Good evening!
I hope I'm not repeating some old topic about the subject, but I couldn't find any, so I decided to ask (also consider that I'm a newb in here, so any stupidity of mine "should" be taken with a smile
)
After building some TL loudspeakers in wood, I became a little bit more ambitious and started to plan building a spiral pipe around a pole, that would work as the QWTL tube for the L22 Seas Woofer, in a 3 way loudspeaker (the other 2 being, most likely, Seas W15CY and Vifa XT25). That said, a lot of ideas on how to make this pipe came into mind, notably building a structure in fiberglass (or any other fiber that suits this). The bigger problem is the mould for that.
I thought in 2 possibilities, both needing to be explored (and eventually abandoned if not suitable). One of them would be doing a pipe structure using wicker, this one rolling around the pole. This "wicker pipe" would be later covered by fiberglass (being some kind of endoskeleton). The other one would be using a regular PVC pipe, again covered by fiberglass, and removed (have no idea how) when the fiberglass is firm.
Can you please help me with this?
(here's a very ugly sketch)
Thank you very much!
(and since I'm on the season, Happy New Year to all of you!)
I hope I'm not repeating some old topic about the subject, but I couldn't find any, so I decided to ask (also consider that I'm a newb in here, so any stupidity of mine "should" be taken with a smile
)After building some TL loudspeakers in wood, I became a little bit more ambitious and started to plan building a spiral pipe around a pole, that would work as the QWTL tube for the L22 Seas Woofer, in a 3 way loudspeaker (the other 2 being, most likely, Seas W15CY and Vifa XT25). That said, a lot of ideas on how to make this pipe came into mind, notably building a structure in fiberglass (or any other fiber that suits this). The bigger problem is the mould for that.
I thought in 2 possibilities, both needing to be explored (and eventually abandoned if not suitable). One of them would be doing a pipe structure using wicker, this one rolling around the pole. This "wicker pipe" would be later covered by fiberglass (being some kind of endoskeleton). The other one would be using a regular PVC pipe, again covered by fiberglass, and removed (have no idea how) when the fiberglass is firm.
Can you please help me with this?
(here's a very ugly sketch)
An externally hosted image should be here but it was not working when we last tested it.
Thank you very much!
(and since I'm on the season, Happy New Year to all of you!)
Hi and Welcome.
When I saw your design sketch I couldn't help but think of this:
http://www.t-linespeakers.org/projects/davidduke/index.html
It's not fiberglass but it looks so much like yours I thought you might get some ideas from it. You won't need to make molds but it looks like a good bit of work nonetheless.
The rest of the site might help too, if you haven't seen it.
When I saw your design sketch I couldn't help but think of this:
http://www.t-linespeakers.org/projects/davidduke/index.html
It's not fiberglass but it looks so much like yours I thought you might get some ideas from it. You won't need to make molds but it looks like a good bit of work nonetheless.
The rest of the site might help too, if you haven't seen it.
I really like your design idea.
If I were to make this out of fiberglass, I would start out with a block of extruded polystyrene and carve the shape out of this block. You have probably seen this product used as insulation sheets mostly colored pink or light blue. This product is easy to cut, carve, plane, sand and can be easily shaped to whatever. This is sometimes the method used for making surfboards. I would not use expanded polystyrene, the structure of this (little balls) makes it hard to shape, Nor would I use Polyisocyanurate as it is to brittle, unless you can get high density stuff.
After I had the pattern shaped I would make a mold of the pattern, with the simplicity of this shape a 2 piece mold would most likely work.
Then after the mold is made it is just a matter of making your 2 pieces ( DON'T FORGET TO USE MOLD RELEASE )and fiberglassing them together.
You could purchase sheets of expanded polystyrene and glue them together to get your thickness or you could try to find a block of the product. Often they use blocks of this as flotation under docks and floating homes, I would suggest locating a dock builder or small shipyard.
Usually I would suggest that epoxy resin be used instead of polyester resin, however for this project polyester resin is fine and the beauty of it, it's cheap. Polyester resin will melt the foam however, so you might want to make the mold out of epoxy or wrap the mold in a "saran" type wrap and be prepared to do some sanding.
Good luck.
If I were to make this out of fiberglass, I would start out with a block of extruded polystyrene and carve the shape out of this block. You have probably seen this product used as insulation sheets mostly colored pink or light blue. This product is easy to cut, carve, plane, sand and can be easily shaped to whatever. This is sometimes the method used for making surfboards. I would not use expanded polystyrene, the structure of this (little balls) makes it hard to shape, Nor would I use Polyisocyanurate as it is to brittle, unless you can get high density stuff.
After I had the pattern shaped I would make a mold of the pattern, with the simplicity of this shape a 2 piece mold would most likely work.
Then after the mold is made it is just a matter of making your 2 pieces ( DON'T FORGET TO USE MOLD RELEASE )and fiberglassing them together.
You could purchase sheets of expanded polystyrene and glue them together to get your thickness or you could try to find a block of the product. Often they use blocks of this as flotation under docks and floating homes, I would suggest locating a dock builder or small shipyard.
Usually I would suggest that epoxy resin be used instead of polyester resin, however for this project polyester resin is fine and the beauty of it, it's cheap. Polyester resin will melt the foam however, so you might want to make the mold out of epoxy or wrap the mold in a "saran" type wrap and be prepared to do some sanding.
Good luck.
wow!
i never expected help to come this quickly! thank you!
about the seraphim, i already knew it, but once i did a loudspeaker made out of pipe and the sound of it was not very pleasing (sounded to much like a pipe
) anyway, thank you!
about the sketch, please don't take the proportions seriosly. the dimensions and the exact shape of the tube has still to be worked out.
brewboy, what a great idea! what you're sugesting is that i carve the counter-mould in EP, than make a mould out of it (2 pieces), and then work the fiberglass on it, from the "inside to the outside", correct? can you please direct me to any link where they explain the epoxy technic you mentioned? thanks a lot!
i never expected help to come this quickly! thank you!
about the seraphim, i already knew it, but once i did a loudspeaker made out of pipe and the sound of it was not very pleasing (sounded to much like a pipe
) anyway, thank you!about the sketch, please don't take the proportions seriosly. the dimensions and the exact shape of the tube has still to be worked out.
brewboy, what a great idea! what you're sugesting is that i carve the counter-mould in EP, than make a mould out of it (2 pieces), and then work the fiberglass on it, from the "inside to the outside", correct? can you please direct me to any link where they explain the epoxy technic you mentioned? thanks a lot!
When I saw your sketch first off it reminded me of an eeltrap.
You could do it in woven flexible reeds, then cover it in chopped strand mat and resin it up.
Although I believe that the reeds are now quite expensive in Western Europe, perhaps you could substitute something else.
I imagint the central pipe would need to be substantial too.
Regards
Ted
You could do it in woven flexible reeds, then cover it in chopped strand mat and resin it up.
Although I believe that the reeds are now quite expensive in Western Europe, perhaps you could substitute something else.
I imagint the central pipe would need to be substantial too.
Regards
Ted
There is a 2 part video on u-tube showing a guy making a mold out of a auto sub box
http://www.youtube.com/watch?v=gzng9HeGNPE&feature=related
Some of the things that he does, I would not, however it is a good example of how to make a mold. Working with epoxy is basically the same as polyester resin except the epoxy will not melt the EP. Several coats of paint or pva sealer might seal the foam enough to make the mold or like I said wrap the EP in saran wrap, but then you would have wrinkles that would have to be sanded out.
http://www.youtube.com/watch?v=gzng9HeGNPE&feature=related
Some of the things that he does, I would not, however it is a good example of how to make a mold. Working with epoxy is basically the same as polyester resin except the epoxy will not melt the EP. Several coats of paint or pva sealer might seal the foam enough to make the mold or like I said wrap the EP in saran wrap, but then you would have wrinkles that would have to be sanded out.
interesting...
in this case, since the pipe is simmetrical to it's axis, i suppose the best thing i should do is to divide the continuous pipe in sections, knowing that for each one i would make a mould (for the 2 halves of the pipe, since it's simmetrical), and in the end connect all the sections. do i loose structural rigidity with this?
another thing, i presume that all TLs must be tappered (correct me if I'm wrong), which means that for using the same mould for the outside on all the pipe (regular curves, repeated as a pattern), i could simple thick the wall (thus lowering the effective diammeter) as long as i would progress down on it, right?
another thing, slightly out of topic: even knowing that i'm thinking on the drivers said before (both L22 and W15CY from Seas, plus Vifa XT25), i keep some doubts about the tweeter. thinking on its really low distortion and terribly cheap cost (at least compared to the other 2) makes it tempting, but maybe i'm missing a better integration (Troels, for example, uses ribbon tweeters in this kind of setup).
last but not the least, what do you recommend me? active/passive(is this possible?) crossover before the amps, or passive crossover after the amp?
once again, sorry for all the newb questions! (and thanks a lot for all the patience and enlightening)
happy new year for all of you!
in this case, since the pipe is simmetrical to it's axis, i suppose the best thing i should do is to divide the continuous pipe in sections, knowing that for each one i would make a mould (for the 2 halves of the pipe, since it's simmetrical), and in the end connect all the sections. do i loose structural rigidity with this?
another thing, i presume that all TLs must be tappered (correct me if I'm wrong), which means that for using the same mould for the outside on all the pipe (regular curves, repeated as a pattern), i could simple thick the wall (thus lowering the effective diammeter) as long as i would progress down on it, right?
another thing, slightly out of topic: even knowing that i'm thinking on the drivers said before (both L22 and W15CY from Seas, plus Vifa XT25), i keep some doubts about the tweeter. thinking on its really low distortion and terribly cheap cost (at least compared to the other 2) makes it tempting, but maybe i'm missing a better integration (Troels, for example, uses ribbon tweeters in this kind of setup).
last but not the least, what do you recommend me? active/passive(is this possible?) crossover before the amps, or passive crossover after the amp?
once again, sorry for all the newb questions! (and thanks a lot for all the patience and enlightening)
happy new year for all of you!
hummm... 
i was counting with good luck in the usage of the fabulous Martin J. King's MathCad worksheets, but they are not available . after some email enlightment, and knowing what made this happen, i completely understand his position... even so, i'm in a complicated situation, since i was counting on this magnificent "open-source" resource to calculate my TL.
maybe this is a repetitive question, but since most indicate Martin's excellent work which is no longer available, do any one of you know of a free application that offers good quality for the calculation of a TL? i would prefer not using any stuffing (or few), having no troubles at all with the (big) lengh of the pipe.
to help you to help me (pleeeease), i can even give some T/S parameters of the driver i'm pretending to use (removed from thielesmall.com):
Seas L22RN4X/P
Vas: 72L (vs 93L according to SEAS???)
Qts: 0,32
Fs: 23Hz
Sd: 220 cm2 (according to SEAS)
if you're able and willing to help me (thank you!), and there are some values missing, i will try to get them!
thank you very much!
i was counting with good luck in the usage of the fabulous Martin J. King's MathCad worksheets, but they are not available
. after some email enlightment, and knowing what made this happen, i completely understand his position... even so, i'm in a complicated situation, since i was counting on this magnificent "open-source" resource to calculate my TL.maybe this is a repetitive question, but since most indicate Martin's excellent work which is no longer available, do any one of you know of a free application that offers good quality for the calculation of a TL? i would prefer not using any stuffing (or few), having no troubles at all with the (big) lengh of the pipe.
to help you to help me (pleeeease), i can even give some T/S parameters of the driver i'm pretending to use (removed from thielesmall.com):
Seas L22RN4X/P
Vas: 72L (vs 93L according to SEAS???)
Qts: 0,32
Fs: 23Hz
Sd: 220 cm2 (according to SEAS)
if you're able and willing to help me (thank you!), and there are some values missing, i will try to get them!
thank you very much!
still about the midrange-tweeter unit. after researching a lot about spherical enclosures (which I could eventually build) I ended up finding several models of Sound-E-Motion loudspeakers which come up to my desires.
my problem:
since I already decided on the midwoofer (Seas W15CY) and tweeter (Fountek JP3), I'm having quite some troubles on the crossover (which I do not have any idea on how to adapt). Troels Gravesen has a project in which he uses the "plastic" JP3 (NeoCD3) and this midwoofer, which could be a choice but, like in the picture seen, I'd prefer to use spherical boxes... this means no front baffle... this means a different crossover.
however, and please forgive me if I'm saying something stupid, the idea of building a crossover when there is no front baffle at all seems to be quite mathematical. I perfectly accept if it is not.
another choice would be the usage of an active crossover, but together with the bass box, this would become quite expensive on amplifiers (but maybe much better sounded).
what do you recomend me?
thank you!
my problem:
since I already decided on the midwoofer (Seas W15CY) and tweeter (Fountek JP3), I'm having quite some troubles on the crossover (which I do not have any idea on how to adapt). Troels Gravesen has a project in which he uses the "plastic" JP3 (NeoCD3) and this midwoofer, which could be a choice but, like in the picture seen, I'd prefer to use spherical boxes... this means no front baffle... this means a different crossover.
however, and please forgive me if I'm saying something stupid, the idea of building a crossover when there is no front baffle at all seems to be quite mathematical. I perfectly accept if it is not.
another choice would be the usage of an active crossover, but together with the bass box, this would become quite expensive on amplifiers (but maybe much better sounded).
what do you recomend me?
thank you!
If you can tell me accurately what the shape of your woofer enclosure is from beginning to end (its area along the line's length), I'd be happy to model its TL performance using Martin's software. That said, the fact that the woofer is located at the very beginning of the TL is not likely to have good results; the response curve above 100 Hz or so will likely have some serious dips and peaks in it.
sergiof said:hummm...
i was counting with good luck in the usage of the fabulous Martin J. King's MathCad worksheets, but they are not available
maybe this is a repetitive question, but since most indicate Martin's excellent work which is no longer available, do any one of you know of a free application that offers good quality for the calculation of a TL? i would prefer not using any stuffing (or few), having no troubles at all with the (big) lengh of the pipe.
to help you to help me (pleeeease), i can even give some T/S parameters of the driver i'm pretending to use (removed from thielesmall.com):
Seas L22RN4X/P
Vas: 72L (vs 93L according to SEAS???)
Qts: 0,32
Fs: 23Hz
Sd: 220 cm2 (according to SEAS)
if you're able and willing to help me (thank you!), and there are some values missing, i will try to get them!
thank you very much!
thank you very much pkitt, but your answer/question confused me (you see, I'm a terrible rookie on this). I know of different tapperings you may apply, and as far as I remember, the best possible would be a relationship 1.25:1, am I correct? this way, I'd start with 1.25xSd=275cm2 and end with Sd=220cm2.
The length of the tube would be to its Fs, which goes for 23Hz.
Woofer not on the tip of the pipe - got it! What do you suggest? On the 1/3rd of it?
Still about the crossover, I was thinking... What if instead of building a passive crossover for the midwoofer-tweeter unit I used a modified Behringer DCX2496, and 2 stereo amps? (plus a third one for the bassbox). I know it's expensive, but maybe it "compensates" the costs I would have on buying the passive crossover components (for the midwoofer-tweeter) plus the active crossover components (for the bassbox). The extra costs would be an extra stereo amplifier plus the DCX2496, plus its tweaks. Is it worth it?
About the "gasoline trick"... This would obly me to build to inner moulds in polysterene. I have no experience on it, so can you please tell me how easy it would be to sculpture two simmetrical pipes out of it?
The length of the tube would be to its Fs, which goes for 23Hz.
Woofer not on the tip of the pipe - got it! What do you suggest? On the 1/3rd of it?
Still about the crossover, I was thinking... What if instead of building a passive crossover for the midwoofer-tweeter unit I used a modified Behringer DCX2496, and 2 stereo amps? (plus a third one for the bassbox). I know it's expensive, but maybe it "compensates" the costs I would have on buying the passive crossover components (for the midwoofer-tweeter) plus the active crossover components (for the bassbox). The extra costs would be an extra stereo amplifier plus the DCX2496, plus its tweaks. Is it worth it?
About the "gasoline trick"... This would obly me to build to inner moulds in polysterene. I have no experience on it, so can you please tell me how easy it would be to sculpture two simmetrical pipes out of it?
There is no one "best" taper for all cases. In fact, using a taper much less than ~10:1 isn't worth the effort IMO. Tapering the line does two things; it allows the line to be much shorter in actual length because tapering it makes its effective length longer, and tapering also smooths out the response ripples immensely. This doesn't mean that your suggested taper won't give acceptable results, but from my experience, the results won't likely be anywhere near optimum. Also, depending on the woofer's Qts, the line's quarter-wavelength resonant frequency may need to be higher or lower than the woofer's Fs. As to the woofer's location along the line's length, making it 1/5 the way is a good starting part, then it can be moved as necessary in the modeling for optimum results.
sergiof said:
A tapered TL with driver at 1/5
Just for an example for you to peruse, I modeled a 10:1 tapered TL for the L22RN4X/P with the driver located at 1/5 of the line's length from the closed end of the line. This line is 58" long with a beginning area of 3Sd and and ending area of 0.3 Sd. I used Seas' published T/S values. I've attached a graph of the system response (the solid red line). Due to the low Qts, the system tuning frequency as established by the line's length and taper, seems to optimum for this driver at just above 30 Hz.
Just for an example for you to peruse, I modeled a 10:1 tapered TL for the L22RN4X/P with the driver located at 1/5 of the line's length from the closed end of the line. This line is 58" long with a beginning area of 3Sd and and ending area of 0.3 Sd. I used Seas' published T/S values. I've attached a graph of the system response (the solid red line). Due to the low Qts, the system tuning frequency as established by the line's length and taper, seems to optimum for this driver at just above 30 Hz.
sergiof said:hummm...
(ignorance is NOT a bliss) i think i need to read a lot...
anyway, if it helps, the Qts=0,32 (according to the Thiele Small website)
Attachments
The "other" L22
If I were choosing the driver for this TL, I would prefer to use the L22RNX/P because it models a bit better in a similar TL. I used the same 10:1 taper ratio as well as the 3Sd beginning and 0.3Sd ending areas; however, since this driver's Qts is higher, the line's 1/4-wavelength resonant frequency had to be lower. Consequently I made the line's length 66", resulting in a system tuning frequency of ~28 Hz. You can see the results in the attached graph. In case you're interested, I used a stuffing density of 1 lb/cu.ft. for all three of these example modelings, which would require stuffing approximately the first half of the lines' lengths.
If I were choosing the driver for this TL, I would prefer to use the L22RNX/P because it models a bit better in a similar TL. I used the same 10:1 taper ratio as well as the 3Sd beginning and 0.3Sd ending areas; however, since this driver's Qts is higher, the line's 1/4-wavelength resonant frequency had to be lower. Consequently I made the line's length 66", resulting in a system tuning frequency of ~28 Hz. You can see the results in the attached graph. In case you're interested, I used a stuffing density of 1 lb/cu.ft. for all three of these example modelings, which would require stuffing approximately the first half of the lines' lengths.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.