<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN"> <HTML> <HEAD> <META content=text/html;charset=iso-8859-1 = http-equiv=Content-Type><!doctype html public "-//w3c//dtd html 4.0 = transitional//en"> <META content='"MSHTML 4.72.3110.7"' name=GENERATOR> </HEAD> <BODY aLink=#ff0000 bgColor=#ffffff link=#0000ee text=#000000 = vLink=#551a8b> <DIV><FONT color=#000000 size=2>Richard,</FONT></DIV> <DIV> </DIV> <DIV><FONT size=2>You have served as a wonderful goad to this = discussion, but ultimately there comes a time when you must commit to an opinion of your = own.  What will it be?</FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT size=2>In a nutshell, my argument is that a traditionally = crowned soundboard can be constructed such that the top of the board is in tension.  I am not disputing that with drying and manipulation of = humidity that the top can be made to be under less tension, or even = compression.  What I am arguing is that the final stress distribution DEPENDS on = starting and ending humidity of all parts.  Further, if the top of a board = starts in tension and the bottom in compression, as it is loaded by downbearing, = the tension/compression load will be decreased.  Finally, it is my = opinion, that because of the way it responds to load, a traditionally crowned = soundboard will just sound better (this point is pure opinion, and I won't argue further).  Others have made these same arguments in the past and = have tired of butting their heads against the wall of pedagogy and = sycophancy.  I now begin to understand their fatigue.</FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT size=2>I will not address a situation in which parts are = assembled under differing and controlled humidity.  There are too many = variables, and too many correct answers in such a situation.  Stress in the board = will vary, and it will depend.</FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT size=2>Here is an experiment I conducted in my shop = yesterday.  I selected a room dried strip of sitka spruce, made pencil marks on it = 18 inches from the end.  I bent the spruce to an approximate 3' radius and = glued a rib to the bottom.  When I measured the top of the spruce, I found = that the pencil mark position indicated the top had stretched approximately 1/16 = inch, and the bottom had compressed about as much.  This model indicates = to me that the top is in tension and the bottom in compression.  = </FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT size=2>I then placed the model between constraints at each = end and loaded the top.  As the load increased, the line on the spruce = showed me that tension in the top was decreasing.  I loaded the model to failure.  The rib failed first, and the failure was the most = beautiful textbook indication of a failure in tension on the top and compression = on the bottom (if you don't know what this kind of failure looks like, buy a = box of chalk and break a few pieces in different ways - the most interesting is = when you break it by twisting).  Because of one of Newton's laws, we = might deduce that the soundboard was experiencing equal and opposite loads = when the rib failed.</FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT size=2>If you care to construct such a model, I'm sure your = results would be similar.  If you would like to see pictures, I will = consider putting them up on a photo site.  If the actual physical = demonstration of the main point I am arguing does not convince, then I fear we must = promote this discussion to the level of religion where issues are resolved on the = basis of faith and personality.</FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT size=2>Frank Weston</FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV> </DIV> <DIV><FONT size=2></FONT> </DIV> <DIV> </DIV> <DIV> </DIV> <DIV> </DIV> <DIV><FONT size=2></FONT> </DIV> <DIV> </DIV> <DIV><FONT size=2></FONT> </DIV> <DIV> </DIV> <DIV><FONT size=2></FONT> </DIV> <DIV> </DIV> <BLOCKQUOTE style="BORDER-LEFT: #000000 solid 2px; MARGIN-LEFT: 5px; PADDING-LEFT: = 5px"> <DIV><FONT face=Arial size=2><B>-----Original = Message-----</B><BR><B>From: </B>Richard Brekne <<A href="mailto:richardb@c2i.net">richardb@c2i.net</A>><BR><B>To: = </B><A href="mailto:pianotech@ptg.org">pianotech@ptg.org</A> <<A = href="mailto:pianotech@ptg.org">pianotech@ptg.org</A>><BR><B>Date: </B>Monday, September 27, 1999 1:59 PM<BR><B>Subject: </B>Re: = evaluating sdbd. crown & bridge downbearings in a new piano<BR><BR></DIV></FONT>  <P>Frank Weston wrote: <P>Frank... I would ask you again to consider the example below and = simply answer the question. Its a simple enough example. I ask this because = it will clear up a specific principle my own mind wants cleared up. The = paragraph you write below doesnt do that. That much (your paragraph below) is = obvious enough in any case. Please be so kind as to re-read the example = below, taking it for what it is and answer my query. <P>Thanks so much <P>Richard Brekne <BR>I.C.P.T.G.  N.P.T.F. <BR>Bergen, Norway <BLOCKQUOTE TYPE = CITE> <FONT color=#000000><FONT size=-1>Richard,</FONT></FONT> <FONT size=-1>Forget = humidity.  Forget made up numbers.  Do this.  Tape a flexible = steel rule onto the end of one side of a spruce stick  (a wooden yard = stick will do for demo purposes).  Mark the opposite end of the = steel rule on the spruce stick.  Bend the spruce stick.  = Notice that as you bend it, the mark on the spruce changes position relative = to the steel rule.  If the rule is on the outside of the bend, the = spruce gets longer.  If it is on the inside of the bend, the = spruce gets shorter.</FONT> <FONT size=-1>If the spruce is elongated = it is in tension, if shortened, in compression.  Now, bend the = spruce stick and glue another piece on the bottom.  Nothing has = changed.  When the glue dries, the assembly will hold a curve.  The = outside curve of the spruce is still longer than it was originally, and = it is still in tension.</FONT> <FONT size=-1>I hope this is = simple enough for you.</FONT> <FONT size=-1>Frank = Weston</FONT>   <BLOCKQUOTE style="BORDER-LEFT: #000000 solid 2px; MARGIN-LEFT: 5px; = PADDING-LEFT: 5px">Lets say we have a panel exactly 1 meter across grain, and a rib = that is 998 cm long in a room with 50% relative humidity. The panel = is put into the oven and dried to the point that it is also 998 mm = long. Now if we take the panel out and simply let it re-adjust to = the room humidity it will "grow" back to 1 meter in length. = The top and the bottom of the panel will both shrink the same, and = grow back the same amounts... ok so far ??  But if we quickly = attach the ribs while the panel is at 998 and then allow it to re-grow = to what ever length the room humidity and this constraint from the = rib allows for, its length will be less the 1 meter. (This = following what I think I got from Dels description.) Ok.. assuming = this is a correct picture so far, stretching a measuring line across = the <I>top</I> of this  ribbed panel after its re-adjusted = to room humidity shows that it measures less then 1 meter. <P>If, (Frank) you accept that this is correct so far,  = I would appreciate it very much to know the reasoning behind why = this top half of the panel is not (if it is not) to be considered in compression. As initially stated, please follow Dels kind "laymans language" approach to demonstrating the reasoning. <P>Richard Brekne <BR>I.C.P.T.G. N.P.T.F. <BR>Bergen, Norway</P></BLOCKQUOTE></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>