hi guys i know this question might have been asked before but i am a bit too rushed / lazy to search. forgive me dave (planet 10) for I have sinned....
I am considering using a dipole bass for my next system. never mind that i have not yet chosen the drivers. all i know now is that the max system dimensions should be 10" wide x 20" deep. and that is max i would prefer ti to be 8" Wx 18"D. height i guess will be limited by the fact that the ear is about 32-36" off the ground
question: given my space limitations can i even consider dipole bass (2 x 15" would be my choice). or should I look at TL. I have already built various bass reflex and sealed box systems so I am now looking at clean fast non boomy bass.
a prelinnamry study tells me that dipole are very lossy
http://www.diysubwoofers.org/dipole/
this para scared me..."Drawbacks
Dipole bass systems tend to be rather large, employing multiple drivers, primarily to make up for the output reduction due to the 6dB/oct baffle loss. This is not the type of system to use if you've got a small living room, and it's certainly not suitable for car audio!"
thanks.
I am considering using a dipole bass for my next system. never mind that i have not yet chosen the drivers. all i know now is that the max system dimensions should be 10" wide x 20" deep. and that is max i would prefer ti to be 8" Wx 18"D. height i guess will be limited by the fact that the ear is about 32-36" off the ground
question: given my space limitations can i even consider dipole bass (2 x 15" would be my choice). or should I look at TL. I have already built various bass reflex and sealed box systems so I am now looking at clean fast non boomy bass.
a prelinnamry study tells me that dipole are very lossy
http://www.diysubwoofers.org/dipole/
this para scared me..."Drawbacks
Dipole bass systems tend to be rather large, employing multiple drivers, primarily to make up for the output reduction due to the 6dB/oct baffle loss. This is not the type of system to use if you've got a small living room, and it's certainly not suitable for car audio!"
thanks.
Dipole, the misunderstood animal....
Konnichiwa,
Here are some seeming paradoxes for you. A dipole of fairly modest dimension (65X105cm) can support a measured in room response in the region of 50..60Hz, if the driver fitted is suitable, meaning a Qt > 0.5 with an Fs in the 50-60Hz region.
Now this is observed fact. The old theory says the rolloff should start with -3db @ around 130Hz. Of course, the old theory is WRONG. May I suggest to read up on some more modern evaluations? A really good read which Planet 10 has at hand IIRC is "J Backman - Numerical Analysis of Open Baffle Loudspeakers".
This interestingly shows a test evaluation of a 400 X 300mm open Baffle including a Driver with no LF rolloff. The Baffle correctly analysed introduces some mild (<+/-2db) ripple and adds some level in the range between 100Hz & 200Hz while having a -3db point of 100Hz!!!
That is a 12" X 15" Baffle folks!!!
According to a recent german study the -6db point for a dipole is at 0.07 * Lambda/W, where W is the average width of the baffle, meaning take width plus hight, divide by 2 or more precesiely, measure from driver center to edge into all main (8) directions, add the measurements and divide by 4. Lambda is BTW 345m/S, but that is obvious.
Now, if we look at our 65 X 105cm baffle we have average 0.85m width. So 0.07 * 345/0.85 = 28Hz....
Obviously the last word is not yet spoken as to dimension, but quite finite size baffles when combined with suitable chassis (Qt > 0.5 up to 1) can have a rather low reaching LF response.
Of course, you cannot rely on either rrom gain or the pressure chamber effect of the room to give you a hand with the LF output.... ;-) And in most normal rooms large speakers with a flat LF response will show a lot of LF boost, which will be missed with a dipole....
Sayonara
Konnichiwa,
Here are some seeming paradoxes for you. A dipole of fairly modest dimension (65X105cm) can support a measured in room response in the region of 50..60Hz, if the driver fitted is suitable, meaning a Qt > 0.5 with an Fs in the 50-60Hz region.
Now this is observed fact. The old theory says the rolloff should start with -3db @ around 130Hz. Of course, the old theory is WRONG. May I suggest to read up on some more modern evaluations? A really good read which Planet 10 has at hand IIRC is "J Backman - Numerical Analysis of Open Baffle Loudspeakers".
This interestingly shows a test evaluation of a 400 X 300mm open Baffle including a Driver with no LF rolloff. The Baffle correctly analysed introduces some mild (<+/-2db) ripple and adds some level in the range between 100Hz & 200Hz while having a -3db point of 100Hz!!!
That is a 12" X 15" Baffle folks!!!
According to a recent german study the -6db point for a dipole is at 0.07 * Lambda/W, where W is the average width of the baffle, meaning take width plus hight, divide by 2 or more precesiely, measure from driver center to edge into all main (8) directions, add the measurements and divide by 4. Lambda is BTW 345m/S, but that is obvious.
Now, if we look at our 65 X 105cm baffle we have average 0.85m width. So 0.07 * 345/0.85 = 28Hz....
Obviously the last word is not yet spoken as to dimension, but quite finite size baffles when combined with suitable chassis (Qt > 0.5 up to 1) can have a rather low reaching LF response.
Of course, you cannot rely on either rrom gain or the pressure chamber effect of the room to give you a hand with the LF output.... ;-) And in most normal rooms large speakers with a flat LF response will show a lot of LF boost, which will be missed with a dipole....
Sayonara
Re: Dipole, the misunderstood animal....
I was actually there when Juha gave that presentation ... i have to do some more digging and find the ROM with the preprints -- i don't know if that ever got into the journal.
The much smaller baffle size rquirement jives with many people's expressed empirical results with baffles much smaller than suggested by Olsen's work.
dave
Kuei Yang Wang said:
I was actually there when Juha gave that presentation ... i have to do some more digging and find the ROM with the preprints -- i don't know if that ever got into the journal.
The much smaller baffle size rquirement jives with many people's expressed empirical results with baffles much smaller than suggested by Olsen's work.
dave
Nobody has mentioned it yet but the dipole "reference" on the net is Sigfried Linkwitz's site. If you do some digging, there's lots of good reading there about theory and actual speaker designs. I haven't heard them yet but I've heard nothing but good reports about the sound of his latest design, the Orion.
Linkwitz lab
Linkwitz lab
No side panels ?
Are we talking about a flat panel with no "side panels " ?
If this 'low' 28Hz roll off is correct , how come the dipoles that we are seeing - like those on the Linkwitz site - need so much bass equalisation. I don't quite get what the practical implementation is like. Could you give more details ?
Thanks.
Are we talking about a flat panel with no "side panels " ?
If this 'low' 28Hz roll off is correct , how come the dipoles that we are seeing - like those on the Linkwitz site - need so much bass equalisation. I don't quite get what the practical implementation is like. Could you give more details ?
Thanks.
Navin;
Methinks you need something like
this... The Rick Craig line source
called Excelarray.
http://www.madisound.com/cgi-bin/discuss.cgi?read=260191
Rumor has it there is a version in
the works that is 2x taller.
Or something more affordable like
the Jim Griffin arrays
http://www.madisound.com/cgi-bin/discuss.cgi?read=260091
The links are provided to show pics
only.
If I were you, I would not finalize
any design until you made some
test boxes using cheap particle board
to verify that it's the design for you.
One common use for TL is to get
low end extension out of small or cheap
drivers, but there are some who have
large woofer TL's, in which case they
are big... and you said you don't have room.
The line source .. while would sound great in a large room, isn't so bad in
a medium sized room. Referencing
the pics, do you think a design like
that fits in the house? These systems
will give you the "fast bass" and
not be boomy bass.
Sounds like you need to audition some
systems first to get a feel for what
might work for you.
The quest for sonic perfection never
ends - heheh
Methinks you need something like
this... The Rick Craig line source
called Excelarray.
http://www.madisound.com/cgi-bin/discuss.cgi?read=260191
Rumor has it there is a version in
the works that is 2x taller.
Or something more affordable like
the Jim Griffin arrays
http://www.madisound.com/cgi-bin/discuss.cgi?read=260091
The links are provided to show pics
only.
If I were you, I would not finalize
any design until you made some
test boxes using cheap particle board
to verify that it's the design for you.
One common use for TL is to get
low end extension out of small or cheap
drivers, but there are some who have
large woofer TL's, in which case they
are big... and you said you don't have room.
The line source .. while would sound great in a large room, isn't so bad in
a medium sized room. Referencing
the pics, do you think a design like
that fits in the house? These systems
will give you the "fast bass" and
not be boomy bass.
Sounds like you need to audition some
systems first to get a feel for what
might work for you.
The quest for sonic perfection never
ends - heheh
thylantyr said:Navin;
Methinks you need something like
this...
Or something more affordable like
the Jim Griffin arrays
If I were you, I would not finalize
any design ...but there are some who have
large woofer TL's, in which case they
are big... and you said you don't have room.
The line source .. Sounds like you need to audition some
systems first to get a feel for what
might work for you.
The quest for sonic perfection never
ends - heheh
1. thanks for the links. will look at them.
2. line source...now that is an option. lets see...
3. at the hi fi show in Neu Isenberg (Frankfurt) in May-June an italian company that also demoed a 300kg 200W amp had a rather "small" dipole not much larger than a WATT/Puppy using 2 x 12". the top end was a 6" 2 way. the room was not large but the bass was tight and not boomy. what i notice is that in smaller rooms dipoles do not boom as easily as sealed or bass reflex. having not really tried TL I cant speak for that design.
I'd rather have no bass than bass overhang. The bass resonance distracts from the music. if there is no bass the brain usually processes it. Mr Bose and all the mini speaker guys were on to something there.
BTW, presently I am using 2 x 12 (per channel) in a 130 liter sealed box.
line arrays are getting interesting (just saw the links)....
Q1: would you consider dipole line arrays?
Q2: what is a good ribbon tweeter? the BG neo 8? for the line array?
Q3: what is a good mid for the line array i am considering PHL 8" or a Scan Speak or Excel or one fo the new Focal 7/8"
Q1: would you consider dipole line arrays?
Q2: what is a good ribbon tweeter? the BG neo 8? for the line array?
Q3: what is a good mid for the line array i am considering PHL 8" or a Scan Speak or Excel or one fo the new Focal 7/8"
Re: No side panels ?
Konnichiwa,
Of course.
I think it is slightly optimistic, also I may have mixed up (on second thought) -6db & -9db. The point is those baffles go a lot lower than the old theories suggest.
Unsuitable drivers. A driver for a Dipole MUST have a Qt of > 0.5, better at least 0.7. With a large enough baffle and a Qt of 0.7 you have -3db at the drivers fundamental resonance, assuming the baffle does not lead to additional problems.
BTW, folding the large baffle changes the whole system and behaves differently, putting wings on a baffle does comparably little to extend bass, as the issues we need to consider are not primarily/just acoustic shortcircuiting due to the finite pathlength.
In case of a friends system, Baffle 65cm X 105cm (flat), driver is mounted according to the golden ratio. Driver is Qt 0.7, Fs = 55Hz. In the baffle Fs drops. The result, in room LF down to around 50Hz (-6db) with flat LF response (no lift as usual) and more interestingly, when measuring the system the RT60 does not rise to low frequencies anywhere nearly as fast and as much as with normal speakers. BTW, there is NO equalisation.
Sayonara
Konnichiwa,
ashok said:
Of course.
ashok said:
I think it is slightly optimistic, also I may have mixed up (on second thought) -6db & -9db. The point is those baffles go a lot lower than the old theories suggest.
ashok said:
Unsuitable drivers. A driver for a Dipole MUST have a Qt of > 0.5, better at least 0.7. With a large enough baffle and a Qt of 0.7 you have -3db at the drivers fundamental resonance, assuming the baffle does not lead to additional problems.
BTW, folding the large baffle changes the whole system and behaves differently, putting wings on a baffle does comparably little to extend bass, as the issues we need to consider are not primarily/just acoustic shortcircuiting due to the finite pathlength.
ashok said:
In case of a friends system, Baffle 65cm X 105cm (flat), driver is mounted according to the golden ratio. Driver is Qt 0.7, Fs = 55Hz. In the baffle Fs drops. The result, in room LF down to around 50Hz (-6db) with flat LF response (no lift as usual) and more interestingly, when measuring the system the RT60 does not rise to low frequencies anywhere nearly as fast and as much as with normal speakers. BTW, there is NO equalisation.
Sayonara
Q1: would you consider dipole line arrays?
Personal taste. you pick.
Check this out, pretty wild ?
http://www.alonbyacarian.com/products/exoticagr_b.htm
Q2: what is a good ribbon tweeter?
The DIY line sources made by folks
who contribute on madisound.com
forum have use the HiVi (or equivalent)
ribbon, $25 ea. I have eight of these,
they are ok. Eight in parallel increases
SPL.... For a budget system, it works
well. The Jim Griffin Linus array uses
these..
The Esg or Raven array is expensive.
But you could make an array similar
to the Jim Griffin "needles" which
uses the TB midrange drivers and one
quality tweeter in the center.
scroll down to see needles.
http://www.creativesound.ca/
the BG neo 8? for the line array?
People have been reporting that these drivers are not really tweeters, more
like high midrange. The top end roll off
is nasty. I would avoid.
The Rick Craig Excelarray uses the
newform ribbons and Seas Excels,
a system that would interest even me
for casual listening. But I have not
tried these drivers myself.
The next link below, you can find the
ribbons used in the Excelarray and I
believe they sell complete systems too.
I don't remember which model number
was used, but you can ask on madisound.com forum.
http://www.newformresearch.com/
For some reason R45 sticks in my mind.
91 db sensitivity, 45" ribbon looks interesting.
http://www.newformresearch.com/factory_direct_prices.html
You should research madisound.com
forum for recent and archived discussion on line arrays, there is good
stuff there to read.
For the line source project you would
need a combination driver capable
of performing good midbass and midrange. Then add the subwoofer.
I haven't tried Seas Excels, but everyone on the other forum speaks
highly of these drivers for line arrays,
but you have to deal with the nasty
break up modes. One online person I
know used four of these Seas Excels
in their car, in a ported box and crosses
them over at 1.2khz 4th order. /hehe
He said the midbass is awesome and
the nasty gremlins are avoided. He will
use a horn driver to mate with the low
crossover point (a non standard design).
I'm sure you can get some crossover ideas or designs for the Seas drivers.
PHL would be sweet sonically in a line
source such as the PHL1120, but the
driver is only good down to 200 - 300hz, this creates a problem. You
would have to use the bass version
of the driver, I think it's the 1240, but
I have no experience with the 1240.
My line source idea was to use
SA planars (eight per channel),
PHL midranges (eight per channel)
and Lambda TD15's (six per channel)
This is one big and nasty system
that cost an arm and two legs.
If you ditch the 15" woofers and
PHL, you can insert the Seas Excel
driver as the compromise and it's
more down to earth design, but
won't have the SPL levels I seek.
SA planar are way too overpowering
for the low sensitivity Excels, the newforms look like a good match.
In other words, the Excelarray design
is a good choice, if you want higher
SPL you would have to go to extreme
measures.
Personal taste. you pick.
Check this out, pretty wild ?
http://www.alonbyacarian.com/products/exoticagr_b.htm
Q2: what is a good ribbon tweeter?
The DIY line sources made by folks
who contribute on madisound.com
forum have use the HiVi (or equivalent)
ribbon, $25 ea. I have eight of these,
they are ok. Eight in parallel increases
SPL.... For a budget system, it works
well. The Jim Griffin Linus array uses
these..
The Esg or Raven array is expensive.
But you could make an array similar
to the Jim Griffin "needles" which
uses the TB midrange drivers and one
quality tweeter in the center.
scroll down to see needles.
http://www.creativesound.ca/
the BG neo 8? for the line array?
People have been reporting that these drivers are not really tweeters, more
like high midrange. The top end roll off
is nasty. I would avoid.
The Rick Craig Excelarray uses the
newform ribbons and Seas Excels,
a system that would interest even me
for casual listening. But I have not
tried these drivers myself.
The next link below, you can find the
ribbons used in the Excelarray and I
believe they sell complete systems too.
I don't remember which model number
was used, but you can ask on madisound.com forum.
http://www.newformresearch.com/
For some reason R45 sticks in my mind.
91 db sensitivity, 45" ribbon looks interesting.
http://www.newformresearch.com/factory_direct_prices.html
You should research madisound.com
forum for recent and archived discussion on line arrays, there is good
stuff there to read.
For the line source project you would
need a combination driver capable
of performing good midbass and midrange. Then add the subwoofer.
I haven't tried Seas Excels, but everyone on the other forum speaks
highly of these drivers for line arrays,
but you have to deal with the nasty
break up modes. One online person I
know used four of these Seas Excels
in their car, in a ported box and crosses
them over at 1.2khz 4th order. /hehe
He said the midbass is awesome and
the nasty gremlins are avoided. He will
use a horn driver to mate with the low
crossover point (a non standard design).
I'm sure you can get some crossover ideas or designs for the Seas drivers.
PHL would be sweet sonically in a line
source such as the PHL1120, but the
driver is only good down to 200 - 300hz, this creates a problem. You
would have to use the bass version
of the driver, I think it's the 1240, but
I have no experience with the 1240.
My line source idea was to use
SA planars (eight per channel),
PHL midranges (eight per channel)
and Lambda TD15's (six per channel)
This is one big and nasty system
that cost an arm and two legs.
If you ditch the 15" woofers and
PHL, you can insert the Seas Excel
driver as the compromise and it's
more down to earth design, but
won't have the SPL levels I seek.
SA planar are way too overpowering
for the low sensitivity Excels, the newforms look like a good match.
In other words, the Excelarray design
is a good choice, if you want higher
SPL you would have to go to extreme
measures.
Another interesting thread regarding dipole sub size is over here:
http://www.audioasylum.com/forums/HUG/messages/52059.html
Compare the size of the dipole bass to the BR bass, that the reviewers rated quite similar to the performance of the dipoles.
http://www.audioasylum.com/forums/HUG/messages/52059.html
Compare the size of the dipole bass to the BR bass, that the reviewers rated quite similar to the performance of the dipoles.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.