I want to open a discussion about the gain of horn subwoofers compared to other enclosures.
First of all - the frequency range of interest for a SUBwoofer is below 100Hz in my book. You need most acoustical power in the frequency area of the kick drum, that's in most music around 60Hz. You will also need a frequency response below that but with less maximum SPL demand (except some special electronic music).
Many self called horn subwoofers in the PA world don't work as a horn in that frequency range or as a horn at all! You often just have a front resonator or nicely formed reflex port.
So - does anyone here do measurements of a subwoofer in reflex cabinet vs. horn/-like cabinet? A real 1:1 comparison so you can determine the efficiency of the cabinet?
To start there is this measurement from horn guru Ralf Limmer (Germany). I use his 022 and 033 and these are the best horns in this size I ever measured. He really knows about horns ... and has a new bass horn design! https://www.limmerhorns.de/tango/
So I'm confident this is about the maximum you can get with a good bass horn of that size and in this frequency range. And here is the comparison measurement:
Speaker is B&C 18SW115. For the blue reflex housing they used 130L and 34Hz tuning - which is way to low for that volume! I did some simulations with 40-42Hz tuning and estimated the corrected response.
So we gain about 4dB SPL with the horn at 60Hz. That's all?!
5dB at 100Hz (where we already need less max SPL) and up to 9dB way outside of the userange of a normal subwoofer. But we need at least double the volume and get a 57kg speaker. So we could build a double 18" reflex cabinet and get nearly the same sensitivity and higher max SPL cause of more power can be taken.
So - does anybody know about other MEASUREMENTS and comparisons of horns vs. reflex/whatever? I'm first interested in some technical facts, not wander away with sound and personal experiences.
First of all - the frequency range of interest for a SUBwoofer is below 100Hz in my book. You need most acoustical power in the frequency area of the kick drum, that's in most music around 60Hz. You will also need a frequency response below that but with less maximum SPL demand (except some special electronic music).
Many self called horn subwoofers in the PA world don't work as a horn in that frequency range or as a horn at all! You often just have a front resonator or nicely formed reflex port.
So - does anyone here do measurements of a subwoofer in reflex cabinet vs. horn/-like cabinet? A real 1:1 comparison so you can determine the efficiency of the cabinet?
To start there is this measurement from horn guru Ralf Limmer (Germany). I use his 022 and 033 and these are the best horns in this size I ever measured. He really knows about horns ... and has a new bass horn design! https://www.limmerhorns.de/tango/
So I'm confident this is about the maximum you can get with a good bass horn of that size and in this frequency range. And here is the comparison measurement:
Speaker is B&C 18SW115. For the blue reflex housing they used 130L and 34Hz tuning - which is way to low for that volume! I did some simulations with 40-42Hz tuning and estimated the corrected response.
So we gain about 4dB SPL with the horn at 60Hz. That's all?!
5dB at 100Hz (where we already need less max SPL) and up to 9dB way outside of the userange of a normal subwoofer. But we need at least double the volume and get a 57kg speaker. So we could build a double 18" reflex cabinet and get nearly the same sensitivity and higher max SPL cause of more power can be taken.
So - does anybody know about other MEASUREMENTS and comparisons of horns vs. reflex/whatever? I'm first interested in some technical facts, not wander away with sound and personal experiences.
Addendum: I know the horn area is not big for 60Hz (2,7m circumference vs 5,7m wave length) and it probably gains 1-2dB (?) sensitivity in a stack of 4 or 8. But I didn't found measurements about that. And we are still in the area of a double 18" reflex.
Maybe someone can help out with some mesurements to show the stacking effect?
And that's not at all a rant against horn subwoofers at all! And this is for sure a great design!
There is a significant gain in sensitivity and when you like horn woofers this is for sure a great design!
I personally just expected more sensitivity gain from such a design.
Maybe someone can help out with some mesurements to show the stacking effect?
And that's not at all a rant against horn subwoofers at all! And this is for sure a great design!
There is a significant gain in sensitivity and when you like horn woofers this is for sure a great design!
I personally just expected more sensitivity gain from such a design.
Problem with a bass horn is that it has to be gigantic in size to get very low.
I compared the B&C18SW115-4 in a bass reflex to a tapped horn, both with an Fb around 37Hz, and more similar size between the two.
https://www.diyaudio.com/community/threads/tapped-horn-vs-bass-reflex-case-study.184992/
Tapped horns have a low frequency advantage over FLH horns of similar size.
Nice execution, but nothing new in the design that I can detect.
The 2x18" would gain +3dB sensitivity in the upper range, putting it in the 98dB sensitivity range ~100 Hz.
If it's Fb was the same as the Tango FLH Fc, it would have more low frequency (if we call 50Hz LF..) potential, but the Tango would have a bit more (+2dB) upper output potential.
The Tango, like almost all FLH designed for PA use has a vastly undersized mouth, and a path length of only 1/4 wavelength.
In multiples, the FLH LF may increase slightly more than the usual +6dB per doubling of enclosures and power, the LF would "flatten out" to Fc as the mouth diameter approaches the wavelength of it's Fc.
That said, as can be seen in this overlay of two FLH cabinets measured outside at 10meters compared to a single, you would need a lot of cabinets to see any difference
In this case, there is a reduction of upper frequencies compared to lower, no more than +6dB gain.
At 30Hz, there is ~+7dB gain, one dB increase, though the frontal area has doubled.
I've measured more increase doubling frontal area with sheets of plywood.
Art
I have made a lot of these measurements...bass-reflex and FLH...using 5 different double 18" bass reflex builds, and two different FLH's.
The FLH's were JTR Orbitshifters (OS), and Danley's Labhorn design. The OS uses a single 18"; the Labhorns dual 12".
Bass reflex all used the same 8 ohm 18" BMS 18n862's.
First build was 2 single boxes made for stacking, f-3@ 31Hz.
Next was a dual, PPSL, same f-3 @ 31Hz.
Then came a large normal style, dual 18", f-3 @ 26Hz.
And current, a dual, angled PPSL, f-3 @25Hz.
I only measure and use subs for < 100Hz. I measure sensitivity integrated over the full operating range, not by eyeballing the response curve.
My double 18"s tuned for 31 Hz, have 2.83V @ 1m = 101.6dB.
The lower tuned double 18"s, have 2.83V at 100.6dB.
4 ohm nominal loads, processing in place.
A single OS has a sensitivity of 103.3dB. It's single 18" driver is 4 ohm nominal.
Two OS, 109.5dB. That all i have of the. OS
A single Labhorn has 106.6 dB. It's two 12"s are 6 ohm nominal in parallel.
Two Labs 112.6 dB.
Four Labs 118.4 dB.
I'd put both of the FLH's f-3 in the mid 30 Hz range.
The sensitivity measurements for them did not include processing for flat response. They just had 100Hz lpf, and 24Hz hpf in place.
Way below 100Hz is probably more accurate, probably a subwoofer isn't one unless it goes well below 50Hz.
There's some music with strong bass notes around 40Hz, coming from a standard bass guitar. Then a 5-string electric bass goes as low as 30Hz or so, assuming its supporting speaker cabinet is designed to do so.
Popular electronic music gets down to around 20Hz or so without much trouble. That's true subwoofer territory. See the example below.
The 34Hz tuning is not entirely unreasonable if they were aiming to achieve good bass extension; it does work quite well. The VituixCAD simulation of this particular combination is shown below. This produces a shelved response that has a −3dB point of F3=40.6Hz, placing it right in bass guitar territory.
But if we now tune to Fb=41.4Hz, we get an even lower F3=40.0Hz, while increasing the output above F3 by about 2dB. That's a very worthwhile gain for very little extra work. The benefit of the previous enclosure tuning was that F6 was a few hertz lower.
It doesn't seem all that worthwhile for the added complexity and the peakiness of the response right where we want it to be flat. The horn-loaded cabinets may look a bit more impressively businesslike to passers by.
Size is too small. Scott Hinson's sub of doom is larger but even at that size he says it's still not enough
A question i was going to ask: a bass horn? Or a double tapped horn? Or an ML-Voigt (singke tapped)?
dave
Well, let's see what we can do about that, estimating what we can obtain as far as extra low-frequency extension goes when we increase the enclosure volume significantly.
Below is the VituixCAD simulation for a vented enclosure with Vb = 600 litres and tuned to Fb = 23.6Hz. Here we find that F3 has been significantly lowered to F3 = 22.8Hz, with only a very small ripple in the passband. I wonder if anyone has built such an enclosure?
Not familiar with VituixCAD. Do they also simulate horn response?
This is the hornresp prediction and measurement normalized to 1m in Scott's paper.
The outer dimension is 48" x 48" x [23+(59/64)]"
This is the hornresp prediction and measurement normalized to 1m in Scott's paper.
The outer dimension is 48" x 48" x [23+(59/64)]"
As far as I know, VituixCAD doesn't simulate horn-loaded loudspeaker systems.
The horn-loaded woofer that is described in the paper seems to have more limited bass extension than a vented enclosure. The design tradeoff seems to be higher output, lumpy as it is, versus a significantly lower F3 from the vented enclosure. It seems a poor tradeoff considering the added complexity of the horn enclosure.
If we increase the enclosure size to mimic that of 1/2 the horn-loaded one, which comes to about 400 litres, we get an F3=25.6Hz. The horn appears to be nowhere near that.
The horn-loaded woofer that is described in the paper seems to have more limited bass extension than a vented enclosure. The design tradeoff seems to be higher output, lumpy as it is, versus a significantly lower F3 from the vented enclosure. It seems a poor tradeoff considering the added complexity of the horn enclosure.
If we increase the enclosure size to mimic that of 1/2 the horn-loaded one, which comes to about 400 litres, we get an F3=25.6Hz. The horn appears to be nowhere near that.
Last edited:
So I don't know the point using VituixCAD in horn loaded design here.
The Neman Horn is aiming for maximum SPL.
A vented enclosure won't have that output level.
If maximum SPL isn't the point of this thread why go for horn loaded?
The Neman Horn is aiming for maximum SPL.
A vented enclosure won't have that output level.
If maximum SPL isn't the point of this thread why go for horn loaded?
If we opt for a 400-litre vented enclosure for the SB Acoustics ROSSO-18SW1000D 18-inch woofer, and use a filter-assisted low-frequency alignment, we can achieve the response shown below. The low-frequency cut-off frequency is now lowered to F3=21.8Hz, and the 2nd-order peaking high-pass filter adds a modest 4dB of boost at about 28.5Hz. This system hits Xmax limits with 300W input re 8 ohms. The horn-loaded system seems to be well away from that sort of low-frequency performance.
Talk to a PA guy to trade max SPL of 21Hz for 40+ Hz SPL and see their opinion.
The Neman horn do have ports. But it's for cooling purpose only.
The Neman horn do have ports. But it's for cooling purpose only.
The vented enclosure could be tuned to have a whopping great peak in its response. Would that be a satisfactory solution if maximum SPL is being aimed for?
It's not entirely clear, as the horn enclosure has a significantly peaked response. It's less than ideal and presents challenges in its use, but I suppose it could be EQed. Solving the non-flat response problem seems to be a significant impediment.
The vented-box enclosures described above based on the SB Acoustics ROSSO-18SW1000D 18-inch woofer only had one woofer. Two boxes side by side, with the same combined volume as the Neman horn, would have at least 6dB more output. And their mutual coupling as acoustic radiators would add a little bit of extra oomph too. So, if maximum SPL across an extended low-frequency range is important, then an appropriate vented-box system seems a reasonable choice.
Vented has it's benefit but if size isn't a issue front load horn is the way to go.
I'll wait what OP is aiming for.
Scott's design has licensed to a small PA manufacturer so I think it's at least been recognized commercially.
I'll wait what OP is aiming for.
Some bandpass sub also tuned this to have high SPL at a frequency range. I'm actually interested to see your port design to have 130+dB SPL.
Scott's design has licensed to a small PA manufacturer so I think it's at least been recognized commercially.
Look at Powersoft's EDM Sub or the Cine Sub, both have the M-Force moving magnet transducer. The EDM Sub has ( 2 ) of the 30" M-Force 301P02 transducers.
If you're not fond of the bandpass EDM sub, the Cine Sub is fairly flat from 18hz to 100 hz, and capable of 142dB half space at 1m.
If you're not fond of the bandpass EDM sub, the Cine Sub is fairly flat from 18hz to 100 hz, and capable of 142dB half space at 1m.
Any horn built specifically for low bass reproduction will need to be large to accommodate the long wavelengths of bass notes.
Last edited:
reproduction?
The type of box will affect sensitivity. A real horn should be more ensitive ythan a double tapped quarterwave pipe.
dave