Jump to content

Leaderboard


Popular Content

Showing most liked content since 05/01/20 in Posts

  1. 1 like
    In fact it depends to you work and to your working method. With more than 20 years of FEA I never need this feature
  2. 1 like
    Hello robertavarela and L Moretti, thank you very much both. Will try the suggestions in the upcoming week. Regards, Rok
  3. 1 like
    Rok, I see now what you mean. I suggest the following steps then: 1) Use Compose and its OML Math functions to create whatever inputs you need in Flux: as Flux has a native interface with Python, these arguments could be simply written in a .py file that can be loaded in Flux 2) Create a batch file in Compose calling Flux and passing these parameters: the executable is located at C:\Program Files\Altair\<your_version>\flux\Flux and you can pass a .py file as an extra argument 3) Post-process the results once the simulation is finished using Compose: Flux may output .h3d format that is natively understood by Compose with functions such as readcae. If you have questions about how to import the results, you can load your .h3d file in readvectorbuilder and see how the entities were written: With this, you drastically reduce the post-processing time by loading all relevant information with just a few commands. Regards, Roberta
  4. 1 like
    Rok, The most appropriate coupling in this case would be with Activate, our system simulation tool that allows co-simulation with Flux. One example can be seen here at Regards, Roberta
  5. 1 like
    Hi, In Flux 3D when you use a solid conductor region, you force the mesh to second order. This means that the number of nodes increase. When the number of nodes increase the simulation time increase. You can check the number of nodes when you use the magnetic non conducting region and solid conductor region, you will find that the number of nodes it is not the same. You can use the computation distribution to decrease the simulation time. Best regards.
  6. 1 like
    While there are forces and moments in the FSI_LOADS load collector, there are also imposed displacements on the model boundary as a result of breakout modeling. Running without any loads results in non-zero displacements. ThickComp.rar
  7. 1 like
    Hi, These counterintuitive results can be explained if the displacements are imposed: the thicker structure is stiffer and develops more stress under the same imposed displacements. For similar reasons when optimizing for stiffness: -imposed loads= minimize compliance -imposed displacements= maximize compliance
  8. 1 like
    I made lofts between the two PEC surfaces. This then creates a solid which has a region that you can set to dielectric, Then change the face properties to either PEC or dielectric.
  9. 1 like
    Time until a solution is attained is variable and depends on your time step heavily. If your elements keep distorting, your simulation may start at a 3 hour solve time, then when it calculates the next stage it will estimate 3.5 hours due to the deformation. You can achieve reduction in sim times with DT/NODA/CST or AMS (see video). You might also consider the multi-domain method.
  10. 1 like
    Glad to help. Hypermesh is unitless and it is the user's responsibility to follow a consistent set of units. https://www.altairuniversity.com/wp-content/uploads/2012/04/Student_Guide_55-57.pdf Since the material properties are consistent with the tonne, MPA, mm unit system, the model dimensions and loads should follow the same units. So yes, the moment should be Nmm. You are getting a compliance error because of the unrealistically high pressure load of 24000 MPa. Applying only the moment load in the model you shared, the displacement is actually minimal (0.0068 mm). The pressure load should be calculated according to unit system (1 MPa = 1 N/mm2 = 1000000 Pa)
  11. 1 like
    Hi, You should be able to achieve this by setting the required geometry faces to have their "Face medium" as "Dielectric boundary"... For example: Create a Cuboid With the cuboid selected, expand the "Details tree" Set the cuboid region to be your dielectric Set the required PEC faces to be "Perfect electric conductor" Set the required dielectric-to-air interface faces to be "Dielectric boundary" The same process can be applied to your horn antenna geometry by first creating it as a solid (filled) geometry, setting the internal medium and then setting the required face properties. Kind regards, Johan H
  12. 1 like
    Haha, it doesn't seem too likely that Hypermesh or Optistruct would be wrong. Maybe an understanding of how Optistruct segregates design and non-spaces could elp.
  13. 1 like
    Hi, While nonlinear buckling could be done in Optistruct, it is very is likely the implicit solver will experience convergence difficulties resulting in long run times or even fail due to nonconvergence. Alternatively use Radioss integration to solve your model with the explicit method in Optistruct. Therefore I suggest using Radioss explicit solver instead following this procedure: 1.perform modal analysis in Optistruct 2.in postprocessing create a derived load case>linear superposition>use small scale factor (1e-2 to 1e-3) 3.export the deformed shaped. 4. import the deformed shape into Hypermesh Radioss user profile and set up non-linear buckling analysis. By using the deformed shape obtained from the modal analysis the structure will have geometry imperfection triggering a buckling pattern consistent with modal and linear buckling analysis. Nonlinear buckling analysis is recommended to be performed within Radioss. Post buckling can be solved using nonlinear geometry (Implicit) loadcase. Use any of the Arc-Length methods to solve post-buckling analysis. Buckling.pdf1.46 MB · 130 downloads 2_2_snap_roof___implicit.pdf663.71 kB · 105 downloads There are two tutorials and one example on NL buckling: RD-T: 3030 Buckling of a Tube Using Half Tube Mesh (Hypercrash) RD-T: 3530 Buckling of a Tube Using Half Tube Mesh (Hypermesh) RD-E: 0300 S-Beam Crash RD-T_ 3030 Buckling of a Tube Using Half Tube Mesh.pdfUnavailable RD-T_ 3530 Buckling of a Tube Using Half Tube Mesh.pdfUnavailable RD-E_ 0300 S-Beam Crash.pdfUnavailable
  14. 1 like
    1. The example guide examples D4 (Radhaz example) and F2 (MRI) shows how to create near field requests. For far fields, several examples show creation of far field requests. Just search for "far field". In terms of the number of points/increments, if unsure, you can always request continuous far fields. Example H3 (reflector antenna) shows how to do this. 2. The wire radius should be according to the physical size that would be used, e.g. SMA has a radius of 0.65 mm. 3. Search the Examples Guide for "continuous frequency range" or "linearly spaced discrete points", there are frequency list and logarithmic options too. 4. The student edition limits are provided here. https://altairuniversity.com/feko-and-winprop-student-edition-limitations/ If you need higher limits, contact your study leader for access to the full Feko license. 5. Please clarify what you mean by delayed? 6. Please clarify, maybe give a drawing of what you are planning to do. 7. Yes, see Examples Guide G1 Time Domain.
  15. 1 like
    Hello Chris Coker, yes i am very new to the tools. The CAD feature recognition was exactly what i needed. Thank you very much! Regards, Aiko
  16. 1 like
    Yes and It depends on your FE model (I mean the number of DOFs should be higher ...) When you perform a FE simulation with X solver. It highly recommended to use 2 folders : - The main folder : where you will find the results of the simulation (*out file if you use optistruct). - The scratch or temp folder : where the solver write things to solver your problem. Here is an example : #!/bin/bash module load altair/2019 cd ${PBS_O_WORKDIR} optistruct Job1.fem -ddm -np 4 -nt 4 -core IN -minlen 10000 -scr /scratch/beegfs/RunTime
  17. 1 like
    Hi, perhaps mesh imprint functionality can be utilized. https://altairuniversity.com/wp-content/uploads/2017/10/Mesh_Imprint.pdf https://connect.altair.com/CP/SA/training/self_paced/aero_v13/PDF/chapter5_demonstration.pdf Mesh Edit Panel.pdf
  18. 1 like
    The amplitude of vibration is high because there is no damping in the second step. The damping effect is more pronounced at frequencies near the resonance peaks. The worst-case scenario is when the excitation frequency (ramp-up period) is near the resonant frequency and there is no damping. When you are unsure about the damping, perform a sensitivity analysis by varying structural damping in the realistic range and observe how much the responses of interest are affected. The proper amount of damping should also reduce interface force fluctuation. Unfortunately, I do not have the documentation on how the total interface force is calculated, but it is probably the sum of all contact forces. Your explanation is plausible, but it does not account for the period of vibration. p.s. I would appreciate being acknowledged in your thesis. Actually, your challenging queries also sparked my interest in the subject therefore I would like to have a look at your thesis. My contact: simon.kriznik@yahoo.com
  19. 1 like
    Hi, To fix this problem, you can delete the Unsupported card in HyperMesh in the model browser, Cards, Unsupported, right click and select delete. I'm not sure why the /ALTDOCTAG is showing up as unsupported. It is a HyperMesh only card that Radioss does not need. Thanks, Andy
  20. 1 like
    For non-radiating network sources, the source is applied outside of the schematic by choosing the network port in the voltage source dropdown box. For example, in Example Guide example E-3:
  21. 1 like
    In the below image, I modified the Example Guide model D-02 to remove the monopole and added a Voltage source in the cable schematic to the MTL cable. Seems simple, or did I misunderstood your question perhaps?
  22. 1 like
    Hi, For downloading the tutorial model files, kindly find the like attached: https://altairuniversity.com/model-files-for-student-edition-users-accompaniment-to-the-tutorials-in-help/ Thank You
  23. 1 like
  24. 1 like
    Since the AMS run matches other runs without time scaling, the conjugate gradient is below 30 and energy error stays low throughout we can conclude the observed behavior is not an artifact of AMS and the run can be considered verified. Because I haven't researched sloshing I am not able to comment on the validity (matching physical behavior), but I can provide my explanation of the simulation results. There is not a single peak, because the lateral acceleration is ramped up, is held for .8 s then ramped down. The cargo "sticks" to the wall as long as there is lateral loading- instead of bouncing right off. A single peak would be expected if the impulse was a lot shorter (without hold period) or there was an initial velocity imposed on the model. However, I think the acceleration pulse used is consistent with the railcar cornering. The lateral load is ramped up over a period of .3s which excites the bending eigenmode in the sidewall. In the graph below, the X displacement in the region of the highest amplitude of vibration is plotted. The period of vibration of the sidewall coincides with the period of vibration of the contact force; the hypothesis is as the sidewall vibrates against the cargo the contact forces exhibit pulsating forces in phase with sidewall vibration. Therefore the T01 seems consistent with the expected behavior (given the boundary conditions). But I have no idea why T02 does not overlap the T01 trend.
  25. 1 like
    hi 1. mindim is a minimum member size manufacturing constraint. It penalizes the formation of small members and reduces checkerboarding effect. It is recommended that MINDIM be at least 3 times the average element size for all elements referenced by that DSIZE (or all designable elements when defined on DOPTPRM). The average element size for 2D elements is calculated as the average of the square root of the area of the elements, and for 3D elements, as the average of the cubic root of the volume of the elements. 2. Volfrac is a fraction of design space, expressed in the range between 0-1. Volfrac as the optimization constraint with an upper limit as 0.2 means the optimizer will utilize only 20% volume of the design space. 3. Use LOADADD and SPCADD load collectors to combine multiple loads and spc load collectors, respectively then reference them properly in the loadstep. Free-size Optimization Manufacturability.pdf
  26. 1 like
    The animation file is not available for some reason. Can you share it through Google Drive? Compare the interface forces between AMS and CST runs on the same plot. It is strange that T01 and T02 plots not matching in trend nor the magnitude.
  27. 1 like
    Hi, it would be useful to have the animation (my workstation is too weak for this model)- use the HVtrans tool to reduce the file size to upload limits (only stress, contact forces and displacements; reduce frames if necessary). Verify the AMS run against a CST (or without mass scaling) run with a reasonable energy error (ERROR < +2%) and acceptable added mass (MAS.ER < 0.02) along its simulation time.
  28. 1 like
    Hi, This error occurs when a model is under constrained or incorrect material property definition or contact definitions are incorrect or missing. In your case (the units are tonne, MPA, mm) the pressure loading of 24000 MPA is causing excessive deformations. You should review the unit consistency and/or pressure load magnitude.
  29. 1 like
    The waveguide source reflection coefficient will be the ratio of the incident vs reflected power. So any non-ideal flange (impedance match with free space) would deteriorate this coefficient. If the flange causes higher order modes to be excited, at e.g the junction, you would need to set a higher number of modes to include (capture) at the waveguide port on the advanced tab of the waveguide port dialog. This will however require a finer local mesh setting on the waveguide port face. An aperture is allowed in an infite ground plane (Sommerfeld option, see Example guide A.10.2, "aperture triangles in an infinite ground plane") but you won't be able to use a waveguide source with the infinite ground plane. This is not supported. You would have to continue using the finite ground plane.
  30. 1 like
    Hi, A matrix is singular if and only if it has a determinant of 0. This means that When Flux try to solve the equation, found an error because the determinant is equal to zero. About the fact Flux can’t solve the mesh problem, there is not a command which allows Flux to correct the mesh in the event of a mesh defect. If you want to see the mesh defects, you can do it once the mesh is done. Go to the: General data Select Mesh Select mesh defect >> right click, add to visualization Best regards.
  31. 1 like
    A workaround is to split the model once into two halves. Then Union the two split parts. Mesh. Model cfx attached. Planar_spiral_2D_test_meshed.cfx
  32. 1 like
    Hi, import geometry>solidworks>split components by>part (default is body) Try other formats and/or import options if the issue is not resolved.
  33. 1 like
    So isn't there a way to verify with vectors on the elements, the x-axis (for example) direction after the user assignment? Thank you Simon you are the best!!
  34. 1 like
    Stresses are calculated in the material coordinate system. The material coordinate system may be defined as the basic coordinate system (CORDM=0), a user defined system (CORDM = Integer > 0), or the element coordinate system (CORDM=-1). Edit the PSOLID property and change the CORDM to USER and select the coordinate system which will be the material orientation. Alternatively, material orientation can be assigned to elements directly with analysis>systems>material orientation. In the model you shared the CHEXA elements are oriented with THETA option. The element angle option for Hexa solid (Theta) is very useful if the material system is changing from element to element. Make sure to define only one of these options to avoid confusion. Perform a simple 1 element tensile test to verify material orientation. The like button is a nice way of showing appreciation for the help
  35. 1 like
    Bonjour, Je vous prie de trouver à continuation les réponses à vos commentaires : -Maillage : Les maillage est, habituellement, un point délicat dans une majorité de projets. Si vous voulez un maillage uniforme dans la longueur du dispositif je vous conseil d’utiliser le maillage mappé. De la même façon, vous pouvez définir une relaxation de maillage moins importante dans le volume pour éviter un maillage plus relâché dans la partie centrale. Pour assurer un maillage suffisant dans l’épaisseur de peau, je vous conseil d’utiliser la fonctionnalité « mesh generator : Layers » (dans l’image). Vous trouverez un exemple de comment procéder dans cette vidéo : https://altairengineering.fr/resource/altair-flux-new-feature-overview (minute 1:50). Je vous partage aussi un pdf qu’explique d’une façon concise et complète les différentes possibilités de maillage qui vous sont offertes par Altair Flux. -Symétrie : Pour savoir la raison de cet écart je devrais avoir le projet. Peut-être, il s’agit de la connexion des parties symétrisées (en série ou en parallèle) mais aussi la façon de calculer l’impédance (elle peut être directement indiquée par l’utilisateur ou bien calculée à partir des caractéristiques du matériau). J’espère que cette réponse vous sera utile. Bonne journée. Alejandro 4_Mesh-Altair_Flux.pdf
  36. 1 like
    For shell elements, the contact and tie search considers the shell thicknesses. This implies that the defined search distance is expected to be the true distance between the shell surfaces facing each other. For example, in the case of shells without offset, if the geometric distance between the two shell surfaces facing each other is 5.0, and the shell thicknesses are 2.0 each, then the actual distance between the shell surfaces facing each other is 3.0. If the Search distance field is now set to 3.0, then the contact is generated as expected (in this scenario, if SRCHDIS is set lower than 3.0, then there is no contact generated). For more details, please refer to the attached document. To constrain the plate from sliding: -use symmetry (refer to 10.3) -constrain some nodes in the Z DOF Contact.pdf
  37. 1 like
    The PEC parts you can just delete the faces not directly touching the dielectric, the "top" faces if you may. The dielectric, yes, should remain an enclosed volume ("Region" in details tree). If you look at the patch antenna or microstrip examples in the Examples Guide you will see the PEC faces are infinitely thin, touching the dielectric, but the dielectric is an enclosed volume.
  38. 1 like
    Yes, Feko can certainly model that.
  39. 1 like
    HI, You can request for the free Altair student edition here - https://studentedition.altair.com/
  40. 1 like
    Dear Fady, is there anywhere I can find some documentation for vssBlockOutputUpdateFlag and similar functions? I still not get the flag paramater meaning. Thanks in advance Orestes
  41. 1 like
    Some lines to give you some ideas: foreach eid $elemList { set config [ hm_getentityvalue elements $eid "config" 0 ] #----- RBE2 ----------- if { $config == 55 } { set independentNode [hm_getentityvalue elements $eid independentnode.id 0] set elemNodes [hm_nodelist $eid] eval *createmark nodes 1 $elemNodes eval *createmark nodes 2 $independentNode *markdifference nodes 1 nodes 2 set elemNodes [hm_getmark nodes 1] puts "RBE2: $eid = $independentNode \[ $elemNodes \]" set new_eID [proc_to_make_RBE3 $independentNode $elemNodes] } }
  42. 1 like
    AbaqusのODBファイルを読み込んだ際の、「S-Stress components(s)」と「S-Global-Stress components(t)」の違いをご説明します。 一番の大きな違いは、データのタイプが異なります。  「S-Stress components(s)」は、(s)となっていますので、スカラー値です。=そのまま座標変換は出来ない。  「S-Global-Stress components(t)」は、(t)となっていますので、テンソル結果です。=座標変換が可能。 また、参照座標系や、結果をどこで演算したのかなどが異なります。  「S-Stress components(s)」は、Abaqusが計算時の参照座標系を基にODBファイルに直接ミーゼスや主応力の結果を書き込んだ結果をそのまま表示しています。  「S-Global-Stress components(t)」は、「XX,YY,ZZ,XY,YZ,ZX」の6成分を基に全体座標系に座標変換を行い、HyperViewの中でミーゼスや主応力を計算した結果を表示しています。
  43. 1 like
    HyperMeshやHyperView向けのTcl/Tkスクリプト内で、エラーが発生しても無視して継続させたい場合は、一般的なTclコマンドの『catch』を用います。 例えば、foreachなどを使って、各コンポーネント名から、アンダーバー(_)以降を消した名前に変更したい場合、「コンポーネントによってはアンダーバーが付いていない名前もあるかも知れない」ということがあるかも知れません。 実際、HyperMeshでは*renamecollectorコマンドを使って、「旧名」「新名」を引数で指定しますが、旧と新が同じ名前の場合、エラーとなりそこで処理が中断します。 何か処理に不安があるのであれば、『catch』を追加しておくと、エラーを無視して継続をさせることができます。 -例- foreach compname [hm_entitylist comps name] { set newname [lindex [split $compname _] 0]; catch {*renamecollector components $compname $newname}; }    図.実行前                                     図.実行後
  44. 1 like
    I'm using an element set for the contour and the query as well. It works fine so far. But for large systems it takes a lot of time. That's why I was thinking if there was another way to access the results without needing to create the contour in the first place. Thanks
  45. 1 like
    Complex Plotで表示している内容をX, Y-Mag, Y-Phaseでエクスポートしたい場合、残念ながらFile⇒Export⇒Curvesには良いフォーマットがないので、スクリプトを作成しました。 ※標準機能では、FormatでExcelなどを選ぶと、X, Y-Mag, Y-Phaseが出力できますが、データが縦に並ぶことや、カーブ名が消えてしまいますx_x プルダウンメニューFile⇒Run⇒Tcl/Tk Scriptから、下記のTclファイルを実行して下さい。 HG_datalist_Complex2.tcl
  46. 1 like
    To get the forces (for specific nodes) , can you try using the FBD tool in HV. You would need to have an op2 result file to visualize the same. You could even punch out the forces as a csv file. I am attaching the relevant document FYR. Free Body Diagrams (FBD).pdf
  47. 1 like
    hw.exeをTclスクリプトから終了させるには、「hwISession Close」を用います。 -使用例- hwi OpenStack hwi GetSessionHandle sess sess Close
  48. 1 like
    Are you requesting me to write all codes for you? tkcon is in one of three tabs, and tkcon moves to original position after closing proc SwitchTkCon {WID} {.tkcon configure -use $WID} toplevel .tinh wm transient .tinh . set nb [::ttk::notebook .tinh.nb] pack $nb -fill both -expand 1 foreach i {1 2 3} {$nb add [frame $nb.f$i] -text tab$i} set fTkConContainer [frame $nb.f1.fTkCon -container 1] pack $fTkConContainer -fill both -expand 1 wm protocol .tinh WM_DELETE_WINDOW {SwitchTkCon $::tkcon::PRIV(frameId); update; destroy .tinh} SwitchTkCon [winfo id $fTkConContainer]
  49. 1 like
    You embeded tkcon into a widget which is embeded in tab ==> so you already embeded tkcon in tab Please avoid it. before close your GUI, invoke SwitchTkCon again to move tkcon to original position
  50. 1 like
    Hey to everyone who have the same problem: I finally managed to get the MNF files working with adams. The best way is to do it like in the the Tutorial "OS-1930: Generating a Flexible Body for use in Altair MotionSolve". To find it go to the Hyperworks help home window an enter OS-1930 into the search window. There is an example about an already meshed control arm. With the Eigrl card it doesn't work, so you have to use the CMSMETH Card like described in the tutorial. The Only thing you have to change is in the OUTPUT card you have to choose ADAMSMNF instead of H3D. The other thing is to change the "Global Output Request" Control Card form STRESS to GPSTRESS so that the reduced system will claculate stress and strain. In the CMSMETH Card you can chose your reduction method and how many modes you like to transfer. Notice that only the craig-bampton and craig-chang method can be used for creating mnf files. For more background information you can get some information about the "component mode synthesis". After these informations you will understand how an MNF file is build up.
×
×
  • Create New...