A Study of DMLs as a Full Range Speaker

Hello all
Being not the possibility to experiment currently around DML, I am thinking about the next step for me which adding equalization.
I would like to go to some modular approach having the equalization, the DACs, the amplifiers separated. For the equalization, I am looking to Linux PC or Raspberry solutions.
I found as example this Equalizing audio under linux / Raspberry Pi (for DML) from Paul Wieland (Paul W are you also here on diyAudio?).
Does somebody have already something working ?
Thank you for the help.
As this is not only something interesting for DML, there might be other good threads.
One element of my "specifications" is to have a toslink input.
Christian
 
JohnnoG.
I have been listening to the 60x60cm x 4mm panels full range for a couple of days now.
My 10watt exciters are having no problem driving this 4mm panel at the moment , with no overheating
Although I haven't tried heavy metal yet !
I could Even roll off below 40hz to further improve power handling.
Even though it does have the usual dip in the response in the 100hz to 200hz area , I haven't noticed the dip when listening, so far .
So it is a pretty good full range panel in this size.
I think you will find it very interesting listening to movies on these panels, even when using cheap audio equipment.
The sounds become more real, very real in fact, my ears start twitching listening to the sounds in the soundtrack.
And this is with a very cheap all in surround receiver ,in my front room.
I made a recording of Big Ben on this system using very small card panels, it was very impressive.
So size isn't a problem as long as a sub is used.

I went to the 2019 hifi live show , hopefully my wife has left those days free for this year's show ?
Steve.
 
Hi Steve
I have been through the thread in the recent past and downloaded discovered recordings on offer.
Big Ben was not discovered but many years ago I worked very close to the Tower and heard it live at all hours as we did night shifts as well, in the early morning hours, with no noise pollution is the most memorable.
There were nuances in your recording that made me recollect those early morning bell strikes.

Your endeavours into this subject of the DML is quite advanced and there is many experiences had, plus many shapes and materials used as Panels, it certainly is appearing to be addictive past time.

I will have a listen to the page 179 recording tomorrow, with a coffee in hand.
 
Last edited:
JohnnoG.
As you can see from the photo on page 179 ,these panels have been cut about rather a lot and are basically at their end of life , so not optimum for this setup.
They are also just hanging from tape a few inches from the wall close to where the normal surround speakers are mounted.
But still they brought another dimension to the sound, even though I could not XO them properly with this system.
I would have been happy to leave them permanently in situ, if made prettier.
If space is a problem, The small panels can perform as good as the large panels , with proper XO.
Steve.
 
Christian.
There is no easy answer to your question.
There are too many variables.
it would depend on panel thickness and size, the response of the panel materials, the size of room and cross over point chosen ?
plus mounting methods.
Let alone exciter problems and more ?
Obviously my 1mm card or veneer panels are restricted to smaller sizes .
Likewise I would not recommended a 50mm thick 200mm wide eps panel.
If you have a panel material that you prefer the sound of ,I'm afraid the only way to find the optimum size is to test then yourself in your own home .
JohnnoG s 60cm x 4mm panel seems to work quite well ,I'm not sure what would happen with a larger panel of smaller panel ?
If using 25mm 70 grade EPS , I probably would not go below 60cm,
I would think of using a thinner thickness.
My 5mm epoxy coated xps I thought could probably be used in a smaller size ,but as usual I became side tracked to another panel material.
I would recommend anyone using 5mm xps to coat the panel with epoxy , for a better response.
I have not tried thicker xps .
It is very difficult when you have so many different panels to concentrate on one only.
I am very lucky to have found so many panel materials that I like , in there different sizes.
Johnnogs panel is just another one to add to the list, which is nice 😁
Going too large or too small with a certain panel material , is when you start to have extra problems to solve.
There is no one size to fix all.
Steve.
 
The good news is that the Panels that are being claimed to have offered the most impressive SQ, are not difficult to acquire.
There is the advisory that certain materials will benefit from an additional treatment, but still a relatively easy to achieve application method.

This has been great to allow for the extensive investigations that have been undertaken to date.

I suppose there will always be the question, that if a material did become available, that was quite difficult to acquire and was with a notable increase in additional expense, that did suit the role better as a Panel, would such a material be worthwhile to be used as an alternative to the successful use of materials already in place.
 
Loeb, you might find this interesting.
Here's another great video for anyone interested in making their own composites.
The main thing I learned here is that if you drill holes in the core material, you can pull out excess epoxy from both the face layer and that back layer of your composite. He shows divinyl foam as a core in this example, but I think the same process could be used for other cores like balsa.
What's great about this approach is that one face (the bottom face) can be perfectly smooth and flat, being cured against a smooth surface, while still pulling excess resin from both faces.
Eric
 
Steve,
You are so right. Take Bose, for instance. Some people cringe at the very word, but when I heard the 901s in the 70s, I knew that they were the cheapest speaker on the market, because sooner or later, I would HAVE to have a set, so why delay the inevitable and wasting money by buying some other speaker in the meantime. I absolutely loved them, and even bought a second set. The surround sound was unbelievable! 36 speakers! 18 Hz! Others scoff at their "pedestrian" sound and their appeal to the "masses". Well, I belong to the masses, I guess, and that's alright by me. I have a master's degree, but I'm still a redneck in some respects, and that's ok, too. You make speakers out of CEREAL boxes and packing material, and love their sound. The purists must be gnawing at their $100/foot speaker wire at the very thought! Who would have thought you could get glorious sound from an artist canvas? I admire you for constantly breaking new ground and pushing the edge. I would still be listening to wooden boxes were it not for XRK971 and you. You two have rocked my world, and I am humbly and forever in your debt.
 
Jaxboy.
Thank you .
Only $100/foot ,cheapskates 😂
I started out trying to find a horn speaker to build, and came across ziggy on NXT RUBBISH, he had what looked like a pretty decent horn system.
But he thought that his large panels sounded better, well over 10 years and I'm still playing with them.
So my thanks go to ziggy, R.I.P.😔
Steve.
 
Loeb, you might find this interesting.
Here's another great video for anyone interested in making their own composites.
The main thing I learned here is that if you drill holes in the core material, you can pull out excess epoxy from both the face layer and that back layer of your composite. He shows divinyl foam as a core in this example, but I think the same process could be used for other cores like balsa.
What's great about this approach is that one face (the bottom face) can be perfectly smooth and flat, being cured against a smooth surface, while still pulling excess resin from both faces.
Eric
Thanks Eric, been watching most of his videos and has hopefully digested enough information to start doing some actual panels!
I have everything in place except the workspace, just need to get a mirror to use as working surface.

DML project has been paused due to the many other preparations I have to do for our little event in middle of July. Unfortunately the sound guy we usually use cannot make it this year, so I'm considering if I will manage to source everything I need and construct the plates in time now. I'm thinking 8-16 nomex/carbon plates with a total of 32-64 DAEX30HESF-4 exciters and 4 Hypex FA502, so quite a commitment, both in money and time :\

Right now I'm thinking mostly how to configure each stack to make the most of the exciters and amp. Each stack will have 8-16 exciters (haven't decided if I will need 4 plates per stack, or if 2 will be sufficient). The amp is a bit odd in the power ratings at different impedances. If running two separate channels, each channel is 300w @ 2ohm, 500w @4ohm and 350w @ 8ohm. If bridged it is 1000w @ 4ohm and 1200 @ 8ohm.

That means running 2 plates with 4 exciters per plate I could do on of these configurations:
A) Amp bridged with 2 strings with 4 exciters on each, giving 8ohm load and 150w per exciter.
B) 4 exciters in series-parallel per channel on the amp, giving a 4ohm load with 125w per exciter.

And if I would like double the plates, so total of 16 exciters, it would be:
A) 4x4 grid of series parallel with amplifier bridged, giving total of 62.5w per exciter @ 4ohm
B) One 8ohm string per channel, giving only 43.75w per exciter.

Would be nice to be able to get the full possible 75w per exciter when using 16 exciters in total on the amp, but without using 8ohm exciters that seems impossible. And doesn't seem to be any good 8ohm exciters on the market.
 
Thanks Eric, been watching most of his videos and has hopefully digested enough information to start doing some actual panels!
I have everything in place except the workspace, just need to get a mirror to use as working surface.
Leob,
Do you have an idea yet about the individual panel design? Length and Width? Nomex thickness? Carbon thickness? Panel mounting?
Eric
 
Leob,
Do you have an idea yet about the individual panel design? Length and Width? Nomex thickness? Carbon thickness? Panel mounting?
Eric
Will try out the same dimensions as Tectonic to start with (400x575mm). Don't want larger then that anyway, but might try tweaking down the size a bit. I have 3.2 mm nomex (https://www.easycomposites.eu/3mm-29kg-aerospace-nomex-honeycomb) and 64g/m2 carbon (https://www.carbix.se/product.html/kolfibervav-flattow-62g/m2-plain).
Mounting I have not decided on yet. I might make a construction of printed parts and glassfiber rods to try to make something slic, or just make simple wooden frames.
 
Will try out the same dimensions as Tectonic to start with (400x575mm). Don't want larger then that anyway, but might try tweaking down the size a bit. I have 3.2 mm nomex (https://www.easycomposites.eu/3mm-29kg-aerospace-nomex-honeycomb) and 64g/m2 carbon (https://www.carbix.se/product.html/kolfibervav-flattow-62g/m2-plain).
Mounting I have not decided on yet. I might make a construction of printed parts and glassfiber rods to try to make something slic, or just make simple wooden frames.
That should be very very good from an efficiency perspective. Probably as good as or better than PS foams in that respect.
I would love to try carbon that thin. But at my supplier the 67 gram fabric is twice the cost of the 90 gram, so the 90 is probably what I will try next, even though the 67 is probably a better choice.

You might consider trying the 2 mm Nomex also. I just did some estimates (purely theoretical) to get the expected stiffness and density of your proposed panels, and from there I estimated the expected natural frequencies using FEA and assuming that your frame would create effectively a hinged joint around the perimeter. With the 3 mm nomex and your carbon (about 0.08 mm thick) I estimated a fundamental of about 120 Hz, compared to about 80 Hz for the 2 mm Nomex core.

You should get decent output for each starting at their respective fundamentals, but most likely the FR will be pretty bumpy until 2 or 3 times that. So, depending on where you want to cross over from the sub, you may prefer the thinner nomex and the corresponding lower natural frequencies.
Note of course that these are very rough estimates based on lots of assumptions that may or may not hold very well. So take it all with a huge grain of salt!
Eric
 
That should be very very good from an efficiency perspective. Probably as good as or better than PS foams in that respect.
I would love to try carbon that thin. But at my supplier the 67 gram fabric is twice the cost of the 90 gram, so the 90 is probably what I will try next, even though the 67 is probably a better choice.

You might consider trying the 2 mm Nomex also. I just did some estimates (purely theoretical) to get the expected stiffness and density of your proposed panels, and from there I estimated the expected natural frequencies using FEA and assuming that your frame would create effectively a hinged joint around the perimeter. With the 3 mm nomex and your carbon (about 0.08 mm thick) I estimated a fundamental of about 120 Hz, compared to about 80 Hz for the 2 mm Nomex core.

You should get decent output for each starting at their respective fundamentals, but most likely the FR will be pretty bumpy until 2 or 3 times that. So, depending on where you want to cross over from the sub, you may prefer the thinner nomex and the corresponding lower natural frequencies.
Note of course that these are very rough estimates based on lots of assumptions that may or may not hold very well. So take it all with a huge grain of salt!
Eric
Hello Eric
I am in the opinion the need of stiffness is often overestimated. This even more with light material. A 1st mode at 80Hz with the 2mm core seems for me a better target too.
Christian