Huh, you know I should have made the connection when it was mentioned that the sharp, narrow angle in the XT1086 waveguide was there to increase the horizontal coverage. It's like having a little elliptical horn inside the bigger elliptical horn. Except that the little one was actually stood "upright" to increase the dispersion. Interesting. Well there goes my idea of crossing an asymmetric horn at its vertical cutoff to help the flaring vertical response fill in what would normally be a null between the horn and woofer.
It's interesting to read these last posts and look back on my earlier elliptical waveguide test.
http://www.diyaudio.com/forums/showthread.php?postid=1649609#post1649609
It would be interesting to see what the flip does to vertical lobes.
http://www.diyaudio.com/forums/showthread.php?postid=1649609#post1649609
It would be interesting to see what the flip does to vertical lobes.
There is a point of coverage beyond which a diffractionless waveguide is not possible. It is not possible to "pull" the pattern out to an arbitrary wide coverage for a given throat size and so some diffraction is required. It takes some experience and some knowledge of what is happening to know when this diffraction is required and when it is not. I still contend that diffraction is never desirable, but it can sometimes be necessary for wider patterns.
Tom, this is so true. So much of the misconceptions found here come from the manufacturers marketing about how their products don't have any flaws. If you don't know to suspect them you just don't look.
I was once talking with awell known EV horn engineer (Cliff Hendrickson - a loooong time ago) and asked if he wasn't concerned with the large pattern flip across the oblique axis of a square or rectangular horn. He had no idea what I was talking about, but some years later I saw some published data showing exactly this effect. He had never even looked before and just assumed that he knew what would happen, but it didn't happen that way.
Tom, this is so true. So much of the misconceptions found here come from the manufacturers marketing about how their products don't have any flaws. If you don't know to suspect them you just don't look.
I was once talking with awell known EV horn engineer (Cliff Hendrickson - a loooong time ago) and asked if he wasn't concerned with the large pattern flip across the oblique axis of a square or rectangular horn. He had no idea what I was talking about, but some years later I saw some published data showing exactly this effect. He had never even looked before and just assumed that he knew what would happen, but it didn't happen that way.
mige0 said:
Hi Michael,
Very interesting read. It's making me rethink the ESS Great Heil AMT. I had dismissed it as being too wide to use as a dipole but the horn loading, while crude compared to the soongsc profile, is making me think again.
http://www.parts-express.com/pe/showdetl.cfm?Partnumber=264-600
An externally hosted image should be here but it was not working when we last tested it.
Hi Lynn
Hey thanks much for the links, boy you can see the yet to be “Synaudcon” all over Don’s stuff at Altec, pretty cool.
I think the manufacturing cost was the primary problem with the multi-cell horns at the time, I have to admit I love the way they look. I have a big old 3 by 5 multi-cell on the roof of a shed as a yard decoration.
My first sound systems in the old days used various Altec horns which we rented and later bought.
There are issues using a multi-cell up high but the device wasn’t expected to put out 20KHz, “back then” most people couldn’t hear that high and it was assumed that since there wasn’t any in the program, way high was un-needed which dovetailed into it being very difficult.
At home sized distances, getting to or close to 20KHz should be a goal
I had been collecting horn drivers for at least 15 years for a horn system “someday” and had a set of the slant plate lenses too at one point.
I have to say a 375 on the large lens horn combo is about as close to “perfect” in it’s band pass as I have ever seen from a loudspeaker. By perfect I mean free of secondary things in the measurements, nice smooth simple curves, high efficiency, nearly resistive in band.
That was the problem I guess, I had a fair number of things that worked well over some range but saw no way to combine them into a system without interference due to the distances..
Boy, True sonic, I haven’t heard that name in a while. There used to be a store called Olsen electronics down in Chicago and in high school we used to go down there and drool on the new stuff (not really, wish for might be better) and occasionally buy from the junk area. I found an 8 inch true sonic woofer there and bought it. I had never heard of the company but I was impressed how well made it was.
I wonder where that ever went umm.
Soongsc
What is the driver in the post with the 8KHz lobe?
I can’t see from that picture what you have at the center.
TrueSound linked the complex conic for augerpro.
I don’t know if you can see it or not but you have a narrow vertical angle from the driver to mouth but in the horizontal plane there is little to no expansion out to the point where the horizontal wall angles intersect. Roughly, to me it appears to be a CD diffraction horn BUT instead of rectangular with hard bends everything is as smooth and gradual as possible, in a very much Earl looking way.
Hi Earl
I think if sound were visible, a lot of it would be more obvious to be sure, then you have the business of selling.
Yeah it’s true and I hate what they have done to “hifi” in the interest of selling.
Magic knobs, magic rocks, magic clocks, ugh that is so tiring.
Pro isn’t immune though to be sure, it’s just that a larger percentage have access to measurement gear ‘if” they choose to use it.
On the other hand, pro-commercial has a Boze equivalent where the saying is “you just can’t pay more for sound that good”.
Yes it is an annular radiator, it has a path which leads from the radiator to the center to sort of a collector where it turns 90 degrees and goes forward into a mildly tapered exit ending at one inch.
I would have thought there would be more pictures of them on line but I only saw the one neo version JBL Oem’s . Even that photo doesn’t exactly show how it works.
http://images.google.com/imgres?img...org.mozilla:en-US:official&sa=N&start=54&um=1
Catapult, I am not sure the internals are the same (as in are the replaceable elements the same?), the magnet systems are not. I had tested the neo 4552 and a couple others but I liked the 4550 the best. It may not be the same parts as the hf response is somewhat different between them.
Internally it is certainly at least similar to the neo version, also a ring radiator, has one summation path length ending in a smaller diameter than the exit and a conic taper leading out.
Best,
Tom
Hey thanks much for the links, boy you can see the yet to be “Synaudcon” all over Don’s stuff at Altec, pretty cool.
I think the manufacturing cost was the primary problem with the multi-cell horns at the time, I have to admit I love the way they look. I have a big old 3 by 5 multi-cell on the roof of a shed as a yard decoration.
My first sound systems in the old days used various Altec horns which we rented and later bought.
There are issues using a multi-cell up high but the device wasn’t expected to put out 20KHz, “back then” most people couldn’t hear that high and it was assumed that since there wasn’t any in the program, way high was un-needed which dovetailed into it being very difficult.
At home sized distances, getting to or close to 20KHz should be a goal
I had been collecting horn drivers for at least 15 years for a horn system “someday” and had a set of the slant plate lenses too at one point.
I have to say a 375 on the large lens horn combo is about as close to “perfect” in it’s band pass as I have ever seen from a loudspeaker. By perfect I mean free of secondary things in the measurements, nice smooth simple curves, high efficiency, nearly resistive in band.
That was the problem I guess, I had a fair number of things that worked well over some range but saw no way to combine them into a system without interference due to the distances..
Boy, True sonic, I haven’t heard that name in a while. There used to be a store called Olsen electronics down in Chicago and in high school we used to go down there and drool on the new stuff (not really, wish for might be better) and occasionally buy from the junk area. I found an 8 inch true sonic woofer there and bought it. I had never heard of the company but I was impressed how well made it was.
I wonder where that ever went umm.
Soongsc
What is the driver in the post with the 8KHz lobe?
I can’t see from that picture what you have at the center.
TrueSound linked the complex conic for augerpro.
I don’t know if you can see it or not but you have a narrow vertical angle from the driver to mouth but in the horizontal plane there is little to no expansion out to the point where the horizontal wall angles intersect. Roughly, to me it appears to be a CD diffraction horn BUT instead of rectangular with hard bends everything is as smooth and gradual as possible, in a very much Earl looking way.
Hi Earl
I think if sound were visible, a lot of it would be more obvious to be sure, then you have the business of selling.
Yeah it’s true and I hate what they have done to “hifi” in the interest of selling.
Magic knobs, magic rocks, magic clocks, ugh that is so tiring.
Pro isn’t immune though to be sure, it’s just that a larger percentage have access to measurement gear ‘if” they choose to use it.
On the other hand, pro-commercial has a Boze equivalent where the saying is “you just can’t pay more for sound that good”.
Yes it is an annular radiator, it has a path which leads from the radiator to the center to sort of a collector where it turns 90 degrees and goes forward into a mildly tapered exit ending at one inch.
I would have thought there would be more pictures of them on line but I only saw the one neo version JBL Oem’s . Even that photo doesn’t exactly show how it works.
http://images.google.com/imgres?img...org.mozilla:en-US:official&sa=N&start=54&um=1
Catapult, I am not sure the internals are the same (as in are the replaceable elements the same?), the magnet systems are not. I had tested the neo 4552 and a couple others but I liked the 4550 the best. It may not be the same parts as the hf response is somewhat different between them.
Internally it is certainly at least similar to the neo version, also a ring radiator, has one summation path length ending in a smaller diameter than the exit and a conic taper leading out.
Best,
Tom
As I understand it, there are several variations on the theme but the pic shows the basic concept of the BMS drivers. Point 4 is where the voice coil joins the V-shaped membrane.
Link to the patent:
http://patft.uspto.gov/netacgi/nph-...5,878,148.PN.&OS=PN/5,878,148&RS=PN/5,878,148
Link to the patent:
http://patft.uspto.gov/netacgi/nph-...5,878,148.PN.&OS=PN/5,878,148&RS=PN/5,878,148
Attachments
It's fun to carefully re-read the old Altec, JBL, and Stephens sales literature. You can see where they shade the truth and disguise it with marketing phrases that sound impressive but are meaningless.
I remember reading that the metalworkers who welded the multicell horns were the highest-paid production workers in the Altec plant. (Altec installed A2, A4, and A5 systems in more than 10,000 theaters by 1970, and the large widescreen theaters had three speakers behind the screen, so at a conservative guess, that's 15,000 multicells made between 1945 and 1970. That's a lot of horns.)
There must have been enormous management pressure to cut costs and increase manufacturing speed for the theater-horn line. The engineers held the line between theater/pro applications and everything else, with multicells reserved for theater applications, and the sectorals for studio monitors and home use. But the polar patterns tell the tale of which horns were better-behaved in the critical midrange.
Although both Altec and JBL ballyhooed the Mantarays and Bi-Radials to the skies when they came out in the Seventies, you do wonder how much reducing manufacturing costs had to do with it. Neither manufacturer made a big deal about how bad sectorals were until they had rolled out the Mantarays and Bi-Radials as all-purpose replacements for both sectorals and multicells.
I remember reading that the metalworkers who welded the multicell horns were the highest-paid production workers in the Altec plant. (Altec installed A2, A4, and A5 systems in more than 10,000 theaters by 1970, and the large widescreen theaters had three speakers behind the screen, so at a conservative guess, that's 15,000 multicells made between 1945 and 1970. That's a lot of horns.)
There must have been enormous management pressure to cut costs and increase manufacturing speed for the theater-horn line. The engineers held the line between theater/pro applications and everything else, with multicells reserved for theater applications, and the sectorals for studio monitors and home use. But the polar patterns tell the tale of which horns were better-behaved in the critical midrange.
Although both Altec and JBL ballyhooed the Mantarays and Bi-Radials to the skies when they came out in the Seventies, you do wonder how much reducing manufacturing costs had to do with it. Neither manufacturer made a big deal about how bad sectorals were until they had rolled out the Mantarays and Bi-Radials as all-purpose replacements for both sectorals and multicells.
gedlee said:
Dr. Geddes has drawn our attention to an important point with significant implications. What does it really mean when "below this point the polar response will go wider in the vertical than the horizontal because a narrow dimension diffracts wider than a wider one"?
Earl is telling us that the diffraction is starting a higher frequency in the vertical plane; diaphragm loading is starting to be lost, and diffracted and reflected energy from the horn-mouth is starting to become important. Loss of pattern control means an overall degradation of performance in several domains (reflections in time, ripples in frequency, uneven dispersion, and unwanted diaphragm excursion), not only pattern control.
What are the full implications of dissimilar diffraction-onset frequencies in different planes? In old-school thinking, the attitude was that "everything can be equalized" - but that's not true when the group-delay performance becomes sensitive to emission angle, and EQ-optimization for one angle degrades performance everywhere else.
If you were clever enough, you could stagger the vertical and horizontal diffraction-onset frequencies so the overall diffraction-onset would be smoothed-out, but that would probably result in a horn that was only moderately wider than it was high.
As usual I find the whole diffraction issue put much more complicated than necessary.
Simply - you cant squeeze a horn to your desire - not even in one plane *and* assume stellar operation.
Anything is a trade off - and horns actually never ever were easy to handle (to put it mildly) with respect to multi way due to its sheer room consumption (which can *not* be compensated for by lowering the XO point because the mouth will have to grow at the same time) and due to the limitations of short listening distance in domestic apps.
Its not out of the blue that coaxial designs mostly were and are done with horns – exactly out of that reason – meaning the inter-driver XO distance requirements are *much* easier to meet by direct radiators.
Tom Danley's tapped horns are great designs in this regard – way better than all that "just put a small horn into the mouth of a bigger one".
Having tamed a big EV of that sort – uhh - nothing I'd like to do anymore.
For the classical approach of vertical driver stacking, we can be happy *if* there is a fair homogeneity to gain in the horizontal plain at all – and after that we can look what can be done in the vertical plain . From auditioning so far I can tell that even the vertical plain *already* is a "no issue".
Thus - for me and for now – its not "per se" a clear decision in favour to a conventional 15" mid-bass / 15 " tweeter-horn arrangement versus a 8" low-mid / 5 " tall cylinder horn / 8" low mid arrangement.
Quite in contrary.
Anyway....
----------
Thanks for all your warm-hearted comments on my paper – A lot of "new" input to digest and possibly a lot of "inspiration" to try different things later on.
The sectoral / multicell is a "seductive solution" for the whole bouncing between the horizontal planes and also for the top end irregularity.
The AMT has closely spaced slits anyway – and the bars in between could "easily" be extended to the mouth.
In the vertical plain the folds of the diaphragm could be used as a starting point.
I just don't know what sonic trade off''s to be due ?
Any experiences out there with such smallish multicell patterns?
No, I have no better data than the referenced pix posted by soongsc.
Zoom to your desire, print it out and smooth the contour by selecting the "right" blade for your jig saw (if you have some skills in this regard too - this works pretty well, as anyone confirmed )
Go on and report back !
Though don't expect the ESS AMT to make a great horn by its own...
Michael
Simply - you cant squeeze a horn to your desire - not even in one plane *and* assume stellar operation.
Anything is a trade off - and horns actually never ever were easy to handle (to put it mildly) with respect to multi way due to its sheer room consumption (which can *not* be compensated for by lowering the XO point because the mouth will have to grow at the same time) and due to the limitations of short listening distance in domestic apps.
Its not out of the blue that coaxial designs mostly were and are done with horns – exactly out of that reason – meaning the inter-driver XO distance requirements are *much* easier to meet by direct radiators.
Tom Danley's tapped horns are great designs in this regard – way better than all that "just put a small horn into the mouth of a bigger one".
Having tamed a big EV of that sort – uhh - nothing I'd like to do anymore.
For the classical approach of vertical driver stacking, we can be happy *if* there is a fair homogeneity to gain in the horizontal plain at all – and after that we can look what can be done in the vertical plain . From auditioning so far I can tell that even the vertical plain *already* is a "no issue".
Thus - for me and for now – its not "per se" a clear decision in favour to a conventional 15" mid-bass / 15 " tweeter-horn arrangement versus a 8" low-mid / 5 " tall cylinder horn / 8" low mid arrangement.
Quite in contrary.
Anyway....
----------
Thanks for all your warm-hearted comments on my paper – A lot of "new" input to digest and possibly a lot of "inspiration" to try different things later on.
The sectoral / multicell is a "seductive solution" for the whole bouncing between the horizontal planes and also for the top end irregularity.
The AMT has closely spaced slits anyway – and the bars in between could "easily" be extended to the mouth.
In the vertical plain the folds of the diaphragm could be used as a starting point.
I just don't know what sonic trade off''s to be due ?
Any experiences out there with such smallish multicell patterns?
augerpro said:
No, I have no better data than the referenced pix posted by soongsc.
Zoom to your desire, print it out and smooth the contour by selecting the "right" blade for your jig saw (if you have some skills in this regard too - this works pretty well, as anyone confirmed
)catapult said:
Go on and report back !
Though don't expect the ESS AMT to make a great horn by its own...
Michael