For decades large companies like Bose, JBL and the like show that with the right crossover, even the cheapest chassis can be used in an expensive loudspeaker and be sold to satisfied customers.
On the other hand, even the best and most expensive chassis on the planet will not sound except able, if the crossover is not matched.
A common problem with many first time speaker builders. They spent ages to find the right and best speaker for their build but have no idea how to construct a matching crossover. They think it is just an simple accessory that can be constructed by some online program.
They don't take you serious, if you write about the need to measure, simulate, try and error, building test baffles, mixed with extensive listening is the way to build a speaker.
The fact that a 3-way speaker is about 10 times more complicated to get right than a two way and close to impossible without years of real experience, is also not very popular.
The less you know about speaker building, the simpler it seems. Maybe not the only skill simpler talked about than done in real life.
On the other hand, even the best and most expensive chassis on the planet will not sound except able, if the crossover is not matched.
A common problem with many first time speaker builders. They spent ages to find the right and best speaker for their build but have no idea how to construct a matching crossover. They think it is just an simple accessory that can be constructed by some online program.
They don't take you serious, if you write about the need to measure, simulate, try and error, building test baffles, mixed with extensive listening is the way to build a speaker.
The fact that a 3-way speaker is about 10 times more complicated to get right than a two way and close to impossible without years of real experience, is also not very popular.
The less you know about speaker building, the simpler it seems. Maybe not the only skill simpler talked about than done in real life.
Thanks both for your added contributions.
@Turbowatch2, I understand what you're saying. I'm willing to learn, and get into measuring and things like that. I'd also be willing to make changes to my crossovers. Am I right in saying that as well as taking audio measurements, I'll also need to be able to measure the values of crossover components? And would you be able to point me in the right direction?
@skfisher, do you have any suggestions for drivers (especially midranges), other than what you already mentioned?
@Turbowatch2, I understand what you're saying. I'm willing to learn, and get into measuring and things like that. I'd also be willing to make changes to my crossovers. Am I right in saying that as well as taking audio measurements, I'll also need to be able to measure the values of crossover components? And would you be able to point me in the right direction?
@skfisher, do you have any suggestions for drivers (especially midranges), other than what you already mentioned?
I agree with Turbo, when I started out in the 80's I bought premade crossovers. I thought the speakers sounded really bad. I didn't blame the crossover because I didn't understand them. Later, I realized you could buy really low cost drivers and if you knew what you were doing with a crossover, you could make them sound really good. But it takes years to figure all this stuff out. You will never stop learning.
In the past, measuring was very expensive. In the 80's I traveled 300km one trip, just to get my DIYS build speakers measured and corrected. After many days of tuning by ear! Today, less than 100 $, € or £ will get you into the mesuring game.
First you need a microphone and a software to your liking. To keep things simple, an USB mike with a correction file (calibrated) would do.
Simple, but if you start with serious things, the USB mike will be useless and money lost, as it is not correct in the time domain.
A better, but a little more complicated set-up is an USD sound interface with XLR and 48V phantom power and a corrected measuring microphone.
Anyway, not the hardware, but a program you like is the most important question. I consider only free software.
If you want it all, get ARTA and don't look any further. If you are a quick learner and able to read a few pages at a time (not very common today), Arta will not only measure for you, but educate you about the physics behinf speaker building. Arta can be considered the most complete audio software imo.
If you want fast results, my pick is REW. It get's better, year by year.
Component measuring is not that important. You can do it with a sound card, some wire, crocodile clips and a few calibration resistors (1-5$).
A digital multimeter may help sometimes, but those with good low Ohm measurement are a little harder to find.
With you crossover, you may keep a few parts, like the large coil. You can buy special audio capacitors and resistors or save about 90% and buy the same quality as industry standard from ebay etc. You seem to be in the Brexit islands, makes things a little more limited and expensive. Look for vendors on eBay offering international service, eBay does all custom stuff for you. Many don't know that they can still buy in Europe, some never new...
First you need a microphone and a software to your liking. To keep things simple, an USB mike with a correction file (calibrated) would do.
Simple, but if you start with serious things, the USB mike will be useless and money lost, as it is not correct in the time domain.
A better, but a little more complicated set-up is an USD sound interface with XLR and 48V phantom power and a corrected measuring microphone.
Anyway, not the hardware, but a program you like is the most important question. I consider only free software.
If you want it all, get ARTA and don't look any further. If you are a quick learner and able to read a few pages at a time (not very common today), Arta will not only measure for you, but educate you about the physics behinf speaker building. Arta can be considered the most complete audio software imo.
If you want fast results, my pick is REW. It get's better, year by year.
Component measuring is not that important. You can do it with a sound card, some wire, crocodile clips and a few calibration resistors (1-5$).
A digital multimeter may help sometimes, but those with good low Ohm measurement are a little harder to find.
With you crossover, you may keep a few parts, like the large coil. You can buy special audio capacitors and resistors or save about 90% and buy the same quality as industry standard from ebay etc. You seem to be in the Brexit islands, makes things a little more limited and expensive. Look for vendors on eBay offering international service, eBay does all custom stuff for you. Many don't know that they can still buy in Europe, some never new...
Here's my two cents. I think you're strategy to use a "full range" is viable, but pick a good one (Dayton has some but I would spend for a reference series), that will go down to 100 - 125 Hz smoothly and up to say 10,000 Hz, or spend a bit more like the Tang Band W5-1611SAF 5" (avail at P-E), its about $80, these are quality drivers. Then add a quality 1" dome tweeter, I prefer fabric domes (P-E has several to choose from), and cross it relatively high at say 4,500 Hz. You can use a Dayton Audio XO2W-4.5K 2-Way Speaker Crossover 4,500 Hz, its a second order L-R. At that point, you're avoiding crossing in the midrange and the initial harmonics, so the xover is less audible. Be sure to match the driver efficiencies (about 90 dB) so you can avoid using an L-Pad. So essentially you'll build a 2-way speaker that gets to the mid bass level. Then, augment the low bass with a good powered subwoofer (either a Dayton kit say or buy one), and that will come with a variable xover and you play with it in the 100 Hz range to where you like it. Simple for a DIY project. Your main speakers can then be small on stands say to project into the room, and the sub placed where you like it (and the bass impact sounds best). I like to avoid large conventional woofers (12") as you get more overhang and less transient response, and they are not well suited to midrange, depending on where they are crossed.
If you want a custom xover, then you have to design it, buy all the parts, build it and will cost more. Or you can outsource to an electronics provider and will cost even more. But, say if you built two sets of speakers, one with the pre-built xover and one with a custom, and then A-B'd these in listening tests, you may or may not hear a significant difference, and ALSO depending on: the type of music you listen to, the volume you typically like, speaker placement (the famous spouse approval factor if you're married may limit you
, and your room acoustics/reflections. And quality of mixes/recordings factor in. Some folks will talk about phase issues, but there is a debate about phase sensitivity in human hearing (incl. brain processing) and with reflections in your listening room (unless you have a special acoustically treated one) you're dealing with phase soup anyway.Then you will be boosted by confirmation bias, you will want your project to sound good, so unless you're Mr. Expert, it likely will, the perception (or psychoacoustics) is hugely important. And you're friends are likely to say "wow you built those, they're great!" So don't let the mathematicians dissuade you.
The PC105 is pretty impressive for it's price.... measures good and sounds good.
https://techtalk.parts-express.com/.../1441425-driver-testing?p=1442182#post1442182
https://techtalk.parts-express.com/.../1441425-driver-testing?p=1442182#post1442182
If you really want to do this right, get a copy of the Loudspeaker Design Cookbook: https://www.hificollective.co.uk/books/bk5003.html
Reading it will probably depress you, because it will make clear how much more there is to this than meets the eye. But it'll get you going in the right direction. After that, if you want to try a short cut, get a simple simulation program like Basta (https://www.tolvan.com/index.php?page=/basta/basta.php).
By entering driver parameters, box and baffle sizes, you can then design your own passive crossover. If you build that crossover, it probably still won't work properly - you need measurements for that, and maybe something like Vituixcad. But you'll be further along on the road.
Reading it will probably depress you, because it will make clear how much more there is to this than meets the eye. But it'll get you going in the right direction. After that, if you want to try a short cut, get a simple simulation program like Basta (https://www.tolvan.com/index.php?page=/basta/basta.php).
By entering driver parameters, box and baffle sizes, you can then design your own passive crossover. If you build that crossover, it probably still won't work properly - you need measurements for that, and maybe something like Vituixcad. But you'll be further along on the road.
Hi,
I agree with that, but would like to add some #2c. Often people assume that the quality of a driver is related to it's price and this parameter can be used to judge while making purchasing decisions. From personal experience and having been burned by this assumption in the past, I think it's buguous premise, and drivers should be selected after extensive measurements and listening tests. Here comes the caveat that sometimes cheap drivers happen to be really well made and outperform many expensive siblings. So no surprise that big companies have made successful designs using cheapo loudspeakers : price is not always a predictor for quality.
I think this statement can make an impression that 3-way speaker building is something super complicated. Of course, a determined builder should be prepared to make test baffles, try out different drivers, crossovers, be proficient with DIY & know how to make measurements. But this is not rocked science. While combining measurements, listennings tests, DIY skills and a good common sense 3-way can be a fun and not that(extensively) difficult project.
Regards
I agree with that, but would like to add some #2c. Often people assume that the quality of a driver is related to it's price and this parameter can be used to judge while making purchasing decisions. From personal experience and having been burned by this assumption in the past, I think it's buguous premise, and drivers should be selected after extensive measurements and listening tests. Here comes the caveat that sometimes cheap drivers happen to be really well made and outperform many expensive siblings. So no surprise that big companies have made successful designs using cheapo loudspeakers : price is not always a predictor for quality.
I think this statement can make an impression that 3-way speaker building is something super complicated. Of course, a determined builder should be prepared to make test baffles, try out different drivers, crossovers, be proficient with DIY & know how to make measurements. But this is not rocked science. While combining measurements, listennings tests, DIY skills and a good common sense 3-way can be a fun and not that(extensively) difficult project.
Regards
: sqrt(400*600) or (400*600)^0.5