(Text for PianoTalk #2, February 1995)
This issue of PianoTalk discusses the subject of piano soundboards. I have written it to briefly present our concepts of soundboard design and manufacture. It explains why we build Clarity soundboard panels and ribs the way we do. It also explains why we use Sitka spruce exclusively and how we go about building them--from selecting the lumber to gluing up and processing the final panel. If you have any further questions about Clarity soundboard panels and ribs, or any of our other products and services, please contact us at the phone or fax numbers shown below.
Piano soundboard technology is still surrounded by mythology and ancient "science"
When I began to investigate the different design elements that enabled pianos to sound the way they do, one of the first components I looked at was the soundboard. I discovered there was an almost mystical aura surrounding its design and its function. The basic, and still mostly secret, principles had been worked out decades earlier by The Masters who, for the most part, had all taken their secrets with them to their graves. About the only clues to how soundboard worked were found in books such as Piano Tone Building(1) and A Treatise on the Art of Pianoforte Construction(2) --both originally compiled or written before 1920. And, alas, much of the information contained in these references, while state of the art technology in the 1920, was surely somewhat dated by 1980 when I was beginning my study.
I would read passages such as the following and scratch my head in wonderment! Referring to a so-called "cup-shaped disc" found in the wall of tracheids -- a type of wood cell -- I quote, "In its center is a small membrane septum, like the vibrator on a phonograph reproducer with a somewhat thickened disc in its center... It is this vibrating membrane that gives tonal value to the woods derived from the spruce... and related species... If a log of spruce or pine is not properly cared for after felling, by being promptly barked, or if the lumber is subjected to careless kiln drying instead of being air seasoned, these delicate membranes will be ruptured, thus giving rise to inharmonics or destroying tonal value. All the crowning or pressure put on a sound-board of this material will not improve the tonal effect. Now, wait just a minute here! Tracheids are wood cells that -- depending on the species -- are about 20 to 60 microns in diameter (1 micron equals 0.001 mm, or 0.000039"). So these vibrating membranes are found in a cup shaped disc, which itself is found in the wall of a cell of wood that is about 0.04 mm, or 0.0016", in diameter, and we are supposed to be able to hear these things vibrate? I think not.
No part of the piano is as misunderstood as the soundboard.
Much of the misinformation about piano sound that we treasure still comes from the incomplete understanding these early builders had of the basic engineering and physical principles that apply to the soundboard and its supporting structure. Ideas such as the following die hard:
And the list goes on.
So I began my own search. What I learned is that the soundboard, as it exists today, is a highly evolved (rather than a designed, or engineered) mechanical system that converts the vibrating energy in the pianos strings into sound waves. Period.
When you strip away all of the mysticism the soundboard assembly is a sound producing system that just happens to be made from one of the most incredible engineering materials ever -- wood. Soundboard design and the specific type and grade of wood used have an important effect on how the piano will sound. In fact, the soundboard assembly and its mounting system are the primary factors in determining how much power the piano will have (how loud it will be), how much sustain the tone will have and the quality of that tone.
If soundboards aren't resonators and they aren't amplifiers, what are they?
Soundboards are frequently, though incorrectly, called a pianos' amplifier. In fact they do not amplify anything. In engineering terms, soundboards are transducers(a critical distinction. Amplifiers are devices that take small signals and make them larger; they add energy to the original signal. Transducers move energy from one system to another or change energy from one form to another, and this is what piano soundboards do. Energy is "transduced," or changed, from the mechanical wave energy in the string into sound energy in the air surrounding the piano. No energy is added or created in the soundboard.
When a piano string is struck by its associated hammer, energy is transferred from the hammer to the string and a vibrating wave motion is set up within, or along, the string. When this wave motion reaches the soundboard bridge a certain amount of energy is transferred from the string to the bridge causing it to move. This movement is transferred through the bridge to the soundboard causing it in turn to move. The vibrating motion of the soundboard compresses and refracts the air adjacent to the soundboard creating sound energy which our ears pick up and identify as piano sounds. This cycle is repeated until all of the wave energy in the string is dissipated.
There is another thing that soundboards are not. They are not "resonators." That is, they are not designed to resonate at any specific frequency. Many bellymen will thump on a soundboard with their fists or knuckles, listen for the resulting "boom" and pronounce a soundboard to be either good or not so good. This "test," while giving the bellyman some information, is rather misleading and is really a subject for another paper. Suffice it to say that a "resonant" soundboard is a voicing problem waiting to be discovered.
The soundboard must efficiently convert the string's vibration to sound.
The tone potential and sound quality of any given piano is determined by a number of mechanical and design characteristics over which the piano technician and/or rebuilder will have varying degrees of control, sometimes none. Among these are the following: The overall design and construction of the frame and supporting structure of the piano ( the plate, rim, rim bracing, belly rail, etc. The soundboard scale ( its design and construction, the material used-- and how it is mounted to the rim, its condition, etc. The stringing scale used --the string lengths, diameters, tensions, etc. The type (and condition) of the hammers and action used.In this paper, we're going to limit our discussion to just the soundboard.
The transfer of energy from the string to the surrounding air takes place fairly rapidly and in a more or less controlled fashion. How fast the energy transfer takes place varies both with the design of the soundboard and the frequency of vibration in the string. Low frequency wave energy will transfer at a different rate than will high frequency wave energy, so vibrations at the fundamental frequency and its various harmonics will each have different decay rates. The rate at which energy is transferred from the string to the soundboard ( and finally to the air as sound energy ( during the first few milliseconds following hammer impact determines the pianos initial, or impact, sound envelope. It is this impact sound envelope that pretty much determines our impression of a piano's volume and tone quality. How the soundboard system responds to the remaining wave energy in the string determines the sustain of the piano's tone.
Current soundboard design is influenced by the characteristics of spruce.
To do all this efficiently requires a soundboard with a unique and predictable set of mechanical characteristics. From an engineering perspective, the piano soundboard is considered to be a two-dimensional, edge-supported vibrating plate usually with clamped boundaries. It is not a freely vibrating plate, but a driven plate in which the vibrating characteristics are carefully controlled by the mechanical design of the soundboard.
All practical vibrating plates have mass, stiffness and a certain degree of internal friction. It is the relationship between these characteristics, particularly the ratio of stiffness (elasticity) to mass, that determines how a soundboard assembly will respond to the wave energy being presented to it from the strings.
When the bridge moves it disperses energy over a fairly broad area of the soundboard. Energy from one unison does not travel directly through the bridge to drive the soundboard as a point source. It is actually spread over a considerable length of the bridge before reaching the soundboard. With any luck (and a correctly designed soundboard) it will then be dispersed over a fairly broad area of the board. The soundboard should act as a diaphragm, not as a flexible membrane propagating wave motion. The soundboard panel should move as a single unit, not breaking up into small vibrational modes. That this is not actually possible in the real world should not prevent us from trying to make it so.
Why use spruce anyway? Aren't many other woods "stronger?"
To do this requires an essentially flat, very stiff, yet light-weight panel (we'll leave crown and string loading for another time). Traditionally, this panel has been made out of wood, generally one of several species of spruce. Unfortunately, wood is anisotropic(that is, it does not have uniform strength, or stiffness, properties in all directions. It is stiffer along its grain than across its grain. Consequently, the panel must have a system of stiffening ribs along one or both sides crossing the grain of the panel at approximately 90° to give it approximately the same stiffness in all directions.
Since not even "old growth" Sitka spruce trees grew large enough to cut soundboards from a single plank, the panels are built up using of a number of narrow boards selected for their uniformity of color and certain specific mechanical characteristics. For a variety of reasons it's best not to have too many glue joints in our panel so the boards used to make up a soundboard panel are typically between 75 and 125 mm (3" to 5") in width. Much less than 75 mm and the panel begins to look "choppy." Much more than 125 mm and the board becomes more susceptible to changes in moisture content. Not to mention that spruce lumber meeting our specifications is nearly impossible to find in widths much greater than 125 mm. In the interest of better moisture stability, these boards are always quarter-sawn.
Spruce is spruce is spruce, right? Well, not really...
Over the years several different species of wood have been used for soundboards, but nearly all piano builders end up choosing one of the various species of spruce. There are a number of very good reasons for this:
Some other types of wood possess some of these characteristics; indeed, some actually out-perform spruce in one way or another. Spruce alone, however, has it all.
The case for Sitka spruce. It really does have what it takes!
Normally, one of three species of spruce is used for soundboards. They are eastern white, Englemann and Sitka spruce. The following chart compares three of the most critical mechanical characteristics of these woods for use as piano soundboard material. (Sugar pine is not normally used for soundboards, but is included because several manufacturers have used it for ribs.)
|Species||Specific Gravity||Modulus of Elasticity (million psi)||Compression Perpendicular to Grain (psi)|
|Eastern White Spruce||0.36||1.43 - 1.45||430|
|Englemann Spruce||0.38||1.30 - 1.55||410|
|Sitka Spruce||0.38||1.60 - 1.63||580|
|Sugar Pine||0.36||1.18 - 1.20||500|
Specific gravity is the average weight of the wood compared to water. In general, for both soundboards and ribs, the lighter the better. Specific gravity varies with grain density. Boards with a high "grains/inch" count are more dense than boards with a low "grains/inch" count. They are also stiffer --it's a trade-off, you can't have both. Also, the actual weight of any soundboard and rib assembly will vary with its moisture content. All species of spruce have approximately the same specific gravity.
The Modulus of Elasticity (MOE) is the ratio of stress to strain within the elastic limit of the wood sample. Stress is unit force, or the amount of force or load acting on a unit of area. Strain is unit deformation, or the actual amount of bending resulting from a given load acting on the wood sample. For soundboard wood, the higher the MOE the better. Wood with a high MOE will resist bending under load better than wood with a low MOE. In other words it will be stiffer and better able to support the strings' downbearing force acting against the soundboards crown. Sitka spruce has a much higher MOE than other spruces. In fact, Sitka spruce has one of the highest stiffness-to-weight ratios of all readily available wood.
Compression Perpendicular to Grain is a measure of the ability of a given wood to resist compression perpendicular to grain up to its proportional limit., that is, before fiber failure. Wood is hygroscopic. As it absorbs and desorbs moisture, it will expand and shrink if it can. Once a soundboard is installed in a piano, though, its ability to expand is severely limited so the swelling wood cells create internal compression instead. Woods with higher compression perpendicular to grain ratings will resist fiber damage resulting from internal compression better than woods with lower ratings. Of all the spruces, Sitka spruce has the best compression perpendicular to grain rating, it has a greater ability to resist failure due to fiber crushing.
Studying the chart above indicates that there is one type of spruce that stands out from the others in two of the three important parameters listed. Sitka spruce has the highest modulus of elasticity and it has a higher compression perpendicular to grain rating.
There are several other characteristics of Sitka spruce that single it out as one of the worlds best woods for piano soundboards. Its evenness and uniformity of grain and its warm white to light golden tan color are unsurpassed for beauty. Its low internal friction properties provide near perfect damping qualities. It can be cut into defect-free boards that are wide enough for soundboard panel construction, requiring little, if any, patching.
While Sitka spruce has not disappeared from our forests, it is rapidly becoming an endangered species. Lumber of musical instrument grade is now very hard to find. For a long time I thought this was because Sitka spruce trees were not being planted in any quantity. It turns out they are ( in large quantities. It seems that they grow like weeds unless they are planted in areas where they are heavily shaded, the water supply is limited, etc. With good growing conditions they grow very fast which, of course, means fairly wide grain lines and very low grain/inch counts. These trees make good studs for house building, good pulp for paper production and good chips for fiber and particle boards and other man-made products, but they're not much good for piano soundboards.
We need trees grown in old-growth forests under difficult conditions. They need to grow in areas heavily shaded by a surrounding canopy of existing trees. The best trees for our purposes are found in areas with less moisture available --a condition hard to find in a rain forest (well, they tell me its the ground water that counts). But there are still areas where these trees are thriving. Let's hope someone sees the light and begins planting trees soon for a sustainable yield of musical instrument grade lumber. It's not being done today, but we're working on it.
Besides making fantastic piano soundboards, Sitka spruce is used for some furniture making, mill products, interior house trim & millwork and window blinds. Some spruce is still used in light airplane construction, especially in home-built and kit-built craft. Many sailboat masts and spars are still made of Sitka spruce.
Clarity soundboards and ribs are made exclusively of Sitka spruce. Our plant is located at the foot of the Olympic Peninsula in Washington state. Right at the foot of the Olympic temperate rain forest. This area has been renowned for more than a century as the source of some of the finest Sitka spruce soundboard lumber that the world has ever seen.
We select and purchase only the very best musical instrument grade spruce lumber for Clarity soundboards. This is now some of the most expensive wood grown in North America. Most of the lumber we now purchase was grown on the western mountain slopes of Canada and Alaska. We look for boards with medium to tight grain density, clear straight grain, good uniform color, lack of defects, "bear-claw," etc.
Once we inspect and take delivery of the lumber, it goes into our temperature and humidity controlled conditioning room. It is sorted for grade and color and stickered for conditioning. Even though the lumber has been both air and kiln dried before we take delivery, we condition it for a specified period of time before we begin to process it. (The amount of time required varies depending on moisture content and on the cross-section dimensions of the lumber.)
From this lumber, we then further sort and select only the clearest and most uniform lumber to make into Clarity grand piano soundboard panels. We then cut and trim the lumber to make sure it is as close to defect free as possible. As you can imagine, with three separate selection and sorting steps the reject rate is astronomical! But the final results are worth the effort.
Once the boards are sorted and trimmed, we assemble them in our gluing fixture in the shape and size of the soundboard panel we are making. We clamp and press the board flat using a water-resistant, cold press glue. After the glue has cured and the wood has moisture-stabilized we transfer the panel to another temperature and humidity controlled conditioning room where the panel is allowed to age for a minimum of 14 days while the glue joints stabilize before it is sawn to shape and sanded to its final thickness.
Only custom soundboards are available.
It is our goal to fill all orders in a prompt and efficient manner. Each soundboard panel is built individually, however. We try to maintain a large enough stock of dried and sorted lumber to meet the anticipated demand for our soundboard panels. We can usually begin production of your boards within a couple of days of receiving your order. Including drying and conditioning time, it takes us about three weeks to complete a soundboard panel. So turn-around time will be approximately three weeks from the receipt of your order until we are ready to ship.
Standard Clarity soundboards are of uniform thickness. Since many bellymen prefer to diaphragm the soundboard as part of their bellying operation, we have standardized the thickness for Clarity soundboards at 8.0 mm. If you need a panel that is thicker or thinner than this, let us know. We will try to accommodate your needs. Shipping panels thinner than 8.0 mm is somewhat problematic because they are very fragile and easily damaged. They can also be made thicker -- up to 10.0 mm-- at no extra charge. Our standard thickness tolerance is +0.5/-0.0 mm.
We can diaphragm your soundboard for you if you wish. We recommend what we call partial, or bass diaphragming. This type of diaphragming thins out the perimeter of the soundboard along the lower, or bass, half of the panel. Depending on the size of the soundboard we start the taper in approximately 300 mm to 400 mm in from the edge of the soundboard.
Although we can do it, we don't recommend "full" diaphragming for any piano. This type of diaphragming has been used by Steinway since the 1930's. It is our opinion that this type of diaphragming makes the tenor-treble area of the soundboard too thin giving it too much flexibility (it lowers the mechanical impedance) and makes it difficult to obtain good sustain particularly through what we call the "killer octave." That's the area around the fifth to sixth octave where it is often impossible to get good sustain no matter what you do to the hammers. That's because it isn't a hammer problem, it's a soundboard problem. Anyway, we can do this type of diaphragming if you wish. (If you call to talk to us about it though, we'll try to talk you out of it!)
Before we can do any diaphragming, we must know the actual shape of the soundboard you need. You will need to send us an accurate paper pattern so we can cut the panel to shape. (Unless we already have a pattern ( call us and ask before you go to the trouble of making a pattern we may already have.) We will cut the panel to the shape of the pattern plus approximately 10 mm. This will give you enough extra material to get a good fit to the rim and still provide good diaphragming. There is a charge of $50 to $175 for this service.
We build our standard panels with a grain angle of 50° or 60° as referenced to the bellyrail. Some older piano designs use a grain angle closer to 45°. Years of testing and measuring have convinced to me that a grain angle 50° to 60° (depending on the length of the piano) will improve the treble response of any piano with no sacrifice in bass tone. (Actually, my two most recent new grand piano designs have used a grain angle of 60° as does our current "killer B" package for the Steinway Model B grand. For best results, the rib angle should also be changed.) We will, of course, build panels to any grain angle required for the piano you are working on.
The small print:
We are always trying to upgrade our products and services so the details and specifications described above are subject to change without notice. If you have any questions or comments, please call.