A Study of DML's as a Full Range Speaker

spedge

Member
2009-09-03 3:41 pm
pepe.
also if you missed my recordings on page 116 (yellow, duhast , to rich for my blood,)where i demonstrate how loud the panels can go, now you do not have to download them,they play from the site.
these (and all my other recordings) were recorded at disco levels(yes i know im old) so you have to crank the volume up on you headphones to get the full experience .
and also please use your headphones only.
you will not get the depth and size presentation in the room but within the limitations of my recording technique ,you should get some idea of how they sound, hopefully.
steve.
 
I have not tried CF on its own. It may indeed be good, and I think 2 mm is a good thickness. But I wouldn't expect it to be particularly efficient. According to my model, for whatever its worth, it should behave similar to 5 mm thick plywood, at least with respect to efficiency.

For most applications, the real benefit of CF comes when it's used in a structure, like a hollow tube (as for a bike frame) or a sandwich composite with a lightweight core. I think the same is true for a DML panel. But I don't know for sure the best construction. That is, core material and thickness, cf layup and thicknesss, and size.

Eric
I'm OK with inefficiency- watts are cheap and easy. My main objective is an even, responsive, full range sound. Plexiglass is even and full range, but kind of dull... hoping a CF panel in the same size will cover the same spectrum but with more "life"
 
Pepe.
I just realised how excited I was righting about my ideal system.
I need to stop all this experimenting nonsense, clear my room and set up my system.
I'm going to try and make this my goal for 2022.
I can do this 😲
Steve.
That does sound like an interesting and flexible system!
The dml could be very handy when one is moveing around indoors and the cone+dml can be used during seated listening.
 
Christian,
I heard it exactly the same as you. He says 3.5 to 4 mm, so definitely not thinner, which is what I was thinking.
I expect you are correct that the efficiency information is another clue, but I'm afraid I also don't know how to use it.
Thanks for the link to the soundboard thesis. The comment about the D/mu^3 applying for a fixed thickness is interesting, and something I never read before. I'm going to have to give that closer read, and some of the references as well. Maybe some other factor (D/mu?) is actually a better "index of merit" in general. The old patents by Heron all reference the D/mu^3 criteria, while at least one other (Warnaka) references D/mu.



Eric
Eric,
Thank you for the links. I am adding in the "history file" already posted a sheet with my bibliography about DML. As I didn't store all the links to documents, yours are welcome!
This week I was going in my doc about how pistonic loudspeaker works and I found the criteria E/rho^3 which is also D/mu^3 was proposed in 1958 by Harris as factor of merit for cone materials.
To follow
Christian
 
Christian,
I have mentioned balsa before, maybe it was a post of mine you recalled. I have been playing around with balsa in different ways. First, as a core material with carbon fiber or fiberglass skin layers. And secondly, balsa alone, made up like a plywood.

...

Compared to my revply plywood panels, the cf/balsa panel has the greatest efficiency. Perhaps about 7 dB higher than the revply plywood. The balsa plywood is between them, about 4 dB above the revply, and 3 dB below the carbon/balsa.

The carbon/balsa also has a slightly flatter frequency response, while the balsa plywood has a bit of a hump at 200-300 Hz. However, in spite of the 200-300 Hz hump (or perhaps because of it??), and the slightly lower efficiency, I find the tone of the balsa more pleasing, for reasons I don't understand. I only know that whenever I play music on it, the sound makes me feel good!

I'm currently debating with myself about which of these to use to build my next complete speaker pair.

Eric
Thank you for your answer Eric. You answer exactly to the questions I have. Don't worry about the units. With thread I am now used to convert.
On my side I have in stock balsa for next panels. It is 10cm (4" about) x 1m (3'?) x 1mm thick board. My intention was to make a sandwich with this balsa outside as skins and a core made of 3x5mm balsa rods radiating from the exciter axis and xps (or eps) for the space between the rods (clear enough without drawing?) so reaching 5mm total thickness. Another possibility which needs much much work (if feasible) is to make my own honeycomb. So before investing time, it is interesting to share this kind of ideas.
It is very encouraging to read your experience putting this balsa panel in the top of what you tested. Balsa is a kind of "obvious solution" to improve efficiency, not so obvious from stiffness point of view.
Christian
 

spedge

Member
2009-09-03 3:41 pm
Pepe.
I was looking through some of those recordings from page 115 to 119 starting with some eps and working through my then collection including card with domes and chalk board.
I can always tell the eps panel as they sound so alive and the realism is amazing.

the idea of using the bandor unit was for recordings that can sound odd on dml panels.
usually ,I believe because they have been heavily mixed on normal cone speakers which don't show up the problem if you play them on normal speakers.
Having a choice of the cones and dml covers all variables in the recordings.
Steve.
 
Pepe.
I was looking through some of those recordings from page 115 to 119 starting with some eps and working through my then collection including card with domes and chalk board.
I can always tell the eps panel as they sound so alive and the realism is amazing.

the idea of using the bandor unit was for recordings that can sound odd on dml panels.
usually ,I believe because they have been heavily mixed on normal cone speakers which don't show up the problem if you play them on normal speakers.
Having a choice of the cones and dml covers all variables in the recordings.
Steve.
Oh, I see. Thank you. I didnt think of that recording/mixing variable. Makes sense and allso makes even more sense to have both options then.
 
ok, I'll bite.
Tried corrugated cardboard, 3 ply, several coats of lacquer,brushed on, both sides. To me it sounds very good, but, without sealing the edges, it will warp, due to changing humidity levels.
My next attempt will be plastic corrugated panels, I have sampled a listen, on a size far larger than I will use, and it sounds comparable to card board corrugated.
Sorry, but I'm not getting into the "weeds" about testing, and posting some non-nonsensical, ******** that takes a mathematical genius to decipher, these sound good to me, not some machine telling me "what is correct"
 

spedge

Member
2009-09-03 3:41 pm
Mexjerry , jaxboy.
I did have a look at 1cm hollow polycarbonate sheeting but wasn't that keen on the sound.
But thought the 4mm might do better.
I looked into the prices but unless I bought in bulk it wasn't that cheep,so gave it a miss.
I think there was a YouTube video ?
I felt that the 1cm was too heavy ,maybe for my exciters .
I originally started off years back with I think it was 2mm corrugated cardboard which at the time sounded great.
Everyone else seemed to go for the thicker three of more plys
Some used to inject shellac into the flutes which made the cc bend out of shape.
This also made the panels even heavier.
It would be interesting to know what the 4mm polycarbonate sounded like ,but I'm not that keen on corrugations even if they are in the shape of honeycomb, like the podiums.
I think my 5mm xps with epoxy coating is probably a better option, for me anyway.
Steve.
 
The carbon/balsa also has a slightly flatter frequency response, while the balsa plywood has a bit of a hump at 200-300 Hz. However, in spite of the 200-300 Hz hump (or perhaps because of it??), and the slightly lower efficiency, I find the tone of the balsa more pleasing, for reasons I don't understand. I only know that whenever I play music on it, the sound makes me feel good!

Eric
basically , I think you have hit the nail on the head.
When I bring in my TLS to cover the 300hz and down ,I use my ears, and without fail I always end up with an increased output in this area.
to my ears if the panel is equalized flat or a dip is left in this area ,the dml panel seems out of balance, I don't know why but the panel sounds to me , like It has more energy above this frequency ?
raising the response below about 300hz or so ,brings back the warmth and body to the panel sound.
This is the main reason I am not keen on equalizing a dml panel to have a very flat response, to my ears this sounds wrong.
to be honest I usually find That dml panels have a slightly sloping response(not always) similar to the room curve I have seen in papers , I usually just follow this line below the 300hz and with a little bit more boost depending on the panel.
I think this could be one of the main reasons that some find dml panels hollow sounding or lacking in warmth ?
Steve.
 
ok, I'll bite.
Tried corrugated cardboard, 3 ply, several coats of lacquer,brushed on, both sides. To me it sounds very good, but, without sealing the edges, it will warp, due to changing humidity levels.
My next attempt will be plastic corrugated panels, I have sampled a listen, on a size far larger than I will use, and it sounds comparable to card board corrugated.
Sorry, but I'm not getting into the "weeds" about testing, and posting some non-nonsensical, ******** that takes a mathematical genius to decipher, these sound good to me, not some machine telling me "what is correct"
Relax Mexjerry,
We have all our own way of doing or understanding things. Mine for example is to keep in parallel realizations and measurements and math. It is not math that tell me what works or not but I think that we could find some elements that would help going faster to design that fits to our expectation by limiting the trial/error cycles. Just an example, for a given material, which area to reach the expected low frequency. By some aspects, DMLs remind me the loudspeakers in the 80s (or before?) when there was no calculations of enclosure. I think we all agree to say they sound pretty good and are promising. The different approaches we can have is a possibility to make them better together.
Your post was the opportunity to read what you posted before. Very interesting your experience with aircraft and the techniques you described in stretching fabric (#3131, 3136, 3153, 3159 page 157.
Christian
 
Pepe.
40 Euro for such a small panel seems expensive to me.
Are we talking hollow core ,as I could only find down to 4mm for green house use ?
A 20watt or above exciter should be ok as long as long as it has a full output up to 20k ,I believe the 25mm exciters usually do this ?
Steve.
It is not a hollow panel. It’ a plain 3mm sheet.
Thanks Spedge, I will start my research on exciters. Allways good to know where to start.