Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

In Pursuit of a 20-20k Dipole Loudspeaker
In Pursuit of a 20-20k Dipole Loudspeaker
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 3rd January 2019, 07:36 AM   #21
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Wide range dipoles have two major challenges - low bass below say 60Hz and treble above 2kHz.

The bass issue comes from dipole principle, "loss by summation of opposite polarity of waves" When wavelength grows, it gets very difficult to compensate. Another problem comes when we put the radiator in a room with some walls and other boundaries. Easily it happens that backwave gets boosted when bouncing back from front wall, still polarity remains same (negative) if distance to wall is short relative to wavelength. Then at certain frequency the bounced wave is again in opposite phase to fronside wavefront and nulling happens. Above that boosting-nullings happens periodically.

So, a dipole bass only changes the frequency range of destructive interferences! Member kimmosto has measured cardioid vs dipole bass in a room
and John Kreskowsky tells theory

Basics told:
Speaker Placement 101: How to Fight Boundary Interference (monopole source)
Dan Russell's Acoustics and Vibration Animations

This is why I decided to use monopole bass below 200Hz. When monopole is combined with dipole, there is cardioid pattern in crossover region. This seems to work well in practise. Same principle is used for Gradient 1.x and Martin Logan panels.

Craw woofer
http://www.musicanddesign.com/VariabDF.html
__________________
Radikal aktivist AINOgradient speaker project
  Reply With Quote
Old 3rd January 2019, 05:48 PM   #22
CharlieLaub is offline CharlieLaub  United States
diyAudio Member
 
Join Date: Mar 2007
Location: Michigan
Quote:
Originally Posted by John Busch View Post
OK... Would you consider a 1% (92 dbw) a usable half space efficiency target assuming a 23"-25" net baffle width and a max effective overall depth (path difference) of approximate 15"-16"?

Also will assume 107 db clean output at or above 28-30 Hz half space per speaker would be low enough and loud enough to qualify as a full range residential speaker? At 32 - 33 Hz we can produce 1 acoustic watt per speaker, (109 db/mtr forward radiation) again in residential half space. At 20 Hz a pair can still reach 100 db. Those last 12 Hz (from 32 Hz to 20 Hz) are a real output killer! This I have built before.
Let me give you an example of an H-frame that I built and measured as a test bed early last year (currently disassembled). This used an 18" 90dB/W high-Xmax subwoofer in an H-frame dimensioned 24"W, 24"H and 30"D.

I measured the response at the front and rear openings, in the plane of the opening, in the middle of the opening. From that I can create a model of the dipole without having try and measure it directly. This is because at low frequencies wavelengths are long and you can model the front and rear as two monopoles, the rear one reversed in polarity, and separated by D=30". The first plot shows the SPLs of front (cyan) rear (yellow) and the dipole response (dark blue) at a distance of 2m in front of the H-frame. The levels have been normalized so that the front SPL is 0dB in the region of 30-50Hz. I also compensated for the extra distance the rear wave must travel to the listening position in its SPL, and this results in about a 3dB lower level for it at the listening position as shown in the plot. To make the blue (system) curve I take the FRD files and apply an inversion and delay, plus the -3dB SPL distance correction, to the rear data before summing them.

In Pursuit of a 20-20k Dipole Loudspeaker-frame-front-rear-simulated-dipole-2m-png
The CYAN curve is the measured response at the front opening.
The YELLOW curve is the measured response at the rear opening.
The BLUE curve is the synthesized H-frame response (its dipole response) at the listening position, 2m distant and in front of the front opening.

Looking at the plot above, we can observe the following features:
  • The driver response, e.g. at the front opening of the H-frame, is relatively flat to 30Hz.
  • There are some resonances in the front and rear response between 70Hz and 150Hz.
  • The 1/4 wavelength transmission line resonance of the H-frame creates a null around 375Hz and the response starts falling into this around 150Hz.
  • The H-frame response has a dipole peak around 140Hz, and this is +6dB WRT the 0dB level
  • Below the dipole peak there is a region with a 6dB/oct slope from about 40Hz to 120Hz
  • Below 40Hz the rolloff resulting from the driver's fundamental resonance kicks in and the response falls more quickly but only approaching 12dB/oct from 10Hz-40Hz
  • At 40Hz the sensitivity of the dipole is only 3dB less than the response at the front, which is essential the 90dB/W free air sensitivity of the driver, so 87dB/W@1m
Now I will apply some EQ and other response shaping. This will include a boost at low frequencies to flatten and extend the passband. The EQ curve looks like this:

In Pursuit of a 20-20k Dipole Loudspeaker-filters_and_eq-png

In the above curve for the correction filter I am applying 3dB of boost at 40Hz. Boost increases below 40Hz to lift the response and only a LP filter type rolloff is used to both compensate for the dipole peak and create a LP crossover filter. This is possible because the Q of the dipole peak is not too high and the necessary correction is simple. It is a similar approach to the one that Martin King used in his paper comparing an OB, U-frame, and H-frame.

Below is the resulting H-frame dipole subwoofer frequency response, at the listening position:

In Pursuit of a 20-20k Dipole Loudspeaker-eqd_response-png

The response is flat to 20Hz. Even though the correction filter applies up to 12dB of boost, this happens at low frequencies and the relative program level near 20Hz is not as high as above 100Hz so this much boost is not so difficult to manage. Less boost can be used at the cost of some low frequency extension. The H-frame dimension D and tunnel cross section vs Sd has been chosen so that the dipole peak is about where you want to cross over to the next higher band, and the Q of the peak is not too high.

I did build this subwoofer and I knew the first time I ran an impulse sweep on it that it had some impressive bass. Even if you did not use boost (eg. a passive filter only) the response extends to 40Hz before rolling off. I took the data and did the modeling, and this is what I had planned to create but I had to disassemble the system. I hope to build it again soon with some nice, heavy materials to keep the H-frame in place and resist the cone movement.
Attached Images
File Type: png H-frame front rear simulated dipole at 2m.PNG (62.6 KB, 757 views)
File Type: png EQd_response.PNG (36.0 KB, 748 views)
File Type: png filters_and_EQ.PNG (30.1 KB, 432 views)
__________________
Visit my Audio Web Page <<--CLICK TO LEARN MORE-->> Get my LADSPA plugins
  Reply With Quote
Old 3rd January 2019, 05:59 PM   #23
plasnu is offline plasnu  United States
diyAudio Member
 
Join Date: Jan 2011
Location: NY USA
CharlieLaub, this is an amazing outcome. I'm curious how big your room is, and distance between the wall - sub - listening position.
  Reply With Quote
Old 3rd January 2019, 06:10 PM   #24
Scott L is offline Scott L  United States
diyAudio Member
 
Join Date: Dec 2008
Location: Knoxville, TN
In Pursuit of a 20-20k Dipole Loudspeaker
Quote:
Originally Posted by Juhazi View Post
Wide range dipoles have two major challenges - low bass below say 60Hz and treble above 2kHz.

The bass issue comes from dipole principle, "loss by summation of opposite polarity of waves" When wavelength grows, it gets very difficult to compensate. Another problem comes when we put the radiator in a room with some walls and other boundaries. Easily it happens that backwave gets boosted when bouncing back from front wall, still polarity remains same (negative) if distance to wall is short relative to wavelength. Then at certain frequency the bounced wave is again in opposite phase to fronside wavefront and nulling happens. Above that boosting-nullings happens periodically.

So, a dipole bass only changes the frequency range of destructive interferences! Member kimmosto has measured cardioid vs dipole bass in a room
and John Kreskowsky tells theory

Basics told:
Speaker Placement 101: How to Fight Boundary Interference (monopole source)
Dan Russell's Acoustics and Vibration Animations

This is why I decided to use monopole bass below 200Hz. When monopole is combined with dipole, there is cardioid pattern in crossover region. This seems to work well in practise. Same principle is used for Gradient 1.x and Martin Logan panels.

Craw woofer
http://www.musicanddesign.com/VariabDF.html
Where is the LIKE button ? seriously though, it's nice to see some sanity in the audio world, for a change.
  Reply With Quote
Old 3rd January 2019, 06:45 PM   #25
CharlieLaub is offline CharlieLaub  United States
diyAudio Member
 
Join Date: Mar 2007
Location: Michigan
Quote:
Originally Posted by plasnu View Post
CharlieLaub, this is an amazing outcome. I'm curious how big your room is, and distance between the wall - sub - listening position.
The room was about 20x15 feet. I just tested the sub (made measurements on it) out in the middle of the room. It was never really set up for listening.

Also, keep in mind that I did not measure the response in room, I only predicted it from a model. It should be accurate, but the model did not include the room boundaries, room reflections, relative location in the room, and so on. But I did listen to the un EQd sub during testing and was impressed.
__________________
Visit my Audio Web Page <<--CLICK TO LEARN MORE-->> Get my LADSPA plugins

Last edited by CharlieLaub; 3rd January 2019 at 07:10 PM.
  Reply With Quote
Old 3rd January 2019, 06:58 PM   #26
CharlieLaub is offline CharlieLaub  United States
diyAudio Member
 
Join Date: Mar 2007
Location: Michigan
Quote:
Originally Posted by Juhazi View Post
Wide range dipoles have two major challenges - low bass below say 60Hz and treble above 2kHz.

The bass issue comes from dipole principle, "loss by summation of opposite polarity of waves" When wavelength grows, it gets very difficult to compensate. Another problem comes when we put the radiator in a room with some walls and other boundaries. Easily it happens that backwave gets boosted when bouncing back from front wall, still polarity remains same (negative) if distance to wall is short relative to wavelength. Then at certain frequency the bounced wave is again in opposite phase to fronside wavefront and nulling happens. Above that boosting-nullings happens periodically.

So, a dipole bass only changes the frequency range of destructive interferences! Member kimmosto has measured cardioid vs dipole bass in a room
and John Kreskowsky tells theory

Basics told:
Speaker Placement 101: How to Fight Boundary Interference (monopole source)
Dan Russell's Acoustics and Vibration Animations

This is why I decided to use monopole bass below 200Hz. When monopole is combined with dipole, there is cardioid pattern in crossover region. This seems to work well in practise. Same principle is used for Gradient 1.x and Martin Logan panels.

Craw woofer
http://www.musicanddesign.com/VariabDF.html
Your use of a monopole for the lowest band, and the resulting cardioid pattern in the crossover band to the next higher one, is certainly one viable approach. It doesn't fit into my 20-20k dipole approach, however. That doesn't mean you are wrong or I am right, I am just trying to achieve a different goal than you have done.

I have seen lots made about how bad dipole are at low frequencies because of this and that. But personally I have heard several dipole systems with excellent bass. An (early?) version of John Bush's "Widowmaker" used a large planar baffle and two 18" woofers. It had quite astonishing low bass impact. Siegfried Linkwitz's LX521 uses two 10" Seas woofer in a short M-frame and that has lots of bass, too, even in a small listening room (e.g. at a Burning Amp, not Fort Mason). So it seems that these dire predictions just don't quite measure up in real life from what I have heard.
__________________
Visit my Audio Web Page <<--CLICK TO LEARN MORE-->> Get my LADSPA plugins
  Reply With Quote
Old 3rd January 2019, 07:33 PM   #27
John Busch is offline John Busch  United States
diyAudio Member
 
John Busch's Avatar
 
Join Date: Oct 2007
Location: Hill country, Texas
How to say this.... I have always found the belief that dipoles can not work well below 60 or so Hz to be curious. It just comes down to volume of displacement (VD). A pair of 18" woofers I was referring to in my previous post have, at 32 Hz, a useful one way displacement of 2,695 CC. A stereo pair is of course twice that at 5,390 CC! For comparison, a typical long throw 12" sub woofer is around 400-500 CC. A big OB just overwhelms the losses. I have demo'd OB's at a number of venues, small private to RMAF. Have yet to have anyone comment on the lack of or quality of bass. If anything, a few thought there was a bit too much. There are very few commercial speakers systems intended for residential use that can deliver an acoustic watt at 32 Hz. Most can't even approach that level of output.

Admittedly, a dual 18" OB is a bit of overkill and not for every one. Even my relatively small Manzanita with a single medium throw 12" or short throw 15" provides more and deeper low distortion bass than the average high end speaker. Most traditional box loaded (sealed or ported) start to roll off on the low end around 60 Hz. Very few are flat at or below 50 Hz. The Manzanita is flat at 42 Hz.
__________________
Commercial Website: https://manzanita-audio.com/
  Reply With Quote
Old 3rd January 2019, 10:08 PM   #28
fporegon is offline fporegon  United States
diyAudio Member
 
Join Date: Jun 2011
Agree - it is a matter of volume displacement only. I employ 6 x 12" in a stereo H-Frame tower. Very shallow wings - 3". These drivers cost me $18 each - so I discount their 8mm peak xmax spec. Probably more like 8mm peak to peak. This would put volume displacement around 2k cubic centimeters. If it were not for the need to reproduce movie bass - I'd target smaller drivers (in an array) and get some extra bandwidth.

This is a dual purposed system - high quality audio 90% of the time and the occasional movie. Not for everyone of course but they are flat to 30 Hz with EQ.

The bass is so clean and authoritative that I tend to overemphasize it and set the gain for the low channels too high.
  Reply With Quote
Old 3rd January 2019, 10:12 PM   #29
scottjoplin is offline scottjoplin  Wales
diyAudio Member
 
scottjoplin's Avatar
 
Join Date: Dec 2015
Location: Penrhyndeudraeth
Quote:
Originally Posted by fporegon View Post
Agree - it is a matter of volume displacement only.
Can you clarify what you mean please?
__________________
Woofer Assisted Wideband is the New Testament renounce the anachronistic acronym FAST
  Reply With Quote
Old 3rd January 2019, 10:24 PM   #30
plasnu is offline plasnu  United States
diyAudio Member
 
Join Date: Jan 2011
Location: NY USA
Quote:
Originally Posted by CharlieLaub View Post
The room was about 20x15 feet. I just tested the sub (made measurements on it) out in the middle of the room. It was never really set up for listening.

Also, keep in mind that I did not measure the response in room, I only predicted it from a model. It should be accurate, but the model did not include the room boundaries, room reflections, relative location in the room, and so on. But I did listen to the un EQd sub during testing and was impressed.
I see, thank you for the clarification.
  Reply With Quote

Reply


In Pursuit of a 20-20k Dipole LoudspeakerHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Rockin' the KaZba Dipole (K aperture Z-baffle Dipole) xrk971 Full Range 332 24th February 2018 01:30 PM
Dipole Ribbon tweeter isolation from dipole mid-woofer array Bent Planars & Exotics 8 29th March 2015 09:45 AM
Advice Needed On Dipole/Bipole Wal Mounted Surround Loudspeaker philiprst Multi-Way 9 7th October 2008 07:51 PM
The Eternal Pursuit. devzero Introductions 8 15th August 2008 05:13 AM
Speakers, Ebay, And, The Pursuit Of Winning The Auction OMNIFEX Multi-Way 4 14th March 2003 09:37 AM


New To Site? Need Help?

All times are GMT. The time now is 10:07 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki