I've decided to build a 3-way system from scratch. The only decision I've made so far is that I'd like to use JL Audio 10W0 subwoofers, and probably Peerless drivers (no models selected yet).
My knowledge of speaker design is limited to some equasions I found for determining box volume based on Q and Vas.
What are the general steps that a designer/builder uses to come up with a good design before making sawdust?
And what tuning and measurement steps do you take afterward?
I know each builder might have his own methods, but I'm looking for advice on how to proceed with the intention of having better than average results. Thanks for any insights you can offer.
My knowledge of speaker design is limited to some equasions I found for determining box volume based on Q and Vas.
What are the general steps that a designer/builder uses to come up with a good design before making sawdust?
And what tuning and measurement steps do you take afterward?
I know each builder might have his own methods, but I'm looking for advice on how to proceed with the intention of having better than average results. Thanks for any insights you can offer.
First, read a good book on the subject of speaker design and building such as Vince Dickason's The Loudspeaker Design Cookbook. Also, something like Alton Everest's Master Handbook of Acoustics might prove useful. Second, reconsider your choice of subwoofer. If you know nothing of speaker design, equations, measurement techniques, etc., then how could you have picked an appropriate subwoofer already? If you have several lying around already I suppose that could be a valid reason.
Starting with a 3-way isn't the recommended way of learning the art of speaker design. Most people build a few 2-way systems over the course of a couple of years before attempting their first 3-way.
Starting with a 3-way isn't the recommended way of learning the art of speaker design. Most people build a few 2-way systems over the course of a couple of years before attempting their first 3-way.
OK, so you've already read a book and seem willing to learn. I'm no expert by any means, so I'll just state the obvious to get the ball rolling.
At this point, you need to answer a few questions that will help direct your design.
(1) What are the primary listening uses of the speaker? Music, movies/theater, multichannel or stereo...? What style music?
(2) What is your budget for all parts and materials? Do you intend to include software in that budget, spend extra for software, or try to use only freeware tools?
(3) What size room do these speakers need to fill?
(4) Do you already have an amplifier or receiver that you want to power them? Are you willing to upgrade to a more powerful amp if required?
(5) Will they be located in a bookshelf? Can you position them away from walls? Do you prefer the look of floorstanders, or smaller bookshelf + separate sub modules?
These questions will help you decide whether to aim for high efficiency, or low cost drivers, or small physical size, or a front or rear ported or sealed system, or some combination of these and similar targets. When you answer these questions, I'm sure others can begin to point you in the direction of appropriate kits, or perhaps even a good starting point for a truly original design.
At this point, you need to answer a few questions that will help direct your design.
(1) What are the primary listening uses of the speaker? Music, movies/theater, multichannel or stereo...? What style music?
(2) What is your budget for all parts and materials? Do you intend to include software in that budget, spend extra for software, or try to use only freeware tools?
(3) What size room do these speakers need to fill?
(4) Do you already have an amplifier or receiver that you want to power them? Are you willing to upgrade to a more powerful amp if required?
(5) Will they be located in a bookshelf? Can you position them away from walls? Do you prefer the look of floorstanders, or smaller bookshelf + separate sub modules?
These questions will help you decide whether to aim for high efficiency, or low cost drivers, or small physical size, or a front or rear ported or sealed system, or some combination of these and similar targets. When you answer these questions, I'm sure others can begin to point you in the direction of appropriate kits, or perhaps even a good starting point for a truly original design.
As was said, get the T/S parameters and decide what type of design you're going after. Sealed, ported, TL, etc.. There are several resources on line for doing calculations for box size and ports if necessary.
After that is when things get more involved, frustrating and time consuming. Read this. From the table of contents find "Articles" then "How to Design a Speaker".
RHosch said:
Thank you. That's what I'm looking for. I don't want anyone to design my speakers for me, only to share an outline of what steps people use to come up with a successful design that comes close to expected results when built. Maybe that's a lot to ask...
I very much want to design the system myself. Don't want a kit or someone else's premade project. From what I've read, a sealed enclosure suits my tastes. And I really want a 10" 3-way (just because).
1) Primary use is main speakers in a stereo system. All types of music, mostly rock and classical and sometimes HipHop or rap (my kid's music). I like clean midrange and deep bass. My ears suck (too many rock concerts) for highs, but I don't want harsh highs that annoy friends and family.
Most often the listening level will be VERY low, ambient music. Once in a while it'll be cranked up. I'm sure this is a consideration on driver choices.
2) Prefer to use freeware computer tools, as I don't intend to do this again. I already have electronic test equipment that I've collected for 30 years (I'm an EE and always been interested in audio). Budget for raw drivers is maybe $250.
3) Room is a hardwood floor apartment dining room 12X16, but that can change. I want these speakers to last me 20 years.
4) Amplifier power is not an issue. In fact my reason for making these is to have test speakers that I can use for my DIY amp project. I intend to use passive crossovers for now off a 20 watt stereo, then switch to active when my amp is built.
5) The cabinets will be floor mounted, close to walls.
Do I pick drivers next? Thanks again.
Looks like a standard 3-way tower will suit your application just fine. The comment about relatively low listening levels is important, as it IMO opens up the possibility of using first order crossovers... at least between tweeter and mid. That would have an impact on suitable drivers.
Your next step, having apparently settled on a sealed 3-way floorstander, is to settle on a "general" configuration that you like. MTM with integrated sub? Standard tweet/mid/woofer tower? First, second, or fourth order slopes? (I'm not too fond of 3rd order, but that is my preference).
I'd suggest one of two routes to take at this point regarding the crossover: first order with wide bandwitdh high sensitivity drivers if you want to use that 20wpc channel receiver indefinitely, or fourth order Linkwitz-Riley crossover design with the assumption that the passive network will soon be replaced with a more accurate active one.
Your next step, having apparently settled on a sealed 3-way floorstander, is to settle on a "general" configuration that you like. MTM with integrated sub? Standard tweet/mid/woofer tower? First, second, or fourth order slopes? (I'm not too fond of 3rd order, but that is my preference).
I'd suggest one of two routes to take at this point regarding the crossover: first order with wide bandwitdh high sensitivity drivers if you want to use that 20wpc channel receiver indefinitely, or fourth order Linkwitz-Riley crossover design with the assumption that the passive network will soon be replaced with a more accurate active one.
With a 20W amp, you'll definitely want to reconsider using JL's. My experimenting with car audio subs in a home environment is that they just don't work out too well. They are designed to be used in cars with alot of cabin gain, and it's a different ball game when put in a room with little gain.
asauer said:With a 20W amp, you'll definitely want to reconsider using JL's. My experimenting with car audio subs in a home environment is that they just don't work out too well. They are designed to be used in cars with alot of cabin gain, and it's a different ball game when put in a room with little gain.
RHosch said:
Whut's MTM mean?
What significance is the slope, other than the ability to control overlap of the drivers? I would have picked 2nd order filters because of simplicity, and because I've read that too many stages introduces phase shifts. Dunno what that sounds like acoustically, but I understand the electrical reasons.
I'll also probably design 12dB/Octave adjustable filters into the amp, so I guess 2nd order is appropriate passively as well.
As for power - is it generally true that speakers that can handle high power sound crappy at low levels? Or is it possible to build a system that sounds great both soft and loud?
The reason I ask is that the Dayton and Vifa drivers are pretty low power (under 50 watts) rated, and the Peerless ones I've looked at are more like 100 watts. I was planning to build my amp 50 watts per section, but like I said it'll be way lower than that 90% of the time.
And the JL's have found a home in my nephew's truck... so I need subs too.
MTM = mid, tweeter, mid which is the order of the drivers vertically from top to bottom. Variations, MMT, TMM, etc.. MTM is also known as D'Appolito.
The order of the XO slope helps keep the FR out of the drivers resonance. This is especially important with tweeters. Unless you're using robust extended range drivers a minimum of 2nd order is a good idea. A LR 4th order can give you a lot of control but it takes a lot of pieces which compromises other things such as transient response and efficiency. I've heard the 3rd orders are less preferred because the drivers are 180 degrees out of phase but I don't know anyone that can detect a phase change when blindfolded.
It's possible to build something that sounds good loud and soft. However, in most cases some bass boost is needed at low volume levels (Back you purists, BACK!)
Don't get hung up on power ratings. They mean little, maybe nothing. In general, if your amps rated output is higher then that of the drivers that's a good thing (a.k.a. "headroom"). Consider efficiency ratings.
Ding dong, the JLs are gone! You don't really want to use them anyway.
The order of the XO slope helps keep the FR out of the drivers resonance. This is especially important with tweeters. Unless you're using robust extended range drivers a minimum of 2nd order is a good idea. A LR 4th order can give you a lot of control but it takes a lot of pieces which compromises other things such as transient response and efficiency. I've heard the 3rd orders are less preferred because the drivers are 180 degrees out of phase but I don't know anyone that can detect a phase change when blindfolded.
It's possible to build something that sounds good loud and soft. However, in most cases some bass boost is needed at low volume levels (Back you purists, BACK!)
Don't get hung up on power ratings. They mean little, maybe nothing. In general, if your amps rated output is higher then that of the drivers that's a good thing (a.k.a. "headroom"). Consider efficiency ratings.
Ding dong, the JLs are gone! You don't really want to use them anyway.
Just a quick note, for a fisrt attempt I wouldn't try a 3 way design, as all aspects of the crossover interact, it just just about impossible to build straight from theory. You need experience in design and testing, swapping components etc. to fine tune the design. Try a two way first, it's a lot less stress. 
Thanks for the advice... but I already have a 2-way set that sounds pretty good.
So, assuming I'm settled on a 3 way system with 2nd order crossovers, what criteria do you use for choosing drivers (besides cost)?
I'd like to make the enclosure compact (not a tower). A 10" woofer might need .75 cu. ft. of box, that's not much. How do you choose between a 5" or 6.5" mid, and all those zillions of tweeters out there?
So, assuming I'm settled on a 3 way system with 2nd order crossovers, what criteria do you use for choosing drivers (besides cost)?
I'd like to make the enclosure compact (not a tower). A 10" woofer might need .75 cu. ft. of box, that's not much. How do you choose between a 5" or 6.5" mid, and all those zillions of tweeters out there?
maylar said:
For a start, until you have built the boxes and measured the selected drivers response curves, you can't specify the crossover order!
As for picking drivers, you need to do a lot of research. Firstly, pick your budget, bearing in mind you are likely to spend at least 50% more than you think overall.
Then pick a bass driver, as this is the driver that will give you all the box size issues to worry about. Individual driver stats are not important at this stage, as you can play with different drivers to get a bass response/box size that you like. However, you should look at the FR curve to make sure there are no nasties to worry about, and IMHO, the higher the sensitivity the better.
The mid range should be picked to give a good overlap with the bass driver, to allow flexibility in xover design, and again a good FR curve makes life a lot easier, as does matching the sensitivity. T/S parameters are not that important.
The tweeter is selected in the same manner, but matching sensitivity is usually more of a problem, as they tend to be much more efficient by their nature.
This is by no means exhaustive, but should get you going.
my $.02
I also think it would be wise to conseder a 2 way for first design... but I think what you want would be a 2.5 way ( 2way with built in sub for very low freq). This would be the best way to "get it right the first time" and avoid frustration
You could start by selecting a nice combo of midbass and tweeter that suit your musical taste and budjet. figure that the midbass will go down to 60-80hz or so (this is your design so final numbers are up to you and testing)
then pick a sub to power with a seperate 100-200w plate amp either built into one of the towers or in a seperate box.
You have a "3-way" look and feel with the control and "ease of design" of a two way.
A 2.5way design is also going to already be 1/3 active. So, when you complete your amp, you only have the midbass and tweet to worry about.
or not ...just my two cents.
Whaddya think?
I also think it would be wise to conseder a 2 way for first design... but I think what you want would be a 2.5 way ( 2way with built in sub for very low freq). This would be the best way to "get it right the first time" and avoid frustration
You could start by selecting a nice combo of midbass and tweeter that suit your musical taste and budjet. figure that the midbass will go down to 60-80hz or so (this is your design so final numbers are up to you and testing)
then pick a sub to power with a seperate 100-200w plate amp either built into one of the towers or in a seperate box.
You have a "3-way" look and feel with the control and "ease of design" of a two way.
A 2.5way design is also going to already be 1/3 active. So, when you complete your amp, you only have the midbass and tweet to worry about.
or not ...just my two cents.
Whaddya think?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.