Fluid inertia in tubes scales linearly with length, but inversely with the square of the diameter. If you double the length and diameter of a tube, you get half the inertia. I'd expect loading from the throat extensions to get worse if you just scaled up one of the models.
Direct scaling isn’t an option because the throat needs to stay the same. Otherwise this would be too easy to do. The problem is that yet I haven’t figured out how to draw my own version and it would always be nice if you don’t have to do the same that someone else has already done. But it would be nice to learn new things too so maybe I’ll just go to my room and have “my own time” without family 😀
Here's some BEM data. I took A460G2 and changed just the R to 260 and 360 mm (i.e. mouth diameters 46 / 52 / 72 cm).
Throat impedances with the 25-STD-1 adapter:
Polar maps for mouths 46 and 72 cm (with a -6dB marker line):
I would be also interested how to calculate an on-axis SPL gain from this. Maybe it's not so easy after all, I don't know.
Throat impedances with the 25-STD-1 adapter:
Polar maps for mouths 46 and 72 cm (with a -6dB marker line):
I would be also interested how to calculate an on-axis SPL gain from this. Maybe it's not so easy after all, I don't know.
I am playing around with source definition script to make a LF-horn matching the A460G2. So far successful to scan the real driver, extract the geometry and build the source contour. But I have the feeling something is wrong as I cannot get any beaming in lower frequencies. It might be the source contour itself and all true.....or....there is still a mistake. I am worried about those lines forming a plane right at start of the horn. They truncate the source contour. What is this (marked in blue)?
I guess it depends on your expectations 🙂
- The lines you see are probably the back side of the horn shell as seen from the front. You can try to move these (two?) nodes further back manually in nodes.txt, but I don't think this should be an issue. (You can display individual nodes in ABEC Drawing - "Views - Nodes", and turn off "Symmetries".)
- The lines you see are probably the back side of the horn shell as seen from the front. You can try to move these (two?) nodes further back manually in nodes.txt, but I don't think this should be an issue. (You can display individual nodes in ABEC Drawing - "Views - Nodes", and turn off "Symmetries".)
For my eyes that directivity looks quite nice. If there’s still some directivity left at 300Hz it should increase gain too. Is it enough is another question of course
Of course it will. How much, I can't say. My guess would be "not much" 🙂
Yeah, as one sees on the 72cm device sound at 300Hz to 180deg is about as loud as in front, which means it's about 2pi radiation. Increased directivity, SPL to 120deg (90-180) is reduced due to sound doesn't come to that direction just from the closest side, the shortest path aka direct sound, but all around the device, with extra path length making destructive interference. This makes the characteristic "ridges" to polar map that start from lower frequency and angle, and extend toward higher frequency and angle, due to wavelength getting shorter while the object size stays static of course 🙂
Last edited:
@mabat I did get the A460G2 Kit and was wondering what the parameters arcterm and trunc do? I have the published R-OSSE formulas set up in solidworks. Unfortunately I can not load and compare the stl files with the curve in solidworks as perfomance is very limited with big meshed files.
If possible could you also provide step files as they do not represent a triangulation working with them is much nicer.
If possible could you also provide step files as they do not represent a triangulation working with them is much nicer.
- arcterm finds the highest curvature along the profile and continues with that radius until the given angle
- trunc cuts off a throat part of the specified length
Regarding the STEP file, I'll see what I can do.
Last edited:
Is there any guide how to draw profile by excel or how to use Akabak with this horn profile? I tried to make csv file but I cannot get it to work and I'm loosing my mind 😀 Chatgpt doesn't help enough
BTW, these are the throat impedances of the 46cm and 72cm Gen2 horns with the 4554-EXT adapter (17mm throat) -
Here's a considerable "lift" centered around 700 Hz - this is the kind of increase of the radiation resistance to actually make a difference.
Otherwise the two impedances are virtually the same.
Here's a considerable "lift" centered around 700 Hz - this is the kind of increase of the radiation resistance to actually make a difference.
Otherwise the two impedances are virtually the same.
I'm finishing a "Tritonia-S" kit, to be built in an enclosure, or even perhaps as free standing -
1" throat, front dimensions 206 x 178 mm (Tritonia-"S" for small).
1" throat, front dimensions 206 x 178 mm (Tritonia-"S" for small).
Last edited:
JBL2482s (Phenolic dia) on a 200Hz cutoff horn will reach down to 350Hz.
On a 150 horn you'd see 300.
They're big though!
I used the 2482s because I loved the tone - didn't play them very high, 1000-1200Hz or so..
Might I suggest a flange around the perimeter to make it easier to flush mount in a rebate on the front panel?
I'm trying to figure out what I'm doing wrong. I installed ath software and copied ath script from r-osse pdf to separate r-osse.ath file without any changes. For some reason I'm gettin warning messages like this. What would be wrong?:
Info : Reading 'mesh.geo'...
Warning : Start point 2 and end point 10 of GEO line 3 are closer than the geometrical tolerance, at position (0, 0, 0)
Info : Reading 'mesh.geo'...
Warning : Start point 2 and end point 10 of GEO line 3 are closer than the geometrical tolerance, at position (0, 0, 0)
GEO files are generated only for 3D models. The example from the R-OSSE pdf (p.28) uses circular symmetry - it thus can't be as you say. If something is wrong, I have no idea what it can be.
Make sure you keep 'Output.STL = 0', as it often doesn't work properly.
Make sure you keep 'Output.STL = 0', as it often doesn't work properly.