SB 00036 (cracked horiz stab spar) Installation Notes

[plehrke]

Originally Posted by Dean Pichon

If I can accept the legacy design is almost good enough, is there a solution space that brings me to good enough using only the spar web. A one-piece, steel hinge assembly/doubler may distribute the load (even more so if bonded) over a greater portion of the web. If the ears are integral to the steel doubler, more of the bending loads will go into the entire steel doubler rather than the two small footprints of the legacy hinge brackets.

The issue is that the elevator hinge creates a point vertical load and its associated fore/aft moment needs to get reacted into the skin with a fore/aft couple into the skin. The spar web has flanges that attach to the skin and transfer the load into the skin forming a closed cell torque box. (With bigger loads this would be helped out by a rib backing up the fitting and spar web with multiple fasteners to get that couple into the skin.) It is not a “bending” load as you describe. It is a vertical load and moment couple between upper and lower fasteners that attach the fitting. “Thicker spar web” may help with the load at the rivets attaching the fitting but then that load is transferred in the spar web through the rivets and goes around the 90 deg bend in the spar web to get to the skin attach fasteners. Going around that bend is the material thickness of the spar web and therefore a prime location for crack. A doubler may solve the current crack location in the vertical web of spar but will probably cause new cracks at edge of the doubler in the corner of the 90 deg flange that attaches the skin. If you think this through it is pretty intuitive. In my opinion that corner is the fatigue detail that Vans second iteration in the repair SB solved by having the new doubler nest and pick up skin fasteners.

Putting a large steel plate on the spar web may have another negative consequence related to horizontal tail bending. Bending due to lift on the tail is a reacted into the skin in a inbd/outbd couple (not fore/aft couple like i explain above for the point hinge fitting load) The spar is aluminum of a particular web thickness and a particular stiffness. All of a sudden the bending inbd/outbd in the spar web hit a very very high increase in stiffness of a steal doubler. Rapid increases in stiffness is prim location for cracking. This is why spar always taper in thickness changes. Think of tree branches always break at intersection to thicker branches.

No good engineering is ever presented without its caveats.
-A freebody diagram would be helpful for both of these I explain above but I am on vacation with only an iPhone to write this response.
-I am a structural design engineer so some strength engineer my give a better explanation, since they always want to prove us designers wrong.

[agent4573]

My first attempt at this didn't go well. 5 out of the 8 cherrymax rivets came out bad. Luckily the stems weren't retained and they were easily knocked out so I could drill out the cherrymax and try again. Both my hand puller and the pneumatic puller had too wide of a tip and were side loading the rivets. I bought an extended nose hand puller, modified it to fit without hitting the brackets or skin, and with lubed up rivets was able to successfully pull the 5 bad ones that were on the right side, and all 8 on the left side without issue.

I ended up with a single oops rivet on the right lower skin due to a mis-squeeze. Now I just have to do some touch up paint work.

[Dean Pichon]

Originally Posted by plehrke

The issue is that the elevator hinge creates a point vertical load and its associated fore/aft moment needs to get reacted into the skin with a fore/aft couple into the skin. The spar web has flanges that attach to the skin and transfer the load into the skin forming a closed cell torque box. (With bigger loads this would be helped out by a rib backing up the fitting and spar web with multiple fasteners to get that couple into the skin.) It is not a “bending” load as you describe. It is a vertical load and moment couple between upper and lower fasteners that attach the fitting. “Thicker spar web” may help with the load at the rivets attaching the fitting but then that load is transferred in the spar web through the rivets and goes around the 90 deg bend in the spar web to get to the skin attach fasteners. Going around that bend is the material thickness of the spar web and therefore a prime location for crack. A doubler may solve the current crack location in the vertical web of spar but will probably cause new cracks at edge of the doubler in the corner of the 90 deg flange that attaches the skin. If you think this through it is pretty intuitive. In my opinion that corner is the fatigue detail that Vans second iteration in the repair SB solved by having the new doubler nest and pick up skin fasteners.

Putting a large steel plate on the spar web may have another negative consequence related to horizontal tail bending. Bending due to lift on the tail is a reacted into the skin in a inbd/outbd couple (not fore/aft couple like i explain above for the point hinge fitting load) The spar is aluminum of a particular web thickness and a particular stiffness. All of a sudden the bending inbd/outbd in the spar web hit a very very high increase in stiffness of a steal doubler. Rapid increases in stiffness is prim location for cracking. This is why spar always taper in thickness changes. Think of tree branches always break at intersection to thicker branches.

No good engineering is ever presented without its caveats.
-A freebody diagram would be helpful for both of these I explain above but I am on vacation with only an iPhone to write this response.
-I am a structural design engineer so some strength engineer my give a better explanation, since they always want to prove us designers wrong.

Hi Philip,

Thanks for the suggestion of a FBD. Since the pivot cannot transmit a a moment, the applied elevator load is limited to the reaction force applied thru the bolt. Of course the magnitude and direction will vary depending upon elevator deflection/loading. This force can put both a shear and tensile load on the rivets holding the hinge to the spar. In addition, when the force is anything but horizonal, a couple is introduced into the spar.

The issue of the abrupt change is spar stiffness is a good point and not one I have not fully considered. I assume Van's uses the "fish mouth" design of the aft doubler to minimize this. I carried the same idea to the one-piece bracket I am considering, but may need to do more given the higher modulus of the 4130. I'm looking at tapering the thickness of doubler portion. I'll compare the relative stiffness of various section (locations) of the spar (E x I) and compare them to Van's new design.

Thanks for the help,

Dean

[Sam Buchanan]

Putting a mod hat on for a moment....

This thread was started for the purpose of sharing notes about installing the SB kit provided by Vans. Engineering discussions that don't apply to the Vans kit need to be in a different thread so helpful hints about the Vans installation don't get lost in the clutter.

Thank you.

[Dean Pichon]

Originally Posted by Sam Buchanan

Putting a mod hat on for a moment....

This thread was started for the purpose of sharing notes about installing the SB kit provided by Vans. Engineering discussions that don't apply to the Vans kit need to be in a different thread so helpful hints about the Vans installation don't get lost in the clutter.

Thank you.

Understood! Thanks to all for the input.

[Jpm757]

Are there any alternative access points for borescope inspection for those with glassed in end ribs? What about small access hole on bottom skin?

[Ironflight]

Originally Posted by Jpm757

Are there any alternative access points for borescope inspection for those with glassed in end ribs? What about small access hole on bottom skin?

On our RV-3, we don’t have balance weights, so with glassed-on tips, there is no way to go in from the tip end…so we took off the empennage fairing and went in from the root end. Worked great - however, there are no intermediate ribs on the RV-3 (just root and tip) and the distance is only about two feet to the outer hinge…. But it might get you thinking!

[RV8JD]

Originally Posted by Jpm757

Are there any alternative access points for borescope inspection for those with glassed in end ribs? What about small access hole on bottom skin?

Originally Posted by Ironflight

On our RV-3, we don’t have balance weights, so with glassed-on tips, there is no way to go in from the tip end…so we took off the empennage fairing and went in from the root end. Worked great - however, there are no intermediate ribs on the RV-3 (just root and tip) and the distance is only about two feet to the outer hinge…. But it might get you thinking!

See this post:

https://vansairforce.net/community/s...d.php?t=215119