Cheap Coax Speaker Build

[JCJR]

Should anyone be interested, I wrote up three rather long and boring articles about building and testing some cheap homemade coax speakers. So far, one page about construction, one page about driver selection, and one page about REW test results. To be followed sometime by a final page of a few incidental details and listening impressions.

Probably too uninteresting for most folks but it covers questions I was curious about. Was written to include details that I would want to know were I reading about someone else's experiment.

It starts here-- http://errnum.com/html/coax_speaker_build.html

[justin3am]

Thanks for posting! Lots of interesting information there. I'm really impressed with the cabinets you made. They look great.

There is a shop near my office called Speaker City and in addition to boutique speakers they also sell parts and tools for building. I've always thought about making a pair of speakers but figured the math is too hard.

[JCJR]

Thanks for posting! Lots of interesting information there. I'm really impressed with the cabinets you made. They look great.

There is a shop near my office called Speaker City and in addition to boutique speakers they also sell parts and tools for building. I've always thought about making a pair of speakers but figured the math is too hard.

Thank Justin

Designing and building a speaker "the proper way" would be a lot of work. I read about it some over the years but am too impatient to spend so much time on a project. Especially with ported or horn-loaded cabinets. Selecting the drivers, using pencil-and-paper or spreadsheets or speaker design programs to match cab characteristics to the drivers. Designing a "first estimation" passive crossover, and then after the prototype has been built, measuring the speaker and identifying issues which might be solvable by tweaking the crossover values and details. Passive notching out driver peaks, matching levels between tweeter and woofer, etc.

My ignorance is extreme enough perhaps to border on delusional, but I tried to shortcut a lot of that with rules of thumb, electronic crossovers and EQ. Like what a fella might do putting together a good loud PA system. The big bass bins, mid-bass bins, mid horns and super tweeters tend to have rather rough frequency responses with individual quirks. Each section has its different efficiency and such. So you just wire them up and then gross-balance levels between the different bins, and then fine-tune EQ on each section until the total system is working together as best possible.

So far as I can find out, woofers with a QTS in the ballpark of about 0.5 tend to work purt good in sealed cabs. If you "lightly fill" a sealed cab entirely full of fluffy fiberglass or polyester stuffing, it overdamps the cabinet enough not to have extreme resonant peaks unless it is caused by the driver. In other words, full of fluffy, but not compressed over-full of fluffy.

There are recommendations for sealed cab volume which seem fairly reliable, for the QTS = 0.5 speakers. For instance, a 12" speaker with QTS about equal to 0.5 ought to do "OK" in a sealed cab of 1 or 2 cubic feet volume. It will probably work better if the different cab dimensions are not identical. IOW, something that is not a perfect cube. It will work better if the cab is built very non-resonant.

It would not be a good way to design something to get optimum amplitude or optimum flatness or whatever. If you wanted to build a concert system as loud and inexpensive as possible, or a speaker as flat as possible directly connected to an amp with no EQ, it wouldn't do at all. But I (perhaps erroneously) figure if the application doesn't need real loud levels and you can afford to throw in over-specified drivers and amps, then there is enough wiggle-room to tweak out whatever oddities are encountered.

For instance if a certain woofer in a certain sealed cab turns out not quite loud enough at 100 Hz compared to 200 Hz, it is a big problem if the woofer and amp can't handle a 100 Hz boost. But if there is plenty of woofer capacity and amp capacity to handle the boost, it ought to work fine. And in a practical residential small treated room, it will probably need some EQ even if you bought a caviar speaker costing thousands of dollars. Room problems are worse than speaker problems even in a fairly-well-treated small amateur treated residential room.

That was the approach anyway. If the average nearfield listening level target is 90 dB (as in my case) and you put in speakers that ought to be fairly clean with no stress up to 120+ dB, then there ought to be lots of wiggle room to tweak out irregularities.

[Laguna Rising]

This is very interesting.
I couple of years ago I built a bass cabinet, but that was a far simpler business than this project of yours.
Thanks for posting a detailed report

Cheers