Plywood: Some Observations and a Report on the Use of Laminated Wood in Lutherie

[Originally published in American Lutherie #73, Spring 2003]

R.M. Mottola

Copyright © 2002 by R.M. Mottola

Considering that most stringed instruments are made in factories by large companies, and that the instruments at the low-priced end of these companies' product lines are sold in great quantity, and that these low-priced instruments are frequently constructed of laminated wood (i.e., plywood), it is not at all unreasonable to assume that most of the instruments made are probably constructed from plywood. Further, as plywood is known to be a physically robust material, it may be reasonable to conclude that the majority of the stringed instruments extant are made of plywood. For some reason, plywood is associated only with cheap instruments, although there are some exceptions.A cursory review of the lutherie literature reveals not too much in the way of scientific experimentation that would change that association, but it does provide a number of interesting anecdotes and observations that may indicate that plywood could be a far more useful material for high-end instruments than generally thought.

The very first issue of The Guild of American Lutherie Newsletter (vol. 1, #1) contains a letter by R.E. Brune describing the construction of classical guitar ribs made of rosewood/maple laminate, a construction which the author claims to increase volume. In his review of a harp kit in American Lutherie #69, John Calkin compared the tone of two dissimilar harps, one with a solid soundboard and the other with one of plywood, and found the plywood one to sound "bolder and a bit louder." An interesting observation can be found in the bible of flattop guitar making, Cumpiano and Natelson's Guitarmaking Tradition and Technology. They opine that lateral stiffness of a guitar top will greatly influence the tonal response of the instrument, and state that superior lateral stiffness will allow the top to be worked thinner, thus reducing mass. Now, they're not talking about plywood here per se, but a top made of, say, three ply spruce with the center ply oriented at 90 degrees to the outside plys should be stiffer across the grain than a solid top of similar thickness.  Kevin B. Reilly described small bodied guitars he made using birch ply for the tops and backs in American Lutherie #61, and found these instruments to have considerable volume and sustain.

Probably the single most influential research on wood for lutherie to date was performed by Daniel W. Haines and published in two parts in the Catgut Acoustical Society Newsletters #31 and #33 (also reprinted in the Journal of Guitar Acoustics). In the second part, entitled "On Musical Instrument Wood - Part II - Surface Finishes, Plywood, Light and Water Exposure" Haines summarizes his research with respect to plywood as follows:

"Provided equally careful and skillful instrument construction practice is followed, the results show that plywood may stand on nearly equal acoustical footing with solid wood."

A pretty tantalizing conclusion for those with an interest in using this material in the construction of musical instruments.

Here are a couple of instances of the use of plywood (actually molded laminated wood veneer) in respectable instruments. Jazz guitarmaker Roger Borys builds highly regarded instruments with laminated ribs and plates. The plates are cold molded out of sheets of veneer and wood glue, pressed between arched male and female molds. Low priced double basses have been made in this manner for decades, and there is a general feeling among bassists that some of these instruments sound very good, albeit a little quieter and often somewhat "bassier" than carved instruments. Jazz aficionados may need little more convincing of the value of plywood basses than the fact that the great Slam Stewart played a laminated bass for most of his career.

Photo 1 - The low tension of the rubber strings of the acoustic Ashbory style bass make a flexible and lightly braced top a requirement.

My own interest in plywood probably parallels that of the early laminated bass manufacturers, who found that even back then top and back sets big enough for double basses were rare and expensive. In the process of designing an acoustic Ashbory style bass, I ran into the usual-for-bass-builders problem of too small wood from the lutherie suppliers. With top dimensions of 25" x 22" the plate sets for this instrument would have to be custom cut, even if I was willing to use four piece plates. Money really does talk so this problem could certainly be dealt with at some expense, but the real clincher was going to be the bracing. The large plates needed to be lightly and sparsely braced, requirements dictated by the bass range of the instrument and the fact that the Ashbory's silicone rubber "strings" are under very low tension when tuned to pitch. My fear was that bracing that was appropriate acoustically for this instrument would not offer the physical robustness needed to survive the non-musical stresses that instruments must endure - the short falls, the bumps into the wall, or even the stabbings of an over active player's thumb. Plywood offers obvious advantages in the structural integrity department and so I used it for the plates of this instrument. The top is made from 0.125" 3 ply Sitka spruce plywood, purchased from Musicmaker's Kits (http://www.musikit.com , 800-432-5487), while the back is made of 2.5mm 5 ply Baltic birch ply from Anderson International Trading (http://www.aitwood.com , 714-666-8183). Ribs were of solid soft maple - plywood bends just fine, but it doesn't un-bend fine, a quality this hot pipe bending guy needs in a side material.

While working with these two materials in the context of this project I was hoping to at least partially answer some questions about the use of plywood in musical instruments. How does it compare in cost to solid wood? How does it compare in working qualities? And last but not least, how does it compare musically in the finished instrument? My experience shed some light here but did not definitively answer any of these questions.

Photo 2 - The spruce ply top looks like a bookmatched four piece top, albeit with obvious glue lines.

The spruce ply could have been manufactured with instrument soundboards in mind. In fact, according to Jerry Brown at Musicmaker's Kits, it was. They have this made up for them by two different suppliers, one to make up the veneer sheets and the other to press them into plywood. The material is composed of 3 void and knot free vertical grain spruce plys of equal thickness, with an obvious "show" face on one side. This side is composed of approx. 8.5" plain sliced strips of about A or AA grade vertical grain spruce (on the "standard" lutherie wood scale of A to AAAAA) butt joined together. There are obvious glue lines between the strips. On the piece I got, three of the strips are actually bookmatched (not untypical of finish grade plywood), and two of these were a full 8.5" wide, making it possible to get a 17" wide top from the piece that, for all practical purposes, would look like bookmatched solid wood. Surfaces were sanded to about 100 grit, but there were a number of spots (on the good side, no less) that exhibited the deep chunky tearout that spruce is prone to when its surface is worked without the utmost care as to the direction of the runout. No doubt the sheet went through the sander during manufacture and, since some of the top strips are bookmatched, no matter which way the sheet goes through the sander at least one strip is going to end up being sanded into the runout. Now, if you were to see this depth of tearout in solid wood you wouldn't think twice about it, but in laminated wood it can be a cause for concern.Nothing that can't be sanded out, but more on that particular issue later. The price for a 29" x 29" sheet was $70 delivered. For this kind of money, if I ordered more of it I would see if I could get the supplier to send me a piece with a wider bookmatched area on it and less tearout.

The birch ply was composed of 5 void and knot free birch plys of equal thickness, all rotary cut and thus seamless, albeit with repeating grain patterns.The veneer is hot pressed into panels using waterproof phenolic glue. The veneer used for the outer plys is from the outside of the log. Both sides were equally good but rather plain looking, flawless and even in color, but I can't really rate it as one would lutherie wood, as lutherie wood is never rotary cut. The surfaces were sanded to about 100 grit.  The price for a 61" x 61" sheet was $56.30 ($79 delivered), and it was cut in half so as to be shippable by UPS. Note just for price comparison to the spruce above that this is roughly four times the wood for the same money. This is at least partially explained by the fact that the birch is mass manufactured in huge mills out of large veneer sheets, whereas the spruce is custom made in relatively small quantities out of small, joined strips.

Note that plywood is an extremely efficient use of wood. The veneers are sliced from the log with a blade. There is no sawing and no sawdust. The good looking veneers go on the outside, the not so good looking but structurally sound ones on the inside. Veneers with defects are retained for use in lesser quality panels.

The price of these materials does seem steep but note that these are both high grade products made from high grade materials throughout. The difference between these and a sheet of fir CDX from the home center is like night and day. Note also that panels like these sanded to medium grits must be handled rather more carefully during manufacture, shipping, and construction so that they don't get scratched or dinged up.

I performed some simple stiffness testing on the spruce ply, comparing it to scraps of solid spruce in my shop. Similarly sized samples were supported at the ends and weighted in the middle. A dial gauge was used to measure deflection. Note that when grain orientation is mentioned for the plywood, I am talking about the grain of the outer plys. As expected cross grain stiffness was better in the spruce ply than in any of my solid samples. This is due to the fact that the grain of the center ply is perpendicular to the grain of the outer plys. Stiffness along the grain was about the same in the plywood as in the solid wood. This may seem counter intuitive, since the perpendicular grain of the center ply should reduce stiffness along the grain. If you consider that in a beam under load the central area is not as stressed as the upper and lower surfaces (which are under compression and tension, respectively), it does actually make sense that stiffness along the grain is not reduced too much by the grain orientation of the center ply. And the stiffness of the glue itself adds to the overall stiffness of the panel. As it is generally considered that a desirable quality of soundboard wood is a high stiffness-to-mass ratio I would have liked to weigh the samples, but unfortunately I didn't have access to a scale with high enough precision. In Haines' paper he does indicate that the spruce plywoods he examined were denser than solid wood due to the added weight of the adhesive.

For all practical purposes the plywood behaved as well or better than solid wood during the building process. Both types cut well and put up with the abuse of binding channel routing with perhaps a little less tearout than would have been the case with solid wood. Of course you can't thickness plywood - the thickness you buy is the thickness you get. And sanding has to be done with much care lest you cut through the top lamination. In general this was not as much of an issue with the birch, even considering its plys were only 0.5mm thick. I chalk this up to the fact that the surface was in fine shape as manufactured and a minimum of additional sanding was needed in preparation for finishing. I started sanding at 120 grit with the birch. Things were a little trickier with the spruce. As mentioned the good face of the spruce had a number of areas of deep (relative to the thickness of the outer lamination) surface tearout which needed to be sanded out. If I was working with solid wood this operation would have been done in a rather mindless fashion using a random orbital sander, but was performed by hand with a block, out of fear of cutting through. Starting grit for sanding the spruce was 100.

Plywood offers one obvious potential advantage to solid wood when it comes to bracing. As it is not nearly as subject to grain line cracks caused by shear stresses parallel to the grain or tension across the grain, and since a plate can be made in a single piece, it is likely that grafts can be completely eliminated for plywood plates. I didn't use any in my instrument and if I still have access to it in ten years or so it will be interesting to see if that was a reasonable design decision. For the same reason it is probably not necessary to pocket brace ends into the linings, which may offer more options for bracing configurations. The possibility also exists that bracing configurations could be substantially simplified where plywood is used, insofar as part of the purpose of at least some braces is to effectively graft areas of the plate.

In the end I get to add my experience with plywood to that of others, contributing just a little additional information. The price of the spruce ply did not compare favorably with that of solid wood and considering the quality of the bookmatching, jointing, and surfacing, and the care that must be taken with sanding, I doubt I would use it again unless I again needed a more robust material for a lightly and sparsely braced top. On the other hand it is possible that the additional cost of this plywood could be offset if bracing for an instrument using it could be simplified enough. It is difficult to assert whether or not the price of the birch ply compared favorably to that of solid wood, as rotary cut birch is not often used in lutherie. But I will definitely use this material again (if for no other reason than I have most of a whole sheet of it left over from this project). If I was to get serious about the use of plywood for plates and sides I would probably take Borys' approach and laminate them myself out of veneers. That way you get any kind of wood you want for the surfaces. As far as working properties go, as mentioned both materials behaved as well as or better than solid wood. The exception here was with sanding, where the birch was easier to sand than solid wood, simply because it needed a lot less sanding. The spruce needed less sanding than would have been required for solid wood too, but dealing with the surface tearout was tedious and thus time consuming.

This project unfortunately did not reveal any useful information on the comparative suitability of plywood from a musical perspective. The finished instrument sounds just great (says its modest builder), but as the only acoustic Ashbory style bass in existence there is nothing to compare it to tone wise. This suggests a number of areas of future research of course. One well designed experiment might involve the construction of a number of instruments from the same plans, some built completely from solid wood and some built using a single plywood component of the same wood species. A double blind listening comparison could then provide some data on the suitability of plywood for that component in that type of instrument. It is likely that such an experiment would yield results that suggest thickness and bracing modifications more suitable for plywood. This is probably the kind of research only a large shop or a plywood manufacturer would be interested in taking on at this time, but as the price of solid wood goes up and the availability goes down there may come a time when even smaller shops may need to know how plywood stacks up (a little plywood joke there) to solid wood.

Let me close with a teaser for those who may be interested in experimenting with laminated wood in their own building. This style of panel construction may offer some intriguing possibilities for the construction of materials ideally suited for musical instruments. This is all pure speculation, but here are just two thoughts. Although the spruce ply as described above offers a number of advantages over solid wood it does so at the cost of increased mass due to the additional weight of the glue. It may be possible to design a three ply panel that used, say, balsa or cedar for the center ply, retaining all desirable characteristics and keeping mass the same or even less than that of a solid wood panel. Another possibility, this one from the powerful brain of fearless leader Tim Olsen, is for a top panel composed of four spruce plys, the center two of which would be rotationally offset slightly to the left and to the right. The resulting panel might offer all of the mechanical properties of a conventional solid wood top and also be much less prone to grain line fractures.