I tried Nidaplast honeycomb with carbon fibre (130kg/m3 CF tissue) skins. But raw Nidaplast is already 3x denser/heavier than EPS, and once you've added resin then it becomes unusable for my particular application. Maybe I can use this Nidaplast for some kind of "puck" in the future.
I've been trying to get hold of Nomex honeycomb. It's aramid fibre paper internally coated with phenolic resin. I will try Kevlar/UHMWP skins or maybe even CF skins. Apparently NXT/Tectonic etc used it with CF skins. Very very expensive...
This is exactly what I've done. I used small polycarb Twinwall pucks through tightly-fitting holes in the EPS, and covered with Kevlar skins both sides to keep material impedances as low as possible. You can clearly see the sagging HF...
In my previous experiments, I'd used a much larger puck and connected it to the panel via "legs" without connecting through the skins. Paper in this case.
This gave a beoootifully flat response, but kak sensitivity. Nice for Hi-Fi, but useless for PA.
Time for me to go back to drawing board metinks. I can easily get hold of Twinwall Polycarb here. So maybe I'll bark up that tree for a while and see what falls out...
Thanks Eucy,
This sounds like the problem I initially had with EPS where I was using properties parameters for EPS in unexpanded form.
I have tried Correx (Corflute?) which I believe is Proplex. And as mentioned previously somewhere, I could not use it. Maybe it's too highly damped.
You may have noticed from the Joppe's videos 5,6 years back, #8,591, the vibrations from "the exciter" had been brought across the panel, across the angled cut edges to the front. The paper that brought the vibrations to the front might have (or not) glued also to the cut angled edges. And, that little edge(s) brought the high frequencies out. He was using heavy foam board or such like. The motor came from the old rubanoide testing. (You can read about this motor etc in this thread in discussions between @sergiu2009 and @WrineX and others.)
According to the NXT patents, the transducer doesn't have to be an exciter as we know here, a round one. Anyway, what if you glue a thin paper tube (a slitted one, or cone like tube) to the perimeter of the exciter's coil former/bobbin and the other side to the front of the panel through the hole to transfer the vibrations?
Last edited:
This polypropylene dual skin insulation is branded differently in different countries. I tried 2mm, 3mm, 4mm and 5mm such sheets (left overs after a new floor laying). They were held at edges so they won't flap. Two people, who have better hearing than I, said they sounded gloomy. For me, they sounded dull.
Last edited:
I bought mine at Home Depot called "Coroplast". (4mm) Not sure if its the same as Correx/Corflute unless those are just brandnames.
Eucyblues 99.
6mm proplex has round flutes not square, so is heavier and more damped.
efficiency is fine for home use.
I will make a new recording at some time ,using a higher quality rate, to prevent HF distortions.
I let the recordings speak for themselves.
Steve.
6mm proplex has round flutes not square, so is heavier and more damped.
efficiency is fine for home use.
I will make a new recording at some time ,using a higher quality rate, to prevent HF distortions.
I let the recordings speak for themselves.
Steve.
Yep Steve... I realise that your 6 mm sheet differs from the 'norm' (presuming there is one)
Using a dome on the corflute lifted the mids/highs and brightened the slightly 'soft' tone...I can't seem to get anyone to try it though 😔..oh well...que sera..(I know you've tried lots of domes.. (but not the coke can one I believe?))
Eucy
Yes it should increase the highs. But only on-axis. A dome will operate pistonically and destroy the DML benefits. You might as well put in a normal tweeter.
For the use of the "soundboard" material, it is either,
- paper,
- natural wood,
- plywood, paper honeycomb, if possible already covered with paper on both sides,
- cardboard, if possible two-ply in between corrugated both ways,
- aluminium honeycomb, already covered both sides with aluminium thin sheets,
- aramid honeycombs, whatever that is,
- carbon fibre sheets,
- heavy grade EPS, as thin as possible, say 3mm, 5mm,
- glass
- or, an experimental composite from natural elements, which has to be thought out later.
As far as I understood from the many NXT patents/documents that the idea is to transfer the "pistonic" motion only at the point of contact at 90 degrees or at an angle from the transducer to the panel, but not allow the said "pistonic" motion to make the panel wobble, or keep it down to the minimum. In other words, the said panel should not hang freely. #8,552
Last edited:
I completely agree that at least some of the panel edges should be fixed, especially where the panels resonate uncontrollably at the edges. The fixing or clamping or damping will affect the response at lower frequencies.
I haven't paid any attention to this on my designs yet (sub-woofers will have to do the work for the time-being.) Once I get the HF sorted out, then I will refine the bass side.
But as far as dome tweeters go on DML panels, maybe I don't understand the constructional or theoretical details correctly...
If the dome is merely adhered to the front of the panel, on top of the existing panel material, then it cannot increase the vibration of the panel surface beneath it, and any perceived increase in response might simply be imaginary.
On the other hand, if the dome is somehow afixed to the exciter behind it, much like the pucks are, then maybe there could be an increase in HF. The thing is that the pucks still operate as bending-wave transducers, but I battle to see how domes could be of any benefit in a DML system unless they were made of a flexible material (paper, cloth, silk etc.) Inflexible materials (aluminium, Carbon Fibre, Kevlar, in such small sizes) will have break-up modes especially higher than 10k, which might come across as additional HF, but which are in actual fact just distortion since very few measuring systems will pick up even 2nd harmonic distortion above 10khz. (2nd harmonics at, say, 10khz can only be identified by sampling beyond at least 20khz to identify distortion as distortion and not signal.) Say nothing of 3rd...9th-order harmonics which need (microphone) sampling rates and (microphone) flat responses from 30k...90khz to identify the problems as such.
I still need to be convinced that domes 1. Increase HF frequencies, and 2. Operate in omni-directional mode like the rest of the panel does.
Am I making sense?
Theoretically that may be correct,
but that's not my experience... And it certainly doesn't destroy DML benefits.
Theories abound in this thread but until collaborative testing occurs there will never be consensus or real progress, and therein lies the problem
Eucy
Last edited:
Have you done any testing on this subject, Eucy?
I'd be interested in your findings.
I certainly have, but my findings are on a laptop which is 1250 ks away from where I am presently and likely to be so for quite a while yet.
It's not so hard to test, and it would be better to get a second or third opinion.
Eucy
Hi Thomas,
Ok, so I've tested this...
There's a difference! Red trace is the front side of the panel. Blue is the driver side on the back...
Oh, I should have deleted the "dome tweeter" stuff, before replying to you. 🙂
In Joppe's tests in those videos, he talks about why he did that and what he hears. It may help to listen to his other videos on the rubanoide testing (5, 6, 7 years ago). Doesn't matter what is the motor, but he brought the "pistonic" motion vibrations to the front and got the HF. There's a certain area in front of the rubanoide paper membranes, just before the "pistonic" motor that brings out the HF. Further away from the center (about 30% away), gives mids and lows. The discussions, testing between members in that rubanoide thread gives lot of insights. That too is a DM panel, with the vibrations edge transferred to the panel.
You see, the DM panel doesn't have to be exactly flat, can be curved, but shouldn't have sharp bends. There, they brought the thickness down to minimum, by using paper. Georg von Arco did that with Arcophon in 1924-32, Paddock did that with Linaeum with other thin materials.
We can have a flat panel, but we have to have it as thin as possible. But thin materials wobble, so we have to make them somewhat stiffer, by adding corrugations, honeycomb in between the surfaces, which only complicates the problem/solution. We can, by the way, slightly curve the panel/membrane get that stiffness, and fixing the panel/membrane at least on two sides. This would be good, if we are using the convex side for listening, which would diffuse the sound.
Yes, I suppose so. EPS/XPS are actually sound insulation materials, that is, their duty is cut the sound than enhancing it. So, the idea to keep the sheet as thin as possible. Also, the idea is to get the front surface start getting the bending waves (physical, not sound) running along it, so thinner the better. The back side would have waves running along it a few milliseconds before. This would be the same with any material, honeycomb or not.
In Joppe's tests in those videos, he talks about why he did that and what he hears. It may help to listen to his other videos on the rubanoide testing (5, 6, 7 years ago). Doesn't matter what is the motor, but he brought the "pistonic" motion vibrations to the front and got the HF. There's a certain area in front of the rubanoide paper membranes, just before the "pistonic" motor that brings out the HF. Further away from the center (about 30% away), gives mids and lows. The discussions, testing between members in that rubanoide thread gives lot of insights. That too is a DM panel, with the vibrations edge transferred to the panel.
You see, the DM panel doesn't have to be exactly flat, can be curved, but shouldn't have sharp bends. There, they brought the thickness down to minimum, by using paper. Georg von Arco did that with Arcophon in 1924-32, Paddock did that with Linaeum with other thin materials.
We can have a flat panel, but we have to have it as thin as possible. But thin materials wobble, so we have to make them somewhat stiffer, by adding corrugations, honeycomb in between the surfaces, which only complicates the problem/solution. We can, by the way, slightly curve the panel/membrane get that stiffness, and fixing the panel/membrane at least on two sides. This would be good, if we are using the convex side for listening, which would diffuse the sound.
Last edited:
Great find. Looks like a promising option – but at a price, as you mentioned. They will be only half the size of your current panels. At least the shipping will be free for you. Keep us posted - I am very interested in your findings.
Do you know what core your current Valcucine panels have? I suspect aluminum. Here we can get two kinds of honeycomb core – aluminum or Nomex, but both are unaffordable.
Art conservators use panels like these. In the past I could get various kinds of aluminum skinned panels over various cores (e,g., foam core, aluminum honeycomb, paper honey comb). One could get them in various thinknesses of cores and skins. Many such panels had wood edges (covered front and back with the aluminum skins) suitable to staple canvas or paper. A couple of suppliers will still produce custom lay-ups at very expensive prices.
I'll try this AlumComp panels as rear surround speakers.
Sign blanks (in the US) seem like similar the the AlumaComp panels. A couple of suppliers: Sign blanks & Maxmetal
My Valcucine panels have aluminum honeycomb cores, a 4-5 mm aluminum edging around the perimeter (covered by the skins), and the aluminum skins.
Attachments
That is a very useful measurement Andre, many thanks for sharing
Thanks for the info. So, you used the panel as you found it (as a cupboard door) - no cutting done? I was wondering if this panel could be easily cut with a table daw and the correct blade?