So I went and bought the DIY Perks plans for those QWTL speakers he made. I have access to a great 3d printer in work which means I can do insanely long prints like these.
Got most of the way through printing one when I thought, "This doesn't really take advantage of a 3D printer does it"
So here's my concept for a QWTL that would need to be 3D printed I might test with these little tangband woofers.
After the first fold the transmission line is annular. In this model the cross sectional area of the TL is maintained constant with each fold.
Anyway, I'm a noob, so before I spend a bunch on filament to print this thing, it'd be great if someone can tell me why this won't work. Are there particular considerations for annular ports? what do they do to sound?
Also if someone knows what software might be best for simulating a shape like this I'd be much obliged. Could hornresp do a useful job in a case like this?
Got most of the way through printing one when I thought, "This doesn't really take advantage of a 3D printer does it"
So here's my concept for a QWTL that would need to be 3D printed I might test with these little tangband woofers.
After the first fold the transmission line is annular. In this model the cross sectional area of the TL is maintained constant with each fold.
Anyway, I'm a noob, so before I spend a bunch on filament to print this thing, it'd be great if someone can tell me why this won't work. Are there particular considerations for annular ports? what do they do to sound?
Also if someone knows what software might be best for simulating a shape like this I'd be much obliged. Could hornresp do a useful job in a case like this?
Use Vtc/Atc and Ap1/Lp to describe that ‘chamber/big tube’ shape and then the s1-s2/L12 as the long skinny slow tapered ‘port/tube’?
then adjust ‘path’ to the distance Between the vent and driver outputs
then adjust ‘path’ to the distance Between the vent and driver outputs
This looks like an end-loaded TL - one of the worst TL configurations for full-range use, unless the taper is REALLY large, driving that troublesome 1st resonance high enough for it to be well damped by any internal damping in the path. The design shown also wastes a LOT of cabinet volume.
Still, who knows, it might actually sound pretty decent.
Still, who knows, it might actually sound pretty decent.
Folded concentric TL. I would (did) make the annulus the input and the central axis the output, of highly tapered line. Mine was actually an umbrella stand (tall bucket) with concentric pipes forming the 3-fold line. My drivers were fiberglass honeycomb 8" firing upward and 4" widerange of the very same make, XO-less, each resonance-suppressed using automotive "plast" stretchy tape. (Search "LXmidi" for pics.)
addendum: https://www.diyaudio.com/community/threads/1-order-crossover-help.391332/post-7151405
addendum: https://www.diyaudio.com/community/threads/1-order-crossover-help.391332/post-7151405
Last edited:
Cool thanks! I'll have a look at that math. Right now the total path length is simply quarter wavelength of the free air resonance of this particular driver.
Believe it or not, the skinny tubes aren't tapering here, but I would add a taper I think. If we sliced it across the front you'd see each fold is an annulus/ring shape. Here the cross sectional area of any annulus is the same cross sectional area as the 'chamber/big tube", they get thinner as the radius gets bigger to compensate.
It definitely does waste space! Cylindrical is a lot better.
Yeah, I need to parameterise it better next so that it'll solve for different tapers.
Probably wouldn't go full range with this, but maybe if I can sort it out.
Was more wondering if anyone thought there were different considerations for designing an annular as opposed to cylindrical or prismatic shaped TL
Is there any particular reason you would choose the annulus as the input and the tube the output?
You already alluded to the reason: tapered line and space-vs-material favor large-area annulus as input, beginning with the driver cone of course. Tapering the TL increases the effective line length hence lowers the quarterwave frequency, by quite a lot (30-50%). I wrote about this (Martin King's work) in a couple of threads: TLonken 17L 1.5m 8" Supravox and 3.75L 1m 2" Markaudio.
Dave?:
Looks cool, but full of lots of resonances like all ‘expanding path TL’ ?
Looks cool, but full of lots of resonances like all ‘expanding path TL’ ?
Last edited by a moderator:
The ‘tapered shape’ further seperates the fundametal from the rest of the pipe resonance junk you’ll want to get rid of. it’ll be much easier if it’s HF instead .
it’s a win/win I think?
Any undamped pipe will have resonances. In a typical TL they ar edealt with using damping driver offset, mass loading, other tricks.
In the Woden horn shown there are a number of low pass filters (the initial throat opening, and every bend in the pipe) to eliminate what is not wanted from coming out the mouth.
dave
In the Woden horn shown there are a number of low pass filters (the initial throat opening, and every bend in the pipe) to eliminate what is not wanted from coming out the mouth.
dave
Offsetting the driver is usually used to cancel the first harmonic. This will decrease the fundemental slighly but since you can now use less damping you end up with more.
An end loaded line will need to be heavily stuffed on most cases. The microTower is an example, end loaded since otherwise the drivers would be at knee height.
Speaker design means juggling a whole bumch of compromises.
dave
An end loaded line will need to be heavily stuffed on most cases. The microTower is an example, end loaded since otherwise the drivers would be at knee height.
Speaker design means juggling a whole bumch of compromises.
dave
I suppose in ‘lossless’ mode in horn response or math cad models a lot of these look really bad compared to what they actually ‘sound’ like.