar (or close) until it reaches the mouth, where it suddenly breaks into a spherical wavefront. And some other horn shapes can cause the wavefront to be bent and distorted in other weird ways.
So while the hard edges are obvious sources of discontinuity, even flare profiles without sharp edges can have problems. You can't necessarily tell just by casually looking at the device.
Note that sharp edges aren't the only problem that can cause resonance or diffraction. Even a smoothly shaped flare can cause these problems if its profile is wrong. The wavefront propogation transitions from planar to a spherical section when progressing through some waveguide/horns. That's the ideal. But some horns will cause wavefront propogaton to remain planar (or close) until it reaches the mouth, where it suddenly breaks into a spherical wavefront. And some other horn shapes can cause the wavefront to be bent and distorted in other weird ways.
So while the hard edges are obvious sources of discontinuity, even flare profiles without sharp edges can have problems. You can't necessarily tell just by casually looking at the device.
It seems to me that the SEOS guys are getting pretty great measurements- I've seen you refer to a throat discontinuity with these, my understanding was that they are a OSWG style throat transitioned to match the DE250- what specifically would you point out as their shortcomings, as would be reflected in measures?
I've seen some measures of people crossing to a horn while the direct radiator is still omni- not a problem, the pattern just collapses following the horn, more slowly with shallow XO, faster with steep filters. I think the only reason we need to match directivity is because we're trying to do 2 ways, and are limited on how low we can push the horn. With a CD horn that plays that low, one would surely need a supertweeter, but we'd only need the 34" if we wanted to extend the pattern-controlled range (finding a 34" cone that behaves well at a few hundies is another issue) a little bit lower than the horn cutoff.
Yeah, frankly, I think it looks stupid. Kind of like the vertical array horns- why on earth wouldn't one use the extra space above the woofer cabinet to properly terminate the horn?
Perhaps because JBL is at the mercy of Harman Int'l..... a sad slow end for one of the greatest brands in HiFi.
I would love to see some examples of smooth-looking horns that misbehave- the measures I've seen for the smoother profiles all seem to fall along a continuity curve- lecleach at one end, with collapsing directivity, and OSWG/conical at the other with constant directivity. With smooth contours, they all seem to measure pretty well, just not necessarily what someone wants (I like the idea of consistent reflected/direct tonal balance so I'm more of a CD kind of guy).
Grab a copy of Hornresp and model horns of all three shapes - LeCleach, Catenary (OS) and Conical. Try various wall angles, mouth areas and lengths. You'll see some that are nice and you'll see some that are resonant. There's a lot of literature about the reasons, and I think probably you're familiar with much of it. But modeling is pretty easy and it will show you, and as they say "a picture is worth a thousand words."
Actually, this was one of the devices I was alluding to in my last post. Its physical features look smooth, but its response isn't. It's fairly resonant, with 5dB peaks and dips pretty much all the way through the passband. Any time you see a smooth curve of a loudspeaker using a SEOS horn, it either has a lot of EQ or the measurements are smoothed, or both.
The SEOS profile is actually only catenary for the first couple inches, then it deviates from that flare profile. So it is not an oblate spherioid, nor is it a prolate spheroid. It is something else.
I had considered using a SEOS horn at one time, but as I learned more about it realized it wasn't shaped as I would want. Pity, 'cause it's close, and I think if the guys that designed it had more experience with horns, they would have hit that thing out of the park. But I do think they kind of "winged it" on the profile shape, and ended up with a device that's pretty resonant.
I'd much prefer an older radial horn, frankly. Some of those radials are pretty darn smooth, actually, much smoother than a lot of horns sold today as "waveguides". The truth is many of them - including the SEOS horn - don't measure all that great. Of course you can EQ them to get a flat response curve, but you can do that to just about anything.
The difference, of course, is that the smoothness CAN be EQd when the polar behavior is consistent, where if there is diffraction, the artifacts are not consistent in 3d (I know this dead horse has been beaten). To me, the smoothness is more a function of "do the off-axis responses track to the on-axis", which tends to be the case with smoother profiles. We're just arguing semantics here, I respect not wanting to overuse EQ, but I'd say that any of those can provide smooth response, if shaped right.
I'll play around with Hornresp.
I'll play around with Hornresp.
Uh, yup, goofy horn and all. Swapping the horn to something like a DDS ENG90 with a squared flange might not be a bad way to improve the system
The drivers are the real ZOOT of that setup methinks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.