Stephen Birkett wrote: > > Del wrote: > > Creep is creep..... > > maintained for a period of time there will be some additional > > time-dependent deformation. That is creep. And it doesn't matter if it > > occurs is a wood beam, a soundboard panel or a glue joint. > > > I'll restrict this reply to just creep...will get to the rest when I can, > plus I've asked Bill Jurgensen to give us his input. > > Creep may be creep but it is a definite misnomer (albeit not Del's) when > it comes to wood...that is the source of the confusion here. I agree > completely with Del's description of what occurs when wood is stressed: > initial elastic deformation, subsequent time-dependent deformation due to > sustained stress. (Visco-elastic deformation is no doubt a mouthful hence > the use of "creep"). To this I will add that increased moisture content and > temp. accentuate the effect. In addition the deformation is aysmptotic > with time. Thus the final deformation has a maximum, for all intents and > purposes reached after probably a time period of days or weeks. The > maximum visco-elastic deformation is generally of the same magnitude as the > initial elastic deformation, for quite high stresses..thus for practical > purposes we can imagine a worst case scenario of initial deformation > perhaps doubling over the first year of life of a piano, then staying > exactly the same for the remaining 149 years. (If the stress is removed, > about half of the visco-elastic deformation is usually recovered). > > Now what about glue joints...say a face joint, like that in the frame of > the 1814 Streicher I'm making, subject to shear stress from string > tension (i.e. parallel to the joint). An aliphatic glue joint, even > Titebond II, will exhibit creep. In this case the creep really *is* creep > in the sense that anyone would mean...the molecular links that determine > the glue bond will *continually* re-arrange themselves, however slowly, to > form new links. Eventually the two surfaces will move with respect to > each other. This process will continue *as long as the stress is > maintained*, there being no equilibrium condition reached. In 150 years I > can imagine such a joint might eventually self-destruct...no matter how > small creep is for aliphatics it is definitely not zero. > Long time * small effect = big problem. > > What about hide glue? Hide glue exhibits ZERO creep. It is a crystalline > structure when cured that cannot re-arrange itself without shattering. > Even after 150 years the two face-jointed surfaces will be in precisely the > same positions with respect to each other. > > This is why I said we will not witness the final test of the aliphatics in > our lifetime. Long term stability for highly stressed hide glue joints *has* > been demonstrated and proven. Long term stability for highly stressed > aliphatic joints is questionable for the reasons above. If I spend 1000h > on a piano I want to be sure it won't fall apart. > > Big wood framed pianos always bend a bit when strung, then a bit more > over the next weeks as the wood settles into equilibrium...then they stop. > Self destruction, cheek disease, etc. either result from inadequate > or badly designed bracing (such as English pianos) or replacement of > strings with inappropriate modern gauges, increasing tensions beyond the > original design specs. "Creep" failure of not an issue for either the wood > or the glue joints. > > Nuf said. > > Stephen > > Stephen Birkett Fortepianos > Authentic Reproductions of 18th and 19th Century Pianos > 464 Winchester Drive > Waterloo, Ontario > Canada N2T 1K5 > tel: 519-885-2228 > email: birketts@wright.aps.uoguelph.ca Stephen, Very nice analysis of "creep." I do, of course, take exception to your assertion that "...the deformation is asymptotic with time. Thus the final deformation has a maximum, for all intents and purposes reached after probably a time period of days or weeks.” Certainly, the rate of deformation does decline over time, but in a load carrying beam some level of deformation continues as long as the applied load exists. It will never reach a point of complete stability. If the applied load is great enough, or the time long enough, any wood load carrying member will ultimately rupture. Of course, if the beam is made strong enough in relation to the load, the creep-rupture process can take quite a while. But the process in a stressed wood member really is quite similar to that in an adhesive joint—the adhesive joint you describe being stressed more highly relative to its ultimate strength characteristics. And it is surely stretching reality a bit to claim that “...for practical purposes we can imagine a worst case scenario of initial deformation perhaps doubling over the first year of life of a piano, then staying exactly the same for the remaining 149 years.” I have a problem with phrases like “final deformation has a maximum” and “staying exactly the same for the remaining 149 years.” Final deformation does not have a maximum, unless you consider creep-rupture to be a maximum, and no wood member under constant stress will remain exactly the same for even one year, let alone 149 years. (OK, I promised to drop the subject of glues, but you’re implying in the above that I have been advocating using aliphatic adhesives in ways that will ultimately result in the failure of a piano’s structure.) You write, > This is why I said we will not witness the final test of the aliphatics in > our lifetime. Long term stability for highly stressed hide glue joints *has* > been demonstrated and proven. Long term stability for highly stressed > aliphatic joints is questionable for the reasons above. If I spend 1000h > on a piano I want to be sure it won't fall apart.” My point exactly. Over the thirty-plus years that I’ve been in this business I have witnessed—and repaired—far too many failed hide glue joints to trust my structural work to its inconsistencies. When I glue something together, I want to be sure that it stays glued. Which is why I prefer to use an adhesive that is appropriate for the wood being used and the type of joint in question. Have I implied anywhere that aliphatic resin adhesives—even the latest generation of them—are the only adhesives that I use? Or that these are the only adhesives that should be used in piano work? This whole thread started out by my questioning the assertion that animal hide glue has some magical acoustical qualities when used to glue hammer heads to hammershanks—an assertion that I have tested and found to be unsupported by fact. (It’s a claim that is still unsupported by fact, by the way.) As I have been trying to say all along, I believe in using adhesives that are appropriate to the task at hand. Aliphatic resin adhesives are not appropriate for all types of wood joints and/or applications. Neither is animal hide glue. —ddf
This PTG archive page provided courtesy of Moy Piano Service, LLC