What do you suggest for light but slop free control linkages?

From : Don Stackhouse

The Wizard, like our other glass fuselage HLG's, has conduits installed already for music wire pushrods to the ruddervators. Some of our Chrysalis builders have tried pull-pull cables with mixed results. Personally I don't care for them. They are very light, slop free when properly tensioned, but they are usually less stiff than pushrods, which invites flutter. They also put some interesting compressive loads on hinges, which in some situations can lead to chronic problems.

Before the pull-pull cable fanatics light up their torches, let me explain:

Let's consider three pushrod assemblies:

1. 1/8" birch dowel (with short music wire ends, which we will ignore for this analysis)
2. .025" music wire
3. .020" Kevlar pull-pull cable

The birch dowel has an elastic modulus (also called Young's modulus, it's a measure of the inherent stiffness of the basic material) of about 2.1 million psi, a cross sectional area of .0123 sq. in., for a stiffness of 25,800 pounds per inch/inch. It's weight is about .0045 ounces per inch of length. Stiffness/weight is therefore about 5.7 million (the units on this number are goofy, but adequate for this analysis). BTW, the main reasons we don't use this type on the Monarchs or the Wizard are because it's more parts, more work to build, and somewhat bulky. Trying to fit two of those into the tailboom of a Monarch might get rather interesting!

The music wire pushrod (AISI 1060 steel) has a Young's modulus of 30 million psi and a cross sectional area of .00049 sq. inches, for a stiffness of 14,700 pounds per inch/inch. It's weight is .0022 ounces per inch of length. The conduit adds from .00037 to .0015 ounces per inch depending on the type used, so the total weight of the system is about .0026 to .0037 ounces per inch. Stiffness/weight is 6.7 million, and it is about 57% of the total stiffness of the Birch pushrod.

The Kevlar pull-pull cable has a Young's modulus of 8.8 million psi and a cross-sectional area of .00031 sq. inches, for a stiffness of 2,800 pounds per inch/inch per cable. The weight per cable is .00026 ounces per inch of length per cable, but there are 2 cables, so the total is .00051 ounces per inch. Since only one of the cables contributes to stiffness in either direction ( HLG structures are not generally strong and stable enough to support sufficient cable preload to allow for both cables to contribute to stiffness simultaneously, and thermal expansion, linkage non-linearities, etc. make such preloading even less practical), the stiffness/weight is 5.5 million (ok in comparison to the steel), but the total stiffness is only 11% of the stiffness of the birch pushrod! That's a flutter problem looking for a time to happen (it's already found the place!).

If you wanted to make a Kevlar pull-pull system with the same stiffness as the steel pushrods, you would need to use .045" diameter cable, which would weigh .0027 ounces per inch, about the same as the steel pushrods (and this doesn't include the weight of the extra control horn required for the second cable, or the ballast in the nose to offset it, or the extra hinging and structure you may need to support the cable tension).

Spectra has about twice the stiffness of Kevlar and about 2/3 the weight, but it is subject to "creep", or "cold flow" (which is why it never caught on in full scale aerospace structures), so you would have a lot of trouble keeping it tensioned properly. Graphite cable is also better than Kevlar in terms of stiffness/weight, but it's very tricky to attach to the horns and servo arms, and can be very sensitive to chafing. At best the pull-pull systems usually offer little or no benefit over conventional pushrods, with a significant increase in complexity. A good, light pushrod system with well-fitted ends usually works better for me.

Don Stackhouse
DJ Aerotech