A Study of DMLs as a Full Range Speaker

This remind me I have noticed the wires I use have a stiffness that probably modify the magnet resonance during test... Some portion of soft or reduce section wires are required to let the magnet as free as possible and more important to avoid any strength from the wires on the magnet/voice coil position.

Those 19mm exciters have small magnets. I can't find the weight of a DAEX19CT to compare but the force factor of the XT19 is higher which is good for the efficiency. Even if better than the 19CT is about 40% less than a 25mm...
By curiosity, I just run very quick simulations with PETTaLLS without any fine tuning to compare a DAEX19CT on a 30x50cm light EPS panel to a DAEX25VT on a 3mm plywood of the same dimensions. Both with exciters at 0.37/0.37, 1W, free magnet, free edges (FFFF). I did that to have an idea of the resulting efficiency (light panel with a low BL exciter, heavier panel with a higher BL)
The simulation shows the global efficiency are about the same.
The output I didn't expect is the roll off of the 19CT has a more important slope... what I can't explain. May be because of the higher stiffness or the lower magnet mass of the 19mm model
This reminds me I observed similar behavior while testing panels with a 13CT at home. I was expected with them a better HF extension than with a 25mm, which was true by rejecting higher in frequency the effect of the voice coil diameter but I was more surprised by the effect on the bass cut off.
Christian

PS: by the way, I would like to encourage all the DML builders to use such a tool. I think it is reaching a very good level even if all the possible configurations, exciters or materials are not available. There are 2 possibilities : its output are confirmed by the experience so we gain in confidence in its use, it is not confirmed so some aspect is maybe not covered or not with an appropriate model. In any case it is part of the learning experience. Thanks again to Dave (@EarthTonesElectronics) for the free availability of PETTaLS (follow the links to download and install) with its clear interface.

Dark blue : DAEX25VT on 3mm plywood, light blue : DAEX19CT on 10mm light EPS
View attachment 1478923
The wires also have their own resonances.
Especially at low frequencies, they will buzz and shake in combination with the added stiffness and weight.
This is another reason I do not use my exciters much below 100hz .
Mounting the exciters on to a spine will cure this problem though.

Did you also do a test of the DAEX25VT on the 10mm light eps, for comparison?
Steve.
 
Hello Eucy.
What part of my frequency response plots do you think are a fault?
Steve.
Hi Steve... I'm not saying they're at fault. You are questioning why you're getting strong results down to 40 Hz. while no one else is able to do that, particularly with small panels. You're the only one measuring with a Behringer...ipso facto...it would seem important to verify the results by another method... REW being the most easily available.
Eucy
 
Hello Eucy.
I have made and posted recordings and videos of my panels for people to listen to for themselves and hear the full frequency range.
The strong low frequencies are even stronger than on the videos and recordings as the microphones in my phone roll off above 50hz.
This is the confidence I have in my panels performance and sound.
I am not questioning why I am getting strong frequencies down to 40hz, I am questioning how I am getting strong frequencies down to 40hz.
I have measured other speakers using my behringer including my transmission line speakers. with no problems.
And did use it for room correction for a while.
The behringers are hooked up permanently in my system 100%of the time for the last at least 20 years with only one mains capacitor replacement.
But as I have said before, I would not use my panels down to 40hz or below , I use my transmission line speakers or reflex speakers as they give me more power at these frequencies, without all the problems of the panels at these frequencies.
Steve.
 
Bold of you to post.
Hello Eric.
Not really, I just don't care anymore 🤷
I have posted different recordings of my various panels to Christian Using voice from the microphone and music from the computer at the same time to see how they could be used for small PA work .
He probably thinks I am a right nutcase 🤪
My mouth was too close to the behringer microphone which had some foam wrapped around as a wind prevention.
So it was heavily distorted.
Steve.
 
  • Like
Reactions: Veleric
I think that this can be a pretty good mounting method. I've been calling this the "bridge" suspension, because the suspension bridges across the gap between the panel and the frame. I know I posted these images before but I don't think I shared the frequency response. Far from perfect but decent output down below 50 Hz. And about the best looking spectrogram I've generated. This was an outdoor measurement so really minimal reflections compared to most.
Eric

View attachment 1478228 View attachment 1478229

View attachment 1478235

View attachment 1478234
Hello Eric.
This is a very good frequency response, with only the 3 humps in the midrange, this is very easy to reduce without causing problems.
Although I am a little suspicious about the peak at 2.5k and maybe the one at 1.5k, but this is a little low in frequency.
The 2.5k could be exciter noise of some sort?
What would the flow chart look like after these alterations?
You could try my method of using thin film to reduce the peak at about 13k and extend the frequency response a little.
All you have to do is stretch the thin film over about 8 inches or so of the exciter area using masking tape.
This will not damage your panels as other methods do and is very easy to implement.
This should also help with more detail in the upper midrange as well because of the very light weight of the thin film passive radiation.
You could try different sizes or even cover the whole surface.
I did not carry on with these experiments as I should have, but the proplex panel and the other hard surface panels I used this on sounded much more informative and natural.
Steve.
 
I can't suggest what is the best material or design to be used as a Peripheral Interface on a DML Panel.

I can suggest that if a Soft Adhesive Foam Tape is wanted? There are products available that are supplied with pertinent information to show how they are design to function, from being very very supple and yielding, through to being with a hardness and unyielding.

The Foam shown in the Link is very very supple and is available as a Thickness ranging from 2mm - 50mm, with a Width of up to 1000mm. I know this product and it is extremely soft and low resistance.

My knowing EKI Foam through usage in the past, allows for me to inform, the softness is very very different to other Foams and Neoprene Type Materials having been handled.

EKI in comparison has no real indication if offers a resistance, a Panel Suspended on this material will be as free as it can possibly be to react to Amplitude within the DML and the migrating energies transferring to the periphery of the Panel.

If other resistances at the Periphery are of interest, there are other Foam Tapes and other Tape Types, the Higher Number with the Shore 00 will add hardness to the material and increase resistance to move freely, Shore A might be too much harness and resistance.

Shore 00 is like a uncured Silicone Sealant as a Softness, Shore A is like a well chewed Chewing Gum as a Softness.

https://www.ekibv.com/foam-rubber/epdm-foam/eki493
 
Last edited:
  • Thank You
Reactions: moray james
@moray james I chose to present this, as there are from individuals doing DML Panels and seen as those with the most experience, that are quite specific about a DML Panels base material, dimension and thickness, along with other Panel Treatments that can be added. All is quite easy to comprehend and are easily suggested as being standardized.

Methods for Supporting a Panel are not a Standardized Practice, there are many used, I am not sure if this is just individuals having access to different materials, or the designs are ones that the user feels has added benefits for the DML Panel when producing sound.

Having a Material that can be used at the Periphery as a Suspension for the Panel that is a Standardized Material and anybody can see how to achieve it for being used is a Win.
Having a Material that is very easy to identify, and being one that has intrinsic properties that may prove to be very beneficial when a Panel is suspended and producing sound, brings opportunities for all producing Panels to experience.
The Material suggested can be supplied in a Variety of Thicknesses and Widths.

A Fact First before a Hypothetical.
I used Forex Foam as Turntable Platter Mats, I purchased the Foam in 3mm and 5mm cut as a 300mm Diameter.
My first experience was with the 3mm, and it was rejected immediately, with the material not being considered any further.
At a later date a Selection of Metal and Ceramic Platter Mats were being tried out and for some reason, I decided to enter the 5mm Forex Foam Mat, which was its first ever use.
The 5mm Foam was a real treat to be used, it preserved all the important info allowing it to become sound as well as any Mat used and valued for its contribution.
I then added the 3mm Forex Foam and Kapow - Ugly colouration with the Mat being Not Wanted and immediately rejected??

Hypothetical
Suggesting the Peripheral Frame has a Eki Foam used to Suspend the Panel, where the Frame has a attachment of 20mm of Eki and the Panel has a attachment of 10mm of Eki.
The Foam used being Shore 00 @ 2mm x 50mm, this design leaves 20mm of free moving foam and might be seen to do the job required.
Another Panel is produced where the Frame @20mm attachment and Panel @ 10mm attachment is still maintained.
The change being the Eki is now Shore 00 30 @ 4mm x 150mm, giving 120mm free movement of an 100% increased in thickness foam that has a slight increase resistance to the freest movement, and there it is the Eureka moment :idea:.

There are always many many questions, but experimentation is the DML Way, with Objective and Subjective evaluation being the underpinning for options being suggested as a result of experiences had.
 
  • Thank You
Reactions: moray james
I was going to send this to Christian, but then I thought, no, I will post it here as it could be of interest to others.
This is a 1cm low density EPS panel.
I was playing pink Floyd medal on the panels only , and the low frequencies sounded week ,first picture.
So I tried holding the side of the panel each side of the exciter position, picture 2.
In that picture you can see that the dips in the response have been filled from 40hz to 80hz, but the frequency now drops off at 40hz (not shown in this picture).
So I decided to place some weights in these positions, I was in a hurry and did not have any weights to hand so I just clipped on a couple of pegs that I had at hand.
As you can see in the 3rd picture the low frequencies below 40hz have dropped off considerably.
But the strong output between 40hz and 80hz is still there ,this produced a strong rumbling sound in the room, not outstanding but better than some small bookshelf speakers.
Picture number 4 is of the panel and microphone positions, with the microphone just dangling from a speaker.
I was just interested in the low frequency output performance, so it's position was not critical.
This was just a quick listen nothing more.
But also, what was interesting was the frequency response of the whole panel was smoother with the pegs in place.
I have noticed this at times in the past ,but I am not sure if I have mentioned this or shown it ?
This panel can go very loud for long periods, without heating the exciters at all, as the exciters move very little even at high volumes.
I did send a video to Christian of me playing RAMMSTEIN at very high volumes, with me trying to shout above the "noise" 😱
Steve.
 

Attachments

  • 20250708_173021.jpg
    20250708_173021.jpg
    210.6 KB · Views: 23
  • 20250708_173132.jpg
    20250708_173132.jpg
    193.2 KB · Views: 9
  • 20250708_173808.jpg
    20250708_173808.jpg
    208.1 KB · Views: 24
  • 20250708_175155.jpg
    20250708_175155.jpg
    193.3 KB · Views: 24
But also, what was interesting was the frequency response of the whole panel was smoother with the pegs in place.
I have noticed this at times in the past ,but I am not sure if I have mentioned this or shown it ?
Hello Steve,
I remember you mentioned the role of the side mass as a possible way to act on the very first edge reflection or even a possible fixation point. It is one of my expectation to the simulation to see the role of such mass.
Christian
 
  • Like
Reactions: spedge
It is one of my expectation to the simulation to see the role of such mass.
Hello Christian it occurred to me that when Steve placed the wooden pegs on the sides of the panel (horizontal axis) the added mass would act to lower the standing wave in that axis and help to bring its resonant mode more in line with the longer axis of the panel which would have a lower resonant mode.
hopefully I will get back into the swing
looking forward to you getting back into the experimental swing of things. The overall improvement of the speaker with the pegs in place is of interest and makes me wonder about not only spot clamping of the panel edges with compliant edpm foam but also full peripheral panel damping as Eric has done. Have you any thought on this Steve?
 
as long as you know why you are clamping the edges of the panel ,and the consequences,
Well I am still most of the time thing of pistonic action in direct radiators and forget that we are dealing with a DML with its multitude of modes. The added mass of the pegs will go some way toward a closer balance between the resonant frequency of the shorter panel axis Vs the longer panel axis which would reduce torsional flex/twist of the panel but I don't know just how that would impact the performance of the panel without seeing a before and after frequency response measurement.
If the clamping action of the clothes peg impacts the hardness or density of the diaphragm due to material compression (at the contact point of the pegs) and that is part of what is accounting for the difference of performance (with Vs without the pegs in place) then a simple experiment could shed some light on the matter. Cut a hole in the diaphragm at the same location where the pegs clamp the diaphragm. With zero material now where once there was compresses diaphragm material you can listen and see how the sound compares. This assumes that clamping the diaphragm alters the ability of the diaphragm to express a modal point forming at the clamping positions on the diaphragm. Having zero material at those locations will eliminate the possibility of a mode forming there so if the issue is one of a mode Vs no mode you can listen and see what happens. I suppose that another way of altering the diaphragm density would be to press a deep dimple (deep enough to remain after the compression) into both sides of the diaphragm at the same clamping locations of the pegs. This will increase somewhat the density of the EPS foam while at the same time altering somewhat the balance of the diaphragm. This is making me think that this process of cutting holes or compressing parts of the diaphragm is similar in a way to balancing a wheel to remove a shimmy at a given speed/frequency. Just thinking out loud here. If any have thoughts please jump in. Thanks.
 
I've finally located a source in the US for Proplex. Entrypoint ImpactBoardFR1500.

The company calls the versions with round flutes "ImpactBoard FR".
They carry it in 5mm and 6mm thicknesses.

As I understand it, the 5mm ImpactBoard FR 1500 version comes in at 60% of the density of the 6mm version does.
Likely more efficient?

Located in Boston, Entrypoint does not have geographic distributors for this material.
One has to arrange pick up. They might ship a whole pallet.

Still awaiting pricing and how much scale they need to sell at a time.

I'll update as I have more information.

ImpactBoard FR 1500 (5mm) details

Material Construction: Polypropylene Copolymer​
Description: Flame-Retardant, Antistatic, O-Profile Fluted Sheet​
Intended Use: Temporary Protection from Heavy Traffic and Impact on Horizontal and Vertical Surfaces​
Appearance: Translucent / White​
Unit: Sheet​
Dimensions: 2.25 m x 1 m (~89” x 40”)​
Thickness: 5 mm (~0.2”)​
Weight: 3.4 kg (~7.4 lbs.)​
Stocking Units: Individual Sheet or 200 Sheet Pallet​
Surface Resistance: 10^11 Ohms​
Storage Recommendations: Store away from heat sources, open flames and strong acids and/or bases. Store indoors within recommended storage​

I wonder if the "Surface Resistance" has any bearing on its performance as a DML panel?

Skudo Board

I found another product... Skudo Board that might prove interesting
This stuff has a circular honeycomb like structure perpendicular to the surfaces (see photo).

This material would seem more balanced across the panel axes than ImpactBoard.
Whether an improvement over ImpactBoard, I leave to discussion and empirical evaluation.
Thoughts?

They have distributors across the US.

Dimensions: 4’ wide x 8’ long
Square Footage: 32 sq. ft./board
Thickness: 5mm (1/4”)
Color: Grey

1752268220603.png
 
  • Like
Reactions: moray james
In relation to Entrypoint, they will have Distributed this Product to various locations.
I would ask them if you were to give them your home area, if they could supply a List of Deliveries made in the past 18 Months that are the closest to this area, this will then enable a inquiry to learn if any redundant Sheets are available in either Used / New condition.

The round flute board I have supplied, was waiting to be dumped stacked in piles about 5 foot high, an inquiry to have a selection will not be met with a 'No Answer' if at that stage of the usage life, hence my suggestion of 18 Months since delivery.