Mantarays look pretty rough from the measurements I've seen. Not nearly as nice as the EV or JBL contemporaries. The JBL 2360 is hard to beat from a response smoothness and variation off axis point of view.
Multicells, really? Nice vintage look but does anyone have any response curves?
David S.
There is a big difference in size and price between an 811 and a 1005, Bear! I love my 1005s, but not everyone wants a horn that big. And the driver will be bigger, heavier and cost a lot more for the 1.4" throat horns.
The bigger horns do sound great, tho. Can't argue that. For me the 500Hz horn worked better than the 300Hz, tho both sound great. Here are mine.
Yes David, multicells. I'd own the 1505s if I could afford them.
The bigger horns do sound great, tho. Can't argue that. For me the 500Hz horn worked better than the 300Hz, tho both sound great. Here are mine.
Yes David, multicells. I'd own the 1505s if I could afford them.
Attachments
Yes. I have measured them - they are as flat as the driver that is on the back end?
Of course they are not "perfect" nothing is. But as horns go, they are very good. Can't post any curves because I have none in electronically transmittable format, and those on old thermal graph'd paper have gone the way of the dodo bird. _-_-bear
PS. problem with the Mantarays is that they are based on diffraction for starters, and the apparent acoustic center varies with frequency - easy to hear nearfield, they might be ok fine at 100 yds - talking about the BIG ones now, and even the medium ones (about 20" sq. iirc). They just sound weird.
Last edited:
Each cell has a narrowing pattern at HF as any single exponential horn would have. The midrange pattern is OK as I recall, but above around 5K they start to "finger" badly, at 16K the finger lobes are quite deep, assuming the polars I recall had 5 dB increments, probably 15 dB drops between cells.
They sound better than the pattern would indicate.
Like Pano, I'd prefer them over the 511 horn, I'd also prefer to be shot to death rather than hung to death
.
It's strange, I read so much about the HF lobing with the multi-cells, I have even seen at least one plot that showed it. But I just don't hear it. And no one I've ever seen in front of a pair has ever mentioned it, ever. It should be there, but just doesn't seem to be a problem in practice. I do notice the HF beaming with the 811 and 511 horns, as well as the round types. Usually not a big deal on axis or near it.
Okay, now the fingering or lobing is interesting.
Keele shows a polar at 10k ("Whats so sacred about exponential horns") and you clearly see 6 lobes. Unfortunately the multicell was 2 x 5! Rather than it being a simple case of the polar taking on individual beams from each cell, it is closer to what I saw when modeling line arrays. Lobes form at the sides and as frequency goes up they fold in and extral lobes join. The number of lobes has noting to do with the number of cells, so at high frequencies it isn't the case the broad frequency pencils of sound come out from each cell.
Look at any polar curves and you will see it.
They do have more constant directivity than radials do (so a more rolled off top end on axis) but the horizontal and vertical directivities are not very smooth. Still, I'm sure with a little bit of EQ they can sound pretty good.
Regards,
David S.
Keele shows a polar at 10k ("Whats so sacred about exponential horns") and you clearly see 6 lobes. Unfortunately the multicell was 2 x 5! Rather than it being a simple case of the polar taking on individual beams from each cell, it is closer to what I saw when modeling line arrays. Lobes form at the sides and as frequency goes up they fold in and extral lobes join. The number of lobes has noting to do with the number of cells, so at high frequencies it isn't the case the broad frequency pencils of sound come out from each cell.
Look at any polar curves and you will see it.
They do have more constant directivity than radials do (so a more rolled off top end on axis) but the horizontal and vertical directivities are not very smooth. Still, I'm sure with a little bit of EQ they can sound pretty good.
Regards,
David S.
Fwiw, The narrowing of each lobe with increasing frequency (way up high) and the interference pattern an array of sources produces are two separate issues. The narrowing was more audible at large distances at high frequencies, don’t forget, most of this stuff was used in large scale sound for ages before it found It’s way into a tiny living room.
I have had a number of large old horns in my system over the last 40 years and obviously have a fondness for the approach.
My personal guess is that the flaws present in good multi-cell horns were not the reason for their demise. Rather, once it became possible to mold polyester / fiberglass horns and once it became possible to address directivity with new shapes, the huge relative cost of the multicell was it’s doom.
The direction I chose to pursue was the straight conical horn, for commercial sound, this gives a constant radiation angle down to the pattern loss F set by the mouth size and horn wall angle. This produces an essentially constant sound field as it acts like it is a single point source driver at the apex.
Here is the CLF file for an SH-50 at 10KHz, the beam width is 50 degrees.
Tom
I have had a number of large old horns in my system over the last 40 years and obviously have a fondness for the approach.
My personal guess is that the flaws present in good multi-cell horns were not the reason for their demise. Rather, once it became possible to mold polyester / fiberglass horns and once it became possible to address directivity with new shapes, the huge relative cost of the multicell was it’s doom.
The direction I chose to pursue was the straight conical horn, for commercial sound, this gives a constant radiation angle down to the pattern loss F set by the mouth size and horn wall angle. This produces an essentially constant sound field as it acts like it is a single point source driver at the apex.
Here is the CLF file for an SH-50 at 10KHz, the beam width is 50 degrees.
Tom
Attachments
In fact they represent the classic manufacturing trade off. The multicell horn has almost no tooling cost but very high labor content. As labor gets more expensive then the expensive tooling (and low piece price) of a radial becomes justified. The old pictures show multicells being soldered up from hand snipped tin over simple wooden forms. Highly skilled work but okay if labor is cheap.
David S.
The crew who built the multis were said to be some of the best paid at Altec, so you can imagine the savings by dropping the format. And the marketing spin to justify the replacement of the multi-cell.
I think that Tom has also hot on a good point. At long throws, there may be better ways to do it.
I think that Tom has also hot on a good point. At long throws, there may be better ways to do it.
I have a pair of 291-16As driving 311-90 horns and I have to say that even though the Symbiotik diaphragms are said to be less faithful than the aluminum ones, they represent some of the most timbrally correct sound I have ever heard. They are not multicell, so they're a little OT but they sure do sound good, even at fairly close range. I got one for $50 from a local want ad; got its mate for $575 plus shipping from an eBay seller.
If I am recalling properly you really can't get a multicell of the Altec type to go past about 14kHz because of the throat geometry (iirc).
There's no doubt that the requirements for large spaces is very different than the requirements for most (even the very very big) listening rooms. The lobing thing has been fought against for a long long time now... seems like every pro mfr of PA gear has a version of a "line source array" now. DSP, etc... Now being the perspective of looking back a few decades.
Tom, could not the "tapped" in midrange concept be applied to other expansions? And do you see IM/modulation effects on the HF driver's diaphragm??
_-_-bear
There's no doubt that the requirements for large spaces is very different than the requirements for most (even the very very big) listening rooms. The lobing thing has been fought against for a long long time now... seems like every pro mfr of PA gear has a version of a "line source array" now. DSP, etc... Now being the perspective of looking back a few decades.
Tom, could not the "tapped" in midrange concept be applied to other expansions? And do you see IM/modulation effects on the HF driver's diaphragm??
_-_-bear
Not really, but I have heard them sound pretty darn good on 811s on a couple of occasions. Crossover is key.
Bear,
Tom may be too modest to tell it like it is, but as one who has employed his Synergy and Paraline concepts in my own cabinets, I can say that they are a significant improvement over the “old school” separate horn approach like the A7, or a horn above a front loaded speaker.
I have been copying design features of various horn cabinets, and doing incremental upgrades on them for decades.
Tom Danley has done the same, and has made many of his own innovations.
Once I was aware of them, I tried and tested his concepts and listened to the results.
My posts in #2 and subsequent posts come because I have heard what those innovations can do sonically, copying a “good old” design when there are design concepts that take up similar space and have far more accurate acoustic response makes no sense to me.
Tom’s concept of coupling mid drivers through an an acoustical LP filter (the small port holes) near the HF driver horn throat entrance could be applied to other expansions, but HF dispersion is largely dependent on the horn wall angle set relative to the wavelength traveled.
In other words, an exponential, hypex or tractrix “Synergy “ type horn would “beam” the HF, since the initial throat would be a narrow angle compared to the latter expansion.
In a conical expansion the sidewall angle determines the directivity pattern down to a wavelength approximating the mouth area, constant directivity is a very good thing if even coverage is desired over a wide area.
As far as seeing IM/modulation effects on the HF driver's diaphragm, the pressure generated by other drivers going in to the HF driver is so much less compared to the pressure generated by the HF diaphragm that it is not a problem. If it were a problem, any co-ax design would suffer from it, empirically we know that they don’t.
Art
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.