Firstly, I am not too clued up about speakers, so please reply in understandable language if that is remotely possible.
I have been looking around on the weg for some nice speaker designs.
I am realy pleased with what I heard about TQWT's (Tapered Quater Wave Tube), but I always like to make things a little more personal, rather than just coppying designs.
I don't like the look of the straight pipes, they're too bulky.
But the folded pipes realy appeal to me, and I am going to build a pair anyway.
During my varios serches, I came accross B&W's TL (Transmission Line) design called The Nautilus, because it is coiled up in a spiral, like the Natural spiral of a Nautilus Shell.
This gave me the idea of making a coiled TQWT.
It would be similar to B&W's Nautilus, only I would put the driver, somewear along the middle of the tube, rather that in the opening at the end.
As I said I'm not exactly an accomplished speaker designer, and perhaps I'm missing some flaw in the design.
Also, to do a propper job, I am going to have to learn a bit of the theory behind the TQWT.
Can anyone give me any advice, or direct me as to where to find some. I think with some help, I could design some great speakers.
I have been looking around on the weg for some nice speaker designs.
I am realy pleased with what I heard about TQWT's (Tapered Quater Wave Tube), but I always like to make things a little more personal, rather than just coppying designs.
I don't like the look of the straight pipes, they're too bulky.
But the folded pipes realy appeal to me, and I am going to build a pair anyway.
During my varios serches, I came accross B&W's TL (Transmission Line) design called The Nautilus, because it is coiled up in a spiral, like the Natural spiral of a Nautilus Shell.
This gave me the idea of making a coiled TQWT.
It would be similar to B&W's Nautilus, only I would put the driver, somewear along the middle of the tube, rather that in the opening at the end.
As I said I'm not exactly an accomplished speaker designer, and perhaps I'm missing some flaw in the design.
Also, to do a propper job, I am going to have to learn a bit of the theory behind the TQWT.
Can anyone give me any advice, or direct me as to where to find some. I think with some help, I could design some great speakers.
Interesting idea you have there. Look at Martin King's website if you haven't already for everything you'll ever need to design a killer TL of any description, but particularly ML (Mass Loaded) TQWT from the man who cracked the mysteries of the transmission line. You'll find it here:
www.quarter-wave.com
Download MathCad Explorer 8 and Martin's TL worksheets, and play around. They look frightening, but thy're actually very easy -remember, Martin's done all the difficult Math. You just need to plug in the T/S driver parameters and then play around with the dimensions of the enclosure you want to build until you reach a satisfactory frequency response.
Also look at Bob Brines' site at http://www.geocities.com/rbrines1/
for some excellent articles on designing TLs using Martin's software.
Beware, or at least be deeply suspicious of most of the earlier works (as in pre-Martin) written about TLs -not that's there's much, because no-one really knew how they worked -it's all rule of thumb, trial and error tuning, and 99.99% of the theories are completely wrong. Lynn Olson's 2 way Ariel is one of the very few really good TLs that existed before Martin's work. For historical interest, generally the best that was possible using the old-fashioned way was that the line-length used to be tuned to the Fs of the driver, the driver being mounted 1/3 of the way along the line. It worked passably, with some hefty tweaking, but you can do a whole lot better now!
Best
Scott.
www.quarter-wave.com
Download MathCad Explorer 8 and Martin's TL worksheets, and play around. They look frightening, but thy're actually very easy -remember, Martin's done all the difficult Math. You just need to plug in the T/S driver parameters and then play around with the dimensions of the enclosure you want to build until you reach a satisfactory frequency response.
Also look at Bob Brines' site at http://www.geocities.com/rbrines1/
for some excellent articles on designing TLs using Martin's software.
Beware, or at least be deeply suspicious of most of the earlier works (as in pre-Martin) written about TLs -not that's there's much, because no-one really knew how they worked -it's all rule of thumb, trial and error tuning, and 99.99% of the theories are completely wrong. Lynn Olson's 2 way Ariel is one of the very few really good TLs that existed before Martin's work. For historical interest, generally the best that was possible using the old-fashioned way was that the line-length used to be tuned to the Fs of the driver, the driver being mounted 1/3 of the way along the line. It worked passably, with some hefty tweaking, but you can do a whole lot better now!
Best
Scott.
There's a very good copy of the nautilus here:
http://www.pcpaudio.com/pcpfiles/proyectos_altavoces/Odisea1/Odisea1.html
Gives a good idea to construction techniques required.
http://www.pcpaudio.com/pcpfiles/proyectos_altavoces/Odisea1/Odisea1.html
Gives a good idea to construction techniques required.
inertial said:ShinOBIWAN,
IMO, This is FAR from the original Technical concept by L. Dickie.
Anyway it is very nice .
Cheers
Agreed, I was talking from merely a construction point of view.
That design linked to has non of the internal labyrinths to cancel out rear wave dispertion from the mids and tweet, like in the B&W original.
Still its a looker.
Thanks Guys
I was realy looking for advice on the theory side of things.
Construction is never a problem for me.
For this particular project, I'm not too worried about how it looks, and I think I can get a good compact, aesthetic design, if I make the laberynth itself out of polystyrene, and cast concrete, or ceramic around it, and then disolve out the polystyrene with a suitable solvent.
However this idea could change when the time comes. But I'm not too worried about that sort of thing yet. Construction comes after design.
I was just not sure weather it would actualy work. perhaps the straight or folded tubes have an esential characteristic which a coil doesn't. Also I'm not too interested in TL, I want to make a Voigt Pipe, where the driver sits in the middle of the tube, not at the end.
What are the essential factors, for TQWT?
Total Volume?
Length?
Area of Cross Section?
How do I convert A Straight pipe into a coil with the same characteristics.
Another thaught I had, was to make the cross Section of the tuge Pentagonal, rather than round or rectangular. This would eliminate any parallel sides.
Thanks.
-Raphael
I was realy looking for advice on the theory side of things.
Construction is never a problem for me.
For this particular project, I'm not too worried about how it looks, and I think I can get a good compact, aesthetic design, if I make the laberynth itself out of polystyrene, and cast concrete, or ceramic around it, and then disolve out the polystyrene with a suitable solvent.
However this idea could change when the time comes. But I'm not too worried about that sort of thing yet. Construction comes after design.
I was just not sure weather it would actualy work. perhaps the straight or folded tubes have an esential characteristic which a coil doesn't. Also I'm not too interested in TL, I want to make a Voigt Pipe, where the driver sits in the middle of the tube, not at the end.
What are the essential factors, for TQWT?
Total Volume?
Length?
Area of Cross Section?
How do I convert A Straight pipe into a coil with the same characteristics.
Another thaught I had, was to make the cross Section of the tuge Pentagonal, rather than round or rectangular. This would eliminate any parallel sides.
Thanks.
-Raphael
Re: Thanks Guys
Dealing with the last point first; remember what Siegfried Linkwitz says regarding room acoustics: "Some people think that by making the room other than rectangular or using curved surfaces, that they can eliminate standing waves. They merely change frequencies, shift their distribution and make their calculation a lot more difficult." The same principle also applies to a loudspeaker cabinet. Non-parallel sides can work, but they are not a panacea, and can often be more trouble than they're worth. Also, remember that you've already coiled the thing anyway -anything else is going to be overkill.
A TL of any type is probably the hardest of all cabinets to design, apart from a horn. The best way forward is usually the Mass-Loading technique pioneered by Martin King. I would suggest that you model an optimum straight pipe, then simply coil it up, keeping all the measurements -length, areas, driver position etc the same. Don't worry -coiling the thing up won't affect the frequency response, providing you don't alter the vital dimensions (see below).
There are numerous factors which need to be taken together, but, initially, you need to select an appropriate driver for pipe loading. I think the best thing to do here is point you to the different, easily digestable articles written by Bob Brines on TLs of all their different varieties (highly recommended), and how to approach designing them using Martin's MathCad sheets. This one is a good startinging point, and there are numerous others on his site. They help make Martin's MathCad sheets and the best way to approach and use them much, much easier. (Oh, by the way, I depart from Bob's view on only one point: that the straight and conventional taper TLs always give better results than a TQWT: that depends upon what you want in the first place, and the circumstances. For example, I have found a TQWT loaded driver can perform better than one loaded by the conventional or straight types in a larger than average room.)
http://www.geocities.com/rbrines1/Pages/Design_Procedure/Small_TL.html
Once you've selected a driver suitable, you need to decide upon an approximate cut-off point. With this done, you need to consider:
1) Length of the pipe
2) The area of the top of the pipe (sealed end)
3) The area of the bottom of the pipe (open end)
4) The area and length of the port
5) The position of the driver in the pipe
The trick is to ballance all of these factors. Martin's mass-loading technique basically adds mass to the open end of the pipe by restricting it's size. If you look at his first full-range driver project on his site -a ML TQWT using the Fostex FE164, you'll notice that by blocking off the open end, except for a 3" x 1" port, the cut-off point of the pipe plunges from around 70Hz to 35Hz. Not bad!
Bob goes into all this in easily understandable detail in his articles; a few general design guidelines here though:
1) Don't try making too-long a pipe. As Terry Cain points out, Bigger is Better, and longer is stronger, but unless you know exactly what you're doing, this can backfire, and it's only really necessary when using single driver systems.
2) Never use a top pipe area (sealed end) less than the surface area of the driver you are loading the pipe with. In some very specific circumstances a smaller top-pipe area might be used, but this is caused by a pipe / driver mis-match; i.e. the driver is not really suitable for TL loading. beware the 'Classic Voigt Pipe'. Voigt never advocated using a top that comes to a point, and for good reason.
3) A larger open end area can give more bass, but if you go too far you'll get or increase a null elsewhere.
Hope some of this helps.
Raphael Shaw said:
Dealing with the last point first; remember what Siegfried Linkwitz says regarding room acoustics: "Some people think that by making the room other than rectangular or using curved surfaces, that they can eliminate standing waves. They merely change frequencies, shift their distribution and make their calculation a lot more difficult." The same principle also applies to a loudspeaker cabinet. Non-parallel sides can work, but they are not a panacea, and can often be more trouble than they're worth. Also, remember that you've already coiled the thing anyway -anything else is going to be overkill.
A TL of any type is probably the hardest of all cabinets to design, apart from a horn. The best way forward is usually the Mass-Loading technique pioneered by Martin King. I would suggest that you model an optimum straight pipe, then simply coil it up, keeping all the measurements -length, areas, driver position etc the same. Don't worry -coiling the thing up won't affect the frequency response, providing you don't alter the vital dimensions (see below).
There are numerous factors which need to be taken together, but, initially, you need to select an appropriate driver for pipe loading. I think the best thing to do here is point you to the different, easily digestable articles written by Bob Brines on TLs of all their different varieties (highly recommended), and how to approach designing them using Martin's MathCad sheets. This one is a good startinging point, and there are numerous others on his site. They help make Martin's MathCad sheets and the best way to approach and use them much, much easier. (Oh, by the way, I depart from Bob's view on only one point: that the straight and conventional taper TLs always give better results than a TQWT: that depends upon what you want in the first place, and the circumstances. For example, I have found a TQWT loaded driver can perform better than one loaded by the conventional or straight types in a larger than average room.)
http://www.geocities.com/rbrines1/Pages/Design_Procedure/Small_TL.html
Once you've selected a driver
1) Length of the pipe
2) The area of the top of the pipe (sealed end)
3) The area of the bottom of the pipe (open end)
4) The area and length of the port
5) The position of the driver in the pipe
The trick is to ballance all of these factors. Martin's mass-loading technique basically adds mass to the open end of the pipe by restricting it's size. If you look at his first full-range driver project on his site -a ML TQWT using the Fostex FE164, you'll notice that by blocking off the open end, except for a 3" x 1" port, the cut-off point of the pipe plunges from around 70Hz to 35Hz. Not bad!
Bob goes into all this in easily understandable detail in his articles; a few general design guidelines here though:
1) Don't try making too-long a pipe. As Terry Cain points out, Bigger is Better, and longer is stronger, but unless you know exactly what you're doing, this can backfire, and it's only really necessary when using single driver systems.
2) Never use a top pipe area (sealed end) less than the surface area of the driver you are loading the pipe with. In some very specific circumstances a smaller top-pipe area might be used, but this is caused by a pipe / driver mis-match; i.e. the driver is not really suitable for TL loading. beware the 'Classic Voigt Pipe'. Voigt never advocated using a top that comes to a point, and for good reason.
3) A larger open end area can give more bass, but if you go too far you'll get or increase a null elsewhere.
Hope some of this helps.
Hey, cool pic.
Thanks.
its inspiring to get input and support like i've found here.
Was the driver going in the mouth?
Why the horn shape at the mouth?
... just out of interest.
I'm afraid I've been on holiday, and I'm about to write some exams, so time are few. But I'll be on the design as much as I can.
Thanks.
its inspiring to get input and support like i've found here.
Was the driver going in the mouth?
Why the horn shape at the mouth?
... just out of interest.
I'm afraid I've been on holiday, and I'm about to write some exams, so time are few. But I'll be on the design as much as I can.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.