Leob,
All what I share at the moment is oriented for my understanding of how DML work and how to get the best results for home use. I am not thinking really about power handling. The power requirement should make appear new constraint like the needed displacement according to the lower frequency to reach, how the material can accept it and also how the voice coil behave under such strength (there is al least on paper about the compression of the voice coil)... many additional things.
I don't remember if it was mentioned before but one interesting characteristic of DML is the level decreases with the distance more slowly than with a cone... at least in a living room. I have line arrays and it is about the same : -3db when doubling the distance compare to -6dB. Probably DML are better to that than line array because there is not the problem of comb filtering, interference because of the distance between the cones. the condition is probably to have only one panel with several exciters close to each other.
Below : SPL vs distance (up to 2m) yellow and green are standard loudspeaker (FRS8). the other are panels.
Christian
Eric,
From my tests with the different materials and their efficiency, I have an heuristic where efficiency = 5*log10(D/µ³) + 83 (nothing shows for now it is usable!)
When I test it with the data from your CF/balsa, it gives about 6 dB in favor of the CF/balsa compare to plywood so close to your 7dB observation which ok.
To have the balsa in between, the Young modulus of the balsa plywood should be 200 to 400MPa. Does this value make sens for you?
Christian
Yes, it will be a lot of testing to see how materials fare when it comes to power handling, but your data is still very helpful...thanks for sharing!
Can I ask, what signal did you use for that graph?
It makes sense that if you measure right in front of a speaker, the SPL is a result of the travel of the cone or panel. So if the cone moves 1mm at 1khz and the panel does the same, the SPL will never be higher on the panel due to a larger surface. But even if the waves are not stronger there will be more of them, resulting in more air moved even if the compression of the waves are the same.
The drop-off at 2 meters looks like around 8dB for XPS20 and 13dB for the FRS8. 5dB is a massive difference over a still relatively short distance!
Outdoors with a big area with many people, the effect visible in that graph will probably be multiplied many times.
I'm also interested in DML for home use, but I guess it can hard to quantify the effect of reducing the panel size. There I care more about perhaps relatively small differences in the perception than the loudness. If one want to feel like one is bathing in sound, perhaps a larger panel will help? Until I get some XPS to do my own experiments I'm just speculating, but my guess is that the smaller you go the more like a cone it will sound? My 500x400 PC panels certainly sound larger than my 300x300 printed, even if the printed sounds tighter and louder, but of course that is probably mostly due to the materials.
What is quantifiable and seems clear from your graphs over different size panels is that the small panels has a more jagged FR, so one would think they would sound a bit worse.
Concerning re-use of exciters:
I have been using the shiny metallic tape used for duct work.
1. Apply a thicker coat of PVA on the panel where you decide to place the exciter.
2. Take a piece of the tape large enough to accommodate the exciter but make it a bit longer to fold over on itself a tiny bit along one edge for help in removal later. (You can also stick this piece of tape to your pants leg a time or two to lessen the stickyness a bit which will help with removal. Test for yourself to see what works best for you on some scrap first because not all glues are made equal.)
3. Place the tape on the thick PVA location and burnish it secure.
4. Cut yourself a donut out of same type of tape and apply to the exciter mounting area burnishing it as well.
5. I have been using just a thin film of the 5 min. epoxy to adhere the two.
6. After testing, get under that flap you created on the panel tape and carefully peel it away from the PVA . I use a dull, rounded end exacto knife. Peel tapes from exciter and you're ready to move on.
I've done this many times with only one incident where I had sanded my EPS panel super thin at the exciter placement point and it broke the EPS during extraction. Hopes this helps.
I have been using the shiny metallic tape used for duct work.
1. Apply a thicker coat of PVA on the panel where you decide to place the exciter.
2. Take a piece of the tape large enough to accommodate the exciter but make it a bit longer to fold over on itself a tiny bit along one edge for help in removal later. (You can also stick this piece of tape to your pants leg a time or two to lessen the stickyness a bit which will help with removal. Test for yourself to see what works best for you on some scrap first because not all glues are made equal.)
3. Place the tape on the thick PVA location and burnish it secure.
4. Cut yourself a donut out of same type of tape and apply to the exciter mounting area burnishing it as well.
5. I have been using just a thin film of the 5 min. epoxy to adhere the two.
6. After testing, get under that flap you created on the panel tape and carefully peel it away from the PVA . I use a dull, rounded end exacto knife. Peel tapes from exciter and you're ready to move on.
I've done this many times with only one incident where I had sanded my EPS panel super thin at the exciter placement point and it broke the EPS during extraction. Hopes this helps.
Christian.
I have posted many pictures of the hump in the 10k to 20k on various ply panels over the years.
even when I used the strip of paper to tame this response in the coil area ,the edges of the panel were still radiating the 10k to 20k hump ?
with a hard 4 foot materials such as ply the hf seems to travel longer distance through the material to the edge of the panel so that the whole panel is radiating up to 20k
a four foot 1mm card panel would have no hope of doing this.
These are two extremes and are easy to measure.
Steve.
I have posted many pictures of the hump in the 10k to 20k on various ply panels over the years.
even when I used the strip of paper to tame this response in the coil area ,the edges of the panel were still radiating the 10k to 20k hump ?
with a hard 4 foot materials such as ply the hf seems to travel longer distance through the material to the edge of the panel so that the whole panel is radiating up to 20k
a four foot 1mm card panel would have no hope of doing this.
These are two extremes and are easy to measure.
Steve.
KoAP .
Thanks for the tip.
I was thinking of using the tech ingredients method of applying the epoxy and then laying the cloth onto the epoxy ,they say this prevents bubbles ?
This is for the 5mm xps.
Also I need the thinnest cloth ?
Do you have any idea what grade or name this would have in the UK ?
it difficult to know what to order, Unless you know what you are looking for ?
I don't want to make the panel too rigid, just with the epoxy seems about fine.
But until I try it , I won't be sure ?
Steve.
Thanks for the tip.
I was thinking of using the tech ingredients method of applying the epoxy and then laying the cloth onto the epoxy ,they say this prevents bubbles ?
This is for the 5mm xps.
Also I need the thinnest cloth ?
Do you have any idea what grade or name this would have in the UK ?
it difficult to know what to order, Unless you know what you are looking for ?
I don't want to make the panel too rigid, just with the epoxy seems about fine.
But until I try it , I won't be sure ?
Steve.
Christian.
I have not read the paper on voice coil compression, but I can imagine this being part of the problems I can hear.
today I was just thinking about very rigid panels and the hf responses I have seen, when it occurred to me that maybe they use the soft tacky coil rings to try and damp out this problem ?
Although this doesn't help with panels that do not suffer from this problem ?
Voice coils on cone drivers are designed to move a light cone back and forth ,if pushed too hard or excessive use of EQ, the coils can come loose( I know this from personal experience).
But when a coil former is attached to a 5mm or 10mm hard panel such as ply, the coil former is literally banging on the panel, similar to, if you were to bang your head up against a brick wall !!
I noticed this year's ago but never really thought much about it, other than I shouldn't push my exciters too hard as they were starting to sound distressed.
Probably the former and coils were starting to buckle ?
This was on a very percussive track with lots of banging , if I remember correctly.
Steve.
I have not read the paper on voice coil compression, but I can imagine this being part of the problems I can hear.
today I was just thinking about very rigid panels and the hf responses I have seen, when it occurred to me that maybe they use the soft tacky coil rings to try and damp out this problem ?
Although this doesn't help with panels that do not suffer from this problem ?
Voice coils on cone drivers are designed to move a light cone back and forth ,if pushed too hard or excessive use of EQ, the coils can come loose( I know this from personal experience).
But when a coil former is attached to a 5mm or 10mm hard panel such as ply, the coil former is literally banging on the panel, similar to, if you were to bang your head up against a brick wall !!
I noticed this year's ago but never really thought much about it, other than I shouldn't push my exciters too hard as they were starting to sound distressed.
Probably the former and coils were starting to buckle ?
This was on a very percussive track with lots of banging , if I remember correctly.
Steve.
Leob,
Those tests were done with a pink noise probably (sorry I didn't keep note of that) with a limitation in the low frequencies to limit the possible effect of the different loudspeakers bandwith or of the room modes.
The difference in the drop-off comes from the different nature of the loudspeakers. A DML behaves more like a line array. It is interesting because for a given listening level, a lower power is needed.
I should have perhaps made a measurement at 3m in addition (this living room is relatively small) to show the level in far field.
I have no experience of outdoor acoustics but my understanding of those curves is they are linked to how the pressure wave expand (spheric expansion for a true point source) this linked to the nature of the source. No link with the space or how it is occupied? This should be checked...
About the dimension of the panel, a bigger panel for a given material, a given thickness, will have a lower first mode so a higher mode density after which is good for the smoothness of the FR. With a bigger panel, the waves travel for a longer distance so get a higher attenuation before reflection. It might help in the smoothness. I have no evidence of that.
But I think the first main effect of the dimensions is on the bandwidth in the low frequencies which is visible on the plots. The smallest panel has the shortest bandwidth. For now, I have no good idea of the mechanism behind.
Christian
Sounds like a good idea to make your own double side tape using the quality of each glue. Smart.
Christian
Steve,
I would suggest either of the thinnest 2 fabrics here, that is, the 25 gram or 100 gram fabrics. They are not expensive, so may you could try both. But if you choose to try ony one, I would most suggest the 25 gram fabric.
https://www.easycomposites.co.uk/glass-cloth
You may be surprised at just how thin and light the 25 gram fabric is. It will be very light and delicate. The 100 gram fabric will be more substantial for handling purposes, but might add too much stiffness to you 5 mm PS foam.
I have never purchased from this supplier, but their video tutorials are excellent, so I assume their products are equially good. You may want to check some of them out.
I assume you plan to apply the fiberglass/epoxy composite to both sides of the panel, right?
Some tips:
- Safety: epoxy resins create harmful fumes, at the very least, be sure to work in a well ventilated space, or better yet use a mask with cartridge filters desined for this purpose.
- Probably you should sand any skin off the PS foam before applying the fg/epoxy. Epoxy generally does not bond well to plastics, but if it can get into the pores of the foam it should be able to key into the pores and stick well.
- When you have applied enough epoxy, the fabric will become transparent. Adding additional epoxy only adds unnecessary weight. Any less epoxy and the composite will not work as it should.
- Before cutting the fabric, apply a strip of masking tape around the perimeter that you plan to cut, and cut down the center of the tape. Make the fabric piece larger than the panel, and cut off the extra only after the epoxy has cured. The tape will keep the edges of the fabric from fraying, and reduce the amount of stray fibers that get airborne.
But one question. What do you hope to achieve? To me, the main benefit of a composite with a light core and stiff skins is to provide high efficiency. But PS foam already does that quite well alone. Is there some other benefit you hope to get from the fg/epoxy skins? Or are you just curious to see what happens? Oops, I guess that was three questions...
Eric
Eric.
Thank you for the great link and the tips.
Your safety advice makes me think I will wait until the weather conditions improve , it's far too cold at the moment for proper ventilation 😬
The epoxy sticks to the xps skin with no problems, I did think about sanding the xps at the time but thought the epoxy would soak into the foam and absorb more epoxy, more weight ?
While I was looking for some old 5mm xps responses, I came across .I post 3944 on page 198.
This shows the 7inch veneer responses ,very flat above 1.5k to at least 20k .
Applying the epoxy to this panel has created problems that were not there before, I believe pva would have been a far better option for this panel, even though the weights do sort out the response plot , the problems still exist.
It's a bit of a nightmare trying to find old plots , it was great over on audiocircle with my gallery, which held all my plot and photos in one place, for easy reference.
I believe the 5mm xps and probably other thicknesses benefits from a harder skin, the sound comes more alive and has a far better high end response.
The very low end is not important to me as I use the low frequency driver to support the panel below 300hz.
As long as the panel I am using (whatever size) shows no signs of distress I run the panel down as low as possible.
I can adjust the ox while playing music , and if I feel the panel is still contributing to the performance I will go all the way down to 60hz.
But only if the panel can handle the worst I can throw at it !
And to clarify your last THREE questions.
when Adding heavier skins to an xps or even an eps panel, I would expect a reduction in efficiency.
I presume you ment high efficiency ,not hf efficiency ?
what I hoped to achieve was an improvement in the sound of the xps panel.
Xps might be more efficient on its own, but the sound and response is not good enough.
It shows signs of early hf roll off and is over damping itself.
The stiff skins ,as you say, sort out these problems giving a more lively sound and higher HF extension.
The slight drop in efficiency is of no concern to me as my 10watt exciters still have no problems driving the extra weight.
In my opinion the epoxy and xps go well together ,more so than the pva , it's probably just a question of getting the size and thickness correct?
And sorting out the exciter panel interface ?
Steve
Thank you for the great link and the tips.
Your safety advice makes me think I will wait until the weather conditions improve , it's far too cold at the moment for proper ventilation 😬
The epoxy sticks to the xps skin with no problems, I did think about sanding the xps at the time but thought the epoxy would soak into the foam and absorb more epoxy, more weight ?
While I was looking for some old 5mm xps responses, I came across .I post 3944 on page 198.
This shows the 7inch veneer responses ,very flat above 1.5k to at least 20k .
Applying the epoxy to this panel has created problems that were not there before, I believe pva would have been a far better option for this panel, even though the weights do sort out the response plot , the problems still exist.
It's a bit of a nightmare trying to find old plots , it was great over on audiocircle with my gallery, which held all my plot and photos in one place, for easy reference.
I believe the 5mm xps and probably other thicknesses benefits from a harder skin, the sound comes more alive and has a far better high end response.
The very low end is not important to me as I use the low frequency driver to support the panel below 300hz.
As long as the panel I am using (whatever size) shows no signs of distress I run the panel down as low as possible.
I can adjust the ox while playing music , and if I feel the panel is still contributing to the performance I will go all the way down to 60hz.
But only if the panel can handle the worst I can throw at it !
And to clarify your last THREE questions.
when Adding heavier skins to an xps or even an eps panel, I would expect a reduction in efficiency.
I presume you ment high efficiency ,not hf efficiency ?
what I hoped to achieve was an improvement in the sound of the xps panel.
Xps might be more efficient on its own, but the sound and response is not good enough.
It shows signs of early hf roll off and is over damping itself.
The stiff skins ,as you say, sort out these problems giving a more lively sound and higher HF extension.
The slight drop in efficiency is of no concern to me as my 10watt exciters still have no problems driving the extra weight.
In my opinion the epoxy and xps go well together ,more so than the pva , it's probably just a question of getting the size and thickness correct?
And sorting out the exciter panel interface ?
Steve
Thank you Eric for all those advice. There is experience behind!
For Europe (sorry Steve!) : here is the link www.easycomposites.eu
Eric,
Composite is a completely new domain for me but it seems really promising for DML: controlled stiffness, light material... So some questions :
- Is there an (affordable) alternative to epoxy with lower risk for the health?
- In the process you described here, is there a need for a vacuum bag as you used for the CF/balsa panels?
In parallel to that I have doubts the PS (EPS or XPS) to be the right material for the core. Its Young (tensile) modulus is low so the bending wave speed will be limited (see in post above 4613 p231 and 4632 p232). In post 4610 p231, I mentionned the Rohacell which is the material listed in M Azima's patent. It is among the core material proposed in by easycomposites. Do you have any opinion/experience with it?
Christian
@Veleric
Eric,
Some posts ago, we have spoken about an unused data from the Tectonic panels : the efficiency. With the heuristic proposed in post 4682 (efficiency = 83 + 5.log10(D/µ³) ), 91dB is a D/µ³=40
So to keep simple values : µ=0.5kg/m², D=10N.m give fc = 5800Hz, A.fo = 8Hz.m², D/µ³=40. For a 378x478mm area, f0 = 44Hz and 90g for one membrane and if I remember, the thickness is said to be 3mm in a video.
Christian
Eric,
Some posts ago, we have spoken about an unused data from the Tectonic panels : the efficiency. With the heuristic proposed in post 4682 (efficiency = 83 + 5.log10(D/µ³) ), 91dB is a D/µ³=40
So to keep simple values : µ=0.5kg/m², D=10N.m give fc = 5800Hz, A.fo = 8Hz.m², D/µ³=40. For a 378x478mm area, f0 = 44Hz and 90g for one membrane and if I remember, the thickness is said to be 3mm in a video.
Christian
Possibly dumb question (and maybe already addressed, but I'm not up for sifting through several hundred pages of this thread):
It appears that the panels used for DML speakers are flat -- has anyone looked into curved (particularly dome shaped) panels of some kind of sandwich construction? It seems like that would be better for stiffness, assuming that max stiffness is what is wanted.
It appears that the panels used for DML speakers are flat -- has anyone looked into curved (particularly dome shaped) panels of some kind of sandwich construction? It seems like that would be better for stiffness, assuming that max stiffness is what is wanted.
Yes hundred of pages and years also! So difficult to extract a synthesis but an incredible sound scene quality to complexity ratio. The technical field addressed is completely different to what I know from more standard loudspeaker.
Max of stiffness probably not exactly but more as light as possible with the right stiffness. The 2 leading characteristics are the stiffness (bending stiffness) and the aerial density. For a given aerial density, if the stiffness becomes to high, the area to reach the targeted low frequency becomes to important.
I don't have in mind realization or paper about curvature in one or more directions.
As there is still possibilities with flat membranes like for example wood or composite or even other material, a curved surface sounds to me like adding building complexity without element to understand the benefit. I even don't know the consequence on the sound wave.
But you are right, using the shape was used in the Walsh driver for example with a very thin conic membrane which. It is also a bending wave loudspeaker.
Did a little more experimenting 3d printing trying the same material in same size, but with different density and thickness.
Nothing conclusive...have a measuring mic on the way and will see if it tells me anything I don't hear already. The PLA+ is bad...if I make it really thin it sounds a bit weaker and less defined, and if I make a thicker plate with more density it sounds a bit tighter but also starts getting a tone of its own. Not the same kind of dampable ring as I had with the polycarbonate though.
Tried a slightly different material, the Extrudr BioFusion which basically is a "Silk PLA", a PLA with added copolymers.
For both materials being PLA variations the difference is surprising. Silk PLA is mostly used for its shiny visual quality, and specs are similar to regular PLA, but known to be a bit brittle in comparison. However it actually has noticeably better HF response, so might be some hope to find a decent material.
Also been reading up on wet layup with vacuum bag. It has startup cost that is a bit off-putting with pump and additional materials, but when set up the actual expoxy, carbon and nomex would not be that much per panel, and obviously it would give better results that anything I can print.
Been using PVA based glue stick to fix the exciters, and been holding my hefty 200g exciters without problem, but today I got the idea to try hairspray. Have a bottle standing here since I tried using it as adhesive for 3d printing. It is also PVA based, and seemed like it could be a nice quickfix option.
Sprayed a thin layer, let it quickly dry so it was a little sticky and placed exciter. Did it a couple of hours ago, and have it hanging without support playing music already and seems stuck fine. Well see how it holds up, and if I can remove it later, but for a quick test it is excellent it seems.
Nothing conclusive...have a measuring mic on the way and will see if it tells me anything I don't hear already. The PLA+ is bad...if I make it really thin it sounds a bit weaker and less defined, and if I make a thicker plate with more density it sounds a bit tighter but also starts getting a tone of its own. Not the same kind of dampable ring as I had with the polycarbonate though.
Tried a slightly different material, the Extrudr BioFusion which basically is a "Silk PLA", a PLA with added copolymers.
For both materials being PLA variations the difference is surprising. Silk PLA is mostly used for its shiny visual quality, and specs are similar to regular PLA, but known to be a bit brittle in comparison. However it actually has noticeably better HF response, so might be some hope to find a decent material.
Also been reading up on wet layup with vacuum bag. It has startup cost that is a bit off-putting with pump and additional materials, but when set up the actual expoxy, carbon and nomex would not be that much per panel, and obviously it would give better results that anything I can print.
Been using PVA based glue stick to fix the exciters, and been holding my hefty 200g exciters without problem, but today I got the idea to try hairspray. Have a bottle standing here since I tried using it as adhesive for 3d printing. It is also PVA based, and seemed like it could be a nice quickfix option.
Sprayed a thin layer, let it quickly dry so it was a little sticky and placed exciter. Did it a couple of hours ago, and have it hanging without support playing music already and seems stuck fine. Well see how it holds up, and if I can remove it later, but for a quick test it is excellent it seems.