Need help making speaker from scratch.

Dear friends,

I understand the principle of induction.

I understand how a mechanical wave (sound) is converted (via mic) into an electrical signal & thus turned back (via loudspeaker) into mechanical wave (sound).

I understand all the rules that follow (Fleming, Lenz, etc).

I have made many loudspeakers before, though I have always guessed on the magnet & voice coil. I literally just grabbed any magnet I could find & wrapped as many turns of wire around it as possible. Not enough turns & the speaker wouldn't function. Too many turns & my cone literally sagged down on the suspension/spider & it wouldn't function. There was no math involved, lol.

I assume there is a proper way to do this. Perhaps a function?

Additional,

The amplifier I will be using puts out 20wrms/ch. @ 4 ohms. It will run up to 16 ohm. I assume I need to match the impedance of the speaker with that of the amplifier. Not sure how though.

I know it's a bit unconventional, but I will be winding the voice coil around the magnet. This is for a presentation & I want the vc to be visible! (not hiding inside the ring magnet)

I will be ordering from this company: https://www.kjmagnetics.com/products.asp?cat=13

The link brings you directly to the cylindrical neo magnets. I wish to use these. My budget is $10/magnet.


Any tips for winding the voice coil? I assume the goal here is getting as much copper within the lines of flux as possible. Is it okay to "overlap" turns? Is there a convention as far as which way to mount the magnet? For instance, the N pole should always be facing down. I assume this doesn't matter as I could just as easily switch the polarity of the voice coil.

Thoughts?

The important thing is to avoid driving your amplifier into too low a resistance. Speakers generally are rated by their DC ohmic resistance. I have never seen much discussion about the voice coil Henry value.

For the fun of it, I'm running simulations which put 5W (peak) into loads of 4, 8, and 16 ohms. I chose a smaller figure than your stated 20W, since your speaker will not carry the full amount for more than a fraction of the time.



I went by ampacity ratings at the links below.

Unlike you I have not constructed speakers, but I have dismantled a few, and none had anywhere near as much wire as the tables indicate. I suspect you could make a 4 ohm speaker using 26 gauge, 100 feet. (But don't quote me.)

Links to charts of wire gauges showing safe amp-carrying capacity:

http://amasci.com/tesla/wire1.html



http://www.powerstream.com/Wire_Size.htm

The traditional loudspeaker is a ring magnet with a magnetic short across one end of it with a rod which sticks out of it around which the voice coil sits. on the other end there will be a large washer so its inner hole is a close fit to the voice coil. The effect of this is that the voice coil whose thickness is in the order of 1mm sits inside a gap of 1.5mm. The actual VC formmer is some sort of varnished paper, making it extremely light and rigid, with a couple of layers of windings on, which again are much thinner then you would imagine, Music is very Peaky , so driving a 20W amplifier to its distortion point will only give 4 or so watts of RMS power on music. The magnetic field must be linear over the total length of the voice coil travel (3mm?). For the cone, why not use a piece of flat polystyrene, this was the basis of the "poly planar " loudspeaker that was marketed , some 40 years ago. You will still need to build some sort of edge suspension for it though.
Frank