Hello! I am new to DIY speakers, and I see many warnings against building a 3 way if you're a beginner. Why is that? I know the crossover design is inevitably going to be more complex, but is that it? Thanks!
Member
Joined 2009
Paid Member
Because you're trying to integrate 3 drivers instead of two.
If you were learning to ride a bike, would you start out on the steepest hill around with the most traffic? Or would you start on level ground with a minimum of traffic and obstructions?
If you were learning to ride a bike, would you start out on the steepest hill around with the most traffic? Or would you start on level ground with a minimum of traffic and obstructions?
Problem begins long before you get to the crossover design, that really is much harder to do right. You need to make good set of gated measurements with reflection free zone up to about 10ms so you would know what's going on with woofer and midrange down where you need to cross them over. Woofer nearfield measured response or sim of low end response should then be merged between 100 and 200Hz, bellow baffle step, with 1m measurement you've made. Upper crossover point usually is not an issue. Component values for a three way are huge on lower crossover point so, naturally, the price is high. No room for experimenting with topologies or it can easily end up being more expensive than the drivers itself.
DSP and multiamping are making things much easier, of course.
DSP and multiamping are making things much easier, of course.
Last edited:
I don't see why building a three way would not be a good idea for a DIY beginner. You might take a look here.
If you want to design a three way, then yes, a three way might be more difficult than a single driver loudspeaker or a two way -- but not necessarily always. No matter how many drivers there are, it is difficult to design a speaker from scratch. Not something you'd choose for a first time project as a DIY beginner.
If you go with active crossovers, i don't think a 3-way is too much for a beginner...or even a 3-way with a sub for that matter.
Anyway, don't set the target too high at the first few times! Use cheap drivers and make the box as easy and ugly as you can. Good drivers are ok if you are careful and don't burn them while testing...
Learning to measure and simulate, even with dsp takes lots of time and errors! After say 3-5 project you start to know your abilities and can start investing on good drivers, xo parts, pretty and heavy cabinet work and finish too.
Do you have previous experience with electronics, carpentery, acoustics, measurement tech, simulation software, paintjob etc?
Learning to measure and simulate, even with dsp takes lots of time and errors! After say 3-5 project you start to know your abilities and can start investing on good drivers, xo parts, pretty and heavy cabinet work and finish too.
Do you have previous experience with electronics, carpentery, acoustics, measurement tech, simulation software, paintjob etc?
Member
Joined 2009
Paid Member
Correct measurements (and ditto assumptions or theoretical knowledge) in the low midrange are the real challenge for a beginner. Simulating and crossover configuring isn’t any harder than with a standard two way design.
2-way is easiest to design if target is bad sound. More ways make some things easier and sound more healthy and clean. XO design is simple with modern methods.
Guys, he asked about building a loudspeaker, not about designing one. Building a tried and tested three way design is just as easy / difficult than a one/two/four/whatever way if you ask me.
Member
Joined 2009
Paid Member
Contrary opinion here because you can do what-ever you want. Not being able to afford the good drivers needed to make a good 2-Way speaker my first DIY was a 3-Way. Mind you I did a lot of reading and bought some books and used the good general guidelines about such speakers before I started cutting wood. I actually find the woodwork aspect of building speakers far harder than designing crossovers etc and I don't measure yet/still I rely on my partners golden ears and the ears of my grand daughter and I am prepared to do a lot of tinkering A lot of tinkering. If you are prepared for less than perfect then just go for it
That's exactly my point. The OP wrote "I know the crossover design is inevitably going to be more complex". My understanding is that he refers to an existing design, which may have a more complex x-over than a two way. I don't think his idea is to develop his own design.
Member
Joined 2009
Paid Member
Well, in that case, one option is to get a kit from the likes of Troels Gravesen. He has done all the hard work to design the cross over and you can buy all the parts. I've read people say the parts are rather pricey though but the quality of the result is not in question so it may be a case of you get what you pay for.
But not from him, unless of course you want to pay a huge premium for overpriced parts.
Total agreement here. It just depends on what you're after.
If you're looking to have some fun and get your feet wet in a new hobby, give it a try. Don't spend much at first and don't worry about fancy, expensive brands of components. Build, listen, learn, rebuild, etc- Just enjoy it.
However, if you expect to drop a significant (to you, not us) amount of money, or have high expectations for a first build, then start with plans for a known design or a kit.
Check out the introductory sticky threads in this forum.
You can build a very simple, but effective 3 way, but at a cost. Driver cost & crossover parts cost. Oh yeah, you will need a fairly large enclosure too. And, limit the usable low end to around 40 Hz. The efficiency to around 88 db/watt. It comes down to driver selection and retaining a minimal number of crossover parts. Using natural driver roll offs as much as you can. A wide range mid range is the key. Along with a large enough woofer able to move adequate air that is clean to 2.5 Khz. Sealed box. No traps required on any of the drivers. Maybe one Zobel. Yes, I am teasing but serious. As is usually the case, it comes down to effective, efficient engineering. With all that said, for a beginner to understand all of the issues, choices and work required to pull it all together, it is a major, I mean MAJOR task.
I will post additional info about what I have intimated if anyone is interested. Do not want to give away the farm, but interested in seeing what direction this build will take without my intervention. John
I will post additional info about what I have intimated if anyone is interested. Do not want to give away the farm, but interested in seeing what direction this build will take without my intervention. John
Member
Joined 2009
Paid Member
for most of the projects the crossover details are not public, you have to buy from him or you don’t buy at all