I did not get involved with the conversation about this panel material suggested by Mark51.
But I did think the 2mm panel mentioned in the next post sounded interesting.
I did look on the Internet and it wasn't too expensive, not sure about postage.
This material might behave itself a little better than my 1mm panel.
Steve.
Using the matchsticks did increase the distance between the spider and the panel on the normal ring coil exciter.
But if using this exciter,in the picture, which does not seem to mind the extra mass of a panel or maybe extended shaft.
Extending the central shaft thread to get a greater distance is possible.
In the picture , luckily I had a piece of grey foam that fitted nicely in between the coil spring and the exciter foot.
This helped damp the springs and central coil area a little.
Steve.
Sorry, picture will not load again.
Steve,
In other tests I haven't shared, I tested something close to the matchsticks arrangement. The exciter was on small "piles". The picture below is while gluing
After are the gated FR from front (red) and rear (green). The "high SPL area" of the rear is here in the 3/4k range. The peak at 10k is probably a mix of drum effect and coincidence frequency.
The good point is the IR is not as spiky as with the other tentative of interface.
The FR (long time window = similar to a power response) was not that bad (see third picture) except the HF peak. The panel is 25cm x 125cm
If I remember the sound was a bit better with this configuration than without but it was not day and night.
It could give us an idea of the minimum contact points to have on the 20/25mm diameter
Christian
Hello Christian.
It's been a long time since I attempted to use x shapes (and other shapes) in the centre of the voice coil area on my ply panels.
I had limited success using different materials.
I preferred to do as little as possible in this area, I might have posted this on the other forum?
Steve.
It's been a long time since I attempted to use x shapes (and other shapes) in the centre of the voice coil area on my ply panels.
I had limited success using different materials.
I preferred to do as little as possible in this area, I might have posted this on the other forum?
Steve.
Christian,
Yes, this is a good point. The spine is essentially another DML on the back. It would be fun to see what happens if instead of a regular spine you used a second panel attached to the back of the exciter!
My own approach so far has been to use a thick, short, spine and clamped to the frame to push the lowest resonance frequency of the spine as high as possible. That may be why your 9x55 vertical mm spine works pretty well, as it is the stiffest of the spines you described. I am not so sure that the width of the spine is really much of a factor. I say that because we have previously found that the area of a panel has little effect on the SPL, and I don't see why the spine would be any different.
Ideally the spine would be stiff enough that it's own resonance would be above at least 10k, and ideally over 17k (for those under 20 years old). But that may be difficult to achieve. But I suspect the other things that would be good ideas would be to (a) use spine materials that don't radiate sound very efficiently and (2) use good damping materials between the back of the exciter and spine.
I have generally used 18 mm thick wood for a spine, but that is probably not the best choice, as wood has moderately high radiation efficiency. In retrospect, something like MDF or PMMA, both of which have higher density and lower stiffness would be better in that respect. Probably there are even better materials if we think about it, or even composites (say MDF with a steel weight added?).
But maybe that isn't even needed if there is a good damping material between the exciter and the spine. As you may recall, I have tried several such materials between the exciter and the spine. In my permanent builds (i.e not prototypes but final speaker pairs), I have used 3/8" thick EPDM foam between the exciter and the spine. To be clear, my main objective for doing that was to minimize any excess offset of the voice coil relative to the magnet, rather than to damp the vibration of the the spine. But it could be helping a lot in that regard.
Eric
easy to establish the amount of output radiation of a given spine by mounting a n exciter to it and measure or just have a listen to it. you could do the same to see the effect of different damping materials used between the exciter and the spine.
Just to add to the number of "knobs" to tweak, I had a thought regarding the current discussion about the acoustic impact of a spine.
I have been thinking about attaching an additional mass to the back of the exciter, basically thinking about increasing its inertia. It might improve the bass response (perhaps not). But another potential application would be to attach another panel to the back of the exciter. Attaching that panel to a different location would excite different modes so the overall effect might smooth-out the frequency response -- mids on down.
This system would add a lot of complexity to the model because the panels would interact with each other through their common exciter connection.....and also through the air that is trapped between them. The exciter body is a lot more mass than the VC so it would behave like a low pass filter. So I'd expect it would affect the FR on the lower end.
One practical consideration is the added weight that's hung on the back of the exciter. It might displace the VC too much, unless some additional support rig is added. It wouldn't have to be a rigid spine, but does imply that the panel is mounted on some kind of frame.
Since I already have made a frame It should be relatively easy for me to play around with some of this. The downside for now: I don't have a calibrated microphone to measure the results. At least, not yet....
I have been thinking about attaching an additional mass to the back of the exciter, basically thinking about increasing its inertia. It might improve the bass response (perhaps not). But another potential application would be to attach another panel to the back of the exciter. Attaching that panel to a different location would excite different modes so the overall effect might smooth-out the frequency response -- mids on down.
This system would add a lot of complexity to the model because the panels would interact with each other through their common exciter connection.....and also through the air that is trapped between them. The exciter body is a lot more mass than the VC so it would behave like a low pass filter. So I'd expect it would affect the FR on the lower end.
One practical consideration is the added weight that's hung on the back of the exciter. It might displace the VC too much, unless some additional support rig is added. It wouldn't have to be a rigid spine, but does imply that the panel is mounted on some kind of frame.
Since I already have made a frame It should be relatively easy for me to play around with some of this. The downside for now: I don't have a calibrated microphone to measure the results. At least, not yet....
Eric,
What we see in those tests are in my understanding acoustic effects of the spine (= effect of the spine on the trajectory of the rear wave). Those "test spines" or obstacles were not mechanically linked to the exciter. They were just hand held few millimeters behind so I don't think they got enough airborne energy to excite their own modes.
Christian
Mark,
Maybe something to test but I am afraid it is difficult to predict or act on the results due to as you say the complexity of the new system which is getting additional degrees freedom( more variables).
In an other hand, I think a target for any open back loudspeaker is to have a back wave as similar as possible to the front one in order to create an omnidirectional radiation. There are example of open baffle loudspeakers which were said improved by their designer by adding a back side driver. This has maybe a limit in the treble (Gobel says the rear radiation of its DML is limited to 4kHz or something like that)
About measurement, no need of a calibrated mic. A USB mic is very handy but many DIYers have been working with a simple low cost omnidirectional electret. They have of course some dB error (from 5k?). You can have an idea of the error using a classical full range or a tweeter.
Christian
I'm trying to understand the focus on ensuring that the radiation to the back is uniform. It seems extremely rare with setups where the panel actually needs to radiate to the back as well. One scenario could be an open space like an exhibition area where you can spread out speakers in the middle of a crowd, or you could make them double up as monitors for stage application I guess.
But in most cases you want to be killing backwards radiation rather than tuning it? For hifi or cinema the ideal is to mount the plate in the wall ,or otherwise absorb as much of the reflections as possible. For a dancefloor, you don't usually want to put speakers in the middle, meaning you cannot make use of radiation in both directions either.
Is it some preference that open baffle enthusiast have that having reflections from a real wall somehow improves the sound in a hifi setup, or are you actually using them in scenario where you make use of the omnidirectional sound?
If not using the plate to deliver direct sound to a bigger area, will a more uniform radiation to the back actually make much of a difference? It would be similar to when you are producing music and mix in a little bit of reverb and you have to make quite drastic filtering to hear the difference. If you hearing the difference of the width of the spine in your reflections, that is a lot of reflections, and I would think it is better to treat the walls a bit with absorbers or experimenting with making absorption built in to the speaker.
Not dismissing the great work you are doing mapping and trying to understand what is going on, but just curious if and why you would find it important in practice?
But in most cases you want to be killing backwards radiation rather than tuning it? For hifi or cinema the ideal is to mount the plate in the wall ,or otherwise absorb as much of the reflections as possible. For a dancefloor, you don't usually want to put speakers in the middle, meaning you cannot make use of radiation in both directions either.
Is it some preference that open baffle enthusiast have that having reflections from a real wall somehow improves the sound in a hifi setup, or are you actually using them in scenario where you make use of the omnidirectional sound?
If not using the plate to deliver direct sound to a bigger area, will a more uniform radiation to the back actually make much of a difference? It would be similar to when you are producing music and mix in a little bit of reverb and you have to make quite drastic filtering to hear the difference. If you hearing the difference of the width of the spine in your reflections, that is a lot of reflections, and I would think it is better to treat the walls a bit with absorbers or experimenting with making absorption built in to the speaker.
Not dismissing the great work you are doing mapping and trying to understand what is going on, but just curious if and why you would find it important in practice?
There is a fascinating aural/effect that can take place with open baffle speakers > sound radiating in great similarity from both front and rear.
It generally requires some room EQ and the speakers having a reasonable distance from a rear wall.
The Hybrid DML's I built [ DML + open baffle woofer + tweeter ] have tricked/fooled me on numerous occasions as to whether a sound has
come from the speaker or from outside [ with my front door open ] > I'm sure this has a lot to do with placement, but it is strange when it happens.
This is an example of a sound so true that you're not sure that it came from the speakers 🙂
It generally requires some room EQ and the speakers having a reasonable distance from a rear wall.
The Hybrid DML's I built [ DML + open baffle woofer + tweeter ] have tricked/fooled me on numerous occasions as to whether a sound has
come from the speaker or from outside [ with my front door open ] > I'm sure this has a lot to do with placement, but it is strange when it happens.
This is an example of a sound so true that you're not sure that it came from the speakers 🙂
Hi Leob,
It is a good question and sometimes I wonder how far to go... I started this investigation with the list of what could be a problem as it was identified in the DML threads and what I usually see in my measurements.
The front radiation of an open back DML is linked to the rear radiation independently of the reflection on a back wall by what propagates in the air surrounding the panel. I don't have a good model in mind but I think in a software like the Edge, the rear is modeled as sources at the perimeter of the baffle radiating to the front. Does it enter in diffraction mechanism, I am not sure.
The sound of an open back DML in a room is the result of the direct sound which is already the result of the mixing of the front and the rear waves and, a bit later in time the reflected sound from the room.
What I am pretty sure is that some peaks in the front FR (a time gated one that excludes reflections) might have this rear emission as cause ("might" because I have no evidence of that)
Due to our sensitivity to the 2 to 5k frequency range where it occurs, its possible (very) high SPL, its seems better to push further the topic to get get designs where the effect will be limited or even (I don't know if it is possible) not present.
The difficulty is we can't have for now listening test results of a panel free of this problem to know if it is a real problem or not.... so I prefer to follow Steve's experience with in mind in this case a kind of ideal panel fully symmetric.
Christian
Leob and Christian.
DML is no different from planar, electrostatic, and open baffled speakers.
The rear waves are not constrained within a box, so they do not suffer from box colorations.
I think this is the main reasons people love no box baffles.
I don't think it is possible to have a uniform sound using the exciters we use.
As long as the rear wave is not producing bad sounds with large peaks and dips, that is the best we can hope for.
These bad sounds are very obvious with very small panels, especially if you are only sitting 10ft away in your armchair.
How much the rear wave affects the front sound depends to some extent on the panel material and size .
A large heavily damped panel would block most of the direct sound coming from the exciter area, but not the reflected sound from the sides and back wall.
Then there is mechanical noise, how much is passed on through the panel.
You can easily alter the sound of the noise from the exciter by placing your hand over or near the exciter magnet, this usually makes it worse.
Exciters are not perfect, but is all we have at the moment, until maybe high powered piezoelectric drivers finally arrive.
Steve.
DML is no different from planar, electrostatic, and open baffled speakers.
The rear waves are not constrained within a box, so they do not suffer from box colorations.
I think this is the main reasons people love no box baffles.
I don't think it is possible to have a uniform sound using the exciters we use.
As long as the rear wave is not producing bad sounds with large peaks and dips, that is the best we can hope for.
These bad sounds are very obvious with very small panels, especially if you are only sitting 10ft away in your armchair.
How much the rear wave affects the front sound depends to some extent on the panel material and size .
A large heavily damped panel would block most of the direct sound coming from the exciter area, but not the reflected sound from the sides and back wall.
Then there is mechanical noise, how much is passed on through the panel.
You can easily alter the sound of the noise from the exciter by placing your hand over or near the exciter magnet, this usually makes it worse.
Exciters are not perfect, but is all we have at the moment, until maybe high powered piezoelectric drivers finally arrive.
Steve.
Christian,
Thanks for clarifying that. When you referred to them as "spines" I just assumed that they were attached to the exciter. I do have to admit that I wondered why your hand was holding them! Of course I agree in this case that the spines/obstacles themselves were probably not being excited enough to produce sound. Did you also measure the front to back difference with no spine/obstacle? It would be interesting to see the various results in an overlay plot.
I must admit though that I am still wondering how much the excitation/vibration of a real spine (i.e. attached) might contribute to the front/rear difference. I would not be surprised if in some cases it would be a bigger influence than the "blocking" effect you are showing here. But I have never tried to measure either, so I really could not say.
Eric
Mr. "Flat-coil" would have disagreed, if he had not been banned! Haha.
Eric
Eric,
Here are examples of front and back with no spine. For each, there 4 curves : front / back, long time window (100ms), short time window (3..5ms). With the long time window, the reflections of the room are included (kind of power response), with the short time window it gives a quasi anechoic response.
First is a poplar plywood test panel, second is the same with the exciter glued "on piles", third is a light XPS Depron.
To confirm with more measurements, the exciter mask effect seems to occur in the 1 to 3kHz range when the cavity resonance might be in the 4 to 6kHz range.
I think at least for my taste and purposes, a DML sounds open enough without real wall reflections. I think, like most speakers, they sound best outdoors where there are no rear reflections.
But I do use them indoors, and if I would have noticed the response issues from the rear, my primary approach would be to reduce the rear reflections to be less obvious. At least according to any conventional sound engineering perspective, that should result in a better sound than any small adjustments made to the rear response.
I somehow doubt that I would think that an open baffle speaker would not sound as good if the rear walls would have some absorption and diffusion, and would guess that they also sound best outdoors with any rear reflection whatsoever. But having some diffused reflections could give a bit of DML character to the sound perhaps? DML is already diffuse though, and I think that if you wall mounted a DML panel to make it act as a cardioid, it will still sound very open, since diffusing the already diffuse signal doesn't do that much to change the overall impression.
My current thought for using plates indoor is to cover the back of the frame with some material that dampens the emitted sound without actually closing the back. So something like acoustic foam or rockwool, which will not impede the movement of the plate since it doesn't completely block the waves, and that will not reflect a lot, but reduce the SPL to the rear. It will not be effective over the whole range of the speaker, but should make it better and can be complemented with some acoustic treatment.
But I do use them indoors, and if I would have noticed the response issues from the rear, my primary approach would be to reduce the rear reflections to be less obvious. At least according to any conventional sound engineering perspective, that should result in a better sound than any small adjustments made to the rear response.
I somehow doubt that I would think that an open baffle speaker would not sound as good if the rear walls would have some absorption and diffusion, and would guess that they also sound best outdoors with any rear reflection whatsoever. But having some diffused reflections could give a bit of DML character to the sound perhaps? DML is already diffuse though, and I think that if you wall mounted a DML panel to make it act as a cardioid, it will still sound very open, since diffusing the already diffuse signal doesn't do that much to change the overall impression.
My current thought for using plates indoor is to cover the back of the frame with some material that dampens the emitted sound without actually closing the back. So something like acoustic foam or rockwool, which will not impede the movement of the plate since it doesn't completely block the waves, and that will not reflect a lot, but reduce the SPL to the rear. It will not be effective over the whole range of the speaker, but should make it better and can be complemented with some acoustic treatment.
I note that the rear hf boost is missing in this case... Why would that be I wonder?
Eucy
Hi all, I am planning to build some DML speakers and wonder if someone could give me some input/suggestion regarding panel materials. I am German, based in Goa, India and there are not many materials in the market to choose from. Apart from the DML's sounding as good as possible, the esthetics play a big role for me as well. The materials I can get here are XPS (24"x36"x1"), hollow PP sheets and signboard materials which I could cut to size. If any of these materials sound better than plywood, I would hang them in a frame and hide them behind a thin speaker fabric. Regarding plywood, the thinnest and lightest one I can source are sheets (8'x4') of 6mm thick Baltic birch ply or 3.2mm thick aero light plywood in 24"x36" sheets. I could cut them with a laser to a recommended size with straight or round edges. One more idea is to make my own composite panel by laser cutting the 3.2mm aero ply into a honey comb panel and gluing 0 .6mm thick paper veneer on either side with epoxy. I do have 4 Dayton DAEX32EP-4 exciters to experiment with and would be happy to get any feedback, Thanks shido
Yes and yes - Try 300x600 or 300x900 panels - don't bother with the rounded corners unless you like the look - It's more about how the edges are supported (or not) than the round corners.
The lase cutting is worth a try but you'll need lots of trial and error, esp with shrinkage and humidity control - particularly in Goa (I've been there).
You'll need to find a good lightweight sealant - there's a lot of info on this site - avoid water based ones though.
Cheers
Eucy