Easier said than done.
Car engines are designed to be disassembled and reassembled, down to the tiniest part, so they are held together by screws, bolts, pins, clamps, or simply click in.
You can open an engine, disassemble, replace a tiny worn bearing in its deepest area and fully reassemble it using all other parts ... and it will be exactly as before, except the replaced part.
Now speaker moving parts must meet 3 conditions: light, light and light, because they must move back and forth up to 20000 times a second, with minimal inertia, which means minimal mass.
You can simply NOT use bolts and nuts to fix parts together !!!!!!!!
So parts are made out of light flimsy components : paper, cardboard, cloth, foam, etc. and voice coils are wound on *thin* paper, Kapton, Nomex, best case paper thin aluminum, and are glued together.
Typical modern glue is Epoxy (very little of it) or some high temperature cyanoacrilate ... both stronger than the parts which they are holding together and, once hardened, they never ever redissolve or melt.
Trying to remove the voice coil means *destroying* it and usually damaging the cone beyond usability.
Being in Argentina, a few times I have repaired expensive speakers where no replacement parts are available, specially cones (original or aftermarket) by carefully grinding Epoxy out with a Dentist type tool, destroying the coil in the process and trying not to damage the cone ... not a job for the faint hearted, so I could fit a new coil (usually handwound by me).
In USA you would be better off by simply getting the proper speaker or at least, full reconing yours, since coil is glued to cone and spider, which are glued to frame , so in general all you can do is to cut them away from the frame with a hobby knife, clean residue, and install *all* new parts: full reconing.
I suggest you obtain a book dealing with the subject; Martin Colloms's High Performance Loudspeakers is a decent primer.
Baldly, VC inductance creates a nominal HF low-pass limit which in the case of wideband drivers may or may not be factored into the response (i.e. compensated for by the cone / sub-cone resonance). The VC mass affects efficiency, but this may not necessarily be a uniform impact with a wideband driver, which are resonant structures for most of their BW; this may depend on how & where it is coupled. Then you have the physical length of the coil, both relative to the VC gap, and also in establishing the centre of gravity which in turn may alter the response due to the differing loads, especially under dynamic conditions. The material of the coil, gauge & winding affect electrical behaviour in addition to the mass, which further impacts a variety of behaviours including FR & extension. And so on.
Baldly, VC inductance creates a nominal HF low-pass limit which in the case of wideband drivers may or may not be factored into the response (i.e. compensated for by the cone / sub-cone resonance). The VC mass affects efficiency, but this may not necessarily be a uniform impact with a wideband driver, which are resonant structures for most of their BW; this may depend on how & where it is coupled. Then you have the physical length of the coil, both relative to the VC gap, and also in establishing the centre of gravity which in turn may alter the response due to the differing loads, especially under dynamic conditions. The material of the coil, gauge & winding affect electrical behaviour in addition to the mass, which further impacts a variety of behaviours including FR & extension. And so on.
Voice coil has huge influence on speaker parameters:
a) moving mass: coils can be wound on different formers, from paper to kapton to aluminum to reinforced fiberglass.
Out of copper or aluminum wire.
Different wire diameters and winding length.
Round or ribbon.
All change weight and so Fr ... and many more parameters.
b) conductive/non conductive former has huge influence on damping.
c) winding turns and geometry directly influence series inductance, which in turn modifies impedance above, say, 1 kHz, and high frequency response.
d) shorter/longer winding directly influence X max, linearity and efficiency.
I design and build my own speakers, so all these factors (and many more) must be considered.
Hi Fi or Guitar?
Not sure about full kits (including frame and magnet) but there are a few sellers of aftermarket parts, you might get a couple dead speakers for peanuts or nothing, clean and measure them to get needed cone depth, suspension diameter, voice coil gap which is very important because tolerance there is measured in thousandths of an inch, not kidding, and start experimenting.
Copper ribbon voice coils are not very common, a couple Pro woofers use them, think JBL and EV.
Not sure about full kits (including frame and magnet) but there are a few sellers of aftermarket parts, you might get a couple dead speakers for peanuts or nothing, clean and measure them to get needed cone depth, suspension diameter, voice coil gap which is very important because tolerance there is measured in thousandths of an inch, not kidding, and start experimenting.
Copper ribbon voice coils are not very common, a couple Pro woofers use them, think JBL and EV.
HI-FI.
My goal is to make the driver impedance about 100 miliohms or less.
I'm curious to know what the FR and performance characteristics of a near dead short driver is.
I've always been fascinated by current drive.
Copper ribbon seems to be the easiest way to achieve this. I just need to find a way to build a driver or replace the coil in an existing one.
My goal is to make the driver impedance about 100 miliohms or less.
I'm curious to know what the FR and performance characteristics of a near dead short driver is.
I've always been fascinated by current drive.
Copper ribbon seems to be the easiest way to achieve this. I just need to find a way to build a driver or replace the coil in an existing one.
Why such a low value, and you are aware that impedance on moving coil designs is dynamic with the signal content, and depending on how you measure / rate the nominal figure can dip significantly below that ? Will you have an amplifier comfortable operating at the current levels that less than 1 ohm implies?
If this was a really good idea, I’d imagine that major players would have long since had entered the game.
If this was a really good idea, I’d imagine that major players would have long since had entered the game.
Because with a near dead short, voltage amplification is not required in the system and capacitive parasitics become a non existent issue. Also the voltage axis of device linearity becomes practically irrelevant.
Yeah but I have no idea what causes the shift in impedance. I'm guessing maybe if I use a single turn of thick copper foil as the VC then the inductance will be so low that the impedance shift will be minimal?
Yes.
We ( you) want Joule !
Or you should shift to something new, not magneto-dynamic. Something like this USound - Technology
Or you should shift to something new, not magneto-dynamic. Something like this USound - Technology
Oh trust me I have a lot of ideas for unique driver tech. Most of which is at least loosely based on proven concepts. But I can only do one thing at a time and I have a large amount of things to do on my sequential list of R&D.
Currently I'm just wondering if dynamic driver design belongs on that list.
IMO current drive is superior to voltage drive and as far as dynamic drivers go it seems like an obvious step to replace the long thin winding of a VC with a short, wide, thick VC.
Of course this presumption is based upon me knowing next to nothing about how the VC effects FR and impedance and such.
Which is why I'm asking people in-the-know how such a driver might perform.
Currently I'm just wondering if dynamic driver design belongs on that list.
IMO current drive is superior to voltage drive and as far as dynamic drivers go it seems like an obvious step to replace the long thin winding of a VC with a short, wide, thick VC.
Of course this presumption is based upon me knowing next to nothing about how the VC effects FR and impedance and such.
Which is why I'm asking people in-the-know how such a driver might perform.
It's very hard to find commercial midrange and high frequency drivers below 4 Ohms nominal. You could use half that fact to argue there's probably no performance benefit in going lower, and the other half to argue it's because there are almost no commercial hi-fi amps available that would perform best below that.
The single turn thing is easy to reject just on the basis of physical problems in getting symmetrical coil field with possible lead in/out, athough it might be handled well enough if just a few turns are allowed.
You can rely on basic equations to determine how BL runs agains inductance. Without any calculations you could see that at some point passive crossover design gets weird with tiny inductors and giant capacitors, or that all the benefit quickly goes to pot unless the amplifier output is strapped directly to the coil. An 100 Amp, 500 mV amplifier seems hard to make with better specs than a normal amp but if you did manage to solve it then at least you'd have a product you could sell to the ribbon tweeter crowd.
Maybe you could try rewinding some tweeters with replaceable diaphrams and see where it goes.
The single turn thing is easy to reject just on the basis of physical problems in getting symmetrical coil field with possible lead in/out, athough it might be handled well enough if just a few turns are allowed.
You can rely on basic equations to determine how BL runs agains inductance. Without any calculations you could see that at some point passive crossover design gets weird with tiny inductors and giant capacitors, or that all the benefit quickly goes to pot unless the amplifier output is strapped directly to the coil. An 100 Amp, 500 mV amplifier seems hard to make with better specs than a normal amp but if you did manage to solve it then at least you'd have a product you could sell to the ribbon tweeter crowd.
Maybe you could try rewinding some tweeters with replaceable diaphrams and see where it goes.
Oh yes it is, just to a lower voltage than before, say 1V instead of 20V, but in any case , what is the problem with voltage amplification?
Any signal you feed your amp already went through a dozen voltage gain stages, from microphone to your preamp output.
They are already irrelevant in an 8 ohm speaker.
No, it´s exactly as before, speaker will behave the exact same way: will have a resonant frequency, an impedance peak there, minimum impedance point, practically raw DCR somewhere between 200 and 400 Hz, and rising impedance above that, due to coil parasitic inductance.
Only "everything impedance related will be scaled down" , as, say, instead of 6.5 ohm DCR you will have 0.1 ohm , instead of 40 ohm at resonance you will have 0.7 ohm, impedance at 10 kHz will rise to 0.2 ohm instead of 16 ohm and so on.
Basic behaviour and sound will be the same.
Notice speaker parameters; Fr, Q, Vas, etc. , do NOT include DCR in the equations.
Inductive behaviour, or to be more precise, reactive behaviour, both electrical and mechanical.
No, percentage wise change will be the same as before.
Impedance curves will have the same shape, but smaller numbers written on the vertical scale.
As I said above: "roughly the same but scaled down".
Agree that current drive changes speaker behaviour big time, but it will do so in 8 ohm speakers and 0.1 ohmones.
In fact I *do* use lots of current drive in my Guitar and Bass amplifiers, go figure.
Last edited:
"Bose had car systems with 1 ohm voice coils and amps to drive them "
Not so silly when you are only looking for automotive grade performance and can get rid of a switching converter. Their 901 basically ran them all in series to get around needing a special amp. How significant is it that this arrangement was throughly tried but never generically took off in commercial hi-fi?
Not so silly when you are only looking for automotive grade performance and can get rid of a switching converter. Their 901 basically ran them all in series to get around needing a special amp. How significant is it that this arrangement was throughly tried but never generically took off in commercial hi-fi?
Good point. A few turns would be necessary.
Still though. How would such a configuration perform compared to normal configurations?
JMFahey said coil length and geometry have a large effect on the drive response, but how?
Eew, why would anyone use a passive crossover?
My crosovers would be at the source.
Yes I have made such an amp.
Performance exceeds typical amp specs. Not quite capable of 100A though. What transducer possibly needs 100A running through it? Talk about an inefficient driver.
I'm currently finishing my work on another dac/amp that I intend to use as a base to start a business off of. It has the best sound and performance I have ever seen or experienced so it is wasteful not to do so.
But as soon as I've got my business up and running I'm going to dive deeper into my R&D of "zero" ohm drivers and other experimental driver tech.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.