Hi Jude Thanks for your post. The thing is... there are several arguments out here being used to justify the use of RC&S boards over compression reliant boards. Central to some of the most important of these is the issue of panel compression. I dont think I need to get into all the reasonings here as they've been well discussed already. The self-destruct thing, predictablity thing... we've been through that all. So if degree of compression is to be blamed I think there for its more then fair... and hardly chasing foul of any sort to establish what is actually known. Coming up with a decent guestimation of how much compression builds up because of the panel taking on humidity is fairly easy... tho as Hoadly warns the actual variance of compression strength from panel to panel is pretty large. Still we have usable ways and means there and they are these that are used as the basis for the argumentation against compression reliant panels. Knowing how the glued and crowned board behaves with respect to what compression degrees occurs when the assembly is subjected to downbearing is at the least a bit more complicated matter. You are looking at at least a biaxial if not triaxial loading of both the ribs and the board. Its not like you are just pushing down on a set of ribs. The ribs are of interest in this question of panel compression only because the degree of tension and compression they find themselves in before the panel is loaded is needed in order to know what degree of panel compression will build up for any given load. Likewise the degree of panel compression before load needs to be known. I'll just sketch out two rough examples to illustrate. In each it is assumed that all relevant data about rib dimensions, panel thickness, MC at glue up for both etc are known ok ? In a CC board ready for string loading one has ribs which are bent by way of the compressed panel tensioning the top half of the rib. Isolated, a downwards load force alone would of course simply relieve both the bent state of the rib and panel... essentially removing the compression/tension moments that have to do with the actual bent form of both of these. But we know that thats not at all what ends up happening. The glue joint between the panel and ribs causes a second load to be applied as a result of the downward pressure. This load goes along the surfaces of the adjoining panel and rib surfaces causing a net increase in panel compression and at the very least less relief of rib tension then if rib was not subjected to this increase in panel compression. Just how much resultant net compression there is in the panel and tension there is in the rib is from what I can see anyones guess. In an RC&S board you have ribs that are machine crowned, and a panel that is bent over their surface. No problem figuring amount of compression / tension in each as the only thing going on here is the panel in bent form. As it takes on humidity one still can with ease reasonably calculate the amount of compression in the panel. The rib will experience the same kind of tensioning as in a CC board but to a much less degree. But application of downbearing changes things entirely here. If one has as a goal to press the assembly downwards to say 50 % of its unloaded height then the ribs are essentially bowed DOWNWARDS . From their static unloaded state, the top half of the rib is then under compression and the bottom half is under tension... the reverse of a CC board. So what happens to compression in the panel then ? It of course has to increase as result of the load placed along the surfaces resulting from the downwards pressure... which will exert a tensioning stress on the rib. But the top half of the rib is in a compressed state now... because of being bent downwards. It will readily comply with any tensioning force applied by the panel and would continue to do so until it eventually reached its unloaded state. So how much net compression ends up being in this board ? Unless one can answer these questions and at least show the numbers that relate to each of the type of boards then one is essentially guessing and as a result stand on very weak grounds indeed when making claims about which kind of panel is more stable, more predictable, more durable etc etc all attributed to a degree of compression in the panel one in reality doesn't know. If on the other hand one CAN answer these questions... then at the very least the whole predictability argument dissapears... at least to the degree with which the strength of the panel across the grain itself is predictable. And again... the arguments against compression reliant boards is at least on that point weakened. I've been told by a few here I've been given plenty of answers and information and I should be greatful. I would like to point out that this is really the one and only question I have raised in the recent discussion, and that anything close to an answer is still lacking. Frank at least implied he has software that may be up to the task. JD and yourself seem at least also interested in the question. Heck guys... its just a question that needs answering... and as I say... none of this has no bearing on the validity of the various approaches to building soundboards. All have proven themselves plenty viable. Personally, thats where my the boarder of how thick a hair to split goes. Cheers RicB Ric, Symantecs aside, it seems like a fair and interesting question to ask to me. I've built my rib data spreadsheet to calculate the sag by entering the pressure of the string bearing force, rib dimensions, modulus of elasticity of rib material, bending and resisting moments & moment of inertia. Many of these factors come from static tables in old industrial standard texts for the given material (ie sitka spruce, sugar pine etc.) and are based upon some sampling once upon a time. I know it's been brought up that the modulus of elasticity, for example, will vary from rib to rib even with the same species and dimensions (this by the way is definitely one of the advantages of the laminated ribs in that the elasticity coefficient is averaged out). Here are some potential drawbacks: 1.. Even with all this data, some of the static values and constants are averages or aproximations. 2.. Everyone I know, including myself, uses the formula for center loaded beams, which isn't exactly the case in the piano, where there may be two to three different loads on a given rib and even when there is only one load it is not necessarily in the center. This does make a good case for the symetrical design though, if for no other reason than to make the math easier. 3.. A judgement still has to be made as to how stiff you want your assembly to be. Nevertheless, it's still a pretty good tool, a point of departure shall we say. I don't know how much compression this translates into but I would sure like to know. And I would like to experience how this affects the sound. All in due time I hope, unless this is my Moby Dick. Ron N. makes some very good points about staying out of the "swamp of details" and staying aware of the diameter of the hair we are trying to split (definitely goes in the Ron N. Greatest Hits of List Quotes), but there's no harm in asking questions. Cheers,
This PTG archive page provided courtesy of Moy Piano Service, LLC