crossover design

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I know crossover design is tough. For the novice, it's best to use an existing design. But assume I want to use a driver combo for which there's nothing available, and assume also that I am not afraid of failure. What's the best fighting chance for someone who doesn't have a lot of experience or test equipment?

I've got the Weems book Great Sound Stereo Speaker Manual. It came with an old DOS program called CMP.exe. It uses driver-data files with .FRD and .ZMA extensions. Is it any good? Are there data files on the web or something for drivers other than those in the book? Is there a reasonable way to generate the files from the manufacturers' frequency response graphs?
 
Depends on what you find acceptable. Your chances of doing something really good without test equipment and a decent CAD program are pretty slim, unless you've got a LOT of talent, experience, and time.

Test equipment and speaker CAD software are cheap these days, the cost of a couple of good drivers. Why go fight the fight unarmed?
 
SY said:
Depends on what you find acceptable. Your chances of doing something really good without test equipment and a decent CAD program are pretty slim, unless you've got a LOT of talent, experience, and time.

Test equipment and speaker CAD software are cheap these days, the cost of a couple of good drivers. Why go fight the fight unarmed?

Okay. How do I arm myself?
 
I assume you've got a computer and a soundcard. Now you need Speaker Workshop or something like that for FFT/MLS measurement of frequency response and impedance. Speaker CAD program like the pay version of Speaker Workshop or CALSOD. You need a test mike (easy to make from a Panasonic electret capsule; see, for example, the article at www.linkwitzlab.com/sys_test.htm#Mic). And you need a copy of Dickason's Loudspeaker Design Cookbook and d'Appolito's Measuring Loudspeakers.

There are also some freeware spectrum analyzers out there, which when used with your test mike and a pink noise source (test CDs are wonderful things!) will tell you a lot about spectral balance in-room.

Doing proper from-scratch design of speakers looks intimidating, but it isn't. There's a pile of stuff to learn, but none of it is conceptually difficult, and it's very, very rewarding.
 
SY said:
And you need a copy of Dickason's Loudspeaker Design Cookbook


I found this book very overrated.
I got one, after many advices here. And I didn't learn much things.
This book has a biiiig chapter on bass reflex, with equations and everything that is now done with every software. And everything else isn't very well explained, especially for crossovers

I was expecting more from this book, that some people condider as a reference.
 
Well?

Can someone please point me in the right direction here? If I need software and test equipment to have a fighting chance, what software do you recommend? The LEAP 5 crossover software costs $800 and then I need microphones and I don't know what-all else.

Is the answer, "Give up. Mere mortals like you have no hope"? (I don't care for that answer.) :)
 
I use Calsod (it ran about $65) for CAD. Clumsy and DOS-based, but it works. LEAP is top-of-the-line and more than you probably need to get started.

The mike will run you a couple of dollars.

And there's TONS of inexpensive measurement software; Speaker Workshop has both CAD and measurement capability and is much less expensive than LEAP. You can put together all the tools you need for maybe $300. Compare that to the price of a couple of ScanSpeak midranges or a good table saw.

BTW, Dave, you're only about an hour drive away. If you want to see a demo on how powerful these cheap tools can be, drop me an email.
 
Loudspeaker Design Cookbook....

It doesn't answer all the questions but still a very good reference. I find it useful to go back over a chapter after building a speaker - locks into the grey matter better.

I would have liked some examples using the formulae in the book to help the understanding instead of giving a description of the computer modelling process - OK if you have the same software. A speaker designed completely manually would have been great.

But this is part of the fun of the DIYer.
 
There's no need to give up!

Speakerworkshop is free although slightly rough and unintutive (www.speakerworkshop.com), a jig is recommended (about $20) and so is a mic pre amp (another $10). A basic Panasonic mic capsule is about $3 from Digikey or a calibrated version can be had for about 30 from Kim Guardin (sp?). A calibrated capsule is not essential IMHO as the basic mic is more or less flat (no worse than +/- 2dB IIRC).

I have my own unfinished jig schematics at...
http://www.apcl43.dsl.pipex.com/SW_jig.htm

... but Eric Wallins site has the 'official' version and he also includes a tutorial:
www.gti.net/wallin

Good luck,
David.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.