AnsweredAssumed Answered

CosmosWorks/ motion for vibration/ earthquake

Question asked by Knut Sand on Apr 3, 2008
Latest reply on Apr 21, 2008 by Knut Sand

Like • Show 0 Likes0
Comment • 4

Hi

This is an variant of another thread inside here, "following a closed path".
However; my vague idea was to use a modelled frame (A skid with some equipment on; the object to check), place this frame on a "floor" and under this , attach a "ring", through this ring a 3d sketch/ line with horisontal/ vertical bend, if this line is pulled in a determined speed, in horizontal movement, the floor, and the frame will (have to) move like during a simulated earthquake....
If I have understood this correctly then, I then can check the stress in the design, the legs, the stiffeners...?

Or am I just influenced by too much heavy medication now...?

Any/ all ideas will be appreciated. But please restrict ideas to the Cosmos program(s) that is included in the SW premium package.

No one else has this question

Outcomes

4 Replies

Ian Hogg Apr 3, 2008 3:55 PM
Mark Correct
Correct Answer
Hi Knut,

In the SolidWorks premium package, you have no realy way of modeling the response to an earthquake type vibration. In COSMOSMotion, you can give an osciallting displacement as a function of time to mimic the movment, but the parts are rigid (ie they do not deflect). You can look at results at given time steps from the motion simulation on individual parts, but you will not get the true response of the entire structure. It will most likely underpredict the stress and deflection since at the extreme travel of motion, the actualy movement of the parts will not be cumulative through the struction.

To be honest the best was is to model this is as a dynamic analysis in COSMOS Advanced Professional. This product is tailored to this type of analysis where you can provide excitation via displacement, velocity or acceleration and observe the response of the structure as a function of time.

So, probably not the answer you were hoping for, but the reality.

Cheers,

Ian
Like • Show 0 Likes0
Actions
- Knut Sand @ Ian Hogg on Apr 4, 2008 8:38 AM
  Mark Correct
  Correct Answer
  True, not the answer that I hoped for....
  Not even if I assume very conservative limitations, like; no deformations?
  
  I may end up talking to my accontant on this... (is that the name? for the guy i hire part time to hold me down when I go shopping ("Huh?! I can buy that, I have my own company" (more correctly; a one man band, and truly not too much money involved...)))
  Like • Show 0 Likes0
  Actions
  - Ian Hogg @ Knut Sand on Apr 4, 2008 10:15 AM
    Mark Correct
    Correct Answer
    Hi Nnut,
    
    With Motion, you will see the rigid body dynamics. So any accelerations will be an combination of the base excitation and the mechanism dynamic movement. It may give you a ball park figure for the accelerations acting on each part and from there to see the stress, but it is hard to guess how accurate as that heavily depends on the design.
    
    If you are doing this for a consulting/contracting like job, and you have to meet certain standards. For example there are ASTM, Mil Spec, ISO standards for vibration testing, then you should be using. Not sure if this is an electronic package or not, but normal physical tests involve a shaker table and some pretty severe loading. I used to work in an engineering office above test labs where they'd run the shaker table and you could feel & hear it through the whole building. But you should really be repeating those tests on a computer if your doing that line of design work.
    
    Cheers,
    
    Ian
    Like • Show 0 Likes0
    Actions
    - Knut Sand @ Ian Hogg on Apr 21, 2008 5:56 PM
      Mark Correct
      Correct Answer
      Well here's the problem; I have some skids, no big stuff, no HiSci nor anything, just some skids/ cabinets,with some pipes/ tubing. Now they want to know if the equipment is useable or repairable after an earthquake.
      
      The peak Acceleration occurs after 0,2 sec.
      Mean Spectral Acceleration Horizontal; 1,81 (g).
      Mean Spectral Acceleration Vertical ; 1,29 (g).
      After 2 sec its down to approx 0,2 (g).
      
      Is it possible to get a "usable" evaluation to this problem by just adding these (peak) acceleration forces to the leg's connection points? Measure the stress by probes, nothing really can happen if not minimum 2 legs reach full yield in at least 2 cross sections each? (Ehh, or cutting off some of the existing bolting then...)
      
      I did try this with one extremly simplifyed skid, and it seem ok... (tm). However I do not quite see where this "ball park figure" method may be too much way out of course.
      
      Any (easy) ideas will be appreciated. Especially if they don't involve purchasing of an extra Cosmos package.
      
      Regards
      KnutS
      
      Like • Show 0 Likes0
      Actions

CosmosWorks/ motion for vibration/ earthquake

Outcomes

Related Content

Recommended Content