19 Replies Latest reply on Jan 31, 2013 3:27 PM by Jerry Steiger

    Static analysis of a Telescope structure [PLEASE HELP!]

    J. F

      Hi Everyone!

       

      We are a group of students working on a project and in our school they don't teach us solidworks, so we have to learn all of this stuff on our own. I have two problems that I will list here:

       

      I will be giving you a short explanation below and will provide you YouTube links, so that you can better understand the problem. Please let me know if you need anything else. My directo contact is Xtremophilelor@yahoo.com

       

      We have two different designs so far for the telescope, the first one is a Hexapod platform, and the second one is a Tensegrity structure. Please look at the two links below:

      Hexapod: http://www.youtube.com/watch?v=vg2LcMhnt0s

      Tensegrity: http://www.youtube.com/watch?v=21p7bQ9KSLc

       

      Our problem with the Hexapod: As you can see the upper ring is holding the secondary mirror at the center, and  the ring is supported by the 6 compression tubes. I need to find out the forces on each of the compression tubes when the entire assembly is being subjected to 20g of gravity (20 * 9.81 m/s^2). The results tool just doesn't give us the result! And what is the free body force?!  I am using bonding to hold all of the tubes in place, because I don't know how to use bolts!

       

      Our problem with tensegrity: We are using 9 strings, or spokes (like bike spokes) to hold the secondary mirror in place in the middle. These spokes MUST be pre-tensioned so that we can have a rigid structure with high frequency. I'm unable to put these spokes in pre tension. And in the video they are moving like noodles!

       

      Please help us with these issues as we have tried everything we could.

       

      We are students, so if there are any fees that we need to pay, please let me know.

        • Re: Static analysis of a Telescope structure [PLEASE HELP!]
          Justin Strempke

          J.F.,

           

          If you are interested in basic reaction forces, you need to run a static analysis.  Both of your videos display frequency analysis, for which are meant to only show you the different modes.  When you want to subject the Hexapod to 20 g's, you just add the gravity force and alter the value (assuring the correct direction).  Just be sure to use the correct fixtures/restraints.

           

          The Tensegrity model it a bit more tricky if you want any results from the cables themselves.  You can always use temperature, determining your tension and distance and altering your cable length, then removing heat from just those parts so they retract.  I know you can use bolts with a pre-tension, but unsure if you can use them over such a length (may be able to add two concentric cylinders on each parts and use a REALLY long bolt!)  Bolts are generally difficult, but if you need a pretension they may be worth your looking into.

            • Re: Static analysis of a Telescope structure [PLEASE HELP!]
              J. F

              Justin,

               

              Thank you so much for trying to help with this. It was my mistake, I did not provide you with the screenshots of my Static analysis. I have done that, and that is indeed where the problem is. When running the staric analysis, I used bonded connections to hold the tubes inside the holes, and I also tried rigid connections to do the same. Both gave me the same results, which seem to be correct. However, I need to find out the Stress in each part of the telescope, and especially in the tubes. I have included a few screenshots for you to see.

               

              As for the botls, I don't know what preload to use! 100N? 10N? in torque or axial? (probably torque). And Which surfaces to click.

              I am trying to use your advise on tensegrity now. The Hexapod is the very important one however, so I hope we can get that fixed.

               

               

               

               

              Assem1-Static FEA under 20g vertical-Stress-Stress1.jpg P2.jpg

              P3.jpg

              P4.jpg

                • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                  Justin Strempke

                  If you have the static analysis results, you have the tube stress - unless you're interested in something different.  You can probe the mesh (under Simulation > Plot Tools > Probe) on the tube to find certain results.

                   

                  It may also be easier to create a new, single part model for the analysis.  This is common to have a different simplified model that the construction version.  Just as well in your assembly analysis you could treat the rods as 'beam' elements - a bit quicker and more reliable since you *should* have at least 2 element thickness on beam shapes when doing solids for proper bending stress results.

                   

                  For the bolts, again it would be a somewhat different model, where you cut most all the center part of the tubes away, leaving two concentric tubes.  Use a tight connection bolt to attach the two, with the given preload set to what your cable tension would be.  I'm not certain this would work mind you; I haven't used bolts across spans like that but it would allow for easier preloading.

                    • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                      J. F

                      Ok. I used the probe tool, but it just gives me minimum and maximum, and RMS (what is RMS?). If I want to know what it the total sress on one of the tubes due to a force, I can't find it that way. It kind of lists all the nodes and the stresses on them. I'm looking for let's say the total force on the tube, that's it. What is the SUM for? It sums up all the stresses?

                        • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                          Bill McEachern

                          if you tubes are separate bodies you can set the compute free body forces flag (usually on by default), make an exploded view, use the forces result tool - pick the mating face onthe end of the tube say and it will give you the free body forces/loads. I ddi not read teh whole thread so maybe I am missing something.

                            • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                              J. F

                              Can you please tell me the step by step? It is already selected. Now how do I find out the force on each tube? Clipboard01.jpg

                                • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                  Justin Strempke

                                  Are you wanting the forces or stress on the tube?  I'm a bit lost that your probe doesn't list the stress value...  Did you just use 'At Location' option and pick a point on the tube or use Selected entities?  If you want the max you can select a tube and read the Max value.  If you want forces you can do as Bill suggested.  After your sim has run, RMB the 'Results' in your sim tree, and select 'List Result Force...'  You can then look Reaction forces at your fixtures, Free Body forces at any given location, or Contact forces.

                                    • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                      J. F

                                      If you look at the picture below, I am selecting the component to find the forces on the tube. I want to know the load on the tube. so is the load 100.75N ? I don't think that makes sense. You know how you do a static analysis by hand, and then you can pick the load on the tube by solving for the components? Let's just say, you put a 400N on a chair, and you want to know the loads on each leg. Let's say it will be something like 100N on each leg. So how do I do that in this analysis?Clipboard01.jpg

                                    • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                      Justin Strempke

                                      I took a look at your tensegrity video, and am curious what exactly you're after.  Are you wanting to know the structure stress/forces (as with the Hexapod) due to the cable tension?  If you know what tensile value to use, then I would exclude the cable parts from the analysis, and use spring connectors.  You can use extension only, and apply a preload to your desired value. 

                                        • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                          J. F

                                          Ok thanks! I will give that a shot for the Tensegrity. I will remove the cables and use springs to tighten everything up. Please look at the picture below. The secondary mirror's weight is 1.54 Kg and I'm putting a force of 196 N on the whole structure (gravity*20). So I should see a resultant force of 1.54 * 20 * 9.81 soewhere here...but I dont. And I'm not getting what's whith all of the forces listed on the mirror, why not just the resultant? One last thing about the Probe tool...when I select an "Entity", it will only select one face of it! I don't know why it is so hard to just select the tube, and see what the total force is on it!!! in X, Y, and Z!

                                          Clipboard01.jpg

                                            • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                              Justin Strempke

                                              All the forces shown are all associated face loads for the selected body.  The middle two are simply the top and bottom being subjected to gravity, whereas the edge is your contact load and *should* register your desired 302 N.  If it's not, I would check your mesh density, it's likely it will need to be more fine at that surface to reflect the contact more appropriately.

                                               

                                              When you use the probe, it's largely designed to do just that - probe a certain location.  You want to only be able to see certain node stresses or face stress as having the entire part stress will not mean anything.  I'm thinking to get what you want for the tubes, use the free body tool and select the surface on the *top* ring where the tube slides into.  This will display the connection/contact force between the two parts.  They should all be uniform, and again you may need to use a finer mesh at those locations.  Your part is also axisymmetric, so you could split it into thirds and reduce your work/sim time.

                                               

                                              I assume you are required to simulate this as well as use hand calcs, do you have the reference number that you should be seeing for the tube force?  It's just a matter of the lens/holder mass *20g, divided by the number of tubes, then resolved to their angles as you've mentioned with the chair.  If you can chop your model, as is needed at times, remove your lower base and apply fixtures to the bottom face of each tube and run.  You can then view Reaction forces at each, and should give the same results.

                                                • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                                  J. F

                                                  Hi Justin, thanks for that.

                                                   

                                                  You see, here I went ahead and turned my mesh to the finest possible. the upper plate + the little mirror in the middle weight 8.923 Kg. So as you said, (8.923 * 9.81 *  20)/6 = 291.78 N. But I'm not getting that anywhere. If you look at the picture, I'm selecting the inside surface of the hole where the tube slides in, and select a plane that is prependicular to the tube...but nothing. It is as though the update button doesn't work!   I'm scratching my head here,. I'l try to just simulate one rod and one tube, to see if that is gonna get anywhere.

                                                  Mesh.jpg

                                                  force.jpg

                                                  reaction.jpg

                                                    • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                                      Justin Strempke

                                                      You have a few things going here - first is if you're trying to measure the force for just the mirror, you need to refine your mesh at the mirror-holder interface (use a mesh control).  Secondly, you're trying to measure the reaction force of the tube when there is no fixture to measure.  You need to remove your base and use a fixture to hold the bottom of the tubes.  Go to your parts like, and RMB > Exclude for the base to remove it.  Then you will be able to utilize the Reaction force.

                                                       

                                                      If you (can) post your model, I can take a look when I have a second.

                                  • Re: Static analysis of a Telescope structure [PLEASE HELP!]
                                    Rick McWilliams

                                    I usually model and build a telescope truss using spherical bearings. There is no bending of the turss elements. The nose piece and secondary holder is important. I use wide vanes to increase stiffness in the hope that vibrations will die out quickly.  The static bending is usually so small that it simply changes forcus and pointing by a small amount. Strength was always secondary to stiffness.

                                     

                                    I think that you will get excellent results for the mirror on its suspension system. Are you using a 18 or 27 point balanced support? I designed a 30 inch thin  mirror folded telescope before solidworks. Everything needs to be kinematically supported. Fun project.