Boring old Dayton DAEX32EP-4 Thruster 32mm's Andre. I first got an extended top end from these on the build I did for my brother. I had expected to have to EQ the top end as I have done in the past but his panels measured the same which came as a bit of a surprise. I have used the same exciters on free hung Birch ply panels and the Scandi Blonde experiment and had to correct the top end drop on both of them. I am not sure why these are so extended but they have measured consistently in two versions of the same type of build now. The difference in build suggests the material is giving a better response but I am not sure of the mechanism yet. Picture of the design for reference.
Burnt
Burnt
Attachments
Bass:
Bass is limited by the type of panel material, the panel size, and its edge treatment. Some very soft materials prefer to be clamped solidly all around; stiffer materials might benefit from a loose edge. The sweet spot for me seems to be a large panel area corresponding to 2 x 15" cone speakers, and supported by a soft rubber support along the long edges.
You can get very extended bass out of a properly sized and mounted panel, but it might not have the transient "kick" that you get out of a front-loaded cone speaker.
Treble:
As per Burnt's most recent measurements, you can certainly get very decent output above 10khz depending on your choice of panel material and drivers.
There are many designs that cheat a bit. Some use piezo's as HF drivers, others use shorting caps across some drivers (if you're using multiple drivers) (@homeswinghome , Christian I think this was one of your suggestions. According to my simulations it's pointless, but I've tried it on my PA panels and it actually works very nicely indeed!) I've also done my own version of a "mechanical" cross-over by placing the driver on a CD-sized disk in a hole cut into a less-stiff panel, but it takes a lot of trial and error to get it right. If all else fails, then a digital signal processor will certainly flatten your response to recording-studio-monitor grade, and it will do so without introducing phase problems.
All of the above solutions retain the bending-wave action of the panel and its omni-directional characteristics. But be careful, there are other poor attempts at getting better HF, such as gluing aluminium tin-cans and whizzers to the front of the driver, and which carefully destroy the enormous advantages of DML panels.
However, if whizzers and tin cans seem like good solutions, and if one thinks they "sound" better, and especially if one does not care to properly measure their system to aim for a decent, audiophile-grade frequency response, then it would be better to go for a cheap point-and-spit cone speaker.
There are some who just do not do not understand the fundamental differences between bending-wave panels and pistonic cones. Maybe it's beyond their grasp. They happen be the same people who misunderstand the most basic issues regarding the audio reproduction chain.
Be careful who you take your advice from.
Just to add to Andre's comments on bass performance, another reason to use separate subs is you can position them independently of the panels so you can take advantage of Earl Geddes work on bass reinforcement. Earl used to post on this forum but I don't know if he still does.
xsuper9988,
I and several others have detailed how to build our DMLs at least once. I have detailed mine several times. Like all of us, just do a little reading on this forum, There is even a chart compiling our efforts.
I and several others have detailed how to build our DMLs at least once. I have detailed mine several times. Like all of us, just do a little reading on this forum, There is even a chart compiling our efforts.
So your CD technique doesn't destroy the enormous advantages of DML panels while Spedges tin can technique does?🤣
Then you go on to say if one thinks they sound better hmmmm you mean like how you think your CD technique sounds better then the tin can or whizzer cone?😵
Audiophile grade frequency response?😵🤣🤣🤣 Hey bob your frequency response doesn't look like audiophile grade.😆🤣🤣🤣🤣
Some people dont understand that Bending waves once they reach the lower (100hz and below) frequencies become pistonic, while above 100hz they become modal in nature. Maybe its beyond your grasp?🙄
Like I said many times before some newbies come on here and after 3-6 months they are the experts. 😆🤣
Last edited:
Perfect example Burnt... Thanks.
I am a bit surprised at the plunge below approx 160Hz for a panel that size... Any thoughts on that??
Eucy
My experience with Thrusters is that they're not the best option for hf extension. I put it down to the larger voice coil diameter (32mm vs 25)
My answer to the good hf extension is Poplar vs Birch. Poplar SG is around 0.4, and Birch can be between 0.55 to over 0.7, assuming the ply core is also Birch (may easily/commonly not be).
Lighter is better for hf.
Perhaps interestingly, some while ago I tested 3mm Tasmanian Blackwood panels which are harder and denser than Poplar by some margin, and as expected, the efficiency was down compared to Poplar, but it produced perhaps the most even frequency response I've seen so far, maybe because of the damping factor of the material.
When I get time I intend to revisit those panels.
Eucy
@Eucyblues99
"My experience with Thrusters is that they're not the best option for hf extension. I put it down to the larger voice coil diameter (32mm vs 25)"
I understand that is the case as well Eucy.
"
My answer to the good hf extension is Poplar vs Birch. Poplar SG is around 0.4, and Birch can be between 0.55 to over 0.7, assuming the ply core is also Birch (may easily/commonly not be).
Lighter is better for hf."
Yes I would think that is probably the reason as well although I have not tested it.
"I am a bit surprised at the plunge below approx 160Hz for a panel that size... Any thoughts on that??"
In my experience when DML's move out of their operating window with bass their response plunges. 160hz seems a bit high but I am not to concerned as the sub fills in nicely. I spent a lot of time a year and half back trying to coax more bass out of DML's but in the end I find IB subs do a good job.
Burnt.
"My experience with Thrusters is that they're not the best option for hf extension. I put it down to the larger voice coil diameter (32mm vs 25)"
I understand that is the case as well Eucy.
"
My answer to the good hf extension is Poplar vs Birch. Poplar SG is around 0.4, and Birch can be between 0.55 to over 0.7, assuming the ply core is also Birch (may easily/commonly not be).
Lighter is better for hf."
Yes I would think that is probably the reason as well although I have not tested it.
"I am a bit surprised at the plunge below approx 160Hz for a panel that size... Any thoughts on that??"
In my experience when DML's move out of their operating window with bass their response plunges. 160hz seems a bit high but I am not to concerned as the sub fills in nicely. I spent a lot of time a year and half back trying to coax more bass out of DML's but in the end I find IB subs do a good job.
Burnt.
Hey BC , You mentioned your current panels are 3mm Poplar , why do they look glossy in the photos ? , Did you use a coating on them ?
thankyou.
Last edited:
@Earborne 963Hz
Hi. They are not glossy. The day I took the picture it was very sunny. What you are seeing is the effect of the sun shining through. In my experience I find Poplar ply doesn't need any treatment. Others may have tried treating the surface and there is an argument for stiffening the skin by applying a thin layer of epoxy on both sides but that would add weight which is to be avoided.
On materials for DML @homeswinghome has produced a very nice chart of the different materials people have tried with a 'sweet spot' for optimal performance. Search for his posts, he is a very capable and thorough developer who has added some good observations to this thread. His chart will show you that Poplar is already too heavy and has poor efficiency compared to EPS, EXP and composite materials, but I still like the sound of Poplar and don't find volume a problem. I am using 40W exciters though so that compensates.
Burnt
Hi. They are not glossy. The day I took the picture it was very sunny. What you are seeing is the effect of the sun shining through. In my experience I find Poplar ply doesn't need any treatment. Others may have tried treating the surface and there is an argument for stiffening the skin by applying a thin layer of epoxy on both sides but that would add weight which is to be avoided.
On materials for DML @homeswinghome has produced a very nice chart of the different materials people have tried with a 'sweet spot' for optimal performance. Search for his posts, he is a very capable and thorough developer who has added some good observations to this thread. His chart will show you that Poplar is already too heavy and has poor efficiency compared to EPS, EXP and composite materials, but I still like the sound of Poplar and don't find volume a problem. I am using 40W exciters though so that compensates.
Burnt
Andre, If you're including a dome in the category of aluminium cans and whizzers, then I suggest you try it as I have before you ridicule it..it certainly doesn't "carefully destroy" any of the DML advantages.
Eucy
Hey Eucy,
I note you have had success with 3mm Poplar which i can get from Plyco for $ 19 per 1200 x 600 sheet.
I am contemplating a design concept of using two different densities such as Poplar for one DML pair , and Poplar covered with a 0.3 mm Blackwood or Walnut veneer for the second set with my thinking being that the different densities will give a richer tone and frequency balance for presence although it might need some EQ.
Whats your thoughts on that versus just going to a 3 mm Blackwood ? (Plyco are currently showing nil stock of the 3mm blackwood ply or veneer)
One reason i am trying to source a quality veneer is for aesthetics , Having great sound is one feature but if they also have a beautiful dark grain that will make them an architectural feature in the room then its a big plus as long as sound quality is not lost.
PS , I guess if it hasnt been done i can experiment .
If pursuing range and contrast from two pairs of DML's , I would probably have to include EPS as one pair and poplar ply as the other too.
Last edited:
Hello Burnt,
I am also wondering the mechanisms at play in the HF. It is strange to see how it can be different from sample to sample. I have good result with poplar too. Recently, I made quick tests with a piece of light plywood coming from a packaging, the result was poor... because of a too poor double side tape? The theory identifies 2 low pass filters, one on the electrical side between the inductance and the resistance of the voice coil that limits the driving force in HF (reduction of the mechanical force) and one on the mechanical side between the moving mass of the voice coil and the panel mechanical impedance. But this is not enough to explain. Do you have in mind the adhesive used in each case?
Christian
Larger voice coil = higher inductance = lower HF ?
I can't explain but I think it more complicated
Good idea. I would be interesting to know how is the HF extension. Can a material with a smooth FR (so a high damping) have a good HF extension?
Christian
Hi Christian.
Yes that is a very interesting possibility. As I was reusing the exciters I had opted to use contact aggressive instead of tape. To use the adhesive you are supposed to apply to both sides and let the adhesive dry out before applying the components to be glued. My method of application is to apply the adhesive to the exciter pad, push that against the panel, then remove again. That way you get a very thin layer on both. The adhesive still provides a strong bond even though the layer is thin. I wonder if this provides a ‘stiffer’ mounting than tape?
Burnt
Yes that is a very interesting possibility. As I was reusing the exciters I had opted to use contact aggressive instead of tape. To use the adhesive you are supposed to apply to both sides and let the adhesive dry out before applying the components to be glued. My method of application is to apply the adhesive to the exciter pad, push that against the panel, then remove again. That way you get a very thin layer on both. The adhesive still provides a strong bond even though the layer is thin. I wonder if this provides a ‘stiffer’ mounting than tape?
Burnt
+ @Earborne 963Hz , @xsuper9988
The link to the post of the "material chart paper" is in post #1 with other links
The "history file" is in the github page AudioDIY_DML with papers from Burnt an Eucy
Christian
Based on the way you applied it, it must be stiffer IMO as most adhesives are stiffer than foam tape (VHB before?) especially in thin layers. Less damping as well.
Was that one of the differences between your previous build and later build for your brother, and be a likely explanation for extension of the high frequency response? It makes sense to me physically. In a related question, did you notice any significant deviations in the overall sound, discounting the EQ adjustments or lack thereof?
This is rather anecdotal but could be useful none the less.
This is rather anecdotal but could be useful none the less.