Beaming … directivity … horn types and their typical directivity
As we‘ve already got two comments from guys who seem to see things differently … imo we need to make sure we all are thinking of the same definition/ threshold
I take it to be - 6dB or more. Do you agree … anyone not?
Although I’ve not heard anyone say it, there’s an argument that the threshold for horns have more latitude.
As it’s harder for a number of horn types to achieve “acceptable coverage”. What is acceptable varies a lot
If I could ask one slightly off topic question: does anyone know of a webpage or book that has a list/ table of horn types with their typical directivity.
Width of angle at which theyre say -6 dB
of course, there’s more variables than type. So AFAP with other factors being equal or close to equal.
And variables within type.
so to be meaningful stating assumptions mad. And/ or likely range.
The 15” woofer is in a 58cm baffle, crossed at 600hz. I always sit in the same spot. Maybe if it was crossed at 800 or higher it might sound more beam-y, which was my subjective opinion at the time I was dialing things in, but the 15PR400 seems to play very well up to 600hz without audible issues to my ears. It sounds very smoothly coupled with the horn (active 4th order LR). I’m actually adding another small 6” Azura with a Faital HF10AK, crossed somewhere between 3k-5k, hoping to hear a little more sparkle by breaking out the highs.
Not really it depends especially the horn type and crossover point. If you compare 2 way 4430 to say 4 way 4344 the 4430 doesn't beam in comparison. Assuming you are talking Horizontal as all systems have limited Vertical in comparison.
Rob 🙂
There are two possible issues here. One is the overall directivity of the system - how narrow the beam of sound it produces is. Horns as typically used in this thread (LeCleac'h style horns) with large woofers will produce higher directivity (narrower beam width) systems. The biggest effect I notice with these systems is the narrower imaging. You'll have a narrower soundstage and central sounds will sound more in the room / between the speakers when compared to a system with lower directivity. The second issue is rapid changes in the system directivity vs frequency such as when a woofer's beam width narrows at high frequency before crossing over to a tweeter with a wider beam width. These big woofer / horn systems actually do a pretty good job at avoiding this problem as the woofer and horn are roughly the same size so their beam width is about the same over a range of frequencies. For example, see the horizontal plot of my system in the lower left here: http://jhsaudio.com/images/170219_4wayJBLsystem/150104_4wayMonoData.png This is a 15" mid bass crossed over at 800Hz to a 385Hz LeCleac'h horn. You can see another issue with this type of system in the vertical plots - the large vertical spacing between drivers makes it tough to avoid issues in the vertical plane. However I don't notice those being too problematic.
Regarding beaming, directivity, woofers and horns for the 2 way design. It is always the horn that is the problem or the most difficult problem to solve. The woofer is never the cause of beaming in these designs. The large cone of the woofer is large enough to control the radiation pattern of some of the mid bass which can improve subjective impact, but not so large as to cause beaming near the Xover of these designs. Woofer choice is about sensitivity, HF breakup and tonality. To match the horn you want a light, fast cone that is a nice match to the horn. You'll still need to attenuate the horn and also watch out for HF breakup of the woofer above your Xover point.
You can verify the woofer directivity by going all the way back to Harry Olson's "Elements of Acoustical Engineering" published in 1947. You will find his polar plots of various geometries and drivers copied exactly into many modern texts. It is accepted science and for cone drivers you really don't need much more than the fundamental work on directivity by Olson. Add to that the additional research on diffraction done by Vance Dickason in "Loudspeaker Design Cookbook" and you have just about everything on the radiation of a cone driver in a box, IMHO.
I think it is always the horns that are the problem. Jean-Michel Le Cleac'h described that he got the best sound from his horns by toeing them in such that they cross about 2' in front of the listener. This can be a significant toe in for a small room. While the Le Cleach horns are quite nice in many ways, I do think they are a little hot in the midrange on axis and Jean Michel's recommendation appears to make a big improvement.
This problem is to be expected in the Le Cleac'h horns as it is a problem that is well documented and also the motivation for a lot of other work. We can see in Olson's polar plots and others that followed, that many horns suffer from a midrange pinch in radiation, often around 1,000 hz. I think Don Keele's paper "What's So Sacred About Exponential Horns?" (https://www.xlrtechs.com/dbkeele.com/papers.htm) is a really great study of this problem and he shows that even multi-cell horns have this problem. He then goes on to document his solution to the problem he developed for the Electro Voice HR9040 and others. These were the first constant directivity horns.
Many audiophiles prefer bi-radial horns for this reason and the Yuichi A-290 horn is a typical favorite. Call it the one to beat. Joseph Crowe's ES horns are a design which very explicitly tries to solve this directivity problem of axi symmetric horns like the Le Cleach and he publishes polar plots of all his designs and others including the Yuichi for this reason. It is the problem we are all trying to solve, I think.
You can verify the woofer directivity by going all the way back to Harry Olson's "Elements of Acoustical Engineering" published in 1947. You will find his polar plots of various geometries and drivers copied exactly into many modern texts. It is accepted science and for cone drivers you really don't need much more than the fundamental work on directivity by Olson. Add to that the additional research on diffraction done by Vance Dickason in "Loudspeaker Design Cookbook" and you have just about everything on the radiation of a cone driver in a box, IMHO.
I think it is always the horns that are the problem. Jean-Michel Le Cleac'h described that he got the best sound from his horns by toeing them in such that they cross about 2' in front of the listener. This can be a significant toe in for a small room. While the Le Cleach horns are quite nice in many ways, I do think they are a little hot in the midrange on axis and Jean Michel's recommendation appears to make a big improvement.
This problem is to be expected in the Le Cleac'h horns as it is a problem that is well documented and also the motivation for a lot of other work. We can see in Olson's polar plots and others that followed, that many horns suffer from a midrange pinch in radiation, often around 1,000 hz. I think Don Keele's paper "What's So Sacred About Exponential Horns?" (https://www.xlrtechs.com/dbkeele.com/papers.htm) is a really great study of this problem and he shows that even multi-cell horns have this problem. He then goes on to document his solution to the problem he developed for the Electro Voice HR9040 and others. These were the first constant directivity horns.
Many audiophiles prefer bi-radial horns for this reason and the Yuichi A-290 horn is a typical favorite. Call it the one to beat. Joseph Crowe's ES horns are a design which very explicitly tries to solve this directivity problem of axi symmetric horns like the Le Cleach and he publishes polar plots of all his designs and others including the Yuichi for this reason. It is the problem we are all trying to solve, I think.
Attachments
While the Yuichi isn’t biradial it’s shape is almost/ a close variant; and Crowe’s ES horns are basically a variation on the Yuichi
Where are the graphs that show the considerable beaming?
This appears to come back to my point that people vary in at what point they call a dispersion pattern beaming
If you can just look at the directivity curves. That tells you what you need to know. If it's rising in the last octaves it's going to beam. It should also be relatively uniform and smooth with no abrupt discontinuities.
Rob 🙂
Where are the graphs that show almost constant directivity? Where are the graphs that show perfect acoustic loading?
My understanding is that in a two-way system such as one using this horn and compression driver
https://www.athosaudio.com/2021/01/01/tad-th-4001/
https://www.usspeaker.com/radian 745neoBepb-1.htm
and this midwoofer
https://josephcrowe.com/blogs/news/altec-416-8b-in-100l-sealed ,
aberrations will arise from the woofer being directional (beaming) below the crossover frequency, whereas the horn does the opposite (broad coverage).
Thus, isn't beaming caused by the large difference in diaphragm size between the two drivers, and not the filtering?
Last edited:
I'm just the messenger here; never claimed that the plot of that horn/driver combo produced anything approaching constant directivity, which even way less than well schooled ones like me can see that it doesn't. But about the ES450 horn directivity pattern, might the narrowing with increasing frequency be more due to the driver than the horn? Here’s the off-axis response of that driver.
https://cdn.shopify.com/s/files/1/0..._0_15_30_45_Off-Axis_480x480.png?v=1697218820
Last edited:
I just tried the toe-in you mentioned, I had not heard this. Wow it actually works with the Le Cleac'h horns, they disappear more completely, and while the soundstage doesn't change much in breadth, it creates a really large and deep "cloud" from the far edges of the horns to the center. It sort of brings everything into focus, and cymbal taps and snare hits sound more in the room where before they sat back a bit. Great presentation.
Bryan, I am so pleased that tip helped. It certainly helped me and has been a continuous source of learning. It is also one of those things we discover where we say "Why is no one talking about this, this is so obvious, am I crazy?". And, of course you're not, and that's the beauty of looking back to people like Jean Michelle who were perfectly honest and curious about these phenomenon and only wanted to forward our learning.
I found a much more explicit polar plot of a 16" woofer by Olson. You'll notice that a 16" dia. cone driver has some narrowing appearing at 1k, but it's still quite nice and is very similar to the directivity of horns in this region, except for the darn midrange pinch. This is one of the benefits of the 16", that it's directivity should be similar in the Xover region as the horn, at least for most horns in consideration here.
Also see enclosed a chart of the peak amplitude or excursion plot of various diameter cones at 1 watt. This really shows why a 16" woofer is what we use in this two way design. To cover something like 30 to 1k, the limiting criteria is cone excursion and only a large woofer can cover this range without excessive cone excursion distortion. If you search back through this thread for Gary Dahl's explanation of his design, you'll see he further limited excursion by using a somewhat small sealed box for the woofer. This increases low end response and really controls excursion. He adds another woofer I think or sub for the lowest frequencies.
I found a much more explicit polar plot of a 16" woofer by Olson. You'll notice that a 16" dia. cone driver has some narrowing appearing at 1k, but it's still quite nice and is very similar to the directivity of horns in this region, except for the darn midrange pinch. This is one of the benefits of the 16", that it's directivity should be similar in the Xover region as the horn, at least for most horns in consideration here.
Also see enclosed a chart of the peak amplitude or excursion plot of various diameter cones at 1 watt. This really shows why a 16" woofer is what we use in this two way design. To cover something like 30 to 1k, the limiting criteria is cone excursion and only a large woofer can cover this range without excessive cone excursion distortion. If you search back through this thread for Gary Dahl's explanation of his design, you'll see he further limited excursion by using a somewhat small sealed box for the woofer. This increases low end response and really controls excursion. He adds another woofer I think or sub for the lowest frequencies.
Attachments
Several years ago and lacking way too much theoretical knowledge and DIY skills, I was nevertheless highly intrigued by Gary’s and Lynn Olson’s description of the sound and performance of this driver-and which then sold for ~25% of what it does now. https://greatplainsacoustics.com/pr...uAhu3kamXt8H2t-RzVfpslRh_-w6O1dmSZNvG17jzCje7
With Gary’s kind sharing of his plans, I then hired Jim Salk to clone Gary’s midwoofers. Here they are measured by Troy Crowe. https://josephcrowe.com/blogs/news/altec-416-8b-in-100l-sealed
Jim also built me Rythmik F12 sealed subwoofers. https://www.salksound.com/model.php?model=Rythmik+12+Subwoofer
So where am I still at now for how to use these lovely 416 midwoofers? Stuck in Crossroadsville with these options:
1/) Ask Troy to use the ES450 horn
https://josephcrowe.com/products/3d-cad-files-horn-no-1994-es450-biradial-for-jbl-2446-2-throat and a Fostex 900A tweeter with this anti-diffraction waveguide https://josephcrowe.com/products/horn-lens-no-2474-for-fostex-t90a?_pos=1&_psq=fostex&_ss=e&_v=1.0
For the midrange, either https://bcspeakers.com/en/products/hf-driver/archive/2/8/DCM50 -which Troy praised and described here between 18:58 and 19;10 as have a warm vocal presence.
OR https://www.sbaudience.com/index.php/products/compression-drivers/rosso-65cdn-t/
which Troy found outclassed the $1485. 18Sound1480Be, when fed by Troy’s Don Sachs preamp-which I have for driving my First Watt F4 or J2 amps. https://josephcrowe.com/blogs/news/...river-test-review?_pos=2&_sid=5c8354bca&_ss=r
There's also this add-on device for the 65CDN-T driver.
https://josephcrowe.com/products/3d...ence-65cdn-t?_pos=2&_psq=65cdn-t&_ss=e&_v=1.0
2/) Use block milled versions of the TH-4001 horns https://www.athosaudio.com/2021/01/01/tad-th-4001/ to build Pierre’s perfectly equalized speakers; see posts 15266, 15276. https://www.diyaudio.com/community/threads/beyond-the-ariel.100392/page-764
Troy claims that the ES horns produce zero coloration. But as its seems clear that the ES450 horn will not produce constant directivity (regardless of the driver used???), could I be correct in assuming that the narrowing in the projection of the midrange band will tend to produce a “They are Here” soundscape-much like what Gary said he receives from his JMLC 425/Yamaha JA6681 combo? https://galibierdesign.com/wa-trip-01/
Alternately, to my ears, I can’t know how much if any coloration the 4001 horns produce. But Pierre did mention that he receives a “You are There” soundscape from his speakers. Conversely, could that be because the 4001 horn produces much more constant directivity (regardless of driver used???) than the ES horns?
Perhaps a group survey of these options would provide a long overdue ticket out of Crossroadsville!
With Gary’s kind sharing of his plans, I then hired Jim Salk to clone Gary’s midwoofers. Here they are measured by Troy Crowe. https://josephcrowe.com/blogs/news/altec-416-8b-in-100l-sealed
Jim also built me Rythmik F12 sealed subwoofers. https://www.salksound.com/model.php?model=Rythmik+12+Subwoofer
So where am I still at now for how to use these lovely 416 midwoofers? Stuck in Crossroadsville with these options:
1/) Ask Troy to use the ES450 horn
https://josephcrowe.com/products/3d-cad-files-horn-no-1994-es450-biradial-for-jbl-2446-2-throat and a Fostex 900A tweeter with this anti-diffraction waveguide https://josephcrowe.com/products/horn-lens-no-2474-for-fostex-t90a?_pos=1&_psq=fostex&_ss=e&_v=1.0
For the midrange, either https://bcspeakers.com/en/products/hf-driver/archive/2/8/DCM50 -which Troy praised and described here between 18:58 and 19;10 as have a warm vocal presence.
OR https://www.sbaudience.com/index.php/products/compression-drivers/rosso-65cdn-t/
which Troy found outclassed the $1485. 18Sound1480Be, when fed by Troy’s Don Sachs preamp-which I have for driving my First Watt F4 or J2 amps. https://josephcrowe.com/blogs/news/...river-test-review?_pos=2&_sid=5c8354bca&_ss=r
There's also this add-on device for the 65CDN-T driver.
https://josephcrowe.com/products/3d...ence-65cdn-t?_pos=2&_psq=65cdn-t&_ss=e&_v=1.0
2/) Use block milled versions of the TH-4001 horns https://www.athosaudio.com/2021/01/01/tad-th-4001/ to build Pierre’s perfectly equalized speakers; see posts 15266, 15276. https://www.diyaudio.com/community/threads/beyond-the-ariel.100392/page-764
Troy claims that the ES horns produce zero coloration. But as its seems clear that the ES450 horn will not produce constant directivity (regardless of the driver used???), could I be correct in assuming that the narrowing in the projection of the midrange band will tend to produce a “They are Here” soundscape-much like what Gary said he receives from his JMLC 425/Yamaha JA6681 combo? https://galibierdesign.com/wa-trip-01/
Alternately, to my ears, I can’t know how much if any coloration the 4001 horns produce. But Pierre did mention that he receives a “You are There” soundscape from his speakers. Conversely, could that be because the 4001 horn produces much more constant directivity (regardless of driver used???) than the ES horns?
Perhaps a group survey of these options would provide a long overdue ticket out of Crossroadsville!
Last edited: