[World Leader Pretend]

Hey, I have been around audio for a while an I have built my own subs and stuff, but I want to design my own speakers and I am clueless when it comes to crossover design.  I understand the principle but I want to know how to build them and how to separate crossover frequencies, etc...

Can someone give me a hand and describe or link me some advice? 

Thanks!

[bpape]

http://sound.westhost.com/lr-passive.htm

[World Leader Pretend]

I can search Google too you know, don't feed me the first hit! 

[bpape]

There's a reason it's first....  It's one of the most comprehensive, most linked to sources for that information.  There is good theory, good practical examples, a design spreadsheet to help you 'play' with options, etc.  Short of asking a professional to design one for you for free...

Bryan

[sts9fan]

That is one of the best sites on the web for DIY information imho.  Don't act so ungrateful when someone answers one of your questions. 

[S Clark]

Xover design can be done approximately fairly easily, or with precision with a great deal of expertise and measuring equipment, and difficulty.  If your goal is an excellent pair of speakers, build a design created by those more experienced and willing to share.  Also, GR-research, Madisound, Parts Express, all have kits that may be just what you need.

[hdspeakerman]

I recommend you visit RJB Audio and also the following link.
http://www.pvconsultants.com/audio/frdgroup.htm
Roman has a tutorial to help you design your own crossovers.  After that it is just sticking your nose in there and keep trying.  A good measuring rig and the knowledge of how to use it will go a long way.  Crossovers are a real DIY thing.  YOU have to just get in there and keep trying.
Howard

[TerryO]

That is one of the best sites on the web for DIY information imho.  Don't act so ungrateful when someone answers one of your questions. 

I agree, I was tempted to reply earlier today when I first saw this but felt that as a relative newcomer to this site I'd just back off. On the old Bass List, a response like that, toward somebody trying to assist, would have had a dozen people handing his head to him. Great bunch, with a *LOT* of expertise, but they didn't suffer fools.

My advice to our friend: Learn some manners and ask nicely. If the answer isn't quite what you'd like, ask for clarification or in a different manner. These forums make it hard at times to know what is on a person's mind, they may be in a hurry or assumed that your understanding may be more extensive than it actually is. The important thing to remember is that someone took the time to respond, which I would take as an obligation, on my part, to be polite.

Having said all that, my suggestion would be to construct a single driver "full range" speaker or construct an available kit (which can run from plans to the whole ready-to-assemble package).

Best Regards,
TerryO

[jules]

Yes, Rod Elliot's site is great and yes, the FRD consortium site is also fantastic! There are also useful books like Vance Dickason's Loudspeaker Design Cookbook to help you get started.

I'd also suggest that if you can, you should find a mentor with a experience in the field. To DIY, you're going to need some quality  measuring equipment and access to some computer programming software at a minimum but there are many variables that that are more easily solved by the wisdom of experience than by innumerable tests [and components] ... do you use a 1st., 2nd., 3rd., or 4th. order Xover? How do you choose a box volume to relate to the characteristics of your drivers? How do you avoid nasty box resonances? What are the best xover points for a given set of drivers?

This is a huge topic and I suggest that you find someone in your vicinity who you can visit and talk to directly to avoid lots of flailing around in the dark [and lots of discarded expensive capacitors  ]

jules

 
 

[World Leader Pretend]

That is one of the best sites on the web for DIY information imho.  Don't act so ungrateful when someone answers one of your questions. 

I apologize, I had already read through that entire article before posting this thread.    It just seemed rather convenient that someone would quickly link me to the first hit, it felt like a slap in the face.  Kind of a "get lost, and come back after you have searched" type of thing. 

Anyway, since you agree that it is a good reference I am grateful for your help, please understand that I was acting in haste... 

As for sounding ungrateful, I actually wasn't trying to (notice the smile). 

*sigh*

Back on topic, it looks like I would need lots of measuring equipment and a PC (I have an excellent Mac but few of the programs made for measurements work on it) and a set-up for testing.  I don't have a whole lot of cash and I would rather be able to actually build speakers rather than test crossover assemblies.  I originally planned on building GR kits, and that might be what I'll end up doing.  I love to design speakers, and modify existing designs.  How would the GR designs be impacted if I re-aranged the driver look but kept the enclosures the same sizes.  Would the kit still work fine, or would I have to modify some of the components? 

Again, I apologize for my harsh comments.

[Daryl]

I would not recomend the ESP site at all.

The FRD I would recomend.

Speaker design is very interesting but I would warn you that it covers many varied subjects.

[Russell Dawkins]

I think you assume speaker design is straightforward.
You say this:
I want to design my own speakers and I am clueless when it comes to crossover design.  I understand the principle but I want to know how to build them and how to separate crossover frequencies, etc...
and then you post this:

Back on topic, it looks like I would need lots of measuring equipment and a PC (I have an excellent Mac but few of the programs made for measurements work on it) and a set-up for testing.  I don't have a whole lot of cash and I would rather be able to actually build speakers rather than test crossover assemblies.  I originally planned on building GR kits, and that might be what I'll end up doing.  I love to design speakers, and modify existing designs.  How would the GR designs be impacted if I re-aranged the driver look but kept the enclosures the same sizes.  Would the kit still work fine, or would I have to modify some of the components? 

These questions are, frankly, hopelessly general.

I'd say the shortest path towards "mastery" of speaker design would be to buy a good kit, a GR kit would probably be a good choice, and then spend a year or two to try to improve it. If you succeed, then you'd be that rare and brilliant exception.

If you don't, don't feel bad - maybe, like most of us, you're not cut out for speaker design.

If it were easy we'd be surrounded by perfect speakers.

[Daygloworange]

If it were easy we'd be surrounded by perfect speakers.

Ain't that the truth! 

Cheers

[World Leader Pretend]

Thanks Russell, as you can tell I am hopelessly lost so far, and I don't expect to be enlightened overnight.   

I think I'll order a GR kit or two and work on building through July. 

I have some cool looking designs that I would love to share, but I don't have a scanner 

Thanks for the help, I guess I was making sure that there wasn't some "magic bullet" that made crossover design easier. 

WLP

[bpape]

If you want a leg up, you may want to talk to Rick Craig from Selah Audio and have him work up a custom design for you to build.  Ask him to send you the interim iterations as he goes through the xover design so you can follow along with the plots, decision making process, why things are changed, etc. 

I'm not dissing Danny's kits at all.  They're just proven finished designs where you only see the end product and have no insight into the process along the way.

Bryan

[markC]

This is corny as hell, but.../. Grasshopper.... WHEN you can snatch the pebble from my hand...
(and perhaps a bit dated).

[Vapor Audio]

Thanks Russell, as you can tell I am hopelessly lost so far, and I don't expect to be enlightened overnight.   

I think I'll order a GR kit or two and work on building through July. 

I have some cool looking designs that I would love to share, but I don't have a scanner 

Thanks for the help, I guess I was making sure that there wasn't some "magic bullet" that made crossover design easier. 

Sure there's a magic bullet - experience!

Building a finalized kit is a good place to start.  I had a bit of a setback when my first attempt was a complete DIY design and build, it didn't come out very well.  So my next attempt was going with a kit, which of course did turn out very well.  That gave me more confidence, and since I've come a long way. 

The point is you don't want disappointment on your first try.  It's a learning process, and the act of assembling all the parts and seeing what's involved will teach you a lot.  From there you could get the measurement gear and start thinking about your own designs, but you gotta start somewhere. 

Other advice, don't go with this current trend of 'judging' speakers based on nothing more than their FR curves and distortion profiles.  That's a BIG mistake if you ask me, sure some drivers simply fail and shouldn't be used - but if you measurebate between a driver with .1% distortion at xxx hz and one with .15%, you're not gaining anything in the final design.  Those measurements when so close don't tell you how a driver sounds, and that's the tricky part, using your ears as the best engineering tool.