The configuration people are discussing isn't with the N and S poles touching, though. Instead it is with the N pole of one magnet pointing up and the S pole of the neighboring magnet pointing up. Even with the magnets touching, field lines will emerge from one pole and return to the neighboring pole (more or less).
Another good reason to separate the magnets (besides room for air to move) is that the change in the magnetic field with distance above the gap can be lower as a function of height above the magnets. Although increasing the gap reduces sensitivity, it also may reduce distortion.
Jeremy
Jeremy,
Thanks for getting involved and commenting. (But read the thread, please.)
The configuration being discussed was with the N and S touching at their edges. So yes, they were touching. Everyone here already realizes that the N and S are both facing the same way, and are the large flat faces of the magnets, which are all parallel to the mylar. And they realize that even if the edges are touching, there will still be field lines that emerge, to some relatively small and much less useful extent. But it is unhelpful to just state that field lines will still emerge, without any regard for the goals that are implicit in the thread, i.e. without noting that it would likely be a much less desirable configuration.
Since you seem to possibly be knowledgable about magnetics, we would welcome any help in learning to better-characterize the effects of the magnets' spacing and the effects of the height of the conductors above the magnets, with respect to the force vector exerted on the mylar when there is current in a conductor that is attached to the mylar.
Presumably there would be combinations that produce maximum force, and combinations that produce the best-behaved (most-orthogonal) forces, etc, etc. For example, I think that we would usually want the field lines to be parallel to the mylar, at the mylar. (But, of course, the mylar changes its distance from the magnets, as varying current amplitudes are passed through the conductors that are attached to it. So we would be looking for the "best" range of distances, and the best at-rest distance given the other variables.)
And it seems likely that the field strength, at some distance in front of the magnets' front plane (e.g. at the mylar), would increase as the columns of magnets start to separate, then reach a maximum, then start to decrease as they get even farther apart. At the same time, I'm guessing that the distance from the magnets' plane to the most-useful region where the strongest part of the field is most-parallel to the magnets' plane would also change, but maybe not in the same way or at the same rate.
It might be useful to be able to calculate (or simulate, e.g. with FEMM), in order to determine some sort of quasi-optimization, or at least just get a feel for how to guess at a satisfactory configuration, of the combination of 1) the foil width, 2) the foil's distance from the magnets' plane, and 3) the distance between the columns of magnets.
Ideally, maybe, it would be good to be able to characterize or even predict the maximum displacement of the mylar membrane to which the conductors are attached, for a given current amplitude, so that the resulting configuration would not cause the mylar to actually touch the magnets. My Calculus of Variations skills are basically non-existent so it would be a whole lot of work to get to the point where I could construct and then solve the forced-membrane equations. BUT, maybe we could come up with some very simple guideline, such as "tension the membrane while it's horizontal, until placing X grams in the center makes it sag by Y mm", if we could calculate or simulate to get the force on each conductor given the other variables.
In the end, we would also want the most-accurate reproduction of sound that we could possibly get for a reasonable cost, and a reasonably-large maximum sound pressure level from a typical rated maximum amplifier output power.
I was actively planning and preparing to design and build some planar magnetic speakers, a long while ago. But life got in the way and I had to put it aside. And in the meantime I have purchased some Magnepan MG-3.6/R speakers. But I wouldn't mind helping to figure out how to do it and would still love to be able to come up with something that seemed too good to not build. I do also still have the DC Gaussmeter that I bought, back then, which has a directional probe.
Cheers,
Tom
Thanks for getting involved and commenting. (But read the thread, please.)
The configuration being discussed was with the N and S touching at their edges. So yes, they were touching. Everyone here already realizes that the N and S are both facing the same way, and are the large flat faces of the magnets, which are all parallel to the mylar. And they realize that even if the edges are touching, there will still be field lines that emerge, to some relatively small and much less useful extent. But it is unhelpful to just state that field lines will still emerge, without any regard for the goals that are implicit in the thread, i.e. without noting that it would likely be a much less desirable configuration.
Since you seem to possibly be knowledgable about magnetics, we would welcome any help in learning to better-characterize the effects of the magnets' spacing and the effects of the height of the conductors above the magnets, with respect to the force vector exerted on the mylar when there is current in a conductor that is attached to the mylar.
Presumably there would be combinations that produce maximum force, and combinations that produce the best-behaved (most-orthogonal) forces, etc, etc. For example, I think that we would usually want the field lines to be parallel to the mylar, at the mylar. (But, of course, the mylar changes its distance from the magnets, as varying current amplitudes are passed through the conductors that are attached to it. So we would be looking for the "best" range of distances, and the best at-rest distance given the other variables.)
And it seems likely that the field strength, at some distance in front of the magnets' front plane (e.g. at the mylar), would increase as the columns of magnets start to separate, then reach a maximum, then start to decrease as they get even farther apart. At the same time, I'm guessing that the distance from the magnets' plane to the most-useful region where the strongest part of the field is most-parallel to the magnets' plane would also change, but maybe not in the same way or at the same rate.
It might be useful to be able to calculate (or simulate, e.g. with FEMM), in order to determine some sort of quasi-optimization, or at least just get a feel for how to guess at a satisfactory configuration, of the combination of 1) the foil width, 2) the foil's distance from the magnets' plane, and 3) the distance between the columns of magnets.
Ideally, maybe, it would be good to be able to characterize or even predict the maximum displacement of the mylar membrane to which the conductors are attached, for a given current amplitude, so that the resulting configuration would not cause the mylar to actually touch the magnets. My Calculus of Variations skills are basically non-existent so it would be a whole lot of work to get to the point where I could construct and then solve the forced-membrane equations. BUT, maybe we could come up with some very simple guideline, such as "tension the membrane while it's horizontal, until placing X grams in the center makes it sag by Y mm", if we could calculate or simulate to get the force on each conductor given the other variables.
In the end, we would also want the most-accurate reproduction of sound that we could possibly get for a reasonable cost, and a reasonably-large maximum sound pressure level from a typical rated maximum amplifier output power.
I was actively planning and preparing to design and build some planar magnetic speakers, a long while ago. But life got in the way and I had to put it aside. And in the meantime I have purchased some Magnepan MG-3.6/R speakers. But I wouldn't mind helping to figure out how to do it and would still love to be able to come up with something that seemed too good to not build. I do also still have the DC Gaussmeter that I bought, back then, which has a directional probe.
Cheers,
Tom
Last edited:
Finished building my experimental full range planar headphone, very pleased with the outcome, used a cheap headphone as the frame. A bit fiddly on one half the left which contains resistor and 4 wires, the right one only contains 2 wires. 2 different neo layouts were used and 2 different sized neo magnets. One side is a little louder I think! My ears are not as good anymore, can only hear up to 8k. This makes it difficult to tell by listening alone, one of the drawbacks of old age. I made it about 29 ohms although the Koss Pro/4XTC measures 105 ohms. I thought about reducing the impedance to about 6 ohms ,any thoughts on the matter from you technos?
thank you, have started another rebuild of a cheap headphone, which sounds terrible in comparison to my first rebuild. Can't take my planar headphone off, so good, it's like listening to speakers not like headphones at all!! Very happy with them.Have taken the next headphone apart. Supposed to be the same model, but is entirely a different setup and shape, oval in fact. A new challenge for me.
Henry,
Cool!
My gut instinct would be to have both sides identical.
Also, if the side of the planar "speaker" that is away from your ear is open to the outside air, the sound will be diferent than if it's all sealed up. It might be much better with them open. But it depends on the details of the design. Something to try, though! (I guess in that case you might be able to just make your own open frame, instead of tearing open the existing headphone shell, if it happens to be the closed type.
Tom
Cool!
My gut instinct would be to have both sides identical.
Also, if the side of the planar "speaker" that is away from your ear is open to the outside air, the sound will be diferent than if it's all sealed up. It might be much better with them open. But it depends on the details of the design. Something to try, though! (I guess in that case you might be able to just make your own open frame, instead of tearing open the existing headphone shell, if it happens to be the closed type.
Tom
PLANAR HEADPHONES.
Finished at last, a bit fiddly made a few goofs on the way, but very much worth the effort very happy with result. Don't sound like headphones, more like speakers inside your brain. Wore them for hours and hours never wore headphones for such a period of time, unbelievable really. The next 3 pics are of my new headphone rebuild, using a cheap chassis so to say!!
This is a more oval shape than my first one, and a different design and fit. I hope I don't make too many goofs this time, altho' I have made one already and broke 2 of my magnets already, clumsy aren't I
Finished at last, a bit fiddly made a few goofs on the way, but very much worth the effort very happy with result. Don't sound like headphones, more like speakers inside your brain. Wore them for hours and hours never wore headphones for such a period of time, unbelievable really. The next 3 pics are of my new headphone rebuild, using a cheap chassis so to say!!
This is a more oval shape than my first one, and a different design and fit. I hope I don't make too many goofs this time, altho' I have made one already and broke 2 of my magnets already, clumsy aren't I
Attachments
this time!!!
Hopefully.http://www.diyaudio.com/forums/attachment.php?attachmentid=349024&stc=1&d=1368710119
Hopefully.http://www.diyaudio.com/forums/attachment.php?attachmentid=349024&stc=1&d=1368710119
ANOTHER REBUILD OF A BOUGHT HEADPHONE USING NEOS.
PICS showing work in progress,the headphones in pic 8 are my previous rebuild, which sound marvelous, well worth the effort! Just a few things to change used 2 different layouts to see which was the best, the one on the left is slightly louder. So must redo the one on the right will see how my second rebuild goes, vis a vis the impedance before I change the first ones impedance, if I do!
PICS showing work in progress,the headphones in pic 8 are my previous rebuild, which sound marvelous, well worth the effort! Just a few things to change used 2 different layouts to see which was the best, the one on the left is slightly louder. So must redo the one on the right will see how my second rebuild goes, vis a vis the impedance before I change the first ones impedance, if I do!
Attachments
Last edited:
do you mean my headphone rebuilds? If so they sound brilliant, have worn both pairs for hours on end, have never before worn headphones so long. Usually they grate on your ears after awhile but these full range planar designs are so easy to listen to. They don't sound in the ear as so many do. Like listening to speakers inside your head! You don't realize you have headphones on. Very strange feeling or should I say hearing! These spoil you for any other method of listening period. Certainly glad I finished both pairs. Even if the second rebuild was harder than the first. Different shape and layout, I used two different shells, of cheap headphones I bought from Amazon, so I didn't have to build the shell as well. Which would have been beyond my woodworking skills!!
do you mean my headphone rebuilds? If so they sound brilliant, have worn both pairs for hours on end, have never before worn headphones so long. Usually they grate on your ears after awhile but these full range planar designs are so easy to listen to. They don't sound in the ear as so many do. Like listening to speakers inside your head! You don't realize you have headphones on. Very strange feeling or should I say hearing! These spoil you for any other method of listening period. Certainly glad I finished both pairs. Even if the second rebuild was harder than the first. Different shape and layout, I used two different shells, of cheap headphones I bought from Amazon, so I didn't have to build the shell as well. Which would have been beyond my woodworking skills!!
well that's good!
I'm tempted to try something like this myself but maybe won't have enough time till 2014
I shall read the rest of this thread when I get the chance. I take it the loudspeaker sounds pretty good too?!
can't you make time go bed 1 hour later or get up 1 hour earlier, it wouldn't take you long to build if you use a cheap pair of headphones for the frames. The best frame to use would be PRO-LUX PRO 20 headphones. Just the right size to build the diaphragm and wood surround with the perforated sheet to fix magnets to. My pics show all the details you would need. The work needed is well worth it when you listen to them. They sound better than any headphones, that I have ever heard. Marvelous and sweet, they must be because I have never worn headphones for so long before, usually they irritate after a short time, but with these I have even watched television wearing them, which I have never done before.
I suppose so!
If I get up 1 hour earlier I'll be yawning all day. If I go to bed 1 hour later, one thing will lead to another and the next thing you know it will be 4am!
This month I'm seriously skint! Don't even think I can spare £20. Besides, I'm still busy working on my invention. Still got to set up some spectrum analyser software and work out how to test it. I've still got a few products, permutations, designs I haven't even built yet. Everything hand made!Of course I'd have to try my invention with the headphones
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.