[zygadr]

Both have bass....................framed has more bass

Podiums are framed and have suspension.............they have bass.

[tommus]

Hey there foamheads!

I finally got my panels together and I have some pictures to share.

These panels are 2x7 feet each, of 3/8" (about 10mm) HD EPS, with six exciters per channel wired in a series parallel arrangement for about 4+ ohms per channel. 

The black tabs are loops of stretchy nylon spandex fabric glued to the edges of the panel with tacky fabric glue.  When stretched and stapled to the wooden frame they suspend the panel quite nicely and allow it to dance while damping edge noise. (I tried a full perimeter fabric suspension and it killed the high end!) 

The stiffener fins on the back are lap-jointed with each other and glued to the panel with Gorilla glue (next time I would use the plastic resin - see below).

I prepared a circular pad of several coats of urea-formaldehyde (plastic resin) glue for each exciter (see brown color under exciter foot in pix) and then glued the exciter on with rubberized cyanoacrylate (super) glue.  Plastic resin glue is the hardest stiffest drying glue you will find.

The wires are set into lengthwise slits in the edges of the stiffener fins and secured with more tacky fabric glue, which dries clear and rubbery.  Also I used this glue, cut with some water, to damp/seal all cut edges of the foam.

I have to say the sound is very very nice.  Acoustic instruments and human voices are astonishing!   
The sheer "size" of the music is amazing (well for the size of the panels it better be!)  Some crude EQ made them sound a lot better.  I include a photo of my EQ settings, which were done only by my (untrained) ear.  It would be nice to get some test equipment. (BTW, I am running the panels from 100w class D monoblocks and a tube pre-amp that I got from Arjen Helder.  Not to start an amp discussion but I think it is pretty good stuff...)

On the down side, while bass notes are present down to maybe 40hz or so, they are nowhere big enough to impress, and attempts to EQ them produce a lot of distortion.  A lot of that comes from the speaker wires flapping and buzzing, so there is room for improvement here I think.  However, some of it is the panels themselves... maybe 1/2", 3/4" or thicker EPS my be better! Also, there could be more done with the placement of the fabric suspension.  Playing deep test tones while touching different parts of the panel's edges made for varied bass responses.

For now I have no plans to move these from my workshop.  They sound too good!
I do plan to make a smaller pair, maybe 1x4 foot each.  Mated with dipole woofers it could be something I could keep in my living room! 

[CLS]

Hi,

Thanks for sharing. Well done! Congratulations

The arrangement of excitors and fins (ribs) are interesting. What's the exitors? (look like PE cheapo?) And, are they equally spaced?

I don't see any support for the magnets. They are just floating by themselves, aren't they? Are they working OK like this? Maybe some portion of the bass distortion comes from VC rocking (?)

I can't read the numbers in the picture of EQ. Is the major notch set between 1.5~2kHz? And it seems the very HF extension is OK (not much boost).

Please share more.

CLS:)
 

[tommus]

CLS, the exciters are the cheap PE ones in the white plastic casing.  They are equally spaced along the length and centered on the width of the panel.

The feet are just glued to the surface with no other support.
As far as I can tell they are OK, but you may be correct about the distortion.
There seems to be a fine line between preventing wire flap and possibly skewing the coils' position with the wire tension.  Lighter gauge wire perhaps?
I could try fitting a wooden back brace and silicone goop, but I will need to pull off the center fin to do this.

Sorry for the bad photo, my camera didn't want to focus on the laptop screen for some reason...
That notch is centered at about 1.5khz to smooth a "shouting" tendency with some (mostly live) recordings.  On better recordings I think it's not really needed.
The HF is pretty nice, but without this bump (i think it's about 5-6db) there is a definite flatness to the sound, especially with electric guitar, etc.  Once again it's a "one size fits all" approach because a lot of the stuff I listen to varies in sound quality, but this seems to be the minimum needed to bring life into the upper harmonics.   
Honestly I am still learning to assess the frequency response by ear.  I haven't heard enough really good systems to have an ideal to aspire to.  It would certainly help to get a cable long enough to make real time changes while being outside the nulls in between the speakers!  It's amazing how quickly my brain "resets" during the stroll from the EQ control to the listening position.

[CLS]

Hi there,

I'm wondering if it should maintain "distributed mode" (chaotically excited) throughout the whole spectrum, or the panel might be (should be) transferring to piston motion under a certain frequency, especially on those relatively rigid materials?

And in the case of using multiple excitors, it's easier to provide evenly spreaded drive to the panel, thus the tendency of piston motion. (if the suspension allows)

I guess there are complex factors, and it's also hard to decide an optimal frequecy point (zone) for this transition. What would happen at this transition?

Any thoughts?

Sorry if this has been covered in the previous posts. Although I've read through all posts in this thread, I really don't remember...

[CLS]

I think I should rephrase the 'piston motion' to 'the first mode of vibration'. I mean, there's only one big 'bump' on the panel at such situation.

I found this:
 http://www.speech.kth.se/music/5_lectures/wogram/modal.html

So, there'd be a lowest frequency that a single big 'bump' is generated on the panel. Below that frequency, the bump is too big to be completely supported by the panel. (If the bump is infinitely big, the whole panel would look like moving for-aft as a whole, then we can call it piston motion... ) Above that frequency, there'd be more than one bumps (and dips) on the panel.

Apparently, a penal of smaller area or more rigidity, this frequency would be higher.

Also, I guess it's affecting the radiation pattern. When there're many bumps and dips on the panel, it's more like a diffused omni source. While if there's only one bump, it's more like a dipole.

And the previous questions remains the same - there'd be a transition, where should we set it? And how?

[jackman]

This is one of my favorite threads!  Do you guys know how I can affix the little NXT's to a couple acoustic guitars without making perminant marks on the finish?  I'd like to use them as speakers in my office and quality sound is not as important as not ruining my guitars.  Will they work on solid body guitars also?  I've got several that might be good candidates because they have tough poly finish and I'm not worried about scratches. 

If I decide to use them on a guitar with lacquer finish, what should I use to make them stick to the guitar without making marks?  I wonder if they will help the guitars loosen up and sound better (like those devices they use on violins)?  It's a commonly held belief that the vibrations of years of playing will make a guitar or violin sound much better.  People believe this is one of the factors that contributes to the sound of very old and classic violins.  I believe the other factor is age (slow drying of wood and chrystalization of sap within the wood of the instrument).  Unfortunately, poly doesn't allow wood to breethe like nitro (lacquer). 

Anyone ever try this?  Thanks!


Jack

[tommus]

 
CLS wrote:"there'd be a transition, where should we set it? And how?"

Maybe trying to engineer a frequency-based "crossover point" for these behaviors is counterproductive - maybe the distinction between them is not relevant to a "coherent" source where pistonic and DML are concepts at the ends of a continuum. (couldn't you model a DM panel as a bundle of pistons?)  Is the answer to the question simply "make the panel as big, stiff and light as possible!"

Theory aside, I have been debating whether it is better to have the exciters fixed only to the panel (more bending waves) or backed by a rigid spine (more pistonic).  Ideally, it seems you would have both conditions at the same time!  With a 14 square foot panel, maybe you can...

...how about varying the stiffness of the spine over the height of the panel?  I have 6 exciters in a vertical line on my 7 foot panels.  Exciters in the one and six positions would get the stiffest backing to the spine, #2 and #5 get a softer one, and #3 and #4 just a little bit or maybe nothing at all.  Now the panel can still exhibit DML resonance (from about 80 hz based on a 7 foot half wave) but the top and bottom regions can kick it pistonic.  The foam stiffener fins on the back of the panel can be likewise tapered in depth or position.  Would this be the coherent panel equivalent of a WMTMW?  Doesn't a electrostatic manufacturer (Magnepan?) do something like this?  I am going to get pretentious here and call this "compliance tapering"  

Any thoughts?

[CLS]

The soundboards of various musical instruments are amazing, and also very inspiring. I've been thinking of building something similar to those structures.

Aside of the biggest piano soundboard, the ones of guitars are also very interesting (and beautiful). For exmaple, see this: http://russellguitars.com/rgblog/2009/06/deep-lake-bracing/

(how beautiful is that woodworking!)

I'm not saying we should mimic the making of musical instruments, but some (of many) factors are indeed in common. Here is the content of the complete lecture about acoustics of the piano I've quoted above:
http://www.speech.kth.se/music/5_lectures/contents.html

In this page, I saw interesting comments on the tradeoffs:

...The modal frequencies are determined by many factors, the primary ones being the material, size and shape of the soundboard, its thickness and grain direction, and also the material, dimensions, and placement of its ribs. Secondary factors include the characteristics of the rim or case to which the soundboard is attached. In general, the thicker the soundboard, the louder the piano but the less the duration of its tone. Soundboard design is often a compromise.

That highlighted sentence is quite the opposite of what I've thought. I thought the thicker the panel, the heavier it'd be, then it needs more force to be excited. So why is it louder? hmm....


And, the 'wavelength' on the panel is not the same as in air, because the sound speed propagating along the panel is different from that through air, mostly faster. So the wavelength should be longer.

[zygadr]

It seems to me that as I have previously found, bracing the back of the panels(whether they be foam or C.C.) presents another field of problems that are difficult to overcome.

What we need is what has just been mentioned : a large, but light, yet RIGID panel.

I believe that thicker EPS, chosen thick enough to hold it's own weight when in a vertical position is the next step in this massive experimental thread we have here.

There is no point debating over distributed modes against pistonic movements because we have both at work here as do the Podiums.We also have multiple exciters on each panel which once again is a massive deviation from NXT's typical DML theory papers of one exciter per panel.

On the subject of piano board thickness, I have noted that a 25mm thick EPS VH piece of Poly was MUCH louder than the same size in only 10mm..............interesting........eh?

[CLS]

 

Ah! I must see it from the wrong angle.

Transition high or low, is just the result of the combined characters of the panel(s).

As to large, light, yet rigid panel, it seems there's no better material than composite honeycomb board, huh? No wonder it's the choice of Podiums.

I've found a local provider of honeycomb board made of fiber glass suface/paper core. But the largest size is only 45 by 45 cm, much too small. I've asked him if it's possible to make it larger. And the reply was the paper core won't be a continuous structure when larger than that - it had to be joined by another piece. And no guarantee for the flatness. Pity.

It's a nice finding about the thickness. I'll keep that in mind.

And, what's wrong about the bracing? I remember it's good, no? All soundboards of musical instruments have ribs. My understanding is this is indeed for light and rigid.

[bobloblob]

Jackman-

There is a major fly in the ointment with using a musical instrument as a reproducer.  Musical instruments all give their own colorations to sound, which is what makes them sound different from each other, and why one chooses one particular instrument over another.  This is why, using steel strings, a piano does not sound like a guitar does not sound like a mandolin does not sound like a banjo, and why brass instruments made of the same materials all sound different from each other.  The very purpose of them is to add a particular color to the sound they produce.

This is the antithesis of a musical reproducer, which attempts to add as little of itself as possible, so that the sound coming out of it represents as closely as possible the sound that was recorded.

The bridge on a guitar works in fundamentally the same way as one of these exciters does.  It transfers the sound of the plucked string to the soundboard and resonant chamber of the guitar.  Likewise the bridge on a piano soundboard.  On a solid body guitar, you have Peter Frampton, who sang through the guitar.  A fun thing to listen to, but it was not intended to sound like his voice.  Most likely if you mount an exciter to a guitar the music will sound a lot like musical instruments played through a guitar, which could be weird indeed.  And where you mount it will affect the sound a lot.

This could be fun to experiment with, but I don't think the result will be a very close match to what you are feeding into it.

On the other hand, NXT says that you can put advertising on the front of the panels, so I don't see why you could not do the same with these wither by painting or by mounting pictures/posters of your choice.  Depending on how it's done, it may or may not affect the sound very much.  If this is just for an office system to have fun with, it may not matter that much.  I think you will have to look beyond this forum for thoughts on that, though.  I have seen some pictures and suggestions elsewhere for this, but can't remember where.

[jackman]

Jackman-

There is a major fly in the ointment with using a musical instrument as a reproducer.  Musical instruments all give their own colorations to sound, which is what makes them sound different from each other, and why one chooses one particular instrument over another.  This is why, using steel strings, a piano does not sound like a guitar does not sound like a mandolin does not sound like a banjo, and why brass instruments made of the same materials all sound different from each other.  The very purpose of them is to add a particular color to the sound they produce.

This is the antithesis of a musical reproducer, which attempts to add as little of itself as possible, so that the sound coming out of it represents as closely as possible the sound that was recorded.

The bridge on a guitar works in fundamentally the same way as one of these exciters does.  It transfers the sound of the plucked string to the soundboard and resonant chamber of the guitar.  Likewise the bridge on a piano soundboard.  On a solid body guitar, you have Peter Frampton, who sang through the guitar.  A fun thing to listen to, but it was not intended to sound like his voice.  Most likely if you mount an exciter to a guitar the music will sound a lot like musical instruments played through a guitar, which could be weird indeed.  And where you mount it will affect the sound a lot.

This could be fun to experiment with, but I don't think the result will be a very close match to what you are feeding into it.

On the other hand, NXT says that you can put advertising on the front of the panels, so I don't see why you could not do the same with these wither by painting or by mounting pictures/posters of your choice.  Depending on how it's done, it may or may not affect the sound very much.  If this is just for an office system to have fun with, it may not matter that much.  I think you will have to look beyond this forum for thoughts on that, though.  I have seen some pictures and suggestions elsewhere for this, but can't remember where.

Thanks Bob, I was only thinking about using these with guitars because I have guitars on the walls of my office and it would be fun to freak people out when they come in here and can't figure out where the sound is coming from.  I agree guitars or pianos would not be a good choice for accurate music playback.  Thanks for your comments!

J

[bobloblob]

Jackman

A thought I had was perhaps making a 2-dimensional cutout  painted to look like a guitar, and mounting an exciter or two to that.  I suppose you could even have some guitar hardware on it - hardware that won't resonate too much - so that to someone who isn't looking too closely it would look like the real thing.  It would not play much bass, but may cover much of the range of a guitar without changing the sound the way a real guitar would  It would be easy enough to test - initially just use some cardboard cut to the shape and dimension you want, and mount an exciter.  You could add other guitar things to it until it started to affect the sound, so you would know when to stop, particularly if you did a pair, and kept one without additions as a check.  Might be a fun project, and worth it for the reactions you would get.

[bobloblob]

Another thought:  For this project, you might wish to leave the feet on, since your goals are different, and it may be easier to remove the exciters from the guitars later if you choose to go that route.  Don't know about that, though.  If the exciters are easy enough to remove, there is no reason not try them on your guitars.  The sound may be good enough for your purpose.  Cheap thrill if it works.

[CLS]

Again:

....

In this page, I saw interesting comments on the tradeoffs:

...The modal frequencies are determined by many factors, the primary ones being the material, size and shape of the soundboard, its thickness and grain direction, and also the material, dimensions, and placement of its ribs. Secondary factors include the characteristics of the rim or case to which the soundboard is attached. In general, the thicker the soundboard, the louder the piano but the less the duration of its tone. Soundboard design is often a compromise.

Using the whole musical instrument as a NXT panel is surely not right (nuetral sounding), but if we take only the soundboard, I think there's some possibility to get pretty good result.

As the quote above (and the experience from zygadr), thicker panel is louder. Yet there is another interesting character is the decay is shorter. It's reasonable to be shorter because the internal damping would be higher in thicker material.

We need some inertia in the panel so the excitment can be distrubuted throughout the whole area, also we need some damping so it'd stop vibrating just in time.

So, we are indeed building soundboards, just in slightly different set of compromises from musical instruments.

[jackman]

Interesting!  You could get some thin wood like a soundboard of a guitar and use it for a panel. I've seen acoustic guitars for very cheap on eBay. What if you cut the back out and take the strings off and hang it on the wall with a guitar hanger. That would be cool!

[CLS]

I've mentioned this in previous post, someone in diyaudio built panels of (almost) the same structure and material of piano soundboard:  http://www.diyaudio.com/forums/planars-exotics/133711-piezo-nxt-type-panel-56.html#post2529266

But that person just disappeared and never answered my question That's a wrong place for this topic anyway... Or, that's only a teasing fraud?   

However, I do believe that can be good, as my experiment on a big piece of glass on a dry wall. A large and heavy wall (=panel) can be effectively excited and sound good, too. I don't keep pursuing that approach because of the cost, and the trouble in building big and heavy things... (big and heavy was my old style I've abandoned)

I own a cheap, old, unused guitar. By the available area (and without the whole box), I guess it's only good down to about 300~400Hz. A mid-high unit for a multiway ? 

[zygadr]

It would be nice to make our panels out of wood but unfortunately the material is very heavy compared to what we are used to.
Not only that, stick an exciter on a piece of balsa wood plank..................sounds horrible! .

I don't know what ''exotic wood'' sounds like, but can imagine that the wood like resonances will be a big problem to overcome?

[CLS]

Maybe it's because the balsa is too soft to effectively give and take higher frequencies...

As to the heavy materials, I don't know much. But in the experiment of glass/dry wall, I stuck only 2 cheapo. Yes the SPL is limited, but the sound quality is not dumb (or muffled, lifeless as those ordinary speakers of low efficiency). It's quiet but very clear sounding. At the resonance of the whole wall (near 30Hz), it's like earthquake, very scary. But it's a flaw, not a strong point. And the internal structure of the dry wall is squeaky when pushed. So it's not usable anyway, I have to filter it out.

Besides that too strong LF resonance, I don't feel obvious drawback in the heaviness of this whole dry wall. Yeah of course there are oh too many drawbacks in a dry wall, it's difficult to have a proper size and angle, at a proper position. And it's not movable and probably coming with some squeaky noise when playing loud...

Another killer drawback - heavy things tend to be expensive. Heavy, large and rigid usually means a lot of money.