And it still impresses me every time I look at the curved walls of my acoustic guitars. 🙂
One of them has a cutaway body, so the body walls have to curve quite tightly in that region. Amazing.
As far as I know, guitar walls are made with special "bendy plywood". Instead of alternating veneers oriented cross-grain to each other, all veneers have the grain in the same direction. That makes it possible to bend this type of wood, but only in one direction.
Long ago, I built balsa model aircraft, and you can curve a thin sheet of cross-grain balsa like that, particularly if you soak it in diluted ammonia first (Windex works).
For the loudspeaker enclosure, I would be tempted to use some other material, not wood. Maybe heavy cardboard, curved to shape and then given a coat of epoxy resin?
I know, DIY loudspeaker purists will be upset. But a curve like that is a very strong shape, and it's supported at both ends where it joins the sidewalls.
-Gnobuddy
One of them has a cutaway body, so the body walls have to curve quite tightly in that region. Amazing.
As far as I know, guitar walls are made with special "bendy plywood". Instead of alternating veneers oriented cross-grain to each other, all veneers have the grain in the same direction. That makes it possible to bend this type of wood, but only in one direction.
Long ago, I built balsa model aircraft, and you can curve a thin sheet of cross-grain balsa like that, particularly if you soak it in diluted ammonia first (Windex works).
For the loudspeaker enclosure, I would be tempted to use some other material, not wood. Maybe heavy cardboard, curved to shape and then given a coat of epoxy resin?
I know, DIY loudspeaker purists will be upset. But a curve like that is a very strong shape, and it's supported at both ends where it joins the sidewalls.
-Gnobuddy
It's actually a really small enclosure, for a 40mm driver (that's "mounted magnet out" for an omnidirectional dispersion), the curve was simply done on a bandsaw, if I wanted to make a bigger one, I'd probably just cut a few curved bits of wood and glue them together. I've just received a pair of Visason BG17s and trying to decide on the best encloser, we seem to be going off the topic of foil cables, perhaps there's enough interest in this driver for a separate thread?
When I said my friend bends wood for his speaker and amp cabinets, I meant he BENDS them 😱 , I was amazed.
Last Saturday I was in an Entrepreneur Fair with him, and he had an open unfinished cabinet to show construction, I forgot to take a picture, but here are the finished ones:
Mind you, it´s 12/15 mm plywood! 😱
looking at it again, it might be solid wood instead of plywood 😱
Doubly amazing 🙂
Last night I finally got around to tinkering a bit more with these, based on car drivers (to flatten the frequency response I had previously used an interconnect that reduced everything bellow 10,000 Hz, but I made a new on that was more pronounced), and I spent the rest of the night reviling in what these speakers could do, admittedly used quite far from how the manufacturer originally intended. The don't do transients, but then omnis don't anyway, so nothing is lost, but as used, they have (almost) no overhang, (4.5kg of iron bolted to the magnet, a double wall upper box mechanically isolated from the lower helps) so what transients are there aren't hidden in mush. More "horses for courses", using each driver in a way that shows its strengths, and hides its weaknesses.Have anyone experienced the following design of Silversmith Fidelium Speaker Cables?
https://silversmithaudio.com/fidelium-cables/
https://silversmithaudio.com/fidelium-cables/
Bend layers of 6mm plywood & bond together.
The further from the neutral axis, the greater the change in length.
Limiting the thickness makes it do-able.
the pioneers of bent or formed plywood for furniture were Aalto, Eames and Saarinen. nice stuff
Back in the day Richard Marsh described a foil umbilical for a moving coil pre preamp that he designed to be a high capacitance/ low inductance cable ideal for transmitting power from a regulated supply. He also suggested thin flat cable taped to the chassis for ground conductors.
Back in the day Richard Marsh described a foil umbilical for a moving coil pre preamp that he designed to be a high capacitance/ low inductance cable ideal for transmitting power from a regulated supply. He also suggested thin flat cable taped to the chassis for ground conductors.
The thin flat cable idea is derivative of published analysis based on RF characteristic impedance. However, the usual equations for that (which can be derived from Maxwell's Equations) presumes the model of a plane wave impinging on a cylindrical conductor. The problem is that at lower audio frequencies the model, and thus the usual equations do not give correct impedance values (because conductor thickness is not large compared to skin depth at LF). Instead, Bessel functions are needed. Or alternatively, characteristic impedance can simply be measured at different audio frequencies, and the measured numbers used. Other than that, material use to construct cables can have some influence on the sound. Things like electrostriction, magnetostriction, proximity effect, skin effect, dielectric absorption, triboelectric effect, connector materials, etc., may have small numerical values, but in some cases may influence sound.
Last edited:
But one wonders, why would the Radio Frequency Characteristic Impedance of a loudspeaker cable matter at audio frequencies?
There is some work done on that. There is the effect on amplifier stability, of course (Bateman).
There was also some experimental work done in a thread called "zip cord for speaker test"
Hans Polak measured its impedance with a VNA, then used lumped compensation to lower the impedance at low frequencies. It did have some subjective beneficial effect on the sound for everyone who tried it. IIRC, something about stored energy in the cable due to the mismatch at LF causing inefficient power transfer. Settling time was too long. Don't know how much it had to do with the amplifier's interaction with the reflections.
Also, IIRC there is an old AES paper on cables which describes characteristic impedance. Some cable designers have relied on that article only to produce cables that sound bad. Of course, the article didn't take into account the equation inaccuracy at LF.
There was also some experimental work done in a thread called "zip cord for speaker test"
Hans Polak measured its impedance with a VNA, then used lumped compensation to lower the impedance at low frequencies. It did have some subjective beneficial effect on the sound for everyone who tried it. IIRC, something about stored energy in the cable due to the mismatch at LF causing inefficient power transfer. Settling time was too long. Don't know how much it had to do with the amplifier's interaction with the reflections.
Also, IIRC there is an old AES paper on cables which describes characteristic impedance. Some cable designers have relied on that article only to produce cables that sound bad. Of course, the article didn't take into account the equation inaccuracy at LF.
Last edited:
Yep, he found that with rare combinations of power amp, speaker cable & length and loudspeaker, it is possible to have radio frequency ringing or oscillation.
But it is more about the amp than the cable.
Sure, but the amp doesn't blow up until you try to use the cable. Didn't NP say the best speaker cable is six-inches of Romex? Probably true. But if taking a piece of zip cord and lump compensating its impedance at LF to affect the sound, then maybe the amplifier is also implicated in the result.
It's the amps problem not the cable's or the loudspeaker's. The cable is just a small part of a radio frequency tuned circuit.
No such thing as "taking a piece of zip cord and lump compensating its impedance at LF" of a loudspeaker cable.
Even compensating a 100 mile long cable at audio frequencies is real messy.
* * * * * * * * * * * * * * * *
You need to read the Jim Brown paper"
Transmission Lines at Audio Frequencies, and a Bit of History
http://www.audiosystemsgroup.com/TransLines-LowFreq.pdf
No such thing as "taking a piece of zip cord and lump compensating its impedance at LF" of a loudspeaker cable.
Even compensating a 100 mile long cable at audio frequencies is real messy.
* * * * * * * * * * * * * * * *
You need to read the Jim Brown paper"
Transmission Lines at Audio Frequencies, and a Bit of History
http://www.audiosystemsgroup.com/TransLines-LowFreq.pdf