diyAudio (
-   Multi-Way (
-   -   Loudspeaker project from scratch - driver construction, cabinet, & crossover (

UncleJessie 26th March 2003 10:16 PM

Loudspeaker project from scratch - driver construction, cabinet, & crossover
Loudspeaker gurus, your help/advice is requested:

I'm looking to engage in a project which basically takes me from "scratch" to building a pair of speakers. The steps would be described in a forum like this one, in a book, or somewhere on a website. Let's say I am to build a stereo pair of passive subwoofers. Here is what I'm referring to:

1. Reference an existing off-the-shelf woofer with certain specs (something reasonable - very average design) and note it's various T/S parameters. This is just for reference. My goal is actually to build a "raw" woofer.

2. Run a simulation(s) using speaker driver design s/w in order to choose an appropiate magnet, voice coil, cone, etc. for use a given enclosure.

3. Buy the various driver parts and assemble them. I'm referring to actually assembling the woofer, in this step.

4. Perform some tests on the woofer to verify the specifications (Xmax, etc..). Replace parts, tweak, if necessary.

5. Build the cabinet and install the woofer.

6. Build and install a passive crossover.

7. Make more measurements.

8. Tweak the cabinet.

I acknowledge that there are probably more steps and other things to consider. Please bear with me, I'm obviously new to this.

My goal is to get ground level experience in a building a speaker system. The initial focus is more on the driver design than the cabinet & crossover, for the time being. I'd like to see how accurately I can build a woofer to meet design specs.

Vance Dickason steps the reader through a few transducer design
(speaker driver) examples in his book, although the s/w he uses is pretty expensive. Anyone know of alternatives? Any insight from you guys will be great.



Nappylady 26th March 2003 10:47 PM

If you're going for ground level, don't start building your own drivers with the intent of actually using them in a box!

Building raw drivers is fun and exciting, but also extremely challenging. Getting it right takes a LOT of knowledge.

Trust me. I made a CD-subwoofer with homemade voice coils and everything. It was a disaster! I still have all the parts...

The only use I can think of for making your own woofer is doing something extremely esoteric in design--perhaps something with a HUGE moving mass and lots of excursion, with a very high-power amp and a voice coil made of maybe 24guage wire--this would give nice very low frequency reproduction.

You might try making your own tweeters or midranges first. Definitely go with the self-assembling design to begin with. It makes things lots easier. Having to deal with hanging suspensions (surrounds and spiders) takes a lot of practice.

Seriously though. Buy some drivers and build a box for them. Don't think of it too seriously, and don't spend too much on them; think about making this project, and then making your next project better because of what you've learned from this one. Make sense?

pinkmouse 26th March 2003 11:02 PM

Hi UncleJessie

Dickason makes it sound so easy, doesn't he!

In reality, because he is a well known designer, all his prototypes will be built to his specs by driver manufacturers to high tolerances and on specialist machinery.

A DIYer is unlikely to get the same results, or be able to buy the different components to construct their own drivers. Homemade components are unlikely to have the same specs, and the tolerances required for assembly and manufacture difficult in the extreme.

Not to put a total downer on your ideas, it is possible, but it could prove very expensive and frustrating to get a reasonable end result.

UncleJessie 27th March 2003 12:14 AM

Some clarification
I just want to clarify that I'm not attempting to manufacture a magnet, spider, basket, etc.. However, maybe I could make a voice coil?

A previous post suggested that it might be easier to build a tweeter than a woofer. Is this the general consensus? If so, why?

Once again, I'm looking at this project from a transducer engineering viewpoint. So, at the very least I'd like to assemble a driver and perform some tests on it. I realize that this won't be a trivial task and there will be a lot of stumbling blocks. Knowing this, I'm still ready to get my feet wet if the project doesn't break the bank.

I welcome your posts. Please keep them coming.


Nappylady 27th March 2003 04:23 AM

Okay, we are at the nitty-gritty here.

If you want to get into DIY audio, there are several levels to it.

One is going to Radio Shack, buying a setup, and installing it yourself.

Then there's installing the Radio Shack setup and running your wires through the walls.

Then there's building your own cabinets.

Then there's building the drivers.

Then there's building the parts for the drivers.


There are places online where you can get raw voice coils and cones and replacement surrounds and enough fun stuff to make your own drivers, but it's almost guaranteed that you won't do as well as the pros in driver construction.

I've thought about actually building the parts for drivers, and I've gone so far as to make some really crappy voice coils on my own (using an M&M Mini's tube as a former--this was a BAD CHOICE--and 30-guage wire from Radio Shack. Unless you need high power handling, use smaller wire... I think 40-guage is good, and since it reduces the moving mass, you can get a more efficient speaker.)

My favorite idea is making the surrounds. There's nothing quite as cool about a speaker as watching the rubber surround contort as the woofer moves an inch at a time to make your chest thump. It's just really cool to see. :) At present, though, I've never succeeded in making one. While I prefer cloth surrounds over everything else, I haven't the slightest clue as to how you would go about making a cloth speaker surround.

Cones? Same thing. I think you could paint a salad bowl with a paper pulp and get a fair result though.

Spiders? I'm clueless as to how to make these.

Gaskets? Cardboard. (At least something is easy!!)

Dustcaps? This could be really fun--copper on a lathe, for example--and you can make those really cool Seas-style copper plug heatsink dustcaps.

Baskets? You'd need experience in machining for this, and probably the easiest thing is to stamp an aluminium frame. This is probably rather expensive though.... I don't know much about machining so I'll leave this to the experts. This is another area where I am truly clueless. The best I ever did was to secure my magnets inside a Lego fixture. :-P Even this didn't work too well.

Anyway. Bounce some ideas off us! The odds are hard against us that we'll get anything even close to pro level, but I think this is more about fun than about success!

Dave Bullet 27th March 2003 05:38 PM

How about learning about transducers from dis-assembly?

I mean, you could learn a lot by taking a working driver and "mucking about with it" (sorry for the technical description). Changing the mass of the cone, altering the spider resistance, removing and reattaching the voicecoil with extra windings

At least this way you can play with the motor and see the effects.

It seems to me (no I'm not an engineer) to build something that is working that has fine tolerances without the equipment and someone to mentor / with experience will be very difficult to get a working product.

If your end goal is to get something you built from scratch that you will enjoy listening to, I'd buy a driver and muck around with crossovers and cabinets.

I can't imagine a cheap way of building a driver with all the parts and bits and pieces you'll need... and actually have it work.

If you do have a go - would be great material for a website. I'd be interested!

all the best,

Nappylady 27th March 2003 05:42 PM

Uh-oh. Another David.
here we go. :-)

Disassembly of drivers is a lot of fun! Go for it! Find as many old, blown-out, single, or otherwise useless drivers, and rip 'em up! Learn about how they're constructed, what kinds of materials are used, and get ideas about how you want to do it.

Yeah, that's something I forgot to mention, but driver disassembly is one of the most educational things I've ever done. TONS of fun. Don't miss it.

Topgun 28th March 2003 02:27 AM

Dissasembling anything is a great way to learn how it works, as long as you don't kill yourself in the process or forget how to put it back together. Which is another way to get yourself killed. Ask my dad about all the things i took apart when i was a kid. Learned alot though at the expense of a couple groundings and missed allowances.

I would go ahead and try to build a driver, forget about the naysayers. It would be fun, and i bet you would learn a ton. This is something that i would also like to try but i don't have as much dissposable time or money anymore.

Good luck

Hamish 28th March 2003 02:46 AM

yep. i like the idea. try using the baskets of old drivers to get started. maybe just redo the voice coil first, then try a bigger magnet. if all that works start looking at cones. carbon fibre or fiberglass are really easy to work with for this, but if you are extra keen (and there are metal spinners near you) try getting them to spin one up out of aluminium. i don't know if it is possible, but they may even be able to do the coil former and cone one piece. push bike inner tubes for surrounds.....

the list goes on.

i hope you make a go of it.

keep us posted.

Nappylady 28th March 2003 04:19 AM

Ooh!! Neat idea! Voice coil former same piece as the cone... I like your thinking!

I think fiber-reinforced silicone sealant would be a nice surround material too, though forming it well would be a challenge. I think two pieces of plastic tubing arranged in a circle of the right size would to the trick, or maybe just one. The hard part would be getting the part where the piece of tubing joins itself to be smooth.

Bike tubes are very big. I don't know if I want that large of a driver. :-)

I have been playing with some drivers and I've discovered that the neodynium magnets I have lying around make wire move sideways instead of up and down. I'm going to see if I can capitolize on this behavior.

All times are GMT. The time now is 06:48 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio