I'm a newbie and I want to see more sharing from other people and discover experiences. I recently saw Arylic's amp board and wanted to make my own smart speaker. That's cool.
Trying to learn...found a conversation on the parts express forums...Thankful this group is not that group 😂
Sebastian,
Your posts are the opportunity to learn about bracing techniques :
- which wood is it ? which section?
- the hand planner is it to give the concave shape we see at end of bracing like on a piano sound board? If yes why to it after gluing?
I think I have found a way to put a bit more Life into the canvas panel without the ply.
I think I prefer the canvas panel without the ply, but obviously it will not be full range anymore.
It will not be far off from a full range unit, maybe 200hz .
I will let everyone know how my little upgrade goes (if At all) when I am sure it works ?
Also I was a little bored ,so decided to put a disc on the ply art panel, I'm afraid I did not notice much difference in sound ?
I then stuck the metal disc(a tin can lid) on to the front of the exciter area with a blob of blu-tack.
This created a large peak in the 7k to 11k area, cymbals were very loud and clear, and out of proportion , with a strange sound.
The CD might do better or worse , but I think I'll stick with the canvas for now.
I do actually have a CD around hear somewhere with an egg dome in the centre, but never got around to attaching an exciter to it.
If I get bored again I might attach the exciter.
Steve.
I think I prefer the canvas panel without the ply, but obviously it will not be full range anymore.
It will not be far off from a full range unit, maybe 200hz .
I will let everyone know how my little upgrade goes (if At all) when I am sure it works ?
Also I was a little bored ,so decided to put a disc on the ply art panel, I'm afraid I did not notice much difference in sound ?
I then stuck the metal disc(a tin can lid) on to the front of the exciter area with a blob of blu-tack.
This created a large peak in the 7k to 11k area, cymbals were very loud and clear, and out of proportion , with a strange sound.
The CD might do better or worse , but I think I'll stick with the canvas for now.
I do actually have a CD around hear somewhere with an egg dome in the centre, but never got around to attaching an exciter to it.
If I get bored again I might attach the exciter.
Steve.
Sebastian,
Thank you for the links.
I just go quickly through it but I won't say it in the same way for the double exciter configuration.
If as I did in the test you go from one exciter connected to the amp to two identical exciters in series keeping the same volume (so voltage), the SPL are the same. The small differences come for me in that the panel is not vibrating exactly in the same in both tests because of the exciter added.
In addition both waves, the front and the rear are modified.
The exciter are force generator. From the current sourced from the amplifier, it applies this force to the panel. The current is the effect of the voltage (the amplifier is a voltage source) applied to the exciter impedance.
For a given voltage, with two exciters, the force from each is divided by 2 compared to one exciter situation. So twice a half force same SPL.
Having 2 exciters in series is the possibility of applying a double power (ie 20W to 40W) so to increase the force... if the panel can support this force staying in its elastic limit.
The good news is the FR is almost the same.
OK for polyurethane. To be classified in the false good ideas.
Christian
Very rich post Sebastian! First time I see the second graph with the wording. It is the kind of card to share between DIYer to share the vocabulaty. With a simple EQ, it should be possible to create each situation on our favorite tracks. I remember that Harman proposed a training to detect the loudspeaker problems. Is it from that? They even selected tracks with a large spectrum for the detection. At the top was pink noise (yes!) and I think "fast cars"
- to go for 2 identical exciters that match your amp impedance capability sounds wise at this step... enough adventures in the panel itself.
- at this step, I won't conclude from the almost flat guitar spectrum that its bracing will give a flat response for a DML. May be, may be not if the bracing is here to compensate some other characteristics... More evidences would be welcome.
- the 2/5-3/5 points are related to a rectangular plate without taking into account any bracing or even impact of the exciter itself (this second aspect might be not critical for heavy panels). So not surprising you found other points.
- to have the possibility of measurements is in my view essential. Not easy to make conclusion but a good support. Many measurements in changing speaker or mic positions are often necessary. The FR change with the distance from the panel. you will see but I would say being at at least twice the panel large dimension might work. REW is a very good tool.
- Veleric is developping a tap test (see here). It will be interesting to have a feedback from your design with this technique.
Some months ago I probably wouldn't say that but now I see more and more the DML builder as a special instrument maker...
At this time I was ignorant of what string instrument makers do to tune the instrument.
Is a DML an untuned instrument?
Christian
PS : I can't find for now where you mentioned the technique adding some metal disc. For now to me it is sorcery in the sens I don't see which method can leads to the right design (position, thickness, gluing, material...). Hopefully some pleasant coloration?... You have seen probably that I am strongly technically oriented... nobody is perfect!
Last edited:
Got the DAEX25FHE-4, and connected 4 of them in parallel-series, attempting to position them in a similar way to Tectonic, with them grouped in the centre of the 500x330x25mm 20kg/m3 EPS panel.
The 50w exciters I have been using sure sounds much duller. Judging by the difference in impression I would have expected a bit flatter response in the top end, but still see a quite clear slope.
I do want a bit of a slope and not too bright sound, and compared to previous results I think I can use quite little EQ to get a very good graph.
Gonna do tests for power handling and SPL later. Impressions are that plate starts distorting at about 40% on the amp, and then it is as loud as when the 2x 50w started distorting at 60%....very loud. I'm starting to be convinced that plate is reaching the limit before the exciters and amp now, so it probably can handle around 70-80w as a rough guess, but at that level you do have a fair bit of distortion. Although less unpleasant than with conventional speakers, it probably would be better to split the plate and have 50w per plate instead and not have to drive them as hard.
The 50w exciters I have been using sure sounds much duller. Judging by the difference in impression I would have expected a bit flatter response in the top end, but still see a quite clear slope.
I do want a bit of a slope and not too bright sound, and compared to previous results I think I can use quite little EQ to get a very good graph.
Gonna do tests for power handling and SPL later. Impressions are that plate starts distorting at about 40% on the amp, and then it is as loud as when the 2x 50w started distorting at 60%....very loud. I'm starting to be convinced that plate is reaching the limit before the exciters and amp now, so it probably can handle around 70-80w as a rough guess, but at that level you do have a fair bit of distortion. Although less unpleasant than with conventional speakers, it probably would be better to split the plate and have 50w per plate instead and not have to drive them as hard.
Hello Steve,
I had also bad experiences with noise coming from the computer it is why I came to a hand recorder (not only the mic) which by the way is a good alternative to a soundcard. Mine (Zoom H1n) which not a costly one has a little bit too noise. I have some idea of a small preamp to improve that but it is another story. I will think to make so records I could share so that you have an idea of the performance (good or not...). A really good point is the possibility to connect directly an headphone so you hear directly what you can record.
Your technique and experience with phone is interesting. I think that basically the mics of a phone have at least the quality of standard electret capsules that are entry level mic for measurements. The trap is more in what the app does after.
So a simple app collecting both mics with almost no filter and running at 44.1 or even 48kHz is what we need.
Do you know such app?
Christian
Leob
The mystery remains... Your FR are really different of what I have experienced. In the bass there is no hump somewhere below 100Hz and in the treble no hump around 10kHz. We share now the same exciter model.
Is there in the use of 4 exciters?
Saying that, I wonder if I got such treble from EPS for now...
I think I have a piece of EPS I could cut to your dimension to see what I get. It is a lighter one (15kg/m³) and thinner (20mm).
As I think about it : for the high power application you are designing, the fatigue resistance of the material might be a question. Will the material be able to survive to a very high number of deep cycles.
Christian
I went to 4 exciters right away, but will start with one for the next plate I prepare to see how big the difference is in FR.
Through my experiments so far I have the impression that more panels and/or exciters almost always sounds better. At least when placing the exciters close to each other, which is what I have tried. Since each panel and/or exciter will have somewhat jagged response on its own, adding up several slightly different sources makes a smoother overall response.
For the 25 kg/m3 Peripor and Neopor it feels so solid that I'm not really worried about material fatigue, although the thought struck me as well. Not sure how well they age, and maybe after a few years they might to start crumble. Also, I will be using a HPF, probably around 150 Hz, which makes a very big difference in how much stress the plate experiences.
On that note... I've been experimenting with my panels with and without 150Hz HPF and the overall sound seems to benefit from the panel not being loaded on the low end. In particular I seem to get a weakness around 4x the low frequency peak that resolves when that low frequency mode is lessened. I suspect there is some complex interference there.
A couple of things I noticed about bass reproduction is that with a single exciter and no suspension the plate goes wild. I have mine hanging, and hold on to the lower corners to simulate suspension for now when testing. Sound seems much more focused with compliant suspension on four corners, especially the bass. However, when using multiple exciters the plate actually shakes a lot less, and the sound changes less when I hold the bottom corners.
The do sound nice and punchy, and at normal listening levels if I boost the bass a bit I do like the sound as a full range, but I cannot feel the bass as needed for electronic music.
Been trying to do some mixing, EQ and compression on DML, and it is very interesting experience. Somehow individual sounds and frequencies are a bit harder to pinpoint, so first it just seemed really bad. But actually you get a very useful perspective. Regular monitors will by making the sounds more separated make it harder to hear what frequencies collide in different sounds playing together, so instead of hearing the whole, you focus on the sound your changing.
DML are forgiving in a sense that things doesn't easily become really harsh and overall makes the mix sound a bit extra glued together like a nice saturation effect, but somehow at the same time they make it possible to hear when things become unclear and messy. I would say that instead of a regular monitor that exposes things that sounds bad in the mix, DML encourages things that sounds good. Not sure if I would do a mix on only DML, but seems very useful as an additional perspective for sure.
Christian.
this guy on YouTube used discs on the ply on his canvas panel to increase treble.
he is now using a CD on the front.
I decided(in my boredom)to see if this made any difference.
the one disc I put on the back did not seem to make much of a difference ?
So I stuck the disc over the coil area as in the video, but just used a piece of blu-tack to hold it on.
this made a big difference, but not in a good way, it produced a very large broad peak at around 8k.
Steve.
this guy on YouTube used discs on the ply on his canvas panel to increase treble.
he is now using a CD on the front.
I decided(in my boredom)to see if this made any difference.
the one disc I put on the back did not seem to make much of a difference ?
So I stuck the disc over the coil area as in the video, but just used a piece of blu-tack to hold it on.
this made a big difference, but not in a good way, it produced a very large broad peak at around 8k.
Steve.
Leob and Christian.
the mystery remains for me too.
no resonance hump below 100hz and a very heavy smooth roll off after 4k ?
Steve.
the mystery remains for me too.
no resonance hump below 100hz and a very heavy smooth roll off after 4k ?
Steve.
NaRenaud,
I totally agree. Wow. That's awesome. I've been wondering how to do nomex without prepreg. I have similar nomex, and similar fiberglass. Hmm...
Great post. Thank you!
Eric
I'm not mixing or throwing a party...but I actually like the bass sound from my DML better than my 10 inch powered sub. It's cleaner and has a better attack and less sustain making it sound less mushy...but also it has a steep peak and a steep dip in frequency response which makes it a bit unnatural for certain sounds. My sub is there because it has a more level output and also making the panels not produce below 150Hz makes them work noticably better in the kids.
Ive actually considered trying to make a dedicated DML sub and have it take everything below 300 and then my panels will sound even better I think, plus my ear has gotten really good at picking up what is NOT DML. I put my bookshelf speaker back in service for a bit just to remember where I started and it's disappointing. Yes, it's more even in response which is good, but the clarity from about 400-1500 Hz is now so bad to my ears.
I know this thread is "full range DMLs" but if you were going to make a panel specific for bass, what material would you choose?
Narenaud.
If you are not using for mixing or partying,
The quick and easy option would be the canvas art panel, but using a larger ply panel for better LF control.
The down side obviously would be that it is a full range dml panel 😄
You could try a small bass shaker on The panel I suppose ?
At a few hundred HZ the panel would be heavily into piston mode already, I would think ?
thinking about it , I would probably use my art canvas panel without the ply for everything above about 200hz and the art panel with the larger ply for everything below this.
This could be quite flexible as you could switch them easily to full range if needed.
I have not as yet tried using a larger panel on the art panel so can not guarantee it's performance.
Steve.
There are a few other options , but this would probably be the quickest and easiest option, I believe ?
If you are not using for mixing or partying,
The quick and easy option would be the canvas art panel, but using a larger ply panel for better LF control.
The down side obviously would be that it is a full range dml panel 😄
You could try a small bass shaker on The panel I suppose ?
At a few hundred HZ the panel would be heavily into piston mode already, I would think ?
thinking about it , I would probably use my art canvas panel without the ply for everything above about 200hz and the art panel with the larger ply for everything below this.
This could be quite flexible as you could switch them easily to full range if needed.
I have not as yet tried using a larger panel on the art panel so can not guarantee it's performance.
Steve.
There are a few other options , but this would probably be the quickest and easiest option, I believe ?