TRUSS design, boolean issue

michael_stalls

Hi all,
here are some images of a truss for the fuselage of an aerobatic aircraft i have been working on.



Making this truss has been somewhat tedious but i got there in the end. i made it by creating a box that represents the four outer longerons and the sketching side profiles on the faces of the box. i then swept the tube profiles along these side profiles. i made them mostly as seperate sweeps and then joined the using boolean union. When i tried to add yellow and blue tubes to the rest of the truss it comes back with an error "cant produce a valid part'' why would this be?

brucebartlett

I'd love more robust tools for producing part's like this. I can see those junctions are going to become quite complex and that would be my guess as to why its not working, a small bit on an edge somewhere. I'd be trying to trim bits back to simplify the whole model and find the problem area's 1 by 1. 

What the end goal? This can make a difference as to how far to persist with problems like this. 

Manufacturing drawings?

Models For FEA?

CAD data for CAM? 

bobt

The most common cause of Boolean errors of this sort is trying to join tubes of identical diameter. A slight tolerance error in position, and now there is no longer a valid intersection. If possible I would make the cross struts slightly smaller than the lateral struts.

I can't tell from your images, but hollow tubes also make things more error prone, doing all the booleans on solid rods and then shelling the result, or worse case boring out the center, may also help.

brucebartlett

I'd love more robust tools for producing part's like this. I can see those junctions are going to become quite complex and that would be my guess as to why its not working, a small bit on an edge somewhere. I'd be trying to trim bits back to simplify the whole model and find the problem area's 1 by 1. 

What the end goal? This can make a difference as to how far to persist with problems like this. 

Manufacturing drawings?

Models For FEA?

CAD data for CAM? 

bobt

The most common cause of Boolean errors of this sort is trying to join tubes of identical diameter. A slight tolerance error in position, and now there is no longer a valid intersection. If possible I would make the cross struts slightly smaller than the lateral struts.

I can't tell from your images, but hollow tubes also make things more error prone, doing all the booleans on solid rods and then shelling the result, or worse case boring out the center, may also help.

michael_stalls

Cheers thanks. Thinking about it, it's not really a huge issue. It should still come out alright for the drawing and i only need the coordinates for the intersections to make the FEA model. Would there be any CAM applications that would help fabricate a truss like this?

All the tubes are hollow and when i use translucent the intersections are pretty complicated...i figured it might be something to with this. Ill try making them solid next time i make a truss