I have a crazy idea that's probably a misguided mishmash of speaker design, and I'm no audio engineer (but am well versed in electronics), so please excuse anything here that sounds idiotic. I'm just trying to filter out what doesn't make sense and make something cool.
That said, I'm thinking of designing a subwoofer with the following considerations:
And yes, subs aren't directional in lower frequencies, but individual floor sections shaking would be, as would frequencies up into 120Hz.
But that's just the low end. I was watching the Tech Ingredients video on Voigt tubes for full range speakers, and thought, what if I made that but tuned it to extend a subwoofer's range in both directions? Now I'm imagining the aformentioned setup with 3 or 4 drivers at the floor on a large, slanted column that takes up the corner up to the vent at the ceiling.
Is this a crazy/stupid idea?
Here's a quick mockup of what I was thinking it would look like. The gray lines are the floor joists set up such that they don't inhibit air movement.
That said, I'm thinking of designing a subwoofer with the following considerations:
- As low of response as possible, hopefully 10Hz or less.
- Upper response into 120Hz.
- Multiple drivers (3 or 4) for ability to move big air and get really low while also being smaller (I'm thinking 12") and cheaper with good fidelity into the higher frequency.
- Large, almost infinite baffle tuned to the low frequency response target.
And yes, subs aren't directional in lower frequencies, but individual floor sections shaking would be, as would frequencies up into 120Hz.
But that's just the low end. I was watching the Tech Ingredients video on Voigt tubes for full range speakers, and thought, what if I made that but tuned it to extend a subwoofer's range in both directions? Now I'm imagining the aformentioned setup with 3 or 4 drivers at the floor on a large, slanted column that takes up the corner up to the vent at the ceiling.
Is this a crazy/stupid idea?
Here's a quick mockup of what I was thinking it would look like. The gray lines are the floor joists set up such that they don't inhibit air movement.
Last edited:
That guy. Sigh. He has the basics down, he has a lot to learn. As much somehat misleading as informative.
A voigt has been shown to significantly improve with mass loading.
But in your situation a pr of BIBs that take up the entire corner(s) of the room and are large enuff to, with the aid of the corner finishing the horn), brings a simple, effective voigt-pipe/horn.
dave
A voigt has been shown to significantly improve with mass loading.
But in your situation a pr of BIBs that take up the entire corner(s) of the room and are large enuff to, with the aid of the corner finishing the horn), brings a simple, effective voigt-pipe/horn.
dave
Last edited:
A labyrinth is a real good solution. Piece of cake to make a long folded pipe with the driver on the floor and the exit port near the ceiling (so they don't interact in the assumed text-book way).
https://www.diyaudio.com/forums/subwoofers/322418-17-foot-pipe-sub-12-230-hz-5db.html
The essential limitation of bass extension is the resonance of the drivers. That's the key parameter.
https://www.diyaudio.com/forums/subwoofers/322418-17-foot-pipe-sub-12-230-hz-5db.html
The essential limitation of bass extension is the resonance of the drivers. That's the key parameter.
Yeah, I've been running into a lot of misinformation affecting my ideas for this project (an HT room). And for some reason, I thought he had left a small gap/port at the top, so I mispoke earlier about the port being at the ceiling.
But how would I model such a monstrosity? I can't find any software that models Voigt tubes/BIB other than some rumored Mathcad things I can't find and probably wouldn't be able to figure out easily from the sounds of it, and from what I've read, it might not be accurate anyway? Add in multiple drivers per box, using the floor as the horn portion, etc, and I don't think it's even going to be possible from what I'm seeing. So should I just go for it and see what happens, lol? Or should I model it as a sealed box, only paying attention to the volume, with a very large passive radiator?
Although, I wasn't really planning on using the floor portion as a horn per se, so perhaps this isn't quite a BIB or Voigt... unless I was to cut ports into 2 of the edges of the floor frame facing away from the sub. But since I was hoping to suspend the floor for tactile movement, it's something of a very large, heavy passive radiator. Technically, since mass loading is not only done with port restriction but also moving component mass, does that count? That's a lot of mass, and then there's seating and people on top of that. I figured hydraulic principles would dictate that a large air movement from several drivers of smaller surface area could move a large floor area commandingly if only a small amount, but plenty enough to feel it without being bouncy.
Last edited by a moderator:
BIB is easy, Godzilla has a calculator.
Martin King’s no longer available Mathcad modeler is the gold standard. HornResp is most commonly used today for modeling these, but there is this unsupported modeler. A few more have been popping up.
https://www.diyaudio.com/community/threads/transmission-line-modelling-software.220421/
BIB is easy, but trade big for finesse and really needs that corner loading. Used as just a woofer means that its largest deficit no longer plays much role.
dave
Martin King’s no longer available Mathcad modeler is the gold standard. HornResp is most commonly used today for modeling these, but there is this unsupported modeler. A few more have been popping up.
https://www.diyaudio.com/community/threads/transmission-line-modelling-software.220421/
BIB is easy, but trade big for finesse and really needs that corner loading. Used as just a woofer means that its largest deficit no longer plays much role.
dave
Thanks, I had just found that but hadn't played around with it much because I was planning on multiple drivers per box and wasn't sure how that affected the parameters.
But then I got to thinking that, since this is meant to boost the bass response of a single, full range driver, doing so for a large woofer, multiple drivers would be a bit overkill, especially with 12 of them in the room. So I went looking for a good driver to plug into the Godzilla calculator. 18" made way too big of a box, even if folded. 15" still too big. 12" just right, which I was already planning to use! And this one from Eminence has a very low Fs of just 23Hz, decent excursion of 13mm, and says it's good for horn loading. Plugged that in and it's about as close to perfect as I can probably get with the horn folded up under the floor exiting under each corner and side of the seating.
So, now I'm thinking I'll get some 14" sonotube, cut it at the appropriate angle, build a concrete form in a corner with the flat driver panel covering the whole corner on the now angled sonotube, terminating at the upper corner, and pour concrete to form the back (leaving the sonotube in place) with a gradual radius to direct the rest of the horn into the floor. This puts my driver 3' off the floor which is nice, and leaves plenty of room for the large magnet which otherwise wouldn't fit in a rectangular box... although perhaps it would be better to keep the large volume of the driver outside of the box as much as possible and build up a raised mount or just use a very thick panel?
So, all 6 subs will have a horn exiting into a large space under their respective portions of the seating area which should limit the amount of area they act upon, thus increasing the tactical effect in each corner and side so ULF will be felt from that direction.
Now I'm getting excited because it's really looking like this will work as expected and be fairly cheap and easy. Yay! Now to mock it up and test it. So far, all my calculations are looking good. Just figured out the change in area as it goes down the tube (0.911n) and calculated the area needed at the driver and the base, and how that relates to... well at first a rectangle until I noticed the driver doesn't fit, then, since it's in a corner, just using a triangle shape, but then figured it would be good to make it round, and started calculating for a sliced tube which makes for even more driver space. Perfect!
:^)
You should be able to move a lot of air. With 6 channels of woofer, 6 channels of amplifier and some sort of room EQ for them. Toole has a good section on woofers well worth reading.
dave
You should be able to move a lot of air. With 6 channels of woofer, 6 channels of amplifier and some sort of room EQ for them. Toole has a good section on woofers well worth reading.
dave
Actually, if I'm doing 6, might as well do the LFE channel too (for the front center) for 7 channels. I have an Outlaw 770 that should power them nicely off of the pre-outs on an NAD T758 AVR. The only question there is if I can get BEQ to pull out the ULF for all 7 channels.
I've so far been told that it cannot, but if BEQ is just a filter, why couldn't it be applied to each channel? But I don't fully understand BEQ, so I'm still researching that.
Do you mean the book by Floyd E. Toole,
Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms?
Thank you very much @planet10 . What a great suggestion. I spent the weekend perusing Floyd's book, and the very first reference I found in it was to a youtube video of his conference in 2015... which led to an interview of course, and just wow. That was really eye opening. I feel like I have fell into a very nice rabbit hole, so thanks again! I also got a whole bunch of other books... that were all nothing but disappointments with very little explanation for WHY or HOW things work, just plans and ungrounded explanations.
Using the BiB calc, you enter the Fs of the driver (my current choice being the Dayton Ultimax 15" with Fs of 19.5Hz). But the point of this idea is to stretch down to 10Hz or lower. Are voigts and BiBs limited to the resonant frequency of the driver, or should I enter a lower value into the calc? And if mass loading is good for these boxes, that lowers the resonant frequency right? So therefore, should I just stick to the calculator and lower the resonant frequency with mass loading?
I'm thinking of mass loading by covering the horn opening with AT spandex fabric and placing weight on it in some fashion, sort of like a hear through passive radiator. With such low frequencies, it's all about the air movement, so I thought, since the fabric WILL respond at least somewhat to the air movement (perhaps even covering some of the area with foam) I could get the mass loading without sacrificing air movement. Does that make sense or am I overthinking things?
I'm thinking of mass loading by covering the horn opening with AT spandex fabric and placing weight on it in some fashion, sort of like a hear through passive radiator. With such low frequencies, it's all about the air movement, so I thought, since the fabric WILL respond at least somewhat to the air movement (perhaps even covering some of the area with foam) I could get the mass loading without sacrificing air movement. Does that make sense or am I overthinking things?
Basic decision is whether you are tuning or not. Since, in my opinion, tuning is a bad idea for reproducing the sound of other sound sources that are tuned such as a piano, you should always aim to use untuned elements in a HiFi.
Forget about the elaborate guesstimate sims, just funnel the rear wave into a labyrinth that emits just the lowest notes your driver can muster in free air and in a way that interacts with the front wave incoherently.
B.
No! Fs/2. Note that originally there was just the math/instructions, so here's a small collection of postings of BIB variations/design considerations with the rest 'buried' in BIB threads, but the forums' 20 k limit blocked it, so searching the 'Loudspeakers' Forum will likely answer the vast majority of questions not addressed here.
Terry Cain's 40-1354 BIB
Outside box dimensions are 7"x14"x70" Driver height from base is 39" Baffle is centered at 5-1/2" between outer walls and floor of cabinet
http://www.diyaudio.com/forums/full...se-fostex-craft-handbooks-59.html#post1190733
Vb = 20*Vas*Qts^1.25
http://www.diyaudio.com/forums/full...se-fostex-craft-handbooks-59.html#post1191596
Fs = 70 Hz
Vas = 0.29 ft^3
Qts = 0.445
Vb = 20*Vas*Qts^1.25
L = 13464.54/2/70 = 96.175"
folded height = 96.175"/2 = 48.088"
Vb = 20*0.29*0.445^1.25 = 2.10804 ft^3
Sm = 2.10804*1728/48.088 = 75.751"^2
depth = SQRT((75.751*SQRT(2)) = 10.35"
width = 75.751/10.35 = 7.319"
zdriver = 96.175*0.217 = 20.87"
a-b-c = 10.35/2 = 5.175"
Z = 0.217, 0.349, 0.424, 0.6, 0.714, 0.848
Rev. BIB Excel SS: https://www.diyaudio.com/forums/att...5-begins-fostex-fe166en-bib-bib_166_gm_v2-zip
Godzilla BIB site w/ V1 & 2 Excel SS: https://speakerprojects.wordpress.com/cabinet-types/bib-loudspeakers/bib-calculator/
Godzilla BIB site w/ BIB layout, pictures: https://speakerprojects.wordpress.com/cabinet-types/bib-loudspeakers/bib-pictures/
Instructions:
https://www.diyaudio.com/forums/full-range/313921-bigger-bigger-badder-bib-post5571336.html
The BIB default alignment is Fs/2, so input Fs, Vas, Qts' and it auto calculates everything you need to input it into HR/whatever.
The basic math is: Vb = 20*Vas*Qts'^1.25
Qts' = Qts + any added series resistance [Rs]: http://www.mh-audio.nl/Calculators/newqts.html
Click test:
If it's too short or too long, i.e. too far from the ceiling or too close, then change 'Fs' until you get the desired folded height.
For really high Fs drivers or just want a smaller floor loaded model, just change 'Fs' to suit.
If it's sheer size/'footprint' is too big for the intended app, then change Vas for relatively minor adjustments and/or Qts' for major ones, though may not adequately protect the driver enough in high power apps.
L*0.217 is the original/standard driver offset with L*0.41 being MJK's alternate when a flatter overall response is desired, though may not adequately protect the driver enough in high power apps.
For compression loaded, i.e. very close to the floor or ceiling, then one of these ML-horn/ML-TQWT driver offsets may be better overall, though again, the further down the less the driver protection at higher power: L*0.349, 0.424, 0.50 - 0.60
GM