A Study of DMLs as a Full Range Speaker

I recently found 2 aluminum honey comb panels on the street in Manhattan.


They additionally appear to have an internal frame around the edge.
Light weight. Very rigid.

Would these work?
Recommended exciter?

They would make some very cool monolithic DML's

They just might! Yes, they are pretty darn heavy but with that construction they are also really stiff, I'm sure. The closest thing to that I've ever tried is Alumalite, which is aluminum skins on a corrugated polypropylene core.
It was actually among my favorites of the 20 or so panel materials I tried. As I recall it was a bit lacking on the low frequency end, but otherwise quite good. Not nearly as loud as PS foam, but very clear, clean sound. I'm eager to hear what you think when you get them singing!
Eric
 

Attachments

  • Alumalite.jpg
    Alumalite.jpg
    73.5 KB · Views: 458
Stimper,
You absolutely can. I've done exactly that to make some Balsa core prototype panels with CF and FG skins. I have not tried it with the nomex but I see no reason it would not work.
Using the clothes bags is super simple and easy. Previously I was using FoodSaver bags and they work great too, but they are limited in size. You can get the clothes storage bags in much larger sizes. As far as a vacuum pump, when I tried the "Hoover" it pulled about 6 in Hg, which is a pretty weak vacuum, but probably enough. What I do now is use my Foodsaver pump to pull the vacuum on the clothes storage bag and it pulls about 21 in Hg.
Here are some pics of my CF/balsa/CF prototype, made using the storage bag method.
Eric

Thanks, I will try this method with just applying adhesive to the honeycomb first to make up one more CF panel with the leftover CF and buy one more piece of nomex honeycomb. I’m tempted to try aluminium honeycomb, but thinking about making some lxmini +2 for my living room and use the CF panels as monitors for my digital piano. This was what put me on my original DML quest, as the piano doesn’t sound that great using Mackie 824 mk2’s. If they sound good on the piano, I’ll sell the mackie monitors to buy a sub for the panels.
 
Eric --

Thanks for the thoughts. The synergies of this discussion thread and the panels leaning against my wall have me thrilled in a way that I first had when I started building my first Nelson Pass A/40 amps from the Audio Amature article back in the 1980s or so ;-)

Some thoughts

With panels 70 3/4" high x 20 1/4" wide I've got 13.88 square feet of surface area.
I conjecture that this should push a lot of air and give me pretty extended low frequency response.

The rigidity vs weight should give me speed of response to get the higher frequencies.

Progress

I've already started taking apart my main system to start modifying it to hold the aluminum panels.
It uses 2" x "2 aluminum square tube uprights that have 1/4" think walls and liking composite materials, I forced 1 1/2" poplar into the tubes. Super rigid. (see attached)

Still working out the suspension/mounting of the panels.


Exciters on their way!
I think I have everything else I need.

More thoughts

I've started looking for a model that would enable me to input factors like panel weight, rigidity, size, dimensions, etc.
I just need the relationships, then I can program the model in Mathematica and maybe make it available to the community.

Any suggestions?

Also, I suspect that rigidity or maybe more accurately tensile strength relative to weight might stand as the most import factor.

This has me wondering if a solid panel of aluminum or even titanium (not nearly as expensive anymore) of just sufficient thickness to recognize as a panel/sheet/plate rather than "foil" might perform very well in a DML.

All the honeycomb cores just separate the skins where the tensile strength resides. The thicker the cores, the more rigid. Just a continuous beam.

But doesn't something get lost with such cores? If the cores differ from the skins won't they resonate differently? Some of the cores may hardly resonate at all.

Does the excitement travel through the panels or does it travel across the skin (faces) of the panels?

You get where these musing have taken me.

Would a solid piece of aluminum or titanium work even better?


What about the edges?

Seems like whatever goes on at the edges and corners of panels scan degrade sound quality.
Some commercial DMLs laser cut into the edges of the panels.
Other round off the corners of panels.

Can with think of the issue as turbulence messing around with the rest of the generated wave(s)?

Airplane wings will often have vortex energy diffusers (see attached).
I wonder if one could stick on something similar on the edges of DML panels to diffuse or control the edge issues?

Thoughts?
 

Attachments

  • Screen Shot 2021-02-17 at 3.23.30 PM.jpg
    Screen Shot 2021-02-17 at 3.23.30 PM.jpg
    139.4 KB · Views: 561
  • Screen Shot 2021-02-17 at 3.47.47 PM.png
    Screen Shot 2021-02-17 at 3.47.47 PM.png
    199.9 KB · Views: 471
Those are vortex generators, they do not diffuse but excite the laminar (if attached) flow at low airspeed or high alpha (angle of attack), giving the wing panel or control surface energized flow.

A strip of foam around the edge of the baffle could diffuse......hard to say if it would actually be audible. It would be fun to use weather stripping foam and keep stacking it to see if it helped......
 
More thoughts

I've started looking for a model that would enable me to input factors like panel weight, rigidity, size, dimensions, etc.
I just need the relationships, then I can program the model in Mathematica and maybe make it available to the community.

Any suggestions?

Concerning modeling, I've done a lot of searching for the factors that determine DML performance. One of the best sources IMO are the patents by Heron like this one:

https://patentimages.storage.googleapis.com/dc/fb/7b/46e5a02efc68f3/US6058196.pdf

According to this and other authors, the ratio:

(bending stiffness per unit width)/(weight per unit area)^3

determines radiation efficiency. And the ratio:

(bending stiffness per unit width)/(weight per unit area)

is one of the factors that determines the low frequency limit of the panel.

In both cases, higher stiffness and low weight are favorable.

Note, it's bending stiffness and not tensile strength that determines performance.

Because solid panels of any material (including titanium and aluminum) are quite heavy compared to composites with foam or honeycomb cores, the latter are preferable. That said, I think plywood actually performs pretty decently, despite being solid. The same patent also includes equations for the relationships between panel size and minimum frequency, and panel thickness and maximum frequency.

Regarding cores: The main concern with lightweight cores is not that they might resonate differently than the skins. Rather, it's that if their shear stiffness is too low, the composite may not end up with the stiffness that simple engineering estimates might suggest.

Regarding excitement: What happens (in theory) is that excitation causes the panel to bend as a unit (that is, the back face of the panel bending exactly the same and simultaneous with the face), and that these bending waves on the faces generate sound waves in the air. There's a pretty cool animation of it here (showing just one side of the plate).

Evanescent Waves (sound radiation from a plate)

In fact, this animation is really showing the importance of the factor (bending stiffness per unit width)/(weight per unit area), because that ratio determines the panel's "coincidence frequency" and hence effectively determines (in part) the frequency below which the generated sound waves become "evanescent", that is, decay with distance.

Hope this helps,

Eric
 
I may not be able to get to test/build/experiment till mid-late summer. Sharing my brain dump, hoping that it may be helpful. Maybe flesh out more ideas via your collective wisdom and experiences. At least notes for me to look back at.

For the EPS/XPS panel designs:
High-density EPS/XPS "Phase Plug" on the front panel with the center mirrored with the driver in the back. Can this reduce the HF brightness EPS/XPS panel can have? Base radius of the plug being 1x-1.5x, of the exciter ring attached on the back. Extending the thought vary thickness on the face of the panel; highest thickness around the part mirroring the exciter and thinnest along the boundary of the panel. Spitball guesstimate 2"-0.5" tapered as a curve; carved on a CNC. Maybe 2" to 0.5" and closer to the edges, an inch or so close, go back up to 1" or 1.5"; would this act as a reflection absorber/dampener?

Looking at specs
- XPS FOAMULAR®/FOAMULAR® NGX™ 600,ASTM C578 Classification Type VII
- EPS ASTM C578 Classification Type XV
Would these grades would allow one to use thinner panels ie using higher density panel but thinner so lighter? I guess finding the right balance is the key.

For composite/sandwich core construction designs:
Would extending the core material (nomex/rohacell/EPS/XPS/Alu or paper Honeycomb) beyond the bounds of the composite so as to allow to be mounted to a frame, improve FR. (Some stabilizing intermediary maybe needed for some type of cores.) Idea being the panel is not directly damped at the frame rather a light dampening at the end where the composite terminates and slowly dampens waves that propagates to the frame. Will old school cone speaker doping liquids work here?

Correct me if wrong, ideal panel needs to be lightweight, with high compression strength and flexural strength. And hence composite panel designs, to harness the different material properties. Can say a material like Dragonplate paper-thin (.010") Carbon fiber veneers sandwiched with polypropylene/nylon mesh (like the kind used in water filter cartridges/window screen/Chiffon fabric) with an ideal glue (possibly using intermediary technique illustrated by the double top guitar maker few posts back) reduce the panel thickness and result a panel with ideal material properties. Possibly thinner than nomex composites?

Idea I would like to explore : {someday}
An Organic composite. Simple_Weave/Mesh Hemp fabric sandwiched between thin hardwood veneers (how thin do they make it?, do they come in large sheets?) glued double top guitar style. Designed as mentioned earlier with the core extending beyond the panel. Maybe some Permatex Super High Tack Gasket Sealant on the extended parts for some damping. Mounted on a frame (crude illustration attached). Would this composite preserve the wood tonality (all materials being wood fiber based)?

Would hardening the veneer improve anything? Possible options:
- Minwax® Wood Hardener [resin reinforcement]
- PVA adhesive modified with sodium silicate pdf
- Rapid petrification
- Other Commercial Products like

If hardening like above works reasonably well, can we harden/fortify the faces of those Kraft pallet boards (honeycomb core).

Modeling panel sounds like a swell idea. [although I argue with my lizard brain, how can I get the right parameters of a diy composite, or how can i scale 'em, without actually testing the diy-composite's specific parameters]

Would this be of any help aagas
https://ocw.mit.edu/courses/electri...s-spring-2009/readings/MIT6_013S09_chap13.pdf
 

Attachments

  • sandwich.png
    sandwich.png
    62.7 KB · Views: 378
Hi Eric -- Very helpful indeed. Thanks.

Also, I came across this desktop sized DML project on a woodworking site: Wooden Speakers - by TheFritz @ LumberJocks.com ~ woodworking community

See attached photos.

This project hangs the panels from the exciters. No other suspension of the panel required.

Likely unworkable for anything much larger, but someone might take the idea somewhere.
 

Attachments

  • Screen Shot 2021-02-19 at 11.43.59 PM.png
    Screen Shot 2021-02-19 at 11.43.59 PM.png
    527.1 KB · Views: 404
Quarter.
Easy Tricks Using Fiberglass Strengthening! - YouTube

you might find 5minutes into this video very interesting?

If you are referring to the harsh sound in the 10k ish area on say eps.
Uploaded Pictures
These show some measurements of a 40cm vh grade 1cm eps panel before and after modification.
In the bottom ,room photo with the small panel on the right, you can just about see the concave area in front of the exciter.
In the centre of the coil area you might be able to see a yellow dot?
This is a small hole with a very thin slither of light foam ,to brake up harsh distortions in the coil area,which can and does cause frequency spikes and dips.
I also used this method on my card panels,but now I use a dome in the centre of the card,so this is not necessary anymore,the dome also helps reduce cavity noise within the coil,magnet and panel area,which I guess the hole and foam plug idea reduces as well ?

As for rigidity,this would depend on panel size,(that is what I believe anyway) the smaller the size the less rigid it needs to be.
A large 6ft ply panel can sound very good with plenty of low end ,but reducing the size to say,12inches or so ,it can start to sound glassy hard and unpleasant.
In my opinion obviously.
Hope this information is of help.
Steve.
 
Last edited:
First speaker up and running.
Photos attached.

My wife, an architect, just unleashed all sorts of vitriol at me throughout the entire assembly process.

A combination of a year sheltering in place due to COVID and she had decided that this weekend she had had enough of me.

Can't necessarily fault her judgement ;-)

The first thing I played, U2s, With or without you.
I hadn't realized I had selected the most appropriate possible recording.

I continued with a range of other vocal, classical, and jazz pieces.

My wife walks into the living room, looking at the panel, looking behind the panel, looking at me. Utterly confused. Completely astonished.

To quote her, "I don't have any idea what you've done, but this is f____ing amazing!"

How does it sound?

To repurpose the most eloquent of descriptions from earlier in this very long thread, ...

WOW!

Volume
Highs
Midrange
Bass
Dynamics
Dynamic range
Air
Bloom

It seems to have them all.

Most importantly, the single channel playing, the music engages you.

Single exciter. I just stuck it on the center of the panel.

Frame cobbled together for parts at hand.

I've driven the panel with a single amp of a Hypex FA123.
I'll likely bridge the amp when I get around to it.

No crossover over.
No equalization.

Not certain what equalization could even do with these panels, but the Hypex software might give ways to optimize the output.

Thoughts?

Onto the right channel!
____

Total out of pocket project cost at this point about $46.
 

Attachments

  • Screen Shot 2021-02-20 at 5.21.58 PM.png
    Screen Shot 2021-02-20 at 5.21.58 PM.png
    839.8 KB · Views: 510
Steve,

After re-watching the TI video, I have to say, this could very well have inspired my thoughts on exploring different composite options.

I agree, varies with material and thickness. Scouring through various forums over the years, my observation, if my memory serves me right, is brightness/sharpness/distortion/{choice of audiophile lingo}, seems to be in the 3k-5k range for thicker panels around 10k+/- on thinner panels. I do recall having seen that panel with concave design mirroring the exciter, but did not know/recall/notice the foam dot until you mentioned it. However you got to that design choice, I admire the ingenious simplicity. In the 10k range the wavelength is in the range of 1.3", so is the concave region changing the radiation pattern to a focal point, creating some destructive interference reducing the brightness/ sharpness/ distortion. Extending that logic, thicker panels with ~4k brightness would need a bigger concave area around the exciter. My thought on using the "phase-plug" concept i.e. a solid convex face in front of the exciter is, rather than having a destructive interference, create a dispersion, widen the beaming; could be more pleasant audibly? Conceptually akin to your dome card panel. Is that your experience? Is the dome solid or hollow?

Another reason for my thoughts on convex construction on the EPS/XPS panel (ie carving out as a one) is the idea of diffraction. As I understand it, sound just like light, diffracts as it changes medium. That could explain the yellow sponge in your design. Depending on the frequency could be absorbing (possibly internal reflection within the foam) or changing the angle of radiation due to diffraction different than that of the panel. Resulting in overall smoother response? Also recall a design with 3d printed "Whizzer Cone", to improve freq. response.

Diffraction in different medium, is also the reason in thinking along the lines of thinner composite and possibly similar type of materials wood, plant fiber based. I am inclined to believe too much flexural stiffness on smaller panel can contribute to the glassy / ringing, dampening might reduce some of it but without some ease in flex the bandwidth of FR will be limited. I recall smaller carbon fiber sheet panels, acrylic panels, glass panels etc perform as you describe. As the size increases there is room for flex inherent in the material, improving the freq bandwidth. Too much flex and lower freq responses would get floppy and go incongruent to the rest of the waves in time. Imagine an exciter attached to a sheet of paper. I suspect similar glassy/ringing on smaller hard epoxy based composites and smaller metal and metal based composites as well. Ideal flex would be the balance of the maximum bandwidth that can be usable for a given size and material, with dispersion and dampening used to refine the panel.

Absolutely Steve, thank you very much for help.

I know, what I do not know, far supersedes what think I know, and there are times when I AdamSavage it; please do not let that stop you from correcting my errs.


.
 
First speaker up and running.
Photos attached.

My wife, an architect, just unleashed all sorts of vitriol at me throughout the entire assembly process.

A combination of a year sheltering in place due to COVID and she had decided that this weekend she had had enough of me.

Can't necessarily fault her judgement ;-)

The first thing I played, U2s, With or without you.
I hadn't realized I had selected the most appropriate possible recording.

I continued with a range of other vocal, classical, and jazz pieces.

My wife walks into the living room, looking at the panel, looking behind the panel, looking at me. Utterly confused. Completely astonished.

To quote her, "I don't have any idea what you've done, but this is f____ing amazing!"

How does it sound?

To repurpose the most eloquent of descriptions from earlier in this very long thread, ...

WOW!

Volume
Highs
Midrange
Bass
Dynamics
Dynamic range
Air
Bloom

It seems to have them all.

Most importantly, the single channel playing, the music engages you.

Single exciter. I just stuck it on the center of the panel.

Frame cobbled together for parts at hand.

I've driven the panel with a single amp of a Hypex FA123.
I'll likely bridge the amp when I get around to it.

No crossover over.
No equalization.

Not certain what equalization could even do with these panels, but the Hypex software might give ways to optimize the output.

Thoughts?

Onto the right channel!
____

Total out of pocket project cost at this point about $46.

I found eq to make the bass richer. I used REW and a Minidsp 2x4 HD.
 
I've got both channels operating now.

Astonishing.

The capacity of this technology to reproduce music humbles me.

I finished setting these things up 45 minutes ago. I've done nothing to optimize them. I've done nothing to move them around in the room.

Astonishing.

They do everything I mentioned earlier and so much more.

The sound stage completely changes with each recording.

I have never had an experience of recorded music in my home that so effectively recreates the venues of live recordings from intimate jazz clubs, to house concerts, to orchestral halls, to stadiums.

Astonishing.

I want to listen to lots more, but at this point these things have handled anything from delicate jazz vocals to Japanese drumming to orchestral works to rock-n-roll.

Listening to a live recording of Chick Corea (sad that we lost him last week) and John McLaughlin at the Blue Note in NY 2016 - a performance that I attended - these speakers so immersed me in the room, that I turned my head to my right when I heard the person next to me pull out their chair. The recording had and these DMLs reproduced all the auditory information that had me physically respond in the same way I would have at the table, center stage, in the club.

Astonishing.

Maybe tomorrow, I'll give some thought to why I think these simple panels to what they do, but tonight all I've got left is...

WOW!
 
A product ideally suited to satisfy these weight, composition, and rigidity requirements is sold under the trademark KLEGECELL #33, by the American Klegecell Company. KLEGECELL #33 is a substantially rigid, polyvinylchloride material which is lightweight (2 pounds per cubic foot), non-conductive, and acoustically impermeable

And that is lightweight?
With no mention of response?
Another method of making something very simple into something very complicated. :confused:
Alas, I'm afraid I have found very few patents that have actually helped me with the design of dml or should I say good sounding dml ,
Steve.