[Kame2]

If you look at all the professional manufacturers of DML/BMR loudspeakers from Bertagnis, Sony's APM's, Technics SB flat driver series, Yamahas Ear speakers to Podiums you will notice they ALL use some sort of frame and spline....There is a reason why all of these professional loudspeaker manufacturers utilize a frame and spline.

The spline is needed to support the magnet on higher excursions as it needs something to push against....with nothing to push against to support the magnet on higher excursions the magnet will rattle and make a lot of noise and also some percentage of energy will be lost resulting in lesser/weaker vibrations going all into the panel....not sure if that made sense but oh well. LOL

I don't wanted to disagree. I just do not have the experience to say something about the sound quality with or without a spline. In my little experiment I noticed no difference. That's neither for nor against it.
Though, as I explained, I can imagine that it might improve bass when the mass of the magnet is not enough to withstand the force pushing against the panel.
But what I see is that it is a good idea to take the burden from the light plate. So I support your statement to use a spline. Just more from the technical part where I have more experience with.

[Bendingwave]

In doing many side by side A/B comparisons of DML panels with a spline and panels without a spline I have noticed this difference.

Panels with a spline tend to sound a tad more tighter/accurate and pronounced in all frequencies with a more forward like sound effect.


Panels without a spline tend to sound more diffused and less pronounced with a more (recessed) laid back like sound effect.


DML panels have a very diffused sound to begin with especially the "Free Floating Panel designs", so it can sound unnatural when compared to the conventional cone type of drivers. Which sound one prefers is personal preference but it seems that the high end speaker manufactureres choose the one with the spline and a frame.


Free floating panels sound very diffused.


Panels utilizing some type of Frame and Spline sound less diffused.

[Kame2]

Ok, note to myself: When building the next panel, save one measurement without a spline and one with and compare the direct sound components.

Next topic: Maybe if I have enough prototypes I try DMLs for a home theater installation. But that requires direct wall mounting for the surround speakers which kills the sound. I tested a bit with 5 mm sound insulation (XPS, the thing you put underneath laminate) as panel and noticed an improvement when bending it (looks a bit like the plywood panel someone constructed for wall mounting). Thinking about this I come to the conclusion that the non uniform distance to the wall prevents destructive interference with reflected waves. So it might be a good idea in general to (slightly) bend a panel to reduce negative effects of walls nearby. What do you think?

[Bendingwave]

It depends on how one mounts there surround speakers...I use the 9 inch L-brackets to mount my surround DML panels so there is about 7 inches from the speakers to the back wall.


I haven't experimented with bending of the panels so I cant really comment on it but JMC does bend there panels since there panels are (if I am not mistaken) mounted close to or directly on the wall.....So you could be right in bending a panel slightly if mounting it directly to the wall.

[koko007]

Hi Bending and Kame,

did you see:

http://www.tornadoacoustics.ru/forum/25-1026-4

[koko007]

and this:

http://www.tornadoacoustics.ru/forum/25-1026-3

Sorry for 2 posts, i cant find way to edit first. 

[Kame2]

I tried my best with google translater. What was the most interesting for me was the panel material discussed on page 2. It is on my todo list to try to laminate thin XPS with fabric. Mainly for protection. My test-panels all have notches from my wife and the kids. For whatever reason no cat has harmed the panels so far...

[Odal3]

Thanks for the historical article Odal3.

Someone on Audiokarma said that Yamaha licensed the technology from Bertagni/BES....which is probably why its shaped that way like an ear. If you look at bertagnis design its shaped very similar.

https://img.usaudiomart.com/uploads/large/1096124-bertagni-electroacoustic-systems-bes-sm100-speakers.jpg

Thanks for sharing too - I had never seen those before, so had to do some googling. Fun to see that the thread is still going on!

[koko007]

On the third page there are links (now broken) with the movie of the speaker in the work. I can tell you, that the panels look quite flexible and that the bass panel has a great extrusion (they made their own exciters). Panels, seem to me, were made of thin xps with laminated fabric, something like 2-5 mm sound insulation. Also, i think bending  panels is good way to go like in this case (what i did with antinox like material):


https://www.youtube.com/watch?v=FNugB9DZzK0&frags=pl%2Cwn

and

https://www.youtube.com/watch?v=Kdz4A0Ibg9A&frags=pl%2Cwn

I was tried material like antinox but in my country i cant find 2mm thick material so i used 3 mm which is to much for my exciter (much quieter than xps), although the sound was more natural than xps.

And this is measurement of Russian speaker:



Impressive!

[Kame2]

Let me give you a small update on my experiments. First I added a small exciter to my panel and take a frequency response for each:

Next I will order some components for a simple crossover at about 6 kHz.

Additionally I started the construction of an wall-mounted dml-speaker with a curved panel. As I read on another forum that someone made enclosures for dml-speakers out of foam board I thought I test some paper honeycomb board. I bought it for my dml experiments but it sounds not good enough.

[Bendingwave]

Paper honeycomb does NOT vibrate like foam as that is one of the main reasons why EPS is my choice of panel material as well as XRK's foam speaker enclosures.


When EPS/XPS foam is used as an enclosure (or DML panel) it vibrates like a instrument especially in the lower frequencies......I made a DML panel sub using T.E. frog exciter (incorporating XRK's foam enclosure) and compared it against the conventional cone type of subwoofer (polk psw10) using the same sub amp and my DML sub sounds better with better detail, accuracy and it actually plays deeper/lower....The only thing the polk psw10 did better was it hit harder with more punch due to the conventional pistonic movements of conventional cone type of drivers. Sure the polk sub is a entry level sub at only $100.00 but my DML sub cost me only $35.00 to build.

Also a DML sub blends better with DML panels then with a conventional cone type of sub.

[Kame2]

Paper honeycomb does NOT vibrate like foam as that is one of the main reasons why EPS is my choice of panel material as well as XRK's foam speaker enclosures.

So you suggest to use the same material for enclosure as for the panel?

When EPS/XPS foam is used as an enclosure (or DML panel) it vibrates like a instrument especially in the lower frequencies......I made a DML panel sub using T.E. frog exciter (incorporating XRK's foam enclosure) and compared it against the conventional cone type of subwoofer (polk psw10) using the same sub amp and my DML sub sounds better with better detail, accuracy and it actually plays deeper/lower....The only thing the polk psw10 did better was it hit harder with more punch due to the conventional pistonic movements of conventional cone type of drivers. Sure the polk sub is a entry level sub at only $100.00 but my DML sub cost me only $35.00 to build.

Also a DML sub blends better with DML panels then with a conventional cone type of sub.

A DML sub is interesting! Do you have any pictures to share? Which dimensions do you use and how deep does it go?

[Bendingwave]

Yes....Even I wouldn't have thunk it until I read XRK's foam core speaker enclosure on the DIY forum.


A DML sub is just a panel utilizing a sub amp so that only the low frequencies saturate the panel....sorry don't have any pics....panel size is around 16inchX24inch.....I cant tell you exactly how low it goes as I don't have the means to measure it accurately but the polk psw10 states it can go as low as 40hz so I am guessing my DML subs are lower like in the 30hz range.

Right now I am tweeking my DML satellite panels that are only 10inchX16inchs. Which uses only a single Dayton 13mm exciter which is around the size of a nickel....Currently am using the receivers high pass cross over at 100hz for the DML sats…..I could later on just put a non polarized electrolytic 250uf capacitor directly to the DML sats.

[Kame2]

Do you think a single 3W exciter is enough to handle a moderate playback volume?

How thick are your panels? And how thick are the walls of your enclosures? What material do you use for the spline? Do you take any care to secure your foam from damage? Or do you just don't allow kids and pets (and the wife) to get in the same room?

[Bendingwave]

It would depend on how big ones room is and how loud you like to listen to music.....In my small 10ft.X12ft. room it can play at sufficient moderate levels as I don't listen to music at very loud levels but you will need at least a 100hz high pass crossover for the Sats.


This is just my prototype DML satellites as I can make them have greater SPL by utilizing a higher power handling bigger/larger exciter....With these panels you can custom tailor every aspect of it.

My DML satellite panels are 1/2 thick high density EPS......enclosure is the same 1/2inch thick....Material for spline is wood.


I dont have to worry about damage of the foam as no one touches my panels, not even my chihuahua. lol

[Bendingwave]

Finally finished tweeking out my DML sat panels to where I am completely satisfied with there sound quality...They sound friggen amazing, in fact they sound more accurate then my bigger 2 way DML panels....Like I've always assumed that a single exciter per panel is the best way to go for the most accurate of sound.....with that said my next project is to make a 2 way with two separate panels utilizing 1 exciter per panel instead of 2 exciters on one panel....At $2.00 per exciter on sale you are not going to find a better performing sat speakers at that price. Price to sound quality is unbelievable.....OF COURSE IT ALL WILL ALL DEPEND ON ONES DESIGN.

Calling my DML sats "BASTAT"....paired with any type/brand of powered sub with sub cross over set at 100-160hz they sound phenominal….only down side with utilizing these small exciters is they wont play that loud.

I wish there was a way to let you all hear them I bet yawl be blown away and even more so when you realize it cost almost nothing to build them. LOL

[Kame2]

I have build a crossover for my panel. The highs are way better now. Relatively smooth up to 10 kHz, than falling rapidly.

I see this 10 kHz falloff on all my XPS 2cm panels. Seems to be material specific. The bended panel I am working on uses XPS with 5mm thickness. First tests by just holding an exciter against it suggest that they may play up to 20 kHz without falloff.

Like I've always assumed that a single exciter per panel is the best way to go for the most accurate of sound.....with that said my next project is to make a 2 way with two separate panels utilizing 1 exciter per panel instead of 2 exciters on one panel....

I second this. The panel I built sounds horrible when increasing the volume. The tweeter produces strong resonance distortions at ~470 Hz. Seems like the shaking of the panel is to much for it. I think I try to add a stick of wood diagonal at the corner to separate the tweeter part. Inspired by the bertagni speakers.

I wish there was a way to let you all hear them I bet yawl be blown away and even more so when you realize it cost almost nothing to build them. LOL

That is the negative part of this hobby. You can capture an image of your speaker and show it to the world but you cannot let the world hear it. The only chance is to document your journey to encourage people to try it themselves.

[Bendingwave]

Thicknesses in panels play a significant factor....2cm (1inch) compared to 5mm (1/4inch) is a big difference so it will effect the way each panel sounds.

One way to decrease tweeter resonance is to use a non polarized electrolytic capacitor (4uf) at around 5khz.


Yep I agree...The closest way to let everyone hear the sound is through a well recorded audio/video...Even though its not realistic its better then nothing, but at this point in time I do not have the proper equipment to do so.

Tip/one of my techniques...........DML panels have a very very wide (dispersion) soundstage in which it can sound very diffused or unrealistic....to control dispersion one can use the "FRAME", as sort of like a guide to control and focus the sound so it sounds more foward and prominent....Depending on the depth of your frame, instead of aligning the panel to the front edge of the frame move the panel 1/4inch into the frame so there is a 1/4 inch lip that will act as a guide to focus the sound....The further back the panel is moved the more focused the sound becomes.


I would like to add that its not just one technique that will drastically change the sound of a panel but the combination of many techniques (in my case 6-7 techniques) working together....I see a lot of people get discouraged after one technique did not drastically change the sound and most tend to give up and quit the project.

[Kame2]

One way to decrease tweeter resonance is to use a non polarized electrolytic capacitor (4uf) at around 5khz.

As mentioned I build a crossover. The takeover is at 6 kHz. But it is not an electrical but a mechanical problem. The movement of the panel introduced by the bigger exciter is to heavy for the tweeter. This is happening the same way when I disconnect the tweeter.

Tip/one of my techniques...........DML panels have a very very wide (dispersion) soundstage in which it can sound very diffused or unrealistic....to control dispersion one can use the "FRAME", as sort of like a guide to control and focus the sound so it sounds more foward and prominent....Depending on the depth of your frame, instead of aligning the panel to the front edge of the frame move the panel 1/4inch into the frame so there is a 1/4 inch lip that will act as a guide to focus the sound....The further back the panel is moved the more focused the sound becomes.

Good tip, thanks. At the moment the wide dispersion is a thing I like about the panels. You can walk the room with very small changes in frequency response. I plan to use some outcomes of my experiments for background music in some rooms. In this case the diffuse, backgroundish sound of the panel matches perfectly.
Another use may be for a home cinema system. There it might make sense to focus the sound a little bit to the area where the listeners are. An inward bended panel may be another option to try.

[Bendingwave]

Even with the frame acting as a focus guide its dispersion/soundstage is still remarkable....for the very little dispersion that is lost you gain a more prominent sound, instead of a diffused sound....Of course it all depends on your taste and or particular application.


For me my main goal is to create hi fidelity DML panels for music so accuracy and good all around sound is a must....I also use my panels in a 5.1 set up to watch movies which is secondary and they sound amazing.