The Jbl 236x series horns can play to 500hz and the Comparable width 140hz tractrix horn I simed with a 4800sqcm mouth can be used to lets say 300hz....In elliptical form the height of the tractrix is about 15", not sure what Constant Directivity horn has similar height but with the wide width....I do have a pair of EV HP420's...
referring to the jbl 2366a just above in post 9280....it would be nice to have an impedance and FR measurement's for these horns... Im going to look shortly...I looked at the 2360....the supposed tractrix has more directivity 500hz and above. The 2360 and 2366 have the same length as my elliptical tractrix, somewhere around 30 inches.....I think my horn maybe 35".....the jbl 2366 is 54inches long. If we made a horn with 54inch inch axial length maybe the directivity would be more comparable? The CSA before the slot of the Jbls probably have very steep walls, in its advantage towards high Di. Still the tractrix plays low as the 54inch version of the Jbl's.....
So in order to get the DI higher than the tractrix horn we need to increase the axial length it appears....The width of all the 2360 horns is 31"...the width of the Tractrix horn I simulated was 30" wide and 37" long. The 60 and 65 are 32" long and the 66 is 54inches long.....so far my theory about directivity is still holding together, Maybe what I implied isn't being understood.
These are Horns, not waveguides....because all horns are waveguides but not all waveguides are horns, so it makes sense to just call it a horn since thats the separating factor right? I made the comment that if you increase the axial length a waveguide eventually becomes a horn....I think...that is accurate. I would like to see a 36" deep waveguide with a practical width mouth that doesn't load. So yeah its true mouth size is not the final factor but thats only technically speaking, practically speaking lets think in that space.....Practically my mouth/baffle size is limited....lets say its limited to 30" wide and tall.....we can control directivity down to a point limited by the width of the waveguide.....the waveguide at this point has directivity in a part of spectrum that a compression driver or any other suitable tweeter can't recreate....pointless point to say a waveguide can create higher directivity than a horn, unless the driver can support playback in the spectrum that is desired....in this case we are referring to compression drivers from 300hz-20khz
The challenge was
I am reading and understanding and not disagreeing so I don't feel like I am confused, So I leave you with this
So in order to get the DI higher than the tractrix horn we need to increase the axial length it appears....The width of all the 2360 horns is 31"...the width of the Tractrix horn I simulated was 30" wide and 37" long. The 60 and 65 are 32" long and the 66 is 54inches long.....so far my theory about directivity is still holding together, Maybe what I implied isn't being understood.
These are Horns, not waveguides....because all horns are waveguides but not all waveguides are horns, so it makes sense to just call it a horn since thats the separating factor right? I made the comment that if you increase the axial length a waveguide eventually becomes a horn....I think...that is accurate. I would like to see a 36" deep waveguide with a practical width mouth that doesn't load. So yeah its true mouth size is not the final factor but thats only technically speaking, practically speaking lets think in that space.....Practically my mouth/baffle size is limited....lets say its limited to 30" wide and tall.....we can control directivity down to a point limited by the width of the waveguide.....the waveguide at this point has directivity in a part of spectrum that a compression driver or any other suitable tweeter can't recreate....pointless point to say a waveguide can create higher directivity than a horn, unless the driver can support playback in the spectrum that is desired....in this case we are referring to compression drivers from 300hz-20khz
The challenge was
And so far only more Horns have been suggested because a large horn is the practical solution for my particular goal.
I am reading and understanding and not disagreeing so I don't feel like I am confused, So I leave you with this
Ok then, I look to be corrected....show me a horn with low DI...or...show me a Horn, that has lower Di than a waveguide that has the same expansion profile=) Looking at the polar of a 1000hz tractrix I think my theory is flying....
Last edited:
Well put. @camplo there is a free, easy to read white paper on horn directivity (for constant directivity devices) here which shows the function of horn mouth on directivity control:
http://excelsior-audio.com/Publications/Understanding_Horn_Directivity.pdf
On an asymmetrical constant directivity horn, the mouth dimension has to be larger for the angle having a wider nominal dispersion to maintain pattern control to the same frequency cutoff.
Last edited:
I've been down this road, as shaky as my understand may appear to you...it is enough to be applied successfully...I may have been literally wrong but practically I am good to go...
I have a limitation on mouth size....and so does pretty much everyone else....as I slowly recall the schooling Marcel gave me in this thread previously...I keep preaching practicality... practical only applies to whats practical for my design goals right......300hz-20khz is the goal for my tweeter.....In too many words, I said that the only way to achieve this goal with a compression driver is a large horn....So I focused in on mouth size...and Im referring to horns (Remember he asked me how is MY system considered high directivity not just ANY system).
"The horn is 33-36" wide......thats higher directivity than any 15" CD waveguide you could make." - the statement that got me in trouble maybe?....technically isn't true but practically I think it is. No one ever said anything I said was wrong in particular outside of the inclusion of dynamic radiators vs aspects of cone profile. That I do not claim to know about. Gedlee made a general statement but never was specific, and Fluid is providing general core knowledge on the topic.
If my mouth size is limited to 30 some odd inches.... and I want my tweeter to play to 300hz....what is the alternative to a compression driver and a large horn....that causes as much or more directivity...there is no easy answer that I know of.
The technicalities of waveguides/horns vs directivity I may not have memorized to a T, but I solved the above challenge with a large horn and compression driver and would like to know the situation that would causes more directivity, and fits the criteria....since thats the point I was trying to make.
I give precedence to Mouth size because, it is a the limiter on the LF of control...and one of the first dimensional limitations one will encounter in the design, so I am not off in my thinking. Mouth size says nothing of the axial depth or expansion profile.....yep.....obviously not going to choose the wide DI causing parameters for axial length and wall angle.
I have a limitation on mouth size....and so does pretty much everyone else....as I slowly recall the schooling Marcel gave me in this thread previously...I keep preaching practicality... practical only applies to whats practical for my design goals right......300hz-20khz is the goal for my tweeter.....In too many words, I said that the only way to achieve this goal with a compression driver is a large horn....So I focused in on mouth size...and Im referring to horns (Remember he asked me how is MY system considered high directivity not just ANY system).
"The horn is 33-36" wide......thats higher directivity than any 15" CD waveguide you could make." - the statement that got me in trouble maybe?....technically isn't true but practically I think it is. No one ever said anything I said was wrong in particular outside of the inclusion of dynamic radiators vs aspects of cone profile. That I do not claim to know about. Gedlee made a general statement but never was specific, and Fluid is providing general core knowledge on the topic.
If my mouth size is limited to 30 some odd inches.... and I want my tweeter to play to 300hz....what is the alternative to a compression driver and a large horn....that causes as much or more directivity...there is no easy answer that I know of.
The technicalities of waveguides/horns vs directivity I may not have memorized to a T, but I solved the above challenge with a large horn and compression driver and would like to know the situation that would causes more directivity, and fits the criteria....since thats the point I was trying to make.
I give precedence to Mouth size because, it is a the limiter on the LF of control...and one of the first dimensional limitations one will encounter in the design, so I am not off in my thinking. Mouth size says nothing of the axial depth or expansion profile.....yep.....obviously not going to choose the wide DI causing parameters for axial length and wall angle.
Last edited:
Once mouth size is fixed....or in the designers situation, once max mouth size has been achieved.....you can only increase directivity by wall angle and axial length. Since those are usually open choices with acceptable dimensions, we can not think about them so much, but we must live with the idea that we will not achieve directivity much past the mouth dimension regardless of axial length or wall angle....to me that makes Mouth dimension a little more important to speak about, though it is not the whole picture.
I compared the jbl CD horn (2360 and 2365) to the simulation of the comparable tractrix and the tractrix achieved higher DI over more of the spectrum
What is an example of a Horn with a low DI???
I compared the jbl CD horn (2360 and 2365) to the simulation of the comparable tractrix and the tractrix achieved higher DI over more of the spectrum
What is an example of a Horn with a low DI???
Last edited:
Mouth size will not ever increase directivity only allow directivity control to a specific frequency. Wall angle and length are tied together when the mouth size is fixed, they are only independent when the mouth size is allowed to vary.
In the examples you are giving if the mouth size is fixed extending the axial length forces the wall angle to narrow, which increases the directivity.
One could argue that the mouth size and other horn geometry factors would ideally be chosen in concert with the potential options for midrange or LF driver size, placement and the like to ensure a smooth transition without discontinuities at (or either side of) the acoustic crossover region…
A fixed mouth size does fix the lower frequency limit for pattern control - the simple formula (just four variables) for which is clearly explained in the paper I linked, with example plots of DI - but the choice of coverage angle is still important to match it with what you can or want to achieve below the horn’s frequency range.
A quick example; again assuming a constant directivity horn profile - your 30 inch mouth dimension sets a loss of pattern control at 370 Hz, if you’re aiming for a 90 degree nominal coverage in that dimension. If you’d prefer 60 degrees coverage, to keep sound of the walls, then the loss of directivity occurs at 555 Hz.
For the 60 degree example, at 300 Hz your 30 inch mouth dimension would give a coverage angle of 110 degrees. Aiming for narrower coverage in that dimension using multiple LF drivers would create an inconsistent polar pattern, and by nature an inconsistent set of reflections that would affect the CTA 2034 measures.
I forget if you’re making an axisymmetric horn or not, since you’ve recently shared images of speakers using bi-radial horns, but the other thing to be really aware of is pattern flip. This is also illustrated in the paper, and is a bigger issue for studio monitor use with a horn or waveguide playing to a low frequency, and especially an issue with bi-radial horns. An example from the paper, showing a limit of 2000 Hz for a 12” by 6” horn with a stated 90 x 45 degree coverage pattern:
You should also try not to forget the effects of mirror image sources on directivity, especially at LF where acoustic treatment is less effective or feasible.
A fixed mouth size does fix the lower frequency limit for pattern control - the simple formula (just four variables) for which is clearly explained in the paper I linked, with example plots of DI - but the choice of coverage angle is still important to match it with what you can or want to achieve below the horn’s frequency range.
A quick example; again assuming a constant directivity horn profile - your 30 inch mouth dimension sets a loss of pattern control at 370 Hz, if you’re aiming for a 90 degree nominal coverage in that dimension. If you’d prefer 60 degrees coverage, to keep sound of the walls, then the loss of directivity occurs at 555 Hz.
For the 60 degree example, at 300 Hz your 30 inch mouth dimension would give a coverage angle of 110 degrees. Aiming for narrower coverage in that dimension using multiple LF drivers would create an inconsistent polar pattern, and by nature an inconsistent set of reflections that would affect the CTA 2034 measures.
I forget if you’re making an axisymmetric horn or not, since you’ve recently shared images of speakers using bi-radial horns, but the other thing to be really aware of is pattern flip. This is also illustrated in the paper, and is a bigger issue for studio monitor use with a horn or waveguide playing to a low frequency, and especially an issue with bi-radial horns. An example from the paper, showing a limit of 2000 Hz for a 12” by 6” horn with a stated 90 x 45 degree coverage pattern:
You should also try not to forget the effects of mirror image sources on directivity, especially at LF where acoustic treatment is less effective or feasible.
Last edited:
It is not the same, Earl tried to explain it to you above. There is a big difference between the beaming that occurs from increasing the size of the radiating area to the directivity of a horn. A wider baffle constrains the radiation to a lower frequency just like a wider mouth does.
Maybe this is you using your own interpretation of already established words. If Earl's, Kyle's and my explanations are not enough I am not sure what will be.
Camplo, directivity can seem to increase with increasing transducer size if you look at it on particular frequency, say 3kHz. Directivity will vary with frequency, and quite a lot since low frequencies don't fit into a room and the smallest ones almost fit into average belly button. High or low DI can be on any sized transducer / horn depending on which frequency you look at. Check out 20Hz on any device and the DI is low, or 20khz and the DI is high.
Wavelength interacts with physical objects similarly as long as they are relatively sized to each other, no matter if they can be held at your hand or barely fits through the door. 15" wavelength will interact with 15" objects like 3" wavelength with 3" objects.
If you relate the wavelength and directivity to transducer size, then the directivity is rather constant. Ideal 15" driver on infinite baffle has same directivity index at ka=3 as 4" with ka=3, where the ka is related to diameter of the driver and thus is at different frequency for each of the drivers.
This you must have forgotten, it was discussed at least 450 pages back and maybe
https://www.diyaudio.com/community/...w-distortion-with-a-2-way.334757/post-5778117
edit. your plots say profile is the same, so both depth and mouth size is just a scaled? Names and visual data suggests directivity is about the same for both about octave apart. You could make it even bigger this way to get 30 degree pattern to 300Hz, just scale it up. Notice the mouth size gets pretty nasty, as well as the depth
Wavelength interacts with physical objects similarly as long as they are relatively sized to each other, no matter if they can be held at your hand or barely fits through the door. 15" wavelength will interact with 15" objects like 3" wavelength with 3" objects.
If you relate the wavelength and directivity to transducer size, then the directivity is rather constant. Ideal 15" driver on infinite baffle has same directivity index at ka=3 as 4" with ka=3, where the ka is related to diameter of the driver and thus is at different frequency for each of the drivers.
This you must have forgotten, it was discussed at least 450 pages back and maybe
https://www.diyaudio.com/community/...w-distortion-with-a-2-way.334757/post-5778117
edit. your plots say profile is the same, so both depth and mouth size is just a scaled? Names and visual data suggests directivity is about the same for both about octave apart. You could make it even bigger this way to get 30 degree pattern to 300Hz, just scale it up. Notice the mouth size gets pretty nasty, as well as the depth
Last edited:
I agree still, I'm not disagreeing with the information being given, it must be the way I am relaying it back...But your predictions of directivity don't line up with Hornresp, not sure why. I like your numbers better.
Regarding mirror images are you suggesting using redundant woofers to control direcitivty? like a MtM?
Im just gonna take a breather, I've been up since yesterday...that post I just deleted, directed to fluid had vertical polar maps when it should of been horizontal.Camplo, directivity can seem to increase with increasing transducer size if you look at it on particular frequency, say 3kHz. Directivity will vary with frequency, and quite a lot since low frequencies don't fit into a room and the smallest ones almost fit into average belly button.
Wavelength interacts with physical objects similarly as long as they are relatively sized to each other, no matter if they can be held at your hand or barely fits through the door. 15" wavelength will interact with 15" objects like 3" wavelength with 3" objects.
If you relate the wavelength and directivity to transducer size, then the directivity is rather constant. Ideal 15" driver on infinite baffle has same directivity index at ka=3 as 4" with ka=3, where the ka is related to diameter of the driver and thus is at different frequency for each of the drivers.
This you must have forgotten, it was discussed at least 450 pages back and maybe
https://www.diyaudio.com/community/...w-distortion-with-a-2-way.334757/post-5778117
edit. your plots say profile is the same, so both depth and mouth size is just a scaled? Names and visual data suggests directivity is about the same for both about octave apart. You could make it even bigger this way to get 30 degree pattern to 300Hz, just scale it up. Notice the mouth size gets pretty nasty, as well as the depth
Though I am apparently wrong anyway, also consider that when i say "increases directivity" I can be referring to the lowering of directivity control to a lower frequency....because at those frequencies the directivity was lower and then when you increase mouth size....directivity sure does increase in those areas where directivity was not before....yet we can say "extends" to be more technically correct for those who want to hear it that way....words....at least with a set horn profile and expansion rate, increasing mouth size, forces you to increase axial length, thus increasing directivity, so practically correct there.
Heres at least the two horns polars I wanted to share...
Last edited:
Agreed. That seems to be the simplest way to express 'the name of the directivity game'.
Anyone working with horns and waveguides needs Keele's classic formula burned into their melon, methinks haha!
I've read to keep H to V aspect ratio below 1.6:1, or pattern flip becomes rather inevitable. If any thoughts on this, thx.
As a related aside, one of the things i like about the MEH/Synergy/Unity design.......is that if it is made to reach down to a sub at around 100Hz or so, where the sub is still pretty much omni..... the MEH's automatically provide a smooth transition from the onset of its loss of pattern control, down to an omni handshake with the sub.
You may be interested in this paper which explains why JBL introduced fast flair horn along with snout-less driver. Slow flair rate horn is beamy at high frequency and it adds more distortion, but WE based older drivers had to be used with those deep horns. When they published this paper, they still did not use the term "waveguide", it was a transition period.
https://jblpro.com/en/site_elements...zed-aperture-horns-and-low-distortion-drivers
By our calculations, the initial flare rate in the older driver design was approximately 160 Hz, reflecting the need to drive the very large horns that were used in early motion picture systems. Today, we can double or quadruple that flare rate, inasmuch as many horns are now intended for nominal crossover at 800 Hz.
Yep, that's right. This new paper is about JBL's first series of the products finally called "Waveguide", and the paper I linked was about the last series of the products that JBL called "horn".
The interesting part is their first waveguides and their last horns are the same size and interchangeable (PT-F95HF and 2322), but they were comparing their newly introduced waveguide with much older long horns instead of their last generation of horn, most probably because the difference is too small to show on the paper. Based on those papers, it seems like the most drastic difference between horn and waveguide was caused by cutting the length of snout, at least in the history of JBL.
The interesting part is their first waveguides and their last horns are the same size and interchangeable (PT-F95HF and 2322), but they were comparing their newly introduced waveguide with much older long horns instead of their last generation of horn, most probably because the difference is too small to show on the paper. Based on those papers, it seems like the most drastic difference between horn and waveguide was caused by cutting the length of snout, at least in the history of JBL.
Last edited:
It's funny about the sizes. When I switched for 2234's to PTH1010 both 100X100 the PTH fit right in only difference was the bolt pattern! If you look at a 2234 vs a PTH the 2234 is significantly deeper. Added attachment more info on horn design.
Rob
Attachments
If I could do cardioid I would, Im not sure I could pull it off. Lets say Im trying to create as much direct energy to the listener as I can without being too elaborate. I seen guys design passive cardioid with vituix cad but I am not so technically able... though I used to be scared of unity horns and the like and now see it as possible...
With the large horn I will have, practically, the most directivity I can create using a single driver from 300hz to 20khz.. within my size constraints. Jmlc actually has a little more directivity and spherical wave as well but they are in the same ballpark jeebus....
"Slow flair rate horn is beamy at high frequency and it adds more distortion" - So I look at the parts of spectrum where DI gets high, and I get happy. For example when I compared the Tractrix to the Jbl236X series...the 2360 and 65 where the fair comparison, the dimension are very close, outside of profile. I aimed to simulate a horn with a similar wideness to replicate the horizontal pattern within Hornresp. The DI was higher in tractrix horn than the Constant Directivity horns of the similar dimensions for more of the spectrum that can be created with said devices.
"A quick example; again assuming a constant directivity horn profile - your 30 inch mouth dimension sets a loss of pattern control at 370 Hz, if you’re aiming for a 90 degree nominal coverage in that dimension. If you’d prefer 60 degrees coverage, to keep sound of the walls, then the loss of directivity occurs at 555 Hz." - My horn resp says I loose pattern control at about 330hz so I guess we are close there. Ballparking it anyway. What I don't understand is why a 30inch wide horn should loose pattern control at 370hz yet the jbl 236X series all claim pattern control to 200hz....All the plots show about a 130-110degree pattern at 200hz and rising DI increasing above 200hz..is this inaccurate info from Jbl? Seems like every above a 370hz or so would be at 180degrees....
The tractrix looks like from ~500hz and from ~750hz the tractrix has a higher DI than the 2360 and the 2365....and loads to 150hz to properly support a 300hz XO. Akin to polar, I feel that a smooth transition from high to low DI is a priority.
VituixCad 15"+32"baffle polar vs Hornresp ~28" wide 140hz (fta65.84) tractrix horn, at 300hz the Di within 1db of each other at 60degrees and at 403hz the Di was identical. So the polar is matched at Xo or potential XO's, what have you.
Last edited: