Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

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 9th March 2013, 01:15 PM   #11
guangui is offline guangui  Puerto Rico
diyAudio Member
 
Join Date: Feb 2013
What? No listening impressions yet!!! All these nice pics, and you left us hanging in there, waiting for more...Nice work, look forward to your conclusion, and measurements.
  Reply With Quote
Old 9th March 2013, 04:29 PM   #12
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä
Nice and unconventional work!

I see the heritage of Carlson here. There is a lot of very early reflections from the wall and structure behind the mid. You might need a session or two to settle with them.
__________________
AES Associate Member / My DIY speaker history: -74 Philips 3-way, -82 Hifi 85B, -07 Zaph L18, XLS10+PR/Hypex, -08 CSS125FR, -08 Hifitalo AW-7, -08 TangBand FR, -09 MarkK ER18DXT, -13 PPSL470, -13 AINOgradient

Last edited by Juhazi; 9th March 2013 at 04:42 PM.
  Reply With Quote
Old 9th March 2013, 10:43 PM   #13
diyAudio Member
 
Join Date: Mar 2010
Location: Stockholm
A little bit about my listening impressions.


I’m really pleased the basic principle works as intended. The stereo image stays between the speakers in a large area of the listening room and does not collapse as you move closer to one speaker. Sure, the stereo image is most natural when you sit in the middle between the speakers, but you still experience a sound stage between the speakers even when you sit right in front of one speaker at normal listening distance. And the sound keeps the same character as you move around in the whole room, which helps both the perception of the direct sound and the reflections.


The wall behind the speakers disappears and the sound stage is built up between, behind, to the outside and above the speakers, depending on the recording. I think it is hard to identify the speakers in the image. My better half says it sounds as for real. And I promise, this is nothing I have ever said she should say to make me happy.


Since the dipolar speaker never will pressurize the room in the same way as a conventional speaker, I have to do without the strong disco punch until I drag in a sub woofer or install a bass shaker to the sofa. Bass is for sure there when it is present in the mix, but never over emphasized or physically overwhelming. Instead, what I get is a very even pitch balance throughout all registers, where I find it simple to follow any instrument or what the musicians are doing irrespective of loud or soft in the mix. And I basically don’t experience any problems with room nodes, standing waves or comb filtering effect from the front wall in my listening room.


The force cancellation between the bass drivers in the W-baffle is also a cool experience. It’s fascinating to see the units pump air at full excursion and the chassis remains completely calm.


Another thing that has surprised me somewhat is that I find it hard to achieve a real sense of loudness from these speakers. Sure, they are not the most sensitive things out there and they need a fair amount of watts in the bass to compensate for the dipole cancellation. But I can in reality measure sound pressures well over 100 dB in my room before my amplifier gives up and it still doesn’t sound loud. I try to still my audiophile diy itch with that this is a good sign of low colouration from my speakers, without a lot of distortion that might give a initial impression of loudness. But tire you long term. What are your experiences around this out there?


I’m very pleased with how the individual drivers perform in my application. I would have appreciated increased sensitivity from the Dipole 10’s to counteract the dipole cancellation, otherwise they perform to my full satisfaction. The Seas Nextel 8” has an edge reflection around 800 Hz which has to be equalized, but still to my ears remains neutral throughout their operating range. The angling puts strong requirements on good dispersion in the upper registers, but I believe the Illuminators and Seas midrange live up to those expectations in my design.
  Reply With Quote
Old 9th March 2013, 11:14 PM   #14
diyAudio Member
 
Join Date: Mar 2010
Location: Stockholm
Juhazi,

Sure I have some early reflections from the structures for the backwards projection. But that energy is basically aimed to assist the diffused sound and I'm not too concerned with diffraction effects there as long as the basic balance is ok. Or what do you think?

I believe the frontal projection is reasonably free from disturbing diffractions and early reflections and this is what I have prioritised in this design. At least that is what I interpret in my measurements and what I hear as I listen.
  Reply With Quote
Old 10th March 2013, 09:06 AM   #15
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä
Sondek, I am just a hobbyist. Reflections can't be avoided and by looking at some discussions here at diyaudio, the topic is open and diffuse (sic!)

I am glad that you got the soundfield (-experience) that you wanted!

I have owned a pair of Sonab OA-13 for 15 years now. I know how they sound when placed in various location and environmets. They are now at my summer cabin that has all-wooden structures - the right environment!

In my new dipole project AINO GRADIENT the acoustical room space is different. My side walls are roughly 2m from the speakers. I thin that generally speaking, the room available gives outlines for speaker type and properties, specially sound dispersion patterns (directivity).
__________________
AES Associate Member / My DIY speaker history: -74 Philips 3-way, -82 Hifi 85B, -07 Zaph L18, XLS10+PR/Hypex, -08 CSS125FR, -08 Hifitalo AW-7, -08 TangBand FR, -09 MarkK ER18DXT, -13 PPSL470, -13 AINOgradient

Last edited by Juhazi; 10th March 2013 at 09:16 AM.
  Reply With Quote
Old 11th March 2013, 12:28 AM   #16
diyAudio Member
 
Join Date: Mar 2010
Location: Stockholm
Here are some measured results, to illustrate some of my design ambitions. I have found a limited down shelving like on the Linkwitz Orion, roughly -3 dB between 500-5000 Hz, produces the most natural spectral balance for my design as well.

Click the image to open in full size.
Frequency response gated 3.5 ms. Freefield on axis (red), intended position in room with floor bounce partly damped (blue) and freefield standing (green).

At 3.5 ms gate time the measurement includes the influence from the wall behind the speakers on the direct sound. As you see the response remains in character for the intended room placement, where the integrated damping material and dipolar dispersion pattern helps to reduce the effect of diffraction and early reflections on the direct sound. Green curve is standing position at normal listening distance, so you don’t notice any dramatic change of balance if you sit or stand up during listening.

The effect of dispersion pattern and damping material can be seen even clearer in the next picture, where I compare my design with a conventional reference under the same conditions – freefield versus intended room placement.

Click the image to open in full size.
Reference in red and my design in blue. No extra damping used in the room measurement.

The remaining wiggles in the room response are most likely from the residue of the floor bounce, which you can see in the above measurement.

Click the image to open in full size.
Freefield frequency response for the direct sound within the listening window, where 0 degrees represent on axis listening between the speakers in an equilateral triangle.

The ambition for the direct sound has been to have reasonably similar balance within the listening window between the speakers. The dipolar dispersion helps with the time intensity trading when you move within the listening window and the stereo image remains between the speakers even when you change positions. The biggest compromise is around the crossover frequency at 2 kHz. But I have a hard time to determine which consequence this has on the end result, since the time intensity trading is an approximate psychoacoustic effect for phantom sources already from the start and not a well defined and easily verified quality of any sound source.

A few examples on total sound energy curves in my listening room.

Click the image to open in full size.
Octave band curve at the listening position with 500 ms gate time.

Click the image to open in full size.
Long term average spectrum curve at the listening position.

Finally some time domain measurements.

Click the image to open in full size.
Burst decay free field time resolution gated 4 ms. Freedom of resonances and reflections in the main operating area

My ambition has been to minimize resonances and reflections in the design, by restricting the drivers to work in their linear range and suppress early reflections to not impact the interpretation of the direct sound. And the result can maybe be illustrated by a couple of decay measurements done in the intended position in the living room, compared with measurements on a conventional reference under the same circumstance.

I’m particualarly satisfied with the relative lack of strong early reflections from floor and the wall behind the speakers. The first strong reflex appears more than 8 ms after the direct sound. Earlier reflections are suppressed around 20 dB below the initial transient.

Click the image to open in full size.
Burst decay at the intended position in a living room, without any additional damping. Reflections are 20 dB below initial pulse.

Click the image to open in full size.
First strong reflection appears after more than 8 ms in my living room.

For the conventional speaker you can see strong early reflections already after a few milliseconds, with the risk to influence the perception of the direct sound.

Click the image to open in full size.
Burst decay at intended position in a listening room without additional damping. Strong early reflections might impact perception of the direct sound.

Click the image to open in full size.
Strong reflections continue after 10 ms.

Click the image to open in full size.Click the image to open in full size.
So in the end, after three and a half years, finished speakers in my living room.
  Reply With Quote
Old 11th March 2013, 01:39 PM   #17
diyAudio Member
 
Join Date: Mar 2010
Location: Stockholm
Juhazi,
Love your Aino gradient project! What a coincidence we post almost at the same time talking about dipolar/controlled directivity projects with references to Stig Carlsson and Salmi's Gradient work. Looking forward to follow your progress.

I believe my design would fit well into the acoustic properties of your own listening room. It seems just like the kind of space I would like my speakers to sound good in. I also have similar reverberation time curves in my listening room. Would have loved to try it out...

I will post comments to the Aino project in your own thread.

/Mats
  Reply With Quote
Old 16th March 2013, 08:40 AM   #18
diyAudio Member
 
Join Date: Mar 2010
Location: Stockholm
I’ve been reflecting a little bit lately, while enjoying music through my speakers. Here are some findings and lessons learned from this open baffle project on Stig Carlsson’s ortho acoustic design ideas.

I believe the basic design idea is definitely worth to continue pursuing, confirmed both in my listening and measurements. To angle the dipoles like Benjamin Bauer suggests in his 1960 AES paper broadens the area for good soundstage perception. Phantom images off axis are of course not as precise as in the dead center, but still gives a clear sense of soundstage width and the images never collapse into the closest speaker.

The angling also directs more of the radiated energy away from the front wall and towards the side walls. This helps to decrease the influence on the direct sound from early front wall reflections and instead focus more sound energy on the side walls for good illumination of the room by later reflections, assisting the sensation of apparent source width and envelopment in the reproduced sound. There is no problem to create a believable reproduction of the original event with conventional placement of the speakers close up against the front wall in my living room.

Another finding is how the open baffle concept gives a more balanced perception of bass than conventional speakers in my living room, where it is always easy to follow all the details in the register without disturbance from room modes and other resonances. My theory is that the dipolar absence of room gain results in a more natural reproduction of bass, more in line with what we experience in a live event which usually takes place in a larger venue and hence is free from audible room gain effects. There is also less boundary gain with normal conventional room placement, since the gain effect is naturally more evenly distributed over frequency due to the +/- controlled directivity dipole nature of the speakers.

I initially thought I would achieve the same boundary gain as for conventional boxed bass drivers in this project, but realized dipolar radiators act differently in that respect when I sat down and worked out the theory properly. Which led to the decision – the lack of bass sensitivity in my design requires active bass equalization for a proper overall balance in this system. If you aim for a fully passive open baffle system you need really large panels and loads of radiating surface, which ends up in a domestically unacceptable solution for my situation. For the time being I will stick with active solutions in my projects, which also has other advantages to be exploited and some drawback as well. You can’t have it all.

Two additional things in this project have given me new insights. First is how freedom from distortion reduces cues to sensation of loudness. These speakers sound very much the same all the way up in level until the power amplifiers start clipping or audible feedback in the turntable is heard through the system. It just doesn’t sound loud as you increase the volume. It’s not until someone starts to speak or another sound enters the room you realize the how loud the sound is. Sure, you have a level where the reproduction sounds most natural, but it never sounds loud. Almost like the speakers lack dynamics.

The other thing is how sensitive voicing mid and tweeter levels are to get just the right balance in the sound. Often the adjustments are fractions of a decibel between sounding natural or giving you a sensation of some kind of exaggeration. Changes often smaller than what I believe you find in normal sample variation between speaker units. You wonder how commercial speaker manufactures manages their QA.

Finally I want to mention what a great development tool the miniDSP solution is. So much plug’n play capabilities and it makes it dead easy to evaluate design changes with just a flick of a switch. But my analog active filter improves the resolution substantially. The miniDSP is great for the design process, but I will stick with active filter for the final implementation until I find a completely transparent DSP solution.

… and then some ideas possibly worth exploring in forthcoming designs:
  • How can I reduce midrange and tweeter bloom to improve dispersion uniformity even further? Possibly investigate alternative baffle structures and waveguide solutions.
  • The tilted baffle structure in this design results in reflections affecting the measured direct sound of the back radiation. What happens with a solution with a more uniform measured direct sound?
  • What happens with improved physical time alignment between the units?
  • How can woofer sensitivity be improved without scarifying other parameters for dipolar designs and still find its way into a domestically accepted solution? Current design requires more power for the lowest frequencies than I have available, to reach the intended maximum sound level. Should investigate alternative drivers, loading and amplifier technologies.
  • Further investigations into W-baffle calculation and equalization – cavity dimensions, loading and resonances.
  • And next project must be lighter and easier to build.
  Reply With Quote
Old 19th June 2013, 09:09 PM   #19
diyAudio Member
 
Join Date: Mar 2010
Location: Stockholm
I repost this post due to a broken link to one of the pictures above.

Here are some measured results, to illustrate some of my design ambitions. I have found a limited down shelving like on the Linkwitz Orion, roughly -3 dB between 500-5000 Hz, produces the most natural spectral balance for my design as well.

Click the image to open in full size.
Frequency response gated 3.5 ms. Freefield on axis (red), intended position in room with floor bounce partly damped (blue) and freefield standing (green).

At 3.5 ms gate time the measurement includes the influence from the wall behind the speakers on the direct sound. As you see the response remains in character for the intended room placement, where the integrated damping material and dipolar dispersion pattern helps to reduce the effect of diffraction and early reflections on the direct sound. Green curve is standing position at normal listening distance, so you don’t notice any dramatic change of balance if you sit or stand up during listening.

The effect of dispersion pattern and damping material can be seen even clearer in the next picture, where I compare my design with a conventional reference under the same conditions – freefield versus intended room placement.

Click the image to open in full size.
Reference in red and my design in blue. No extra damping used in the room measurement.

The remaining wiggles in the room response which you can see in the above measurement are most likely from the residue of the floor bounce reflection.

Click the image to open in full size.
Freefield frequency response for the direct sound within the listening window, where 0 degrees represent on axis listening between the speakers in an equilateral triangle.

The ambition for the direct sound has been to have reasonably similar balance within the listening window between the speakers. The dipolar dispersion helps with the time intensity trading when you move within the listening window and the stereo image remains between the speakers even when you change positions. The biggest compromise is around the crossover frequency at 2 kHz. But I have a hard time to determine which consequence this has on the end result, since the time intensity trading is an approximate psychoacoustic effect for phantom sources already from the start and not a well defined and easily verified quality of any sound source.

A few examples on total sound energy curves in my listening room.

Click the image to open in full size.
Octave band curve at the listening position with 500 ms gate time.

Click the image to open in full size.
Long term average spectrum curve at the listening position.

Finally some time domain measurements.

Click the image to open in full size.
Burst decay free field time resolution gated 4 ms. Freedom of resonances and reflections in the main operating area

My ambition has been to minimize resonances and reflections in the design, by restricting the drivers to work in their linear range and suppress early reflections to not impact the interpretation of the direct sound. And the result can maybe be illustrated by a couple of decay measurements done in the intended position in the living room, compared with measurements on a conventional reference under the same circumstance.

I’m particualarly satisfied with the relative lack of strong early reflections from floor and the wall behind the speakers. The first strong reflex appears more than 8 ms after the direct sound. Earlier reflections are suppressed around 20 dB below the initial transient.

Click the image to open in full size.
Burst decay at the intended position in a living room, without any additional damping. Reflections are 20 dB below initial pulse.

Click the image to open in full size.
First strong reflection appears after more than 8 ms in my living room as placed in the picture below.

For the conventional speaker you can see strong early reflections already after a few milliseconds, with the risk to influence the perception of the direct sound.

Click the image to open in full size.
Burst decay at intended position in a listening room without additional damping. Strong early reflections might impact perception of the direct sound.

Click the image to open in full size.
Strong reflections continue after 10 ms.

Click the image to open in full size.Click the image to open in full size.
So in the end, after three and a half years, finished speakers in my living room.
  Reply With Quote
Old 19th June 2013, 09:21 PM   #20
Banned
 
Join Date: May 2008
Location: Switzerland
Quote:
Originally Posted by sondek12 View Post
To angle the dipoles like Benjamin Bauer suggests in his 1960 AES paper broadens the area for good soundstage perception. Phantom images off axis are of course not as precise as in the dead center, but still gives a clear sense of soundstage width and the images never collapse into the closest speaker.
There's a paper by Kates which gives a very precise description how the optimum radiation pattern utilizing "trading" should look like: Kates, "Optimum Loudspeaker Directional Patterns", JOURNAL OF THEAUDIO ENGINEERING SOCIETY, 1980 NOVEMBER, VOLUME 28, NUMBER 11
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
open baffle project sqrivi Full Range 56 26th May 2012 06:43 PM
FS: Acoustic Elegance 12 Dipole one of the best open baffle mid bass drivers hum4god Swap Meet 5 24th March 2012 01:52 AM
Acoustic Elegance TD15LO Dipole Best Driver for open baffle NEW!! hum4god Swap Meet 3 22nd September 2011 01:57 AM
FS: Acoustic Elegance 12 Dipole one of the best open baffle mid bass drivers hum4god Swap Meet 0 10th September 2011 03:43 AM
Vented / Open baffle hybrid design and ideas Spike415 Multi-Way 2 23rd September 2010 05:53 PM


New To Site? Need Help?

All times are GMT. The time now is 05:23 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2