Jump to content

Chris Coker

Moderators
  • Content Count

    18
  • Joined

  • Last visited

Everything posted by Chris Coker

  1. depending on how the coupler is defined, it's either: Rotation of steering input (in radians)/travel of steering rack (in mm) or travel of steering rack (in mm)/Rotation of steering input (in radians)
  2. In general, it's a better practice to apply your initial conditions to the Bodies, rather than the Joints. In the real (non virtual) world, F=m * a. The forces and accelerations act on the mass. The joints add constraints to the degrees of freedom to the bodies. Applying an initial displacement to a joint implies that you want an initial translation to occur immediately at time =0, because the joint is not initially in the desired location for the start of the simulation. Numerically, this can lead to problems achieving a robust solution. You can also apply a Motion to a Joint as a function of time if you need to move that joint to a particular position. This is a more reliable way to position the joint, as you are applying an enforced displacement. Chris
  3. Are you saying the model doesn't run when you change the joint from non-compliance to compliant? OR are you saying the BISTOP is active, but not creating any reaction forces. It's possible that the references you are using to define the BISTOP are removed when changing the JOINT type from non-compliant to compliant. It's also possible that the torsional rate of the bushing it too high to allow the BISTOP to function as expected, i.e. it's not rotation enough, due to the stiffness of the bushing. Investigating the .log file for any warnings when model behavior has changed, is always a good first step. Chris
  4. If you will look in our documentation, the is a very detailed explanation on the difference in normal force calculation. You can search for "MotionView Contact Best Practices", and that should give you the right location. To answer your question about which one to use....The answer is "use the one that works best for your specific problem" Sometimes I find that Impact works well for some kinds of problems. If I have trouble, say with either model robustness, or too large penetrations, then I will sometime look at the Volume method. Ultimately, the Volume method seems to be my personal favorite for more complex problems, but I have successfully used all three choices (Impact, Poisson, Volume).
  5. An advanced joint should work to define the mechanism, but I don't think you will get the correct stress (on the surface of the cam contact)
  6. You have to be very careful about how you set up an enforced motion on a problem like this. Using Marker Motion should work, but you have to take care to attach the marker at the correct location, so that you do not force the car off of the path at the motion progresses through the simulation.
  7. There is likely some incompatibility between your user/custom wizard, and the way the default library model/events work. I would suggest that you study how the default library and event topology is configured, and use that as a pattern for designing your own custom wizard. There is specific code (and logic) for how the events communicate with the vehicle models. If you do not match this logic flow in your custom wizard, the events will not know now to connect to your custom models.
  8. Direct 3d contact onto flexbodies like you are attempting is currently not supported. There is a complicated workaround for this, but in general this is not recommended. Essentially, you create many interface nodes over the surface of the flex body. Then create tiny rigid bodies at each one of these location, at attach them with fixed joints. Then, you would define contact for each one of the small rigid bodes that were created.
  9. Did you create the vehicle model from the existing wizard library, or did you create this from scratch? One thing I have noticed in your screenshots, is that the icon for "Altair Driver" does not look like a driver's helmet. Also, the modeling entities for the "Altair Driver" do not get exposed to the user when everything is set up correctly.
  10. I think the issue is that you do not have the correct preference file loaded. Under File -> Load -> Preference File. Select MBD - Vehicle Dynamics Tools, and then press the "Load" button. You may have to reload your model after loading in the preference file. If everything is loaded correctly, you should see a "Vehicle Tools" menu option at the top menu, and the icon beside the text "Altair Driver" in the project browser should look like a driver's helmet rather than the generic System/Event icon. Chris
  11. you need to add an event to the model. With 2017.2, right click in the project browser over the "model" folder. Then you should get a pop-up menu with a selection option to "add events". Once you add an event, you will have control over the specifics for that event.
  12. Hello Adit- Did you build your vehicle using the standard vehicle wizard library? Use analysis -> Task wizard and you will see full choice of events. Note, if you did not build your model with a powertrain, some of the tasks will not show, since the required logic is not included if you do not include the right systems when you initially build the model using the wizard. Also, it would be helpful for you to provide information on what version of Motionview you are using. The worfklow has been evolving a little over the past few versions.
  13. Part of the issue might be your file naming convention. "filename.mdl..xml" is likely going to create issues with the workflow. A MotionSolve run file should look like "filename.xml" Also you are using a very old version (13) of Hyperworks. I would recommending updating as our tools have improved a lot since version 13. Chris
  14. Yuvaraj, Just to add some more detail to Felipe's great advice. It appears that maybe you are using joint friction, as well as friction on your contacts? These are both very computationally complex, utilizing a LuGre friction model for joint friction, and in general it is common practice to leave these features turned off, unless we can answer two questions: 1) Is including this friction absolutely essential to capturing the required physics of my problem? In most cases, the answer is "no", or can be achieved using computationally more efficient methods like damping. Also, in the joint friction models, there are different levels of complexity that can be switched on and off. Only use the friction features that are required. For 3D contacts, static friction can add a lot of compute time to your runs. Ensure you need these feature before activating them. 2) Do I have all of the correct input data to populate the friction model correctly? The friction models required details about the size of the joint, friction properties, etc. Do not assume that you can simply turn on the joint friction model, and use the default values. Based on the size of your joints, and units of your model, the default values will usually not be a good starting point. If the answer to either of these questions is "no", but you need to damp your model's response to achieve a more physically accurate solution, a better choice is to add damping to the model. This can be easily accomplished by adding a bushing on top of an existing joint, but don't assign it any stiffness properties. Just give it some damping, in the desired directions of interest. This is computationally much more efficient that the LuGre model, and can be used with great success. It's also curious that you are using the VSTIFF integrator, and not DSTIFF. Also the integrator is taking a very, very small time step, with your max step size set to 1e-08. If this is required to get your models to run, there is likely some other issue with your model configuration. Chris
  15. It's also possible that the model didn't run at all. Did you get a message window pop up when launching the run? If not, then it's almost 100% you don't have the solvers installed or correctly installed. Note, MotionSolve does not write out an .xml file, that is the input to MotionSolve
  16. Sushanth- The parameters you are changing only influence the calculated parameters seen in the SDF outputs (for parameters such as anti-lift, anti-dive, etc). If you are running static ride analysis, and the vehicle weight changes, then your spring rate (and spring preload) will likely need to change, to account for the change in vehicle weight. Changing the preload will change the load into the suspension, which will change the displacement, stress, strain in the flexbody.
  17. Hi Abdullabh- To answer your second question, the math behind flexbodies is based off of linear deflection, and linear material properties. You will not get good results attempting to model a rubber component as a flexbody. If you need to model something like a rubber component, you should investigate using the non-linear finite element feature in MotionSolve (NLFE). For your first question: MotionSolve does not currently support 3D contact with flexbodies today. There is a workaround to this, where you attach many small rigid bodies to your flexible body with fixed joints. These small rigid bodies can then be used with 3D contact.
  18. Hi- The SDF file is created by a custom user subroutine, and is specific to vehicle suspensions. This is not quite what you want. You could go this route, but would be overly complex. A better approach would be to use the "TextView" client in HyperWorks, and some templex math to calculate your metrics. There should be tutorials in our help that guide you if you search for "TextView" "Report Templates" and "Templex" Regards, Chris Coker, Technical Manager - Altair
×
×
  • Create New...