Home-built Speakers from scratch

[murphyslabrat]

I am not sure if I should be posting this here or in the A/V forum, but I am interested in playing around with building a speaker from scratch. I wanted to try something small, using materials easily (see: cheaply) obtained: I was planning on using a canning-jar lid (regular width, meaning about 65mm) for the frame, sandwich bags for the diaphragm material, and some neodymium magnets from a local electronics supply store.

There appears to be very little on the web about this topic (or I am just asking google the wrong questions), but most of what I've seen is low-sensitivity speakers built using cardboard and iron magnets. My plan was to extend these projects to a smaller design, using a more flexible material, and a stronger magnet.

The intended use is for un-amped speakers/headphone speakers. And, while I am doing this primarily for fun, I would enjoy having something useful at the end. Hence, I am willing to get Obsessive-Compulsive about the quality, all else being equal.

Here are the questions:

0. Can anyone recommend to me a project or thread that dealt with this already, or a resource for this kind of information (bear in mind, I am no EE).

1. Would the plastic in sandwich bags deform too easily to be used as a diaphragm? And, if so, what would be a better material be, that could be still be obtained easily?

1.b How do variables like diaphragm weight, size, flexibility, tightness, etc. affect frequency response from the finished speaker?

1.c I am not really using a spider, but I am using silicone caulking to secure the diaphragm. For this small of a size, and considering the material, would one be necessary?

2. Am I right in assuming that more wire-length/higher magnet power makes for better SPL with a given diaphragm material at a given voltage?

3. for the coil, does the total mass of copper have an affect, or is it just about the un-wound length? What strength magnets do typical headphone speakers use?

4. Is it necessary to have an axially polarized magnet?

 

[CycloWizard]

I have been designing and building voice coils recently. It seems that Dr. Klippel is the recognized authority on how to address distortions, so I'd give his website a gander. He has everything from technical reports to application notes that should be just what you're looking for. I've read just about everything on his site, as well as many other papers on the subject very recently, so I can probably answer specific questions. The link should address #0 for now.

1. probably, though it depends on how much travel the coil will have and how much tension the plastic is under when mounted. Most spiders seem to be made from rubber.
1b. Weight is very important because the force required to accelerate the coil depends on the total mobile mass of the speaker. Mass is probably one of, if not the, biggest concern for correct frequency response.
2. Yes, more or less. The general rule of thumb is that F=B*L*I, where F is the force put out by the motor (counteracted by the inertia, among other things), B is the radial magnetic flux density, L is the length of wire within the field, and I is the current. If you PM me your e-mail address, I can send you a detailed derivation showing how all of this fits together.
3. The total length is important for controlling the resistance, but only the wound portion is important for most other things (i.e. having a little extra wire hanging off of the coil isn't going to hurt you, as long as it's not hitting anything while the coil moves).
4. Depends on the design, but pretty much every design I've seen uses axially polarized magnets to get the desired flux path. Details of flux paths are also in my derivation if you're interested.