DIY 1 Inch Exit Compression Driver

Hey all,
My school just scored a CNC machine plus a 3D printer and suffice to say that I REALLY want to do something with it and the thought that came to mind was a 1 exit compression driver. Now DIY'ing a compression driver, let alone your average dynamic driver, is crazy complicated but I figure I can learn on our collective expertise. Let's get started with my questions, because, well, there are quite a few.

A. Diaphragm: well... I could machine an aluminum diaphragm but the CNC machine is supposedly pretty expensive and now there's questions if there's even money in the budget for the CNC machine. If it does go through, then Ill probably try something like this:
http://www.ygacoustics.com/userfiles/files/PDF/Technologies/YG_BilletCore_Oct13.pdf
Interestingly they use ribbing on the back of the cones which is claimed to increase stiffness and that will be important because I'm hoping to get to 20K (more on that in a second). Anyway, assuming we get the machine, it will be have resolution to the um, so an ultra low mass yet stiff diaphragm should be very possible. Assuming the worst though, I could in theory print a plastic diaphragm using the printer and also get very good results (I don't know though because I tend to like metal dome compression drivers). The issue with that is though that obviously plain old ABS plastic won't make really work out to 20khz, so I will have to source the plastic myself, prototype each diaphragm and then determine whether or not there is better to be had from another plastic. The one plastic I would hope to find which fit the printer would be Teijen Endumax (?). There was an article in AudioExpress about how Dr. Kurt Mueller is working with it and has been for awhile, perhaps a call to him would help...


B. Voice Coil Former: Full disclosure: I stole this idea from the Vitavox S2 when I researching this topic but its pretty nifty and should in theory reduce eddy currents in the motor while also lowering weight. Anyway, the idea itself would be to wind the voice coil onto the voice coil former, which is one piece along with the diaphragm.


C. Voice Coil: I hope to use pure copper wire but I would be open to using copper clad aluminum wire for mass reduction.


D. Motor: I will hopefully use pure soft iron as a pole piece but the printer obviously doesn't do that but I may have an idea on that... What if I were to use plastic formulated with iron particles impregnated in the plastic itself? I wonder if that would give me enough flux density in the gap though? I will likely use an AlNiCo ring magnet from a reputable supplier that can provide the necessary flux density required to reach 20khz with good sensitivity. Otherwise I will attempt to source a neodymium magnet from KJ magnetics or perhaps a field coil, although I don't know how I would wind the coil and I would need someone to help me calculate the numbers involved?


E. Phase Plug: this will almost certainly prove to be a major sticking point and I will really have to lean on you guys to help me get it right? The options range from the classic phase plug of the WE 555 (effectively a bullet in front of the diaphragm, surprisingly it is still in use from companies like ALE, GOTO, Onken)
to the modern radial phase plugs of today's pro drivers. Now remember when I said I would like to reach 20khz and how I promised more on that later, well here it is:
https://www.google.com/url?sa=i&rct...hZzcF0q2YsdZW_W8l6xuYyeg&ust=1457665174885199

Ok, assuming this puts up past 20khz, we face an even bigger challenge in the phase plugs because now the path lengths from any given points on the diaphragm vary even more. I may or may not have FEA software to assist me but that would certainly help immensely...


F. Format: The WE 555 is designed so that the voice coil attaches to the rear of the dome, as seen here:https://www.google.com/url?sa=i&rct...gfh2Yud5eKadL5DjAwLqGlbA&ust=1457665402842679

Well, that's obviously one way to go about it and the companies that produce high end compression drivers in Japan swear by it, but also in Japan we have TAD, who's drivers attach the voice coil to the front of the dome:
https://www.google.com/url?sa=i&rct...g21wrtR13X3ejKcCMGJwfW7w&ust=1457665563622719

I suppose it comes down to whether it is better to push or pull the diaphragm. :confused:


G. Surround: If I want 20khz with an aluminum diaphragm then I need an aluminum surround, or am I wrong? I can't print the surround so assuming I used the plastic diaphragm how would I reliably attach, say, a Mylar surround to the dome. One note: if it becomes obvious that 20khz is not possible with the given materials, then I will let that goal go, but I also will not get to 20khz the way TAD did in the 4001/4002. That is with using a series of surround resonances to extend the FR.


H. Body: the body of the compression driver will be 3D printed and hopefully allow for phase plug, diaphragm and magnet material to be changed in just a few minutes. If and when I conquer my wondering heart and admit this is the best I can do then I will (hopefully) build a more permanent body to minimize internal resonances.


I. Throat: ideally the throat would match the expansion profile of a tractrix horn, as the tractrix profile is the one I find to be the most natural on voice. As for the ratio, is 1:10 a good start (BTW, I hope to make the throats interchangeable too...), or should be higher or lower?

Well, I think I got everything, if I missed something please let me know.
Thanks in advance,
Clashing
 
Last edited:
https://www.matterhackers.com/3d-printer-filament-compare
Well this helps immensely, it looks I will probably use carbon fiber reinforced plastic for the diaphragm (prototyping will be done in another material because this stuff is not cheap). The thinnest and therefore lightest diaphragm I can produce is 50um (thinner diaphragms are possible but I would have to purchase a thinner nozzle). I can and I may use a type of conductive plastic (seen as the Fill Series on the link) as the voice coil, you can see the electrical measurements on the link so this would be suitable for a voice coil or should I try and use pure copper? The trouble with that though is that copper is more expensive and difficult to wind. For the surround, I hope to use TPU plastic as it has very good bonding properties to prevent separation but I worry about it will do in high humidity. For the body, I will likely use cheap material on the body and Magnetic Iron PLA plastic as pole piece.
I will hopefully post pics of the printer sometime tomorrow,
Clashing
 

ICG

Disabled Account
Joined 2007
Your post is horrible to read, could you please edit in some paragraphs/empty lines to make it a bit easier? I'm sure that will hep to get some replies.

G. Surround: If I want 20khz with an aluminum diaphragm then I need an aluminum surround, or am I wrong?

No, the opposite. If you want to CNC your diaphragm out of an uncompromised block of aluminium, it would be a very bad to introducing bending, flexing at the edge, which introduces stress to the membrane - which you wanted to avoid.

I can't print the surround so assuming I used the plastic diaphragm how would I reliably attach, say, a Mylar surround to the dome.

Mylar would be a good choice to start with. Laser cutting gives you by far the best results. Mounting the surround works best with liquid glue or possibly evaporated glue while masking the diaphragm.


One note: if it becomes obvious that 20khz is not possible with the given materials, then I will let that goal go, but I also will not get to 20khz the way TAD did in the 4001/4002. That is with using a series of surround resonances to extend the FR.[/QUOTE]
 
https://www.matterhackers.com/3d-printer-filament-compare
Well this helps immensely, it looks I will probably use carbon fiber reinforced plastic for the diaphragm (prototyping will be done in another material because this stuff is not cheap). The thinnest and therefore lightest diaphragm I can produce is 50um (thinner diaphragms are possible but I would have to purchase a thinner nozzle).

I can and I may use a type of conductive plastic (seen as the Fill Series on the link) as the voice coil, you can see the electrical measurements on the link so this would be suitable for a voice coil or should I try and use pure copper? The trouble with that though is that copper is more expensive and difficult to wind.


For the surround, I hope to use TPU plastic as it has very good bonding properties to prevent separation but I worry about it will do in high humidity. For the body, I will likely use cheap material on the body and Magnetic Iron PLA plastic as pole piece.

I will hopefully post pics of the printer sometime tomorrow,
Clashing
 
Interesting project.
A conductive plastic voice coil and magnetic iron plastic pole pieces will result in low efficiency and low power handling.
You need a low resistance material for the voice coil, and high magnetic permeability for the pole pieces.
It is really hard to get to 20khz, everything has to be optimal to get there.
A ring radiator might give you a better chance to reach 20khz, it can be done even with a softer diaphragm material like mylar.
 
Cloning the S2's voice coil is a good idea.

The rest leaves much to be desired and is unlikely to provide you with the desired results. Why go through all the hassle when you can buy complete motor structures with flux density up to 2.2 Tm in China?
It's a DIY project, I know and I respect the effort.

The magic is in the diaphragm and phase plug design.
You'll need tooling to form a diaphragm.
 
Last edited: