Anybody seen this new Crossover from Marchand yet. Check out the options line below, it's about time someone offered this in a Home Hi-Fi kit. $399 and $499 for the 2-way and 3-way versions, not too bad.
Available 2-way, 3-way or 4-way.
Steep 24 dB/octave 4-th order slope (Linkwitz-Riley) standard. Up to 48 dB/octave optional.
Asymmectric or symmetric slopes.
Crossover frequency 20Hz - 20 KHz.
Subwoofer summing option.
Heavy duty gold plated RCA connectors for input and output.
No transients during turn-on or turn-off.
Options: Baffle step compensation, notch filters, delay sections
http://www.marchandelec.com/xm44.htm:)
Available 2-way, 3-way or 4-way.
Steep 24 dB/octave 4-th order slope (Linkwitz-Riley) standard. Up to 48 dB/octave optional.
Asymmectric or symmetric slopes.
Crossover frequency 20Hz - 20 KHz.
Subwoofer summing option.
Heavy duty gold plated RCA connectors for input and output.
No transients during turn-on or turn-off.
Options: Baffle step compensation, notch filters, delay sections
http://www.marchandelec.com/xm44.htm:)
Rob Elliott's Project 9
Cant find anything in the archives about "Project 9 of Rod Elliott's". Sounds interesting, whats it all about?
Cant find anything in the archives about "Project 9 of Rod Elliott's". Sounds interesting, whats it all about?
I guess I'll be the curmudgeon here.
First, these are not phase coherent. If we put a square wave
in and sum the outputs, we will not get a square wave.
Second, I dislike 24 dB/octave slopes. I've been fooling
around with these things for about 30 years, and I count
on one hand the number of times 4 pole slopes have proven
to be the best solution. I don't know why this is, but I'm
not the only one who has experienced this.
First, these are not phase coherent. If we put a square wave
in and sum the outputs, we will not get a square wave.
Second, I dislike 24 dB/octave slopes. I've been fooling
around with these things for about 30 years, and I count
on one hand the number of times 4 pole slopes have proven
to be the best solution. I don't know why this is, but I'm
not the only one who has experienced this.
Thank you, Mr. Pass, for your comment.
Can you elaborate on what aspect you didn't like about LR4? Too harsh? phasey? lobind issues, etc...Does this apply to both passive and active? What kind of topology do you recommend?
I once compared a passive LR2 /w LR4 with some friends. We preferred the LR2 because the LR4 sounded harsh, but I didn't know whether it was the slope or some other issues.
Thanks again,
Duc
Can you elaborate on what aspect you didn't like about LR4? Too harsh? phasey? lobind issues, etc...Does this apply to both passive and active? What kind of topology do you recommend?
I once compared a passive LR2 /w LR4 with some friends. We preferred the LR2 because the LR4 sounded harsh, but I didn't know whether it was the slope or some other issues.
Thanks again,
Duc
Don't mistake me as commenting on any particular product,
as my remarks were intended generally.
There have always been a lot of loose references to "phase
coherent" as with any other commercially valuable phrase.
I subscribe to the phrase "phase coherent" as meaning the
ability to reassemble a square wave at the output. I
refer you to the article I wrote on phase coherent crossovers
which was in TAA, and can be seen at www.passlabs.com
Other such phrases?
Anything to do with Class A
Anything to do with feedback
patented
balanced
The list goes on and on.
As to 4 pole slopes, it's not like they can never sound good, but
in my opinion that occurence is rare, and you are much better
off exploring lower pole responses first.
as my remarks were intended generally.
There have always been a lot of loose references to "phase
coherent" as with any other commercially valuable phrase.
I subscribe to the phrase "phase coherent" as meaning the
ability to reassemble a square wave at the output. I
refer you to the article I wrote on phase coherent crossovers
which was in TAA, and can be seen at www.passlabs.com
Other such phrases?
Anything to do with Class A
Anything to do with feedback
patented
balanced
The list goes on and on.
As to 4 pole slopes, it's not like they can never sound good, but
in my opinion that occurence is rare, and you are much better
off exploring lower pole responses first.
My Marchand adds a veil to the system that i find i can't live with anymore. It is going up on the block and i'm going to build something simplier.
1st trials will be a 2nd order subtractive.
dave
1st trials will be a 2nd order subtractive.
dave
Nelsons comments from the subjective view point are interesting.
My experience is similar but I have found that lower order, particularly AllPass (nothing to do with Passlabs) yield great results for lower frequency transitions say below 300 hz and in particular 150 and below for subs. They tend to allow a more gradual transisition which I think sounds more coherent and have less group delay and other strange phase characteristics.
The higher order slopes tend to come into favour above 1 khz, but it depends on the total driver integration and acoustic interaction. This is because soft roll slope create broader polar response irregularities due to greater overlap which effect the amplitude response at differing listening positions, particularly on the vertical axis.
The chip buffers are an issue for sound quality and it is better to have discrete buffers and the least number of stages if possible
See Vance Dickason Cook Book for the teck talk.
There is no rule book for any of this however and to own the new PassLabs active crossover which has almost infinite adjustment would allow a great deal of practical experimentation.
PS How is this diy version progressing Mr Pass.
Cheers
macka
My experience is similar but I have found that lower order, particularly AllPass (nothing to do with Passlabs) yield great results for lower frequency transitions say below 300 hz and in particular 150 and below for subs. They tend to allow a more gradual transisition which I think sounds more coherent and have less group delay and other strange phase characteristics.
The higher order slopes tend to come into favour above 1 khz, but it depends on the total driver integration and acoustic interaction. This is because soft roll slope create broader polar response irregularities due to greater overlap which effect the amplitude response at differing listening positions, particularly on the vertical axis.
The chip buffers are an issue for sound quality and it is better to have discrete buffers and the least number of stages if possible
See Vance Dickason Cook Book for the teck talk.
There is no rule book for any of this however and to own the new PassLabs active crossover which has almost infinite adjustment would allow a great deal of practical experimentation.
PS How is this diy version progressing Mr Pass.
Cheers
macka
As to 4 pole slopes
The fact that the driver resonant frequency(on the low end) or high frequency rolloff (and directionality changes) are often less than a octave from the crossover frequency greatly contributes to the actual slope. The driver response is very much a part of the total acoustic response of the filter. To sum it up, you usally wind up with total response slope greater than that of the active filter contribution.
This get's even more convoluted since a 4th order slope is chosen to get even closer to the point where the drivers out of band response starts to fall off. You can wind up with something closer to a sixth order response. Mr. Dickason's loudspeaker books outline this fact very well.
Mr. Pass has given the world the first product to allow one to really exploit this design aproach and the manual for this product is required reading for DIY active crossovers.
XVR1 Owner's Manual
http://www.passlabs.com/pdf/XVR1MAN1.PDF
Loudspeaker Design Cookbook
http://www.mfr-eng.com/ldc.htm
Thanks again to Nelson for shining a bright light in another dark corner.
Fred
P.S. further reading:
http://www.rane.com/pdf/note147.pdf
http://www.rane.com/pdf/linriley.pdf
http://www.rane.com/note107.html
http://www.rane.com/note119.html
http://www.pvconsultants.com/audio/frdgroup.htm
The fact that the driver resonant frequency(on the low end) or high frequency rolloff (and directionality changes) are often less than a octave from the crossover frequency greatly contributes to the actual slope. The driver response is very much a part of the total acoustic response of the filter. To sum it up, you usally wind up with total response slope greater than that of the active filter contribution.
This get's even more convoluted since a 4th order slope is chosen to get even closer to the point where the drivers out of band response starts to fall off. You can wind up with something closer to a sixth order response. Mr. Dickason's loudspeaker books outline this fact very well.
Mr. Pass has given the world the first product to allow one to really exploit this design aproach and the manual for this product is required reading for DIY active crossovers.
XVR1 Owner's Manual
http://www.passlabs.com/pdf/XVR1MAN1.PDF
Loudspeaker Design Cookbook
http://www.mfr-eng.com/ldc.htm
Thanks again to Nelson for shining a bright light in another dark corner.
Fred
P.S. further reading:
http://www.rane.com/pdf/note147.pdf
http://www.rane.com/pdf/linriley.pdf
http://www.rane.com/note107.html
http://www.rane.com/note119.html
http://www.pvconsultants.com/audio/frdgroup.htm
Exactly, Fred.
And don't forget:
http://www.kbacoustics.com/ygbsm/2waycrossovers/learnhowsealed/index.html
provides an excellent solution.
And don't forget:
http://www.kbacoustics.com/ygbsm/2waycrossovers/learnhowsealed/index.html
provides an excellent solution.
- Status
- Not open for further replies.