Thursday, 20 April 2017

Wing Rigging - Wire Tension Progression

Before doing the W&B, I rigged the wings using the recommended procedure in the Christen Eagle AFM.

From past experience, a biplane wing rigging is never a straightforward process. Since I was crazy enough to buy a digital torque wrench, it gives me the possibility to quite quickly and accurately read the wire tension (along with the tool from Holloway Engineering). I therefore made an experiment: record all wire tensions for every iteration of the tensioning process on one side. This way I was going to hopefully capture the converging tendencies and learn as much as possible to hopefully ease future riggings.

The following graph summarizes the results I got during the tensioning of the RHS wing wires. Each bar height is proportional to the wire tension. When a label exists above a bar, it means this wire has been adjusted compared to the last iteration. For example, on iteration 2, the fwd landing wire has been tightened by 1/2 turn. Negative values mean wire loosening. The brackets on the bottom indicate the results of the readings from the level I placed over the top wing front spar.


I could note/confirm the following trends for my aircraft:
  1. When tensioning one wire in a pair, the other wire of the same pair often looses tension (see landing wires during iterations 1,2 and 3)
  2. Adding tension can be done relatively gradually whereas removing tension can easily be drastic (see iteration 7)
  3. Adding tension on the flying wires does not significantly add tension on the landing wires (see iteration 4)
  4. Adding tension on the landing wires can have more of an effect on the flying wires (see iterations 5 and 6)
  5. Building tension on the flying wires has to be done with care since it can bend the wing and cause undesireable anhedral. In this example, I caused a slight anhedral up to iteration #6, then I removed tension on the flying wires on iteration #7 to remove the anhedral. In other words, it seems to be preferable to tighten the flying wires only when there is already quite some tension on the landing wires.
After iteration #13, I switched to other side to tension the LHS wing, then when back to RHS and did small adjustments to get within tolerances. After all that, I measured the wing incidences with the rigging boards and quite happily confirmed they were within tolerance too! This seems easy to summarize, but the whole process took me around 2 days of work.

I do not pretend the above is applicable to any other biplane, but perhaps a potion of the above findings can help others : )

Next step: the first engine startup. Stay tuned!

2 comments:

  1. Hi, which digital torque wrench did you buy ? Thanks!

    ReplyDelete
  2. Sorry I missed your comment! It is a snap-on 3/8" flex-head Techwrench.

    ReplyDelete