Replies

1. 
   StanleyNiemiec | Jun 28, 2003 11:33am | #1

   Ofda!  Do you ever ask simple questions?  And this one is truly and actually a knotty question.   (BOO, HISS)

   If you are referring to bending strength (as a 2x4, applying a load in the center of the board supported on the ends such that you are testing the strength of the 4" dimension by applying a downward force) the position of the knot is extremely critical.  If the knot is in the center of the board (in the area termed neutral shear), it will have considerably less effect than if it is at or near the edge.  Knots on or near the edge disrupt the ideal straight grain and therein because of the change in grain angle are much more prone to either compression (top surface) or tension (bottom surface) failures.  The key factor in the strength reduction is the immediate grain angle.

   If you are bending a board on the two inch plane and the knot direction is perpendicular to the baords face (as if the knot were running throught the board), you might as well subtract the entire knot dimension from the width of the board and consider the board strenght to be that of a therein narrower board. 

   The type of knot also makes a big difference.  There are live knots and dead knots.  The former is associated with a living branch while the latter is generally that of a dead branch.  A live knot is generally tight because both the branch and the bole wood were living and in a sense were growing together.  There is not that growth interrelationship between stem wood and a dead branch and as such dead knots are often loose.  Live knots are also termed tight knots.  Wood member strength is considerable greater for live knots than it is for dead knots. 

   Knot size is also a big factor; the larger the knot the weaker the board.  For simplicity sake, the strength effect of knot size is geometric rather than arithmatic.  Therein you might consider, as a simple rule of thumb that a two inch knot will have one-fourth the strength of a one inch knot.

   Knot direction is also an important factor.  Grading rules are more preferential to knots running through flat sawn boards and exceedingly negative regarding knots running through VG or radially cut boards.  Knots of this type are referred to a spike knots and are very weak at the knot.  This is obvious because the knot at that particular point is such a large percentage of the overall wood at that point.

   In compression (perpendicular to the grain) knots do not have much effect. 

   I could find no information regarding shear strength (splitting) but from the amount of fire wood that I have tried to split that contained knots, I would have to say that straight grained wood splits more readily than knotty wood.

   Did I miss anything?  I hope this helps and is both readable and understandable!  Maybe Jon should take a shot at it!

1. 
   forestgirl | Jun 28, 2003 08:05pm | #2

   Thanks for responding, Mr. N.  It'll be fun to take a copy of your response out and identify which kinds of knots I placed in the scrap area of the cuts.  Actually, I only included a couple of very small knots in a couple of seat slats.  But since the back and seat braces are curved, and there were many knots, I ended up including a bit of the surrounding wood (not the knots themselves) in a couple of places.  This is the wood I referred to -- the grain lines are compressed in these areas.

   This will be a fun way for me to commit to memory information about knots, something that simple reading doesn't do these days.  Question:  when the live knots age, do they shrink disproportionately to the surround wood and become loose?  Seems like that might happen with knots more than 1/2" or so.

   Ofda -- are you Swedish?forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl   ;-)

1. 
   StanleyNiemiec | Jun 28, 2003 09:26pm | #4

   Mr. N  --  please, it's either Stanley or Lee.

   Am I Swedish?  No not really.  Niemiec (shortened from Niemiecz) has its origins across the Baltic in a southeasterly direction from Sweden.  But what with all those marauding Vikings from the days of yore, there is no doubt that I have some of that Norsk blood running through my veins.  Actually, I picked up ofda from my midwestern days especially those times spent in Central Wisconsin.  In some instances, it is seemingly the only appropriate response one can make.

   Live knots will generally have a solid perimeter and very likely intercellular/molecular bonding within the overall ligno-cellulosic mattrix and as such do not tend to fall out or become loosened.  They will however tend to check like you would expect to see with boxed heart pith.  Dead knots (aka loose knots) tend to have a non-bonded perimeter (unless there is pitch inclusion) because the branch was dead and the two areas could not "grow" together.

   IN the pines, dead knots tend to get very pitchy likely because they are no longer "part" of the living tree and to sterilize this now foreign and potentially infectious material, resin is accumulated there.

   I also wanted to add a couple of things to my earlier comments.  The first relates to slope of grain.  In the Wood Handbook is the following quote --  "To provide a margin of safety, the reduction in strength due to cross grain in visually graded structural lumber should be about twice the reduction observed in tests of small, clear specimens . . . "   Thus if you cut it too close, while you have eliminated the knot itself, there is still a level of inherent weakness associated with curvy grain.  That the grain is "compressed" indicates in my thinking two possible strength reduction concepts  --  that this wood may contain more residual growth and drying stresses and that there may be grain deviation in more than one direction.  I cannot recall specifically with regard to WRC but when DF grows around included branches, it tends to develop a conical or rounded growth pattern certainly visible in the bark and also in the wood itself.  As there are not additional growth rings in this area, the implication to me is that multi-planed grain deviation.

   Secondly, total grain area variation (per unit area around the knot) increases significantly with increasing knot size.  Grain angle deviation, off the top of my head, is may 50 to 75% of the knot radius on each side before straight grained growth returns.  Thus relative to strength reduction, you have a reduction associated with the angled grain and also for the non-structural aspect of the knot itself.

   Your a smart person and you have good common sense so it is likely you are not going to include a piece that is completely inappropriate.  Just watch out for the really obese ones.  Maybe you should put a little sticker on the piece, as often appears on lamps, not to exceed a certain bulb wattage! 

2. 
   jonsherryl | Jun 28, 2003 09:18pm | #3

   Nice summary on the structural significance of knots, Stanley. I don't think you've missed anything important.

   About the only thing I'd add with respect to this particular seat application relates to grain direction. Since the grain in a knot is perpendicular to the grain in the slat...and since wood doesn't shrink hardly at all longitudinally...the knot will constantly be rising above or receding below the surface...especially in exterior applications. This isn't necessarily a structural issue, but it sure raises hell when it comes to snagging clothing. Also, if the species is resinous, knots sometimes bleed through the finish...which isn't the best news for your wardrobe either.