JBL horn?

Status
Not open for further replies.
Jump to Latest
#41
Patrick Bateman said:
For instance an oblate spheroidal waveguide with a 1" throat, a 15" mouth and a depth of 7.5" will have less output on axis than a 15" horn with a 1" throat, a depth of 7.5", a 15" mouth and a diffraction aperture at the throat.
Click to expand...

Again John, I just don't think that this is true if both devices are CD and have the same coverage. It simply can't be, otherwise some power is being created somewhere. And don't try and tell me its "loading". The "loading" is simply not that much different. The diffraction device has large peaks and dips in its loading, but its average has to be about the same as the OSWG.
 
#42
buzzforb said:
I was aware that it did not have an exponenetial flare, but thought it may be generally included. Thank you for the tidbit on his horns. Very interesting. I guess that leads me to a reasonable question. How great a role does the listening enviroment play on the type of horn that should be used. For an average listening room of 15 x 18, does one horn provide advantages over another?
Click to expand...

Big time! That's why I've been building all these silly unity horns for my cars. In a situation where you're listening off-axis, it's really tough to improve on a constant directivity horn. (Check out Geddes' paper on how to set them up for off-axis listening.)

And JBL seems to be listening; I've measured some of my JBLs, and they're clearly designed for listening off axis. (You'll see a rise on axis, and the flattest response is OFF axis not ON.)

Or another way of looking at it is that the primary axis of a speaker like the Summa isn't on-axis; the primary listening axis of a speaker like the Summa IS off axis. These are speakers that should be listened to with a significant 'toe in.'


If you don't need to move around the room, and if you can sit with your head in a vise, there are some merits to other flares. On axis efficiency in particular. An oblate spheroidal waveguide makes an engineering tradeoff of lower on-axis efficiency for constant directivity and low diffraction.
 
#43
Patrick Bateman said:
Le Cleach has argued that his loudspeakers are optimized for European homes, which are typically smaller, where the listener basically sits equidistant between the speakers. In that type of arrangement the room sound is minimized, because you're so close to the loudspeaker.
Click to expand...

They are optimized for EU tastes. The fact that they are not CD is a problem for them so they argue that it is not important. The room reflections are never unimportant - granted, less important when you sit in the direct field in a dead room. If you like that kind of sound. I don't.
 
#44
Patrick Bateman said:
Or another way of looking at it is that the primary axis of a speaker like the Summa isn't on-axis; the primary listening axis of a speaker like the Summa IS off axis. These are speakers that should be listened to with a significant 'toe in.'
Click to expand...
That's why I use the term "listening axis", and not "on-axis".
An oblate spheroidal waveguide makes an engineering tradeoff of lower on-axis efficiency for constant directivity and low diffraction.
Click to expand...

I see the tradeoff being more size than efficiency. OS wants to be large because they flare so rapidly. Diffraction horns are the only way to get decent CD in a small size. The price you pay for this is smoothness. To me that's the tradeoff.
 
#45
gedlee said:
If you smooth them enough then everything will look the same.

Some smoothing makes sense. I use critical band smoothing, as does Dr. Farina (Italy, inventor of log-sweep testing). He did quite a bit of research into this and found that EQing using critical bands was found to sound the most natural in a test of listeners. It also makes the most sense. Critical bands widths change with frequency, so some complex programing is required, but for my money it is the right thing to do. They are about 1/3 octave @ 100 Hz and about 1/20th octave at 10 kHz., but the change is not linear. Its a know function however.

As to all the hypothesizing about how the "horns" work, I am sorry but none of what is being said rings true to me, and I've spent a little time looking at the problem. Its just not as simple as you guys make it out to be. First, what John (Patrick, John is his real name) says about axial gain versus directivity simply cannot be true. That's because a compression driver delivers a basically flat power and if one widens the beam then the axial response has to fall regardless of how short or narrow or whatever the throat section is. The driver can deliver the power over a narrow beam - higher SPL, or a wider beam - lower SPL. How one gets to that beam width has no effect on the end result SPL - it can't. This assumes that one actually does have constant directivity of course. That's where most people cheat - the directivity is not constant.

It is true that an incoherent wave front at the mouth will mitigate any nulls seen on axis from mouth diffraction (but so does a large radius). But is this the "right" thing to do? Is an incoherent wave what is desired? To a certain extent I think it is (I did some work on this years ago), but too much can be a problem. I found that small incoherencies have audible advantages, but large ones don't sound right. The foam that I use does this to a certain extent - it randomizes the wave propagation, but not excessively. I know that the axial nulls would disappear if the mouth were random instead of round. Round is a worse case. With numerical control and injection molding one can easily do this. Its tough for a small shop however.
Click to expand...

Clipboard01.jpg


I don't disagree with this. The first stage of a CD horn like the ones from the 80s increases the *on axis* SPL. The first horn in the pic above is an example of this.

You state that "How one gets to that beam width has no effect on the end result SPL." But narrowing the walls will *definitely* increase SPL on axis.

It's just the same amount of sound power narrowed into a cone that's smaller.

If anyone wants to explore this, take a look at the spec sheets of some of the Danley horns. You'll notice that the narrow angle horns have much higher SPLs in the measurements than the wide angle horns, even when configured with the exact same compression driver.

The higher SPL is mostly due to narrower coverage. Same amount of sound output over a narrower angle.
 
#46
Patrick Bateman said:
Clipboard01.jpg


I don't disagree with this. The first stage of a CD horn like the ones from the 80s increases the *on axis* SPL. The first horn in the pic above is an example of this.
Click to expand...

The length allows for a smaller mouth, nothing more- The same pattern without the slot expansion initially (the vertical expansion in these horns is typically a straight-line expansion through the slot section, without diffraction edge in the vertical). The horn would not be any more efficient than an OSWG of the same mouth size, in the middle of the response. It WOULD however have higher efficiency in the lower end of the range, where the increased depth allows the given mouthsize to perform lower in freq.
 
#47
Wayne Parham said:
Kinda thought you had seen 'em ....
To tell the truth, I've seen a lot of waveguides with charts that look kind of like the SEOS, having 5dB ripple pretty far up into the passband. It's the hardest part about constant directivity horns and waveguides - Many of them have lumpy response.
Click to expand...
Hmm, guess I missed it or forgot. I'd prefer to look at it like this, I think.

Now, somebody go get JBL measurements. Nobody has even mentioned it's a full-fledged DSP crossover, have they? Perhaps the averaging is smoothing the spec. sheet lines a lot, sure, but they could have made pretty much everything that isn't inherent to the polar response as smooth as wanted, after all.
 

Attachments

  • temp.gif
    temp.gif
    89.4 KB · Views: 442
Last edited:
#52
I was alluding to the fact that there was evidence of active correction to make the response that flat. Apparently the correction that was done was enough to greatly impress te few that had heard them. If it sounds great, does it matter how you got to that point?
 
#53
Well, now we're back to a common theme, where a problem must be consistent to correct with EQ- horn artifacts are almost always 3d and not consistent across multiple points, and thus, a correction for one angle/distance would be a further deviation from another.


Of course, a high q notch for particularly problematic resonances/colorations might not be too problematic, since we're less sensitive to dips, but still, I'd expect some artifacts from the weird oblique profile arrangement.
 
#54
Patrick has openly admitted to loving the Summas and using them as a refernce point, at least in his home. What are the systems of the other members who apparently look upon the M2 with disregard for the possibility of good sound?
 
#55
buzzforb said:
Patrick has openly admitted to loving the Summas and using them as a refernce point, at least in his home. What are the systems of the other members who apparently look upon the M2 with disregard for the possibility of good sound?
Click to expand...

Yikes, I have nothing against the M2s at all.
I've long argued that the main reason I bought the Summas, instead of going DIY, is that I simply don't have the skillset to make the Summas at this price point.

The M2 and the Summa is similar, but the M2 is basically double the price once you factor in the amps. Is it 2X as good? Is it as good?

If the M2s and amps were less than the Summas, then I think they'd be competitive.
Having said that, I think the sticker for the Summa should be somewhere around $15,000.
 
#56
Of course they are ridiculously expensive, but as I grow older, i don't mind anyone getting paid good money as long as they have something to offer. Rather pay JBL $20k than my local politician a dime. I wish the Summas could fetch that price and I could pay it. Alas, not going to happen, at least on my end. I applaud your admittance to liking the Summas. It is actually pretty rare for folks to say what they listen to. Come on the rest of you. Pony up. Thanks Patrick.
I tihnk Ill add Earl to the Summa list and Wayne to the Pi list.:devilr:
 
#58
buzzforb said:
It is actually pretty rare for folks to say what they listen to. Come on the rest of you. Pony up. Thanks Patrick.
I think I'll add Earl to the Summa list and Wayne to the Pi list.:devilr:
Click to expand...

I've had a tendency to ramble on with sentimental detail about the speakers that influenced me over the years, and the designs I have made hoping to take the best features from each and improve upon them. I'd bore you if I listed them all, but here are a few posts like that:

 
#60
Last edited:
Status
Not open for further replies.
Top Bottom