Framework calculations in 3-D 2

[MECHIS]

Hi!
I'm currently doing a project where i'm constructing a framework for an engine mount connecting an airplane with an engine. The project involves both calculations by hand and with CAD(Creo), and i have no problem with the CAD part as i have done the simulations.
The part with doing calculations by hand have caused some problems where i'm doing the calculations in 3-D, and its causing some problems.
The top and bottom diagonals are not to be considered in the calculation by hand. Therefore you have 6 symmetrical beams.
So far what I have done in the calculations is that i've created a equation system with help of directional cosines which seems to get me the wrong answers. So i'm looking for some help in creating the equations system for this problem.

 

[3DDave]

Virtual work is a good method for such things.

 

[rb1957]

why are the top and bottom diagonals neglected in hand calcs ?

sure their primary job is lateral load, but they'll work under vertical load, what I guess you're hand calc'ing.
sure they complicate things, but that's what truss member do to you ... create loads in unintended directions.

direction cosines should work.

but the structure is redundant so that'll be your hand calc difficulty.

you could do something iterative ...
neglect the top and bttm diagonals (for vertical load; negect the side diagonals for lateral load)
each side reacts 1/2 the vertical load and one side of the couple from moment.
direction cosines should distribute the load between the three members,
direction cosines should complete the loads in the members.
now how have the ends moved, and what loads does this create in the diagonals?
rework to remove these "unwanted" loads ...

of course you'll FEA for the proper loads



another day in paradise, or is paradise one day closer ?

 

[Sparweb]

Simplify first!
I would start with the assumption of pin-joints on all truss members. Analyze the mount as if it were a pure truss with no beam or bending elements at all.
Furthermore, you would be giving yourself a model that DOES permit consideration of the diagonal members (and wouldn't work without them, in fact).
Will the tensile/compressive loads on all of the members, you can then do a reality check on a second model, using beam members, if you want.
Good frame design depends on alignment of all member centerlines to common points - which not only prevents secondary moments in the joints, but also makes the truss correspond to the pin-joint model, too.

STF