I can see the difficulty with getting good sounding ceiling panels laying more with the placement and room response than deficiencies in the panel design itself. Being a true Biolar DMLs usually perform at their best when they room to "breathe" from hard boundaries. Ceiling panels are right up against a reflective surface, and depending on the room size, shape, surface materials and contents, you have all kinds of things happening with the waves that you wouldn't necessarily have to deal with if they were facing horizontally in the same space rather than vertically. It's a very non-optimal placement scenario.
It would kind of defeat the point of a DML to enclose it, but you're not wrong. I would think you would begin to run into a lot of the separation and potential frequency loading issues that OB speakers have when you start implementing perpendicular wings along the baffle. If you watch the excellent "Open Baffle Basics" YouTube videos by GR-Research you'll get an understanding of how critical the effects can be. The mechanics wouldn't work exactly the same I imagine, but I would think it wouldn't exactly be a simple process and could make the final footprint cumbersome.
NegativeGhostrider
Fully agree with you about ply versus polystyrenes. The loss of efficiency is not a so big problem considering the final sound.
I try to keep a kind of synthesis of the different DML built and shared along this long long this thread. See my post #4211for the last update. BurntCoil's Tall Blonde is in the file. Could you remind me your post(s) describing your panels or a little bit more detailed description (see the file in 4211)? Thanks
Christian
Sure thing. Post #4120 gives my build summary and photos. Happy to answer any questions as well. I think I just used the height and width ratio of the Tall Blonde project, or something similar. It gave me more surface area but allowed me to keep the footprint of the panels down. It also helped with low end response over panels of similar surface area but a more square size ratio. I expect it's a result of the anisotropic nature of ply, as Valeric pointed out back in the day. I recall Dave Lang had good results with tall panels as well.
Last edited:
Eric,
You are completely right, the factor 2 is missing! Sorry...
An Fc = 5kHz leads to a D/mu around 13.5. D/mu = 10 gives Fc = 6kHz which a bit higher what I infer at 6.5. Then F0 is 50 (D/mu = 13.5) to 57Hz (D/mu = 10)
Looking at the hemispherical contour plot of the DML 500 spec you pointed, it makes sense to assume an Fc between 5 to 6kHz rather than 7kHz (beginning of beaming).
About the thickness, it is what I understood in the video How DMLs Work Part 2 (see at 24sec). Better if you check.
About the mu, there is something already not used but I don't know how to do. It is the efficiency at 91dB which is reached with 8Ohm impedance and 4x32mm exciters. I posted a bit before (post #4237) I found an efficiency of 91dB about with a single Dayton Audio DAEX25FHE-4 (I guess the BL will be in the equation) and an XPS 20mm which has a mu = 0,81kg/m². Does it give you some idea?
Here is the link about the Piano Sound board thesis
NegativeGhostrider,
I have to apologize : your panels are for sure already in the file. You were the first to answer to my question "which DML are currently working"! Really difficult to keep the different exchanges in mind. Sorry again.
Christian
You are probably right- I will see if I can repurpose them in a more optimal space. Unfortunately the room they are in won't allow speakers anywhere else without cutting up the ceiling, which I don't want to do. I think/hope some OB drivers will work better for my situation
OB drivers are going to suffer from the same negative effects if you place them too close to a rear surface. Since they're dipole vs bipole it may be even worse, since they aren't getting enough space to be truly "open" behind the baffle. Rear reflections and frequency loading will likely be a major issue. I wish you good luck though.
Sheesh, seems I have to spell things out.
It’s not in a box of course. It’s as one side or wall of a box!
The main problem with DMLs is low efficiency and/or a large panel to reach low.
Using a resonator would help to couple certain bands better to the air.
Sheesh. It seems like I have to spell things out. I know it wasn't in a box, of course. I was talking about using baffles on one or both of the long sides to try to reinforce certain frequencies like open baffle speakers do. But with that wave manipulation can come other compromises.
Hmmm.....?
What about a DML in a kind of H-Baffle?
Maybe with a free hanging/suspended mid to upper range DML above?
That's what I was trying to get at. Not sure how well it would work. A U baffle or Asymmetrical H might be better to reduce how much you're forcing the omni pattern that makes DMLs so special into a hypercardioid or cardioid. I'm not sure that you would be better off trying to get low gain with baffles than just using the room for reinforcement by corner loading or something. It would be an interesting experiment. I think I recall seeing a guy in the PE forums who did something like this using multiple exciters on a large panel against a wall and a sort of front facing U baffle to make a sort of subwoofer DML. I know the mechanics wouldn't work the same as an OB speaker with cone drivers but OBs need pretty deep wings for low boost, and those wings cause loading of higher frequencies and lead to a boxy sound. So yeah, would likely need to do a 2 way array with a separate DML dedicated to highs to avoid that issue.
That’s totally missing the point. That’s no resonator (a very ineffectual one then). There is no backwave to stop canceling the front wave.
A normal cabinet has two functions. Containing, eating or delaying the backwave and secondly resonating across the lower octaves in various ways to better couple the driver to the air.
Bass reflex and pressure chamber is out of the question with a non pistonic driver.
But a non ported, straight resonating is not.
Last edited:
Eric,
Here is the link about the Piano Sound board thesis
Looking for a document I found among the pdf I collected about DML : Analysis of flat panel speakers by Luke Humphry. This document was pointed at least by gazzagazza in his post #1296 (pdf available) or under the link https://web.archive.org/web/2013031...6/files/290707 Final Report - Final Draft.pdf
Going quickly through it, there is page 23 the energy radiating by an infinite plate :
P = rho_air/(4pi.c_air).(F/mu)^2 where F is the force applied so F = BL.i or F = BL.U/Rcoil
So to make the link with previous exchange, it leads to a radiated power proportional to BL/mu... Hmmm. Not so clear for me. I think to make some test in the next weeks with different materials or thickness to come on a practical approach.
Any comments/advice welcome as usual!
Christian
OK, so I grabbed an 18x24x1/4" acrylic (Plexiglass) sheet in combo with 2x Aiyima 20W exciters and 1 channel of a Zoudio board amp
They are really interesting. First of all this is truly a full range setup. They go from about 40Hz up to 15kHz based on my ear + a test tone generator. And that is butted up against my desk- far from ideal. This is my current seating position as well- off axis response is excellent; the sound does not change at all.
I'm not quite sold on the overall sound though. It's a little dull and lacking in subtlety. I don't know how to explain it.
Interestingly they seem to have better bass than the 10 & 12" GRS woofers I bought. If I have time tomorrow I will just measure everything. Maybe they will wake up with a little EQing, but I hope they don't need too much. They def have the ends of the spectrum covered which is impressive, but it feels like something is missing. We'll see tomorrow
They are really interesting. First of all this is truly a full range setup. They go from about 40Hz up to 15kHz based on my ear + a test tone generator. And that is butted up against my desk- far from ideal. This is my current seating position as well- off axis response is excellent; the sound does not change at all.
I'm not quite sold on the overall sound though. It's a little dull and lacking in subtlety. I don't know how to explain it.
Interestingly they seem to have better bass than the 10 & 12" GRS woofers I bought. If I have time tomorrow I will just measure everything. Maybe they will wake up with a little EQing, but I hope they don't need too much. They def have the ends of the spectrum covered which is impressive, but it feels like something is missing. We'll see tomorrow
any idea regarding those metal discs attached to the plywood board attached to the canvas..
they are saying its for improved treble response .. do we need such metal discs on the canvas DML speakers..
Many years ago over on NXT RUBBISH I used two or three small eps panels in line array.
The panels were very close together , so I decided to join two panels together with a small piece of blu-tack.
I was surprised at the output of both panels with only one exciter driving the main panel and the other only being driven from the edge , being joined at one single point at the panel edge.
the disc is similar but being metal and only a small disc , it will only produce the higher frequencies.
The canvas does damp the ply a little so the disc might bring a little life to the sound ?
the disc could be mounted on a bolt at the back or front (or both) to be driven from a single point ?
would a disc on the side of an ordinary ply panel improve things ? Or mounted on a bolt in a positive position ?
I would have thought that if it this was wanted , it might be better to have the ply on the front side of the canvas ?
Although you might find you have more HF anyway if front mounting ?
I'd also probably cut a hole in the canvas so as to glue the exciter directly to the ply, for more HF ?
these are just some quick thoughts on this subject.
better stop before my head explodes😁
Steve.
The panels were very close together , so I decided to join two panels together with a small piece of blu-tack.
I was surprised at the output of both panels with only one exciter driving the main panel and the other only being driven from the edge , being joined at one single point at the panel edge.
the disc is similar but being metal and only a small disc , it will only produce the higher frequencies.
The canvas does damp the ply a little so the disc might bring a little life to the sound ?
the disc could be mounted on a bolt at the back or front (or both) to be driven from a single point ?
would a disc on the side of an ordinary ply panel improve things ? Or mounted on a bolt in a positive position ?
I would have thought that if it this was wanted , it might be better to have the ply on the front side of the canvas ?
Although you might find you have more HF anyway if front mounting ?
I'd also probably cut a hole in the canvas so as to glue the exciter directly to the ply, for more HF ?
these are just some quick thoughts on this subject.
better stop before my head explodes😁
Steve.
Last edited: