a few posts back most people indicated that the highest freq. that can be managed using a array is determined by the center to center distance between 2 drivers. I figured it would be center to center distance between THE 2 FURTHEST drivers. for example for a 32 x 3" line source the center to center distnace would be in feet (something like 7.5 feet) pushing the highest freq down quite a bit - approx 150Hz. yet in one example this was quoted as being 4520Hz.
Since 7V has really worked on this i would like to know his opinion on this in real world. Sorry for putting you on the spot 7V.
Since 7V has really worked on this i would like to know his opinion on this in real world. Sorry for putting you on the spot 7V.
navin said:
Navin,
Did you not want to follow my previous post and read Dr. Griffin's
white paper? I think you'll find it quite informative.
Regards,
Dan
according to this paper the max freq for 2x3" w3871 or jx 92 would be about 4-5k. above which a tweeter would be needed to avoid comb effects. alternately one would have to roll off one driver and sacrifice senstivity and power handling.
the w3 871 cant handle much power to begin with my object of using multiples (4 or more) was to increase power handling, spl, and sensitivity.
7V, do you run only one driver full range in your speaker?
the w3 871 cant handle much power to begin with my object of using multiples (4 or more) was to increase power handling, spl, and sensitivity.
7V, do you run only one driver full range in your speaker?
Navin likes the line source ? 😎 😎 😎
expensive but........
**********************
Spend more, build once.
vs.
Spend less, build ten times.
**********************
http://www.selahaudio.com/excel.html
http://members.aol.com/bvirgil311/Excelarray.htm
** addition research **
Go here;
http://www.audioasylum.com/index.html
There is a search function on the right side of the page.
Select "all forums" and search for excelarray.
I like the idea to use larger diameter midranges as shown above
(vs. 3") and use ribbons for sonic nirvana, no cutting corners.
Actually, those Seas are woofer/mids, I call them combo drivers.
Feel the line source bass and midbass, let your subwoofer do the
rest.
The best suggestion is to contact Brad since he's been
listening to these beasts for a while now. I'm sure he
has a more detailed opinion.
expensive but........
**********************
Spend more, build once.
vs.
Spend less, build ten times.
**********************
http://www.selahaudio.com/excel.html
http://members.aol.com/bvirgil311/Excelarray.htm
** addition research **
Go here;
http://www.audioasylum.com/index.html
There is a search function on the right side of the page.
Select "all forums" and search for excelarray.
I like the idea to use larger diameter midranges as shown above
(vs. 3") and use ribbons for sonic nirvana, no cutting corners.
Actually, those Seas are woofer/mids, I call them combo drivers.
Feel the line source bass and midbass, let your subwoofer do the
rest.
The best suggestion is to contact Brad since he's been
listening to these beasts for a while now. I'm sure he
has a more detailed opinion.
planet10 said:
Funny, I had a major realization about that yesterday. I had driven the new Nonsuch speakers to a friend who uses the original System IVs, in order to make a comparison. To my great relief the new ones are much better than the old.
While listening to both pairs I realized that I like the comb effect when I stand up.
The speakers are quite attention grabbing when you're sitting down. I find it quite nice to be able to move out of the "show" a little when I stand.
It's the distance between the two furthest drivers that needs to be reduced. Minimizing distance between the drivers helps.navin said:
I run them all full-range***. With these particular speakers that's perfect when I'm sitting down. I lose the highest frequencies (over 10kHz) when I stand up but, as I just explained to Dave, I find it pleasant and quite handy to be able to move out of the experience a little when I stand.navin said:
Steve
*** I am experimenting with filter networks to slope off the high frequency response of the outer two drivers. In theory this could compensate for the diffraction step and reduce combing. I feel that I must try out this approach to hear what the pros and cons are (the caps and inductors should arrive any day now).
I must admit though, I'm a purist by heart when it comes to speaker design. For me to use any components (other than high quality cable) between the drive-units and the amplifier, they would have to be give a significant improvement without losing any of the things that I love about the speakers at the moment.
Line Source Misconceptions
Steve,
Let me comment on some of your statements in your posting.
You said
"It's the distance between the two furthest drivers that needs to be reduced. Minimizing distance between the drivers helps."
Spacign drivers closer together is a good thing. For a line source you want to create an elongated constant phase source which means spacing the drivers at no more than one wavelength at thier highest frequency of operation. Now you can get away with placing the comb lines in the 10-20 kHz region--the Flectcher-Munson curves show that the ear is less sensitive to sounds in this region versus the mid-range area. Hence, the spacing could be a far apart as 2 wavelengths at 20 kHz so the first cancellation occurs at that frequency. That says the spacing could be a wide as 1.35" apart--very small diameter drivers placed very close together.
Bottom line is that you want the drivers in the line to act as a single elongated driver and avoid comb lines if you can.
You said
"I run them all full-range***. With these particular speakers that's perfect when I'm sitting down. I lose the highest frequencies (over 10kHz) when I stand up but, as I just explained to Dave, I find it pleasant and quite handy to be able to move out of the experience a little when I stand."
Some people strive to have crisp highs that don't fall off or dull as frequency goes up and also have speakers that can be appreciated in both sitting and standing positions. Your design has drivers too far apart to prevent comb lines and you get the resultant loss of highs, uneven vertical dispersion, and reduced sensitivity as the comb lines are formed.
While it is nice to be a purist, you can expect better results if you follow solid acoustical concepts.
I think we had this same discussion before on any earlier thread at my posting:
http://www.diyaudio.com/forums/showthread.php?postid=195455#post195455
You can reread that posting and the subsequent replies.
Jim
Steve,
Let me comment on some of your statements in your posting.
You said
"It's the distance between the two furthest drivers that needs to be reduced. Minimizing distance between the drivers helps."
Spacign drivers closer together is a good thing. For a line source you want to create an elongated constant phase source which means spacing the drivers at no more than one wavelength at thier highest frequency of operation. Now you can get away with placing the comb lines in the 10-20 kHz region--the Flectcher-Munson curves show that the ear is less sensitive to sounds in this region versus the mid-range area. Hence, the spacing could be a far apart as 2 wavelengths at 20 kHz so the first cancellation occurs at that frequency. That says the spacing could be a wide as 1.35" apart--very small diameter drivers placed very close together.
Bottom line is that you want the drivers in the line to act as a single elongated driver and avoid comb lines if you can.
You said
"I run them all full-range***. With these particular speakers that's perfect when I'm sitting down. I lose the highest frequencies (over 10kHz) when I stand up but, as I just explained to Dave, I find it pleasant and quite handy to be able to move out of the experience a little when I stand."
Some people strive to have crisp highs that don't fall off or dull as frequency goes up and also have speakers that can be appreciated in both sitting and standing positions. Your design has drivers too far apart to prevent comb lines and you get the resultant loss of highs, uneven vertical dispersion, and reduced sensitivity as the comb lines are formed.
While it is nice to be a purist, you can expect better results if you follow solid acoustical concepts.
I think we had this same discussion before on any earlier thread at my posting:
http://www.diyaudio.com/forums/showthread.php?postid=195455#post195455
You can reread that posting and the subsequent replies.
Jim
And yet it does, all the way to the mass corner, or ~(2*Fs)/Qes, which due to the added 'drag' is slightly raised while Fs is slightly lowered. If the driver is used above its mass corner, then there's the thermal power compression that occurs on transients due to the stuffing's insulating effects, increasing with increasing power....... Really, it affects it to some extent all the way up till it starts transmitting either by TL or breakup modes.
Mount a small, moderate Q wide BW driver on an open baffle and staple several 'blanket' layers of stuffing over the rear of it to mimic an overstuffed cab, using EQ to BW limit its LF/midbass. I can hear quite a bit of difference, and since there's no LF/midbass in this configuration.......
GM
Re: Line Source Misconceptions
I had hoped that correspondence had ended with my posting: http://www.diyaudio.com/forums/showthread.php?postid=196370#post196370
Perhaps you could reread that posting and the postings preceding it.
This time you've pitched in with:
People's views of "solid acoustical concepts" can differ.
For example, you believe that it is necessary to hear the full frequency spectrum whilst standing up, while I believe that one can't achieve satisfactory coherence, integration or imagery by putting a crossover in the mid-range and moving from a cone bass device to a ribbon tweeter.
Personally, Jim, I find our exchanges rather tiresome. Can't we just agree to disagree?
Yes, I remember. That was the posting I started called Pictures of my range of speakers. You posted the following:Jim Griffin said:
You may remember my reply where I suggested that actually listening to the speakers, was a necessary precedent to judging them.
I had hoped that correspondence had ended with my posting: http://www.diyaudio.com/forums/showthread.php?postid=196370#post196370
Perhaps you could reread that posting and the postings preceding it.
This time you've pitched in with:
Jim, on rereading this pronouncement, would you come to the conclusion, as I did, that it is just a teeny bit arrogant?Jim Griffin said:
People's views of "solid acoustical concepts" can differ.
For example, you believe that it is necessary to hear the full frequency spectrum whilst standing up, while I believe that one can't achieve satisfactory coherence, integration or imagery by putting a crossover in the mid-range and moving from a cone bass device to a ribbon tweeter.
Personally, Jim, I find our exchanges rather tiresome. Can't we just agree to disagree?
Steve,
FWIW I think your speakers from a craftsmanship point of view
are lightyears beyond most of the DIY kits I have seen. From a
sound point of view I would never judge a speaker without
hearing it, reguardless of what acoustical laws say.
IMHO there really is no right, no wrong.. If there was a "right"
then every speaker would be built as such, in audio absolutes
absolutely do not exist. Keep up the great work!
FWIW I think your speakers from a craftsmanship point of view
are lightyears beyond most of the DIY kits I have seen. From a
sound point of view I would never judge a speaker without
hearing it, reguardless of what acoustical laws say.
IMHO there really is no right, no wrong.. If there was a "right"
then every speaker would be built as such, in audio absolutes
absolutely do not exist. Keep up the great work!
Spot on Brett. Isn't it this that holds our interest - that's there's still room in this business for maverick designers with original concepts? I've heard speakers that break several "rules" only to sound so good in one or two aspects that they are really beyond criticism. Isn't it a question of taste?Brett D. said:
And anyway, does anyone know of a speaker that has no compromises and doesn't break any "rules"? Look at all the rectangular boxes out there, the two-ways with mid-range crossovers, the mini line-arrays that you have to sit down to listen to (tee-hee), etc., etc., etc. Our task as designers is to choose which rules we break. If we think that we're not breaking any rules, we don't understand speaker design.
Thanks for your comments on the look of my speakers. I hope that you can get to listen to them some time.
thylantyr said:Navin likes the line source ? 😎 😎 😎
expensive but........
**********************
Spend more, build once.
vs.
Spend less, build ten times.
**********************
http://www.selahaudio.com/excel.html
http://members.aol.com/bvirgil311/Excelarray.htm
** addition research **
Go here;
http://www.audioasylum.com/index.html
I like the idea to use larger diameter midranges as shown above
(vs. 3") and use ribbons for sonic nirvana, no cutting corners.
1. i am trying to build a simpler system 1.5 way the excel array is a 3 way considering you need a sub for the lowest octave
2. i am hoping the system using the w3 871 is what i call proof of concept speaker. it will be usable and operational as a complete system but other will be able to build souped up versions using better drivers such as a jx 92 / 15"x2 combo or a system using a ribbon or small planar driver from 100Hz + and a couple of good 12" woofers.
BTW another pic of my prpoposed sub is below only the drivers in my proposed system will be firing sideways.
w3 871
interesting little driver for $12, individually they
probably wouldn't stand out in a crowd,
87db
xmax 0.5 mm
15w
but line array like needles looks cool and cost effective.
wonder how it sounds? for $12 you can find out if you
have an imagination.
😎
What is your project budget ?
interesting little driver for $12, individually they
probably wouldn't stand out in a crowd,
87db
xmax 0.5 mm
15w
but line array like needles looks cool and cost effective.
wonder how it sounds? for $12 you can find out if you
have an imagination.
😎
What is your project budget ?
well i am attempting to do a speaker that will look like 7V's nonsuch on top a dipole bass with the woofers firing sideways.
Navin;
/warning
/excessive rambling
/this is just fyi
I did an experiment with my assembled
SA drivers, four in a box.
Pretty crazy indeed.
I though one tweeter per channel
was good based on prior tests..
but had visions of a monster system...
well, four of these beasts is pretty
insane.. The spl level will drive someone
mad..
The old lady seems to prefer two
drivers per channel -- oddly enough
she wants more than one.. That is unusual -- perhaps she is getting mad
like me - haha
I've also compared this setup to my
eight line sourced HiVi planars ($25 ea.)
and I still think one SA planar beats
the line source of cheaper tweeters.
Two speaker cabinets -
Cost
16 low cost planars - $400
2 expensive SA drivers- $1200
In this case, less drivers is best.
Midranges?
hmm.. The mating midrange for
the SA cost $169, 100 db sensitivity,
you need two - $340
How much does 16 low cost drivers
run ? prolly the same as above,
but typically these drivers are 87db
sensitivity...
But, if the single SA/midrange solution
... for some crazy reason .. doesn't
yield enough SPL, then more is
better Even the small
line source of four is sufficient.
************************
So.. my crazy plan to use four tweets/mids is still going forward,
but I wonder if three 15's per channel
can keep up with this setup at full power. /haha I think this is a case,
the bass subsystem may be
weak.. I guess I need to add four
18" subwoofers later.... doh!!
I think experimentation is more rewarding that theorizing on paper,
building a final product -- then listening... repeat 10 times.
With some driver sampling, cheap
box experimentation and good old
fashion listening, you can mold a system tailored for your listening habits
which may not even be what you
theorized.. But the catch is... it takes
longer to build the final design -- it may
cost more, but you may save money
by not rebuilding again and again
and again....... I really don't want to build another system after this one,
my brain hurts
/warning
/excessive rambling
/this is just fyi
I did an experiment with my assembled
SA drivers, four in a box.
Pretty crazy indeed.
I though one tweeter per channel
was good based on prior tests..
but had visions of a monster system...
well, four of these beasts is pretty
insane.. The spl level will drive someone
mad..
The old lady seems to prefer two
drivers per channel -- oddly enough
she wants more than one.. That is unusual -- perhaps she is getting mad
like me - haha
I've also compared this setup to my
eight line sourced HiVi planars ($25 ea.)
and I still think one SA planar beats
the line source of cheaper tweeters.
Two speaker cabinets -
Cost
16 low cost planars - $400
2 expensive SA drivers- $1200
In this case, less drivers is best.
Midranges?
hmm.. The mating midrange for
the SA cost $169, 100 db sensitivity,
you need two - $340
How much does 16 low cost drivers
run ? prolly the same as above,
but typically these drivers are 87db
sensitivity...
But, if the single SA/midrange solution
... for some crazy reason .. doesn't
yield enough SPL, then more is
better
Even the smallline source of four is sufficient.
************************
So.. my crazy plan to use four tweets/mids is still going forward,
but I wonder if three 15's per channel
can keep up with this setup at full power. /haha I think this is a case,
the bass subsystem may be
weak.. I guess I need to add four
18" subwoofers later.... doh!!
I think experimentation is more rewarding that theorizing on paper,
building a final product -- then listening... repeat 10 times.
With some driver sampling, cheap
box experimentation and good old
fashion listening, you can mold a system tailored for your listening habits
which may not even be what you
theorized.. But the catch is... it takes
longer to build the final design -- it may
cost more, but you may save money
by not rebuilding again and again
and again....... I really don't want to build another system after this one,
my brain hurts
i agree it is often better to use one good driver than many less good drivers. however given that 4 W3 871 cost $50 can you recomend one full range driver that could equall 4 of them in performance (incl SPL capability etc..).
Navin, I'm just posting here to ensure that you've seen my earlier post on the "low-pass filter network thread" (number 14): Careful of the comb, navinnavin said:
i know. if i make a 1.5 way like you propose to do then my system gets more complicated as i already intend to have a dipole bass handling the lower end. in that case my sysmte is more than a 1.5 way but a 1.5.5 way! Yes I am as mad as you think.
I might even use 2" driver (Jordan or Bandor) to reduce the comb. I assume the Jordan 50mm module is still available in some form. BTW did you ever see the speakers mae my ALR Jordan apparently Ted Jordan had a hand in those as well. They used metal cones but were more conventional in design.
So it seems that I might have 5 channels of 1.5 way (usig multiple 50-75mm drivers) driven by my proposed 5 channel SE amp and a dipole sub using 2 x 12" (DPL 12) or 2 x 15" (lambda) driven by a 200-300W class A-B amp.
lets see.
BTW right now I am building a system to transport 300 ton (yes 300,00 kgs) transformers so I am used to extreme engineering.
I might even use 2" driver (Jordan or Bandor) to reduce the comb. I assume the Jordan 50mm module is still available in some form. BTW did you ever see the speakers mae my ALR Jordan apparently Ted Jordan had a hand in those as well. They used metal cones but were more conventional in design.
So it seems that I might have 5 channels of 1.5 way (usig multiple 50-75mm drivers) driven by my proposed 5 channel SE amp and a dipole sub using 2 x 12" (DPL 12) or 2 x 15" (lambda) driven by a 200-300W class A-B amp.
lets see.
BTW right now I am building a system to transport 300 ton (yes 300,00 kgs) transformers so I am used to extreme engineering.
Yes, Ted Jordan is in the top division of all-time drive unit designers. I don't know whether the 50mm Jordan is still available.navin said:
I would have thought that this is a little extreme for your listening room hi-fi but what do I know? How big are the capacitors in the regulator?
navin said:
I'm not sending subliminal messages to you as what
you should do, I was just posting some info based
on my experience with certain drivers. Line sources
are cool, but I think driver selection is still important
as with any design. You should listen to your drivers
before finalizing a design -- too make sure it sounds
good to you. If you configure your line source for spl
by paralleling drivers, the sound you get is essentially
the individual driver with more spl - heheheh
so any nuances of the driver gets magnified...
For $50 you can do a four driver single channel
experiment to see if you like the results.
- Status
- Not open for further replies.