I built it 2 years ago, and we did it as an experiment. We did not take any pictures.
Imagine a rats nest of cable around a short piece of 12 mm pvc pipe, a piece of paper i held in my hand with a small "blob" of Nd magnets glued with superglue. An old Mission Cyrus 1 amp and a 9 watt 10 ohm resistor making a small burned mark on my table. My son was playing Skinny Puppy through spotify on my phone connected to the amp.
The pipe was later transformed in to an electric railgun. It could shoot a 8 x 15 mm Nd magnet several meters when suddenly connected to a 9 volt battery.
This was much more fun for the kids then a really bad sounding crude loudspeaker-prototype.
Cheers,
Johannes
Imagine a rats nest of cable around a short piece of 12 mm pvc pipe, a piece of paper i held in my hand with a small "blob" of Nd magnets glued with superglue. An old Mission Cyrus 1 amp and a 9 watt 10 ohm resistor making a small burned mark on my table. My son was playing Skinny Puppy through spotify on my phone connected to the amp.
The pipe was later transformed in to an electric railgun. It could shoot a 8 x 15 mm Nd magnet several meters when suddenly connected to a 9 volt battery.
This was much more fun for the kids then a really bad sounding crude loudspeaker-prototype.
Cheers,
Johannes
OK, we have widescale production of moving coil loudspeakers, including semi-exotics like Magnepan [and competition?] planar magnetic drivers. But it's still a moving coil speaker.
Just above this message, was discussed a moving magnet speaker. Whatever their merits, you still need a permanent magnet. Or do you?
How about a dynamic driver that uses no permanent magnet at all? Surely this has been done; I think I recall that it was used as a space and money saving trick in an old tube radio (driver had an electromagnet that doubles as an Inductor?). Clearly, I know almost nothing about electromagnetism. However, from a practical standpoint, if you lose the magnets, then you have a much lighter (and perhaps cheaper) structure. Yes, you still need windings of wire, perhaps a lot of it, to make the electromagnet. There must be trade-offs.
My wild poorly-informed brainstorming such as this, is counter-balanced by the knowledge that this is hardly new science, if the vacuum tube is > 100 years old now, then surely the magnet-free speaker has been experimented with before?
Tell me more! I know that Jim Winey built his Magnepan because he wanted to get the sound of electrostatics but using conventional amplifiers. Well, he only went part-way. To make a complete analogy, he would have needed an electromagnet instead of PM.
OK, what's the downside? An electromagnet requiring huge amounts of DC current? An armature winding requiring the annual copper output of Bolivia? What 🙂
Just above this message, was discussed a moving magnet speaker. Whatever their merits, you still need a permanent magnet. Or do you?
How about a dynamic driver that uses no permanent magnet at all? Surely this has been done; I think I recall that it was used as a space and money saving trick in an old tube radio (driver had an electromagnet that doubles as an Inductor?). Clearly, I know almost nothing about electromagnetism. However, from a practical standpoint, if you lose the magnets, then you have a much lighter (and perhaps cheaper) structure. Yes, you still need windings of wire, perhaps a lot of it, to make the electromagnet. There must be trade-offs.
My wild poorly-informed brainstorming such as this, is counter-balanced by the knowledge that this is hardly new science, if the vacuum tube is > 100 years old now, then surely the magnet-free speaker has been experimented with before?
Tell me more! I know that Jim Winey built his Magnepan because he wanted to get the sound of electrostatics but using conventional amplifiers. Well, he only went part-way. To make a complete analogy, he would have needed an electromagnet instead of PM.
OK, what's the downside? An electromagnet requiring huge amounts of DC current? An armature winding requiring the annual copper output of Bolivia? What 🙂
- An electromagnet requiring huge amounts of DC current
- An armature winding requiring the annual copper output of Bolivia
Today, NdFeB supa stuff make Moving Coil even more attractive.
Just decade or two ago, for the very highest flux, you'd use a humongous electromagnet but even this has gone bye the bye. Today, you need superconducters to make electromagnets worthwhile.
____________________________
There's loadsa techniques that will make a noise. Only 2 achieve high quality sound; moving coils & constant charge push-pull electrostatics
There are some techniques that might be viable for good sound if some material specs became 'better' by more than an order of magnitude. Without these developments, the noise makers remain toys.
There are good reasons why Rice & Kellog's 1925 invention is pre-dominant from the cheapest to the highest quality applications. One of their 'inventions' was voltage drive for speakers.
Last edited:
Here you will find something interesting about linear loudspeakers: The Linear Current Loudspeaker
but unfortunately Joe removed the details of the crossover
but unfortunately Joe removed the details of the crossover
http://www.powersoft-audio.com/en/docman/897-m-force-datasheet/file
There is a lot of ongoing development in moving magnet technologies.
From PDF above.
Cheers,
Johannes
There is a lot of ongoing development in moving magnet technologies.
From PDF above.
Cheers,
Johannes
https://www.kth.se/en/aktuellt/nyheter/nytt-material-ger-unika-hogtalare-1.434583
There is more, but this is the two examples i found first.
Cheers,
Johannes
There is more, but this is the two examples i found first.
Cheers,
Johannes
The problem with the nano cellulose moving magnet loudspeaker is that they used a normal voltage amp. This means they had to use a very small fixed voice-coil to keep the inductance low to avoid a very reactive load. If they had though about current drive instead they could easily have increased Bl, lowered Qes and increased efficiency by using a much larger coil (more turns).
Thanks for those links, Circlomanen.
http://www.powersoft-audio.com/en/docman/897-m-force-datasheet/fileThis is a servo motor connected to a conventional cone.
I've played with such devices. They are good for high spl VLF but have serious problems at even 100Hz. Note the moving mass is in kgs. 😱
Efficiency ISN'T determined by motor principles at LF for an optimum design. It is determined by the size of box.
Theoretical and Practical Aspects of Loudspeaker Bass Unit Design
________________________________
https://www.kth.se/en/aktuellt/nyheter/nytt-material-ger-unika-hogtalare-1.434583This has the potential for high quality sound but material specs. have to increase by several orders of magnitude to take it out of the toy category. Sensitivity is 55dB spl @ 5cm for 1W.
Presently, it is in the class of
* An armature winding requiring the annual copper output of Bolivia
Examples of R&K's invention can cover nearly 10 8ves with a single unit at VERY high quality & sensitivity. In fact the ones which do this are VERY similar to their 1925 prototype ... about the size of the prototype Swedish unit.
_______________________
In my previous life, I kept a very close eye on 'new' ways of making a noise including doing & commissioning expensive studies on their feasibility.
Invariably, when you look closer, they are either toys or some material spec needs to be FAR better to challenge R&K.
The best was a novel way of using piezo film ... but even that needed piezo coeffs. 10x better than available today.
http://www.powersoft-audio.com/en/docman/897-m-force-datasheet/fileThis is a servo motor connected to a conventional cone.
I've played with such devices. They are good for high spl VLF but have serious problems at even 100Hz. Note the moving mass is in kgs. 😱
Efficiency ISN'T determined by motor principles at LF for an optimum design. It is determined by the size of box.
Theoretical and Practical Aspects of Loudspeaker Bass Unit Design
________________________________
https://www.kth.se/en/aktuellt/nyheter/nytt-material-ger-unika-hogtalare-1.434583This has the potential for high quality sound but material specs. have to increase by several orders of magnitude to take it out of the toy category. Sensitivity is 55dB spl @ 5cm for 1W.
Presently, it is in the class of
* An armature winding requiring the annual copper output of Bolivia
Examples of R&K's invention can cover nearly 10 8ves with a single unit at VERY high quality & sensitivity. In fact the ones which do this are VERY similar to their 1925 prototype ... about the size of the prototype Swedish unit.
_______________________
In my previous life, I kept a very close eye on 'new' ways of making a noise including doing & commissioning expensive studies on their feasibility.
Invariably, when you look closer, they are either toys or some material spec needs to be FAR better to challenge R&K.
The best was a novel way of using piezo film ... but even that needed piezo coeffs. 10x better than available today.
Last edited:
It is innocent unbounded enthusiasm that gives wild fancies flight. In turn it is reason, the lessons of hard experience and plain old science and engineering that are the figurative anti-aircraft battery that shoots those pie in the sky ideas down to earth 🙂
Meanwhile for us dabblers (me!) current drive is a more accessible tinkering zone than building a superconductive, liquid-nitrogen-cooled electromagnet speaker 🙂
Even with course work looming, I can't resist (pardon the pun) the temptation to tinker with series resistors. Some searching on the internet about DIY resistors led me to try pencil graphites (too high a resistance) and for similar materials: graphite bars (for artists) from a Michael's (art supply store for you foreigners!) that conveniently have very close to 1 ohm resistance each. So I can load down my mains and make that Behringer put out some current for a change 🙂
"WTF" moment: who knew that white graphite exists (at same art store)?
Meanwhile for us dabblers (me!) current drive is a more accessible tinkering zone than building a superconductive, liquid-nitrogen-cooled electromagnet speaker 🙂
Even with course work looming, I can't resist (pardon the pun) the temptation to tinker with series resistors. Some searching on the internet about DIY resistors led me to try pencil graphites (too high a resistance) and for similar materials: graphite bars (for artists) from a Michael's (art supply store for you foreigners!) that conveniently have very close to 1 ohm resistance each. So I can load down my mains and make that Behringer put out some current for a change 🙂
"WTF" moment: who knew that white graphite exists (at same art store)?
Last edited:
kgrlee,
That's it, we all need to go back to electromagnets and we will have perfect speakers! I see this in other threads on Compression drivers where some want to go back to old field coil designs, to bad they don't look very hard at the tradeoffs and why we are using magnets in the first place. With modern high MGOe Neo there really is no reason not to use it, otherwise Alnico seems to be the ticket. I'll leave the ceramic magnets for cheap devices.
I took one look at the speaker that Circlomanen posted and first thing I thought about was the extremely wide surround. To bad people don't seem to see how much surface area that surround has, that will have so much out of phase information emanating from that large surface area. I see this all the time with the one note subwoofers made for cars. When will some learn to look at the obvious before worrying about the esoteric.
That's it, we all need to go back to electromagnets and we will have perfect speakers! I see this in other threads on Compression drivers where some want to go back to old field coil designs, to bad they don't look very hard at the tradeoffs and why we are using magnets in the first place. With modern high MGOe Neo there really is no reason not to use it, otherwise Alnico seems to be the ticket. I'll leave the ceramic magnets for cheap devices.
I took one look at the speaker that Circlomanen posted and first thing I thought about was the extremely wide surround. To bad people don't seem to see how much surface area that surround has, that will have so much out of phase information emanating from that large surface area. I see this all the time with the one note subwoofers made for cars. When will some learn to look at the obvious before worrying about the esoteric.
Please read Don Keele aes paper
35 "Comparison of Direct-Radiator Loudspeaker System Nominal Power Efficiency vs. True Efficiency with High-Bl Drivers," presented at the 115th Convention of the Audio Engineering Society, New York (Oct. 2003)
From AES Papers -- Official website of D.B.Keele
Bl is directly related to efficiency. Not efficiency defined as a voltage driven speaker at 2,83 volts without any regard to watt being a composite of voltage times current.
Cheers,
Johannes
FWIW, 'White Graphite' is a goofy name for Boron Nitride. Its only similarity to graphite is its slipperiness and high thermal conductivity. It is an excellent electrical insulator.
Regards,
I said optimum design.
Today, we can forgo flat freq. resp. and just EQ the hell out of a small box. Gives us a lot more flexibility with sensitivity at MFs but not efficiency at LF which is still governed by Garner & Jackson's Figure of Merit which links box size, 'efficiency' and LF
This is all part of my patented Powered Integrated Super Sub tech. which incorporates all dis stuff including linear & non-linear analogue & digital EQ, amps with twiddled output Zo bla bla to make louder and better sounding boom boxes. 🙂
Understanding the FoM allows you to take full advantage of da supa dupa EQ tricks.
There's a good recent paper by John Vanderkooy on the subject of high Bl drivers too.
________________________
AES Papers -- Official website of D.B.Keele
Thanks for this link.
Brings back many happy memories of using and extending Don's stuff ... and also reminds me what a great and helpful guy he was. 🙂
I see his paper is actually in response to John Vanderkooy's and illustrates my point(s) exactly.
I'll have to get my patent lawyers to talk to Don 😀
Last edited:
Kgrlee,
Since we are talking about current drive here would you increase wire length at the cost of added mass while at the same time increasing the impedance of the driver? I'm trying to see the balance here between the different factors now that current drive has been introduced. I see increasing the impedance and adding wire length could be a very good thing in this concept.
Br back in about an hour to see if there is a response.
Since we are talking about current drive here would you increase wire length at the cost of added mass while at the same time increasing the impedance of the driver? I'm trying to see the balance here between the different factors now that current drive has been introduced. I see increasing the impedance and adding wire length could be a very good thing in this concept.
Br back in about an hour to see if there is a response.
It is dangerous to have a high DCR in the voicecoil. This leads to a thermal runaway. A high out-z amp will increase its power output into a rising impedans when the coil heats up.
This is partly why i believe in huge fixed voicecoil moving magnet design.
An optimum driver for current drive can and should have a very large coil, taking advantage of the current amps ability to drive very reactive loads without problem.
This is partly why i believe in huge fixed voicecoil moving magnet design.
An optimum driver for current drive can and should have a very large coil, taking advantage of the current amps ability to drive very reactive loads without problem.
If we are just talking about LF (ie subs) and loadsa EQ like Don suggests, then having the highest Bl^2 / Rdc and the best Xmax you can get is good.
The EQ required will be different with current drive than if you are using a voltage amp but you should be tuning this for your exact unit & box anyway.
It's not Bl which is important but Bl^2 / Rdc and this is what cost $$ if you want good Xmax too. You need to squeeze the maximum amount of Cu into your gap but this can be any Rdc you like.
The best impedance depends on the Voltage & Current capability of the amp. Don't forget big Bl^2 / Rdc will also give a huge Impedance peak at fo so don't have too high a Rdc.
Don's understanding of the issues is good. One interpretation of his paper is
... if you have a sealed box and are operating ONLY near the fo impedance peak, you might want Rdc=1R 😱 and anticipate your big Bl^2 / Rdc to give sufficient impedance to match the amp.
You have to do the sums for your unit/box/amp, but I would be tempted to have a huge gap and a 4 layer coil with thick wire.
BTW, operating like this will give maximum THD benefit from Current Drive if you have the correct EQ. Roughly, the bigger the impedance rise due to high Bl^2 / Rdc, the more electrical stuff dominates and can be corrected by eg Current Drive.
Low Rdc usually gives a more robust and 'easier to wind' voice coil too.
Don't get too carried away by this unless you can GUARANTEE the amp won't be asked to deliver power far from fo where your impedance will drop to 1R. ie sharp LF & HF filtering. And don't forget your box will reduce the peak a lot.
Last edited:
This is the way B&C 18 and 21 IPAL is designed. A current drive amp can run a dead short without problem, so there is no need for any DCR. DCR is only a problem for a current driva amp.
For a fixed coil moving magnet driver one could use 20 x 50 mm solid strips of copper as a voicecoil. Stuff can be really fun when running 500 amperes through the voicecoil.
500 amperes over 0,1 ohm needs 50 volts. It would make a powerful magnetic field. I am sure 80 small Nd magnets is strong enough for a high Bl with that kind of voicecoil.
and i heard that Bolivia is all out of copper this year... someone is building a loudspeaker. 😉
Last edited: