The Plywood Discussion

[kpcrash]

For those that have been on here for any length of time, the discussion of materials comes up in one of many forms. Can I use XXXX to make a guitar? for example. Then there are the almost flameworthy posts from some who battle over which wood produces the best "tone" in an electric guitar. This will not answer those questions directly.

After reading MANY posts, doing some research, and even building a bass out of plywood, it stands to reason that by presenting some scientific fact into the equation we can have a discussion on "how" not "if" plywoods are best used in making guitars.

I am not disputing "mojo" at all - we all know sometimes things just work that don't have any scientific principles or proof behind it; however, after spending Thanksgiving dinner with someone who is both an extremely competent woodworker and a scientist with ridiculous knowledge, it's time for a good post.

Plywood - in this conversation, plywood is defined as any wooden substrate comprised of 3 or more layers not greater than 2 mm thick per layer, on average, bonded by a resin (typically phenol formaldehyde) or glue, where each layer is laid in opposite directions of the prior. By this definition, if you take 3 pieces of 1/40" veneer, glue them together with the middle piece cross-grained to the top and bottom you have made plywood (technically, a laminate - but now we're splitting hairs).

Plywood is not "chipboard", OSB, nor is it MDF, nor any product with a foam core.

What's typically in plywood? This depends on its' intended purpose. For example, Marine plywood must be extremely durable, and is usually made from like African mahogany, sapele, and okoume, but depending on where in the world you are, you'll get different results...

U.S. -> U.S. Products Standard PS 1-74 specifies that softwood plywood is Douglas fir No 1 or western larch, with waterproof glue and no more than 9 corrections (putty holes) on a 4x8 "A" faced sheet

U.K. -> British Standard Specification BS 1088 allows untreated tropical hardwood veneers with suitable level of resistance to fungal attack

Aside from Aircraft use, which specifies - African mahogany or American birch veneer laminated in a hot press to a hardwood core of poplar or basswood with waterproof glue. (Hmmm.... is it just me or do we see these species frequently?? )

What's wrong with plywood? Well.... in some regards, plywood is the SPAM of the guitar world. Some people really like it for various reasons, even if they're "in the closet" about it - others will simply have nothing to do with it. It's mystery as to what's inside and a conundrum of being satisfied with the end result.

It's also been abused and relegated to a nasty word in some circles of luthiery. Not to mention that since it's open grain on all 4 sides, you have to contend with that as well.

What's good about plywood? Well... it's got a corner on the stability market. Engineered plywood is manufactured in a consistent manner. This means you'll get the same results sheet after sheet. Lucite (a resin) has the feature of preventing unwanted resonance and vibration. (we could have an entire discussion about lucite bodies, but that's for another day). It comes in just about any flavor you're looking for - mahogany, walnut, birch, maple, oak, etc.

What about jointing??? Scarf joints if necessary. Rule of thumb is make the scarf 8 times the thickness of the plywood. Carving is not out of the question either - consider a rotary tool.

The science is that on electric guitars the majority of sound collected by the pickups is produced when the magnetic field over the pickup coil(s) is altered or broken by striking a magnetic object(string). This creates an signal passed through potentiometers and capacitors to an amplifier with additional capacitors, etc. and possibly an effects rack in the middle. The reason for stating the obvious is this - once the signal is grabbed by the pickup - rules of electricity take over (mojo aside) and handle the "sound" of the guitar. Grounding, magnet material, coil material, string makeup all can immediately impact this sound.

Wood, has an infinite number of variables because of nature. Will every piece of mahogany produce the exact same tone? This is scientifically impossible because of the basic laws of nature. Solid wood is used because there are features that are generally accepted to be associated with a species. Maple has good tensile strength, so it makes good necks. Often, necks are reinforced by truss rods and other support rods. Will something else make as good of a neck? If it has comparable properties, yes. This is the difference between science and opinion - there has to be a logical explanation.

Could plywood make a neck? Scientifically? Sure! Martin has done it on several models. Many, many manufacturers have used plywood on even some of the highest-end models - either for looks or practicality.

Hopefully this helps answer some of the questions on using plywood.

In the above, I have dutifully tried to maintain only fact - please amend corrections with sources as noted.

[Keegan]

Alright, I'm going to start with the assumption that density and grain direction are the most important factors in a wood's tone.

Scientifically speaking, metal has the best acoustic properties because it's grain is very uniform and has a fairly flat frequency response.

But our ears don't want to hear a lot of those overtones, so wood is what we use for instruments.

If you look at a chart of wood densities, you can see that the properties we perceive are almost perfectly correlated to density:

http://www.simetric.co.uk/si_wood.htm

Ebony and Maple are very dense, and also very "bright", and the "brightness" goes down with density. These are generally not used for bodies as they would sound "harsh". They are used for necks however because they offer stability and also can make the whole guitar sustain better, though there's not enough neck wood to significantly brighten up the sound.

Plywood could make a good body wood if it was the right density, but I'm not sure the implications of the grain would be good for tone.

Also, plywood is a pain in the ass to work with.

Some sort of molded material would probably be the best alternative to wood, given it had the right density. And indeed, this is what we most often see manufacturers turning to when they try to find alternatives to wood.

The only real benefits of using plywood are the cost and stability. The cost is really a non-issue if $50 more could make the difference between a great-sounding guitar and a poor one. The stability isn't even important to the body of an electric guitar, because the neck or neck join will break far sooner than the body ever will.

I could only see it realistically being used for necks, because it has a similar density to maple, and with lots of plys it can be even stronger than quartersawn wood. But the trouble would be making a good-feeling neck.

I think if I were to take a shot at a wood-free guitar, I would use something molded for the body, like hemp: http://www.guitarsite.com/news/electric_gu...made_from_hemp/. For the neck, I think aluminum would be good, because it has all of the properties you need for a neck: stability, feel(if brushed, that would be a fast neck), and sustain. The fretboard...I'm not sure what the fretboard would be made of, you'd want it to be slick. Perhaps it would be a 1-piece aluminum neck/fretboard with the back brushed and the fretboard polished. The frets would be stainless steel, because nickel alloys blow and wear way too fast.

It looks like someone else already discovered aluminum as an awesome neck material: http://www.bergeronguitars.com/id19.htm

[fryovanni]

You make a lot of good points, but I think some conclusions could use a little extra consideration.

First of all wood is a fiberous material. The orientation of those fibers depends on growth patterns, environment and species. Although there is a lot of variation in these factors they are not without some sense of uniformity(which can be controlled to a degree with selection). Actually this is the tool used to develop the properties of plywood. Plywood as an engineered material uses averaging of variables to limit variation in charictoristics, as well as make use of more of the material which is available(some of which would not be suitable for a given application on its own). Wood that is comprised of wildly twisted fibers can be sandwiched with other material that is oreinted in alternating general orientations to create some sense of predictability. Now that does not mean the wood is the ideal or even close to ideal per. its total volume, but it has been made acceptable at a certain lowered level of performace within a reasonable range of consistency for its intended use. The sum has become better than the weaker parts it is made up of.

If the application requires stability and strength in all directions, and the requirement for strength is in a given range, and weight or efficiency of the material is less critical(Need more strength add more sheets), and of course cost and availability of material is a factor. Using a construction grade plywood for constructing a strong wall or table may be an outstanding solution. Not all the wood used to make the plywood would have met the requirements, and certainly would have not had the same directional strength or stability on alone. A great use of wood for sure.

We have to recognise though that you can select wood that is more predictable, has fewer flaws, and orient it as plywood is oriented and assembled. Then come up with an engineered product that is more consistent and stronger than a common construction grade plywood. It is all about averages.

To expand on strength and stability. Both of these factors relate to the orientation of the woods fibers. If you need strength on two directions equally it makes perfect sense to take advantage of alternating the fibers(same is true with carbon fiber layups), but if your application requires a significantly higher strength in on direction then you are gaining nothing from this alternate orientation(quite the contrary, half your wood is oriented cross grain which holds about 20% as much strength relative to the direction the strength is needed). Stability in terms of expansion and shrinkage with moisture has similar directional relation and you can actually reduce your effective stability in a singular direction if you choose to mix cross grain orientations(longitudinal orientation varies significantly less than either the radial or tangential orientation.

Since the body on a guitar or neck has very directional strength and stability requirements. It would be a wasted compramise for these reasons, although a solid body is VERY over engineered vs requirements to resist string tension(remember an acoustic uses heavier strings than an electric, with a lightly braced, approx. 1/8" thick soundboard and survives) and you can get away with less. So really the advantage to using plywood(again unidirectional and or construction grades) comes down to availability and cost. Lower grades of plywood are cheap and available so this is an advantage.

If you actually have an application that requires unidirectional strength and stability. The plywood may very well hold an advantage. On acoustics the tail block and even neck block have these requirements, and plywood has become very popular. So it all comes down to application to me.

[psw]

I have a lot of interest in this at the moment...I'm glad to see this subject come up...in fact I tried bumping up the best thread on this recently...but it faded pretty quick...so nice to see another one...this is the thread...

LINK

Bit busy right now...but it is an interesting discussion and fryovanni has excellent valid points as always...certainly food for thought.

A few short observations...

Something like this martin neck does not look like it meets the definition of "plywood" that KP has proposed. It would appear to be a laminated neck with a lot of very thin laminations...but all going in the same direction. It is common to make a neck blank form a few layers for stability and appearance...this would seem to be taking things to the extreme of that concept.

My interest is not so much from a "tone" debate...it is true that if you are going to put a lot of work and high quality hard ware and pickups etc. the least you can do is spring the extra $50 for a good piece of timber. Also, cheap plywood will not be good...you don't want any major voids and things like marine ply is expensive.

However...it is a material that I think could be exploited for it's look and it's strenghts with appropriate strategies to good effect...and I have some ideas that make the concept seem a logical choice in some ways if you were going to take the whole guitar as a single piece, over a body with a conventional neck attached.

Anyway...been thinking about this a lot lately, but for a standard guitar build, it may not be that good a material...for something more radical, it could be great. My thoughts on the neck for instance would be to glue up ply with internal layers of CF matting and epoxy...the ply then is much like filler and is similar to the parker fly concept where they can use even basswood for a neck and coat it in an outer layer of glass and epoxy to make it strong. My present thoughts are internal layers that provide all the stiffness and strength and the plywood there to provide form...a lot of which I would propose hollowing out. Using traditional timbers where the bulk of it is removed for chambering seems to be a little wasteful.

Also...a 335 is essentially plywood with an internal mahogany core...so it does have it's uses ...I think there is some merit in it with the right application and potentially it could be made to look very cool...if it is used as a feature. Not a high end guitar by any means...but definitely something to consider...

pete

[kpcrash]

You guys have all added some excellent points - thank you. I plan to conduct a few experiments over the next few weeks to actually attempt to measure resonance, hell full frequency spread available from plywood and solid wood. I'm curious not only to exact measurements of impact between two styles of wood, but also, what is the impact of bridge/nut materials - from a purely scientific standpoint (brass, steel, bone, wood).

Part of this comes from a challenge (I love a good challenge sometimes). The challenge is to prove the following:

1. A solid body guitar makes sound between two breakpoints - the bridge and the nut.

2. A certain amount of sound is made by the guitar itself because of string tension in point 1

What is the exact influence on frequencies produced (based on averages) between solid wood and plywood given a common set of hardware?

For this I'll be using "standard" chrome plated ferrous material for a bridge (since a plywood bridge is beyond this discussion) and a plastic nut, and a bone nut. I plan to use a bass template set recently purchased from zyonsdream for consistency between the builds so there is no influence by body shape. I haven't decided yet on the material for the solid body, so if someone has a suggestion on what (other than mahogany) is a proven stable, ultra-resonant wood, let me know and I'll try and get a chuck in 8/4 as there can be no glue in the body to prove this point accurately.

As for neck, on the plywood build, I'll make a plywood neck (from the same ply as the body) - on the solid wood build a standard maple/rosewood neck. Both will be bolt-on. Some might say "KP! Why the hell would you build two guitars just to prove a point???" Simple, I only have faith in one thing - everything else I need to see rock-solid proof.

I've got a list of models here with me that are plywood based, but that doesn't tell me "why" they are. Plus, on a selfish note, not only do I love this challenge - how great might it be to create a plywood instrument that sounded great at the wood level, then for the electronic side, simply add a psw sustainer? And study those effects on a live audience?

[fryovanni]

I think it is important to look at engineered materials and why or what the intent was. As I was saying, using construction grade plywood as an example. Was that the choice of material and intended application are sensable, efficient, and effective. If you are designing a composite for a neck you should certainly look at the needs and materials you have available. You can then design your material to best suit that application.

The Martin neck is a great example. The material that is used for the neck makes sense. We have fewer large, straight grain, clear trees available today. By laminating more pieces that are smaller we are able to use pieces that are all clear and straight grained, all of which can be oriented in the best way to provide strength and stability for a necks requirements. Very sensable.

Using other components such as heavier yet stronger materials surrounded by lighter materials to take advantage of design principles that can add to the efficiency is sensable(think of how we construct beams, in situations requiring directional strength). You can follow those principles in body construction as well (think aircraft floors, walls, wings). Now the key to a really good design is going to stem from a really good understanding of your needs. Which is kinda where KP is heading with his research.

The trick and where this line of study becomes a challenge is when you get to the subjective points. Basically the requirements will vary with the intended performance. Unlike orchestral instruments we embrace all sorts of distortion from pure tonality in the timbre of our instruments. So your solutions will never be "one size fits all". I think understanding or being able to accept that will help keep your findings more meaningful. By that I mean you will find "soft" material acts in such a way, which is what it is(good or bad? depends). It will also allow you to look at the fact that material is soft, but is also light, and that in a larger volume it has the same weight as another material however it is three times as stiff. A greater unbiased understanding of the materials we can use will give you a bigger arsenal when you are trying to achive a design goal.

P.S. Kp, When you say this-

I plan to conduct a few experiments over the next few weeks to actually attempt to measure resonance, hell full frequency spread available from plywood and solid wood. I'm curious not only to exact measurements of impact between two styles of wood, but also, what is the impact of bridge/nut materials - from a purely scientific standpoint (brass, steel, bone, wood).

If you want to find meaningful information, keep your test simple. You are going to find there are so many variables that may impact your findings in a "full frequency spread available from plywood and solid wood"(not exactly sure what that means) that your results may not have much real usefulness. I have found even the most simple and direct models have variables I did not recognise. Just thought that was worth mentioning.

[kpcrash]

Now the key to a really good design is going to stem from a really good understanding of your needs.

Precisely. I'm just looking to separate the "requirement" from the "desirement" for lack of better terminology.

I wasn't really clear in my prior posting about the testing, what I'm looking for specifically are variances in resonant frequencies (those not attributable to electronics) between the materials. My assumption here is that this eliminates the subjectivity of tone and is measuring the "quality" of vibration produced when two equal strings are struck between two endpoints of equal quality with the only variable being the substrate supporting the end points.

The neck conversation also interests me greatly. My goal is to start with a quality, furniture-grade plywood. Meaning, the face sides are clean of defects and contains a composite of the facewood as the core. Using multiple thin pieces to create a neck seems logical as it would reduce the chances of pockets of air along the length of the neck, thus making it solid. Most readily available locally is birch and maple. The need is to have a wooden substrate as thin as possible, capable of remaining at a given angle under tension sufficient to create 440 Hz waves, produced by 4 strings varying in diameter from .045" to .110" while maintaining plyability sufficient to support arc adjustment via threaded brass rod down the middle line. Can a plywood neck support being thinner than normal? And would that then give an edge because the strength afforded by using engineered materials would create an easier playing environment? Not that any of this was my original goal, but it could be an interesting side effect. Your beam statement does make a whole lot of since - given how we've gone to using the engineered joists for just about everything because of increased strength and overall consistency of product.

This is not really about cost savings or anything like that - purely - proving, scientifically, that there is a difference in using plywood and solid wood and perhaps a measurement of what that difference is.

[Keegan]

Might want to use the method described in this report to get started: http://online.physics.uiuc.edu%2Fcourses%2...inal_Report.pdf

Mahogany balances well with perceived loudness, that's probably why we see it as such a good tone wood:

http://en.wikipedia.org/wiki/Loudness

The curves on that graph represent equal loudness as we perceive it. Ex: If you had a 20dB 1000Hz sound(about where the C on the 2nd string is), a 330Hz sound(low E) would have to be almost 30dB(10 times more intense than 20dB) to be perceived as the same loudness.

Mahogany's resonant peaks are highest in the lower registers, making it sound very full and balanced.

I'm not quite sure how you'd get a setup like theirs, the piezo transducers are easy enough, but I don't know what you'd need to receive the data or create the waves.

[psw]

I'm not too sure about the validity of this test or the results coming from it. A bass has a limited number of frequencies and often someone will like a more direct and immediate "piano" like tone. It also is less likely to be played poly-phonically, and the frequency spectrum is very low compared to a standard guitar say. The results may not apply more broadly, nor to a guitar in general.

I don't think in general that a plywood material has any advantages and likely many disadvantages if used in a traditional manner...unfortunately, a side by side test kind of demands it. Also, a plywood neck is more radical than a body in concept but could have a huge impact especially with the long scale of a bass...this could seriously impact the perceived result with something subjective like "tone".

I think there is an established principle about grain direction and resonance...core layers across the grain may well work as a dampener. Flat ply sacrifices longitudinal strength for overall stability, not necessarily what is wanted.

Then there are the different wood varieties...plywood of birch compared to a solid birch may be a more realistic test for comparison if you know what I mean.

...

There are two ways in which ply could be used I was thinking last night. One is in a flat plane...but another interesting approach is with the wood laminated upright (kind of like that neck if it were 'real plywood')...imagine if the entire neck and body were made like this.

Interestingly, I recently found a plywood manufacturing company...I had been passing them for 2 years once a week now but until this came up, had no reason to investigate...I thought they were an average supplier...but they tell me that (for a price obviously) they can custom make plywood with anything you want...potentially you could order multiple layers, all going in the same direction if you wish...with a mix (or not) of any number of veneers.

I have a feeling that to make best use of plywood, you either need to exploit it's qualities for best effect, or compensate in other ways. For instance, by removing large amounts of internal wood from a plywood body, you could have a lot of the resonate qualities of a semi-solid without as much waste, weight and with significant manufacturing advantages and fully exploiting the stability to have thin strong planes without the risks associated with cracking associated with a solid timber of such a thinness. There is the potential also for bending if you were to use thin laminations (many layers of 1/8" ply for instance) and gluing up a blank under tension in a curved plane...making solid shapes like a curved top with a matching concave back eminently possible. Such a structure would have many more layers that are logitudinal to the cross layers that are doubtful to our cause in a guitar.

I think that it should be possible to make a thinner structurally sound body from plywood also...or one with more radical material removal in the carving and still have a strong structure...these are the kind of things that attract me to the material...how can it be used in new and effective ways that exploits it's qualities...rather than in traditional ways for reasons of cost reduction. It would only be in using something in this way that it would prove competitive.

There are enough plywood strats and solids with bolt on maple necks to prove that there is indeed a difference (even with the strat having the strings mounted into a large chunk of metal) There is also plenty of evidence that saddle material for instance makes a big difference (as does the material of a strats trem block...steel vs. cast)...what there hasn't been is a use of the material to it's best advantage.

So...part of my present mind meaderings is to consider...what forms can plywood offer superior reasons for use over solid wood...and do they apply to the guitar.

...

Anyway...just a few random thoughts. One thing that interested me was that potentially you could make an entire neck and body blank from a single piece and shape something really interesting and eliminate a lot of things like neck joints without enormous waste. You may also be able to exploit alternative timbers like pine (most ply here seems to be hoop pine or hardwood like luan) but with stability that this material often lacks. The glue joining itself may add to the strength with quality material. If laminating up a blank, there is further opportunity to exploit this further by adding things like glass and epoxy, or even CF to provide other qualities to a composite material while still maintaining a lot of the working qualities of wood and without the need to go the parker route and have an external exoskeleton to provide strength, alternate material selection, radical form, light weight and interesting resonant qualities....and all the specialized tooling to do this properly.

Also, I kind of doubt that you could make a plywood guitar with conventional and traditional approaches that is "superior" to a high end guitar made from traditional "tone woods" ...but perhaps you could make a radically different guitar that is a lot cheaper that exploits the qualities of the material that surpasses a mid range instrument and has other advantages (conservation of quality materials for use on high end stuff...or leaving them grow another hundred years so you can harvest something more substantial in future generations. From an environmental point of view, plywood does make very efficient use of the tree...being shaved around it means far less waste compared to sawing of lumber.

In general though, you can tell some of the comparative qualities of materials as "tonewoods" buy hitting it..hit MDF and you can tell it is dead...hit plywood and it is a lot better...hit an equivilent fine timber well dried, and it may well ring like a bell. On the other hand, some very 'resonant' wood not only rings like a bell, but there's a big dent where you hit it...

enough from me already...interesting thread...I'm thinking I might go out and find some scrap plywood and take a sander to it to demonstrate to myself some of the aesthetic and form possiblities ply might have to offer...

pete

[Mattia]

Plywood's great for certain things. It's pretty stable (mostly - although it's not made of premium grade wood, not necessarily of perfectly dry wood, and warps worse than any of the kiln dried lumber I've got, if we're talking averages), it's got a ton of constructional strength, but it has a number of major disadvantages.

1) Price. Plywood is expensive compared to a number of tried and tested hardwoods (basswood, alder, african mahogany).

2) Workability. Construction grade ply is not void-free (and is still more expensive than solid wood. Counts double for quality birch ply), and the alternating grain means you have end grain EVERYWHERE, covering 50% of every single edge of your workpiece. I don't know about you, but I'm no big fan of end-grain. Makes scarfing a fairly unattractive proposition, among other things.

Lamination is something we all do quite a bit; in necks for stability, in bodies usually for looks. Basically, I don't see any reason to go with an overengineerd, more expensive, more annoying to work with product, and I'd like to point out that you can probably do a lot of the testing you want to do without building all the guitars yourself: buy a cheap plywood strat or tele, build one out of wood (or get a good wooden one), and swap out the hardware and pickups. Voila, you've done the testing.

Pete: the 'plywood' applications you seem to be talking about that are most useful are ones that require you to make your own ply, essentiall, by forming the layers over a mold. Perfect way to make material-saving arched tops and backs - although the acoustics aren't the same as a carved instrument, plenty of experiential data on that from the archtop world. Alternative shapes in different directions would work as well, although again, likely not cross-grain, and bent laminations need to be bent and laminated into a shape, can't just grab ply and force it into a shape.

I use ply in places I think it makes sense - to make molds, templates and fixtures (although I prefer MDF's workability for templates, and transfer to phenolic/trespa once I have a 'perfect' template), and as end blocks (home-made ply, with scraps of hardwood, because I have them anyway) in acoustic instruments, where the impact and crack resistance is superior to any solid block of wood.

[psw]

Yes...some very valid and strong points there mattia. I don't think it is a great substitute for traditional woods at all and there is plenty to support that...as well as guitars you can prove it to yourself with. There is also not a lot of plywood qualities that immediately lend themselves to this kind of thing if used as a direct replacement.

I do think there is potential though...but only if the material could be used for best advantage. I think there may be potential there, if it could be used creatively.

Bear in mind also that low grad construction ply is not really what is being proposed here void free should only be considered.

The more I think along these lines, the more it does become making my own plywood doesn't it...to be competitive it needs to justify itself and it's shortcomings compensated for unlike what is being proposed by KP and is typical of a "plywood strat" which is seeking purely a cheap alternative.

There may well be potential though...but your points are very valid and worth taking into account for sure. While the solid body has been around in pretty much the same form for a long time now (this discussion really isn't about acoustic building)...and people have tried many alternatives...it is surprising how persuasive the traditional timber products have survived the test of time in application...it is something that shouldn't be ignored

pete

[fryovanni]

I wasn't really clear in my prior posting about the testing, what I'm looking for specifically are variances in resonant frequencies (those not attributable to electronics) between the materials. My assumption here is that this eliminates the subjectivity of tone and is measuring the "quality" of vibration produced when two equal strings are struck between two endpoints of equal quality with the only variable being the substrate supporting the end points.

It is a good thought, but there are a couple things you should consider.

manufactured Plywood is very inconsistent. The orientation, grain structure, potentially voids, density of material used, moisture content and such is significantly less predictable than solid wood(at least you have visual clues with solid wood, or even ply you assemble yorself). If you are being true to your testing you have to acknowledge this, and I am not sure how you would quantify the wider range of variance.

How you deliver the energy to the string. This is a variable you have to control and repeat consistently, or your results will vary with input.

How you collect your data samples. Again consistency, and control over contamination.

Range of data and how relevent it will be to a real world instrument application. An open note may produce a bit of info, but instruments have a range of possible positions. How one material acts at a given low range frequency is one thing, but how it acts as the frequency is raised may vary differently from material "B". We understand materials potentially respond differently at different frequencies, and that may have to do with density, stiffness, structure, oils and other factors common to a given bit. Also moisture, temperature and other factors will play a role.

It would seem to me that a more productive approach would be to try to understand the corrilations between things like density, stiffness, orientation and so forth. These are common properties that can be translated to a broader range of materials. Bigger tools so to speak.

If you limit the study to plywood vs solid wood, and draw conclusions. What about a very dense bit of plywood vs a less dense example, or even a couple bits that vary in stiffness. Bits will vary, so that will be the next step in your study and you will have to start exploring basic properties.

Edit; FWIW, I recall a little testing I did a while back. There was a discussion relating to the effect of grain slope on ultimate strength and stiffness in Spruce. I thought I had a reasonable test. I simply used a piece of wood to cut a handful of samples. I split the wood taking great care to control slope. I used the most basic test rig I could think of with attension to consistency. The results clearly supported my beliefs, and I did take measures to blind myself to what material was being tested. However, the test was only as good as the size of the sampling(which is of course pretty limited, as I only used one billet of spruce). It was not a real world model(as in a functional part of an instrument), so other factors may very well alter the results or significance. Even the directionality of forces did not mimic a real world situation. The test had to be simple for control, but that is the catch 22 of much testing. I did walk away with data that showed basic properties, and that is a tool I will use for what it is worth.

Edited December 1, 2008 by fryovanni

[psw]

I guess a lot of this kind of thing is inspired by the common belief that in a solid body, a lot of the "tone" is a result of pickups, a lot to do with the kind of hardware, a lot to do with construction methods (things like bolt on vs set vs neck through) and only a very small amount a result of the timber used. As a result, alternative materials if used creatively with an even playing field (good pickups and such) may well produce surprisingly good results.

However, if there are no benefits in manufacturing or requirement for the properties that plywood can offer...then perhaps there is no reason to even consider it. You are right, good plywood is not "cheap"...anything less like construction grade could not really be considered due to voids and poor quality control regards to moisture and inner layer integrity.

I still there may be some aesthetic reasons to consider it, but perhaps the best approach would be effectively a multi laminated affair of choice thin woods and lots of them...making your own plywood with no endgrain or a super-blockboard kind of thing...this could end up even more expensive than a solid piece of wood of course.

I will be getting some scrap ply and making some shaps and carvings up in the next day or so to see how it works and what it looks like...maybe that will kill or inspire further with the idea...

[kpcrash]

fryovanni makes some excellent points as I had been kind of mapping out in my head exactly how I would have control in this test aside from the obvious constants of hardware. I'm up in Pennsylvania for the next couple of days and know there are some hardwood/plywood suppliers in this area. Perhaps I can contact one of them not so much for quality of wood - but to see if they have any information on the consistency of their product so that I can establish a baseline before pursuing any additional metrics.

[j. pierce]

I think it'd be interesting to attack the problem from as scientific a method as possible. A few things that stick out are sample size, and the fact that we're talking about playing the instruments by hand, which induces variability. Also, are we removing as many variables as we can?

Here's what comes to mind - there's certainly the possibility to go into this in more depth and approach pure scientific experiment more, but I really have little interest in trying myself - I enjoy working with wood, but don't really joy working with plywood!

If we're looking at how the body tone affects the instrument, let's choose a good bolt on neck, preferably with bushings in it, as we'll be removing it frequently. This should help take the neck out of the equation.

Build several "surrogate" bodies out of various materials. I'm imagining something akin to what I've seen Erlewine use in the fretting book I have to facilitate the use of the neck jig on a bolt on neck while keeping the body of the guitar out of harms way. We probably want something more akin to a guitar body in shape though. Bodies should be machined to be as equal in size and thickness as possibly. CNC is probably the ideal way to go here. I wouldn't worry about finish sanding or work, we just want rough slabs, as that's what we're testing. Neck pockets could be avoided by elevating the bridge, but this may take us too far from the ideas we're testing.

We want a lot bodies. As many as we can afford to make and test. And not just a lot of species or materials, but a lot of each species or material we want to test. This lets us get some repeatability in our results, so we're not colored by finding that one magic slab or block.

The pickups should be the same between instruments. Ideally, we would have pickguard harness we held over the strings. This would isolate them entirely from the body, if what we're really concerned about is changes in string response because of the affects of the wood. Some types of pickups with no potting or lesser potting may respond differently depending on mounting and whether sypathetic vibrations induce some coil movement or not, and this may or may not be an important factor, but I think focusing on just the strings for now is a good idea. String ground is not a necessary thing to worry about. Pickups should be measured to be precisely the same distance from the strings in all cases.

An optical system, (lightwave pickups are a start, but I'm really thinking about fancy test gear) may be ideal if we have the resources to spend, as it may be able to tell us more about the vibration of the string, like specifics of amplitude and frequency that may get lost in a pickup. But since we're mostly concerned about what an electric pickup can hear, that ought to be part of the test regardless, and is probably fine for our purposes. To avoid having to test with multiple pickups, we need a coil that detects a very large frequency range relatively flat, and we need to know it's response range so we can take that into account. This may involve more testing if that information is not available from the manufacturer.

So then we need lots and lots of tests. If we have humans involved, we need double blind tests - which get to be hard to do - single blind is easier, you pluck different guitars, and someone else listens to them. Double blind is harder, because we need to keep the player from seeing the guitars as well. And ideally, you should be able to switch between two identical instruments quickly. The problem is that we need to do lots of tests to determine that all those pickups are the same, and all those necks are the same and things to ensure consistency and reduction of outside variables across the tests. And then, just to be safe, we need really large sample sizes and repeated tests to be sure. And multiple listeners and all sorts of things.

So the best idea would be to remove the human element entirely. As this induces repeatability into the mechanics of the plucking or strumming or whatever, it allows us to focus purely on measurable responses. We could rig up a repeatable mechanical plucking motion, and a repeatable mechanical clamping motion. An easier solution may be to drive the strings with a sustainer fed a pre-recorded signal for a predetermined amount of time. And we measure the output from the pickups on scopes or frequency analyzers or what have you.

Of course, at this point you need to sit down and determine what you think is important in creating a strings sound, find ways to test that theory to see if it's valid, (or locate supporting research) and then use that information to construct your tests on different body materials. And you may end up having multiple factors to test. Is testing how long a string vibrates after reaching a predetermined amplitude at a fixed frequency via sustainer method and then having external forces removed important? Is the "pluck" important, negating our use of the sustainer as a cheap driver? Is crosstalk between multiple strings important? It could very well be that there are thing we can measure as easily having differences, but that have no audible effect. And the opposite could be true - that the things we have a hard time measuring are making the most audible effect, and then we have to go back and look at how we're measuring.

With this sort of test, it's harder to prove that "plywood makes a body sound different/better/worse" than it is to say "plywood helps a string sustain more/less and that helps to sound" or "plywood absorbs more of the frequencies in range of 1KHz to 10KHz" or something. Often with these types of studies, you go in with one theory, and find that's wrong. You think plywood and hardwood sound different, so you measure, and find that across all the tests you did, there was not appreciable difference. Your gut or your ear tells you there's still a difference, so you need to measure something else to prove that difference exists and why it exists and figure out what you need to be measuring to take that perceived difference into account.

And of course, when it comes down to music, (which was the whole point wasn't it?) the perceived value of an item can certainly have an impact on how we perform with it, or how we choose to hear it. There's a whole spectrum of psychosomatic issues in here too...

I don't know, that only touches on a lot of things that make a scientific study a scientific study . .. and already it's looking pretty ominous.

Does that negate the use of anecdotal evidence such as obtained in building a couple of similar instruments? Hardly. We can learn something from those kind of experiments. But unless someone goes really into depth on it, there's no way to be definitive. And even when you have the data and the definitive answer, there's always a group of folks who'll refuse to believe it.

Eh, I'm going to bed, sorry that whole post was rambling and not really well written. If you ever read a real scientific study's paper, you realize how incredibly boring and monotonous science can be, and even then, the conclusions aren't always clear.

[Mattia]

One has to ask; given the best possible outcome would be a consensus on what sounds 'best' (defined in no small part by the personal preference of the guitarist/listener in question, the amp of choice, etc.), is this really worth all the effort? I mean, really? Grab a random two dozen guitarists with various musical backgrounds (metal, country, rockabilly, rock, indie, bluegrass, britpop) and ask them to define the 'best' sound. All kind of depends on the situation, doesn't it? That warm, resonant, easy feedback vintage pickup ES 355 may not be the ideal candidate for master of puppets riffing.

Designing a proper experiment isn't primarily about methodology - first and foremost you need to decide what it is you want to measure, how you want to measure it (and then validate the way you're measuring it), and then set up a nice blinded experiment to 'prove' whatever you're setting out to prove, with sample sizes of sufficient size to power the study for the effect you expect to find. Having a robot arm pluck a string consistently and analysing the waveform isn't going to do a whole lot of good if the ultimate application is, y'know, playing music on a guitar.

[kpcrash]

So then the question becomes - what to prove? Is proving what to prove worth proving?

My original intent was to prove without a doubt that there is an audible resonant difference between plywood and solid wood. After thinking about this some more, the plywood would have to be derived from the same species as the solid wood. There are no pickups involved at this point as I'm looking to prove that there is an audible difference in electric guitars at the pre-processor level. There are certain known averages such as frequencies audible to the human ear. By measuring the frequencies, or spectrum of frequencies released by a controlled striking of a string we can measure what is audible and what is not. Not what is better.

It is not unfathomable to collect the data that represents the full range of open strings on a given guitar with a given method of construction - this allows you to have a control group by using manufactured instruments, for example, a strat with a bolt-on, a les paul with a set neck and something with a neck through.

Now, construct these instruments using plywood of a matching species to the original piece - we know what's available from the original because that's the first test.

Is there an audible (as mentioned above) difference? I think there will be, but need to start by proving that in a measurable metric that follows the basic testing algorithm. Blind testing to eliminate subjectivity is barely necessary as the striking of the strings is done mechanically with a measurable force and direction. While this may sound rather specific, we're not looking to prove that if kpcrash plays a plywood strat, does it sound different that if psw plays a solid wood strat - of course it does - there's too many variable there to be able to prove anything beyond pete's playing ability being superior to mine. And again, pickups are being used yet - that's the next test.

CNC routing would be ideal to provide as exact a replica as possible - at the same time, does body shape really matter? If I change the bottom curve of a strat out by 1/2", would it really sound different? I cannot afford a CNC router, but can gain access to a replicator for purposes of this experiment.

On the flipside, is that getting too specific in the test? Could this be done by comparing a solid alder strat to a ??? ply squier strat? This, to me, wouldn't prove much since we don't know the exact makeup of the ply.

Once it's proven whether or not there is an audible difference at the pre-processed level, then - each on can be plugged in to an amplifier and be measured again. Are the audible frequencies different, or not - that is the question. Nothing subjective - just proving that when someone says a plywood guitar "sounds" different from a solid guitar there is something backing it up other than perception.

[psw]

I am sure there will be differences...and in general we could assume that solid wood will produce more pleasing sounds...given the same structure.

How much importance it is to the functioning of the instrument as an electric instrument...whether you are looking at the guitar as simply a platform for the strings to vibrate and the pickups to be mounted...is the most important question really, possibly hard to tell. On a bass, often a really tight piano like sound is preferable and ply could do a good job of this with a less complex tone. There are things you could do like a hardwood block under the bridge, or extensive hollowing that might make for an interesting response and liven up a piece of ply. There are all kinds of strategies that could be employed in a layered structure for hollowing out or perforating the inner layers that might be interesting...there are qualities in plywood that might allow some radical carving not suitable with many solid woods. There may be an opportunity to use soft woods not suitable for traditional guitar building to plywood's increased stability and strengths.

There are a lot of down sides, the end grain and potential for shrinkage in these layers in the opposite direct where they are exposed is often seen below a finish on a play strat in time...but perhaps it is more apparent with a cheap finish that tries to hide it.

I have a cheap ply strat here that I am considering sacrificing to see what can be done with it as far as carving...it is a bad guitar however and suitable only for it's parts...and not the ideal kind of thing for what I was thinking of making...but it is a big bit of ply...so it is probably time to get out the sander and go to town on it...hahaha

Overall though, I am starting to be convinced by the arguments against...definitely as a direct replacement in a traditional wood with the same structure...I kind of doubt this kind of test thing would provide meaning results...a much easier approach would be to mount hardware on a length of ply and wood and string it up and see how it sounds over making an entire guitar. In the early days of the sustainer development I did this with a bit pine and a single string...just so I could test the things on the bench without messing with a real instrument.

I still think there could be some good things that could be done with ply of some sort (even a lot of laminations made up yourself like that martin neck)...but to make it really worthwhile, it might take some thinking outside of traditional forms.

[Keegan]

Well, you can string anything up with EMGs and play metal.

[kpcrash]

Well, you can string anything up with EMGs and play metal.

Which is the point that got me into this trouble in the first place.

[fryovanni]

Kp,

I think your search for an answer *

My original intent was to prove without a doubt that there is an audible resonant difference between plywood and solid wood.

is something that is reasonable. You are starting to look down the road a bit though, and are seeing that you are going to have to identify what you are compairing(thus your comment about the wood needing to be of the same type). Given the wood was the same species, the next question would be what are the differences in the two samples of the same species. Is the plywood sample denser than the solid sample? Then is one sample stiffer than the other sample and if so is it the difference related to the plywoods assembly of the wood itself. Hopefully you would be able to confirm moisture content was very close in both samples, but that may take some aclimation time. You would also of course have to be dead sure both pieces were equal in dimension. After you identify the differences(and they will be different) you will have to figure out if the different performance relates to the difference in the properties of the wood, or the by product of the laminate construction. This will be a very daunting task to say the least, especially with limited control over the manufacturing of the ply.

It may be that instead of jumping into model analysis it would be more orderly to start with material analysis. That way you can understand the materials, then move to modeling with control or recognition of variables. This way you could for instance insure two bodies used in a test have similar density, even if variables in stiffness are just a by product of the plywood construction, or vice versa.

After playing with spruce used for soundboards, and really focusing in on natural ranges of properties from bit to bit. I have really come to realise some significant potential for variation. I have also recognised the significance of certain attributes that account for much of this variation(such as grain orientation, growth patterns, density and such). I would certainly say it is not a total crap shoot as to what properties you will get if you account for these notable features, although you still have smaller variations that are difficult to nail down. Actually, many of those features are also used to develop grading criteria for plywood to ensure some minimum performance garentee(although within much wider range of tolerance, for most grades of ply). I think with musical instrument wood there is a history of importance placed on selection of materials for this reason, at least with acoustic instruments where pushing the limits of efficiency and performance are highly valued.

[fryovanni]

Well, you can string anything up with EMGs and play metal.

And there may be your answer

Certainly, does seem like we are wasting a lot of time building electric guitars when we should just be swapping pickups

[psw]

I think it is really very unfair to bring in "acoustic instrument lore" into this. The proposal is for a solid or at least electric instrument. An acoustic top requires flexability and the whole sound is produced by the materials ability to flex and react in complex ways and the art of bracing to control it. In a solid 2" piece of solid wood...much of this means nothing...and with so much relying on the electronics...the differences may well be extremely subtle.

Ferrington made a guitar that ry cooder bought (he liked it so much) that was a single piece of ply, extremely rough and thin.

...

If the pickups really do make such a difference, does it really matter what a guitar is made of at all. In whcih case...perhaps the testing should be focused on the pickups and not the material.

I'd like to hear if there really is much relevance (obviously there is a little) between the laws of a good acoustic top as referenced by fry and a solid body guitar...or are they essentially different beasts.

But there is a lot against working with ply in many respects and I am cooling on the idea a little in that regard.

The question of the importance of a solid bodies characteristics is very interesting and perhaps the real point here. If we disregard the material of a moment...what is the ideal characteristics of a body and neck structure for a purely electric guitar. Is it stiffness and strength, is it only longitudinal strength, is it something to do with the resonance...does chambering help, what does it do...are these things even significant in view of the importance the electrionics seem to place in the scheme of things. I am not sure that the answers will be found by referencing acoustic instruments...though maybe I am wrong...

pete

[psw]

Ok....here's an interesting site of timber info with good samples of all kinds of solids that people will find interesting as well as some of the new composite "plywoods" that are being made these days.

http://www.piecesofwood.com/woods.html

This product is called dymond and has permanently dyed layers in various forms...this is a more subtle example...

here's another...

Colorbond is a similar kind of thing...

This is similar to the stuff martin now uses which are laminated and dyed birch veneers it would seem...but with the grain all in one direction unlike 'plywood'...

But perhaps the most interesting of these alternatives is this Dakota burl made from a high pressure epoxy and sunflower seed husk material...

on further investigation this "burl" probably has little structural strength and is more akin to chipboard with a better "look"...hmmm

All interesting stuff though

...

I have a catalogue of (single) veneers about here somewhere that has unique manufactured veneers that are pressure bonded and made from dyed splinters and stuff...extremely consistent and a unique look...very expensive however...

Anyway...thought it might be of interest. Unidirectional plys would seem to be a better alternative for the making and aesthetically than common plywood and very much mimic real wood but with increased stability and consistancy and the potential to use some alternative materials and techniques...and exploit some unusual exotic 'natural' looks....

maybe we could make a guitar out of matchsticks...hehehehe

p

[kpcrash]

If the pickups really do make such a difference, does it really matter what a guitar is made of at all.

Exactly my point. There are hundreds, maybe thousands, of entries on this forum and others where the mere mention of using plywood to construct a body is complete heresy. This makes it a very emotion laden discussion sometimes with the majority stating that unless a guitar body is made from (enter favorite wood here) it will not sound good. I think in my plywood build from zyondream's thread, I proved that a plywood bass anyway can sound good - but in all honesty, there's a good set (Wilkinson I believe) of pickups in there so what you hear in the recording is the processed sound of pickups and amplifier. In some regards - that might be considered a parlor trick.

In my professional world, there are two facets to every discussion, emotion and logic. In order to have fact, you must remove emotion. Know that this flaw in my personality drove my parents crazy as "because I said so" had no credit. As good sound and tone are extremely subjective values and thus too laden with emotion to produce too many facts, you must stick with what the facts are - an electric guitar makes sound based on principles of vibration when unplugged and by principles of electromagnetics when connected to an amplifier. Is the vibration really that relevant? We have nothing to prove by saying that an electric guitar's sound is different based on pickups or amplifiers, that's extremely easy to prove. As stated, EMGs sound like EMGs pretty much no matter what they are in - positioning may have some effect - that's it.

I think with the direction this is going, it won't be too long and we can say with evidence to back it up not only if there is a difference, but what that difference is at least in general scientific terms and not just "well... because plywood sucks"