Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum FAQ - Tuner-Technicians Forum Lack of crown?

Del,

I suspect what you call "compression set" is what wood technologists call "stress related deformation" (SRD). I have suspected this (SRD) to have something to do with tone deterioration in old pianos from stressful climates, and of course it hurts pianos in many mechanical ways as well.

This is the first I have read the same SRD might occur in new sound boards. It's an interesting subject and it would be great if someone really gets a handle on it and can get the rest of the industry acting on the knowledge.

What is the "design feature" you are writing about that is overstressing sounbdboards?

Regards,

Rick Clark

Wood technologists call compression set, well, compression set. It is the result of stress related deformation. This latter phase can also be applied to tension deformation. This is the term used primarily by R. Bruce Hoadley in his book, Understanding Wood.

The USDA Forest Service’s Wood Handbook deals more with the concept of ‘creep.’ When a stress is applied to wood it deforms elastically. In the case of the compression-crowned soundboard panel it creates its own load as it expands by taking on moisture after being ribbed. If this load is maintained (as it is in the compression-crowned soundboard panel) additional time-dependent deformation occurs. This is called creep. It occurs even at very low stresses, particularly when those stresses are perpendicular-to-grain compressive stresses.

If a constant deformation is imposed and maintained on the wood member (in this case, the soundboard panel) the initial stress relaxes at a decreasing rate to about 60% to 70% of its original value within a few months. This is called ‘stress-relaxation,’ or compression set. This stress relaxation continues until some equilibrium is reached and there is no longer enough compression stress to matter. Obviously, if the soundboard system was relying on this compression to form and maintain crown there is going to be a problem.

In general, wood can sustain perpendicular-to-grain compression levels up to about 1% without instantaneous wood fiber failure. Above this and the wood fibers begin to deform quickly and often catastrophically. The typical compressive stress levels within a compression-crown soundboard panel can start as high as 1.5% to 2.0%. This level of compression drops rather rapidly as compression stress takes its toll. Along with it goes some of the crown that was initially built into the system. The rate at which the compression level drops is dependent on a number of factors, including:
— The strength characteristics of the specific wood samples.
— The specific amount of initial compression.
— The range of temperature and, more importantly, humidity the wood sample is exposed to.
— The peak level of compression realized during the highest humidity peaks.

Obviously, there are a number of variables and a system has evolved which works reasonably well most of the time. I suspect, however, it has worked some better in the past as we chewed our way through those wonderful old Sitka spruce giants that were able to grow to enormous heights and impressive ages within the protected confines of the old-growth temperate rain forests. Most of which are now history. Such wood is becoming increasingly rare and we are now working on what is left. And what is left would have been rejected out of hand even as recently as half a century back. Times change.

Del

If a soundboard is 1 meter wide, 1% compression would be squashing from side to side until it is 1 cm. narrower. That's quite a bit. If that's the difference in width from the top of the soundboard to the bottom, that would make quite a curve in it. How do you figure that much stress?

But not exceptional. I once took an unribbed Baldwin 743/5 (the designation for the Hamilton back assembly) soundboard panel and drew a line across it diagonally (perpendicular to grain), dried it down to 4.0% MC (as weighed and compared to an oven-dry sample), placed two marks on it exactly 60” apart and left it to sit in the Trumann, Arkansas plant warehouse for 48 hours (during August). It took about that long for the marks to be just shy of 61.2” apart. That’s an expansion of about 2%. Just about what the books predict.

Putting the same panel back in the conditioning room soon brought it back to 4% MC at which point it was taken to the rib press and immediately ribbed (with flat ribs). Once the glue was dry it was taken back to the warehouse and allowed to reach equilibrium. I don’t recall the exact distance between the two marks but it was something like 60.2” and the assembly was highly crowned. This soundboard assembly was allowed to sit, undisturbed, in that same warehouse for a year. At the end of this year the crown was down to about three-quarters of the original amount (with similar temperature and humidity readings in the warehouse). Even without being glued to a back assembly and held down by any string bearing the soundboard panels internal compression was dissipating.

While this was not the most scientific of tests it was still a graphic illustration of the process and what happens to the wood as a consequence.

Del

However, the stress in a soundboard that wants to be 61.2" wide when it is being held at 60" doesn't depend on how the soundboard is crowned. That is a result of gluing ribs across its grain, no matter how the soundboard is formed. It is how compression ridges get formed.

I'm not sure I understand your point here. The only reason for taking a soundboard panel down to 4.0% MC at ribbing is so that it will subsequently form the crown that results from the resultant stress interface between the (now-compressed) soundboard panel and the ribs. A high level of internal compression is necessary both to form and maintain this crown and that is (usually) how it is developed. That is how the design works and that is why they are called 'compression-crowned' soundboards.

If the soundboard system crown is formed by machining the requisite curve into the rib it is not necessary to reduce the panel’s MC to 4.0%. The system is designed such that there will be little or no internal compression during normal climatic conditions. In general, internal compression is considered to be counterproductive and the system is designed to reduce it as much as possible. We control the MC of our both our soundboard panels and our ribs to between 6.5% and 7% MC as the ribs are glued on. This figure is fairly typical for most rib-crowned soundboard designs.

Del

Del,

I appreciate your writings on this. You know from past threads I have a skeptical thought process so I may probe your assertions or bring up a counterpoint sometimes, but please don't mistake that for criticism. Mostly I'm just trying to get a grasp on these things myself.

Regards,

Rick Clark

I found this two year old thread in the archives and I thought I would "light it up". There is a lot of good information in the archives.