Custom 16hz subwoofer box

[rjsilva]

I have a 10” Klipsch subwoofer driver from the RW-10d model. The amp was roasted a while back and so I trashed it and the box and kept the driver. In the back of my mind I've wanted to use it for something and recently I've wondered how possible it would be to design a custom cabinet that could get a useful 16hz out of the driver.

I realise that I'd need certain parameters about the driver which probably aren't available without testing it, but I wondered where a good place to start would be for approaching a solution.

[Eric Mack]

You might find some useful information in the Transmission Line Speaker Forum. I'm planning to build one when I complete my Walker T3 Clones. https://www.facebook.com/groups/tl.speakers/

[Papus]

Maybe shouldn't have trashed the original enclosure which was designed for that specific purpose....
Curious - why do you need 16Hz? nobody can hear it except certain species of whale.
Certainly no Hammond produces a note that low

[rjsilva]

This thread was moved from the church organ thread :) Church organs need strong 16hz capability for proper 32' stop reproduction.

The driver in the original enclosure could get down decently maybe to 29hz, which is only a couple of notes into the 32' stop range. I don't even know if this driver could do a decent 16hz in a different enclosure. I contacted Klipsch seeing if they'll give parameters for the driver but I have my doubts they'll give it.

The transmission line speakers are definitely interesting. Thanks for recommending it Eric. I'm wondering how large of an enclosure would be needed for this driver to do 16hz (but I'll need the driver details to determine that).

[Papus]

Ah, that makes sense!
Transmission line is probably the way to go.
This technology goes way back to pre-WWII, it is well-researched and proven.
It's a neat way of extending the response of a system driver to be below its natural response by coupling it with matching drivers with precise spacing and phase relationships, in effect creating a single massive subwoofer out of several smaller units - in much the same way astronomers use many smaller telescopes to create a single virtual massive telescope.
Fun fact: digital recording systems usually roll off the bottom end around 25-30Hz, to prevent the lower limit of 20Hz becoming problematic. So unfortunately most digital recordings of great pipe organs cannot reproduce the lowest notes, you'll only get the harmonics.

[rjsilva]

Thanks for encouraging me in that direction Papus. I'll do some research.

My idea, of it ever materialises, is intended for organists themselves. A sub that can produce a relatively flat signal down to 16hz (that doesn't also break the bank) is something that feels unattainable. If I could come up with a design that is cheap enough perhaps it might be desirable for organists who have home consoles producing 32' stops but who don't have the speakers to properly pull it off because those kinds of organ speakers tend to be very pricey. I'm only looking for the range of 16hz with a roll off to 64hz max, perhaps with a built in crossover to easily enhance existing subs.

[Papus]

I wonder if it is possible to take advantage of basic physics and psychoacoustic principles here, by using the same technique that allows "bass foldback" on Hammond organs to produce audible tones a whole octave lower than the tones physically produced by the tonewheels...
I believe some pipe organs employ this technique also, but I forget the technical term for it - it involves blending a fundamental note with a fifth and an octave in certain proportions, the wave-fronts align to produce an audible note one octave below the fundamental. This occurs in the air, not in the pipes or the speaker itself.
I've analysed the notes produced by my T202. The lower manual bottoms out at F2 87.3 Hz. Even with lower manual bass foldback installed, the lowest note physically produced by the organ via direct electrical connection is F2 87.3Hz (not including the pedalboard).
But an analysis of a microphone placed in front of the speaker confirms that the lowest note produced in the air is F1 43.7Hz - one octave below the bottom limit of the organ.

[rjsilva]

Interesting thought Papus. It's called a resultant stop (on a pipe organ at least). But ... even if that were equivalent to a native 32' stop, you can't really change existing stops on most digital/analog organs from a 32' to a resultant :)

[toodles]

Achieving 16 Hz at affordable prices is indeed possible, but includes either very large enclosures or true infinite baffles.

One interesting option for transmission lines is folding the line using concentric cardboard tubes of descending diameter, the inner tubes being of slightly shorter length than the outside tube, staggering the short end between the top an bottom on alternating tubes. As tubes are inherently circular they are extremely strong against flexing, permitting a rather thin wall of lightweight construction.

[rjsilva]

I'm having trouble visualising that tube idea, in particularly where the port would be. But the more I'm researching transmission line the more I'm intrigued. From what I understand, a flat response with a limited range suits it well. If Klipsch gives me the specs on that driver I can try and experiment with little cost to me—that's pretty exciting.

[toodles]

If the line terminates at the bottom of the enclosure, the "port" (and it really isn't a port, but the opening of the line to the acoustic environment) is a cylindrical section of the bottom of the enclosure. If the line terminates at the top of the enclosure, the port is an annulus on the outer edge of the top plate of the port or a cylindrical section of the outer part of the enclosure (as you desire and think is easier to construct).

The concept is the same as mitered organ pipes or Haskell bass pipes, except spun through 360 degrees. The concept provides for very easy, simple, construction and a lightweight, extremely strong structure compared to plywood boxes.

Here's a cross sectional sketch showing the general idea--the arrows show the air path from the speaker cone to the outside of the enclosure, with the port at the bottom.

Obviously there need to be some internal braces to keep the cylinders spaced and some legs to support the short end of the cylinders either to the top or bottom plate depending upon which cylinder end is short. Top and bottom plates can be 3/4 inch plywood--a circle cutting jig provides a means to accurately cut circular rabbits in the top/bottom plate to keep the cylinders in place.

The cylinders can be concrete cylinder molds, readily available in incremental diameters. Outer cosmetic surface could be carpeting or even some type of veneer.

[samibe]

If I'm understanding your drawing and concept correctly, the diagram is a cross-section through three concentric cylinders. The inner one is about the same diameter as the driver. The next one out is a few inches (4"-8") bigger in diameter than the inner tube. The outside tube would be a few more inches (4"-8") bigger than the middle tube (8"-16" bigger than the inner tube).

If you were to cut three or four slotted holes in the ends of each cylinder, would that get around the additional bracing? (Since each cylinder could be embedded a little bit into both the top and bottom plates.)

[rjsilva]

Thanks Toodles!

So in your example you have two ‘ports'? Or am I not seeing it correctly. I like the idea of using cylinders.

I've been researching more. As I understand it, to tune at 16hz I'd need about 17.5ft of tube. I'm not quite understanding the roll off on the bottom end though. Could I tune to, say, 20hz and still get a flat response to 16hz?

I'm also a little unclear about how the Fs fits into it. One source seemed to suggest you're limited in response to that value (so for instance a driver with an Fs of 60hz wouldn't be appropriate for a 30hz transmission line enclosure), but then another source simply used the Fs in calculating the line length/diameter (or some part of it) and it didn't seem to matter much beyond that.

[toodles]

rjsilva: There are not 2 ports--remember each vertical line is the outer edge of a cylinder--imagine the inner cardboard tube of a roll of paper towels sitting an inch above a surface--the "port" is the entire space between the tube and the surface and so it is a thick disc in shape, something akin to a roll of masking tape, though much larger.

The drawing is a cross sectional view--as if you sliced through the assembly along a center line.

You can pretty much forget about flat response down to 16 Hz with any system using standard drivers--look at any driver's response curve, and you'll see it tapers off at low frequencies. The most you can hope for is smooth response and minimal loss from the mid-band response.

Note that the Qts of a driver provides the output at resonance; a Qts of 1.0 provides an output at resonance of 0 dB down from mid-band. The formula is output in dB is 20 times the log (base 10) of Qts. Log of Qts 1.0 is zero, times 20 is zero. Log of Qts 0.7 is -0.15, times 20 is -3.0, so a driver with a Qts of 0.7 is down 3 dB at resonance.

Drivers provide response below resonance, though the higher the Qts of the driver, the steeper the roll off curve. So while a driver with a high Qts would provide stronger response at resonance, it would roll off faster as the frequency goes below resonance; a driver with a low Qts would have a gentler roll off below resonance.

This image shows the effect: https://www.diyaudioandvideo.com/Cal.../Help/boxQ.gif

samibe: you understand the concept correctly. You wouldn't need to cut slotted holes--you could just drill multiple large diameter holes (say, about one inch) instead. Though that provides a little resistance to air motion, that resistance would act as dampening. Usually in a transmission line there is dampening material along the entire length of the line, often with increasing dampening density along the length of the line--sometimes done by tapering the line. Tapered cylinders are not very practical to build, so this approach could use staggered diameters and with dereasing spacing between cylinders to reduce the cross sectional annular area.

[rjsilva]

Yes, I understand now Toodles, thanks for explaining. A neat concept. For a 16hz tuning wouldn't that enclosure need to be about 6' tall?

So for a better understanding, this sub:

https://www.parts-express.com/dayton...4-ohm--295-458

Has an Fs of 29.6hz and a Qts of .4. How feasible would it be to get ‘smooth' (by flat I meant smooth :) ) response in the 16-32hz range using a transmission line design tuned to 16hz (or 20hz, to save a few feet on the tube). Of course the goal is not to have the sound become noticeably weaker as it approaches 16hz.

[toodles]

Yes, I understand now Toodles, thanks for explaining. A neat concept. For a 16hz tuning wouldn't that enclosure need to be about 6' tall?

So for a better understanding, this sub:

https://www.parts-express.com/dayton...4-ohm--295-458

Has an Fs of 29.6hz and a Qts of .4. How feasible would it be to get ‘smooth' (by flat I meant smooth :) ) response in the 16-32hz range using a transmission line design tuned to 16hz (or 20hz, to save a few feet on the tube). Of course the goal is not to have the sound become noticeably weaker as it approaches 16hz.

I would not choose that driver for a TL to get to 16 Hz, simply because its efficiency is very low to start with (about 86 dB) and with that Qts, you'll already be down to 77 dB at resonance--backwave via the TL added to the front wave would bring it back up to about 80 dB. Since the threshold of hearing at 16 Hz is about 70 to 80 dB, it won't be a very strong system in terms of output.

I don't have any modelling software for transmission lines, so I can't give specific advice on overall performance.

The overall height of the concentric cylinder design depends upon the total length you want and the number of nested cylinders--you can always add outer cylinders to get to a length at any specific height, but the overall diameter of the system gets bigger. Note that if you choose to mount a driver facing upward or downward, you'll need to make sure the driver is capable of that. Many of the very heavy cone drivers will sag if mounted facing up or down. For the cylinder sub, you can mount the cylinder horizontally and still use a heavy cone driver, or put a top on the box that holds the driver mounted on a vertical plane, with a "hole in the bottom of the box" to provide air flow into the cylinders--whether you find this aesthetically pleasing or not, I cannot say.

You might like to review this design just to get some ideas: https://spinditty.com/instruments-ge...nder-subwoofer

[rjsilva]

I would not choose that driver for a TL to get to 16 Hz, simply because its efficiency is very low to start with (about 86 dB) and with that Qts, you'll already be down to 77 dB at resonance--backwave via the TL added to the front wave would bring it back up to about 80 dB. Since the threshold of hearing at 16 Hz is about 70 to 80 dB, it won't be a very strong system in terms of output.

Forgive my ignorance. Isn't the 86dB sensitivity rating at 2W (4 ohms)? With the driver being able to handle 300W RMS couldn't it reach quite a bit higher than 80dB?

Thanks for the tips and the link! A fun read.

[toodles]

Usually power limits on a driver are based on the thermal limits of the components used, including adhesives. They don't reflect how much power the driver can accept and produce sound without distortion. To provide a true limit of the power, you have to know the speaker cabinet being used, how much distortion is acceptable. The driver manufacturer doesn't know these things--only the system designer can know them. Often the actual power limits will be lower than the thermal limits.

To know the true maximum SPL output of the system, you have to calculate the Per (maximum power based on the excursion limits of the driver, which will vary with the level of distortion that is acceptable), and from that you can calculate Par (maximum acoustic power at Per). I don't know the forumlae for these for a transmission line system.

Suffice it to say, for every doubling of power supplied to the driver, you gain 3 dB in acoustical output (acoustical output in db = 10 times the power ratio); so 300 watts gives you 24.7 dB over 1 watt; but if you start with 2 W for the 4 Ohm driver, you just get 21.7 dB. Unfortunately, the ultra high power amps used for subwoofers almost invariably include an infra-bass filter, so they don't like to give full output at 16 Hz. If you look at the response curve for that Parts Express driver, you'll see that response at 20 Hz is down about 10 dB from resonance, and the curve is headed downward from there into 16Hz. Look at the equal loudness curves (wikipedia is a good place to start) and you'll see you have to get 16 Hz sources to the range of 85 dB just to be heard.

Sensitivities are measured at 1 watt, so to get that sound into the room very far you need to produce a very loud sound, In the end, the most successful speakers that work down to 16 Hz are those with very efficient cabinets (bass reflex of large size with large driver surface area) and infinite baffles with multiple large drivers.

A 4 ohm speaker will pull about twice the power from an amplifier than an 8 ohm speaker, so sensitivity is given based on 1 watt for 8 Ohm speakers, and on 2.83 V for 4 Ohm speakers to give sensitivity that can be compared. Of course, if your amp can't deliver the extra power, you don't get the benefit of it.

[Goff]

You could also get this. It goes down to 5hz up to 200hz which might connect better to the lower rotor's frequency. You can connect this to a platform or your organ bench and feel rather than hear the low hz.


You can't really hear 16hz. But you can feel it resonate. Like an earthquake resonates the ground and a building before it starts shaking it.

It's only 5.5 inches by 5.5 inches.

[Eric Mack]

The folks on the TL Loudspeaker forum I linked above have solutions that are not enormous that go to 16hz using a 10" subwoofer. The moderator has been doing this for 40 years and has many designs and examples documented. I'm thinking of a small one to put under my grand piano where it won;'t be seen. I encourage you to check out the site. I am also encouraged that the reason the moderator got into this was his love of organs and the need for suitable low frequency reproduction. (I have no experience, just sharing what I have observed there.)

[rjsilva]

Thanks Eric for encouraging me again. I did visit the link but I've been putting my energy into understanding speaking design better. I'll definitely visit the forum!