Hi all
This is my first post here, and as I am only just beginning to learn about the art of loudspeaker design please forgive me if I ask any obvious newbie questions 🙂
I live in Bali, and noticed that the local craftsmen here create a huge variety of furniture, homewares and even entire building structures out of bamboo. Since it seems such a strong and versatile material, it strikes me that bamboo might also be a great material for building loudspeaker enclosures.
Searching the Internet, there appears to be a few speaker models on the market that use fabricated bamboo paneling in the cabinet construction, but none using the giant bamboo in its raw form (see attachment "big bamboo.jpg" for an idea of what this looks like). Perhaps there is a very good reason why this is the case (cylindrical speaker design does not seem to be a popular research topic!), but hopefully, its just because no-one has thought to try it before 😉
The internal structure of bamboo is interesting, as it naturally forms hollow "internodes" that are separated by a solid plate at each node joint (see "bamboo cross section.jpg"). My idea is to see if it is possible to utilise the natural form of one of these bamboo internode sections to create a full-range speaker enclosure.
I've decided to start with a full-range project to begin with, since I want to minimise complexities introduced by needing crossovers etc. I figured a good starting point would be to design a full-range "bookshelf" speaker, since that is roughly the size of one internode.
So to start my project, I found a scrap piece of giant bamboo from a local construction site to grab some measurements. Apparently it is possible to get bamboo poles as big as 20cm diameter, but the piece I found measured 15cm diameter on the exterior, and 12cm diameter to the interior. The internode length is about 49cm (46cm internal gap), which gives an internal volume of about 20.8L.
Sticking with the "natural product" theme, I would like to use bamboo paper cone drivers if possible. The only manufacturer I could find that advertises bamboo drivers is Tang Band (are there others? please tell me!). Given the size of the bamboo tube I'm working with, the 4" Tang Band W4-1879 seems like it might be a good match.
I've attached some Sketchup pics of the kind of design I'm thinking of. DISCLAIMER: I've not done any real modelling of this design yet, other than include measured dimensions in the drawing, so these pics are more to communicate the idea than a final design spec!
As you can see in the third diagram, the actual construction of this should be reasonably straightforward, basically consisting of two bamboo tubes intersected. Although not quite shown in the diagram, the top and bottom seals will actually be the natural bamboo plate within the bamboo itself. Once settled on a design, I will try to engage the local craftsmen to build it for me, as they have the right tools and are capable of creating a lovely finish to the wood.
Unlike typical enclosures, this cannot be designed with millimetre precision CAD/CAM in mind. Given that no two pieces of bamboo will ever have exactly the same dimensions (although we can get close), I need to try and come up with a design that is as forgiving as possible, while still endeavouring to sound good 🙂 My ultimate goal is to create some really decent-sounding handmade speakers using as many natural materials as possible, and make them look good too!
So, question time 🙂
1. Are there any resources I can refer to on designing a cylindrical enclosure? This is something I really haven't been able to find any info on. It seems that the curved shape of the internal cavity might be good at eliminating reflections, compared to traditional boxed shape? Maybe there are some limitations I need to know about?
2. Given the shape of the bamboo, it would be trivial to cut a bass-port in the bottom to vent the enclosure. Would this result in better sound? Alternatively, would that help make it more "forgiving" of differences in bamboo dimensions?
3. In order to get good spatial positioning, I read it helps if the enclosure volumes of two paired speaker match exactly. I was thinking that when making a pair, it would be possible to "fill" one of the enclosures with solid material a little to make the internal volumes match. Is this a good idea/necessary?
4. What about the positioning of the driver? Is the centre ok, or would it be better moved towards the top? I worry about standing waves forming inside the tube, and also about not making the centre-of-gravity of the unit too high.
5. Is 20.8L internal volume too great for the W4-1879? (The Vas on the spec sheet is 9.12L - although I know these are not directly related?). Does the internal volume matter as much provided it is greater than the ideal minimum? (this might be one way to get some flexibility in tolerances?).
6. Are there any design rules of thumb I could draw on that might help deal with the organic nature of this design and still result in a consistent sound?
Anyway, thank you for reading so far. I wonder if I am being crazy with this idea? Is there something obvious I'm missing that means I should not even attempt it?
I would love to hear your thoughts on this project and any tips or suggestions you might have! Pointers on how to start modelling this would also be great too. I am tempted to try the "suck it and see" approach and jump straight to construction, but the inner engineer in me suggests that some modelling first might be wise to make sure I'm not building a lemon 🙂
Thanks in advance!
This is my first post here, and as I am only just beginning to learn about the art of loudspeaker design please forgive me if I ask any obvious newbie questions 🙂
I live in Bali, and noticed that the local craftsmen here create a huge variety of furniture, homewares and even entire building structures out of bamboo. Since it seems such a strong and versatile material, it strikes me that bamboo might also be a great material for building loudspeaker enclosures.
Searching the Internet, there appears to be a few speaker models on the market that use fabricated bamboo paneling in the cabinet construction, but none using the giant bamboo in its raw form (see attachment "big bamboo.jpg" for an idea of what this looks like). Perhaps there is a very good reason why this is the case (cylindrical speaker design does not seem to be a popular research topic!), but hopefully, its just because no-one has thought to try it before 😉
The internal structure of bamboo is interesting, as it naturally forms hollow "internodes" that are separated by a solid plate at each node joint (see "bamboo cross section.jpg"). My idea is to see if it is possible to utilise the natural form of one of these bamboo internode sections to create a full-range speaker enclosure.
I've decided to start with a full-range project to begin with, since I want to minimise complexities introduced by needing crossovers etc. I figured a good starting point would be to design a full-range "bookshelf" speaker, since that is roughly the size of one internode.
So to start my project, I found a scrap piece of giant bamboo from a local construction site to grab some measurements. Apparently it is possible to get bamboo poles as big as 20cm diameter, but the piece I found measured 15cm diameter on the exterior, and 12cm diameter to the interior. The internode length is about 49cm (46cm internal gap), which gives an internal volume of about 20.8L.
Sticking with the "natural product" theme, I would like to use bamboo paper cone drivers if possible. The only manufacturer I could find that advertises bamboo drivers is Tang Band (are there others? please tell me!). Given the size of the bamboo tube I'm working with, the 4" Tang Band W4-1879 seems like it might be a good match.
I've attached some Sketchup pics of the kind of design I'm thinking of. DISCLAIMER: I've not done any real modelling of this design yet, other than include measured dimensions in the drawing, so these pics are more to communicate the idea than a final design spec!
As you can see in the third diagram, the actual construction of this should be reasonably straightforward, basically consisting of two bamboo tubes intersected. Although not quite shown in the diagram, the top and bottom seals will actually be the natural bamboo plate within the bamboo itself. Once settled on a design, I will try to engage the local craftsmen to build it for me, as they have the right tools and are capable of creating a lovely finish to the wood.
Unlike typical enclosures, this cannot be designed with millimetre precision CAD/CAM in mind. Given that no two pieces of bamboo will ever have exactly the same dimensions (although we can get close), I need to try and come up with a design that is as forgiving as possible, while still endeavouring to sound good 🙂 My ultimate goal is to create some really decent-sounding handmade speakers using as many natural materials as possible, and make them look good too!
So, question time 🙂
1. Are there any resources I can refer to on designing a cylindrical enclosure? This is something I really haven't been able to find any info on. It seems that the curved shape of the internal cavity might be good at eliminating reflections, compared to traditional boxed shape? Maybe there are some limitations I need to know about?
2. Given the shape of the bamboo, it would be trivial to cut a bass-port in the bottom to vent the enclosure. Would this result in better sound? Alternatively, would that help make it more "forgiving" of differences in bamboo dimensions?
3. In order to get good spatial positioning, I read it helps if the enclosure volumes of two paired speaker match exactly. I was thinking that when making a pair, it would be possible to "fill" one of the enclosures with solid material a little to make the internal volumes match. Is this a good idea/necessary?
4. What about the positioning of the driver? Is the centre ok, or would it be better moved towards the top? I worry about standing waves forming inside the tube, and also about not making the centre-of-gravity of the unit too high.
5. Is 20.8L internal volume too great for the W4-1879? (The Vas on the spec sheet is 9.12L - although I know these are not directly related?). Does the internal volume matter as much provided it is greater than the ideal minimum? (this might be one way to get some flexibility in tolerances?).
6. Are there any design rules of thumb I could draw on that might help deal with the organic nature of this design and still result in a consistent sound?
Anyway, thank you for reading so far. I wonder if I am being crazy with this idea? Is there something obvious I'm missing that means I should not even attempt it?
I would love to hear your thoughts on this project and any tips or suggestions you might have! Pointers on how to start modelling this would also be great too. I am tempted to try the "suck it and see" approach and jump straight to construction, but the inner engineer in me suggests that some modelling first might be wise to make sure I'm not building a lemon 🙂
Thanks in advance!
Attachments
The circular structire of the bamboo should make a very good enclosure. The "end caps" may need reinforcement.
If the 20-some litres is too large you can always use random solid chunks as filler to reduce the volume if needed.
dave
If the 20-some litres is too large you can always use random solid chunks as filler to reduce the volume if needed.
dave
Cool idea.
I wouldn't worry about bamboo as a material. To my knowledge bamboo is much stronger than wood. Additionally, with reguards to pressure, cylindrical objects actually deal with pressure (much much much) better than cubes or rectangles. There's a reason pressurized hoses/pipes are cylinders 😉
The only reference i know to a cylindrical enclosures effects on frequency response is here. Not really a guide, just was a very generalized experiment to find the optimal enclosure shape. Which is a sphere if your wondering 😛
Anyway, filling a portion up from the bottom would be very easily done with concrete or sand. I'd imagine sand might actually have some damping characteristic that would give it an advantage over concrete. Not to mention it would be cheaper, faster, and easier to do lol.
As far as standing waves and whatnot inside an enclosure, stuffing with polyfill takes care of a good majority of it. Generally 1 to 1.25 pounds per cubic foot of enclosure is recommended and raises apparent enclosure size by approximately 25%. I'm sure coco coir (hair like) would work just as well if you can get it free around you.
As far as enclosure size goes, you can model the effects with WinISD. I do have to admit, it's quite confusing for a novice to use WinISD and understand what's going on though, so here's some general guidelines.
For any driver, a lower Qts will give a worse low frequency response than a higher Qts for a given Fs. Qtc of a sealed enclosure can never equal or be lower than the Qts of the driver. We can use this to our advantage.
Without a ton of technical jargon, I'll just tell you that a speaker driver with a Qts of 0.5 is the most accurate/fastest to the input signal (music). Lower Qts' are slower, and higher Qts ring (both are very generally bad except for certain circumstances).
Remember how I said enclosure Qtc can never go below driver Qts? Well in a situation like yours, when your enclosure is going to be too large for your driver, we can just use a driver with a Qts around .5. That way we optimize low frequency response, and hopefully get somewhere close to between .5 and .707 (maximally flat response).
Basicially, in an oversized enclosure you'd probably be better off using the Tang Band W3-1364SA 3" (don't worry, it's bamboo 😀)
If you wish to calculate sealed enclosure size, here is an easy to use online calculator. Use 0.707 as "Actual System Q (Qtc)".
It's generally not recommended to mount a driver symmetrically along the vertical axis. Any driver putting out high frequencies is best aligned with the ear anyway, so it would be best to put the driver somewhere at the top, where it would be pointing at your ears if it was sitting at your desk (or whatever your putting it on).
I wouldn't worry about bamboo as a material. To my knowledge bamboo is much stronger than wood. Additionally, with reguards to pressure, cylindrical objects actually deal with pressure (much much much) better than cubes or rectangles. There's a reason pressurized hoses/pipes are cylinders 😉
The only reference i know to a cylindrical enclosures effects on frequency response is here. Not really a guide, just was a very generalized experiment to find the optimal enclosure shape. Which is a sphere if your wondering 😛
Anyway, filling a portion up from the bottom would be very easily done with concrete or sand. I'd imagine sand might actually have some damping characteristic that would give it an advantage over concrete. Not to mention it would be cheaper, faster, and easier to do lol.
As far as standing waves and whatnot inside an enclosure, stuffing with polyfill takes care of a good majority of it. Generally 1 to 1.25 pounds per cubic foot of enclosure is recommended and raises apparent enclosure size by approximately 25%. I'm sure coco coir (hair like) would work just as well if you can get it free around you.
As far as enclosure size goes, you can model the effects with WinISD. I do have to admit, it's quite confusing for a novice to use WinISD and understand what's going on though, so here's some general guidelines.
For any driver, a lower Qts will give a worse low frequency response than a higher Qts for a given Fs. Qtc of a sealed enclosure can never equal or be lower than the Qts of the driver. We can use this to our advantage.
Without a ton of technical jargon, I'll just tell you that a speaker driver with a Qts of 0.5 is the most accurate/fastest to the input signal (music). Lower Qts' are slower, and higher Qts ring (both are very generally bad except for certain circumstances).
Remember how I said enclosure Qtc can never go below driver Qts? Well in a situation like yours, when your enclosure is going to be too large for your driver, we can just use a driver with a Qts around .5. That way we optimize low frequency response, and hopefully get somewhere close to between .5 and .707 (maximally flat response).
Basicially, in an oversized enclosure you'd probably be better off using the Tang Band W3-1364SA 3" (don't worry, it's bamboo 😀)
If you wish to calculate sealed enclosure size, here is an easy to use online calculator. Use 0.707 as "Actual System Q (Qtc)".
It's generally not recommended to mount a driver symmetrically along the vertical axis. Any driver putting out high frequencies is best aligned with the ear anyway, so it would be best to put the driver somewhere at the top, where it would be pointing at your ears if it was sitting at your desk (or whatever your putting it on).
Last edited:
... the optimal enclosure shape. Which is a sphere if your wondering
From a diffraction point of view a sphere is amongst the best, but one can do better, from the POV of the inside of the enclosure is is problematic, due to a strong single internal resonance potential.
dave
One thing to be careful of with bamboo is that as it dries out, it will crack and split - this is the natural state of dry bamboo. Which is why bamboo musical instruments are carefully wrapped in thread at the joints and kept in a humidity controlled environment. Shakuhachi flutes have this problem and they are wrapped with thread that is then lacquered, and they are kept in waterproof bags with a small source of moisture when not in use to prevent destructive cracking and splitting. Bamboo is inherently a nice strong material for aero-acoustical devices like flutes and TL speakers though.
From a diffraction point of view a sphere is amongst the best, but one can do better, from the POV of the inside of the enclosure is is problematic, due to a strong single internal resonance potential.
dave
Interesting! Have any studies been on internal enclosure shapes/ratios? Just wondering.
Thanks for the heads up!
Interesting! Have any studies been on internal enclosure shapes/ratios?
Yes, some has. I haven't seen anything as extensive as Olson's study of external shapes.
dave
I would use the 'solid plate at each node joint ' for making a resonating chamber , by making a hole in it. Also the stuffing is important; I suggest tar foils+ waste felt available at automotive stores, it can flex to make 10 cm diameter circle.
The first intersection that holds the driver needs to be cutted when entering the vertical bamboo body, otherwise it forms a first chamber...maybe it could be useful, or not ..!?
And finding decreasing measure ( in diameter) bamboo, you can make a TL...looking like a snail. Ok, the last secret is expansion rate ( tapering) and damping. I'll leave it to you 🙂
The first intersection that holds the driver needs to be cutted when entering the vertical bamboo body, otherwise it forms a first chamber...maybe it could be useful, or not ..!?
And finding decreasing measure ( in diameter) bamboo, you can make a TL...looking like a snail. Ok, the last secret is expansion rate ( tapering) and damping. I'll leave it to you 🙂
Thanks everyone for your feedback. It has been most insightful!
planet10: Yes the end caps seem small in that photo, since the cross-section is of a much smaller piece of bamboo. In the big sample bamboo piece I've got it seems to be similar thickness to the walls (i.e. around 15mm) but I think it gets thinner in the middle. Might be able to reinforce it internally so it is not visible.
Xaborus: Thanks for the link to those diffraction graphs. It does seem that a sphere is clearly the best, however it is interesting that "Shape D Cylinder Side Mount" is actually pretty much exactly what this bamboo design would be. The chart suggests that the response is certainly not as good as a sphere, but not too bad compared to other shapes either (although it would be good to be able to see a larger version of this chart to be able to read the scale). The slight dip in the middle seems to be the only concern.
I was interested to see that "Shape C Cylinder centre-mount" has an utterly disastrous response chart. I guess it makes sense though, since it is going to cause all kinds of standing waves to form.
Do you think this would be a problem for subwoofer frequencies though? I've also got a design idea for a bamboo subwoofer that uses pretty much this exact configuration, i.e. a driver mounted at the end of a bamboo tube, creating a sealed sub unit - kind of like a bamboo version of the standard cylindrical subwoofer design. The trick might be in finding bamboo which has a large enough diameter to house the bigger drivers though.
Yes I was also thinking that coconut coir would make a good enclosure stuffing material - I didn't know that was the correct name for it though! 😉. It can be found all over the place here and it certainly meets the natural materials goal. Not sure what the exact dampening properties are, but I'll give it a try.
Actually, this discussion just gave me another idea for a project to try after this - a spherical enclosure design based on a coconut shell! 😎
Can you please elaborate why you think I might be better off using a 3" driver instead of the 4" with an oversized enclosure? I'm drawn towards the 4" because of the lower frequency reach, which is important as I want these to be able to be used on their own without a sub. I also always thought that the bigger the enclosure the better, since more space for the sound to be dispersed, hence also the use of padding material to increase the effective volume? Running some numbers through that calculator, it actually surprised me how small some enclosures need to be!
xrk971: Thanks for the tip with possible cracking. One of the suppliers I spoke to here said they process and treat all their bamboo before they sell it, and it is definitely not green. I wonder if it is dried slowly and carefully it may avoid cracking? Or perhaps because we are in the humid tropics cracking from drying out might not actually be a problem here? I'll make sure to ask them about it anyway.
picowallspeaker: Do you mean the section that holds the driver needs to be cut so it is flush with the main bamboo structure internally? I don't think that would be too hard to do. It would be interesting to try both ways and see if there was an audible difference. The reason I did it this way was it makes construction pretty much trivial, since it literally is just cutting a hole in the main bamboo and inserting a straight-cut tube with the same diameter as the driver mounting ring.
What does everyone use to assemble their conventional MDF-type speaker enclosures? Wood glue, or something stronger? Just wondering what would be strong enough for constant vibration!
planet10: Yes the end caps seem small in that photo, since the cross-section is of a much smaller piece of bamboo. In the big sample bamboo piece I've got it seems to be similar thickness to the walls (i.e. around 15mm) but I think it gets thinner in the middle. Might be able to reinforce it internally so it is not visible.
Xaborus: Thanks for the link to those diffraction graphs. It does seem that a sphere is clearly the best, however it is interesting that "Shape D Cylinder Side Mount" is actually pretty much exactly what this bamboo design would be. The chart suggests that the response is certainly not as good as a sphere, but not too bad compared to other shapes either (although it would be good to be able to see a larger version of this chart to be able to read the scale). The slight dip in the middle seems to be the only concern.
I was interested to see that "Shape C Cylinder centre-mount" has an utterly disastrous response chart. I guess it makes sense though, since it is going to cause all kinds of standing waves to form.
Do you think this would be a problem for subwoofer frequencies though? I've also got a design idea for a bamboo subwoofer that uses pretty much this exact configuration, i.e. a driver mounted at the end of a bamboo tube, creating a sealed sub unit - kind of like a bamboo version of the standard cylindrical subwoofer design. The trick might be in finding bamboo which has a large enough diameter to house the bigger drivers though.
Yes I was also thinking that coconut coir would make a good enclosure stuffing material - I didn't know that was the correct name for it though! 😉. It can be found all over the place here and it certainly meets the natural materials goal. Not sure what the exact dampening properties are, but I'll give it a try.
Actually, this discussion just gave me another idea for a project to try after this - a spherical enclosure design based on a coconut shell! 😎
Can you please elaborate why you think I might be better off using a 3" driver instead of the 4" with an oversized enclosure? I'm drawn towards the 4" because of the lower frequency reach, which is important as I want these to be able to be used on their own without a sub. I also always thought that the bigger the enclosure the better, since more space for the sound to be dispersed, hence also the use of padding material to increase the effective volume? Running some numbers through that calculator, it actually surprised me how small some enclosures need to be!
xrk971: Thanks for the tip with possible cracking. One of the suppliers I spoke to here said they process and treat all their bamboo before they sell it, and it is definitely not green. I wonder if it is dried slowly and carefully it may avoid cracking? Or perhaps because we are in the humid tropics cracking from drying out might not actually be a problem here? I'll make sure to ask them about it anyway.
picowallspeaker: Do you mean the section that holds the driver needs to be cut so it is flush with the main bamboo structure internally? I don't think that would be too hard to do. It would be interesting to try both ways and see if there was an audible difference. The reason I did it this way was it makes construction pretty much trivial, since it literally is just cutting a hole in the main bamboo and inserting a straight-cut tube with the same diameter as the driver mounting ring.
What does everyone use to assemble their conventional MDF-type speaker enclosures? Wood glue, or something stronger? Just wondering what would be strong enough for constant vibration!
This is my first post on here (or any fur on for that matter) I'm new to speaker building, a friend of mine upgraded his Ascend Seirra-1s and had the old crossovers and speakers that he gave to me for free!
I could not find bamboo sheets so I found flooring at Lowes on Clearance for$25 bucks. I glued them together to make panels and built these towers with them. I am amazed by the sound quality, I have about $49 into these speakers and they sound better than my old Infinity Reference 2006s.
I've included a few pics. I could have done better On the finish but think they will look better in another 20 days when the Lacquer cures and I can rub them out and was them.
I really enjoyed this project and am looking forward to building more. Sub woofer is next.
-Tom
I could not find bamboo sheets so I found flooring at Lowes on Clearance for$25 bucks. I glued them together to make panels and built these towers with them. I am amazed by the sound quality, I have about $49 into these speakers and they sound better than my old Infinity Reference 2006s.
I've included a few pics. I could have done better On the finish but think they will look better in another 20 days when the Lacquer cures and I can rub them out and was them.
I really enjoyed this project and am looking forward to building more. Sub woofer is next.
-Tom
Attachments
Hey/ Hi Tom
the argument was most about bamboo/cylindrical structure rather than finishing
so it's ok BTW Those speakers& crossovers were a good score but why are you thinking of a sub when you've already used much volume in those boxes ? Maybe they're overdamped or something...and it means bad (bass) alignement .
the argument was most about bamboo/cylindrical structure rather than finishing
so it's ok BTW Those speakers& crossovers were a good score but why are you thinking of a sub when you've already used much volume in those boxes ? Maybe they're overdamped or something...and it means bad (bass) alignement .
Bamboo speakers
Thanks for the response.
Sorry about that, just learning how the forum works, didn't realize I hi-jacked another thread.
I built basically bookshelf speakers with the stands "built in". I separated the top 14" from the bottom with 1.25" bamboo divider (see pic) so there isn't as much volume as it appears. Don't get me wrong, they sound great and have decent bass (I was actually surprised how much bass).
I'm thinking of building a sub just because I enjoyed this project so much, and I don't think it would hurt adding a dedicated sub. They are in a large room (17x28).
Hey/ Hi Tom
the argument was most about bamboo/cylindrical structure rather than finishing
so it's ok BTW Those speakers& crossovers were a good score but why are you thinking of a sub when you've already used much volume in those boxes ? Maybe they're overdamped or something...and it means bad (bass) alignement .
Thanks for the response.
Sorry about that, just learning how the forum works, didn't realize I hi-jacked another thread.
I built basically bookshelf speakers with the stands "built in". I separated the top 14" from the bottom with 1.25" bamboo divider (see pic) so there isn't as much volume as it appears. Don't get me wrong, they sound great and have decent bass (I was actually surprised how much bass).
I'm thinking of building a sub just because I enjoyed this project so much, and I don't think it would hurt adding a dedicated sub. They are in a large room (17x28).
Attachments
- Status
- Not open for further replies.
- Home
- Loudspeakers
- Full Range
- Bamboo speaker project