I feel like I know enough about audio to be wrong, but not enough to be right, so thanks in advance for your patience.
I've heard that, if you're willing to sacrifice efficiency, you can more-or-less get as much low extension as you want without resorting to a huge enclosure. Is that accurate? I'm assuming it's incomplete if not outright inaccurate, but I want to understand better.
I'm helping a friend get a live events biz off the ground. Right now we're using a pair of Mackie SR18Ss. They are roughly 2' x 2' x 2' each and give us enough overall output. However, as with most PA subs, nobody's home below 40 Hz without DSP. I'd love it if we could go down below 20 Hz with little or no DSP/EQ, but as I'm sure you all know, there's nothing off-the-shelf in pro audio that plays anywhere near that low unless it's several times the size, and we can't (yet) tote around something that big.
Let's pretend I have access to all the amplification I need. Sky's the limit there. Given that, could I conceivably build a sub not much larger than 2' x 2' x 2' with relatively flat frequency response from like 18-100 Hz without DSP and as much SPL capability as one of our Mackies?
If so, what should I look for in a driver and enclosure design, and what other downsides should I expect it to have (distortion, transient response etc.), if any?
If not, how can I tell how big the enclosure would have to be – again, assuming I can compensate for any lack of efficiency with a boatload of amplification?
I've heard that, if you're willing to sacrifice efficiency, you can more-or-less get as much low extension as you want without resorting to a huge enclosure. Is that accurate? I'm assuming it's incomplete if not outright inaccurate, but I want to understand better.
I'm helping a friend get a live events biz off the ground. Right now we're using a pair of Mackie SR18Ss. They are roughly 2' x 2' x 2' each and give us enough overall output. However, as with most PA subs, nobody's home below 40 Hz without DSP. I'd love it if we could go down below 20 Hz with little or no DSP/EQ, but as I'm sure you all know, there's nothing off-the-shelf in pro audio that plays anywhere near that low unless it's several times the size, and we can't (yet) tote around something that big.
Let's pretend I have access to all the amplification I need. Sky's the limit there. Given that, could I conceivably build a sub not much larger than 2' x 2' x 2' with relatively flat frequency response from like 18-100 Hz without DSP and as much SPL capability as one of our Mackies?
If so, what should I look for in a driver and enclosure design, and what other downsides should I expect it to have (distortion, transient response etc.), if any?
If not, how can I tell how big the enclosure would have to be – again, assuming I can compensate for any lack of efficiency with a boatload of amplification?
Last edited:
There is zero need for 18 hz at a live event. It would be a non-event. If you review the Fletcher-Munson, later Robinson-Dadson, and finally the ISO 226:2003 equal loudness contours, it may help clue you in.
Targeting 100dB outdoors, you would need at least ~130dB capability at 18 hz to sound as loud as 100hz.
Solving for # of drivers for 130dB, 18 hz @ 1m with 25mm excursion drivers, you would need more than 13 in sealed boxes, and that is with NO headroom accounted for. With 10dB headroom? 42 + 18" drivers, moving 25 mm one way, or 50 mm p-p.
https://www.baudline.com/erik/bass/xmaxer.html
Power compression from sealed boxes would be a non-starter, they'd melt in no time given the amount of power you'd need to hit those levels. Large, low tuned ported enclosures with something like a Sundown Audio 18" driver could work, but you would need several, with many large power amplifiers, and a large enough service to run them.
In other words, it's not feasable. I doubt a large 5 ton box truck would be enough to transport just the ~18hz tuned subwoofer boxes to legitimately produce a level equal to 100dB at 100 hz, not counting the amount of amplification for them.
If you want really low playing PA-ish gear, have a looka at Danley. https://www.danleysoundlabs.com/
They do some of the most compact HT like, deep reaching sub systems. If you think they are large, you are wrong, for what they do they are surprisingly small.
I'm just asking my self what you want to achieve, so much acoustical power has to go somewhere, the whole place would resonate, rattle and people get sick if you go that loud, that low.
What kind of music do you think would need that? Is it a closed room or open space?
Maybe ask someone who knows about such things?
Your pair of Mackie's is something people have at home in a living room, so no wonder you do not get much low end out of them. You got to have a healty relationship between chassis area, room size and SPL. Anyway, if you want low and deep, you need volume, power and the right, efficient construction principle. For permanent installations horn constructions like the ones Danley does, are the best you can do.
They do some of the most compact HT like, deep reaching sub systems. If you think they are large, you are wrong, for what they do they are surprisingly small.
I'm just asking my self what you want to achieve, so much acoustical power has to go somewhere, the whole place would resonate, rattle and people get sick if you go that loud, that low.
What kind of music do you think would need that? Is it a closed room or open space?
Maybe ask someone who knows about such things?
Your pair of Mackie's is something people have at home in a living room, so no wonder you do not get much low end out of them. You got to have a healty relationship between chassis area, room size and SPL. Anyway, if you want low and deep, you need volume, power and the right, efficient construction principle. For permanent installations horn constructions like the ones Danley does, are the best you can do.
Last edited:
Pretty much. Hofmann's Iron Law is what it's typically called.
https://en.wikipedia.org/wiki/Josef_Anton_Hofmann
"Hofmann theorized that when woofers are mounted in speaker enclosures, the designer would have to accept that there are three trade-offs. Hofmann argued that the designer had "...three parameters that cannot all be had at the same time. They are low-bass reproduction, small (enclosure) size, and high (output) sensitivity." Hofmann stated that designers could pick two of these three parameters, but in doing so, it would compromise the third parameter."
-----------------
Without DSP, it's tougher these days, since that's what most people resort to to get very low in a small box. You would likely need a custom driver if you want a lot of output or extreme power handling. That may be possible with something like the Digital Designs Z series. As stated above, you still have displacement limited output issues at very low frequencies though, since each octave you go down requires 4x the volume displacement to make the same SPL.
If you just want more extension in about the same volume as you have currently, that can be had with something like the Eminence LAB15, but according to the sheet below it may have less output (assuming Mackie's claim of "Maximum SPL Peak 133 dB" is equivalently measured, which I'm unsure of).
www.parts-express.com/Eminence-LAB-15-15-Pro-Subwoofer-290-575?quantity=1
www.parts-express.com/pedocs/more-info/290-575--eminence-lab-15-cabinet-design.pdf
Last edited:
I've done exactly this: https://www.diyaudio.com/community/threads/fun-in-a-theatre-3x-faital-18xl1800.386725/#post-7034321
Step 1 - pick a modern, high-end 18" PA driver with a lot of Xmax
Step 2 - Try to design a compact cabinet that will maximise output in a compact space
Step 3 - Realise that the port airspeeds are all way too high. You have to choose between a medium-sized ported cabinet (200L net) with lots of output, or a compact sealed cabinet (70L) which means you can EQ the LF response however you like, but will leave some output on the table
Step 4 - Build the sealed box, since you wanted something compact anyway
Step 5 - Drop 3KW into each box
Step 6 - Realise that a ported box with the same driver goes much much louder, or you could have built something smaller and match it in a useful frequency range: https://www.diyaudio.com/community/threads/chasing-output-density-120db-40hz-from-28l-net.416562/
Chris
Step 1 - pick a modern, high-end 18" PA driver with a lot of Xmax
Step 2 - Try to design a compact cabinet that will maximise output in a compact space
Step 3 - Realise that the port airspeeds are all way too high. You have to choose between a medium-sized ported cabinet (200L net) with lots of output, or a compact sealed cabinet (70L) which means you can EQ the LF response however you like, but will leave some output on the table
Step 4 - Build the sealed box, since you wanted something compact anyway
Step 5 - Drop 3KW into each box
Step 6 - Realise that a ported box with the same driver goes much much louder, or you could have built something smaller and match it in a useful frequency range: https://www.diyaudio.com/community/threads/chasing-output-density-120db-40hz-from-28l-net.416562/
Chris
The Mackie (on paper) is rated for 133dB peak, which would be a 130dB RMS rating at one meter.
It's peak output could be achieved only in the upper range of it's pass-band.
It is conceivable to fit 5x24" drivers in a 2' x 2' x 2' box, but 8 cubic feet is too small to be efficient down low.
5x24" drivers with 30mm one way excursion could achieve 130dB at 20Hz.
30mm Xmax is a big ask for a 24" driver, and in the real world the depth of such a driver may preclude fitting 5x24" in a 2' x 2' x 2' box.
Around 50,000 watts would be required, ~98% of that power would be turned to heat, creating an 8 cubic foot oven.
The drivers would require an active cooling system to remove that heat to sustain 130dB SPL at 20Hz for more than a few seconds, or the voice coils and cones would burn.
Your chosen cabinet size and output dictates a sealed enclosure, which would be best in terms of transient response, but requires the most excursion, which is expensive if low distortion is required.
Excursion is key to output, with the power handling to provide it. Each doubling of excursion allows +6dB more output, but requires more than 6dB (four times) the power.
Using real-world drivers and box designs Josh Ricci's Data-Bass site can answer that question:
https://data-bass.com/systems?sort=20Hz:1,mfr:-1,name:-1
The above results are at 2meters, so +6dB can be added for a 1m equivalent.
The VT-218 is slightly more than double your size requirement, ~130dB@20Hz/1m, using ~18,000 watts of power at ~28%THD.
Anyway, low, loud and small are achievable, but not cheaply😉
Art
Guys, sincerest thanks for the thorough and detailed reality check. I really appreciate the time and thought you all put into these posts. I knew I had to be missing something, but it's always a pleasant surprise when what I was missing is as hilarious as a wall of drivers turning the venue into a sauna and then melting. 🤣
Almost all our events have been indoors in smallish venues – dozens of people, often many dozens, but not hundreds or thousands. That's why those two Mackies have been enough so far, volume-wise. Apologies for not making that clear.
The music we play doesn't have a ton of content below 40 Hz, but a lot of tracks do sound better on a system that can play lower. One of our DJs recently bought an SVS SB-1000 Pro for his home setup; I set its crossover in the 40s (matching the natural roll-off in his main speakers), added a few dB in the low-20s, and kept the sub's volume moderate, and we all loved what that did for the sound. It'd be nice if we could bring a similar touch to our events.
I have nothing against big cabinets per se. Personally, I'd be just as happy with bassbins the size of Texas. The only reason for the size consideration is portability. I'd rather not have to add another vehicle to our logistics train, make people have to lug massive cabinets, or risk not fitting through our venues' doors.
I don't think we've ever maxed out our current speakers except at outdoor events, for which I'm not trying to optimize. I'm pretty sure we've never pulled more than 2 kW for the whole system. So, I guess there isn't really a need to match our subs' current max output two octaves down. However, if I'm getting an accurate sense from the sheer insanity of the numbers you guys have calculated, that extra wiggle room would at best move the original idea from preposterous to merely infeasible. Time for a rethink!
Happy New Year to you all!
Almost all our events have been indoors in smallish venues – dozens of people, often many dozens, but not hundreds or thousands. That's why those two Mackies have been enough so far, volume-wise. Apologies for not making that clear.
The music we play doesn't have a ton of content below 40 Hz, but a lot of tracks do sound better on a system that can play lower. One of our DJs recently bought an SVS SB-1000 Pro for his home setup; I set its crossover in the 40s (matching the natural roll-off in his main speakers), added a few dB in the low-20s, and kept the sub's volume moderate, and we all loved what that did for the sound. It'd be nice if we could bring a similar touch to our events.
I have nothing against big cabinets per se. Personally, I'd be just as happy with bassbins the size of Texas. The only reason for the size consideration is portability. I'd rather not have to add another vehicle to our logistics train, make people have to lug massive cabinets, or risk not fitting through our venues' doors.
I don't think we've ever maxed out our current speakers except at outdoor events, for which I'm not trying to optimize. I'm pretty sure we've never pulled more than 2 kW for the whole system. So, I guess there isn't really a need to match our subs' current max output two octaves down. However, if I'm getting an accurate sense from the sheer insanity of the numbers you guys have calculated, that extra wiggle room would at best move the original idea from preposterous to merely infeasible. Time for a rethink!
Happy New Year to you all!
IME, a system that's flat to 40Hz is more impressive than most portable PA systems. A lot of stuff will peak at 60-70Hz and be 6-10dB down in the mid-40s.
Going deeper into the bass is something that requires solid justification. For every octave you want to go down, you need 4x the displacement. That typically means four times as many drivers.
ie, if your current pair of Mackies are, in fact, flat down to 40Hz (I suspect they aren't), then you'd be looking at 8x 18" subwoofers to match them for SPL at 20Hz. You could get that down a bit by using better drivers etc, but there's no replacement for displacement.
Finally, it's worth mentioning that the ergonomics of the subwoofer will dictate portability. I find it easier to move a 2x18" sub, rather than a 1x18". The former will typically have wheels, or if not, it's easy to put on a hand-truck and wheel it around.
Chris
Going deeper into the bass is something that requires solid justification. For every octave you want to go down, you need 4x the displacement. That typically means four times as many drivers.
ie, if your current pair of Mackies are, in fact, flat down to 40Hz (I suspect they aren't), then you'd be looking at 8x 18" subwoofers to match them for SPL at 20Hz. You could get that down a bit by using better drivers etc, but there's no replacement for displacement.
Finally, it's worth mentioning that the ergonomics of the subwoofer will dictate portability. I find it easier to move a 2x18" sub, rather than a 1x18". The former will typically have wheels, or if not, it's easy to put on a hand-truck and wheel it around.
Chris
I've been trying different ways to crack the subwoofer with pro audio driver puzzle for my own project. I've looked extensively for a pro audio sub that can extend down to 50hz in a sealed enclosure before rolling off.
My conclusion is that the only drivers that can hit this target are high excursion subwoofer drivers like what is intended for home theater and car audio. As pro audio drivers extend their low frequency performance, their sensitivity decreases and they look more and more like home theater sub drivers!
I'm using Lab 12 drivers--they have an F3 of ~45hz, up to 120hz in a small sealed box. They are a bit of a unicorn. Larger Lab drivers don't go as low, and no other pro "subwoofer" that I have found specs for can extend as low before rolling off!
Have you considered something like a Lab Horn? It will be bigger than what you have in mind, but might fit the bill.
Otherwise you just need some big high excursion subwoofer drivers, like Dayton's 18" UM II drivers. They have an F3 of 35hz in 4 cubic feet, and handle 1,200watts rms. Sensitivity is low for a pro audio driver at 90.7 dB, but it isn't bad for a subwoofer.
(For real bass, you need some of these! Dave Rat's Super Subs. I understand they are optimized for down to 30hz, but they can be used down to ~20hz. Needless to say they are not your average subwoofer!)
You might find it interesting that a single SVS SB-1000 Pro hits 97dB SPL@20Hz 1m:
The Mackie SR18S has ~-35dB less output at 20Hz than at 40Hz, so if we assumed it could peak at 133 dB at 40Hz (probably 3 to 6dB optimistic..), it's 20Hz output would be at best +1dB more than the SVS SB-1000 Pro.
I'd bet the single SB-1000Pro would be closer to the output of two SR18S at 20Hz.
Note that the 7.8 cubic foot SVS PB13-Ultra, similar size to the Mackie SR18S, has ~+22.6dB more output at 20Hz than the SB-1000Pro .
To put that in perspective, around thirteen (+22.3dB) SR18S would be required to equal the 20Hz response of a single PB13.
Think about that!
Art
The specifications for the Mackie SR18S includes a frequency response curve. This seems to show that the system in "Flat" mode has a −3dB point of around 38Hz, being around −1.5dB at 40Hz. That's quite a respectable level of performance, and handles the lowest note on a 4-string bass guitar with high fidelity (no significant attenuation).
However, below 38Hz we have a very steep roll-off in the low frequencies. This looks like it's a 6th-order rolloff, which implies that this is a 6th-order filter-assisted vented box low-frequency alignment. Hence the compact enclosure size.
That's a very, very big ask, as the implied assumption is PA-type levels. Keep in mind that for 1x 18-inch driver in an infinite baffle or sealed enclosure, it takes an Xmax of about 26mm to produce 130dB SPL @ 1m at a 64Hz, which is 2/3 of an octave higher than 40Hz. That's a great deal of excursion.
Keeping that excursion level constant, with 26mm of excursion and 1x 18-inch driver, by 20Hz we would be producing only 110dB SPL @ 1m. That's 20dB less!
The physics of the situation controls the possible design solutions. Getting a −3dB point of around 20Hz is possible, but it generally requires some quite large enclosures.
The gross enclosure volume of the Mackie SR18S enclosure is 237litres, and the net volume might be around 180–200litres. That's moderately large for a vented enclosure and so should potentially provide some scope for a more extended frequency response in a similarly-sized enclosure using a suitable driver.
Looking at a FaitalPRO 18XL1800 in a 190-liter enclosure, we can get the following results. Note that this is a filter-assisted alignment, which protects the driver below the enclosure's cutoff frequency (a necessity in my view). The filter is a 2nd-order high-pass peaking filter set to 22Hz and Q=1.70, giving about +5dB of boost. With 400W re 8ohms of input power, we reach 120dB SPL, and the peak power required is about 1,400W at 22Hz (due to the peaking equalization), and is about the maximum that this driver can absorb.
If we add a 6th-order Butterworth low-pass filter set to 38Hz to the above response, we get the following result. A pair of such infra-subwoofers could then be added to the existing pair of Mackie SR18S enclosures to boost their bottom end. Although you have 4 enclosures instead of two, you do have almost an octave more low-frequency extension. If the infra-subwoofers have problems, the main subwoofers will still carry on.
Below is a simulation of what can be achieved by combining the infra-subwoofer with the subwoofer. Both are connected with positive polarity, which produces the broad +3dB peak centered on 38Hz. This may be a help or a hindrance, as it will certainly add a bit of bass emphasis to the system, which may need EQing down (but that's easily achieved at the mixing console).
Alternatively, we can switch from a 6th-order to a 4th-order Butterworth low-pass filter on the 18XL1800, and we get a much flatter summed response.
There is a little bit of phase cancellation occurring in the 50–90Hz frequency range, which could be reduced by lowering the cut-off frequency on the 4th-order low-pass filter by a few hertz and adjusting the level on the infra-subwoofer to get a good response. See the example below.
Maxing out driver excursion will of course produce distortion. The high SPLs that are being generated will lead to port chuffing, but that may not be so noticeable amongst the music signal.
The 200-litre enclosure size is probably right on the limit for getting reasonably extended bass at reasonable output levels from a single driver. The larger the enclosure the better when playing that particular game.
Last edited:
For your concept to work in practice, you need a DSP based, dynamic limiter, that prevents the infra (and sub) from getting too much power when their long therm permissable load is reached. These limiters have to be locked, so no DJ or renter in the field can change them. This breed always knows better than the sound man or doesn't care for the gear...
Such a limiter should be set once, through distortion measurements, with the amp that drives them. It is softly cuting the input signal down when some level of distortion is reached. This does not only make them save to handle, but improves sound, too. From some point only the low-mid-high get's louder, the infra stays at a constant level.
Many amps have such a dynamic limiter, even the cheap Behringer D-DSP amps for example. Build in, set once, locked. Very handy. Many active consumer and monitor speaker have such limiter today, which prevents the small woofer from too much excursion, but gives the impression of very deep bass at just a little reduced level.
The down side is that you do not realize when the infra driver is at its limit and may stay there the whole evening. Do not expect even an expensive pro driver to last forever, if permanently driven at x-max. They wear out like any PA driver does, this may just come much sooner. The human ear is quite tollerant to distortion in this low range, so there may no be any audible warning when the drivers are suffering.
So in the end you do not get 130dB at 20 Hz, but an subjectively extended response deep down.
Such a concept will only work if done right, which needs time, experience, some measuring gear and, most important, insight of the owner. Last point often is a problem, when he wants an override option. All this settings have to be done before they are used for the first time...
Such a limiter should be set once, through distortion measurements, with the amp that drives them. It is softly cuting the input signal down when some level of distortion is reached. This does not only make them save to handle, but improves sound, too. From some point only the low-mid-high get's louder, the infra stays at a constant level.
Many amps have such a dynamic limiter, even the cheap Behringer D-DSP amps for example. Build in, set once, locked. Very handy. Many active consumer and monitor speaker have such limiter today, which prevents the small woofer from too much excursion, but gives the impression of very deep bass at just a little reduced level.
The down side is that you do not realize when the infra driver is at its limit and may stay there the whole evening. Do not expect even an expensive pro driver to last forever, if permanently driven at x-max. They wear out like any PA driver does, this may just come much sooner. The human ear is quite tollerant to distortion in this low range, so there may no be any audible warning when the drivers are suffering.
So in the end you do not get 130dB at 20 Hz, but an subjectively extended response deep down.
Such a concept will only work if done right, which needs time, experience, some measuring gear and, most important, insight of the owner. Last point often is a problem, when he wants an override option. All this settings have to be done before they are used for the first time...
Very optimistic, Art. To hit 133dB at 100Hz, we're talking a smidge over 15mm of excursion one-way.
If you have time & inclination, here's an article I wrote back in 2017 on the subject: https://www.prosoundweb.com/spec-wars-looking-inside-loudspeaker-spl-specifications/
Having spend some time abusing that particular driver, I can say it'll take a Powersoft T602, bridged, all the way up to clipping, without making extraneous mechanical noises. The cone excursion was impressive. Heat build-up will be a concern in the longer-term, though.
Chris
That math is crazy!
The fact that HT subs dig so much deeper than PA subs is one of the things that got me thinking along these lines in the first place. I knew there had to be a tradeoff, but I hadn't realized what it was.
From reading these posts, I guess the tradeoff is that a HT sub wouldn't play as loud at audible frequencies and can't sit close to max output for hours on end, yeah?
I thought of this as well. If we could add a pair of infrasubs no bigger than our SR18Ss, that would be workable. And they'd only handle <2 octaves, so probably a bit less hard on the drivers than if they were trying to do everything from 80 Hz down, right?
20Hz is definitely audible at 110dB to an average person.
The ability to play for hours on end really depends on the driver, but in general PA subs are designed to handle more thermal abuse than HT.
Your Mackies can handle down to 40Hz, and there generally isn't much content in the 20-40Hz octave, so on average far less power used, but the duty cycle may be worse when it is.
This is an idea I've pinned in case I do end up trying an infrasub build.
@Rik_bS, what problem(s) would be caused by unused volume?
In dj culture low bass is mandatory and they often build custom subwoofers for that. Low there means close to 30hz, lower than that is not needed and does not sound right on high volume. Even lower 30's on high volume are quiet intense, and not everybody can handle that. On (stepper)dub or old style dubstep parties (those with the lowest bass in general in dj culture) very often people go bad on those basslines and need medical care because they are not used to it and go to close to the sub. So 20Hz or lower is certainly not a good id.
Reggae soundsystems uses groups of scoops (backloaded horns with compression chambers) to go to the low 30's. But today you got more efficient systems actually, like tapped horn or tranflex/paraflex systems. They go very loud very low, but are still quiet big. For your case i would look at the smaller system like that tuned to 30-35hz. You don't need 130dB for the crowd you play for, 120dB will already be to much (but headroom does not hurt).
Reggae soundsystems uses groups of scoops (backloaded horns with compression chambers) to go to the low 30's. But today you got more efficient systems actually, like tapped horn or tranflex/paraflex systems. They go very loud very low, but are still quiet big. For your case i would look at the smaller system like that tuned to 30-35hz. You don't need 130dB for the crowd you play for, 120dB will already be to much (but headroom does not hurt).