doug20 said:
And there was no need to denigrate one of my customers for buying my product either.
doug20 said:
I have posted what I have learned on this topic many times, but I guess that I'll do it again.
When I see a horn with a regular period of peaks and dips in either the frequency response or the impedance curve, but usually both, then you can be sure that it will sound collored. This has to be looked at both on-axis and off axis as very often the on-axis curve will look the best, but looking at all the axes will tell. Peaks and dips at the same frequency in ALL the curves is a serious problem.
I also look at the impulse response as it needs to be as compact as possible. Tails or ringing is a real no-no and we have done extensive research into this. This effect will not always show up in a magintude response curve as it is not minimum phase - the magnitude can be fine, but there can still be a tail.
Last, but not least, the device has to meet its goal of CD or its not really comparable as this is a design requirement that has to be met.
Distortion is, IMO, irrelavent, and there is a lot of data to support this as well.
winslow said:
True. However, the horn is CD and can be run as a two way by closing off the mid holes and crossing the compression driver to the woofer. I've run mine that way, and it has the same basic character as the full Unity. So for this discussion, it still applies. I preferred the full Unity by a little. However, starting from scratch, I'd probably just go two way. For home use, you don't really need the extra headroom that the Unity can provide. It does mean that you need a bass driver with good performance with no nasties well above the crossover frequency.
I don't know whether it's harder to do a good crossover with the mids, or just the two way. I suspect the latter is trickier to achieve, as directivity is changing rapidly there.
Sheldon
augerpro said:
Hi,
I know "horn honk" from a look-JBL-alike expohorn plus lense, I build in the early eighties. Bad. It's signature is an abrupt change in directivity at about 1kHz? I don't know exactly.
A solution is available. In large rooms 18s XT1086 is a great sounding device. It is as smooth as can be with a 18s 1050 driver. I have to admit that I really like the basic concept behind the "summa". But others didn't stand still either.
so long
Sheldon said:
This is a very curious statement because it implies that the "unity" is what it is because of the horn itself and not because of the "optimal" match of the mids to the highs at the throat. This is in stark contrast to the inventors claims.
gedlee said:
Just reporting my experience. Though I do prefer it with the mids, and maybe that has to do with that crossover in the midrange occurring in the horn. I haven't done enough testing to make any causal claims, or even to assert that I would always prefer the full Unity. But it's not a dramatic difference, under my listening situation, to me.
Remember though, that the design was originally intended for PA type application, and does lend itself to higher crossover frequency to the compression driver. So I can see a plausible advantage in high SPL applications. Also, I'd guess, based on limited measurements that I've done, that it's probably easier to get the polar issues worked out at a crossover around 300-400, than up around 1k. I'd defer to your experience here.
My conclusion is that, if one can get the crossover right, the two way system, ala Summa, is more appropriate for home use. As you say, however, getting the crossover right is no trivial matter.
Sheldon
Re: Re: My personal opinions of various design philosophies
OK, well, I am pretty sure you conceded that limited vertical beamwidth would be attractive, but the problem was the elliptical horn is hard to build.
I'm sorry if I misrepresented you, so perhaps you could clarify. This is what I said I thought you agreed with:
For clarification, what part of that statement do you disagree with?
As for the deal about what data is available, what guesses might be made and what is "better", it seems to me that is a bit of a red herring. I mean, what is "better" - a device that offers CD in the vertical but when installed in a loudspeaker creates nulls that cut the pattern in half or even two-thirds? Or a device that loses vertical pattern control at the low end but allows a loudspeaker to be built with a much larger clean forward lobe, having nulls spaced further apart? Who decides what is "better" when both are compromised? I expect you might say the data decides but it isn't that clear when you have competing priorities.
To me, it would be a no-brainer if you had CD through the passband, 90°x40° to 90°x50° beamwidth, and vertical mouth dimension small enough that the null angles could be set outside the pattern. That's the holy grail, in my opinion. However, I'm "guessing" it would be difficult to realize all these competing priorities in one device. Not a real tough "guess" since that's what has been found over and over again for the last few decades. It's part of CD horn history. Maybe that means you and I are both "guessing" the same thing.
Still, it may be possible, and is certainly a worthy goal. Even without vertical pattern control down low, I think the behavior of the device can be optimized for use in a loudspeaker where drivers are stacked vertically on a baffle, and that optimization requires tight vertical spacing. I believe the right design is a balanced approach, not just putting an axisymmetrcial horn on the baffle and saying that's the best I can do.
I am not confident that an elliptical waveguide can be made CD through its bandwidth, but I am sure that a loudspeaker with a 90° round waveguide cannot be made CD in the vertical. In fact, most loudspeakers like that have 60° of their 90° vertical pattern, dirty, outside the nulls and out of phase with the forward lobe. Even well above the crossover region where there's no interference, you still have the issue that high frequencies at large vertical angles serve only to create reflections from the ceiling and floor. For these reasons, to me, the axisymmetrical waveguide/horn is not an attractive compromise and an asymmetrical waveguide/horn can be used to better result.
I'm not saying any asymmetrical horn is better, but one can certainly be made to be better. For example, I do think your round OS waveguide is better than an old sectional horn. No question. But I also think if you make your elliptical waveguide with similar profile and size the proprtions right, it will be a better compromise between pattern control and null angle. I'm hoping you can find a way to make them and to cut your foam to fit, because I know those things are very important to you. I think the build will be the hardest part, probably the only thing that might hold you back. Just a "guess".
Certainly a horn or waveguide with smaller vertical mouth will have a larger clean forward lobe between the nulls. To me, that's a better approach. The fact that you are looking at an elliptical horn implies that you think it may be worthwhile too. I see a lot of products coming out with this general shape, so I think a lot of others are on the right track.
Or if you would prefer to stick with round horns, that's cool with me. The tractrix guys like 'em too.
gedlee said:
OK, well, I am pretty sure you conceded that limited vertical beamwidth would be attractive, but the problem was the elliptical horn is hard to build.
I'm sorry if I misrepresented you, so perhaps you could clarify. This is what I said I thought you agreed with:
Wayne Parham said:
For clarification, what part of that statement do you disagree with?
As for the deal about what data is available, what guesses might be made and what is "better", it seems to me that is a bit of a red herring. I mean, what is "better" - a device that offers CD in the vertical but when installed in a loudspeaker creates nulls that cut the pattern in half or even two-thirds? Or a device that loses vertical pattern control at the low end but allows a loudspeaker to be built with a much larger clean forward lobe, having nulls spaced further apart? Who decides what is "better" when both are compromised? I expect you might say the data decides but it isn't that clear when you have competing priorities.
To me, it would be a no-brainer if you had CD through the passband, 90°x40° to 90°x50° beamwidth, and vertical mouth dimension small enough that the null angles could be set outside the pattern. That's the holy grail, in my opinion. However, I'm "guessing" it would be difficult to realize all these competing priorities in one device. Not a real tough "guess" since that's what has been found over and over again for the last few decades. It's part of CD horn history. Maybe that means you and I are both "guessing" the same thing.
Still, it may be possible, and is certainly a worthy goal. Even without vertical pattern control down low, I think the behavior of the device can be optimized for use in a loudspeaker where drivers are stacked vertically on a baffle, and that optimization requires tight vertical spacing. I believe the right design is a balanced approach, not just putting an axisymmetrcial horn on the baffle and saying that's the best I can do.
I am not confident that an elliptical waveguide can be made CD through its bandwidth, but I am sure that a loudspeaker with a 90° round waveguide cannot be made CD in the vertical. In fact, most loudspeakers like that have 60° of their 90° vertical pattern, dirty, outside the nulls and out of phase with the forward lobe. Even well above the crossover region where there's no interference, you still have the issue that high frequencies at large vertical angles serve only to create reflections from the ceiling and floor. For these reasons, to me, the axisymmetrical waveguide/horn is not an attractive compromise and an asymmetrical waveguide/horn can be used to better result.
I'm not saying any asymmetrical horn is better, but one can certainly be made to be better. For example, I do think your round OS waveguide is better than an old sectional horn. No question. But I also think if you make your elliptical waveguide with similar profile and size the proprtions right, it will be a better compromise between pattern control and null angle. I'm hoping you can find a way to make them and to cut your foam to fit, because I know those things are very important to you. I think the build will be the hardest part, probably the only thing that might hold you back. Just a "guess".
Certainly a horn or waveguide with smaller vertical mouth will have a larger clean forward lobe between the nulls. To me, that's a better approach. The fact that you are looking at an elliptical horn implies that you think it may be worthwhile too. I see a lot of products coming out with this general shape, so I think a lot of others are on the right track.
Or if you would prefer to stick with round horns, that's cool with me. The tractrix guys like 'em too.
gedlee said:
I don't sell speakers. On another site, I have been accused of surreptitiously selling yours. 😉
gedlee said:
It didn't happen. It's his fallacious argument that's at issue.
He likes 'em, that's fine, but it's also irrelevant. We're doing data analysis here, not subjectivist blather.... 🙄
ZilchLab said:
... and data shows GeddLee's offerings as being not that special as one may expect. In some respect they are simply worse than those from the bigger vendors.
To make a horn/waveguide needs some foundations in physics. But when it comes to practical issues all engineers have to use heuristics and simulations and simply try things by doing. The analytical solution of the wave equation ain't it at all.
gedlee said:
Um.....no one has denigrated him at all. If he is going post his subjective analysis then questions will be asked and raise....to me he is just defending your speakers but this thread should have little to do with your speakers and they do not need defending.
I couldnt find anything anywhere that insulted him or denigrated his choice except that he was the one posting the JBL PT is complete crap and rambles on about cars (porsche vs Mustang, etc)....that is total pointless in this thread (IMO). Does he even own a Porsche? (I had one! Traded it for a Mini-van and two kids!)
Saying something is CRAP without data to prove it is combative period and belongs on some site like audio asylum where science means nothing and faith in hearing is everything.
Actually both you and him have had the same tone....You have been very subjective in saying ITS ALL CRAP except your speakers (all waveguides except yours, I wonder how Duke, Mr Parham, Emerald Physics can even survive as your competition 😉 ) Yet you do not prove it using the data or explain properly why the data shows how crappy the JBL is (it is afterall $5). That is all Zilch asks and for some reason no one will give him the straight answer. Im sorry but you sound less like a PhD sometimes then all your great research shows. I do mean great, nothing to me has been more exciting in audio over the past several years and I just wish the masses would understand it better!!
Its funny you say you do not insult other customers, do you not realize how your tone and bluntness does insult people?? Kevin Haskins pointed out on the first page how you error in your blunt opinion but you seem not to care about how you are preceived by many. Let's just say it doesnt help sales 😉
Sheldon said:
Sheldon
Yes, you are correct that the mids would make a big difference in a high SPL situation. And I also agree with you that two-way is the better choice for smaller venues.
Some interesting pages on OB coordinates.
http://mathworld.wolfram.com/OblateSpheroidalCoordinates.html
http://en.wikipedia.org/wiki/Oblate_spheroidal_coordinates
Take a look at this picture. Its quite interesting to me to see such a form in nature.
http://mathworld.wolfram.com/OblateSpheroidalCoordinates.html
http://en.wikipedia.org/wiki/Oblate_spheroidal_coordinates
Take a look at this picture. Its quite interesting to me to see such a form in nature.
Re: Re: Re: My personal opinions of various design philosophies
Hi Wayne
It was more than that, its that I see a lot of tradeoffs in the whole concept of an elliptical waveguide and I'm not sure if the tradeofss will outweigh the advantages. As I said, I am building an elliptical waveguide now and we will see. But I'm not ready to jump to the conclusion that elliptcal is better as you and others are.
Better to me means less vertical aberation with NO and I repeat NO loss in horizontal control. I will do nothing that makes the horizontal control worse even if the vertical is better. We've been over this before.
If 90 x 40 could be done with both axes ideal then yes, this would be perfect. But such a thing is not possible. Which of the tradeoffs is the better set? That remains to be seen, but there certainly is nothing but hypothetical claims that an elliptical HF pattern IS the better set.
I disagree completely with your last statement. I see no evidence that said statement is true.
Wayne Parham said:
Hi Wayne
It was more than that, its that I see a lot of tradeoffs in the whole concept of an elliptical waveguide and I'm not sure if the tradeofss will outweigh the advantages. As I said, I am building an elliptical waveguide now and we will see. But I'm not ready to jump to the conclusion that elliptcal is better as you and others are.
Wayne Parham said:
Better to me means less vertical aberation with NO and I repeat NO loss in horizontal control. I will do nothing that makes the horizontal control worse even if the vertical is better. We've been over this before.
If 90 x 40 could be done with both axes ideal then yes, this would be perfect. But such a thing is not possible. Which of the tradeoffs is the better set? That remains to be seen, but there certainly is nothing but hypothetical claims that an elliptical HF pattern IS the better set.
I disagree completely with your last statement. I see no evidence that said statement is true.
gedlee said:
So the JBL response showed those peaks and dips? and the JBL impulse is not compact enough?
ZilchLab said:
If I misread your statement then I apologize, but you have to admit that
"That's Patrick trying to convince everyone that he's made the right purchase decision."
sounds very much like an acusation that Patrick made the wrong decision.
catapult said:
That only works if the throat has a 0 degree entrance angle. It cannot be done when this angle is 6 dgrees all the way around.
doug20 said:
I really don't know what specific JBL product you are refering to, I'm just answering the question that you claim no one would answer. When you attack people who do answer, its no wonder few do.
- Status
- Not open for further replies.