Hello, I've read many articles on xovers, freeware programs, etc...
I also have ERSE xover calculator... My question is...
If I'm going to go 4 way instead of 3 way, do I do the math for a 2way xover with the higher frequencies and connect it to the 3 way xover tweeter output?
Specifically, woofer 20-100hz, mid bass 100-800hz, high section 800 and above on a 3 way xover. If I were to divide the highest frequencies from 800 to 5000 Hz to a driver and from 5000 Hz and above to a separate tweeter, is it correct for me to calculate these xover points as a 2 way that already has been "high passed" to 800Hz ?
Thanks for any enlightment.
I also have ERSE xover calculator... My question is...
If I'm going to go 4 way instead of 3 way, do I do the math for a 2way xover with the higher frequencies and connect it to the 3 way xover tweeter output?
Specifically, woofer 20-100hz, mid bass 100-800hz, high section 800 and above on a 3 way xover. If I were to divide the highest frequencies from 800 to 5000 Hz to a driver and from 5000 Hz and above to a separate tweeter, is it correct for me to calculate these xover points as a 2 way that already has been "high passed" to 800Hz ?
Thanks for any enlightment.
Have you ever built a speaker before? I am guessing you haven't by your comments. Please build a proven design that has already been planned out and posted online. People have a hard enough time designing a two way crossover sometimes. Designing a three way is way more complicated and you want to build a four way on your first go at it?
Although DavidL's comments are respected for what they may be, they don't answer my question.
To make the question somewhat simpler:
Can anybody comment if this can be done?
The math involved will take forever: change a little variable and everything changes, start all over again. Example: rounded vs square edges...
PS: there are 4 kinds of persons:
Wise persons learn from other person's mistakes
Smart people learn from their mistakes
Dumb people never learn, not even from their mistakes
I would appreciate some constructive coments on how to apply the free software provided by ERSE.
To make the question somewhat simpler:
Can anybody comment if this can be done?
The math involved will take forever: change a little variable and everything changes, start all over again. Example: rounded vs square edges...
PS: there are 4 kinds of persons:
Wise persons learn from other person's mistakes
Smart people learn from their mistakes
Dumb people never learn, not even from their mistakes
I would appreciate some constructive coments on how to apply the free software provided by ERSE.
Thanks FatMarley !
I have that link and have read it several times.
It doesn't fully clear my doubt: for example, if I want to add a super tweeter, I take the xover tweeter output and add a lowpass filter for the tweeter and a high pass filter to the super tweeter. Ergo... 4 way...
If I take a 3 way system (800/5000) and build a 4 way system, I would like to split the woofer range into 2 elements: a low (20-80 Hz) and a mid bass (80-800 Hz) leaving the rest of the system untouched.
My question is: if I build/buy a 2 way 80 Hz xover and add a 800 low pass filter can I feed it from the 20-800 Hz woofer connector from my 3 way crossover ?
Thanks for your help !
I have that link and have read it several times.
It doesn't fully clear my doubt: for example, if I want to add a super tweeter, I take the xover tweeter output and add a lowpass filter for the tweeter and a high pass filter to the super tweeter. Ergo... 4 way...
If I take a 3 way system (800/5000) and build a 4 way system, I would like to split the woofer range into 2 elements: a low (20-80 Hz) and a mid bass (80-800 Hz) leaving the rest of the system untouched.
My question is: if I build/buy a 2 way 80 Hz xover and add a 800 low pass filter can I feed it from the 20-800 Hz woofer connector from my 3 way crossover ?
Thanks for your help !
Hi,
The answer theoretically to both questions is yes but
you clearly have no real idea what you are doing.
The Erse calculators are simplistic garbage and
the best way of using them is not using them.
rgds, sreten.
http://sites.google.com/site/undefinition/diy
(see if nothing else, the excellent FAQs)
The Speaker Building Bible
Zaph|Audio
Zaph|Audio - ZA5 Speaker Designs with ZA14W08 woofer and Vifa DQ25SC16-04 tweeter
http://audio.claub.net/Simple Loudspeaker Design ver2.pdf
FRD Consortium tools guide
Designing Crossovers with Software Only
RJB Audio Projects
Jay's DIY Loudspeaker Projects
Speaker Design Works
http://www.htguide.com/forum/showthread.php4?t=28655
A Speaker project
DIY Loudspeakers
Humble Homemade Hifi
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design
Great free SPICE Emulator : SPICE-Based Analog Simulation Program - TINA-TI - TI Software Folder
The answer theoretically to both questions is yes but
you clearly have no real idea what you are doing.
The Erse calculators are simplistic garbage and
the best way of using them is not using them.
rgds, sreten.
http://sites.google.com/site/undefinition/diy
(see if nothing else, the excellent FAQs)
The Speaker Building Bible
Zaph|Audio
Zaph|Audio - ZA5 Speaker Designs with ZA14W08 woofer and Vifa DQ25SC16-04 tweeter
http://audio.claub.net/Simple Loudspeaker Design ver2.pdf
FRD Consortium tools guide
Designing Crossovers with Software Only
RJB Audio Projects
Jay's DIY Loudspeaker Projects
Speaker Design Works
http://www.htguide.com/forum/showthread.php4?t=28655
A Speaker project
DIY Loudspeakers
Humble Homemade Hifi
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design
Great free SPICE Emulator : SPICE-Based Analog Simulation Program - TINA-TI - TI Software Folder
Last edited:
..."the math alone will take forever..."
This illustrates my point.
No matter how much you read, there's always something else.
I could go Moon Audio at over $1,000,000 per system and not be done.
Or I could use some help and build me something I like.
Something that sounds good to my ears, not a computer program.
Can I use a low pass/high pass filter combo from the tweeter output?
Can I use a low pass/high pass filter combo from the woofer output ?
Can I make my question any simpler?
This illustrates my point.
No matter how much you read, there's always something else.
I could go Moon Audio at over $1,000,000 per system and not be done.
Or I could use some help and build me something I like.
Something that sounds good to my ears, not a computer program.
Can I use a low pass/high pass filter combo from the tweeter output?
Can I use a low pass/high pass filter combo from the woofer output ?
Can I make my question any simpler?
Ready built crossovers and crossover calculators don't work.
Even if you had the frd files (frequency response data) and zma files (impedance data). You still need to simulate the baffle step bump and roll-off (is there software that does this?) and this depends on the baffle size and shape.
In my case, even with measuring gear (calibrated microphone) and posh software (LspCAD) I found it very difficult to get my simple 2-way speakers to sound good. My problems were getting good, clean measurements down to a low frequency, FULLY understanding how baffle step works combined with room gain at low frequencies and using the right crossover point and slopes.
The thing that's probably surprised me the most is how much 1dB difference in the frequency response can make.
There's a ton of very good information on the partsexpress forum - LINK
PS: thanks for the info. Overwhelming. I will download it and keep it in case I venture into a from scratch project, which may be soner than I hope.
Once again, thanks.
Once again, thanks.
To answer your original question
NO
There is too much interaction between drivers and XO components for it to be that simple
The ERSE calculator can get you in the ballpark IF you use the right impedance value, because an 8R driver is only 8R at certain frequencies, but if you have factory graphs and use the numbers at the point where you want to put the XO it will give you a very basic idea of the approximate values of components you MAY need
4-ways are really quite hard
NO
There is too much interaction between drivers and XO components for it to be that simple
The ERSE calculator can get you in the ballpark IF you use the right impedance value, because an 8R driver is only 8R at certain frequencies, but if you have factory graphs and use the numbers at the point where you want to put the XO it will give you a very basic idea of the approximate values of components you MAY need
4-ways are really quite hard
I started off with a simple 2 way, then added drivers in my system as grew it.
I successfully built a passive 5 way crossover for my 5 way active horn system using Erse and another online one that is very similar, a mixture of 15" drivers and compression drivers / ribbon tweeter.
It was widely auditioned and I even played it Scalford Hall Hifi show and was well received - so it can be done.
So, it can be done.
I used the 2 way calculator for the subs (tapped horns). I only used the bass driver part of the calc to provide a 90Hz X/O. 4th order.
I used the 3 way calculator for drivers that needed upper and lower slopes.
You pick the middle driver in the calc output for these...
In my system this was the mid bass crossing at 90Hz and 330Hz. I used 2nd order for both but 4th order might have been better on the 90Hz point.
The mid crossing at 330Hz and 1200 Hz. Both 2nd order.
Using the upper driver part of the calc tool the upper mid only needed a 1200Hz X/O (2nd order again) as the driver / horn trailed off perfectly about 12000Hz.
Again using the upper driver part of the calc tool the Raal ribbon tweeters were calc'd, crossed in with a simple 1st order at ~12000Hz / to taste.
I had learned by measurement that my bass drivers did not respond as expected. The frequency was not being pulled down properly. Rising impedance with Freq so a Zobel was used on both - again calculated first on the calc tools and applied.
I would say you need to acquire freq measurement capability. It's only a hundred $'s or £'s. Behringer mic and phantom pre + Holmimpulse free SW.
It really practically demonstrated by measurement what you are doing and is the best thing for applied learning.
I have subsequently gone Najda DSP - see Digital section😉. What a boon! Superb bit of kit and sound.
So flexible and you can try Butterworth, L-R, Bessel etc and slopes up to 8th order at the touch of a button and for up to 4 ways!
The cost of passive components at 90Hz (big inductors and caps needed) for 4 th order was more than Najda cost built - let alone what it would be for an 8th order test on the bass drivers.
I now have DSP on the lower 3 drivers, pass through set on Najda with a 1st order 3uF cap for the upper ones (Now Vitavox S2's). The tweeter is run off the same line feed to an amp but with ~1uF cap 1st order.
Unfortunately all the early pics of my use of crossover calcs and in room measurements are now lost on my 5 way DIY thread (can't edit them back in myself on this forum).
So, have a go - you might surprise your self. I did.
I successfully built a passive 5 way crossover for my 5 way active horn system using Erse and another online one that is very similar, a mixture of 15" drivers and compression drivers / ribbon tweeter.
It was widely auditioned and I even played it Scalford Hall Hifi show and was well received - so it can be done.
So, it can be done.
I used the 2 way calculator for the subs (tapped horns). I only used the bass driver part of the calc to provide a 90Hz X/O. 4th order.
I used the 3 way calculator for drivers that needed upper and lower slopes.
You pick the middle driver in the calc output for these...
In my system this was the mid bass crossing at 90Hz and 330Hz. I used 2nd order for both but 4th order might have been better on the 90Hz point.
The mid crossing at 330Hz and 1200 Hz. Both 2nd order.
Using the upper driver part of the calc tool the upper mid only needed a 1200Hz X/O (2nd order again) as the driver / horn trailed off perfectly about 12000Hz.
Again using the upper driver part of the calc tool the Raal ribbon tweeters were calc'd, crossed in with a simple 1st order at ~12000Hz / to taste.
I had learned by measurement that my bass drivers did not respond as expected. The frequency was not being pulled down properly. Rising impedance with Freq so a Zobel was used on both - again calculated first on the calc tools and applied.
I would say you need to acquire freq measurement capability. It's only a hundred $'s or £'s. Behringer mic and phantom pre + Holmimpulse free SW.
It really practically demonstrated by measurement what you are doing and is the best thing for applied learning.
I have subsequently gone Najda DSP - see Digital section😉. What a boon! Superb bit of kit and sound.
So flexible and you can try Butterworth, L-R, Bessel etc and slopes up to 8th order at the touch of a button and for up to 4 ways!
The cost of passive components at 90Hz (big inductors and caps needed) for 4 th order was more than Najda cost built - let alone what it would be for an 8th order test on the bass drivers.
I now have DSP on the lower 3 drivers, pass through set on Najda with a 1st order 3uF cap for the upper ones (Now Vitavox S2's). The tweeter is run off the same line feed to an amp but with ~1uF cap 1st order.
Unfortunately all the early pics of my use of crossover calcs and in room measurements are now lost on my 5 way DIY thread (can't edit them back in myself on this forum).
So, have a go - you might surprise your self. I did.
Last edited:
Thanks !
Once again, this is overwhelming. Your experience is clearly way over my head. And you have more toys than I do. I appreciate your encouragement.
I'm thinking of starting with basic internal electronics and fine tune with replaceable external components
Once again, this is overwhelming. Your experience is clearly way over my head. And you have more toys than I do. I appreciate your encouragement.
I'm thinking of starting with basic internal electronics and fine tune with replaceable external components
We all start somewhere😉
I hardly knew one end of an X/O when I started. I had some help from more knowledgeables and went from there.
I am not really a theorist. I like to do a little research / design then I'm a build measure, fix it, build measure fix it, kinda guy. You learn a lot that way you could never learn from books.
I was amazed that I could build a set of speaks from some birch ply and some drivers that would blow a $4000 pair of commercial speakers. From then on it was a slippery slope.
You could get a pro place to measure your drivers so you know more about them. Would not cost that much.
The other thing that no amount design perfection will give you in getting an X/O perfect is a flat in room response. There will be room modes, bass boom, suck out and perhaps too much brightness at some higher freqs.
That is where a DIY project really wins as folk swap whole boxes, talk of bright or dark cables 🙂, trying to get something that works rather than build something that works in the room.
Again DSP X/O's win as you can do room correction in a jiffy...
I hardly knew one end of an X/O when I started. I had some help from more knowledgeables and went from there.
I am not really a theorist. I like to do a little research / design then I'm a build measure, fix it, build measure fix it, kinda guy. You learn a lot that way you could never learn from books.
I was amazed that I could build a set of speaks from some birch ply and some drivers that would blow a $4000 pair of commercial speakers. From then on it was a slippery slope.
You could get a pro place to measure your drivers so you know more about them. Would not cost that much.
The other thing that no amount design perfection will give you in getting an X/O perfect is a flat in room response. There will be room modes, bass boom, suck out and perhaps too much brightness at some higher freqs.
That is where a DIY project really wins as folk swap whole boxes, talk of bright or dark cables 🙂, trying to get something that works rather than build something that works in the room.
Again DSP X/O's win as you can do room correction in a jiffy...
I think I have a simple answer for a starting point. I have nearly finished building my speaker cabinets and I have been investing in very precious woods. Scaling down is not an option. They are a 4 way because initially I was attempting to revive my Infinity Kappa 8. (20-80, 80-800, 800-5k, 5k-20k) then I found out the Kappa xovers are amp killers. And they are very hard ($$) to rebuild.
I did all the math for a 3 way 80/800 xover. The tweeter section ends up with 12.44uF and 3.18mH elements which I deduct are my high pass. Now to the high section...
I am using a 24 db/octave filter because I'm installing BG Neo 8. I've read it's recommended they be cut sharply because they have a big spike at 12k Hz.
If I do the math correctly, can I connect a 2 way 5000Hz 24 db/ octave xover to the tweeter section of my 3 way xover to end up with a 4 way?
THEN, I can start fine tuning with mics and sensors and all the toys.
Any constructive ideas will be appreciated.
I did all the math for a 3 way 80/800 xover. The tweeter section ends up with 12.44uF and 3.18mH elements which I deduct are my high pass. Now to the high section...
I am using a 24 db/octave filter because I'm installing BG Neo 8. I've read it's recommended they be cut sharply because they have a big spike at 12k Hz.
If I do the math correctly, can I connect a 2 way 5000Hz 24 db/ octave xover to the tweeter section of my 3 way xover to end up with a 4 way?
THEN, I can start fine tuning with mics and sensors and all the toys.
Any constructive ideas will be appreciated.
Just to make myself clear:
Woofer 20 - 80 Hz 12db/octave to
Mid bass 80 - 800 Hz (Infinity Polydome 5 inch dome midrange) 12 db/octave to
Mid range 800 - 5000(2) BG Neo 8 PDF 24 db/octave to
Tweeter 5000 - 20000 Dayton Audio PT2C 8
Sometimes I don't understand my own posts...
Any constructive comments, suggestions are welcome.
Woofer 20 - 80 Hz 12db/octave to
Mid bass 80 - 800 Hz (Infinity Polydome 5 inch dome midrange) 12 db/octave to
Mid range 800 - 5000(2) BG Neo 8 PDF 24 db/octave to
Tweeter 5000 - 20000 Dayton Audio PT2C 8
Sometimes I don't understand my own posts...
Any constructive comments, suggestions are welcome.
I will get laughed at by everyone here. So be it.
WOOFER: try both an 18mH and a 15mH LOW RDC Iron Cored Inductor. mh-audio.nl will help you find the capacitor value for the second order leg. You are obviously porting the woofer. Woofers are easy to figure out.
MIDBASS: A 5" midrange has very little output below 100Hz, so why worry about a high pass. Only worry about the x-over at 800Hz. This turns into a 2-way calculation at 800. That's it. Once the measurements start, if you have a bump in the 100Hz range, add a notch filter. This time you use a high Rdc el cheapo inductor in the notch. For the notch you want high resistance.
MIDRANGE: This is where all your problems will start. Your simulations ARE ONLY A STARTING POINT. Every crossover needs to be tweaked. It's only after you start listening that you see how everything interacts. Any sim will give you a starting point. For the capacitor, the higher the quality, the better it will sound. For testing, use cheap caps. The final inductor should be a high quality foil.
TWEETER: Too simple for words. If your crossing in the 5K region, try a 1uF cap and see what happens. Adjust to taste.
Do Not Use Any Multistrand wire. Use only solid core. The tweeter only needs 24guage. Do not use thicker for the tweeter. Smearing exists.
Use Lead Free Silver content solder once all your testing is over.
Buy 50 varying sized alligator clips. Trust me.
Sometimes Zobels are needed, sometimes they are not. Test both with and without.
Remember, it's the journey, not the destination. My speakers took almost 2 years to perfect. The final crossovers used very few of the sims suggested values. By the end of this, you'll understand.
WOOFER: try both an 18mH and a 15mH LOW RDC Iron Cored Inductor. mh-audio.nl will help you find the capacitor value for the second order leg. You are obviously porting the woofer. Woofers are easy to figure out.
MIDBASS: A 5" midrange has very little output below 100Hz, so why worry about a high pass. Only worry about the x-over at 800Hz. This turns into a 2-way calculation at 800. That's it. Once the measurements start, if you have a bump in the 100Hz range, add a notch filter. This time you use a high Rdc el cheapo inductor in the notch. For the notch you want high resistance.
MIDRANGE: This is where all your problems will start. Your simulations ARE ONLY A STARTING POINT. Every crossover needs to be tweaked. It's only after you start listening that you see how everything interacts. Any sim will give you a starting point. For the capacitor, the higher the quality, the better it will sound. For testing, use cheap caps. The final inductor should be a high quality foil.
TWEETER: Too simple for words. If your crossing in the 5K region, try a 1uF cap and see what happens. Adjust to taste.
Do Not Use Any Multistrand wire. Use only solid core. The tweeter only needs 24guage. Do not use thicker for the tweeter. Smearing exists.
Use Lead Free Silver content solder once all your testing is over.
Buy 50 varying sized alligator clips. Trust me.
Sometimes Zobels are needed, sometimes they are not. Test both with and without.
Remember, it's the journey, not the destination. My speakers took almost 2 years to perfect. The final crossovers used very few of the sims suggested values. By the end of this, you'll understand.
Hi,
Very true. You need to understand what you are doing.
You won't get very far with the Erse calculators.
rgds, sreten.
Hi,
Is there a difference between xover calculators?
I thought they were all a good starting point, I have it very clear no particular pre assembled untested xover will fit in "just right".
I will start with a basic 3 way external and tweek them as I go. At some time they will find their way in.
Is there a difference between xover calculators?
I thought they were all a good starting point, I have it very clear no particular pre assembled untested xover will fit in "just right".
I will start with a basic 3 way external and tweek them as I go. At some time they will find their way in.
- Status
- Not open for further replies.