Thank you for the information. There are so many tests to do to that if it is possible to access to a at least nearly satisfying not too expensive common material it might be a good point. For now I am doing some basic tests with some XPS 9mm (depron?) for which I have doubt it can make a not too bad panel.
Hello Sarathssca,
Your 2 choices sounds good to me as among the 4 types of panels I tried they are the 2 really working. My preference is for plywood. This the one currently working now. The membrane is 420x1200 3mm poplar plywood with a 17x17mm foam suspension all around (4 sides). My choice of foam goes for the biggest section available in the DIY store; It is also low density in the idea to have a minimum impact on the panel. I don't know really about that but the panel is really flat even if it warped after the varnish layer.
The choice in those solutions dependent also the money and mainly the time you want to spend on this.
The canvas is quite easy to do. It will work quite nicely with a low effort (a week end DIY?)
The plywood needs more work. For me it gives a sound I have never had with my other DIY speakers.
For the Canvas, I just added a base made of some particle board I had. they are for now on the top of 3 way columns not more used but still in the living room.
Your question about the stiffness is good. I am in the opinion there is a risk for long term. The wood keys at the back of the canvas in the corner allow a tension adjustment. The question becomes more tricky is the exciter is mounted on a spine. How to insure there is no lateral effort on the voice coil?
If you go for the plywood solution with as spine (which I recommend), I think I made a mistake in my design using a 10x10cm piece to maintain the exciter (you should find pictures in post before). This amplifies (a kind of horn) the sound emitted between the panel and the exciter (spider array). The spine has to have a minimum surface parallel to the panel.
Depending how you design it, the plywood might also offer the possibility to make a nice object, a nice styling if you like wood.
Does a nice loudspeaker sound better?
Make your choice, start... and enjoy the sound of DML.
Keep us inform or ask for other questions.
Christian
Sarath.
While the eps was drying, I thought I would try a way of improving the art panel .
In the first picture I inserted a strip of wood in the top and bottom to reduce the size of the open canvas.
This reduces the area so that the inner ply panel is 10cm from the outer edges.
The second picture is the plot without the strips of wood at 3m distance
The third is with the strips of wood, note the higher bass peak, but the panel handles this better .
The problem is the suckout in the 100hz to 300hz.
I was wondering if a larger ply panel was used , one that reaches down to 100hz before attaching to the canvas ?
This is getting very close to a ply panel with a roll surround ?
The fourth picture is of the panel with the strips at a distance of 30cm.
The fifth picture is of my low grade 70grade eps panel at 30cm ,it looks like the glue has dried some more , as the peak at about 13k is now much stronger.
I will be dealing with this problem later.
It does not show a suck out problem at 30cm but at 3m it will.
Steve.
While the eps was drying, I thought I would try a way of improving the art panel .
In the first picture I inserted a strip of wood in the top and bottom to reduce the size of the open canvas.
This reduces the area so that the inner ply panel is 10cm from the outer edges.
The second picture is the plot without the strips of wood at 3m distance
The third is with the strips of wood, note the higher bass peak, but the panel handles this better .
The problem is the suckout in the 100hz to 300hz.
I was wondering if a larger ply panel was used , one that reaches down to 100hz before attaching to the canvas ?
This is getting very close to a ply panel with a roll surround ?
The fourth picture is of the panel with the strips at a distance of 30cm.
The fifth picture is of my low grade 70grade eps panel at 30cm ,it looks like the glue has dried some more , as the peak at about 13k is now much stronger.
I will be dealing with this problem later.
It does not show a suck out problem at 30cm but at 3m it will.
Steve.
Attachments
As can be seen in my plots, poor as they may be , with my exciters and choice of panel materials, 20k is easily attainable.
Maybe now you can understand my dismay when I see panel responses with heavy roll off starting at 2 or 3k ?
The low end is the major problem if you are trying to run a dml full range.
At 30cm you will most always get a good response down to about 40hz (except the ply canvas panel it would seem).
But as you move out into the room that dreaded hole starts to appear, this is also a problem with my transmission line speakers, so it is not just a dml problem.
But combining the TLS and the dml panel in this region , fills the hole ,similar to using multiple subs for the same reason.
An ordinary sub which rolls off at 50hz or 100hz would not solve this problem, still leaving a hole somewhere between 100hz and 300hz (depending on the dml panel ).
Which I believe is the reason why some complain about the hollow sound dml can produce.
many Years ago now, when I moved my 6ft panels into my main room ,they just sounded terrible ,I moved them all over the place but could not get a good sound.
I then, with heavy heart , went into the other room to make a cup of tea and console myself.
As I walked into the next room ,the sound went ,bang , it sounded fantastic, so I rushed back in ,only to hear the same horrible sound !!
When I was standing in the next room ,it sounded like I had a full jazz band playing live in the next room, but as soon as I walked back into the room, the in room suck outs destroyed the illusion immediately.
I was gobsmacked !
It was a weird experience ,standing in the doorway trying not to get too close to the in room problems.
This was in the early days of dml for me, and the thought that I would have to listen to them in the next room for the best sound, for the rest of my life was a bit daunting , to say the least !
Steve.
Maybe now you can understand my dismay when I see panel responses with heavy roll off starting at 2 or 3k ?
The low end is the major problem if you are trying to run a dml full range.
At 30cm you will most always get a good response down to about 40hz (except the ply canvas panel it would seem).
But as you move out into the room that dreaded hole starts to appear, this is also a problem with my transmission line speakers, so it is not just a dml problem.
But combining the TLS and the dml panel in this region , fills the hole ,similar to using multiple subs for the same reason.
An ordinary sub which rolls off at 50hz or 100hz would not solve this problem, still leaving a hole somewhere between 100hz and 300hz (depending on the dml panel ).
Which I believe is the reason why some complain about the hollow sound dml can produce.
many Years ago now, when I moved my 6ft panels into my main room ,they just sounded terrible ,I moved them all over the place but could not get a good sound.
I then, with heavy heart , went into the other room to make a cup of tea and console myself.
As I walked into the next room ,the sound went ,bang , it sounded fantastic, so I rushed back in ,only to hear the same horrible sound !!
When I was standing in the next room ,it sounded like I had a full jazz band playing live in the next room, but as soon as I walked back into the room, the in room suck outs destroyed the illusion immediately.
I was gobsmacked !
It was a weird experience ,standing in the doorway trying not to get too close to the in room problems.
This was in the early days of dml for me, and the thought that I would have to listen to them in the next room for the best sound, for the rest of my life was a bit daunting , to say the least !
Steve.
My thin carbon fiber and foam sandwich board arrived finally...
I was concerned at first because it seemed heavier than expected, mostly because the foam, even though it's only 2 mm is very hard and dense. I suspect it soaked up a good amount of resin in formation, I can't even dent it with my fingernail. I sanded the glossy side a bit and hooked up an exciter...Wow.
I can only compare it to the various incidental objects I tried before (with the thin wood panel stud wall previously giving the best sound...and the two pairs of relatively inexpensive speakers I have here.
I'm blown away. I might leave out the subwoofer and use it in the downstairs setup instead because I don't seem to need it. The sound is fantastic. This is first try, one exciter so far (still planning to add second) but I don't know if my ears would be able to easily detect an improvement from here because it just sounds that good.
I was concerned at first because it seemed heavier than expected, mostly because the foam, even though it's only 2 mm is very hard and dense. I suspect it soaked up a good amount of resin in formation, I can't even dent it with my fingernail. I sanded the glossy side a bit and hooked up an exciter...Wow.
I can only compare it to the various incidental objects I tried before (with the thin wood panel stud wall previously giving the best sound...and the two pairs of relatively inexpensive speakers I have here.
I'm blown away. I might leave out the subwoofer and use it in the downstairs setup instead because I don't seem to need it. The sound is fantastic. This is first try, one exciter so far (still planning to add second) but I don't know if my ears would be able to easily detect an improvement from here because it just sounds that good.
Ugh, well it seems I got lucky on my first listen and now I'm having a hard time reproducing it. I was holding the panel at middle top between two fingers and standing near a corner next to my TV flipping wires...when I suspend it from string it sounds like crap. Leaning it against wall is a bit better but nothing reproduces holding it at arm's length....so far. Looks like I have a lot of testing to do after all. Right now I have the right panel on the ground leaning against the wall and left hanging and the difference is profound. I should switch to some mono music just for better analysis
There is a distinctly loud mode around 85 Hertz. Bass response mostly disappears around 65 Hertz so that's a crazy dropoff. A sine sweep produces a strong inaudible pit of the stomach sort of vibration somewhere in the low 20s which I suspect is that big flat piston at work. Some songs sound great because they just happen to play well with the varying frequency response and others not so much. My edges definitely need to be constrained, maybe all 4
NaRenaud
Let me comment. Hopefully it will help you... at least it help me in sorting what I know (or think to know) from what i don't know!
It is not a surprise that the sound of a panel has a drop off even disappeared in the bass. DML are based on resonances so below the lowest one don't expect much things. The modes above the first seem also inefficient in an acoustic point of view because of their symmetry.
About the difference between the sound the panel at hand and in position with strings :
- the bass level decreases when you increase the distance from the panel. I haven't seem for now a full explanation of this but I currently makes some measurement at 0.25, 0.5, 1m and 2m showing it clearly. So don't judge a panel let say below 1m or better at your listening position. To say the minimal distance is linked to the panel dimensions make sense. To be proven. In the medium, some hump are also reinforced at distance.
- the way the panel is suspended is critical in its capacity to have resonances. Your hands on the sides modified those conditions.
- panel dimensions, weight, strings positions
- frequency response
When I sanded the 70grade panel it became floppy like the lower grade panel, so the rigidity is in the skin.
The lower grade had also suffered some damage that made the panel buzz , probably hair line cracks.
Now that I have coated both panels in a 50x50 mix of pva the rigidity has returned and The hairline cracks have been filled in the lower grade panel, sorted !
The low grade which I did not sand is now as rigid as the higher grade, but much lighter.
I did think I would need to do a second coat ,but it seems to be doing fine ,so will leave for now.
Both now have a similar performance ,peaks are gone, and both extend to 20k, and the sound is very good.
This is with poor quality packing grade eps, with holes in the surface.
I do have some larger low grade panels I saved , some time ago ,these are in the loft, and these are higher quality without holes,
I shall be playing different types of music for a while ,to see if I can pick holes in them 😀 boom, boom.
But I can't fault them , so far !
Steve.
The lower grade had also suffered some damage that made the panel buzz , probably hair line cracks.
Now that I have coated both panels in a 50x50 mix of pva the rigidity has returned and The hairline cracks have been filled in the lower grade panel, sorted !
The low grade which I did not sand is now as rigid as the higher grade, but much lighter.
I did think I would need to do a second coat ,but it seems to be doing fine ,so will leave for now.
Both now have a similar performance ,peaks are gone, and both extend to 20k, and the sound is very good.
This is with poor quality packing grade eps, with holes in the surface.
I do have some larger low grade panels I saved , some time ago ,these are in the loft, and these are higher quality without holes,
I shall be playing different types of music for a while ,to see if I can pick holes in them 😀 boom, boom.
But I can't fault them , so far !
Steve.
This is a 14inch by 8inch art panel that my wife says is too old and asked me to take down.
I've had my eye on this for some time.
I was going to glue a larger ply panel to this panel and see if I could increase the response in the 100hz to 300hz area.
but I do not have that size ply at the moment, so I decided to just quickly glue on an exciter.
5 minutes later and it was done, it's that quick.
I will wait till tonight before listening or maybe tomorrow just to be on the safe side , not sure I can wait that long .
I'm expecting a similar low end response as my larger art panel without ply, which is about 300hz or so I think ?
We shall see.
I might even see if it can go on a wall as the low end will already be restricted ?
At some time I do intend to try a larger ply panel on one of my art panels ,but first I must make sure the panel on its own , goes down to at least 120hz before serious roll off, at 1m let's say ?
If nothing else, it should be interesting.
Steve
I've had my eye on this for some time.
I was going to glue a larger ply panel to this panel and see if I could increase the response in the 100hz to 300hz area.
but I do not have that size ply at the moment, so I decided to just quickly glue on an exciter.
5 minutes later and it was done, it's that quick.
I will wait till tonight before listening or maybe tomorrow just to be on the safe side , not sure I can wait that long .
I'm expecting a similar low end response as my larger art panel without ply, which is about 300hz or so I think ?
We shall see.
I might even see if it can go on a wall as the low end will already be restricted ?
At some time I do intend to try a larger ply panel on one of my art panels ,but first I must make sure the panel on its own , goes down to at least 120hz before serious roll off, at 1m let's say ?
If nothing else, it should be interesting.
Steve
Attachments
Thank you. The distance from the panel does seem to be part of the formula...it sounds better the closer I get to the point that putting my head right next to the speaker produces a pretty cool effect.
I feel like I need sound coming from the TV wall somewhat in line for movie watching...if I ONLY cared about music I would mount these horizontally over my viewing couch on the ceiling.
I have one panel sanded and the other unsanded. Unfortunately that's not my only variable as the exciter setup is different on each so it's hard to pick out for sure the effect, but so far I THINK unsanded is better for this. The sanded panel has some self noise which the unsanded doesn't have. There are tiny bits of loose fiber on the cut edges of both panels but more on the on I sanded because I broke more loose. I might try a bit of glue on the edges.
To keep people from having to search back, here are the panels I am using:
https://www.rockwestcomposites.com/cftfs3-mg-dl
My mistake, my post 4939 on page 247 shows a response plot of the art panel without ply ,after pva treatment going down to at least 130 hz .
not sure this one will get down that low , but the art panel without ply does seem to go lower without the ply suck out at the 100hz to 300hz point at 30 cm, I noticed ?
Steve.
not sure this one will get down that low , but the art panel without ply does seem to go lower without the ply suck out at the 100hz to 300hz point at 30 cm, I noticed ?
Steve.
I should add, handling the CF foam sandwich is a little surprising. I expected stiff obviously but at only 3mm total thickness it's even stiffer than expected. I can't noticable flex it by hand.
It's also MUCH more resonant than I expected. Tapping on the panel with my finger produces sound comparable to tapping on a stringed musical instrument. This concerned me for ringing which I do hear but it's not terrible. Constrained edges should help a lot.
Side note...I wish I could even out the bass levels because I really enjoy the sound of bass through these despite it's unevenness because the attack and roll off is so crisp and there is so little annoying muddiness.
It's also MUCH more resonant than I expected. Tapping on the panel with my finger produces sound comparable to tapping on a stringed musical instrument. This concerned me for ringing which I do hear but it's not terrible. Constrained edges should help a lot.
Side note...I wish I could even out the bass levels because I really enjoy the sound of bass through these despite it's unevenness because the attack and roll off is so crisp and there is so little annoying muddiness.
Narenaud.
Years back I purchased this material in a 1cm thickness , it was far too rigid for my exciters and needed a lot of powering.
tectonic use 4 exciters to power their panels and various techniques to drive the panel more.
https://www.wickes.co.uk/ProWarm-BA...cker-Board-Single---1200-x-600-x-6mm/p/166935
The high frequencies were very strong and hard.
Not sure what to suggest ,maybe edge damping would help ?
or clamping in a rubber frame mount similar to this old patent ?
https://patents.google.com/patent/U...=priority:19610101&oq=inventor:+A+Cohen++1961
Steve.
Years back I purchased this material in a 1cm thickness , it was far too rigid for my exciters and needed a lot of powering.
tectonic use 4 exciters to power their panels and various techniques to drive the panel more.
https://www.wickes.co.uk/ProWarm-BA...cker-Board-Single---1200-x-600-x-6mm/p/166935
The high frequencies were very strong and hard.
Not sure what to suggest ,maybe edge damping would help ?
or clamping in a rubber frame mount similar to this old patent ?
https://patents.google.com/patent/U...=priority:19610101&oq=inventor:+A+Cohen++1961
Steve.
Spedge wrote: Jangling bells and crashing bells already sound good though, eps always excels at this , and the ambience is excellent.
I am currently using high density EPS and agree. Hand bells, tambourines, chimes, triangles, etc sound really good. Wooden percussion instruments stand out as well such as timber drums, wooden blocks, tapping on the guitar, etc. and yes more cowbell!
I am currently using high density EPS and agree. Hand bells, tambourines, chimes, triangles, etc sound really good. Wooden percussion instruments stand out as well such as timber drums, wooden blocks, tapping on the guitar, etc. and yes more cowbell!
Toddincabo.
what size panels do you have and what thickness.
my old 40cm 1cm thick panel is a bit battered about now days, I've done many experiments on it.
I do believe I have a 5mm hd 40cm square panel left somewhere.
Maybe I should give it a try, I ment to do this many years ago, but never quite got around to it.
I might do the same as I did to the 1cm panel to make it out of square.
Trumpets also sound so realistic as well, doesn't matter how loud the trumpet is , eps plays it with ease and realistically.
Nothing comes close.
Steve.
what size panels do you have and what thickness.
my old 40cm 1cm thick panel is a bit battered about now days, I've done many experiments on it.
I do believe I have a 5mm hd 40cm square panel left somewhere.
Maybe I should give it a try, I ment to do this many years ago, but never quite got around to it.
I might do the same as I did to the 1cm panel to make it out of square.
Trumpets also sound so realistic as well, doesn't matter how loud the trumpet is , eps plays it with ease and realistically.
Nothing comes close.
Steve.
My thought was that the shiny surface was bad. Seems not to be, though.
As to treble, it goes silent around 13500 Hertz but my hearing only goes to about 15000 anyway and I don't notice the loss in music so far. If I can tweak it to squeeze past my hearing threshold I will.
I have a plan to try an experiment. I want to get a big 1 inch pink XPS panel and cut a window and mount this panel inside it with all 4 edges secured. I want to see what happens. I suspect crazy modes in the corners will be tamed by partial constraint and also I'll get some sound from the XPS with the whole present panel basically acting as an exciter for the larger panel. I also suspect I'll introduce some distortion but it might be small. Right now these panels are too resonant. Sounds in the 1-2k range have mechanical reverb which is pretty strong and some much weaker individual frequencies higher up.
I'm waiting at the car dealership right now waiting on a minivan Im buying to replace the one that was destroyed last month. Once I have it I will head to the home improvement store.
These panels are small 12"x16" @ 1/2" thickness using DAEX25FHE. (30.5 x 40.5 x 1.3cm). Yep, trumpets, flutes, cellos. Nothing Else Matters cover by Apocalyptica sounds crazy good.