Jump to content

Evangelos Kounis

  • Content Count

  • Joined

  • Last visited

About Evangelos Kounis

  • Rank

Recent Profile Visitors

1321 profile views
  1. The data you provide refer to the behaviour of the material under tension in each direction. The von Mises criterion contains multiple stress components and therefore what you describe in the first post is reasonable. What you should check is whether the individual normal stress components (sigma_xx, sigma_yy) exceed the values prescribed. My guess is that they don't. At least not much.
  2. Thank you for your answer tinh. I should have mentioned that i am using RADIOSS Block 120 and not Optistruct. Thanks anyway.
  3. Does anyone know how stresses are output in the material coordinate system?
  4. Hello fellow users. Just a quick question. I am trying to perform single element tests for material calibration. Doing the tension and compression in both directions was not a problem as you can imagine but the pure shear element is giving me trouble. To simplify things, let me ask the following: Is it possible to restrain the movement of a node so that it moves on the radius of a circle? In other words, i want the distance of a node from a fixed point to remain constant so that no normal stresses can develop. I am attaching a to help you understand what i have done already and where i want to go with it. The white dots are the corner nodes. Currently they lie on the rectangle. I want them to lie on the circle's circumference. A way to resolve this would be the definition of a cylindrical coordinate system where the radial distance "r" would be preserved. Is something like that possible? I am using RADIOSS Block 120 if that helps.
  5. There is no easy way to do it. In fact, scripting left aside, you can't.
  6. If you want delamination and cohesive elements it is highly likely that you are in the wrong place . By that i mean that sadly RADIOSS is not doing a good job helping us (i want that too). There are a couple of things you can try though. Adhesives: First of all take a look at MAT59 (http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/mat_law59_connect.aspx). It is used to model glues and spot-welds, etc. You can also link it to failure keywords such as the /FAIL/CONNECT (http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/fail_connect.aspx). Laminates: What material law are you using for your plies, MAT25? If yes there is the GAMA_INI and GAMA_MAX values which dictate delamination which is also plottable through DAMAGE 3 (http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/anim_eltype_restype.aspx). You got a lot of reading to do, so i am going to leave you to it. There is no easy way around.
  7. By connections you mean adhesives? Are you using shells, solids? You need to be a bit more specific friend .
  8. Good morning. If i understand correctly, i would suggest using the imposed acceleration or imposed velocity which can be founded in Tools -> BCs Manager. Under direction, XX, YY and ZZ correspond to rotations over the x,y and z axis. You can define a curve there and apply it to the tank i suppose. All of the above refer to RADIOSS BLOCK 120. Tell me if you need anything else.
  9. The plies do not get displayed and i think that they should fix that, it is not your fault. To define the thickness, you just add the value you want on the ply card (P19). The total thickness is defined in the laminate (P17). Be a bit careful with the positioning of the plies! To define the orientation, you need to define two angles. " The angle for layer i: = i + where i is defined in the /PROP/STACK for layer i. " The first angle, Phi_i, is defined in the laminate (P19) and the second, DeltaPhi, in the ply (P17). For a 45 degree ply you can either go Phi_i=0 and Delta_Phi=45 or the opposite: Phi_i=45 and Delta_Phi=0. Basically you can choose anything that sums to 45 but the rest do not make that much sense. Also, the value Phi_i is common for all plies in the laminate. I do not get the "...variable thickness whose sum is not equal to shell thickness.". The variable thickness part is understandable but why is their sum not equal to the shell thickness? After all, you define the shell thickness.
  10. What solver are you using? I am only familiar with RADIOSS. The best (and only?) way to do it is with the use of Material law 59 (CONNECT). You will also need Property type 43 and /FAIL/CONNECT. Take a look at the reference guide. material : http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/mat_law59_connect.htm property : http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/prop_type43_connect.htm failure : http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/fail_connect.htm
  11. Hello Chandru. I am using Block120 but they should not be that different. I imagine that you want to use shells. You need to create a material law 25 and a property type 10 or a material law 25 and property types 19 for each ply + p17 for the laminate If the material for the interface between two plies is not simulated, the thickness and material of each ply is the same, the two approaches are identical. For simplicity, lets use approach 1 (M25, P10). In the property card, under the variable "N" you define the number of plies (or sublaminates) which at the same time is the number of integration points through the thickness. Depending on the selection of "N" an equal number of "Phi" entries will appear. That is the orientation of the ply with respect to the LOCAL system. The LOCAL system is defined with the "Vx, Vy, Vz" entries in the same card. For example: (Vx,Vy,Vz) = (0,1,0) it means that direction 1 of orthotropy is parallel to the global "Y" axis. Phi1=45 means that ply number 1 has an orientation of 45 degrees with respect to the local system, so for the example given above, the orientation with respect to the global system would be 135. I think you get it.. Finally in the Material card you enter all the material parameters in the LOCAL system again. Finally, you assign both to the component and you are done. There are also other ways but that's an easy way to start. Hope this helps. Let me know if you got more questions
  12. It turns out that there are limitations on the application of XFEM keywords, the violation of which will give the error described above. Those limitations are the following: Property type 1 or 11 Number of integration points through the thickness is limited to 2 (denoted as "N" in the property cards)
  13. I was aware of that bug since i am experiencing it too but i thought you had a problem with the file format only. I could have helped earlier . Anyway, you figured it out. In version 13.0 i would add a drop down menu for the output variables as well so that the user does not have to write the actual variable himself.
  14. In the most recent update they fixed some errors and bugs with OSSmoth and PCOMPP. Maybe you want to give that a try.
  15. Hello Takky. It is nice to see that the community is getting bigger. I had similar problems in the past. It seems that there is a problem with the T01 file and HyperGraph. I am assuming you are using Block 120. You can create an ABF engine card (Tools -> Create Cards -> Engine Keywords -> ABF). This option generates an *.abf (and a *.tmp) file containing the output variables you have defined. It is basically another file format with which HyperGraph is compatible. In the ABF card you specify the sampling rate (first entry) and the rate at which the *.tmp is translated into the *.abf (second entry) Leave the second one to 0 unless you are having trouble with the simulation (e.g crashes, or starts diverging). More info you can find here: http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/engine_abf.htm Have a nice day
  • Create New...