Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by rjromijnders

  1. Dear all, I'm working with HyperGraph 2017.2 on a Windows 10 desktop PC. I would like to plot a curve in Hypergraph and color the shaded area under the line. Via Google I have found this link, where it seems that there is a check box available from the Curve attributes panel on the Line attributes tab. However, this check box seems to be missing for my Hypergraph: Does anyone have an idea how to solve this? (Or any other method to color the area under the line?) Kind regards, Robbin
  2. I have already found what I am looking for: hwi GetSessionHandle session_handle session_handle GetProjectHandle project_handle session_handle GetDataFileHandle data_handle $input_file foreach val [data_handle GetDataTypeList false] { puts $val } In this way I get a list of all the entities that are saved to the Time History output file. Next, I am plotting the curves for selected entities and automatically exporting the curve(s) to a .csv file. The outputted .csv file can be further processed using any desired scripting language. Kind regards, Robbin
  3. OK, this will work for me as a temporary workaround. Thanks George! One more question: how could I get a list of the available output request from my T01 file? So, the Y-Type, Y-Request and Y-Component, but then before I open the T01 file in Hypergraph?
  4. Hi George, thanks for the response. I get the point of reading the peak value from Hypergraph X-Y plot. What if I do not want to use Hypergraph, but instead I want to use a script that directly processes the T01 file? Or is there any easy way to process the T01 file, like... def processT01(inputfile="runnameT001", param="IE"): 1= Load the T01 file 2= Read the values for the defined parameter, in this case "INTERNAL ENERGY" 3= Convert the values to a comma-separated value, e.g. Time, IE 0.000, 3150. 0.100, 3155. 0.200, 3158. Save in this format to a new file: "my_output.csv" In this way I can further post-process the data. I need to do this for multiple simulations, and for multiple time history output data per simulations (e.g. 5 springs per simulation). That is why I want to automate this process. Kind regards, Robbin
  5. Dear all, I was wondering whether it is possible to directly post-process Time History data output with custom-written Python scripts? What kind of format is the *T01 file? Is it a binary format (is the format known)? A short synopsis of what I am working on is an automated peak detection method on certain Time History Output data. I want to find out at which time a peak value occurs, and subsequently want to look up the Y-value at that same time for a different Time History Output. If a script (different from Python) is available that can directly be applied to the *T01, then I am also happy. Kind regards, Robbin
  6. Excuse me for the late feedback. Indeed, I can plot strains now on my component. thanks! On the property card the value Istrain = 0 (such that it was looked up on the /DEF_SHELL card). On the /DEF_SHELL Istrain = 0 (such that it would take the default value, i.e. Istrain = 2: no) Istrain: Compute strains for post-processing flag (Integer) = 0: default, set to 2 = 1: yes = 2: no I have changed the parameter to Istrain = 1; for both the PROP card and the /DEF_SHELL Thanks for the reply!
  7. Hi there, currently I am running explicit FE simulations with Radioss Block. The simulations are setup using a "master" file, that calls a number of input files. The input files comprise sub-assemblies of the total model. Now, in my results I am able to make a contour plot of the (Von Mises) stresses. Maximum Von Mises stress is 275 MPa. The animation files show indeed excessive deformation of the component in a specific include, which is expected. However, when trying to plot the strains, it seems results remain 0.000e+00 for the entire simulation time for this component. I am able to view both stresses and strains on the other components. In my engine file I do request to output the strain tensor: #----------------------------------- # Animation file control #----------------------------------- /ANIM/DT 0.000000e+000 1 /ANIM/SHELL/TENS/STRESS/ALL /ANIM/SHELL/TENS/STRAIN/ALL Could it be an issue on the material or property card? #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| /MAT/PLAS_JOHNS/14 material_name # RHO_I 2.800000E-06 0.0 # E Nu Iflag 70.0 0.3 # a b n EPS_p_max SIG_max0 0.2 0.3 0.5 0.0 0.275 # c EPS_DOT_0 ICC Fsmooth F_cut Chard 0.0 0.0 0 0 0.0 0.0 # m T_melt rhoC_p T_r 0.0 0.0 0.0 0.0 Is anyone aware of the possibility to loose the strain for plotting purposes, whilst the stresses are still present?
  8. I have downloaded your script, and try to use it via: File > Run / Tcl/Tk script I am given the opportunity to select a component then, and to click Proceed. It seems like something is happening, but it took quite long. Hence, I did setup a small model of just a simple beam that is loaded on one side, and constrained on the other. Again I ran the script, choose the component, but nothing seems to happen. Do you mind sharing your source code, or explain how you would access the values (area, volume, mass)? Kind regards, Robbin
  9. Hi all, I was wondering if it would be possible to generate a table of components, with the area, volume and total mass. It would look something like this: myFilename.txt or myFilename.csv (at least a text-based file format) ComponentID1, ComponentName01, Area, Volume, TotalMass ComponentID2, ComponentName02, Area, Volume, TotalMass et cetera Previously, I have used Tools > Mass calc, and then manually noted down the values in an Excel file. My model is becoming to big at the moment to keep working like that. Kind regards, Robbin
  10. Where could I create the DREPORT card? It does not appear in Tools > Create Cards > "C" Or should I just manually add it in the .fem file, but at which line? Kind regards, Robbin
  11. Thanks Prakash. I also found it in the Optistruct User's Guide, it makes sense now
  12. I do not grasp why the formula reads C = (1/2) F2 (1/K). Why is the force squared? Also in the document of the last post, it says C = FT U, which makes more sense to me. Can anyone help on deriving the above-mentioned equation for C, C = (1/2) F2 (1/K)? Kind regards, Robbin
  13. Hi Prakash, I am looking at a compliance response, yet I am optimizing to minimum mass. I noticed that some load cases lead to quite high compliances (~100), others to low compliances (~10). In the optimization I would like to try to take into account that de constraints for the compliance apply equally for the different load cases.
  14. Hi all, I am trying to do an optimization for a model with multiple load cases. For each load case a constraint is defined, but would it be possible to somehow use weighting factors for the constraints? In that way I hope some constraints will become more important to the optimization. Kind regards, Robbin
  15. Hi all, how can I normalize the results with respect to the max of the results? For instance, I have a contour plot of the Von Mises stress of a mechanical indentation simulation. I want to compare this to a different simulation, so I would like to end up with a normalized contour plot. I have tried using the Expression Builder but haven't succeeded yet. I was trying something like STRESS / max( STRESS ), for all loadcases ==> T3.C7 / max( T3.C7, T3.C7 ) Does anyone have a clue? Best regards, Robbin
  • Create New...