Soliciting ideas for an ESL Soundbar

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I'm wanting to help out a student from Savannah College of Art & Design who approached me recently for advice on building an ESL Soundbar. I'm not exactly the sharpest knife in the drawer here to I'm hoping some of you will jump in with some ideas, please :)

My new friend had drawn up a very slick looking hybrid design using a 3 inch high x 40-something inches wide panel with a 3" mini-woofer at each end supplying the mid bass. And he prefers to make it a stand alone stereo soundbar, complimented by a separate subwoofer.

Unfortunately, I think such a thin-panel dipole would suffer terribly from phase cancellation, aside from its having non-existent stereo imaging.

I first advised that any semblance of stereo effect from such a soundbar would require two separate panels, and that the mini-woofers would be best positioned in the center to allow at least a small amount of separation between the panels.

Next I commented that the dipole roll off may be so limiting that perhaps a monopole would be better. And this could be accomplished by enclosing the backside of the panel and absorbing the backwaves--- or by forming a J-section transmission line behind the panel, such that the curve of the "J" would be directly behind the panel, and the long section of the "J" would extend upward, behind the flat screen TV. The downside is it would not allow the TV to sit flush against the wall.

Personally, I perfer the mono-pole and "J-section" TML; which I think would rebound less energy to the diaphragm than would a sealed enclosure. My new friend prefers the sealed box option, if he concludes that a dipole of those dimensions would be impractical.

Another option would be to abandon the stand-alone stereo idea and just make the soundbar the center channel in a larger system.

OK that's where it stands.... now let's hear your ideas!

thanks,
Jazzman
 
Last edited:
Hi,

think basically of my small esl turned 90°, a curved meatl sheet panel partnered with a bass tower.
The curvature of the panel would have to change too, but ending up still curved horizontally, and still beeing straight vertically.
Segmentation or curvature in the vertical will not be needed as the height of the ears of the listening audience is quite fixed.
Horizontal dispersion will be required though.
The bass tower residing underneath the panel, picking up the curvature of the panel.
Instead of a dipole bass, I´d choose a CB-array, probabely with 5" or even smaller drivers, depending on the panels dynamic range.
One needed to think about either a topboard above the panel as stand for the TV, or a fixation mechanism for it.
Without a top board the panel could still work under ´open´ condiditions.
I´d prefer a topboard though as a kind of open box, which could hold some damping material.

jauu
Calvin

As I find the sound quality of any video quite unnatural and as the optical impression overwhelms by far, the acoustic advantages of ESLs are imho simply not required.
I´d rather go fully dynamic for a centre speaker, leaving only the Front R- and L channels as ESLs.

jauu
Calvin
 
Last edited:
Hi, Charlie!!

Have you showed him my Little panel designs?

Does he plan on making it curve or flat.

My 3.25" x 9.75" panel competes very well with my 8" sub.
I would suggest that he uses no less than a 5.25" with it or even a 6.5" drive would be better.

The natural crossover point using my 5.25' woofer is about 600hz to 700hz and it reaches some very large excursions at less than 40HZ.

But the two drivers sound as one when I listen to heavy drums such as those from Montrose and Foghat on the extreme end of the scale.

And very whole in unison on softer stuff like ELP's "In the Beginning", Just incredible!!
Kansas, Boston and The Spin Doctors would just knock your socks off, and the answer to your next question Charlie would be, "Yes, it would" !!! He,he,he,he ;)

I found that the woofer (5.25" Sony) matched up perfectly as a sealed box rather than its original poorly tuned ported configuration.
This produced a clean tight bass with no muddyness.

He may have to employ a bit of equalization as well such as a variable linkwitz transform for the woofer or something to keep the heavy excursions under control while boosting the bass up a bit.

(I am getting ready to embark on this stuff for my system now as I want to get the amps crossovers and interface all mounted in case before I fire up my latest build so that it will be a complete unit and have HV and wires strewn all over my bench.
And for once call the project done and sit back and enjoy it for a change!!
I will keep every one updated on this as I go.)

Such that I would consider two of them per channel just to keep the package small.
Choose a driver that has the Highest Xmax you can find for the size he chooses.

The little Daytons are quite impressive in this matter but they do come with a cost.
The highest SPL that I could get to with my setup and still sound good due to the 5.25' woofers limitation was about 95db.
The woofer were right to their most extreme excursions of nearly 1/4" to 3/8" peak to peak as measured with a ruler the best that I could.

But they didn't sound bad at all pumping that hard and was not an issue at normal levels of 85db to 90db.
I was surprised that they could do that at all!!

I am working with what I have on hand, but I would consider those little daytons in a second from the videos I have of them in action!!

Have looked into building a mini TL but the length dictates the minimum dimensions for good performance and is no different to using a small driver or a bigger one.

I have though about using a aperiodic enclosure design is this would give the characteristics of being a sealed box but yet the vent will allow the woofer to breathe on heavy excursions.

This may give the effect to help highly dampen the cone movement as well, without restricting it like a sealed box would do.

My lager cabinets are setup in a similar way and the sound turned out very nice.

Kind like a openbox stuffed hard full of insulation or other dampening material.

This is something that I am going to explore once I get going as getting a decent bass out of a small system is a bit of a challenge!

As you know what the consequences are of putting the ESL in a box, I would advise not to do this and don't try to dampen the back wave.
I have tried this and it just destroys the high end detail and finesse that ESL's are noted for.

However, What you can do if it must be in an enclosure is to leave it open backed and line the 4 inside walls with some insulation of some sort.

Make the enclosure bigger than than the panels so that the tunnel is not smaller than the diaphragm size including the thickness of the dampening material.
This will help to enhance the low end (Dipole action F/B is greater) and the dampening on the inner walls will help to reduce any resonance's that would occur in between the inner walls of the box.

I have made a poor example mock up of this back in 2010 but I never took any detailed pictures of it as I was more focused on driving issues at the time, But it did seem to work good.

With 3" high panel and another 1.5"x 2 for the insulation would work out perfectly for a 5.25" or 6.5" matching woofer(s).

Another configuration that I have been debating ( just for some more idea's), is to build a 18" long version of my current 3.25" wide TiG rod design and mount it in the middle of a pair of top and bottom dipole 8" subs that I have.

I have also considered this design with just one sealed bottom 6.5" woofer (or two Top/Bottom sealed using a 12" long panel) for use as standing next to a monitor on either sides.

I hope this helps!!

jer :)

Sorry, I can't post any pictures at this time as Stupid Windows Crashed on me last night and I almost lost everything.
I somehow I got lucky and was able to recover everything in the My Documents folder of my most recent work, even though it has disappeared and cannot be accessed even if it is just a slave drive to a new fresh systems drive!!

Stupid Virus I Think!!!
Probably one of the Cryptolocker's types even though I had no warning, the machine just BSODed and that was all she wrote.
It took 45 minutes for it to boot and about 5 to 10 minutes or more to execute every single mouse stroke and I was able to get the My Doc folder to show up and I copied it to my fresh drive right away.

The copy went fast at normal speed and without a hitch, But any function using windows explorer was extremely slow and the MY Doc's folder disappeared again after that!!

Thats the First time I have seen that one!!

Sorry to be OT!
But beware of this just in case it happens to you or anyone else and not to panic and be patient as you may get lucky, as I did, and be able to retrieve your data before it is gone...Permanently !!!!!!! ;)
 
Thanks all for your comments so far.

I've asked my friend to join the forum and post a concept drawing of his ESL soundbar. I think the initial problems are meeting the size constraints and mitigating the phase cancellation.

As I figure it, the wavelength corresponding to a 3" panel falls around 4,500 Hz. And if the phase cancellation becomes significant around lambda/2 and is 3db down at lambda/4, then the crossover frequency for the naked 3" panel would need to be way up around 1,100 Hz--- not good. (did I get that right?)

Thus, I figure the panel would need to be either a monopole via an enclosure or a dipole mounted on a folded baffle (H-frame or U-frame).

I'm interested in hearing your thoughts on dipole versus monopole and thoughts on enclosures or baffles. BTW, my thoughts on a short J-section TML was for the panel, not necessarily the woofer. And a short TML could perhaps be considered a U-baffle.

Again, keep those ideas coming :up:


BTW, Ger
I support capital punishment for anyone putting viruses on the internet
 
Last edited:
I agree with you Charlie!!!!! ;)


Yes, Your calculations for the 3" wide panel is about right.
I miss quoted when I said it was about 600Hz.

Here is my as measured chart using only the 80HZ and 12Khz shelving EQ's on my mixer to make my system flat and it clearly shows that there natural crossover point for the open panel to be about 1100Hz.

I did however use the sweepable mid to cut the highs out of the woofer but the panel only had the everything below 80 cut only for the amps sake and about -1db to -2db cut at 12khz to flatten it out on the high end.

Even without the 12Khz roll off that I used the difference was about the same and is only something that you can hear.

If I didn't do that then my ears would start to fatigue from the high end bite (NOoooo, I don't listen to music very loud at all........Very Much!! ). ;)

I would use a U-frame for the esl.
Even a 4.5" deep frame would make a big difference and bring your crossover closer to 400hz to 500hz range.

Had I setup my room for measuring this would be extremely flat.
But this was measured as is in front of me on my desk.

jer :)

P.S. Even just a 1/2" to 1" batting on the inside would work well if size is a very stringent factor as long as it does not block the path way or obstruct the paht way toward the opening in any way.
I have been thinking about applying this to the Desktop Version as well to try.
 

Attachments

  • todays test p mic.jpg
    todays test p mic.jpg
    249.2 KB · Views: 180
  • Desktop ESL.jpg
    Desktop ESL.jpg
    74.7 KB · Views: 181
Last edited:
These later designs attempt to address the issue. The circular designs would have a sound damping material to absorb the rear waves. The open designs would direct the waves up or down so they didn't interfere with the forward waves.

Thanks so much for all the help so far!

https://drive.google.com/file/d/0B-XcYUA-nEpmN3RlR2JuWjc1dFU/edit?usp=sharing
https://drive.google.com/file/d/0B-XcYUA-nEpmS2Y2OExzSmdxeGc/edit?usp=sharing
https://drive.google.com/file/d/0B-XcYUA-nEpmUUdFQWNVVU9jcGM/edit?usp=sharing
 
Last edited:
I tried absorbing the rear waves and I was not happy with the results.
You can try it, as it may suit your needs.
And you may find as I did that it is not necessary and would be a waste of material and work, to try and stop the backwave.
All that is required is to increase the path length from the front to back somehow.

jer :)
 
Sorry one more attempt to upload the imaged embedded as instead of [url] from my Google Drive account.
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmN3RlR2JuWjc1dFU/edit?usp=sharing[/IMGDEAD]
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmRDN3WkdJYTZ1MlU/edit?usp=sharing[/IMGDEAD]
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmS2Y2OExzSmdxeGc/edit?usp=sharing[/IMGDEAD]
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmU19nNS1zNnRodWs/edit?usp=sharing[/IMGDEAD]
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmUUdFQWNVVU9jcGM/edit?usp=sharing[/IMGDEAD]
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmaTRXV0ZRNC1paVU/edit?usp=sharing[/IMGDEAD]
[IMGDEAD]https://drive.google.com/file/d/0B-XcYUA-nEpmcmFoaHVxZlNOdVk/edit?usp=sharing[/IMGDEAD]
 
Hi,

The short answer is don't even think about ESL's for such
an application, dipole roll-off is a major issue and trying
to enclose the rear wave of an ESL simply does not work.

Nevermind designing and building ESL panels is no trivial matter.

At a push each end of the bar could have an open backed ESL
tweeter, but in the larger scheme of things a very complicated
and expensive option compared to budget decent tweeters.

FWIW speakers are nearly always a very bad idea for an
Art and Design course unless you also happen to be very
good at engineering, as most of the design work is pure
engineering, balancing the myriad engineering design issues.

Even your typical mechanical or electrical engineering type
can make a complete pigs ear of designing decent speakers,
due to a lack of understanding of acoustics, and quite simply
the level of and type of analysis required is very inappropriate
for an Art and Design course. YMMV, if so good luck.

rgds, sreten.
 
Last edited:
Hey guys,

First of all sorry that I have been so unsuccessful at uploading the images here is a link that hopefully will work:

ESL SoundBar project by Jesse Summers at Coroflot.com

Sreten your absolutely right, I have little to no knowledge of loudspeaker engineering (that's why I'm soliciting help!) Even if the prototype we build ends up not sounding up to par I think it will be worth my time just as a learning experience. I like the idea of using ESL tweeters instead of panels. Also we are thinking that having a separate wireless woofer might be the better option as opposed to integrating mid-bass woofers into the unit.
 
Sreten your absolutely right, I have little to no knowledge of loudspeaker
engineering (that's why I'm soliciting help!) Even if the prototype we build
ends up not sounding up to par I think it will be worth my time just as a
learning experience.

Hi,

Asking for help is a good idea, but realistically you don't understand
the fundamental issues your dealing with so your ideas are simply
fanciful imaginings that are way off the mark of basic reality.

FWIW I don't subscribe to the "you don't know it won't work
unless you build it" mentality. Its inane. I think your also
hopelessly out of you depth in trying to build a ESL panel,
nevermind use it properly, might be wrong, but I doubt it.

rgds, sreten.

However much it might appeal to your imagination if you
don't know much about speakers on a Art & Design
course you will only learn that you don't know enough.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.