I'm building two 3-way boxes, each box will have a crossover board separating the signals for each speaker. I will use the following woofers, midrange and tweeters:
tweeter mid woofer
I would like to know which crossover is ideal for this kit, because when I chose one I found its filter very strange for the midrange (from 1000hz to 7000hz).
What are the ideal frequency cuts for a 3-way kit? Can I find a crossover on AliExpress with the ideal frequency cuts?
From the specifications I find of crossovers on the internet, how can I calculate at what frequency the frequency cut will be?
tweeter mid woofer
I would like to know which crossover is ideal for this kit, because when I chose one I found its filter very strange for the midrange (from 1000hz to 7000hz).
What are the ideal frequency cuts for a 3-way kit? Can I find a crossover on AliExpress with the ideal frequency cuts?
From the specifications I find of crossovers on the internet, how can I calculate at what frequency the frequency cut will be?
No pre-made crossover will work properly for any set of drivers and no crossover calculator will work either, for many reasons. The ideal crossover for any project is one which is custom made for those drivers and the cabinet you're going to use.
Have you had a look for other projects which use those or similar drivers, that might give you a few ideas.
Otherwise, AllenB's advice shows you the way to go.
If you don't have measuring equipment, or only plan to make one set of speakers, I'd suggest using an existing project, if you haven't already bought the drivers
Geoff
Have you had a look for other projects which use those or similar drivers, that might give you a few ideas.
Otherwise, AllenB's advice shows you the way to go.
If you don't have measuring equipment, or only plan to make one set of speakers, I'd suggest using an existing project, if you haven't already bought the drivers
Geoff
The point is that I needed a pre-made board because I have no idea yet and I haven't even studied enough about how to make one for myself
I looked up the drivers you've selected - all from Ali Express, I think - and there's no frequency response graph for the woofer, which doesn't help things.
The tweeter has rather a bright looking response which will need treatment in the crossover and there are no distortion graphs for any of the drivers, so what might look OK in terms of frequency response might not sound right. The mid range frequency response doesn't look that bad, but all the drivers' data is supplied by the manufacturer, so may not be accurate.
The crossover points for any system will depend on the drivers: what you're looking at is an 8" 3 way system, if you haven't bought the drivers there are many projects of that nature out there. They all have different crossover points, depending on the drivers used: for example, some have XO points at 400 and 2,500Hz, others at 300 and 2,000Hz.
If you use a pre-made XO, or make one based on crossover calculator parts values, it will make noise but almost certainly sound less than ideal and you'll think 'this DIY thing is over-rated' and you'll have likely wasted your hard earned money.
Geoff
The tweeter has rather a bright looking response which will need treatment in the crossover and there are no distortion graphs for any of the drivers, so what might look OK in terms of frequency response might not sound right. The mid range frequency response doesn't look that bad, but all the drivers' data is supplied by the manufacturer, so may not be accurate.
The crossover points for any system will depend on the drivers: what you're looking at is an 8" 3 way system, if you haven't bought the drivers there are many projects of that nature out there. They all have different crossover points, depending on the drivers used: for example, some have XO points at 400 and 2,500Hz, others at 300 and 2,000Hz.
If you use a pre-made XO, or make one based on crossover calculator parts values, it will make noise but almost certainly sound less than ideal and you'll think 'this DIY thing is over-rated' and you'll have likely wasted your hard earned money.
Geoff
Do you have any access to speaker building books for beginners?
There really isn't any "IDEAL" crossover because all passive crossovers are a compromise; but the old school telephone bandwidth of 300Hz to 300Hz is what I started with and if I have no other information I use that octave spread as a starting point. A lot has to do with budget and I am assuming that your budget is very limited. Have you had a look at the locally made PRV speakers rather than shopping from China?
No you really will need to make your own crossover, using separate parts secured to a scrap of something and wired/soldered together
We're not trying to put you off, nor being snobs nor perfectionists.
You are unlikely to get a smooth frequency response with some random premade crossover. this is because a premade crossover assumes:
1. Drivers of matched sensitivities
2. Drivers with matched impedances
3. Drivers with uniform impedance (all moving coil drivers have a varying impedance profile)
4. Drivers without any breakup (i.e. smooth frequency responses)
5. No baffle step compensation required (many premade crossovers do not attempt baffle step compensation). This means your speaker will sound bass weak / bright / thin (midrange and treble heavy)
So - in short don't do it (or be disappointed if you do and poorer). Learn how to design a crossover - best with measurements, but you may get away with designing from manufacturer supplied curves if they can be trusted AND you know how to adapt them to your enclosure.
Basically there are no short cuts doing a custom design (2 or 3 way). Much the same as there are no shortcuts to being a fighter pilot.
You are unlikely to get a smooth frequency response with some random premade crossover. this is because a premade crossover assumes:
1. Drivers of matched sensitivities
2. Drivers with matched impedances
3. Drivers with uniform impedance (all moving coil drivers have a varying impedance profile)
4. Drivers without any breakup (i.e. smooth frequency responses)
5. No baffle step compensation required (many premade crossovers do not attempt baffle step compensation). This means your speaker will sound bass weak / bright / thin (midrange and treble heavy)
So - in short don't do it (or be disappointed if you do and poorer). Learn how to design a crossover - best with measurements, but you may get away with designing from manufacturer supplied curves if they can be trusted AND you know how to adapt them to your enclosure.
Basically there are no short cuts doing a custom design (2 or 3 way). Much the same as there are no shortcuts to being a fighter pilot.
I'm going to 'stick my head out' and give you an example of a XO that would work with the bandwidth of many drivers.
The frequencies you wanted to know about are > 700Hz and 3Khz.
With a pre-made XO like this, you could FLUKE sound that you like,
however, there is no guarantee of this.
https://www.newark.com/mcm/50-786/3-way-crossover-network-12db-octave/dp/80R6840
PS.
Some people have Bass & Trebble controls and others don't. These can be useful with a pre-made XO.
The frequencies you wanted to know about are > 700Hz and 3Khz.
With a pre-made XO like this, you could FLUKE sound that you like,
however, there is no guarantee of this.
https://www.newark.com/mcm/50-786/3-way-crossover-network-12db-octave/dp/80R6840
PS.
Some people have Bass & Trebble controls and others don't. These can be useful with a pre-made XO.
In Brasil very good speaker chassis are produced. Import taxes are sky high. So money you need to buy the Chinese stuff you think of, should get you much more value if you go and "buy local". Also, you may still find a workplace in the future, in Brasil, not China.
I don't know how to say this without being rude: One of the many things you don't know how to do by now, is designing a speaker.
Try to find a good, reviewed kit and build the cabinet exactly following the plan. Don't get creative (except for color) and don't improve anything, because "you heard."
Then, after building some kit's and understanding them, you may build your own. In a few years.
I don't know how to say this without being rude: One of the many things you don't know how to do by now, is designing a speaker.
Try to find a good, reviewed kit and build the cabinet exactly following the plan. Don't get creative (except for color) and don't improve anything, because "you heard."
Then, after building some kit's and understanding them, you may build your own. In a few years.
Polarity for some drivers
might need to be reversed.
Often the woofer or tweeter
Sometimes both. Are reversed
So the crossover frequency doesn't cause
peaks or dips.
Even thou as mentioned
I prebuilt crossover not always ideal.
But the generic values can work.
And also likely require one or 2 drivers
to have polarity +/- reversed.
It is possible the midrange sounds strange
because at woofer crossover point causes
cancellation of mid frequencies.
Reverse polarity of woofer and see if it sounds
better.
After listening and find better setting.
Also reverse polarity of tweeter
and find which sounds better.
crossover design in sim is long learning
curve but normal approach without a microphone.
we have Frequency and impedance data for mid
and tweeter. But none for the woofer.
So in future buy drivers with Frequency data and impedance
data. Graph or file. To be able to start learning
crossover design in software.
might need to be reversed.
Often the woofer or tweeter
Sometimes both. Are reversed
So the crossover frequency doesn't cause
peaks or dips.
Even thou as mentioned
I prebuilt crossover not always ideal.
But the generic values can work.
And also likely require one or 2 drivers
to have polarity +/- reversed.
It is possible the midrange sounds strange
because at woofer crossover point causes
cancellation of mid frequencies.
Reverse polarity of woofer and see if it sounds
better.
After listening and find better setting.
Also reverse polarity of tweeter
and find which sounds better.
crossover design in sim is long learning
curve but normal approach without a microphone.
we have Frequency and impedance data for mid
and tweeter. But none for the woofer.
So in future buy drivers with Frequency data and impedance
data. Graph or file. To be able to start learning
crossover design in software.
Just to look at one variable, sensitivity...
According to the makers, the tweeter has a sensitivity of 91dB, the mid 89dB, and for the woofer it doesn't appear to say. We don't know how those are measured or how accurate they are (and in my experience, a lot of Chinese gear has fictitious specifications).
If the crossover isn't designed to match the sensitivities of the drivers, you'll probably get step changes in volume at the crossover points. There's a lot more to matching a crossover to drivers (as others have said, above) than just getting the right crossover frequencies.
By all means pick a crossover with suitable frequencies if you want to build some speakers for fun, but I wouldn't expect them to sound all that great.
(PS: That one with 1kHz and 7kHz crossover points seems absurd)
According to the makers, the tweeter has a sensitivity of 91dB, the mid 89dB, and for the woofer it doesn't appear to say. We don't know how those are measured or how accurate they are (and in my experience, a lot of Chinese gear has fictitious specifications).
If the crossover isn't designed to match the sensitivities of the drivers, you'll probably get step changes in volume at the crossover points. There's a lot more to matching a crossover to drivers (as others have said, above) than just getting the right crossover frequencies.
By all means pick a crossover with suitable frequencies if you want to build some speakers for fun, but I wouldn't expect them to sound all that great.
(PS: That one with 1kHz and 7kHz crossover points seems absurd)
Last edited:
If you are lucky, a pre build crossover will prevent your tweeter from burning through on the first loud sound. That's about all the positive you can say about them.
Now, the usual question: Why are they build and for sale everywhere if they don't work?
Simple answer: Because people can make money by selling them to people that don't know. Like you.
Don't waste your money, the advice you get in this forum is in most cases the best you can get on the whole internet. 100% for free.
Brasil is a huge country with 214 million people. There should be some DIYS speaker builder! Look for local forum's that know about good kit's and plans you can buy at your home. There may be some magazines. Even "my" small Germany still has two of them.
Now, the usual question: Why are they build and for sale everywhere if they don't work?
Simple answer: Because people can make money by selling them to people that don't know. Like you.
Don't waste your money, the advice you get in this forum is in most cases the best you can get on the whole internet. 100% for free.
Brasil is a huge country with 214 million people. There should be some DIYS speaker builder! Look for local forum's that know about good kit's and plans you can buy at your home. There may be some magazines. Even "my" small Germany still has two of them.
To second that, yes, if you want to build your own speakers, I'd definitely recommend finding and using an existing DIY design.
The reason is that speaker design is hard. As well as matching the crossovers to the drivers, the cabinet shape and size are important, as is tuning the bass port (if it has one), bracing, damping, component choice...
I appreciate it might be hard to find the components depending on where you live. But if you can at least find the drivers, could you build the cabinets and crossovers yourself? If you can handle a soldering iron (and you don't need an expensive one), crossovers shouldn't be too hard to build. Doing the cabinet work needs skill, but could you find someone locally to do that for you? (Presumably, you already have a plan for the cabinets - but there's more to it than just making a box).
Overall, I think you'd get a far more rewarding result if you build from a published design than just buying some drivers and crossovers and hoping for the best.
Last edited:
You might hope so, but looking around AliExpress offerings I'm not seeing it - some do have resistors that could presumably be changed, but you'd need the expertise to work out what you need. But yes, at the very minimum, I'd want pre-made crossovers to be adjustable for sensitivity and have clear instructions on how to do it.