Structural Analysis
Since
one spar cap is in tension and the other is in compression,
what you have is something like an earthquake fault line
between the spar
caps. They would like to move relative to each other in a sliding
fashion and
that is what shear webbing was invented to resist. The highest
load on shear
webbing is at the midpoint between the two spar caps. My patient Uncle
and I
examined the shear web loads on my 1/6 scale Cub spar design
and found that
1/64
ply was more than adequate for a spar 
of those dimensions and
loadings (total
depth 1”, width 3/8”, spar cap thickness
1/8”). If your spar dimensions or
loadings go up by a factor of more than
two or three you might want to go to 1/32 ply for the
shear
webbing. Shear
loads are constant from
root to tip. If your spar at the wing
center section is solid top to bottom, along the center 10% of span
(approx.)
it will help resist shear loads.
For the program to be valid the
spar must not be allowed
to buckle as you look at it in cross section from the wingtip. The wing ribs
do this job so they should not be too far apart. If the spar can buckle
it will
fail at a far lower loading. I usually don't go past 3 inches
on rib spacing unless the wing chord is more than 15 inches.
I use this program to gain confidence in the strength of my wing designs. It is a starting point, and should be followed up by actual failure tests of your wood and design, especially if you are cutting close to the margin on weight and load factor. Before testing the sample run the program a few times, using a load factor of one and various weights until you arrive at a 100 percent rating in the design requirement statement. Your sample spar can be about three feet long and should be supported in the middle. Load the sample evenly from tip to tip until it fails. The weight the failure occurs at should be close to the 100 percent rating weight predicted by the program. A variance will most likely be due to 9000 not being the correct stress for your wood sample. You might want to lower this value if your samples consistently fail at too Iow a value, or you can look for some better samples of wood. Another thing that will throw off the result is if your sample is not exactly the size you are putting into the program. Your measurements should be within 1/64 of an inch, especially on small sizes.
To test for errors in the program after you type it, do a test run using the values from the sample run which are for the 1/6 scale Cub spar. Remember, testing is central to the design process. Calculations are only the starting point, and have to be compared to the real world by careful testing.
Page 4 of article ____ Golden Era Model Service Index ____ John Eaton Home Page