Is it possible to cover the whole spectrum, high SPL, low distortion with a 2-way?

Hi @camplo I’m responding to your original question in this thread
“Is it possible to cover the whole spectrum, high spl, low distortion with a 2-way?”

The answer is yes, so long as you are willing to optimize the system for efficiency rather than compact size. I have been intensely interested in the question of “how much juice can you squeeze out of the orange” especially because of the 3 years I spent designing speakers at Jensen where I designed full range drivers for the 94 Ford Probe, 95 Acura Vigor, 96 Honda Civic and 96 Jeep Cherokee.

I have also always liked 2-ways better than 3-ways, feeling that most of the time you create more problems than you solve with a 3-way and it takes a really good designer to solve them.

I love your question referring to the FaitalPro HF146, and asking what woofer to match with it that would yield 115dB+ SPLs. If you match it to any good 15” or 18” in a back loaded horn (sort of like a Lowther but much bigger) you can get all the SPL you want out of a 2-way. This is because the horn solves SPL limitations with efficiency instead of relying on Xmax and cone surface area.

So this morning I modeled the 15” coaxial B&C 15CXN88 in the Leonard Audio horn/TL software program. I use the 15CXN88 as a bass midrange + tweeter in my Bitches Brew live edge dipole design, crossing it over at 110Hz. This driver emphasizes my point because it is not even a subwoofer by any stretch. It is designed to be a bass midrange. But in the right size horn it still can function as a sub as you can see here:

Below is graphs for a 635 liter horn with 4 meter length and 30Hz tuning frequency. You can use it down to about 25Hz and then you’d want a steep high pass filter below that point to minimize excursion. This yields 100dB spl for 2.83 volts at 30Hz:

15cxn88_2.8volts.jpeg


This is the output at 28 volts input (100 watts) which you can see is 120dB:

15cxn88_28volts_spl.jpeg


And this is the excursion:

15cxn88_28volts_excursion.jpeg


At 30Hz it’s about 1mm; peak excursion is under 5mm at 40Hz. So even this bass midrange can put out 120dB at 30Hz.

If you look closely at the right enclosure panel, you can see the settings I used. The driver is mounted about 1/3 of the way from the closed end. I just whipped this up quickly, it’s not particularly optimized. There are midrange peaks from the horn that you can deal with via DSP or with transmission line stubs but it does give you an idea.

I really like back-loaded horns. A few years ago I made one for a Tang Band 4" titanium cone driver and it goes down to 55Hz and plays pretty loud and covers the entire spectrum with one driver. They sound WAY bigger than a four inch full range has a right to sound.

Coming back to this, if you used the Faital HF146 horn with a 15” or 18” subwoofer (let’s say the Faital 18HX1800, with 20mm xmax) you could probably attain 125dB or more, and still get good midrange, plus with the right crossover you can also achieve more or less constant directivity above 400Hz.
 
  • Like
Reactions: 1 users
It's a good sign to me you are able to identify by hear some issues you thoughts you wouldn't.

You cross around 250hz? How does the ppsl sound at this freq?

The ppsl is down -6 at 211hz at the moment. The horn is down -6 at 200hz. On multiple occasions I did my best to A/B test between the 15" and the ppsl, The ppsl is virtually identical to the 15" within the cross and passband. The ppsl had higher thd in some of the lower volume test, in the midbass, but the cone is heavier, and from what I know that is the trend. Still, low regardless, in the grand scheme of things and in the bass area so, a non issue.
After coming to the conclusion that I was struggling to tell which one was "better" or if there was a slot flaw I could hear, when the XO LP was in use. I stress tested the ppsl with heavy bass music at high volume. It took a very low 808 near 30hz to cause modulation on some material I could hear in the upper part the passband, that is was buried in the XO, meaning that modulation that I made out in that situation was also below the passband in spl, due to the low pass. It was a high volume, and THD wasn't an issue, so there is where I started becoming confident that the 2 18's could handle 200hz-30hz at the very least. It was a really high volume nothing near what I would need for normal playback, as well, only one channel.
Hi @camplo
Hello and welcome to the conversation! Just to let you know, You can easily [shift+windows key+s] select an area on the screen, it will copy to the clipboard, and then you can [cntrl+v] right into the post.

It will help you share your measurements clearly.
 
  • Like
Reactions: 1 user
1675754057792.png


Whats going on with Dirac live and active room treatment? I think active cardioid is, in a way, a form of active room treatment. One Idea I thought about was using woofers to create bass directivity but in a way I have not seen. Likely because its a horrible idea but here goes.

Lets say the red circles are the cardioid woofers and the blue is the main. This would be a top view... Does this make any sense lol?
1675754901868.png
 
Last edited by a moderator:
"Here is a sample spreadsheet in Microsoft Excel that generates elliptical tractrix horns based on basic parameter input, including the exit size and desired cutoff frequency:

cssCopy code
A B C D E F Input Formula Output Output Output a =0.05 0.05 2 100 b =0.3*A2 0.015 0.1 c =0.1 0.1 x =ABS(ROWS($A$7:A7)) 1 y =B2*(SQRT(B3^2+(C7/B2)^2)-B3)(1+(B4(C7/B2)^2))^0.5y_exit =D7/E7*y x_cutoff =B2/2/340*F7

You can enter the desired values of "a", "b", and "c" in cells A2, A3, and A4, respectively. The exit size is entered in cell E7 and the desired cutoff frequency is entered in cell F7. The x-values are generated in column C by using the formula =ABS(ROWS($A$7:A7)). The y-values are generated in column D using the formula for the elliptical tractrix horn, =B2*(SQRT(B3^2+(C7/B2)^2)-B3)(1+(B4(C7/B2)^2))^0.5. The formula for the x-value of the cutoff frequency is =B2/2/340*F7.

You can copy the formula down in columns D and E to generate additional y-values and x-cutoff values for different x-values. You can then use these values to plot the elliptical tractrix horn shape and to determine the cutoff frequency."

Anything I could do to make this better? Playing around with openAI.

"
In the above Excel spreadsheet, the following variables are used:

  1. a: It is the minor axis of the ellipse which represents the elliptical tractrix horn.
  2. b: It is the major axis of the ellipse which represents the elliptical tractrix horn.
  3. c: It is a constant used in the formula for the elliptical tractrix horn.
  4. x: It is a variable used to represent the x-coordinate of the points on the elliptical tractrix horn.
  5. y: It is a variable used to represent the y-coordinate of the points on the elliptical tractrix horn.
  6. y_exit: It is the y-coordinate of the points on the elliptical tractrix horn scaled by the exit size.
  7. x_cutoff: It is the x-value of the cutoff frequency, which represents the frequency at which the output of the horn begins to decrease significantly.
  8. exit size: It is the diameter of the exit of the horn.
  9. cutoff frequency: It is the frequency below which the output of the horn decreases significantly"
"
You can use the values generated in the Excel spreadsheet above to create a 3-D model of the elliptical tractrix horn. Here are the steps to do this:

  1. Copy the x and y values from the spreadsheet into two separate columns in a new sheet in the same workbook or in a different workbook.
  2. Open a 3-D modeling software such as AutoCAD, SolidWorks, or Blender.
  3. Create a new 3-D model in the software and select the option to import data points.
  4. Select the two columns of x and y values as the data points to be imported.
  5. Use the data points to create a surface or solid model of the elliptical tractrix horn.
  6. Adjust the visual properties of the model as desired, such as color, shading, and lighting.
  7. Save the model for future reference.
By following these steps, you can generate a 3-D model of an elliptical tractrix horn based on the parameters specified in the Excel spreadsheet."
 
Last edited:
  • Like
Reactions: 1 user
This one might be more useful


Here is a sample Microsoft Excel spreadsheet that generates elliptical tractrix horns based on basic parameter inputs:

markdownCopy code
A B C D E F Inputs: 1. Exit size (cm) 2. Cut off (Hz) 3. Mouth size (cm^2) 4. Axial length (cm) 5. a/b ratio 6. Horn Shape 7. 10 20 100 50 2 8. =B7/340 =C7/(2*PI()) =D7 =E7 =F7 9. 0.029411765 0.062831853 100 50 2 10. =8*9 =8*E8 =COS(9*E8) =SIN(9*E8) 11. 2.9411765 3.14159265 0.86602540 0.50000000 12. =(B11^2+1)/F7 =(F7*B11^2+1)^0.5 13.3.872983346 2.236067977 14. =B12^2-1 =F7^2*B12^2-1 15. 14.872983346 14.236067977 16. =15^0.5 =14^0.5 17. 3.828427124 3.807886553 18. =B16/B13 =C16/C13 19. 0.995037190 1.618033988 20. =19^-1 =ATAN(B19/A19) 21. 1.005016708 59.03624054 22. =20*B21 =20*C21 23. 20.10083417 118.07248109 24. =B22+E8 =C23+D8 25. 70.10083417 148.07248109 26. =25/B8 =24/C8 27. 2.061552813 2.938271605 28. =ATAN(26/27) 29. 63.43494882 Horn Shape: 30. =B8*SIN(29) =C8*COS(29) 31. 0.061552813 0.998629535 32. =30+B7 =31+C7 33. 10.06155281 100.99862954 34. =33^2+32^2 35. 1000.99925928 36. =SQRT(35)/2 37. 31.622776601 38. =37^2/B8 39. 0.1 40. =B7*B8/B39 41. 316.22776601 42. =SQRT(B40^2-B7^2) 43. 316.22776601

This spreadsheet calculates the shape of an elliptical tractrix horn based on the input parameters Exit size, Cut off, Mouth size, Axial length, and a/b ratio. The outputs are the x and y coordinates of the horn.


The "a/b ratio" in the context of an elliptical tractrix horn refers to the ratio of the major axis (a) to the minor axis (b) of the elliptical cross-section of the horn. This ratio determines the shape of the horn and affects its performance, such as its directivity and frequency response. A larger a/b ratio results in a more directional horn with a narrower frequency response, while a smaller a/b ratio results in a more omnidirectional horn with a wider frequency response. The specific value of the a/b ratio depends on the desired performance and application of the horn.

Here is a sample Microsoft Excel spreadsheet that generates Iwata horns based on basic parameter inputs:

cssCopy code
A B C D E Inputs: 1. Exit size (cm) 2. Cut off (Hz) 3. Mouth size (cm^2) 4. Axial length (cm) 5. Horn Shape 6. 10 20 100 50 7. =B6/340 =C6/(2*PI()) =D6 =E6 8. 0.0294117650.062831853 100 50 9. =E7/LN(2) =D7/LN(2) 10. 0.062913907 0.067873385 11. =B9*LN(2+2*E7/D7) =C9*LN(2+2*D7/E7) 12. 46.415888336 46.567729468 13. =B11+E7 =C12+D7 14. 56.41588833656.567729468 15. =14/B7 =13/C7 16. 2.828427124 1.931851653 17. =ATAN(15/16) 18. 63.43494882 Horn Shape: 19. =B7*SIN(18) =C7*COS(18) 20. 0.029411765 0.998629535 21. =19+B6 =20+C6 22. 10.02941176 100.99862954 23. =21^2+22^2 24. 1000.99925928 25. =SQRT(24)/2 26. 31.622776601 27. =26^2/B7 28. 0.1 29. =B6*B7/B28 30. 316.22776601 31. =SQRT(B29^2-B6^2)32. 316.22776601

This spreadsheet calculates the shape of an Iwata horn based on the input parameters Exit size, Cut off, Mouth size, and Axial length. The outputs are the x and y coordinates of the horn.
 
Last edited:
Don't mind me....

"An elliptical tractrix horn can exhibit constant directivity behavior over a specific frequency range, but it is not necessarily a constant directivity horn over its entire frequency range.

Constant directivity refers to a type of horn design that produces a consistent and uniform sound field over a specific frequency range, regardless of the direction from which the sound is coming. This is achieved by carefully shaping the horn to produce a consistent and uniform directivity pattern.

The elliptical tractrix horn is a popular horn design that is widely used in loudspeaker systems due to its excellent acoustic performance and efficiency. When properly designed, an elliptical tractrix horn can exhibit constant directivity behavior over a specific frequency range, making it an excellent choice for many applications.

However, the directivity pattern of an elliptical tractrix horn can change with frequency, and it is not necessarily constant over its entire frequency range. The exact frequency range over which the horn exhibits constant directivity will depend on the size, shape, and loading of the horn, as well as the parameters of the associated compression driver.

In summary, an elliptical tractrix horn can exhibit constant directivity behavior over a specific frequency range, but it is not necessarily a constant directivity horn over its entire frequency range"
 
  • Like
Reactions: 1 user
Came up with a name finally :giggle:

Cynosure - cy·no·sure
1675808060191.png
1675842352192.png


Still a ways to go, but I am so grateful to you all, who have helped, and continue to help me and others like me, in the community, create something as unique and desirable as this. This speaker is not what everyone wanted or wants, but its what I wanted and I am pretty sure I did a good job covering almost every aspect to be criticized, with your help. I've also found a flaw in one of my favorite reference songs :unsure: this system is truly vulnerable and that was what I was aiming for. It has the 2way vide that so many people love for the same reasons you all do. I think this could be called a PPSL 2way maybe? PPSL assisted FR?......PPSLAFR, PPSLAWBH for PPSL assisted wide band horn ;)lol! I did put thought into the design. When I first was introduced to the PPSL it was a sort of eureka moment but with no instruction manual. I knew that if I went this route that I should error on the side of accuracy further up the passband, so a smaller CSA slot was chosen. I initially was inspired by a Ripole recipe that asked for a slot width, 1/4 of the woofer diameter. From there I was inspired to increase the CSA just slightly, higher than a 2:1 compression ratio. Point being, I made this PPSL with context of being mated to other sources, not just a subwoofer to be sat in a corner by itself with a lowpass set at 80hz. I played with Hornresp, paying attention to the placement and character of the slot resonance as opposed to where it might be crossed over at, 200hz. I just so happened to overlap where the Axi could go as if I planned to pair them, but I didn't plan this from the beginning. I call it a blessing😇. I always thought I would run the horn to 300hz, the 15" to 300hz and the PPSL to "just under the horn XO, ie ~200hz. I actually used Vituixcad to show how far down to place the XO, to stop the ppsl from augmenting the simulated response, in the speculated horn xo area, and like I said, about 200hz it was! So the PPSl in a nutshell has been designed to be able to cross at 200hz or lower. Some question'd my strange interest in the low Horn Xo...man I am in love! This system is in a way like my nearfield 3 way, where the 2way portion covered above 130hz and the 12", everything below, I ran the 4" from about 130hz-2.5khz and to me it was pretty awesome sounding midrange. Somewhere I came across the term "naked midrange". This has the same great midrange but the directivity takes it to the next level in turn coming from a single point up through treble. The horn is like a lens, and it has higher directivity than the 18"s unless they were arranged differently but non the less, its one source, not 2 combined, and with the low xo its pretty much Fullrange. 200hz is a good XO territory, there's a lot of energy there, in media, what a great place to have all that efficiency, probably a little directivity and 3 large sources coupled because of 1/4WL. That with all the headroom down to 30-20hz. I haven't even auditioned in stereo yet lol and I am very impressed. Funny thing is with the other system, at about arms length, crossing at 200hz was a problem. I wonder what I did wrong? With this new system, 200hz xo point, it doesn't seem to be an issue. If you are familiar, you basically listen to vocals, and you wait for a low note to draw you're attention down, or for yourself to perceive the vocals to come from somewhere else suddenly with the occurrence of a low frequency in the vocal performance. I'm not getting that with this system and when I mute the horn, theres not a lot of vocal energy there to boot. I think the mechanical lowpass of Slot, on top of the 48db LR lowpass, is causing much more of what would of been more noticeable material, to be attenuated. Still, I should be able to break this illusion you'd think, in particular with my close proximity listening distance. Maybe I need to pull up some Barry White and see what that does. I'm excited because, just as is known, when things are simple, they work together to achieve a great result, much easier. So if it sounds this good now, which is subjective, wait till I hear this baby Phase corrected to my listening spot via AudioLense! Thats objective lol. I've only experienced such a dsp corrected playback once, on an older system using the same 12" bass, but with an 8" midwoofer+1" tweeter. I phase aligned everything best I could to my sweet spot and it was another eureka moment ie, time alignment. The improvement was enough to make me a believer, and not only had me relistening to my catalogue but I remember pulling up pop stuff like 90's Madonna and being blown away by the production, even old Brittany Spears, stuff, I'm thinking, I wouldn't normally listen to this, but the increased SQ gave me proper perspective into the sound engineering which really causes the musicianship, to stand out and I was lovin it. So with this System, I should be able to achieve great things if I use it properly. Ty Ty Ty Ty Ty

OpenAi chat is a really good study buddy. Any one who likes to study anything should be using it, its work flow, its efficacy in transferring knowledge and just narrowing in on specialized knowledge is incredible. I've been pushing my kids on it, and I'm pretty sure, AI, is the best professor you've never had, particularly because it never gets irritated and will take all the time in the world to explain and reexplain again, whatever it is you want to know that it knows. I used it to create the Spreadsheet code above...It even allowed me to create a spread sheet that generated tractrix+exponential hybrids, simply because the math is well known and understood.....Not by me...but by it and other mathematically inclined individuals. Soon, anyone will be able to write programs thanks to AI. Above might be the only elliptical tractrix horn spreadsheet On the internet...... Its not perfect, the AI, but its useful.
 
  • Like
Reactions: 1 user
In all honesty, I need to set some type of limiting to protect myself. I am anxious to getting to the point of working on these monitors. Between the monitors and what I've learned towards Sound Quality and Impulse response, I should be able to work efficiently and effectively.

I'm not sure what to do with the 6 15"s I have.... add active cardioid to the project enclosure, multi sub, or cancel wave at rear of room...
 
I think so. The one Elon is co-founder off.

Its not 100% correct on its acoustics, but it can be trained, and it has lot of knowledge already, it knew what a push pull sub was, and an aperiodic bi-chamber. It new what group delay was etc. it seemed to have dispersion incorrect saying a small woofer had more focused sound and a larger, more diffuse, but... maybe it was thinking something different than I was thinking. I told it the nature of directivity vs source size and it apologized for the confusion 🙃
 
I tried my sub in as a "Ripole", and the FR improved significantly....sensitivity also went up...I've used no damping material in the sub enclosures thus far, and the accuracy has been good.

From a theoretical view, where does this put me, in comparison to a sealed cabinet? Is this now 6th order bandpass? One thing I can tell is that the driver is less in control and it made me think of the comments @gedlee made about 6th order enclosures. The directivity cleans up the response and the resulting FR is better than the Sealed in the same position. I have enough headroom still, but the thing that throws me is that, sensitivity went up. I guess I should consider that I probably just got lucky and that within the positions that could be possible in the room, sealed would still give me the more responses that were desireable. Dipole sub at nearfield is a "thing", apperantly, it seems to have bettered things, in my current situation. I just wondering why not more Dipole sub woofers, in studio settings.

I have not experienced bass this accurate, until I removed the rear panel, I need to re-voice the sub but the removal of peaks and nulls, sounds much fuller, as a ripole it has some mode issues, in a different place, but the overall response seems to be more accurate, and like I said, the increase in sensitivity must be a room effect, because in simulation open baffle/ripole would not perform in said way
 
Last edited:
One of the aspects of design, that has got to be over looked, is the low crossed tweeter axis. I think this effect can be achieved through other methods, for example, I don't recall having issue with the 1+4" section of my close proximitized 3 way. Even in that scenario, the 4" crossed at ~130hz, represents a XO point I do not think I see commonly. In a certain light you might say the Whole Vocal Range, needs to be covered within a certain amount of the baffle??? I did not like what I heard when I moved my XO point upwards, It had to do with the physical separation of spectrum that sounded better centralized. I also would of been looking at ~18" worth of separation between tweet and mid. Thinking back to the 1+4, there would of been only about 4" of separation at ~2.5k xo point. In common waveguide+15" mid, there might be 15" of separation at ~700-900hz? I think in an A/B I would chose the lower 200hz XO point, all things else equal. I would say that the matching of polars at XO is a factor, and I don't think there would of been a match for my horn and 15"woofer at every XO point I checked, but I stayed on Axis so that part should matter less. Literally as I moved the XO up, I was not happy, and it wasn't the voicing, I could tell the treble/midbass was more separated than before. Funny enough, theres likely nothing special about what I am saying because for this project, Close monitoring was always the main objective, and of course at such close proximities as 54-25"s the "cracks" in the illusion of the vertical source spread, are bound to show.


My 2way, not, just for its size, but in general, with a 200hz XO, sums really well, I think only a MTM version would be comparable. I sought out, still, to break the illusion. Vocals are the one thing that I seem to be sensitive to. And this guy right here, In particular, his intro, will likely destroy most systems illusion of Vertical Synergy, especially with a little volume

 
1676851297247.png


I seem to keep finding reasons to add active cardioid. It is the most realistic sub bass horn I'll be able to fit in a house. If I add Matching 18"s on the sides with the motor facing out, I can divide the cabinet space between the drivers, This might limit my ability to create a cardioid effect, because of the geometry in the rear of the woofer.

Is anyone familiar with what those limitations might be?
The volume left over create a Qtc of 0.64, Does this matter in the grand scheme of things being that excursion is very low, and the roll off and FR are able to be changed via eq?
 
Last edited by a moderator: