A Study of DMLs as a Full Range Speaker

I can't picture how gluing a flattened coil to a panel not at right angles, ie, in the plane of the panel if that's what you mean, is going to work.

Anyway, in most cases, the tendency is to support the exciter rigidly, so in effect, only the coil is glued to the panel in any case.

Maybe you can make your idea clearer with drawings

Eucy
 

I can't picture how gluing a flattened coil to a panel not at right angles, ie, in the plane of the panel if that's what you mean, is going to work.

Anyway, in most cases, the tendency is to support the exciter rigidly, so in effect, only the coil is glued to the panel in any case.

Maybe you can make your idea clearer with drawings

Eucy

You can see it in the video @lekha posted...

1729779594239.png


The magnet, held near the coil (on either side of the paper) doesn't touch the coil or panel.
Amplifier connects to the coil, the magnet then activates things.

Interesting idea.

I've got a couple of burned out exciters I can try to repurpose. Essentially crack them open for the coils and magnets.

Not certain how much a panel would move around with respect to the magnet or if it would matter.
 
  • Like
Reactions: Pano
Ok, but I can't see why or how this idea is superior to a traditional voice coil travelling within a magnetic field rather than pulsing towards and away from it. To have any degree of linearity it would surely need to be push pull which immediately makes it more complicated. As I mentioned, provided the exciter is supported by the frame, the only mass attached to the panel is the coil and mounting ring, so no real mass gain there either. And it's not how Magnepan and similar speakers work, they just have linear coils but still between magnets, basically straightened out standard drivers.

Eucy
 
Radial Thoughts




@Andre Bellwood, I had some thoughts in the same direction, mounting coils directly on the panels, but I came out in a different place. You can use flat coils mounted directly to the surface if you use radially magnetised ring magnets. To max out the field strength you can use two sets of coils and magnets on opposite sides which would flatten and intensify the magnetic field. You would have to invert the phase on one set of coils obvs. A manufacturer would be able to get custom ring magnets made up which is too expensive for DIY but you can sort off simulate the same effect using a radial array of bar magnets. Images attached.

I thought there were a number of benefits in that you can split the flat coils into several concentric coils allowing you to control inductance by summing in parallel/series to get the inductance you need which might help the top end performance. Coil heating would also be reduced as the coils are in open air and free to radiate and conduct away heat helping to extend performance limits. As my preference would be to use neodymium magnets there is also the benefit of getting iron out of the circuit reducing eddy current effects, although I am not sure if this is significant. Robustness and alignment tolerance will be much better as the displacement of coil to magnets by a few millimetre is no long a disaster, just a mild drop in performance.

One of the interesting questions for me is how you would treat the voice coil mass now it is directly attached to the panels and the effect this has on mechanical impedance and impulse transmission. This is all very theoretical at the moment, and predicated on being able to get a high enough B. Increasing L will be trivial. I expect there is something I have not thought of that will kill the idea, there usually is, but if I get anywhere I will report back to the group.

Burnt
1729804390499.png


More coil on panel from 2022.
 
I ordered 4 of the Dayton Audio DML25-4 2" Distributed Mode Loudspeaker Transducer 20W 4 Ohm and wired up 2 per each of my center channels.

No testing yet.
By ear they integrate nicely with the rest of my home theater system
I can hear and make sense of voices - clear.
Nothing sounds wrong, harsh, or off.

Parts Express suggested I could wire 4 of them in a series-parallel setup, giving me:
  • 4 ohm load
  • 6 dB higher sensitivity at 88 dB
  • ability to handle 80 watts RMS as a result
I could see running an array of 4 for each of my center channels.

1729992537713.png
 
  • Like
Reactions: Andre Bellwood
Soundimports started a new blog, together with an article written by Dr Ben Zenker (Xcite) as a basic guide to dml panels. Well written and should be very helpful to dml newcomers.

https://www.soundimports.eu/en/blogs/blog/a-basic-guide-to-create-invisible-sound-with-excit/
twocents,
Thanks, I hope he keeps adding to it. The quote below teases an upcoming blog post that could be interesting.

Last tip: a material that is 50% lighter can produce +6dB more output. An upcoming blog will delve into this topic and show you the differences in performance.

I also found these quotes interesting:

If you need maximum output – the High-Power exciter is the better choice, but needs to be combined with another High-Power exciter to get the full frequency range.
and
For a full-range response, use a full-range exciter, combine large and small diameter exciters on the same panel

But I wish he had elaborated a bit more on that, with regard to selection of, or placement of, exciters to "get the full frequency range".

Eric
 
I have read a few zenker papers, but I have seen nothing new.
He seems to me to regurgitate old NXT patents as if he has thought of them himself.
If someone has a paper he has written that has some new useful information, I would gladly read it.
Although I might be sadly disappointed.
I found it strange that he compared his 40watt exciter with a 3watt coin type 😳
Apart from making the 40watt exciter look more powerful, I fail to see the point?
Will his exciter be capable of outputting full power into a panel at 40 hz with realistic sound levels 🤔
If so this could be a good match for my round flueted panels.
But I have my doubts, has anyone experience with these exciters and how do they compare to the usual 40watt exciters ?
Steve.
 
I can see a live chat box at the right left corner of the soundimports link... The Xcite company has it's own website too.

I'd like to see distortion, impulse and directivity measurements. Effect of panel size would be interesting too.
But IRL, exciters are not intended for hifi sound.

There are some TVs that use the screen as speaker, OLED Sony and LG in 2017-19. Poor success... difficult to even find good links.
 
The Zenker paper is quite clear on the comparison in terms of performance, reason and possible application. Also confirmed long ago by your old friend in post #592
My old mate, confirms low power handling and spl for the small exciter 😄
No mention of Zenker exciter.
The 3watt exciter would work perfectly well with a low density EPS panel in a home environment, as long as partnered with a suitable low frequency driver.
I would always use some sort of low frequency driver, even if using a Zenker exciter.
I'm not sure I like the Zenker exciter design, but it could work well with certain panel types, or even stud wall mounting ?
Steve.
 
I found it strange that he compared his 40watt exciter with a 3watt coin type 😳
Apart from making the 40watt exciter look more powerful, I fail to see the point?
Steve,
He didn't actully compare his (XCite XT32) exciter to the coin type. rather, he was comparing two Dayton exciters. And his point was simply that, while the high power exciter has higher sensitivity overall, and good low frequency response, the smaller coin type has relatively better response in the high frequency region.

I'm not sure I like the Zenker exciter design
What don't you like about it? It actually doesn't look to be much different from the Dayton ones.
I bought a few of them. In my tests so far they perform almost identical to the DAEXFHE25-4 exciters. It's not clear to me (yet?) what would justify their cost, (nearly triple of the Daytons) but I see nothing obviously wrong with the design. The plastic housing seems plenty robust. What is it about the design you find questionable?

I do wish they showed a diagram of the "double magnet" to see if/how it's really something different than the rest.

My biggest peeve (besides the price) is that the literature says they are "Xtreme reliable" rather than extremely reliable. haha. But if they are truly better, I can forgive that!

Eric
 
Last edited:
I bought a few of them. In my tests so far they perform almost identical to the DAEXFHE25-4 exciters. It's not clear to me (yet?) what would justify their cost, (nearly triple of the Daytons) but I see nothing obviously wrong with the design. The plastic housing seems plenty robust.
Hello Eric,

Have you compared the distortion figures? I suspect that most of the distortion we can see below 500Hz, which is mainly from the second harmonic with a DAEX25FHE comes from a poor symmetry in the magnetic field.

I find also a bit strange to compare a 40W exciter to a 3W. In fact, The sentence following the graph "The Coin-Type exciter is a good choice, when only little space is available, low maximum volume and less low frequencies are required – e.g. when a subwoofer is available." is questionable. The case of a so little space that a coin type is needed seems to me marginal and the subwoofer argument is not good. Except the canvas, no panels produce real bass... But maybe the targeted bandwidth is not what I have in mind

I did some tests with a CT13 to see the effect of an exciter with a much lower voice coil mass and a lower inductance. More precisely, I compare a DAEX30HFE to a DAEX25FHE to a CT13 this on a 3mm polystyrene glass and on a 20mm EPS panels using PVA glue to glue the exciters... and I haven't found the evidence of a better HF due to lower mass or lower inductance... what I thought to find as it is predicted from papers about DML or from previous tests. So when I read "Smaller exciters generally have less sensitivity but improve high-frequency response thanks to lower inductance and moving mass", I am waiting for evidences (about the HF, for the sensitivity, no question).

Also "the High-Power exciter [...] needs to be combined with another High-Power exciter to get the full frequency range."... We know that to get the HF is a problem but again, the few tests I did with a CT13 haven't shown a small exciter is a possible solution to extend the bandwidth. They have basically a lack of efficiency (much lower BL) compare to a 25 or 40W model.

The paper on the blog is a good initiative but not precise enough... let see the next ones.

Christian
 
Have you compared the distortion figures? I suspect that most of the distortion we can see below 500Hz, which is mainly from the second harmonic with a DAEX25FHE comes from a poor symmetry in the magnetic field.

Christian,
Yes I have looked at the distortion figures. They were virtually identical for the two exciters, including the second harmonic across the entire frequency response. That said, I think the result may be of little practical significance because in both cases the total harmonic distortion essentially matches the noise floor. I don't know for sure, but I suspect that distortion measurements are really only accurate if they exceed the noise floor in the test environment. If they are below the noise floor, I think the distortion measurements shown are simply artifacts of the room noise and not a true indication of the speaker's harmonic distortion. Is this your understanding?
If so, then I think that the only thing I can conclude from my distortion results is that both exciters (on my panel) have harmonic distortion below the level that I can accurately measure in my usual testing environment.

Is there reason to believe that the DAEX25FHE has poor symmetry in the magnetic field? Is that reported somewhere, or is it obvious for some reason? I know Zenker claims the XCite has better symmetry than some other exciter, but do we know what that other exciter is? I have not seen the "other" exciter identified in thier literature (have you?) I know they mention their "double magnet" and I presume that has something to do with their symmetry claim, but I really don't know for sure. Your thoughts?

Eric