<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=Content-Type content="text/html; = charset=iso-8859-1"> <META content="MSHTML 5.50.4134.600" name=GENERATOR> <STYLE></STYLE> </HEAD> <BODY bgColor=#ffffff> <DIV><FONT face="Comic Sans MS" color=#000080></FONT> </DIV> <BLOCKQUOTE dir=ltr style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; = BORDER-LEFT: #000080 2px solid; MARGIN-RIGHT: 0px"> <DIV style="FONT: 10pt arial">----- Original Message ----- </DIV> <DIV style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: = black"><B>From:</B> <A title=Richard.Brekne@grieg.uib.no href="mailto:Richard.Brekne@grieg.uib.no">Richard Brekne</A> </DIV> <DIV style="FONT: 10pt arial"><B>To:</B> <A = title=pianotech@ptg.org href="mailto:pianotech@ptg.org">Pianotech</A> </DIV> <DIV style="FONT: 10pt arial"><B>Sent:</B> November 24, 2002 2:36 = PM</DIV> <DIV style="FONT: 10pt arial"><B>Subject:</B> Re: S&S D = Duplex</DIV> <DIV><BR></DIV>Hi Del... <P>I think I understand what you mean here. Basically you are saying = that if you use flat ribs, whatever crown you end up with is more or less a = measure of how much internal compression the panel is under. Otherwise there = would have to be a difference between flat ribbed / flat cauled and flat ribbed / = curved cauled.... or what ? <P>So please indulge me for a bit so'st I can perhaps get this right.. = all this goes to the flat rib / dished caul variant of Compression = crowning. Follow along and tell me where / if I hop off the wagon. <BR>  <BLOCKQUOTE> <DIV>--If you start off with just a panel in a dished caul and push = it down... then you have a panel thats tensioned on its outside and = compressed on its inside.</DIV></BLOCKQUOTE> <DIV><FONT face="Comic Sans MS" color=#000080>In theory, yes. = there will be some slight amount of tension on the outside of the curve. Keep in = mind, though, that in a practical soundboard the radius of this curve is = quite large (relative to the thickness of the soundboard panel) and the amount of = tension will be very slight.</FONT></DIV> <DIV><FONT face="Comic Sans MS" color=#000080></FONT> </DIV> <BLOCKQUOTE> <DIV>--Push in a few ribs and the sides of the ribs that meet the = panel are tensioned.</DIV></BLOCKQUOTE> <DIV><FONT face="Comic Sans MS" color=#000080>Yes. The top (or = front) of the rib will be under some tension and the bottom (or back) of the rib = will be under some compression. How much tension and compression will depend = on the height of the rib and the radius of the curve.</FONT></DIV> <DIV><FONT face="Comic Sans MS" color=#000080></FONT> </DIV> <BLOCKQUOTE> <DIV>--If panel and ribs are glued thus and allowed to dry, all = stays unchanged as long as the thing is in the caul. But when you take it = out then the whole thing wants to try and return to its flat state but cant. = The compressed underside of the panel and the tensed sides of the ribs = work against each other being glued tight as it were.</DIV></BLOCKQUOTE> <DIV><FONT face="Comic Sans MS" color=#000080>So far, so good. A = stress interface develops between the soundboard panel and the ribs at the glueline.</FONT></DIV> <DIV><FONT face="Comic Sans MS" color=#000080></FONT> </DIV> <BLOCKQUOTE> <DIV>--What does happen tho is that it returns about half way and = you have some crown already before taking on any humidity, and the inside of = the panel is less compressed then it was in the caul, the outside less = tense, and the ribs less less tense.</DIV> <DIV><FONT face="Comic Sans MS" = color=#000080></FONT> </DIV></BLOCKQUOTE> <DIV><FONT face="Comic Sans MS" color=#000080>And now we're on the = edge of the wagon getting ready to hop. Actually, we're probably already on the = bunny trail. I would question that the outside of the soundboard panel would = be just 'less tense.' Even at this point it is almost certainly under some = amount of compression. </FONT></DIV> <DIV><FONT face="Comic Sans MS" color=#000080></FONT> </DIV> <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <DIV>Let the panel take on humidity.. </DIV></BLOCKQUOTE> <BLOCKQUOTE> <DIV>--and the compression on the underside of the panel increases, = the tension on the ribs increases... </DIV> <DIV>--and the top side of the panel... hmmm.. its still under some = degree of tension if I am correct up to this point. So at some point enough = humdity will put the top side under compression.</DIV></BLOCKQUOTE> <P><FONT face="Comic Sans MS" color=#000080>Ah, now we're = definitely hopping on down the bunny trail. And we're not alone out here. For some reason = this seems to be a point of confusion with many. </FONT></P> <P><FONT face="Comic Sans MS" color=#000080>As the panel becomes = compressed to the point it is able to develop a glueline stress interface sufficient = to form or even maintain crown the whole panel is going to be under = compression. And, as it takes on moisture sufficient to increase that crown the amount = of compression is going to increase. And this increase is evident through = the entire panel, not just the surface nearest the ribs.</FONT></P> <P><FONT face="Comic Sans MS" color=#000080>Keep in mind that = amount of wood cell compression required to generate the amount of force -- the = stress interface -- necessary to bend a set of flat ribs into a crown radius = of 18 m (approx. 60') is considerable. There are a lot of variables but it's = going to be something over 1% to 2%. </FONT></P> <P><FONT face="Comic Sans MS" color=#000080>Now it doesn't matter = to the wood if this compression comes as a result of being placed in a vice and = squeezed or if it comes as a result of developed internal compression through = the mechanism of being dried (shrunk), restrained (ribbed and brought = back to some higher MC (expanded). Hence if a soundboard panel starting out = at, say, 1,000 mm across grain, is put in a vice and squeezed -- compressed -- = by 1% it is going to end up being only 990 mm wide. If by 2% it will be 980 mm = wide. </FONT></P> <P><FONT face="Comic Sans MS" color=#000080>Or if a panel is = stabilized at 4% MC and cut to a width of 1,000 mm and then taken back up to 12% it is = going to expand to somewhere around 1,015 mm to 1,020 mm (yes, I've done the experiment). That's an expansion of about 1.5% to 2.0%. Now, if you = dry this panel back down to 4% (shrink it), put it into a fixture that will not = allow it to physically expand, and then take it back up to 12% MC, it = will still be 1,000 mm wide (the fixture sees to that) but it will now have = a considerable amount of internal compression. Something on the order of = 1.5% to 2.0%. At least it will for a while. Since wood cells were only = designed to tolerate compression levels up to about 1.0% they will quickly begin = to fail.</FONT></P> <P><FONT face="Comic Sans MS" color=#000080>So, what happens if = this panel is dried to a MC of 4% and bent in to a set of realistic curved cauls = (let's stick with our 18 m radius). Yes, the outside surface will stretch by = some small amount and place this surface in tension. Probably by some = fraction of 1% -- I'm guessing here, but (assuming a soundboard of 8 mm thickness) = I'd be surprised if the stretch (and tension) was much more than 0.01% . And = the bottom surface will compress by about the same small amount. In other = words, the top surface might want to stretch by about 0.1 mm and the bottom = might want to compress by about 0.1 mm. </FONT></P> <P><FONT face="Comic Sans MS" color=#000080>Now with a set of = flat ribs glued to the back of this panel restricting its expansion as = the panel takes on moisture all of this would-be expansion turns into = compression. And here is where we hopped out of the wagon: The amount of = compression developed will be substantially more than is needed to neutralize that = slight amount of tension and turn it into compression. What you will end up = with is a panel with, for example (and using my admittedly hypothetical = numbers), 1.49% compression on its top (outside) surface and 1.51% compression on its = inside surface. But the whole panel will definitely be under compression. And = when the compression ridges develop they will develop all the way through = the panel, top to bottom and bottom to top.</FONT></P> <P><FONT face="Comic Sans MS" color=#000080>Can we get back in the = wagon now?</FONT></P> <P><FONT face="Comic Sans MS" = color=#000080>Del</FONT></P></BLOCKQUOTE> <DIV><FONT face="Comic Sans MS" = color=#000080></FONT> </DIV></BODY></HTML>