Is it possible to actively crossover a loudspeaker that is originally a 2.5 way design?
My speakers have 4 woofers and one tweeter. The two top woofers run full range up to where it crosses to the tweeter and the bottom two woofers crossover somewhere around the lower midrange. All woofers share the same cabinet.
My guess (when I go active) would be to just eliminate the high pass part of the mid crossover and let the woofers run all the way down just like they do now passively.
Is there anything wrong with doing this? Will active crossovers allow you to omit the high pass part of the mid crossover?
Dave.
My speakers have 4 woofers and one tweeter. The two top woofers run full range up to where it crosses to the tweeter and the bottom two woofers crossover somewhere around the lower midrange. All woofers share the same cabinet.
My guess (when I go active) would be to just eliminate the high pass part of the mid crossover and let the woofers run all the way down just like they do now passively.
Is there anything wrong with doing this? Will active crossovers allow you to omit the high pass part of the mid crossover?
Dave.
Hi,
the simple answer is yes.
I suggest you investigate using three amplifiers.
treble, bass/mid, bass only.
The bass/mid to treble will probably need (benefit from) steep crossover slopes 2, 3 or 4pole in either Butterworth or Linkwitz-Reilly. You will almost certainly need to add some equalisation to either or both driver pass bands. If you're unlucky you may have to add multiple equalisation to both drivers.
The bass/mid should be a low pass only, particularly if it can survive decent low end output without distress.
The treble could be either high pass only or bandpass to reduce supersonic content.
The bass only is likely to sound best if you mimic the existing low pass crossover slope, probably a single pole. But you have the choice with active to experiment with slightly flatter or slightly steeper slopes till you find one that suits best.
I am not sure about your query regarding mid to bass crossover, but thinking how it would be done with two woofers and one treble. The upper woofer and treble operate on low pass and high pass crossover. The other woofer has a low pass single pole crossover (an inductor) set an octave or two below the main X. it means that you have an 8ohm speaker in the treble region, an 8ohm speaker in the mid region, but a 4ohm speaker in the bass region due to both woofers running in parallel.
With your 4 woofer set-up this could equate to a 2ohm speaker in the bass region if all 4 drivers are 8ohm and in parallel. This is unlikely. I need more info.
the simple answer is yes.
I suggest you investigate using three amplifiers.
treble, bass/mid, bass only.
The bass/mid to treble will probably need (benefit from) steep crossover slopes 2, 3 or 4pole in either Butterworth or Linkwitz-Reilly. You will almost certainly need to add some equalisation to either or both driver pass bands. If you're unlucky you may have to add multiple equalisation to both drivers.
The bass/mid should be a low pass only, particularly if it can survive decent low end output without distress.
The treble could be either high pass only or bandpass to reduce supersonic content.
The bass only is likely to sound best if you mimic the existing low pass crossover slope, probably a single pole. But you have the choice with active to experiment with slightly flatter or slightly steeper slopes till you find one that suits best.
I am not sure about your query regarding mid to bass crossover, but thinking how it would be done with two woofers and one treble. The upper woofer and treble operate on low pass and high pass crossover. The other woofer has a low pass single pole crossover (an inductor) set an octave or two below the main X. it means that you have an 8ohm speaker in the treble region, an 8ohm speaker in the mid region, but a 4ohm speaker in the bass region due to both woofers running in parallel.
With your 4 woofer set-up this could equate to a 2ohm speaker in the bass region if all 4 drivers are 8ohm and in parallel. This is unlikely. I need more info.
Is your speaker MTMMM or TMMMM (top to bottom)?
If the former you could fairly easily mimic the slopes as Andrew suggested. There might be room for improvement if you have (access to) measurement equipment. You could do the .5 crossover passively with an inductor and save an amp channel.
If the latter, I suggest making it TMWWW - the Ws rolled off as your .5's and the M run all the way down. I prefer steep slopes for the M-T, and running the tweeter as low as it can handle.
There's an active crossover calculator on my website. I don't have spare boards, though. Jens Rasmussen has the Eagle files on his site, if you are interested.
If the former you could fairly easily mimic the slopes as Andrew suggested. There might be room for improvement if you have (access to) measurement equipment. You could do the .5 crossover passively with an inductor and save an amp channel.
If the latter, I suggest making it TMWWW - the Ws rolled off as your .5's and the M run all the way down. I prefer steep slopes for the M-T, and running the tweeter as low as it can handle.
There's an active crossover calculator on my website. I don't have spare boards, though. Jens Rasmussen has the Eagle files on his site, if you are interested.
If you haven't done so, I would first measure the actual electrical transfer functions for the passive crossover (they will NOT be what you think!) and use that as a starting point. Ideally, you'd want to understand exactly why they're engineered the way they are so that you can change what you want to with your eyes open.
Thanks guys,
To clarify things a bit, the arrangement is a TMMMM top to bottom.
The tweeter is the Vifa XT and the woofers are custiom aluminum cones with nominal impedances of I guess 6 ohms each.
As of now the passive X-over is a progressive array designed by Dick Pierce. Each woofer hands off to the other as in multiple .5 arrays. If you know the speaker I'm talking about it is an Onix product (Rocket RS 750.)
I was thinking if I went active things would sound much better if it were simplified... using the top two woofers as the full range up to the tweeter and the bottom two woofers actively crossed at say 400 or so hertz.
Bob your idea of going TMWWW is interesting. I never thought of that. If each woofer was about 6 ohm impedance how would one hook them up if they are in 3's?
AndrewT You really think I need to get EQ's in the chain as well.
I thought that if you had an active crossover with steep enough slopes that you wouldn't need to no? This puts a different spin on things for me if I have to add equalizers to the equation as well.
Amps and active crossovers will be costly enough let alone well designed EQ's.
Dave.
To clarify things a bit, the arrangement is a TMMMM top to bottom.
The tweeter is the Vifa XT and the woofers are custiom aluminum cones with nominal impedances of I guess 6 ohms each.
As of now the passive X-over is a progressive array designed by Dick Pierce. Each woofer hands off to the other as in multiple .5 arrays. If you know the speaker I'm talking about it is an Onix product (Rocket RS 750.)
I was thinking if I went active things would sound much better if it were simplified... using the top two woofers as the full range up to the tweeter and the bottom two woofers actively crossed at say 400 or so hertz.
Bob your idea of going TMWWW is interesting. I never thought of that. If each woofer was about 6 ohm impedance how would one hook them up if they are in 3's?
AndrewT You really think I need to get EQ's in the chain as well.
I thought that if you had an active crossover with steep enough slopes that you wouldn't need to no? This puts a different spin on things for me if I have to add equalizers to the equation as well.
Amps and active crossovers will be costly enough let alone well designed EQ's.
Dave.
EQ isn't that hard if you are making fixed equalizers. You can make shelving or notch/peak filters with just a couple of opamps.
Without the ability to measure your drivers' responses you're shooting in the dark. You may need EQ, you may not, but you won't know if you can't measure.
If you use the top two covering up to the tweeter XO you are likely to get comb filtering. You might consider a chip amp channel per woofer - a pair of LM4780s would cover it. Otherwise, your're stuck putting three in parallel or series.
Without the ability to measure your drivers' responses you're shooting in the dark. You may need EQ, you may not, but you won't know if you can't measure.
If you use the top two covering up to the tweeter XO you are likely to get comb filtering. You might consider a chip amp channel per woofer - a pair of LM4780s would cover it. Otherwise, your're stuck putting three in parallel or series.
A good advice is to read the “Unofficial Speaker Work Shop” page 203 regarding active x-sover, this is a remarkable piece of work done by Jay Buttermann and Sergiusz Urbaniak.
http://www.audiodiycentral.com/ntutorials.shtml
I am currently using the method to design the active x-over (the recent diyaudio active x-over group buy– thanks to Bob and Jens) for my new dipole speakers, it seems to work fine (I have not yet build it), it's bit tricky – but it can do whatever you want to do regarding designing active x-over.
IMO measuring SPL/Phase is the key to success. Bobs idea - TMWW – is good with some kind of caustic isolation of the “M”. But……… active x-over is not a miracle cure.
Three in series might work just fine.
http://www.audiodiycentral.com/ntutorials.shtml
I am currently using the method to design the active x-over (the recent diyaudio active x-over group buy– thanks to Bob and Jens) for my new dipole speakers, it seems to work fine (I have not yet build it), it's bit tricky – but it can do whatever you want to do regarding designing active x-over.
IMO measuring SPL/Phase is the key to success. Bobs idea - TMWW – is good with some kind of caustic isolation of the “M”. But……… active x-over is not a miracle cure.
Three in series might work just fine.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.