Composite And Non-tradditional Soundboards

[fryovanni]

I am moving this discussion from the topic we started talking about this in as it was way off topic.

So to catch up.Previous discussion

Question was raised about using two different types of wood for a soundboard to improve bass/treble responce. I replied that I didn't think it would work as anticipated (I was way off base as it has been done with success). The discussion has expanded a bit into the overall designs used in these applications, as well as other materials and composites that are being used or possible.

Peace,Rich

[fryovanni]

David, Mattia, Monkey Man

I have been looking into trying composite construction (not split, but maybe now that you guys are enlightening me it may be something to look at). I know LMI sells Nomex, and I have found a supplier of airplane composites that sell the same as well as other skin and core materials. Every since I saw Charles Fox's "Ergo". I have been fascinated by the construction. I am thinking about trying out a wood/Nomex/Wood combo as a first look into how it works and reacts. I am still a little unsure as to the best method for bonding the pieces together (any thoughts?). I may try carbon fiber one of these days but I really don't know how to approch it (or what to expect). I have so many questions as to how the top will react, but I think the best way for me to figure it out is to build the board and see what I get. I hopefully my past experience with regular construction will help me test the composite board and make some adjustments. I would love to hear more. Maybe suggestions on where you would start with a Wood/Nomex/Wood board. Things like top and back thickness. Would you try different wood combinations top/back - side/side. I have Sitka, WRC, Eng Spruce, Port Orford Cedar, Mahogany, Redwood all handy and ready to go. Special treatment of the bridge area of the sandwich. Basically any thoughts you may have or would like to through out.

I may build two boxes side by side for my next project (new jigs/molds being build now). One will be a more traditional design, and the second can be this Hybrid. That way I at least have some sort of baseline.

Peace, Rich

[Myka Guitars]

Cool, this should be a pretty good thread.

As for special bridge area treatment have a look at this page (scroll down): Frederich Holtier, classical builder

As for materials I would choose what you are already familiar with so the subtle changes will be more perceptable to you. I would imagine a significant change in the overall character of the top but with differing materials it might get lost in the mix. I do think a side by side method of experimentation is an excellent way to do this if you do go with unfamiliar materials. I mention this only because I speak with many people who change 5 or 6 variables at once and can't eplain their results.

I am not sure what would work best for gluing the nomex composite but perhaps epoxy is a goo dway to go. I would call LMI and ask them. They should have that information.

I'll try and dig up some more info and post later.

~David

[fryovanni]

Thanks for the link. That picture gives me a good direction to head. I will have to think about maybe a little unique pattern that could be routed into the top that may help with the bass/ treble balance (just a thought). I will see if LMI has any suggestions as to adhesives. Do you know if they develop an arch for these type of tops normally? I could use a pair of thicker plates and sand a radius into them, then route the pattern and glue them up (would be pretty simple). I will figure out exactly what I want for the regular constructed base model and use similar wood for this one (I will be sure to use sequential cut pieces).

Peace,Rich

[Setch]

You might benefit from a search of the MIMF library for nomex - I know at least one guy on there (Randy Reynolds IIRC) has built a number of classicals using nomex skin construction. I'm sure he addressed what glue he used to lay up the tops, and how he shaped the nomex layer.

[fryovanni]

Thanks Setch, I did a quick search and found his site. Pretty good information.Reynolds

The link to supplies he gave is the same company I was looking at before.Supplies

Another source for supplies-another supplier

I also looked over guitar master works tour for a bit of insite.-GMW

Seems like there are some differences in thickness, and "pad" designs. Which makes sense. I noticed some are going braceless and some are using lighter bracing. I can tell I am going to have to slapa few together for comparison.

Peace,Rich

[ToneMonkey]

Would it be worth building a table mounted jig, where just the tops can be changed over. I think that it would be the best starting point. Not truly reflective obviously, but saves building a new one each time.

Count me in on this project.

Can anyone here do Finite Element Analysis? - I haven't done it for a few years and haven't got access to the kit anymore. That would probably be handy in this case. Let me see if I can find anyone in my phone book.

Hell yeah, another project

[fryovanni]

Would it be worth building a table mounted jig, where just the tops can be changed over. I think that it would be the best starting point. Not truly reflective obviously, but saves building a new one each time.

Count me in on this project.

Can anyone here do Finite Element Analysis? - I haven't done it for a few years and haven't got access to the kit anymore. That would probably be handy in this case. Let me see if I can find anyone in my phone book.

Hell yeah, another project

That is something I have thought aboust (a test guitar that is). I had tossed the idea of a rim system that would allow me to clamp on a top and then exchange it. Couple of issues that I couldn't resolve.

1- I have no real good testing equipment to document the results.

2- I do not have a controlled space to conduct the tests in (my recording studio could be an option).

3- I don't have an accurate method of playing it the same way twice.

Even if I did have these things I am probably not qualified to accurately evaluate the data . For the time being I guess I am left with my ears, and passing it around to other fellas for feedback.

Finite Element Analysis . I think you are over my head. I will have to leave that to you smart guys. If I could perform these test or calculations I wouldn't understand the data. I think the only tool I am going to have is what little experience I have from making sound boards. I would be grateful for some input from your findings (if you can explain it to me).

I also noticed one of the supply houses was selling .06" and .125" graphite/nomex/graphite pre-fabricated panels. I wonder how they would perform? Price was not prohibitive. One big question that is still looming in my mind is the Brace/No brace and best methods for developing radius and or go flat (maybe the pre-fabbed graphite panels could be flat?). I really have to see how stiff these will be to know (or make a good guess). I am figuring I will leave one side thick, and compair a bit as I sand it down (that will help me to get a feeeeelllll for the strength and stiffness).

This is good timing for me in the sense I was just about to start a new project. I just hope I don't get bogged down to much in the guesswork.

[ToneMonkey]

These are the blokes that I was on about.

http://www.lboro.ac.uk/service/publicity/n...32_guitars.html

http://www.coolacoustics.com/

[fryovanni]

Interesting ideas. I am not too hip on full on composite guitars, although the high tech materials are very cool. I like the bridge on their guitars (at least the way the strings are captured).

So after doing a little research and some ruff calcs (very ruff and general).

The overall volume of composite tops is compairable to a wood soundboard (maybe composite is slightly smaller .4mm thinner??). The Honey comb volume amounts to about 50% of the composites volume. The weight of the composite is pretty close but slightly lighter than the wood soundboard. However the stiffness and strength should show a significant increase (I don't know if I believe the advertised performance, but believe it should be significant). This leads me to the brace/no brace issue. I believe from my first look into this. A non braced composite (similar in thickness/volume or slightly less) would require some bracing (all be it very light). A non braced soundboard may be possible if the thickness of the honey comb section was increased (say 3/16" instead of 1/8"). That of course is a guess. Trading off bracing which normally would be 8%?ish of the volume of a regular sound board for an increase in volume of 25% and weight of 12%. Seems to be a bit of a wash in terms of weight, but stiffness and maybe more importantly more uniform stiffness may be more significant to the performance. I am hoping I have my calculation correct on the weight/volume on the Nomex (not 100% confident in the numbers), and of course wood I am compairing would be stabalized Sitka (but that could really modify all these %, most likely if a favorable light for the Nomex).

I will keep digging. Please let me know if these assumptions I am developing seem correct or if you see a flaw in my thinking. I am just trying to get my bearings on these new materials.

Peace,Rich

[thegarehanman]

You'll have to bear with me on this, but I have a question. Has anyone ever tried to use a cross section of an ovoid(egg) for the contour of a top as opposed to a single radius? It seems like this could allow you to use a larger radius for the bass side and a smaller radius for the treble side. In theory that could be manipulated to tighten up the treble side of the sound board and loosen up the bass side. I mean, it seems like if just using two different woods for the top, changing one's bridge design, or modifying the bracing would have an impact, than using a top with that geometry could have a noticeable effect as well.

peace,

russ

[Mattia]

You'll have to bear with me on this, but I have a question. Has anyone ever tried to use a cross section of an ovoid(egg) for the contour of a top as opposed to a single radius? It seems like this could allow you to use a larger radius for the bass side and a smaller radius for the treble side. In theory that could be manipulated to tighten up the treble side of the sound board and loosen up the bass side. I mean, it seems like if just using two different woods for the top, changing one's bridge design, or modifying the bracing would have an impact, than using a top with that geometry could have a noticeable effect as well.

peace,

russ

I know some builder (Howard Klepper jumps to mind) only arch across the board, using a variety of different arches. I believe his is a cycloid curve (or something similarly fancy-sounding), but I'm not sure exactly what it does viz. a different arch.

Thing is, I think we have to be very careful when 'dividing' a top in to 'halves'; pluck a string, the whole thing vibrates, after all. One side may let certain frequencies through more easily, I suppose, but it doesn't necessarily have to the one 1" close to the bass strings (as it were).

[Myka Guitars]

I like the analogy of a speaker cabinet. Imagine if all you had was a woofer. Tweeters really add some dimension to the overall sound. This is how I envision the guitar's top. In fact the way I do mine is not in half at all but a gradual increase in stiffenss from bass side to treble side similar in concept to a steel drum.

Of course a single speaker cone does cover everything and illustrates that a uniform soundboard can also work. The trouble with this theory is that as far as I know it has never been tried. Even a Martin soundboard has areas of differing stiffness. The asymmetry is more back to front than side to side.

Russ, Tom Ribbecke makes a halfling bass that is close to what you describe. Check it out.

~David

Edited June 15, 2006 by Myka Guitars

[fryovanni]

I like the analogy of a speaker cabinet. Imagine if all you had was a woofer. Tweeters really add some dimension to the overall sound. This is how I envision the guitar's top. In fact the way I do mine is not in half at all but a gradual increase in stiffenss from bass side to treble side similar in concept to a steel drum.

Of course a single speaker cone does cover everything and illustrates that a uniform soundboard can also work. The trouble with this theory is that as far as I know it has never been tried. Even a Martin soundboard has areas of differing stiffness. The asymmetry is more back to front than side to side.

Russ, Tom Ribbecke makes a halfling bass that is close to what you describe. Check it out.

~David

Boy using speaker and drum analogys really work for me. Especially the drum as a unit. When you talk about optimizing a speaker cabnet with multiple speakers. It makes sense, but really it is two seperate units that are using seperate drivers. In this case we have a single driver and are just trying to improve maybe eq. the respose curve a bit. Any higher efficiency speaker will have a higher output to input energy. As far as power or energy used to produce low frequecy v.s. highs. Well obviously lows require more energy. If we get to loose to allow for more efficient lows we have a top that can't react as quickly or gets sloppy when producing highs. I believe it is not a matter of splitting the top in half per. say. I believe much more weight is placed on low end efficiency. I look at the Dovetail body shape, and it speaks volumes. A larger portion of the box is being optimized for bass, and still a part of the body is shaped to allow for fast respoce of highs. I don't know if that is the best approch? (really well thought out though). Looking at speakers again. Obviosly large drivers move big air, and cabnets can be tuned for optimum air pressure behind the cone as well as getting aditional respoce port tuning and such from the cabnet. Tweeters require very small hard cones. In an effort to get more range from a small package we get 6x9's. So the shape of the body and soundboard does seem as thought it would benifit from some areas that are smaller and other areas that are intensionally made larger. The oval below the upper bouts seems to be the area where the bulk of the work is produced. Maybe a wider oval is optimum, or maybe that should be more eggs shaped (maybe taller on the bass side of the guitar?), or maybe something along the lines of the dovetail? Pretty hard to say what is best, because you really have to balance stiffness of the soundboard and bracing with the shape developed. If the surface size increases. How much do we adjust depth? Do we use a wedge design? or maybe Russ's egg shape? So many factors.

The design of my next acoustic is taking a step in the wider oval direction, and I am going to stick with a moderate wedge slightly shallower. No radical changes. Just another step in the direction I thinkI want to go, but still in my comfort zone.

Man, That was a lot of work (Russ, you know how fast I type)

Peace,

Rich

[tirapop]

Can anyone here do Finite Element Analysis? - I haven't done it for a few years and haven't got access to the kit anymore. That would probably be handy in this case.

FEA would turn into a huge project all in itself. Typically you'd be doing this in conjunction with a lot of experimental testing. You'd need to come up moduli for wood/honeycomb sandwich panels (not in a book and anisotropic), damping coefficients, and work out end fixities/flexibilities. You could spend lots of time feeding the machine and trying to get it to match your test results, so that you could trust it.

[ToneMonkey]

Yeah, it's a bit of an evolved process. I can't remember what package I used to use (didn't use it for very long) but I seem to remember that it had wood features on it where a lot of the paremeters were already in. I know that different woods have different perameters, but I was thinking of more of a general representation showing the sound boards of the two different materials and seeing their interaction.

A few months ago, I was talking to a mate about changing the bridge so that the 4 bass strings were attached to the soundboard by a normal bridge. The treble strings would then have their own bridge which was attached to separte small soundbox within the guitar (by a stiff bar). That way you would have your woofer and tweeter (I used the speaker analogy when I explained it too).

Anybody fancy trying it out and then sending me the guitar........ I'll send it back of course

[fryovanni]

As much as it would be fun to try a split bridge/ dual box / dual soundboard acoustic. I think the chances of me creating a guitar that achives the intended purpose would be oh... like threading a fine needle with rope, while wearing oven mits in a hurricane, blindfolded after slamming a pint of Cuervo (1800 of course). I just wouldn't know how to attack a project like that.

Peace,Rich

[tirapop]

Separate soundboxes? Made me think of tricone resonators.

For FEA, at work, we use mainly NASTRAN and ELFINI. Some folks use ABACUS for non-linear composite analysis and analysis packages inside CATIA V5 CAD system.

I'm used to doing static analysis. I have no experience with dynamic analysis and how to use FEA to design guitar tops. Most of the dynamic analysis of guitars I've seen has been modal analysis: free vibration with mode shapes and frequencies. I don't know how you use that information.

Modal analysis is akin to tapping on an unstrung guitar. Putting on strings and then plucking them is something different. Just the string tension deflects the top, putting you into non-linear FEA. Superimposing a vibration on top of that pushes you near the limits of FEA and possibly into structural simulation software. That's completely off my map.

I'd be interested to know this sort of thing is done. Quiz your mates from school about how they approached it.

[fryovanni]

Separate soundboxes? Made me think of tricone resonators.

For FEA, at work, we use mainly NASTRAN and ELFINI. Some folks use ABACUS for non-linear composite analysis and analysis packages inside CATIA V5 CAD system.

I'm used to doing static analysis. I have no experience with dynamic analysis and how to use FEA to design guitar tops. Most of the dynamic analysis of guitars I've seen has been modal analysis: free vibration with mode shapes and frequencies. I don't know how you use that information.

Modal analysis is akin to tapping on an unstrung guitar. Putting on strings and then plucking them is something different. Just the string tension deflects the top, putting you into non-linear FEA. Superimposing a vibration on top of that pushes you near the limits of FEA and possibly into structural simulation software. That's completely off my map.

I'd be interested to know this sort of thing is done. Quiz your mates from school about how they approached it.

You guys are scary smart!

[ToneMonkey]

Separate soundboxes? Made me think of tricone resonators.

Tricones have a spider comming from the bridge which splits the whole signal over three cones. I don't think it transfers bass signals to the one cone and trebble to another. Could be wrong though. Luckily enough I have the plans for a tricone on my laptop, maybe I can come up with something that would do that though.

[fryovanni]

Well, I am making a "little" progress on my new jigs and molds. I picked up a couple tools and sclaes for a very basic method of recording deflection of the soundboards. I am going to try to document the differences in the boards. It will just be a simple test but at least it may give some reference in future projects. Hopefully if I can catch up on a couple side tasks I can get a bit of time for this project this weekend.

[ToneMonkey]

You could measure the vibration by laser speckle interferometry, but that might be overkill

Keep us up to date on how it's going

[tirapop]

Given the way the bridge straddles the cones and where the strings sit on the bridge, I think there's a bias in how the cones share bass and treble. More of the bass strings' energy is going to be driven into the pair of cones on the bass side and more of the treble strings' energy is going to go into the single cone on the treble side. I've got to believe that's what they were thinking when they came up with that arrangement.

All the cones are the same size, so, they probably haven't been optimized as "woofers" and "tweeters".

[fryovanni]

You could measure the vibration by laser speckle interferometry, but that might be overkill

Keep us up to date on how it's going

Interferometry

I don't know if trying to record vibration would be very productive as the soundboard is not really in place. I seem to recall Mattia talking about some kind of vibration testing (pattern?). After the guitar was put together (I think). Maybe he will pop in with some suggestions for good (simple) testing that may be helpful.

Tirapop- I have never really looked very close at a tri-cone. Do you have a site (link) to some info on them. It sounds like you may be on to something interesting.

Peace,Rich

[ToneMonkey]

Interferometry

Peace,Rich

I was just trying to sound clever

Tirapop - You may actually be on to something. I seem to remember that some tricones have two cones on the bass side and one on the treble side, then some have it the other way round. I'll ask my resonator building buddy when I see him.