Rivets

This is a picture of a mock up section of a helicopter tail boom I did.
The design criteria were that it take 1 day of engineering and 1 day
of assembly to make it. It's about 5' wide. I didn't meet the design
goals.

http://www.metalinnovations.com/images/engineering.jpg

Another design goal was to demonstrate fixtureless matched hole
assembly. That worked out ok. I can't remember how I did it, but I
know it was unpleasant.

It boils down to putting thousands of holes in curved flanges and lofted
bends. When you do that, they aren't round holes anymore, which means
you can't find their centers which means it's really, really hard to
locate (thousands of) holes in mating parts.

What I'd like to do is not use holes, but put an array of points on the
surface and have them unfold with the sheet metal.

Since I can't do that, can anyone offer any suggestions? Anyone else
doing something like this?

--
J Kimmel

www.metalinnovations.com

"Cuius testiculos habes, habeas cardia et cerebellum." - When you have
their full attention in your grip, their hearts and minds will follow.

 

We do a lot of this.

What you do is create the sheet metal part, unfold it, put your desired hole
pattern into the part in the flattend state (try and use the hole
wizerd)then fold the part up again. If you can't select the hole edge in the
folded part to mate a rivet then you can select the axis of the hole to use
as a mate. Using this method gives you a flat pattern with round holes that
you can then use to send to laser cutters etc to produce the parts and get
your "fixtureless" asembly.

A word of caution when unfolding, double check your flat pattern dimensions
especially with cones etc to be sure that it is correct. You may have to
play with the K factor to get it exactly right. SW must be calculating a
stretch allowance when unfolding but with the sheet

Steve R

 

To complete what I was writting....

...... and just rolling it there really is no tradational stretching so you
may find that your flat is not quite correct. The flat result also depends
on what modeling method you are using to create the formed part.

 

Just for curiosity; I tried doing what you wrote, but when I fold up the
sheet the holes go into suspended mode.

 

This works for holes on flat flanges, but aircraft skins, ribs, formers etc have rivet holes
on curved surfaces. If you make a simple "L" bent part with a large bend radius, unfold it,
insert some holes *in the bend* and fold it up again, you'll see that the holes within the
bend zone lose their temporary axes because they're no longer cylindrical.

The best I could do to locate "bent" holes was to insert a ref point using the "Center of
Face" option while selecting the punched surface of the hole (the surface that would be
cylindrical in the flat state). That locates the hole center approximately halfway through
the sheet thickness. (The depth of this point varies with hole diameter and bend radius.

Then another point can be placed using the "Projection" option while selecting the first
point and one of the part faces. This puts the second point on the surface also at the hole
center. A ref axis between the points defines the hole centerline.

Unfortunately, these are "bogus" ref points and axes (for the OPs purposes) because they're
left hanging in space when the part is unfolded. I suppose they're tied to the surface IDs in
the folded state, and these surfaces don't exist when unfolded.

Might be a nice enhancement request to have hole axes transfer between flat and folded
states.


Art W.