I was going to comment, but if I did the moderators would probably delete my post.Any ways I think I will follow Spedge and lay low and let all you Gurus handle this DML thread.
How about here Andre:
The resemblence of the shape of soundboard braces to beams with acoustic black holes is pretty interesting. But I'm not sure if they really act the same way.
Eric
Most people dont that is why it took most of yawl 3 years to figure it out.
Duede,Some engineering, because I am one
Concerning thickness:
- Interestingly, efficiency is virtually independent of thickness. Efficiency is determined by the ratio of stiffness to areal mass^3. Since stiffness is proportional to thickness^3 and areal mass^3 is proportional to thickness^3 the two cancel each other, and efficiency is independent of thickness. Note, also that for good efficiency, stiffness is good, and low density is really good. Hence, high modulus, low density materials are ideal. Wood is pretty decent in this respect, and sandwich composites with stiff skins and light cores can be better. Polystyrene foam (EPS or XPS) is particularly good in this respect, although there are differences of opinion concerning sound quality with PS foam panels. Some love it some hate it.
- The lowest frequency you can expect to get from a panel will be determined by it's fundamental flexural frequency. For a given aspect ratio, boundary conditions and elastic modulus (or moduli in the case of anisotropic materials), the fundamental frequency (and all the natural frequencies for that matter) scale in proportion to thickness divided by area. Hence, a thinner panel will let you achieve the same fundamental frequency with a smaller panel. But also realize that just because your panel has a particular fundamental frequency doesn't mean you'll necessarily get output down to that frequency. The boundary conditions have to be right, and the exciters need to be capable at that frequency too. Boundaries approaching simple supports are better for getting output at the fundamental than free boundaries.
- Thickness also influences the panel's "coincidence" frequency. The coincidence frequency is the frequency at which the speed of the panel's flexural bending wave equals the speed of sound in air. At this frequency, and above, the panel will start "beaming" in the plane of the panel. Coincidence frequency is inversely proportional to thickness, so using a thinner panel may be better in this respect, as you may be able to push this coincidence effect up to or beyond the audible region. But in practice this seems to be an effect which does not seem to be as objectionable as you might think, and perhaps not worth worrying too much about.
Eric
@Veleric I think maybe we are saying the same thing two different ways?
For #1, for a given length and width dimension (cross sectional area) the mass will definitely only increase linearly with additional thickness, but the stiffness increase is cubed. This is one of the reasons why low density infill composite panels (like foam core or honeycomb) are so incredibly stiff for their weight.
Now that being said, after reading the G. Squicciarini et. al. paper linked to earlier in this thread, I do not think I will be increasing thickness and will stick with a ~3mm panel (most likely lightweight wood ply) which is readily available. I have also found a local sign company that will sell 4'x8' sheets of 3mm 'Alumalite' type material (a different brand). But I am going to try the wood panel first. The Alumalite material is great; I used to race cars with a guy that built his air dam and splitter with alumalite and he never had problems with it even at 120+ MPH speeds. But I digress.
For #2 Since you mentioned it I am thinking a bonded foam border approximates a simply supported boundary condition. It leans more toward free than clamped (cantilevered) IMO since there can still be some out-of-plane movement but for messing around it's close enough.
What you mentioned in #3 is also in that paper though I believe they refer to it as the 'critical' frequency or fc.
For #1, for a given length and width dimension (cross sectional area) the mass will definitely only increase linearly with additional thickness, but the stiffness increase is cubed. This is one of the reasons why low density infill composite panels (like foam core or honeycomb) are so incredibly stiff for their weight.
Now that being said, after reading the G. Squicciarini et. al. paper linked to earlier in this thread, I do not think I will be increasing thickness and will stick with a ~3mm panel (most likely lightweight wood ply) which is readily available. I have also found a local sign company that will sell 4'x8' sheets of 3mm 'Alumalite' type material (a different brand). But I am going to try the wood panel first. The Alumalite material is great; I used to race cars with a guy that built his air dam and splitter with alumalite and he never had problems with it even at 120+ MPH speeds. But I digress.
For #2 Since you mentioned it I am thinking a bonded foam border approximates a simply supported boundary condition. It leans more toward free than clamped (cantilevered) IMO since there can still be some out-of-plane movement but for messing around it's close enough.
What you mentioned in #3 is also in that paper though I believe they refer to it as the 'critical' frequency or fc.
Last edited:
I think he left because of 2 reasons:
1 - He said he preferred youtube as it's easier to upload results.
2 - He's tired of the frustration having to spar with and repeat himself about simple common sense things to people who are overthinking the entire thing and think that their way is right and the only way.
I also didn't think we were disagreeing. I was just pointing out the different ways that you can expect thickness to effect (or not) performance.
True dat. And also why those types of materials are good candidates for DML panels, so long as they are not too thick/stiff. Stiffness is generally good, not bad, except that stiffer panels need to be bigger to get output at lower frequencies.
I have tried both the 3 mm and 5 mm versions. I think you should try it, if you can get a good deal on it. The concept is exactly what we want, stiff skin, light core. However, the aluminum layers are pretty thick (about 16 mils as I recall). As a result, both versions are on the "too stiff" side, I think, unless you are okay with a rather large panel. I think it's probably not too far from the alu skin/alu honeycomb panels aagas uses, which are something like 6' x 2.5' if I recall. Probably great if you don't mind pretty large panels and/or a pretty high crossover frequency. But that said, I suggest you put Alumalite, plywood and polystyrene foam on your "short list" for trying out. All are worthy candidates I think.
That's exactly what I would suggest.
Yep, critical frequency and coincidence frequency are the same thing.
Eric
I did a search on here on using the same exciter on different panels. It wasn't clear if a favorite has turned up so thought try my own. Small 5w exciters so stick then to a plastic penny washer with an M4 plastic bolt and nut. Comes out at 1.2g. Exciters weighs 30g but panel weight will vary. To see if the 1.2g has much effect? Add more penny washers and note the effect. This just needs a 4mm hole in the panel. Can't see that having much effect.
Any thoughts on other ways of doing it? I want to try various thing such as edge support/others and this allows me to reuse the same panels. Weight of those - wont know until I buy the materials.
The cheap DML pc speakers I bought proved interesting but do need a woofer box of sorts to cover the tonal range of music such as Heart's Dream Boat Annie. They us a ~A5 sized panel. Amp in them rather crap, NXT patents mentioned.Tried a recording via a phone but didn't work out well. When I am set up I should be able to post sound files as @spedge did.
Any thoughts on other ways of doing it? I want to try various thing such as edge support/others and this allows me to reuse the same panels. Weight of those - wont know until I buy the materials.
The cheap DML pc speakers I bought proved interesting but do need a woofer box of sorts to cover the tonal range of music such as Heart's Dream Boat Annie. They us a ~A5 sized panel. Amp in them rather crap, NXT patents mentioned.Tried a recording via a phone but didn't work out well. When I am set up I should be able to post sound files as @spedge did.
I tried to find him on YouTube but failed..Does anyone know the contract details??
Good idea. But it's not the exciter weight you should be looking at, but rather the moving mass (voice-coil mass plus spider suspension mass.) This could be anything from less than a gram to almost 10 grams.
On the other hand, the panel itself probably weighs an order of magnitude more than your plastic washer etc.
By the basic laws of physics, it cannot work out well.
One cannot, properly, listen to a cell-phone recording as say that 'such-and-such a system' sounds good or bad or better or worse than anything else.
Even if you use a studio-quality recording mike and a 196k/32bit floating point recording, your room sound will completely stuff up whatever it is that you are trying to record. Even if you have a perfectly well-treated hi-fi room there are still reflections and delays and phase interference that will impinge on your recording.
Maybe an outdoor recording, with your speakers raised 5 metres (15ft?) off the ground, and the microphone carefully placed to ignore ground reflections, will give you a better idea.
But it's still not even close to ideal.
Because now that you have a perfect recording (taken with high quality gear in an anechoic or outdoor environment) it still needs to be played back on a perfectly flat, zero phase, anechoic system in order to reproduce what your speakers sound like. Any reflections in my system, for example, will interfere with the recording of your system. It's a chasing of tails.
The only thing a cell-phone recording (or even a studio-quality recording) will do is to prove that you've built speakers, and they have some semblance of sound. I do it too, for the visual aspect, obviously not for audio assessment. As far as a remote quality assessment is concerned, the frequency response, distortion and spectral decay curves might do a better job of communicating whether you have a decent quality system or not.
If you want to start getting get serious, then download yourself a free copy of REW, learn how to use it, and then buy the calibrated little Dayton speaker measurement mic. Seriously, people who call themselves speaker designers or hi-fi gurus cannot do without such basic tools.
I was comparing the sound of the actual speakers to the sound of the recording via my ears. The phone introduced some problems but there is another one as well. The possible need to compare the recording on the same speakers that produced it. I didn't. Youtube based but I can listen to that with both my usual pc speakers and the old NXT based ones and switch between each fairly easily. I can also listen to the recording.
Really this is no different to changing speakers in any audio system. Relative apparent improvements not measurements.
The panels when mounted with the usual double sided tape carry the entire exciter weight and the magnets have to handle that and the panel.
Anyway my initial interest will be frequency response of various arrangements.
AJohnL,
Sorry, but I'm still a little unclear about what exactly you want to test and what you want help with. Can you try explaining again?
Eric
This kinda proves the point, doesn't it?
If one's own recording on their own system doesn't sound like their own system, then it's rather pointless for anybody to post a recording of their system to show what it sounds like to anybody else.
Obviously if this is your method for referencing your own testing, then nobody can fault you. But that's not quite what I'm referring to.
Sorry there seems to be a bit of a misunderstanding. In terms of testing I don't want any help. In respect to that I want to do much the same as others. Different panel materials etc. This involves fitting exciters each time that is changed and i don't want to have to buy an exciter for each change so searched this entire thread for ideas on removing them and reusing them when attached as Dayton intend. Stuck on. Not many posts so wondered if there were any fresh ideas on reusing exciters each time a panel is changed.
My own idea is as I mentioned so will try it. It leaves a bit of a problem. What the added weight does to measured performance. I can easily add another gram and if this doesn't alter that much fine. If it does well I will have to accept that or get them off and stick them on somehow each time the panel is changed.
All for personal interest and nothing else.
Yes and no. For instance I have used a range of pc speakers and different speakers on the TV and amps. The sound directly from the TV I use now via it's speaker is clearly inferior. Some past ones have been much better. I recently used floor standing Yamaha speakers. They clearly have less bass response than another pair I have but those don't fit in well with our lounge.Too much floor area needed. Currently I'm using a 2.1 system. I'd say better than the Yamaha.
Digital sound recording has been around for areas like the BBC for some time. They have a history of developing their own kit. They can probably just buy good enough now. What they did with their early tests was use a panel of people to determine what bit depth was needed. They used people with very educated ears.
Why do some people prefer the sound of vinyl? Complicated subject but some of that will be down to sound processing before it is actually recorded on vinyl. There is info around on that,
Codecs used in audio? There are lossless ones around but but a sound file for instance will be many many times larger than ones that compress. Files can be downloaded and played but what people actually hear will depend on what they listen with. Eg. There will be samples of all codecs around.
https://helpguide.sony.net/high-res/sample1/v1/en/index.html
Netflix have their own standards that appear to be adjusted to suite domestic listening situations. Youtube - no idea but they may do similar
The aspect that has changed for me these days is that it's easy for amateurs to measure performance with rather high bit and sampling rates. Speakers have always been seen as the weak link in the chain. I doubt if that has changed. How for instance could people check distortion? It would need a distortion free mic. How to check that is? A bit tricky. Colourisation? Another tricky one. This area all comes down to our ears and how some sound engineer type feels it should sound when music is initially produced. Also what we choose to listen with.