Ansys Moment determination with Displacement Boundary Condition

[dhrupal93]

I have a parabolic 3D structure (an Antenna) which is mounted rigidly near the center by 8*2 = 16 bolts. I want to deform the structure. So, I have a known deformations at a particular line of CAD model/node (eg say at any edge node along z axis by 1 mm ) (displacement boundary condition). For practically doing this I have a two point moment actuator, which can give equal moments at both points/nodes for deformation.
I need to find that moment ( which will be equal at two points on structure.)(Constrained Moment)

A simple understanding model is shown in image top part.
It may be able to do with parameters for single moment prediction, but overall I have such 8 actuators (hence 8*2 moments) on the overall structure.

 

[rb1957]

do you want to deform the structure or rotate it ?

if you're trying to deform it, then probably the easiest way is to apply 1 lbs along the line of action of the actuator, see the deformation caused, compare to the desired deformation, and scale the applied load.

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

 

[dhrupal93]

To rb1957
I want to bend the structure.
The method you are telling is correctly applicable till one (or few) actuation points. But in the Antenna, their are 16 such points (in the lower part of image, see that I have mentioned M1, M2 ..... M16 all may be different). Hence, response due to all is non-linear and coupled. So, it will not solve my case.
I think some constraint of keeping the unknown moments at all those particular nodes to be equal (in pairs) will do the job. How to do this?

 

[rb1957]

so you're constraining the centre pick-ups and you want to apply a set of forces that's move the rim point 1mm.

with 16 possible forces there are more than 256 combinations that'll achieve the desired goal. Are these 16 inputs constrained together, or 16 "random" numbers ?

in deflecting the chosen point 1mm, do you care what happens to the rest of the rim ?

i think you're beam model is a little too simplistic. it still has a bunch of combinations ... you could achieve all the 1mm deflection with M1 or M2 or Mx alone or in combination. another consideration, are the points tied to actuators ? if they are then the problam becomes much more complicated ... what i mean is, say you have M1 and M2 applying load to the beam; well M1 causes deflection at pt2 and M2 at pt1. now normally we take these into account, but with your question i can see that they might "travel" with the beam.

i think FEA is the only way to determine what i think you are asking.

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

 

[dhrupal93]

1) Yes, I am constraining Center pickups. What set of forces and moments I will have to aaply need to be determined to get displacement of A SINGLE node on rim (not entire rim).
2) 16 random loads (will be bounded by maximum loads at those points). Surely their are more no of inputs than desired, so multiple solutions will be their.
3) Currently, I don't care for other points.
4) Updated my model little bit for better understanding.

 

[rb1957]

1) there are many combinations that'll achieve the same result (of displacing a point by 1mm).
2) start with applying 1 (or 1000) lbs at each of the 16 actuators, and see what happens at your desired point. you might get a table of F1 produces a displacement of d1 in/lbs, etc. i think the combinations may be linear addition.



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