Bending long curve in longeron

[Dave Stroud]

Lads...I'm modifying my fuselage a bit and need to bend about a 3 -3 1/2" deep curve over a length of about 116" in my 5/8" .035 longeron tubing. I've bent plenty of short bends before but never something this gradual over a distance. My thoughts would be to fasten the tube down to the bench and force the bend to the desired shape with at least a 1/2 dozen wood blocks and then apply gentle heat from the torch hoping  the tubing will cool down and relax in the desired shape. Any thoughts ? Thanks...

[scottiniowa]

Lads...I'm modifying my fuselage a bit and need to bend about a 3 -3 1/2" deep curve over a length of about 116" in my 5/8" .035 longeron tubing. I've bent plenty of short bends before but never something this gradual over a distance. My thoughts would be to fasten the tube down to the bench and force the bend to the desired shape with at least a 1/2 dozen wood blocks and then apply gentle heat from the torch hoping  the tubing will cool down and relax in the desired shape. Any thoughts ? Thanks...

Dave, that would be a question that for me to answer, I would start out by saying "it depends"

Meaning, is the arc, from the very tip, to the very end with the 3.5" put in exactly in the middle?  If so, you can almost get it, just my pinning each end, and pulling the center.  Using the truss to then hold this in place.  Or you mentioned method can work to some degree as well, but still would rely greatly on the truss to hold the shape.
or
if you holding some on each end strait, then forming the arc, you block method could work well, but will have to block both sides as you heat certain areas, it will want to flex away from hottest side first. then retract when it cools  (often used on tail post, if it goes out of alignment when welding the top longeron clusters.  (heat on the back side) 

I am guessing you have a drawing in mind, for your change?   Sounds to me like a fixture could be in order, if for no other reason, in that you will be doing two of these, and to have both identical, will be difficult at best, but certainly good enough to be held in a fixture for the next step in frame building.

[Tom XL-7]

http://www.harborfreight.com/tubing-roller-99736.html
 I have not used this
Incremental pressure and multiple push-pulls should do it.
Don't let the tube rotate or you will get some interesting art.
You may be able to make a jig for less if you can turn a few rollers - even wood on a lathe. It won't take much pressure. 
Start with a stick long enough to clamp or grip.
A local fab shop is a possibility also. 
 it won't be as strong as the original tube. It will collapse before a straight tube in compression. 
 Airframes are drawn point to point, cluster to cluster eliminating eccentric loads. The bows and curves are not primary load bearing pieces.
most eagles build in eccentric loads when bending our lower longerons. We build station two and three and then heat and bend the tubes.
We should try to keep the bend as close to the cluster as possible as the change in direction should be in the cluster. I don't see a problem unless someone comes out a few inches before the bend and even then you have the "will get away with it factor" which is true but hard to evaluate.
 There is a trade-off - continuous tube vs a truer cluster. I'll take the continuous tube

Tom XL-7

[Tom XL-7]

Thinking about this I wonder if none of the above is the answer.
If I support a long enough length of 5/8 tube by the ends I will get a sag that equals what you want.
Just jig it into shape then weld it. Say it is your top longeron- with standoffs from the centerline. Midpoints between your diagonals so they are not in your way.
 The only problem is the tube when heated may want to relax a little, straighter than you intend. Minor distortion at welds onto a  straight tube is normal anyway.
 I think Scott was thinking the same way.
" Using the truss to then hold this in place."
Tom XL-7

[Dave Stroud]

Thanks for both your good replies, Tom. Your points are well taken. I think I'll just give the curves a go on the bench with jigging. The bottom longeron ( between the tail post and the area aft of the seat ) curved part is now showing to be about 97" long and the upper longeron goes from the tail post might be about 90" long. I'm just trying for a small curve which might be as little as two inches when I get it drawn out on the table today full size. 

The curves will be slight, certainly not as extreme as a Kitfox for example. Here's a picture of a Velie Model 70 which I'm leaning towards aesthetically.

[Dan_]

I agree with not trying to shape the curves before they are tacked on the ends.  Square or "boxed" fuselages are normally built with both sides being laid out flat, with any curves formed by wood blocks on an jig just like using a rig jig with the truss braces tacked in before taking the side out of the jig.  Also there is going to be a big taper to the tailpost.

To avoid welding heat taking the curve out of the fuselage sides, wood cross braces are placed between the clusters to hold the shape and resist the fabric. 

For this particular 200 mph airplane the prescription is offset the tube 3/8".  That means cut your board 3/4" longer than the distance between the 2 diagonally opposite tubes.  This would be to resist the normal GA weight fabric and welding heat from "squaring off" the tubing between the clusters.

See attached.

[Dave Stroud]

Thanks for help, Lads. It seems in the end only about 1" of curve gives the desired impression. I got one side bent up today, both longerons front to back. Will do the other side tomorrow and start the jigging process from there.