A Study of DMLs as a Full Range Speaker

For anybody that is interested in adding a non-permanent material, to function as damping / dissipater of unwanted energies being transferred from a Exciter of anywhere else ot is manifesting on a Panel.
Newplast Modellers Putty is a non-permanent substance, it has been measured for its performance and has Damping/Dissipation properties that measure very close to some of the most attractive materials used in audio for these purposes.
It has also been used coupled to materials that have useful structural properties, but fall off in the desirable measurements, the addition of Newplast to the material has substantially moved its measurements towards the most attractive ones recorded.
To give an idea, as a non-permanent material that could be considered, such as a Plasticene, this materials damping / dissipation is approx' 60% less effective than Newplast.
 
"(...) Unfortunately, I am very new to the whole subject. What do you recommend as exciter material? I have often read that either EPS or a canvas covered with plywood is highly recommended.

Would you recommend anything else?"

Although i´m experimenting several materials as well combinations in the same honeycomb structure for more powerfull usage, i wanted trasparency to my listening room.
So, not expecting much sound wise, nevertheless because of material ressonance frequencies i did put a polyethylene (syntetic glass) 1000X500x4mm with DAEX25FHE4 exciters because of their top end roll off. And... the "joke" turns out serious.
It`s playing every day with soft eq and xo at 90Hz in a room with 30 m2, 4m apart from each other and hangigng on the ceiling with fishing string.

Just try.
 
  • Like
Reactions: 1 user
So, not expecting much sound wise, nevertheless because of material ressonance frequencies i did put a polyethylene (syntetic glass) 1000X500x4mm with DAEX25FHE4 exciters because of their top end roll off. And... the "joke" turns out serious.
RMAM,
I'm wondering if the material you are talking about is PMMA, rather than polyethylene. PMMA is also known as acrylic or plexiglass. Among the materials I've tested, PMMA seems to have very good internal damping (as your post suggests), based on the impedance results in the link below.
Eric

https://www.diyaudio.com/community/...s-as-a-full-range-speaker.272576/post-7499940
 
Although i´m experimenting several materials as well combinations in the same honeycomb structure for more powerfull usage, i wanted trasparency to my listening room.
So, not expecting much sound wise, nevertheless because of material ressonance frequencies i did put a polyethylene (syntetic glass) 1000X500x4mm with DAEX25FHE4 exciters because of their top end roll off. And... the "joke" turns out serious.
It`s playing every day with soft eq and xo at 90Hz in a room with 30 m2, 4m apart from each other and hangigng on the ceiling with fishing string.

Just try.
Sounds good, I will have a look. How many Exciters are you using on one sheet? Which sub would you reccomend?
 
Last edited:
RMAM,
I'm wondering if the material you are talking about is PMMA, rather than polyethylene. PMMA is also known as acrylic or plexiglass. Among the materials I've tested, PMMA seems to have very good internal damping (as your post suggests), based on the impedance results in the link below.
Eric

https://www.diyaudio.com/community/...s-as-a-full-range-speaker.272576/post-7499940
The vendor specs indicate polyethylene, although is a general store (Leroy Merlin) not famous for their accuracy. So that´s a very good question.
In fact the sheets were mixed on the acrylic/pexiglass zone. I wil ltry to find the manufacturer and did ask my daughter to analise a sample in the laboratory.

I remember your measures in the post mentioned (thanks for your effort).
 
@pixel1
I think the frame doesn't need to be super stiff or strong, and the clamping should be really minimal. I have four small soft pads, 10x20mm, holding the plate in place on each side, so a total of 8 . Heavier clamping will make the panels sound too dead and reduce sensitivity. The deeper bass you want the softer the suspension has to be.

Generally I'm surprised how little of the vibrations are carried over to the frame, and purpose is really mostly to add protection and slightly dampen the plate to avoid ringing.
 
Not sure how you mean with blocking the sides? My frames are font and back of the plate only, covering about a cm around the plate, but the sides are open.

Frame does not seem to affect the sound hardly at all apart from the clamping effect similar to if you just hold on the the plate a bit with your fingers. It will improve bass response a bit.

Got the parts printed to assemble a test plate with my updated design and will take some pics when that is done.
 
  • Like
Reactions: 1 user
Did some sweeps in REW with the X32-4 panels. Both are using the provided VHB tape, and mainly wanted to look at differences between the suspended and free exciters. Unfortunately I did also change position of the plate suspension a little bit, which muddies the result.

First here are som pics of the panel with grill and suspension of the exciters:

PXL_20240210_153423660.jpg
PXL_20240210_153507950.jpg

And here are the TPU pads holing the pate in place:
PXL_20240210_153450621.jpg


It was a bit tricky to get the backplate with cabling in place so had to make a desperate solution with zip ties, but know to assemble them in the opposite order next time so I can make it tidy.

FR is not looking that good though. I already noticed with the first test plate I made without the suspension of the exciters that they where lacking below 150Hz, and suspension did not seem to help. In fact the suspended plates are worse, but that could also be down to the fact that the free exciter plates have been conditioned, or that I moved the plate suspension about 7mm towards the centre of the plate.
Started conditioning the new plate and will measure again later and see if that makes a difference.

The suspended plates generally have more uneven response, but does sound better to my ears. Using glue instead of VHB tape should improve the HF response a little, but not sure what I can do about the lack of response below 150Hz.
I have not had that problem on any of the plates using the Daytons, so I suspect that it could be due to stiffer suspension on the Xcites, in which case some more conditioning might help a bit. Or it is the change to plate suspension, so will test some changes to that. My subs will handle 150Hz without problem, and the plates does sound great despite the rough response, especially with some EQ, so I guess I can live with it.
x32-4 free and suspended.png

Distortion looks very good though:
x32-4 free distortion.png

And even better on the suspended version:

x32-4 suspended distortion.png
 
  • Like
Reactions: 1 users
Japanese DML/Flat panel loudspeaker patents - 1/? (Adding tubes to improve DML performance)

While browsing through the thread's catalog of patents, I suddenly had an idea. Japanese manufacturers have worked with flat panel speakers for several decades, and some good ideas may be stashed away in Japanese patents. A quick search in Japanese led me to several patents from a certain Katsuhiko Umeda on Google Patents. Patent no. JP2009206914A suggests adding a layer(s) of honeycomb tubes to the panel for some benefits. Namely:
Prevent the "wrap-around" of sound waves, which causes poor low-frequency characteristics in small speakers
Provide directivity in the mid-and-low frequency range to reduce unnecessary sound wave diffusion and subsequently, noise in the vicinity outside the intended area of the broadcast (outdoor concerts and annoucement broadcasts are given as examples)
The patent's proposed "flat panel speaker" and "ducting system":
The flat panel speaker of the present invention consists of a diaphragm panel, a vibration exciter mounted on the diaphragm panel, and a number of tubes arranged through a gap against one of the acoustic emission surfaces of the diaphragm panel. The ducts are made of an elastic material that seals the gap between the diaphragm panel and the surrounding ducts.
A diffusion suppression duct device, consisting of numerous tube-like ducts, is attached to the radiating surface of the diaphragm panel with a gap. The area around the gap is sealed with a sealing material.
With this device, the air inside the duct vibrates due to the sound pressure of the diaphragm, sound pressure is introduced into the interior of the duct.
The sound waves S are reflected off the inner surface of the duct, and the sound waves that hit each other interfere with each other, similar to the effect of a common "megaphone". The ducts are parallel, thin, and long, so they do not have the effect that a typical "megaphone" has.
The sound waves are emitted from the back surface B. In this process, the sound waves are oriented in the direction of the duct.
This phenomenon is similar to the principle that light traveling on the inner surface of a honeycomb core with a metallic mirror surface is directed by reflection and interference.
The tubular duct is made of a honeycomb core, and the axis of the tubular duct is formed at an arbitrary angle to the diaphragm panel. Furthermore, the axis of the tubular duct can be formed to have a focal point.
The cross sectional shape of the tubes can be circular, hexagonal, or quadrilateral.
the length and thickness of the tube must be selected according to the intended use of the speaker. If the tube is thin, reflections in the high frequency range, etc., will appear and affect the diaphragm and interfere with sound quality.
The narrower the distance between the diaphragm panels and the diffusion suppression duct devices was assumed to have a large effect, so experiments were conducted by varying the gap in various ways. In the experiment, it was confirmed that the narrower the gap, the less sound waves leaked from the gap, and thus the directional effect was greater. However, the limit is about 2 mm from the amplitude of the diaphragm.
To prevent sound waves from leaking from the gap, a sponge-like material was used around the diaphragm panels to prevent the sound waves from leaking
Example no. 1, a panel for listening to music:
The size of the diaphragm panel is 40 cm in height and 60 cm in width...the core thickness (tube length) was 40 mm and the tube diameter (cell size of the core) was 8.4 mm, with good results...The improvement was about 10db in the range of 50Hz to 300Hz.
You can refer to file F1 before and after. (a) is after the modification, which significantly alters low-end performance and attenuates the mid-range. It also worsens the frequency roll-off and dip between 8-10KHz (likely due to the weight of the tubes damping the panel).
Other examples discussed include:
  • Circular panel, 20 cm in diameter, with 80x8.4mm tubes for outdoor use (with the tubes in front and exciter in the back);
  • Wall-mounted panel with the tubes angled 45 degrees from the wall to reduce wall reflections;
  • More sophisticated angled tube design for a whiteboard or video screen;
  • Panel used for area-specific guidance broadcasts, such as station platform guidance, with tubes on both sides;
  • Curved tubes placed in front of the panel further strengthen directivity in the middle and high-frequency range.
I've attached some screenshots for convenience and reference. There's also a 7zip file that contains the original patent and translation by DeepL.
 

Attachments

  • Whiteboard-Projector Screen Panel.png
    Whiteboard-Projector Screen Panel.png
    17.9 KB · Views: 40
  • Quadrilateral tubes.png
    Quadrilateral tubes.png
    7.6 KB · Views: 35
  • Patent Description.png
    Patent Description.png
    10.9 KB · Views: 35
  • Outdoor Panel.png
    Outdoor Panel.png
    16.7 KB · Views: 38
  • In-door Music Panel.png
    In-door Music Panel.png
    17.3 KB · Views: 38
  • In-door Music Panel cross-section.png
    In-door Music Panel cross-section.png
    5 KB · Views: 35
  • Honeycomb tubes.png
    Honeycomb tubes.png
    7.5 KB · Views: 37
  • Full Tube Panel.png
    Full Tube Panel.png
    4.3 KB · Views: 36
  • Curved Tube Panel.png
    Curved Tube Panel.png
    4.5 KB · Views: 42
  • F1 before-after.png
    F1 before-after.png
    14.8 KB · Views: 40
  • JP 2009206914A patent translation.7z
    428.2 KB · Views: 16
  • Wall-mounted Panel.png
    Wall-mounted Panel.png
    12.4 KB · Views: 43
  • Like
Reactions: 1 user
The Honeycomb Sandwich Design has been utilised for many designs needing rigidity and lightweight. A very very fragile Honeycomb Structure, once bonded to Skins that are producing two fascia's, transforms the structural properties of each part used to produce the final assembly. The structure becomes much more than the sum of its parts used.
I believe a Aerospace Company has shown how much weight this Structure Type can take before destruction, which is usually much much more than the bulk of the guesses that have been made.

The diagrams of the Japanese Patents, all seemingly show only one face of the Honeycomb having fascia skin, which is possibly allowing for a degree of flexion, maybe the depth of the Honeycomb controls the flexion and enables a tuning of the sound.

There don't seem to be any restrictions to trialling a design, the link will show a few readily available options, the Skin Selection will also be interesting, as the Honeycomb will be offering increased rigidity so a usually rejected flimsy very lightweight material might be considered, .

https://www.plastock.co.uk/products...cjdR3XUT_heIJQCzZ8FNg_9wLxmvwelBoCCZQQAvD_BwE
 
  • Like
Reactions: 1 user
Curious, that`s close to what i`m experimenting: the exciter(s) in a carved honeycomb core with several materials combined as skins, although in the back also.
Thanks for the info.
I believe the patent requires an air gap between the core/tubes and the surface of the panel to work. The patent notes:
Further experiments were conducted to try attaching ducts directly to diaphragms and without a gap, but the ducts were not effective...However, since the air in the duct vibrates together with the diaphragm and sound waves do not run through the duct, reflection and interference of sound waves do not occur, and it was confirmed that sound waves cannot be given directional characteristics...Furthermore, the acoustic characteristics in the high frequency range deteriorate due to the increased mass of the vibration system.
P/s: what the patent considers as usable core thickness and cell size, and what happens when the latter is too small:
The results showed that the core thickness (tube length) could be reduced from 20 mm to 90 mm and the cell size from 3.2 mm to 12 mm in various combinations.
In both cases, the honeycomb core and roll core, too small a cell size (thickness of the tubing) can cause the tubing to be too thick.
Too small a cell size (tubing thickness) increased the sound pressure in the tubing and reduced the sound quality in the mid-range, partly due to interference from reflections.
 
Last edited: