Hi All,
I'm sure all this has been covered in previous threads (feel free to provide links if you know them), but I hate searching threads as they take forever to find and weed out...
Anyway, I need to know where to begin learning about designing and building crossover networks, as I know next to nothing about them. I don't know what the components are, what they do, the terminology, etc., so everything I read about them on these pages just goes over my head (among other things, but we won't go there).
Are there books, websites, etc? Also, am I correct in assuming that I will want software for testing/listening, etc? Any suggestions on that too? And where can I buy top end crossover components?
Thanks for your help!
I'm sure all this has been covered in previous threads (feel free to provide links if you know them), but I hate searching threads as they take forever to find and weed out...
Anyway, I need to know where to begin learning about designing and building crossover networks, as I know next to nothing about them. I don't know what the components are, what they do, the terminology, etc., so everything I read about them on these pages just goes over my head (among other things, but we won't go there).
Are there books, websites, etc? Also, am I correct in assuming that I will want software for testing/listening, etc? Any suggestions on that too? And where can I buy top end crossover components?
Thanks for your help!
Exactly!
There are a number of issues that crop up with playing around with crossovers. If your not able to do all the study and understand phase and radiation lobes,then you can stick with the simple stuff.
Simplest
http://www.bcae1.com
Harder
http://www.sound.westhost.com/articles.htm
http://www.linkwitzlab.com/
http://www.woodartistry.com/linkwitzlab/sb80-3wy.zip
Even Harder
http://www.geocities.com/kreskovs/John1.html
http://www.geocities.com/kreskovs/CrossoverdocN.html
Depends on what top end means,if it means expensive components that you think sound better,im sure there are many sites around if you search for 'hifi audiophile capacitor' etc
Passive crossovers will require higher spec components than many cheap systems have.
Active crossovers consist of opamps and components that are less critical and often have good specs anyway.
Note that audiophiles will debate the ability of some of these components and their distortion,but Ive never seen a well carried out ABX test prove it.
How About Some More Details?
I guess I need to buy a book to get started, but let me ask this: What are the basic parts of a crossover (passive) and what do they do? How do you set crossover frequencies?
Here's an actual application I have: I have an older set of a/d/s 3-way x-overs - one has a .47 mh inductor (I figured that one out myself, but I still don't know what it does) broken off, so I need to replace it. Obviously I want to use the same .47 mh value (?), but what would happen if I changed it? Just curious. And, should I use wire (which it has) or foil? And should I use air core (which it has) or solid (hardwood)? How would this change the performance? (Yes if I make changes to one I'll do the same to both.)
I guess I need to buy a book to get started, but let me ask this: What are the basic parts of a crossover (passive) and what do they do? How do you set crossover frequencies?
Here's an actual application I have: I have an older set of a/d/s 3-way x-overs - one has a .47 mh inductor (I figured that one out myself, but I still don't know what it does) broken off, so I need to replace it. Obviously I want to use the same .47 mh value (?), but what would happen if I changed it? Just curious. And, should I use wire (which it has) or foil? And should I use air core (which it has) or solid (hardwood)? How would this change the performance? (Yes if I make changes to one I'll do the same to both.)
That's funny. 
Basic passive crossovers are made up of capacitors and inductors. Capacitors are high pass filters meaning they allow high frequencies to pass but not low ones. The actual frequencies will depend, mostly, on the value of the capacitor.
Inductors are low pass filters meaning they allow low frequencies to pass.
The idea is to use the correct value cap and inductor (using a two-way speaker as reference) to allow the highs to go to the tweeter and the lows to the woofer AND have a smooth transition at the crossover point (the point where both drivers are reproducing the same frequencies.
I second getting David Weems book to start out.
http://www.amazon.com/exec/obidos/A...7/sr=2-1/ref=pd_bbs_b_2_1/102-6790192-7685742
BTW, I think Bill was pointing out that's it may not be necessary to trash the inductor if all that's missing is a very short length of wire. Inductors don't really wear out with age like caps. You may be able to reattach it with no audible change in the speaker.
Basic passive crossovers are made up of capacitors and inductors. Capacitors are high pass filters meaning they allow high frequencies to pass but not low ones. The actual frequencies will depend, mostly, on the value of the capacitor.
Inductors are low pass filters meaning they allow low frequencies to pass.
The idea is to use the correct value cap and inductor (using a two-way speaker as reference) to allow the highs to go to the tweeter and the lows to the woofer AND have a smooth transition at the crossover point (the point where both drivers are reproducing the same frequencies.
I second getting David Weems book to start out.
http://www.amazon.com/exec/obidos/A...7/sr=2-1/ref=pd_bbs_b_2_1/102-6790192-7685742
BTW, I think Bill was pointing out that's it may not be necessary to trash the inductor if all that's missing is a very short length of wire. Inductors don't really wear out with age like caps. You may be able to reattach it with no audible change in the speaker.
Yes thank you sir for some useable information, and I WILL get that book.
I don't mean to be rude to the other poster, but the answer was in the question: "broken." The inductor actually broke at the lead lines, so re-soldering it to the board would be problematic - there was nothing coming out of the inductor to work with.
Plus, he just didn't answer any of my real questions. Part of my post was about possibly replacing it with something "better" - i.e., a higher quality inductor like a foil, etc., if such a change IS actually an AUDIBLE improvement, which opens up a whole new can of worms.
I've read so much about the lengths people go to to build these things (and believe me I salute you guys), but I've really been wondering if all these incredible crossovers I've seen and read about actually make an AUDIBLE DIFFERENCE in the sound of a loudspeaker - not to a microphone or a graph on a computer monitor, but to THE HUMAN EAR.
And the only way to really test this is by doing it BLINDLY: one system, two identical speakers, then fiddling with crossovers and parts, and asking a BLINDED listener to see if he/she can hear any subjective difference when a change is made, but not even being told when or if or what kind of change has been made.
This is how pharmaceutical drugs are tested - it's called a double-blind study - and it should work for anything else. If you spend $100 to replace a part that cost $10, you're probably gonna hear a difference, because you WANT to - but that is subjective, NOT factual. The person who built the system and spent the money is probably not the best person to test for improved performance!
It goes back to the arguments I've seen about loudspeaker cables - jeez you can spend tens of thousands of dollars on these things, and when they were actually properly tested, people couldn't tell the difference from one to the other.
So I'm wondering if the same thing happens with crossovers - I'm sure there are differences, but there must be some point at which further changes (read: spending more money) ceases to make an audible difference. I guess that's one thing I'm trying to get a grasp of here.
Anyway, I dither. Perhaps now that I know a little bit more about a few of the components, you can go on to tell me how to pick various values to achieve desired crossover frequencies? And what's all this first, second, and fourth order stuff, and how do you do THAT?
I don't mean to be rude to the other poster, but the answer was in the question: "broken." The inductor actually broke at the lead lines, so re-soldering it to the board would be problematic - there was nothing coming out of the inductor to work with.
Plus, he just didn't answer any of my real questions. Part of my post was about possibly replacing it with something "better" - i.e., a higher quality inductor like a foil, etc., if such a change IS actually an AUDIBLE improvement, which opens up a whole new can of worms.
I've read so much about the lengths people go to to build these things (and believe me I salute you guys), but I've really been wondering if all these incredible crossovers I've seen and read about actually make an AUDIBLE DIFFERENCE in the sound of a loudspeaker - not to a microphone or a graph on a computer monitor, but to THE HUMAN EAR.
And the only way to really test this is by doing it BLINDLY: one system, two identical speakers, then fiddling with crossovers and parts, and asking a BLINDED listener to see if he/she can hear any subjective difference when a change is made, but not even being told when or if or what kind of change has been made.
This is how pharmaceutical drugs are tested - it's called a double-blind study - and it should work for anything else. If you spend $100 to replace a part that cost $10, you're probably gonna hear a difference, because you WANT to - but that is subjective, NOT factual. The person who built the system and spent the money is probably not the best person to test for improved performance!
It goes back to the arguments I've seen about loudspeaker cables - jeez you can spend tens of thousands of dollars on these things, and when they were actually properly tested, people couldn't tell the difference from one to the other.
So I'm wondering if the same thing happens with crossovers - I'm sure there are differences, but there must be some point at which further changes (read: spending more money) ceases to make an audible difference. I guess that's one thing I'm trying to get a grasp of here.
Anyway, I dither. Perhaps now that I know a little bit more about a few of the components, you can go on to tell me how to pick various values to achieve desired crossover frequencies? And what's all this first, second, and fourth order stuff, and how do you do THAT?
Yes, there are definitely different quality components the biggest difference of which is going to be materials. You won't find many people on this forum that will use eletrolytic caps. More often they will use film and foil and the sonic differences can be very apparent. Perhaps less noticeable is inductors and that can be caused by magnetic saturation. Most will try to avoid iron/steel core inductors except for large values. Air core inductors with low DC resistance are probably the most common used here.
As for the rest of your comments/questions: Beauty is in the ears of the beholder and on occasion approaches the level of religion. I'm not going to go there.
Orders of crossovers:
The idea is to cause the driver to be less responsive as it approaches a certain frequency. This happens gradually and when measured appears as a slope, not a brick wall. The rate of the slope is measured in decibels relative to octaves.
First order = 6db per octave.
2nd order = 12db
3rd order = 18db
4th order = 24db.
As to which is best, that depends on the drivers being used and the designers goals. There are advantages and disadvantages to each and as in all speaker design, the goal is to have the least number of compromises.
Happy reading!
As for the rest of your comments/questions: Beauty is in the ears of the beholder and on occasion approaches the level of religion. I'm not going to go there.
Orders of crossovers:
The idea is to cause the driver to be less responsive as it approaches a certain frequency. This happens gradually and when measured appears as a slope, not a brick wall. The rate of the slope is measured in decibels relative to octaves.
First order = 6db per octave.
2nd order = 12db
3rd order = 18db
4th order = 24db.
As to which is best, that depends on the drivers being used and the designers goals. There are advantages and disadvantages to each and as in all speaker design, the goal is to have the least number of compromises.
Happy reading!
Dude you rock! Thanks for taking the time to go into the detail that you have.
I guess the book is the best plan; most of the time when I ask questions of this nature, the answer I usually get it is "It depends on...." and a whole new batch of variables is introduced into an increasingly complex equation, and I end up more confused. I'm just convinced it should be simpler - the only difference that matters is one you can hear.
I think this is a difficult subject to address - one has to have a more tangible starting point, such as the drivers, speaker type, etc., from which the crossover will evolve. Plus I'm sure I need software to be able to test everything, which I don't have. Maybe a basic rule of thumb is to just start with good quality components and you'll be on the right track.
As for the orders, I have a basic grasp of that, but I have no idea how the various types sound in relation to the eachother. My thinking is that the bottom line should be to achieve the flattest frequency response possible - that seems to be a universal goal, with perhaps many many ways to get there.
BTW - is there any particular software (speaker design, etc.) you'd recommend? Thanks again!
I guess the book is the best plan; most of the time when I ask questions of this nature, the answer I usually get it is "It depends on...." and a whole new batch of variables is introduced into an increasingly complex equation, and I end up more confused. I'm just convinced it should be simpler - the only difference that matters is one you can hear.
I think this is a difficult subject to address - one has to have a more tangible starting point, such as the drivers, speaker type, etc., from which the crossover will evolve. Plus I'm sure I need software to be able to test everything, which I don't have. Maybe a basic rule of thumb is to just start with good quality components and you'll be on the right track.
As for the orders, I have a basic grasp of that, but I have no idea how the various types sound in relation to the eachother. My thinking is that the bottom line should be to achieve the flattest frequency response possible - that seems to be a universal goal, with perhaps many many ways to get there.
BTW - is there any particular software (speaker design, etc.) you'd recommend? Thanks again!
There are a bunch of variables. I just gave a very simple explanation of the basics.
Yes, measurement is very important. As for software, my experience is that you get what you pay for. There is freeware out there that works but isn't very intuitive.
http://www.speakerworkshop.com
You'll need a microphone and probably a pre-amp, a computer and a good soundcard.
Pjay wrote a good summary here.
http://members.aol.com/pjay99site/buildspeaker.htm
A very popular piece of software for crossover simulation is LspCAD. Usually, you would make your measurements, then import the data into LspCAD and use it to model your crossover.
http://www.ijdata.com/
Personally, I got tired of trying to make JustMLS work with a cheap Berhinger mic and preamp and bought Praxis. The combination of Praxis and LspCAD works well but takes a certain amount of committment (read money).
Yes, measurement is very important. As for software, my experience is that you get what you pay for. There is freeware out there that works but isn't very intuitive.
http://www.speakerworkshop.com
You'll need a microphone and probably a pre-amp, a computer and a good soundcard.
Pjay wrote a good summary here.
http://members.aol.com/pjay99site/buildspeaker.htm
A very popular piece of software for crossover simulation is LspCAD. Usually, you would make your measurements, then import the data into LspCAD and use it to model your crossover.
http://www.ijdata.com/
Personally, I got tired of trying to make JustMLS work with a cheap Berhinger mic and preamp and bought Praxis. The combination of Praxis and LspCAD works well but takes a certain amount of committment (read money).
As well as the advice from others, I would also recommend playing around with the spreadsheets and programs here: http://www.pvconsultants.com/audio/frdgroup.htm
And reading the articles here: http://www.pvconsultants.com/audio/frdarticles.htm
And reading the articles here: http://www.pvconsultants.com/audio/frdarticles.htm
sdclc126 said:
Yes it really is. It is apparent that you have a great interest in all of this, you just lack the knowledge at this point. Rather than beat your head against the wall trying to sort it out here, the book will give you lots of info so you can come back less frustrated.
Re: The coil with the broken lead wire:
Chokes are simply magnet wire (copper with a coating) wound around a core which is removed later to leave you an "air core". As Tim said, you can usually just free up a little wire off the coil and scrape off the coating before you resolder it back into the circuit. It won't change anything that much.
Perhaps Bill was unaware of your level of knowledge when he answered the question. He's been at this for a long time and sometimes forgets that not everyone sleeps with a cap and coil under their pillow.
He wasn't trying to be rude. Anyway, get the book, do the reading and come back with your questions. We'll be more useful to you when you have a little better handle on it.
Glad to see your interest in it as it is one heck of a way to spend your free time. Keep us in tune with your progress.
Cal
I know "Bill" wasn't trying to be rude - and I really wasn't trying to be rude back - it's just that none of the questions I was really asking were addressed - so I gave a very short and rather "curt" response - out of frustration that I wasn't getting the info I wanted. Sorry Bill!! Hope I'll still hear from you in the future?
The idea was that I had a broken part, and, as DIYers, we don't just like to fix things - we like to IMPROVE them - so I thought this was a perfect opportunity to ask what would be a better choke/inductor to replace the one that fell off.
I have a really nice set of crossovers I bought off eBay, out of an audiophile (a/d/s/) set of speakers (I don't have the speakers), so it seems I have a perfect opportunity to tinker with these things as a project and maybe resell them. I was just thinking of replacing all the components with components of equivalent specs but higher quality, just to get my feet wet. This seems the simplest way to do it.
Yeah, it won't make much sense to talk crossovers without having speakers to put 'em in, and without a measuring system set up, so I guess I'll be back when I've learned more and am working on specific projects.
But hey, while we're here, are there some basic mathematical formulas for determining the specs of crossover parts to set crossover frequencies at least? Or is this still putting the cart before the horsre?
The idea was that I had a broken part, and, as DIYers, we don't just like to fix things - we like to IMPROVE them - so I thought this was a perfect opportunity to ask what would be a better choke/inductor to replace the one that fell off.
I have a really nice set of crossovers I bought off eBay, out of an audiophile (a/d/s/) set of speakers (I don't have the speakers), so it seems I have a perfect opportunity to tinker with these things as a project and maybe resell them. I was just thinking of replacing all the components with components of equivalent specs but higher quality, just to get my feet wet. This seems the simplest way to do it.
Yeah, it won't make much sense to talk crossovers without having speakers to put 'em in, and without a measuring system set up, so I guess I'll be back when I've learned more and am working on specific projects.
But hey, while we're here, are there some basic mathematical formulas for determining the specs of crossover parts to set crossover frequencies at least? Or is this still putting the cart before the horsre?
Not sure why you want to replace it. Hot glue and a soldering iron work fine. Do you have a picture of the "broken" coil?
I wouldn't bother. You're not really accomplishing anything. It's like replacing the tires on wheels you can't use on your car, just so you know how to replace tires. The crossover isn't something that you buy really good parts for and just throw it at your speakers. It doesn't work that way. First you match 2, 3, or 4 drivers with one another and then you design the XO to match the drivers. The crossovers you have, you might never find a use for unless you find drivers to match them. But I don't expect you are going to try that. That's a dumb way of going about it.
There are many calculators out there. Here's one:
http://www.mhsoft.nl/spk_calc.asp
And lastly, go get the book before your horse trips over the cart. You're full of pith and vinegar, now make good use of it.
Cal
quote:
I have a really nice set of crossovers I bought off eBay
"Your first lesson should be to never do that again unless you're just after the individual components. Ready made crossovers are worse than useless."
I don't think you followed what I was saying, but I didn't give all the details - I ONLY buy a/d/s/ components off eBay (I didn't tell you that part) - I've owned a/d/s/ speakers for years and have used them in my home, car, etc. So when I buy a/d/s/ drivers I buy a/d/s/ crossovers. They are not "ready made" - they are engineered by a high end manufacturer for their particular drivers/speakers.
Since I don't know how to build crossovers this is the best I can do, and I've had great results. This particular set of crossovers was for a project I was doing for my sister, with a/d/s/ drivers. I know there are issues with the individual driver sizes & specs, etc., but again, if I want to just build a speaker, rather than engineering one, which I'm just not equipped to do, this is a reasonable compromise.
I have a really nice set of crossovers I bought off eBay
"Your first lesson should be to never do that again unless you're just after the individual components. Ready made crossovers are worse than useless."
I don't think you followed what I was saying, but I didn't give all the details - I ONLY buy a/d/s/ components off eBay (I didn't tell you that part) - I've owned a/d/s/ speakers for years and have used them in my home, car, etc. So when I buy a/d/s/ drivers I buy a/d/s/ crossovers. They are not "ready made" - they are engineered by a high end manufacturer for their particular drivers/speakers.
Since I don't know how to build crossovers this is the best I can do, and I've had great results. This particular set of crossovers was for a project I was doing for my sister, with a/d/s/ drivers. I know there are issues with the individual driver sizes & specs, etc., but again, if I want to just build a speaker, rather than engineering one, which I'm just not equipped to do, this is a reasonable compromise.
"Not sure why you want to replace it. Hot glue and a soldering iron work fine. Do you have a picture of the "broken" coil?"
Forgot this one other one - I don't have a photo of the broken coil. I suppose it would help clear the air about this whole business if I let everyone know that one of the more significant reasons I want to replace it is that I LOST THE ******* THING!
Forgot this one other one - I don't have a photo of the broken coil. I suppose it would help clear the air about this whole business if I let everyone know that one of the more significant reasons I want to replace it is that I LOST THE ******* THING!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.