Hi chaps,
Here is my problem:
I want to put together a 2 way in a set of boxes I already have using new drivers and crossover.
The unusual thing is I want to use a Scanspeak 10F/8424G00 like a fullrange in the old tweeter slot and a Seas L22RNX-P 8" aluminium bass driver for only the bottom up to 3-400Hz. So the Scanspeak for all the rest over that. So kind of like a FAST but sealed and using a mid for full.
The box is from Studiospares SN-10's and measures 42cm high x 25cm wide x 20cm deep and is made from 18mm MDF, hence has about 17litres internal volume.
'Problems' I am aware of and prepared to accept:
- Directional sound from the Scanspeaks.
- Not a lot of deep bass going to happen in this size closed box
- Not as much high topend without tweeter
Questions I have:
- With that bass driver in that size closed box, what kind of F3 can I expect?
- Will it be possible to get the response pretty straight for the functional range of the components?
The ideal outcome is a studio monitoring tool with fast transients and great detail.
So......I am totally green so designing the crossover myself isn't an option. I'm quite happy to source the parts and put it together, but not design it.
Who here would either be able to pull this off in style or knows someone they'd recommend that I contact about it. I realise Madisound do it, and here in the UK Wilmslow Audio do it, but I wouldn't mind options.
Oh, and I don't expect anyone to do this for free. Hopefully not for buckets of cash either, but not for free.....
Any thoughts appreciated......
Here is my problem:
I want to put together a 2 way in a set of boxes I already have using new drivers and crossover.
The unusual thing is I want to use a Scanspeak 10F/8424G00 like a fullrange in the old tweeter slot and a Seas L22RNX-P 8" aluminium bass driver for only the bottom up to 3-400Hz. So the Scanspeak for all the rest over that. So kind of like a FAST but sealed and using a mid for full.
The box is from Studiospares SN-10's and measures 42cm high x 25cm wide x 20cm deep and is made from 18mm MDF, hence has about 17litres internal volume.
'Problems' I am aware of and prepared to accept:
- Directional sound from the Scanspeaks.
- Not a lot of deep bass going to happen in this size closed box
- Not as much high topend without tweeter
Questions I have:
- With that bass driver in that size closed box, what kind of F3 can I expect?
- Will it be possible to get the response pretty straight for the functional range of the components?
The ideal outcome is a studio monitoring tool with fast transients and great detail.
So......I am totally green so designing the crossover myself isn't an option. I'm quite happy to source the parts and put it together, but not design it.
Who here would either be able to pull this off in style or knows someone they'd recommend that I contact about it. I realise Madisound do it, and here in the UK Wilmslow Audio do it, but I wouldn't mind options.
Oh, and I don't expect anyone to do this for free. Hopefully not for buckets of cash either, but not for free.....
Any thoughts appreciated......
Please do not be taken aback by the following. I only want to show you what you are actually asking.
It depends, do you want the final outcome to be satisfactory? Then the box matters. The crossover can't be done well without looking at the system as a whole. Diffraction, baffle step, internal resonances, placement, your personal tastes. How are you deciding the boxes you have are suitable? The placement on the baffle, how sharp the corners are, wow the box is braced, damping materials, and probably the phase of the moon all change the crossover requirements.
It just is not that simple. It takes me three or four passes for final voicing after I am happy with the active simulation using my DCX. To build a pair of speakers now, after doing it off an on for 35 years. still takes 4 or 5 months. "getting it straight" is an art.
You can look up the TS parameters on the Seas driver and plug them into WinISL (free) and see what the various responses will be. This is not difficult. Please put in a little effort first. Managing the metal cone breakup may take some work.
Now, you will probably get several responses from quick modeling tools. They may be pretty close, they may not be. You have sort of asked for the impossible.
Another approach is to go active and use a DCX or Mini DSP approach, then tune to your ears are happy.
It depends, do you want the final outcome to be satisfactory? Then the box matters. The crossover can't be done well without looking at the system as a whole. Diffraction, baffle step, internal resonances, placement, your personal tastes. How are you deciding the boxes you have are suitable? The placement on the baffle, how sharp the corners are, wow the box is braced, damping materials, and probably the phase of the moon all change the crossover requirements.
It just is not that simple. It takes me three or four passes for final voicing after I am happy with the active simulation using my DCX. To build a pair of speakers now, after doing it off an on for 35 years. still takes 4 or 5 months. "getting it straight" is an art.
You can look up the TS parameters on the Seas driver and plug them into WinISL (free) and see what the various responses will be. This is not difficult. Please put in a little effort first. Managing the metal cone breakup may take some work.
Now, you will probably get several responses from quick modeling tools. They may be pretty close, they may not be. You have sort of asked for the impossible.
Another approach is to go active and use a DCX or Mini DSP approach, then tune to your ears are happy.
the two drivers can NOT share the same internal air/volume/space. They must be acoustically separate and isolated.
The scanspeak, based upon your description, will need a highpass xover filter somewhere around the 3-400 Hz that you want to roll off the other driver.
THEN you need to actually measure them and see what they do in your box, and then base the actual xover elements and slopes and levels to match the real world situation... that's enough to get you close.
After that there may or may not be some more adjustments...
a good idea is to run the speaker terminals to the rear of the box and put the xover externally. then it is easy to get at and tweak as needed.
definitely follow the advice to plug the driver parameters into a simulator and see how it comes out in theory... good place to start.
_-_-bear
The scanspeak, based upon your description, will need a highpass xover filter somewhere around the 3-400 Hz that you want to roll off the other driver.
THEN you need to actually measure them and see what they do in your box, and then base the actual xover elements and slopes and levels to match the real world situation... that's enough to get you close.
After that there may or may not be some more adjustments...
a good idea is to run the speaker terminals to the rear of the box and put the xover externally. then it is easy to get at and tweak as needed.
definitely follow the advice to plug the driver parameters into a simulator and see how it comes out in theory... good place to start.
_-_-bear
Thanks for your input guys!!
Wow, so some issues ahead. And more questions...like:
Why do the drivers need to be acoustically separate ?
I would have run some simulator software already, but am on a Mac and didn't find any. In any case, maybe I will have to just build new boxes if that's what it takes.
I had initially had in mind to do this with two amps and a minidsp, but abandoned the project entirely since. Then the other day before putting the now spare Scanspeaks up for sale I figured I wanted to hear if their topend would have stretched and what it sounds like at least, so I just put them into the old Studiospares speakers instead of the tweeters. The crossover in those is at 3.2khz, so I haven't really experienced the mid quality of it, but what it sounded like over 3.2 made me consider the project again, thinking to try and build it as a passive speaker. Those Scanspeaks really seem sweet!!
I guess I could just pay Wilmslow Audio £12.77 for a box calculation and £33.33 for a crossover schematic, build that volume box and that crossover and hope for the best?
Wow, so some issues ahead. And more questions...like:
Why do the drivers need to be acoustically separate ?
I would have run some simulator software already, but am on a Mac and didn't find any. In any case, maybe I will have to just build new boxes if that's what it takes.
I had initially had in mind to do this with two amps and a minidsp, but abandoned the project entirely since. Then the other day before putting the now spare Scanspeaks up for sale I figured I wanted to hear if their topend would have stretched and what it sounds like at least, so I just put them into the old Studiospares speakers instead of the tweeters. The crossover in those is at 3.2khz, so I haven't really experienced the mid quality of it, but what it sounded like over 3.2 made me consider the project again, thinking to try and build it as a passive speaker. Those Scanspeaks really seem sweet!!
I guess I could just pay Wilmslow Audio £12.77 for a box calculation and £33.33 for a crossover schematic, build that volume box and that crossover and hope for the best?
Last edited:
Seems all that stuff is PC only though......thanks anyway. have found some good info on that one including golden ratio stuff etc on here now.
Interested in what people think about going with say Wilmslow's box and crossover design services as mentioned and about the likelyhood of it creating a good result. Anyone tried it?
Interested in what people think about going with say Wilmslow's box and crossover design services as mentioned and about the likelyhood of it creating a good result. Anyone tried it?
Here's a way to go.
There are a whole lot of similar posts with answers already here. It would be a good idea to search and read, get some ideas and insights first.
When you ask why the drivers need to be acoustically isolated, the need for you to get some additional learning becomes obvious. Nobody here has all the answers or information, but there are certain basic ideas and concepts that are more or less essential to have.
The answer to that question is: the drivers if placed in the same acoustic space in a cabinet will severely interact with each other, and cause bad effects upon each other.
You can also get a program that runs windows programs under Mac OS, maybe a better idea?
But really, look at some other posts for multi-way, simple speaker projects and see what others have done, and how they did it... if you read through I am sure you will find some threads with good information that speaks to some of your design goals... not to mention you will see how others have solved the same sorts of problems by various means...
There are a whole lot of similar posts with answers already here. It would be a good idea to search and read, get some ideas and insights first.
When you ask why the drivers need to be acoustically isolated, the need for you to get some additional learning becomes obvious. Nobody here has all the answers or information, but there are certain basic ideas and concepts that are more or less essential to have.
The answer to that question is: the drivers if placed in the same acoustic space in a cabinet will severely interact with each other, and cause bad effects upon each other.
You can also get a program that runs windows programs under Mac OS, maybe a better idea?
But really, look at some other posts for multi-way, simple speaker projects and see what others have done, and how they did it... if you read through I am sure you will find some threads with good information that speaks to some of your design goals... not to mention you will see how others have solved the same sorts of problems by various means...
I'm sure a lot of people will disagree with you Boscoe any crossover active or passive can be terrible, it depends on the implementation.
Along the lines of bears suggestion, if you have a few spare hours you could read about my journey here http://www.diyaudio.com/forums/multi-way/68301-my-morel-mtm-project.html it is very verbose so you may give up quiclky. post 108 was about where I really made some progress though with fixing what I felt were inadequacies in my original crossover.
There were a number of problems I faced that I didn't anticipate but I overcame them in the end and I am very happy with the results. It was certainly a much harder project than would usually be recommended for ones first foray into speaker design!!
The main problem being the mids were no where near as tame as what I was expecting, and I spent a huge amount of time trying to work out what was wrong, rather than accepting it as being the way it was and designing the crossover acordingly. It's been nearly a year since I finished it, and I haven't felt any need to change anything, so I think that it is right
Tony.
any crossover active or passive can be terrible, it depends on the implementation. Along the lines of bears suggestion, if you have a few spare hours you could read about my journey here http://www.diyaudio.com/forums/multi-way/68301-my-morel-mtm-project.html it is very verbose so you may give up quiclky. post 108 was about where I really made some progress though with fixing what I felt were inadequacies in my original crossover.
There were a number of problems I faced that I didn't anticipate but I overcame them in the end and I am very happy with the results. It was certainly a much harder project than would usually be recommended for ones first foray into speaker design!!
The main problem being the mids were no where near as tame as what I was expecting, and I spent a huge amount of time trying to work out what was wrong, rather than accepting it as being the way it was and designing the crossover acordingly. It's been nearly a year since I finished it, and I haven't felt any need to change anything, so I think that it is right
Tony.
It is probably worth the money if you require it. Many here probably feel it takes longer to design a crossover.
Thanks for your participation chaps!
So the driver interaction is definitely so severe it becomes a problem? I find that interesting, as it seems to me that must a relative statement at best.
As is definitely 'passive crossover are terrible'.
Quite a few speakers would be rendered terrible and useless by these two statements which in reality are good.
I have already discarded the minidsp (bought it, sold it again), and actually I am more concerned about my audio going through an extra set of cheap converters than a well designed passive crossover.
As I am looking to cross at 3-400 and have the very flat indeed Scanspeak do all the mid/top shouldn't a crossover design be relatively simple, given a bass driver is picked which has a good curve from 3-400 down? And given the right volume for the bass driver to naturally reach an F3 of somewhere in the 40's.......having said that, now if I make new boxes that can be bigger, maybe using a 10" instead and aiming for bass to about 40 or high 30's would be an idea.....Found a cabinet maker earlier and waiting for a quote back, so then having new boxes made to achieve the design goal could be on the cards.
I am indeed reading at the same time as this thread is going, so please don't get the idea I am too lazy to find stuff out off my own back. There is only so much time I have to read in a day though, so asking things in parallel seemed appropriate.
Now to investigate the severity of the driver separation thing....is it really that important? Off to read what I can find on the subject.....
So the driver interaction is definitely so severe it becomes a problem? I find that interesting, as it seems to me that must a relative statement at best.
As is definitely 'passive crossover are terrible'.
Quite a few speakers would be rendered terrible and useless by these two statements which in reality are good.
I have already discarded the minidsp (bought it, sold it again), and actually I am more concerned about my audio going through an extra set of cheap converters than a well designed passive crossover.
As I am looking to cross at 3-400 and have the very flat indeed Scanspeak do all the mid/top shouldn't a crossover design be relatively simple, given a bass driver is picked which has a good curve from 3-400 down? And given the right volume for the bass driver to naturally reach an F3 of somewhere in the 40's.......having said that, now if I make new boxes that can be bigger, maybe using a 10" instead and aiming for bass to about 40 or high 30's would be an idea.....Found a cabinet maker earlier and waiting for a quote back, so then having new boxes made to achieve the design goal could be on the cards.
I am indeed reading at the same time as this thread is going, so please don't get the idea I am too lazy to find stuff out off my own back. There is only so much time I have to read in a day though, so asking things in parallel seemed appropriate.
Now to investigate the severity of the driver separation thing....is it really that important? Off to read what I can find on the subject.....
Not sure I understand you, Allen. Are you saying it isn't worth making separate enclosures in the cabinet, or it is?
Just found this as well, which seems interesting:
http://www.diyaudio.com/forums/full-range/167172-4-jordan-plus-woofer-design.html
It seems the guy made the woofer roll off using merely one inductor and left the fullrange running full spectrum. I wouldn't like to try that with the Scanspeak as I fear I'd blow them, but would there be a way to do that the other way round too, as in roll the bottom off the high driver? Could it then be a s simple as getting a bass driver with the same sensitivity as my Scanspeaks and do the relevant inductor rolloffs to all involved and get something useable? Given a good size box for the bass driver to be happy...?
Just found this as well, which seems interesting:
http://www.diyaudio.com/forums/full-range/167172-4-jordan-plus-woofer-design.html
It seems the guy made the woofer roll off using merely one inductor and left the fullrange running full spectrum. I wouldn't like to try that with the Scanspeak as I fear I'd blow them, but would there be a way to do that the other way round too, as in roll the bottom off the high driver? Could it then be a s simple as getting a bass driver with the same sensitivity as my Scanspeaks and do the relevant inductor rolloffs to all involved and get something useable? Given a good size box for the bass driver to be happy...?
More stupid questions... :
Why couldn't using one coil for the bass and one capacitor for the top to create a super basic crossover like this work:
Passive Crossovers
I imagine one reason could be the shallow slope of 6db could make problems with doubling up and phasing? Provided the drivers' sensitivity was matched would this be worth trying?
Why couldn't using one coil for the bass and one capacitor for the top to create a super basic crossover like this work:
Passive Crossovers
I imagine one reason could be the shallow slope of 6db could make problems with doubling up and phasing? Provided the drivers' sensitivity was matched would this be worth trying?
I'm sure a lot of people will disagree with you Boscoe
Along the lines of bears suggestion, if you have a few spare hours you could read about my journey here http://www.diyaudio.com/forums/multi-way/68301-my-morel-mtm-project.html it is very verbose so you may give up quiclky. post 108 was about where I really made some progress though with fixing what I felt were inadequacies in my original crossover.
There were a number of problems I faced that I didn't anticipate but I overcame them in the end and I am very happy with the results. It was certainly a much harder project than would usually be recommended for ones first foray into speaker design!!
The main problem being the mids were no where near as tame as what I was expecting, and I spent a huge amount of time trying to work out what was wrong, rather than accepting it as being the way it was and designing the crossover acordingly. It's been nearly a year since I finished it, and I haven't felt any need to change anything, so I think that it is right
Tony.
I'm sorry but all passive crossovers are terrible. Low orders, large inaccuracies due to component tolerances and it destroys any damping factor an amplifier has over bass driver resulting in terrible bass. Distortion is also extremely bad especially at high power levels as well as low SPL.
Karloff, best to keep them separate.
It will work and it's a good starting point but is unlikely to impress.
I take it you mean keep the drivers in separate enclosures?
If it is unlikely to impress, what would be likely things to do to improve on it from there?
Also, am I understanding it correctly in that the coil will introduce delay? And hence a phase mismatch between the drivers? Could I counteract that by setting the bass driver 'out' a bit from the baffle? How would I calculate the delay/distance to make it up?
You can set the woofer out, but you shouldn't need to. There is a phase difference between the drivers to begin with, and a delay because the voice coils aren't lined up. At higher frequencies the delay turns into more of a phase difference.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.