A Study of DMLs as a Full Range Speaker

Thanks guys, I just ordered 4 of each of the 32U-4 ultras and 25FHE-4s. The FHE's are on back order so it looks like I'll be playing with the ultras first.

With respect to the placement, thanks for the suggestions. For sure I will try lots of different locations. I've seen the 2/5 recommendation before so I did plan to try that for certain initially.

I suppose that the 2/5 recommendation came from someone's trial and error anyway. But I do find the placement question to be an interesting one. My gut feeling would be that you would want to drive as many of the bending modes of the panel as possible. And in that case the panel corner would likely be the best candidate location. But nobody seems to recommend that, so I must be missing something! But I'll try and find out what..

Eric


You can easily experience it for yourself by moving the exciter around the panel. Play a track with a lot of cymbals high hats so that you can easily distinguish which location gives off the best highs. Then play a track with good solid bass so you can distinguish which location has the best lows.

If you can have someone help you ask that person to hold the panel and move the exciter around while you stand back at least 5ft. to listen which location sounds best. Start with the exciter dead center and then start moving the exciter slowly to the edges or corners to hear which sounds the best for you.

Using a single exciter in the corner is not enough to saturate the whole panel with enough modal vibrations. Highest panel saturation comes from placing the exciter dead center in the middle of the panel.

Imagine throwing a pebble into the middle of a pond and watching the (modal waves) ripples expand to the edges.

With this DML technology "HANDS ON" experience is the best teacher.
 
This is a cheap version of a DML from a cheapo guy. Laugh if ya want but they do sound good! If I can master the bass cancellation from front/back I'll have a nice unit! They sound great outside.. More pics in the album.. ImageShack - Cheap DML Panel Speakers

nLb7XK.jpg

Now this pic basically shows you one way of how to mount said panel material to a frame.

If he had added a spline/brace to hold the exciter in place then it would be more similar to a BMR design like Ive mentioned before.
 
Founder of XSA-Labs
Joined 2012
Paid Member
Nice work, OGKG! I like how you mounted the $1 store foamcore panels to the frame. This should be linked in the Foamcore speaker thread.

These are DMLs so should not have front/back bass cancellation. Maybe you are referring to back wall reflection bass cancellation? Like when you out the frames against a wall?

Try adding dampening material like a layer of thick felt or fiberglass on the back side of the frame. The best stuff to use would be the melamine sponges. $6 for a bag of 100 from China. Those are same material as BASF Basotect sound abatement panels. They will crush the back reflection and give you better sound from the front.
 
You can easily experience it for yourself by moving the exciter around the panel.

It seems that some panel types would be more amenable to re-positioning the exciters, depending on the strength of the adhesive and the strength of the panel surface, right? I was thinking I'd start with the 1" thick pink PS foam for my panels, but I can imagine that pulling the exciters off to move them might rip out chunks of foam. Or does applying the glue/water seal on the surface fix that issue?

How about the foamcore stuff like OGKG is using? Does the paper tear when you pull the exciter off to move it? Or is there a way to prevent that?

Gatorboard has a finished wood veneer on the surface that seems like it would be amenable to re-positioning the exciter without damaging the panels. Is that correct?

Is also seems like plywood like Luan or Birch with a finish applied would also allow easy re-positioning of the exciters.

What panels and or panel treatments help to allow moving the exciters without damaging the panels (and/or fouling the adhesive mount)? Or am I trying to solve a problem that really isn't a problem?

Eric
 
I have used a wallpaper knife to ease in between the exciter drive surface and the panel. It works for ply or other hard surface materials. I have not used foam yet. You can re-stick the exciter using more DS tape or contact adhesive which works fine.

As you are using EPS/XPS foam steer clear of solvents, e.g. lighter fluid. These will dissolve the foam very efficiently leaving you a sticky mess.You can use solvent with ply panels.
 
Mother nature probably knows best: golden ratio

....

With respect to the placement, thanks for the suggestions. For sure I will try lots of different locations. I've seen the 2/5 recommendation before so I did plan to try that for certain initially.

I suppose that the 2/5 recommendation came from someone's trial and error anyway. But I do find the placement question to be an interesting one. My gut feeling would be that you would want to drive as many of the bending modes of the panel as possible. And in that case the panel corner would likely be the best candidate location. But nobody seems to recommend that, so I must be missing something! But I'll try and find out what..

Eric
Please try 1 : Phi or 1 : ~1.608 or about 5 : 8
This is the golden ratio which you can find everywhere in nature. It is far more then a ratio for beauty; it also gives the least resonances.
 
Nice work, OGKG! I like how you mounted the $1 store foamcore panels to the frame. This should be linked in the Foamcore speaker thread.

These are DMLs so should not have front/back bass cancellation. Maybe you are referring to back wall reflection bass cancellation? Like when you out the frames against a wall?

Try adding dampening material like a layer of thick felt or fiberglass on the back side of the frame. The best stuff to use would be the melamine sponges. $6 for a bag of 100 from China. Those are same material as BASF Basotect sound abatement panels. They will crush the back reflection and give you better sound from the front.

"IF" the sound from both sides doesn't cancel, then why not use both surfaces? (The Hillbilly is brainstormin' again..) ;)
 
It seems that some panel types would be more amenable to re-positioning the exciters, depending on the strength of the adhesive and the strength of the panel surface, right? I was thinking I'd start with the 1" thick pink PS foam for my panels, but I can imagine that pulling the exciters off to move them might rip out chunks of foam. Or does applying the glue/water seal on the surface fix that issue?

How about the foamcore stuff like OGKG is using? Does the paper tear when you pull the exciter off to move it? Or is there a way to prevent that?

Gatorboard has a finished wood veneer on the surface that seems like it would be amenable to re-positioning the exciter without damaging the panels. Is that correct?

Is also seems like plywood like Luan or Birch with a finish applied would also allow easy re-positioning of the exciters.

What panels and or panel treatments help to allow moving the exciters without damaging the panels (and/or fouling the adhesive mount)? Or am I trying to solve a problem that really isn't a problem?

Eric

No, I am not saying to stick/adhere it to the panel yet, that is why I suggested someone helping you hold the exciter to the panel and move it around until you find the best location.

Adhering the exciter to the panel should only be done when you find the right location that your ears prefer.

When removing the exciter after adhering it to the panel you can use a thin type of tool be it a knife , xacto blade etc to peel off the 3m tape.

When your design is finalized you can permanently adhere the exciter to said panel material by using 2 part epoxy glue. (if there is 3m tape on the exciter make sure to peel it off first.)

Once epoxy glue is used though its very hard to take off without damaging either the exciter or panel material or both.
 
If your room is big enough you can use the rear radiation positively. The soundstage becomes immense if you can free stand/ hang them a couple of metres from the back wall. You can literally walk into the soundstage. If you have to place them close to a back wall sound absorbing material helps a lot. I found I had an annoying colouration from the rear of the panel with mine six inches from the wall which cleaned up nicely with a 35mm layer of insulation material, the sort you use inside cabinets.
 
I've used low temperature hot glue in place of the sticky tape. It's possible to get the hot glue soft enough to let go with a blow drier and some patience. After the driver is out of the way, the remaining hot glue cleans up fairly easy with a little heat put to it. I've removed drivers from the foam sheets without damage. Just take it slow..
 
It’s cancelling with its own reflection from the back wall - not with the sound from the front. If you pull it from the wall, you will notice that the sound is closer to Omni vs dipole pattern.

I've hung them from the ceiling out in the open and both sides make decent sound. Actually, the side the driver is hung on seems to sound a bit better than the intended listening side! Has anyone else noticed that? Of course, my experiments were on the paper covered foam sheets but it tells me the driver uses the paper on the mount side more so than the paper on the opposite side..

Does anyone remember my experiments with the bottom of a beer can? It has merit. It did connect both surfaces to the driver without the need for the sound to travel through the foam and made the treble more.. "Clairfied", per se? Perhaps aluminum isn't the best material to transfer treble but it did tell me something that I plan to work with more. This adds some weight to the voice coil movement and can maybe be "Tuned" like a passive radiator? I did add clay balls in the concave part of the can bottom and the base response changed..

I don't have all of the fancy equipment that many of you do so here's an avenue to go down. I wish someone would just to tell me I'm silly..

ImageShack - DML Super Tweeter
 
Thanks, X..
Yes it did! Actually, the SPL was higher from the rear.. LOL.. Concave.. Convex.. Dunno why but..

I may try some copper or brass. Maybe have a dimple in the middle of 2 flat pieces so to connect the two, front and rear. A quick solder job can achieve that..

I've also pondered two exciters, face to face, on opposite sides, reverse polarity so to work with each other.. That may cure the "One side sounding better than the other" dilemma. Would series or parallel matter in this case? Who knows.. We know what the effect is with standard drivers. Maybe just the impedance difference? Would the SPL increase by a parallel connection? The world's gonna know someday..

I do know if an enclosure is used with a DML, considerable bass can be achieved. From what I've learned from my experiments, I have a design in my head that is totally different from the normal. I haven't done a prototype yet but it's gonna happen..

Yeah.. It's about time for me to get the training wheels off and become more scientific. That'll give me a new function for my padded room.. ;)
 
I do know if an enclosure is used with a DML, considerable bass can be achieved.

Longer panel length, even despite narrow width, increases bass. Multiple exciters increases bass. Splines that cause the panel to move like a piston -- balanced mode radiator -- at lower frequencies also increases bass. I'm not sure enclosures increase bass.

In addition, you don't necessarily want to increase bass in your midrange and high frequency panels. My balanced DMLs generate low frequencies but it sounds terrible because they're located in a small room. Modal behavior takes over at low frequencies.

You're better off separating your bass sources from higher frequency sources. If you want to use a DML for bass design it(them) for bass and nothing else. If you want to use traditional subwoofers you can use regular subs.

DMLs have advantageous characteristics in mids and highs. Those advantages don't apply at low frequencies because you're no longer dealing with reflections/ray acoustics.
 

Attachments

  • Chebychev - Smoothed.JPG
    Chebychev - Smoothed.JPG
    556.9 KB · Views: 428