Remember..
The cotton canvas is the best! I'm playing with a couple of the plastic canvas frames and they just don't have the magic..
I wouldn't worry about how tight its stretched. That may cause some resonances..
The cotton canvas is the best! I'm playing with a couple of the plastic canvas frames and they just don't have the magic..
I wouldn't worry about how tight its stretched. That may cause some resonances..
I wondered about that. I didn't mention the picture was a print. The artists works at a small size then blows the artwork up and prints using a digital printer. The stock he uses for that is a plastic of some kind and it just doesn't hold tension.
I have some new stretched canvas on order. I had no idea until I looked how cheap pre-built stretched canvases are.
One of the things I like about this method is you go up in scale easily. Just order a really big canvas and add multiple discs and exciters. Fun stuff!
I have some new stretched canvas on order. I had no idea until I looked how cheap pre-built stretched canvases are.
One of the things I like about this method is you go up in scale easily. Just order a really big canvas and add multiple discs and exciters. Fun stuff!
Thanks, I did not recall that.
Meanwhile, I found this Azima patent, which covers the damping concept. I think the concept seems to be worth exploring more.
https://patentimages.storage.googleapis.com/9c/65/98/25ed4edea10cf7/US6826285.pdf
In it, among other things, he mentions Sorbothane and "Miers" foam (closed cell PVC) as damping materials.
Eric
Eric this may also be of interest. It suggests profiling the edge of the panel can be used to eliminate reflections.
Acoustics | Free Full-Text | Acoustic Black Holes in Structural Design for Vibration and Noise Control | HTML
Acoustics | Free Full-Text | Acoustic Black Holes in Structural Design for Vibration and Noise Control | HTML
Burnt,
Nice find. Very interesting. I never heard of "Acoustical Black Holes" before. The Azima patent did talk about thinning the edges to slow the wave speed at the edges and reduce reflections, but the paper you linked describes a much more specific edge thickness profile.
Thanks,
Eric
Nice find. Very interesting. I never heard of "Acoustical Black Holes" before. The Azima patent did talk about thinning the edges to slow the wave speed at the edges and reduce reflections, but the paper you linked describes a much more specific edge thickness profile.
Thanks,
Eric
My pleasure Eric.
I think it might be easy to do if you can get a router bit ground to the right profile, which is the hard part. After that it’s just a routing job. It might combine well with using fabric as a mount/suspension in a variation of the method OffGrid is using. That way the thin edges, which will be fragile, could be protected.
I think it might be easy to do if you can get a router bit ground to the right profile, which is the hard part. After that it’s just a routing job. It might combine well with using fabric as a mount/suspension in a variation of the method OffGrid is using. That way the thin edges, which will be fragile, could be protected.
DMLBES, I have followed your posts here, at Audiocircle, and the Parts Express Forum. You have been very passionate and consistent with your advice and design concepts. For you to be this excited about the framed canvas concept - it must really be a solid improvement! Are you still on it? Thanks.
When I get around to it, I'm going to try 2 motors in a rectangular frame. (12" x 24"_14" x 30") *L/R channels on the same canvas..
I noticed placing both of my prototype units side by side increased the bass response significantly. (The stereo effect was still very noticeable) Perhaps if they were both on the same canvas I may get the same effect? Divide the width in 3 equal parts and center the motors in the middle of parts 1 and 3? (Giving 1 full space between the two) It's on the agenda.. 😎
Actually I spoke to soon and got too excited for my first impressions of the framed canvas concept. After more serious testing and experiments against my original EPS concept designs I still prefer my original EPS design over the canvas.
The only thing the Canvas does better is tone because of the paper like cloth it sounds so natural and smooth.
The EPS is still more accurate, detailed with higher spl output.
I also feel that Mylar would sound better then the canvas but I am not willing to spend the funds needed for that route.
The canvas concept has taught me a few things that I have forgotten when I first started. I should of took notes from the start of my venture in case I forget things along the way. When I first started I remember that the lower density EPS had better bass but less highs while the higher density EPS had less bass but more highs. Its like Goldie locks and the three bears where there were three porridges one was too hot and one was too cold but one was just right. The middle density EPS will probably have the best of both worlds. Why do softer density EPS have more bass? Its all about BENDABILITY/FLEXABILITY. This is the reason why honeycomb composites have good bass because its ability to bend/flex. Canvas has decent bass because the cloth can bend/flex. Also like Ive stated many times before that the THINNER the material the better the bass will be because thinner materials can bend/flex more then thicker materials that is why I suggest for EPS to be 1/4-1/2 (5-10mm) inch thick.
The only down fall of DML panels is the bass. Since exciters are very limited in pistonic motion they will never produce the amount of bass as a conventional cone driver. But anyone can just add a conventional powered subwoofer to the DML panels and they will sound spectacular. This also seems to be the case with other electrotats and planars like maggies but I read on another forum where the Magnepan creator is adding a conventional cone sub and wanting customer feed back.
Plus the older I get the more I forget so I am thinking of writing all my past info down for future references.
Or a round bit, stepped to approximate the profile, which should not be hard at all. Seems suited to a thick homogeneous (ish) material like xps or eps. You could probably taper from 1 inch to .1 or .2 inch. Then add damping material to the machined trough. Hmmm.
Eric
Can we talk about impulse response?
I'm readily admit that I'm pretty novice regarding measurements.
But i looked at impulse response on my REW profiles of several DML panels this weekend. Plywood, gator ply, eps, foam core. All took about 300 ms to decay.
But the first post of this thread (foamcore) shows an impulse response that decays in well under 1 ms. Anybody else looked at impulse response? What's reasonable to hope for?
Eric
I'm readily admit that I'm pretty novice regarding measurements.
But i looked at impulse response on my REW profiles of several DML panels this weekend. Plywood, gator ply, eps, foam core. All took about 300 ms to decay.
But the first post of this thread (foamcore) shows an impulse response that decays in well under 1 ms. Anybody else looked at impulse response? What's reasonable to hope for?
Eric
Last edited:
Yes that would work well on thick XPS but I am interested in applying the method to plywood, which I still prefer at the moment. My panels in France are 1/4 inch hence interest in custom profiled router bits.
For contributing to the body of knowledge here are two suppliers, one US, one UK/Europe
Custom Router Bits | Carbide Router Bits | Handrail Tooling
NLS Tools - Contact Info
Burnt
@OffGrid now that is very interesting. I don't know if anyone has tried putting panels side by side before. In theory with a pure DML approach this should not have an effect so you have definitely got something new there, congrats!
I like the idea of a wide canvas panel with two exciters for stereo-great WAF and would look very cool.
I want to try two things now. Butting up a couple of pure DML panels just to see if you get the same bass doubling. If you do then either the theory is not complete or there is something else going on.
Plus I want to try doubling up on the exciters on a larger ply disc. This is to see if its possible to build a hybrid 'big' exciter with increased energy transmission and the ability to combine exciters with different characteristics to balance out the frequency response. I don't think you can do this on conventional DML panels but it might work on the OGKOG method.
Worth a shot!
Burnt
I think when the panels get too big, the bass is diminishing returns. I am guessing that the bass is more pistonic. So something similar to a large 18in woofer is probably all you need. Bigger and. It goes DML for bass.
Another option is to build an angled jig for your router table. A couple of examples on YT. Though most that's there seem to be smaller examples. Not sure I'd want to play with huge bits without significant stabilisation. i.e. lots of feather boards and stout fence behind it all.
I don’t have a router table Jerms. I was planning on clamping the price to the top of an mdf profile piece and using the custom cutter. The profile you need is a power law curve which is quite precise and I don’t trust myself to do anything precise : )
That makes sense. My test frame is much bigger than an 18 inch driver in terms of area so I am going to mess about clamping that to explore the optimal size.
@Veleric. Eric in the papers I have read I get two perspectives on impulse. Azima claims a very fast rise time and a low level but extended decay. From memory he claimed the decay signal was random and therefore did not have a significant impact on reproduction. Again from memory he was claiming circa 25 us for the impulse rise time.
A second paper from the Acoustical Engineering Society claimed a much longer decay time and claimed this carried a significant distortion component. Sorry I know this doesn’t help but it suggests to me that the test examples used were probably quite different. What this suggests to me is that a good performance is possible but not automatic or garunteed.
A second paper from the Acoustical Engineering Society claimed a much longer decay time and claimed this carried a significant distortion component. Sorry I know this doesn’t help but it suggests to me that the test examples used were probably quite different. What this suggests to me is that a good performance is possible but not automatic or garunteed.
Last edited:
Burnt,
Thanks. But I what I was looking for was maybe some actual measurements by some forum members on DML (or other) speakers that showed typical responses, and/or insights that explains why my results on a wide variety of substrates are so different from what was shown in the original post of this thread. Are there any REW experts here that can comment?
Might I be measuring wrong? What might I be doing? Is it more likely the room I'm in than the speaker itself?
Eric
Thanks. But I what I was looking for was maybe some actual measurements by some forum members on DML (or other) speakers that showed typical responses, and/or insights that explains why my results on a wide variety of substrates are so different from what was shown in the original post of this thread. Are there any REW experts here that can comment?
Might I be measuring wrong? What might I be doing? Is it more likely the room I'm in than the speaker itself?
Eric
Again, in the second TI video (at about the 21 minute point), the guy tried thinning out/tapering) to the edges on an XPS panel to see how it effected FR. He showed the panel edge wise and it was clearly thicker in the middle and then thinned out to the edges so it looked kind of like an airfoil I guess. They did a Freq Sweep on it and showed the results...bottom line was that it wasn't chosen as being better either...they were testing it to see if tapering from the center to the edges would increase the panel efficiency...from what I could see of the graph, it looks like it did but the peaks and valleys may have been even more wild... i.e., not enough info to go on really...
geo
geo