hexagon logo
Register Sign in
  • State Not Answered
  • Replies 6 replies
  • Subscribers 11 subscribers
  • Views 87 views
  • Users 0 members are here
Options
Related

Assessing Mode Shapes after a Linear Solve - Kinetic Energy?

jpatterson
I'm performing a LINEAR analysis of a full Tractor-Trailer model, which is generating many frequencies/mode shapes. I was hoping to mathematically assess these modes shapes instead of visually inspecting each one.
 
On the FEA side, I would look at the modal participation in each direction for each mode to do this, where say the mode with the highest vertical content would likely be the bounce mode, etc.
 
I figured out how to write out the kinetic energy table to a text file, but am wondering if this is the right way to go. After an initial look at this data it seems right/logical, but I wanted to hear other opinions or ways to get there.
  • 0
    Hi- modal kinetic energy is a great way to see which 'masses' are moving a lot, or lots of masses moving a little. It is ONE tool to help understand the modes. Another tool is Modal Effective Mass. This is turned on with the MEFFMASS case control command. Take a look at this webinar for good dynamics check out tools:
     
     
     
  • 0
    Hi Don,
     
    Thanks for the reply.
     
    Is "MEFFMASS" for use in ADAMS or NASTRAN? I'm running ADAMS...
  • 0
    Argh- I missed that- sorry... this is in Nastran! I can't speak to Adams...
  • 0
    Sorry I missed this. I've only ever used the kinetic and strain energy tables. I wrote a little script that turns the part numbers into names using the original ADAMS deck to make it a bit more readily comprehensible.
     
    I'm still mostly eyeballing the numbers and making a judgement about big and small, but it does seem a step up from eyeballing animations.
     
    I haven't done any postprocessing of the table itself to automagically identify certain behaviours, I tend to scan the table and have the animations open next to it. I could imagine looking for, say, wheel hop modes this way - but the energy table doesn't tell you whether it's an in-phase or an anti-phase mode, for example.
     
    I've also found that running a modal solution on a model with point follower tyres inside ADAMS is a quick recipe for a morass of confusion; for modal solutions specifically I configure the model with a vertical stiffness and no in-plane stiffness. I use soft springs at the CG to stop it wandering around on the rounding errors and discount the very low frequency modes that introduces.
     
    For motorcycle work I have had literally zero success in over 20 years of sporadically re-trying to believably calculate weave and wobble modes on point follower tyres. Instead I run a time domain test with an impulsive input (hip flick or bar nudge) and postprocess the fit of a damped oscillatory response. This gives me outputs that correspond really well in character with published work in the literature (eg Robin Sharp). I have similar failures with yaw/sideslip mode on passenger cars.
     
    Not sure if that helps?
  • 0
    Resurrecting this topic.
     
    I'm again trying to (mathematically) assess mode shapes from a full vehicle model. I stumbled across an option on the "LINEAR" command called "ENERGY_PER_MODE" but am having trouble getting it to work so I can see what it does. anybody tried this one?
     
    image.png
    
    image
© 2025 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement