Posted on December 11, 2017, by: Ziad MelhemBefore joining PADT last July, I have worked on FEA and CFD analyses but my exposure to ANSYS was limited and I was concerned about the transition. To my delight, the software was very easy to learn; most often than not intuitive and self-explanatory (e.g. mechanical wizard), the setup time was minimized after learning couple simple features (e.g. named selection, object generator etc.) and the resources on the ANSYS portal were very instrumental in the learning process. Furthermore, the colleagues at PADT proved to be very knowledgeable and experienced and more importantly responsive and eager to jump for help. One of the most attractive features that caught my attention was the streamline of the Multiphysics nature that ANSYS has. I have been satisfied with the performance of standalone CFD packages in the past, and same goes for structural ones. But never have I dealt with an extensive software that maintained the quality of a specialized one. The importance of this attribute is showing more and more its powers in recent years given the development of new convoluted products of Multiphysics nature e.g. medical applications. Using ANSYS to simulate medical applications is one of the most rewarding experience I personally enjoy. Even though, it is definitely satisfying to be able to help accelerate innovation in the aerospace, automotive, and a myriad of other industrial areas…the experience in the medical area has a more refreshing taste, probably due to the clear and direct link to human lives. From intravascular procedures to shoulder implants and microdevices, there is one common factor: ANSYS is decreasing the risks of catastrophic failures, improving the product capabilities and shortening the innovation cycle. Editors Note: Ziad is part of PADT's team in Southern California. He is a graduate of USC and has worked at Boeing, Meggit, and Pacific Consolidated Industries before joining PADT. He works with the rest of our ANSYS technical staff to make sure our users are getting the most from their ANSYS investment.
Posted on December 5, 2017, by: Ted HarrisI’m sure most people don’t know the name George M. Low. He was an early employee at NASA, serving as Chief of Manned Space Flight and later as a leader in NASA’a Apollo moon program in the late 1960’s. In fact, he was named Manager of the Apollo Spacecraft Program after the deadly Apollo 1 fire in 1967, and helped the program move forward to the successful moon landings starting in 1969. As most alumni of Rensselaer Polytechnic Institute know, he returned to Rensselaer, his alma mater, serving as president from 1976 until his death in the 1980’s. I still recall the rousing speech he gave to us incoming freshman at the Troy Music Hall on a hot September afternoon. On our class rings is his quote, “Without risk there can be no progress.” I’ve pondered that quote many times in the years since. It’s easy to coast along in many facets of life and accept and even embrace the status quo. Over the years, though, I have observed that George Low was right, and the truth is that risk is required to move forward and improve. The hard part is determining the level of risk that is appropriate, but it’s a sure bet that by not taking any risk, we will lag behind. How is that realization applicable to our world of engineering simulation? Surely those already doing simulation have moved from the old process of design > test > break > redesign > test > produce to embrace the faster and more efficient simulate > test > product, right? Perhaps, but even if they have, that doesn’t mean there can’t be progress with some additional risk. Let’s look at a couple of examples in the simulation world where some risk taking can have significant payoffs. First, transitioning from ANSYS Mechanical APDL to ANSYS Mechanical (Workbench). Most have already made the switch. I’ll allow there are still some applications that can be completely scripted within the old Mechanical Ansys Parametric Design Language in an incredibly efficient manner. However, if you are dealing with geometry that’s even remotely complex, I’ll wager that your simulation preparation time will be much faster using the improved CAD import and geometry manipulation capabilities within the ANSYS Workbench Mechanical workflow. Let alone meshing. Meshing is lights out faster, more robust, and better quality in modern versions of Mechanical than anything we can do in the older Mechanical APDL mesher. Second, using ANSYS SpaceClaim to clean up, modify, create, and otherwise manipulate geometry. It doesn’t matter what the source of the geometry is, SpaceClaim is an incredible tool for quickly making it useable for simulation as well as lots of other purposes. I recently used the SpaceClaim tools within ANSYS Discovery live to combine assemblies from Inventor and SolidWorks into one model, seamlessly, and was able to move, rotate, orient, and modify the geometry to what I needed in a matter of minutes (see the Discovery Live image at the bottom). The cleanup tools are amazing as well. Third, looking into ANSYS Discovery Live. Most of us can benefit from quick feedback on design ideas and changes. The new Discovery Live tool makes that a reality. Currently, in a technology demonstration mode, it’s free to download and try it from ANSYS, Inc. through early 2018. I’m utterly amazed by how fast it can read in a complex assembly and start generating results for basic structural, CFD, and thermal simulations. What used to take weeks or months can now be done in a few minutes. Credits: Motorcycle geometry downloaded from GrabCAD, model by Shashikant Soren. Human figure geometry downloaded from GrabCAD, model by Jari Ikonen. Models combined and manipulated within ANSYS Discovery Live. George M. Low image from www.nasa.gov. I encourage you to take some risks for the sake of progress.
Posted on November 27, 2017, by: Alex GrishinOne of the key outputs from any random vibration analysis is determining the response of the object you are analyzing in terms of reaction forces. In the presentation below. Alex Grishin shares the theory behind getting accurate forces and then how to do so in ANSYS Mechanical. PADT-ANSYS-Random-Vib-Reaction-Forces-2017_11_22-1
As always, please contact PADT for your ANSYS simulation, training, and customization needs.
Posted on November 7, 2017, by: Michael GriesiIEEE Day celebrates the first time in history when engineers worldwide and IEEE members gathered to share their technical ideas in 1884. Events were held around the world by 846 IEEE Chapters this year. So, to celebrate, I attended a joint chapter meeting in at The Museum of Flight in Seattle with technical presentations focused on “Smart Antennas for IoT and 5G”. There were approximately 60 in attendance, so assuming this was the average attendance globally results in over 50,000 engineers celebrating IEEE Day worldwide! The Seattle seminar featured three speakers that spanned theory, design, test, integration, and application of smart antennas. There was much discussion about the complexity and challenges of meeting the ambitious goals of 5G, which extend beyond mobile broadband data access. Some key objectives of 5G are to increase capacity, increase data rates, reduce latency, increase availability, and improve spectral and energy efficiency by 2020. A critical technology behind achieving these goals is beamforming antenna arrays, which were at the forefront of each presentation. Anil Kumar from Boeing focused on the application of mmWave technology on aircraft. Test data was used to analyze EM radiation leakage through coated and uncoated aircraft windows. However, since existing regulations don’t consider the increased path loss associated with such high frequencies, the integration of 5G wireless applications may be restricted or delayed. Beyond this regulatory challenge, Anil discussed how multipath reflectors and absorbers will present significant challenges to successful integration inside the cabin. Although testing is always required for validation, designing the layout of the onboard transceivers may be impractical to optimize without an asymptotic EM simulation tool that can account for creeping waves, diffraction, and multi-bounce. Considering the test and measurement perspective, Jari Vikstedt from ETS-Lindgren focused on the challenges of testing smart antenna systems. Smart or adaptive antenna systems will not likely perform the same in an anechoic chamber test as they would in real systems. Of particular difficulty, radiation null placement is just as critical as beam placement. This poses a difficult challenge to the number and location of probes in a test environment. Not only would a large number of probes become impractical, there is significant shadowing at mmWave frequencies which can negatively impact the measurement. Furthermore, compact ranges can significantly impact testing and line of sight measurements become particularly challenging. While not a purely test-oriented observation, this lead to considering the challenge of tower hand off. If a handset and tower use beamforming to maintain a link, if is difficult for an approaching tower to even sense the handset to negotiate the hand-off. In contrast, if the handset was continuously scanning, the approaching tower could be sensed to negotiate the hand-off before the link is jeopardized. The keynote speaker, who also traveled from Phoenix to Seattle, was ASU Professor Dr. Constantine Balanis. Dr. Balanis opened his presentation by making a distinction between conventional “dumb antennas” and “smart antennas”. In reality, there are no smart antennas, but instead smart antenna systems. This is a critical point from an engineering perspective since it highlights the complexity and challenge of designing modern communication systems. The focus of his presentation was using an adaptive system to steer null points in addition to the beam in an antenna array using a least mean square (LMS) algorithm. He began with a simple linear patch array with fixed uniform amplitude weights, since an analytic solution was practical and could be used to validate a simulation setup. However, once the simulation results were verified for confidence, designing a more complex array with weighted amplitudes accompanying the element phase shift was only practical through simulation. While beam steering will create a device centric system by targeting individual users on massive multiple input multiple output (MIMO) networks, null steering can improve efficiency by minimizing interference to other devices. Whether spatial processing is truly the “last frontier in the battle for cellular system capacity”, 5G technology will most certainly usher in a new era of high capacity, high speed, efficient, and ubiquitous means of communication. If you would like to learn more about how PADT approaches antenna simulation, you can read about it here and contact us directly at email@example.com.
Posted on October 23, 2017, by: Michael GriesiANSYS HFSS features an integrated “history-based modeler”. This means that an object’s final shape is dependent on each and every operation performed on that object. History-based modelers are a perfect choice for analysis since they naturally support parameterization for design exploration and optimization. However, editing imported solid 3D Mechanical CAD (or MCAD) models can sometimes be challenging with a history-based modeler since there are no imported parameters, the order of operation is important, and operational dependencies can sometimes lead to logic errors. Conversely, direct modelers are not bound by previous operations which can offer more freedom to edit geometry in any order without historic logic errors. This makes direct modelers a popular choice for CAD software but, since dependencies are not maintained, they are not typically the natural choice for parametric analysis. If only there was a way to leverage the best of both worlds… Well, with ANSYS, there is a way. As discussed in a previous blog post, since the release of ANSYS 18.1, ANSYS SpaceClaim Direct Modeler (SCDM) and the MCAD translator used to import geometry from third-party CAD tools are now packaged together. The post also covered a few simple procedures to import and prepare a solid model for electromagnetic analysis. However, this blog post will demonstrate how to define parameters in SCDM, directly link the model in SCDM to HFSS, and drive a parametric sweep from HFSS. This link unites the geometric flexibility of a direct modeler to the parametric flexibility of a history-based modeler. You can download a copy of this model here to follow along. If you need access to SCDM, you can contact us at firstname.lastname@example.org. It’s also worth noting that the processes discussed throughout this article work the same for HFSS-IE, Q3D, and Maxwell designs as well.  To begin, open ANSYS SpaceClaim and select File > Open to import the step file.  Split the patch antenna and reference plane from the dielectric. Click here for steps to splitting geometry. Notice the objects can be renamed and colors can be changed under the Display tab.
- Create local variables in HFSS that can be used for both local and linked geometry. For example, create a variable in HFSS for traceWidth = 3mm (which was the previously noted width). Define SCDM_traceWidth = (traceWidth-3mm)/2. Now the port width can scale with the trace width.
- Link to multiple SCDM projects. Either move and rotate parts as needed or create a separate coordinate system for each component. For example, link an SMA end connector to the same HFSS project to analyze both components. Notice that each component has variables and the substrate thickness changes both SCDM projects.
- Design other objects in the native HFSS history-based modeler that are dependent on the SCDM design variables. For example, the void in an enclosure could be a function of SCDM_dielectricHeight. Notice that the enclosure void is dependent on the SCDM dielectric height.
Posted on October 5, 2017, by: Alex GrishinANSYS Mechanical is great at applying tabular loads that vary with an independent variable. Say time or Z. What if you want a tabular load that varies in multiple directions and time. In part one of this series, I covered who you can use the External Data tool to do just that. In this second part, I show how you can alternatively use APDL commands added to your ANSYS Mechanical model to define the tabular loading. PADT-ANSYS-Tabular-loads-2
Posted on September 12, 2017, by: Nicolas Jobert
Guest BloggerWe are pleased to publish this very useful post from Nicolas Jobert from Synchrotron SOLEIL in France. Nicolas is a Mechanical Engineer with more than 20 years of experience using ANSYS for engineering design and analysis in academia and industry. He currently is Senior Mechanical Engineer at Synchrotron SOLEIL, the French synchrotron radiation facility. He also teaches various courses on Design and Validation in the field of structural and optomechanics. He graduated from the Ecole Centrale Marseille, France, and is a EUSPEN member.
As Time Goes ByDo you remember the moment you first heard about ANSYS introducing APDL Math? I, for one, do, and I have a vivid memory of thinking “Wow, that can be a powerful tool, I’m dead sure it won’t be long before an opportunity arises and I’ll start developing pretty useful procedures and tools”. Well, that was half a decade ago, and to my great shame, nothing quite like that has happened so far. Reasons for this are obvious and probably the same for most of us: lack of time and energy to learn yet another set of commands, fear of the ever present risk of developing procedures that are eventually rejected as nonstandard use of the software and therefore error-prone (those of you working under quality assurance, raise your hand!), anxiety of working directly under the hood on real projects with little means to double check your results, to name a few. That said, finally an opportunity presented itself, and before I knew it, I was up and running with APDL Math. The objective of this article is to showcase some simple yet insightful applications and hopefully remove the prevention one can have regarding using these additional capabilities. For the sake of demonstration, I will begin with a somewhat uncommon analysis tool that should nevertheless ring a bell for most of you, that is: modal analysis (and yes, the pun is intended). You may wonder what is the purpose of using APDL Math to perform a task that is a standard ANSYS capability since say revision 2.0, 40 years ago? But wait, did I mention that by modal analysis, I mean thermal modal Analysis?
Thermal Modal Analysis at a GlanceAlthough scarcely used, thermal modal analysis is both an analysis and a design validation tool, mostly used in the field of precision engineering and or optomechanics. Specifically, it can serve a number of purposes such as:
Q: Will my system settle fast enough to fulfill design requirements? A: Compute the system Thermal Time Constants
Q: Where should I place sensors to get information rich / robust measurements? A: Compute Thermal modes and place your sensors away for large thermal gradients
Q: Can I develop a reduced model to solve large transient thermal mechanical problems? A: Modal basis allows for the construction of such reduced problem effectively converting a high-order coupled system to a low order, uncoupled set of equations.
Q: How to develop a reduced order state-space matrices representation of my thermal system (equivalent to SPMWRITE command)? A: Modal analysis provides every result needed to build those matrices directly within ANSYS.Although you might only be vaguely familiar with many or all of those topics, the idea behind this article is really to show that APDL Math does exactly what you need it to do: allow the user to efficiently address specific needs, with a minimal amount of additional work. Minimal? Let’s see what it looks like in reality, and you will soon enough be in a position to make your own opinion on the matter.
Thermal Modal Analysis using APDL MathTo begin with, it is worth underlining the similarities and differences between structural (vibration) modes and thermal modes. Mathematically, both look very much the same, i.e. modes are solutions of the dynamics equation in the absence of forcing (external) term:
[K] is the stiffness matrix [M] is the mass matrix
[K] is the conductivity matrix [C] is the capacitance matrix
- Structural : λ=ω², i.e. the square of a circular frequency
- Thermal: λ=1/τ, i.e. the inverse of time constant
! Setup Model … ! Make ONE dummy transient solve ! to write stiffness and mass ! matrices to .FULL file /SOLU ANTYPE,TRANSIENT TIME,1 WRFULL,1 SOLVE ! Get Stiffness and Mass *SMAT,MatK,D,IMPORT,FULL,,STIFF *SMAT,MatM,D,IMPORT,FULL,,MASS ! Eigenvalue Extraction Antype,MODAL modopt,Lanb,NRM,0,Fmax *EIGEN,MatK,MatM,,EiV,MatPhi ! No need to convert eigenvalues ! to frequencies, ANSYS does ! it automatically ! Done !
! Setup Model … ! Make TWO dummy transient solve ! to separately write conductivity ! and capacitances matrices to .FULL file /SOLU ANTYPE,TRANSIENT TIME,1 NSUB,1,1,1 TINTP,,,,1 WRFULL,1 ! Zero out capacitance terms … SOLVE ! Get Conductivity Matrice *SMAT,MatK,D,IMPORT,FULL, Jobname.full,STIFF ! Restore capacitance and zero out ! conductivity terms … SOLVE ! Get Capacitance Matrice *SMAT,MatC,D,IMPORT,FULL,,STIFF ! Eigenvalue Extraction Antype,MODAL modopt,Lanb,NRM,,0,1/(2*PI*SQRT(Tmin)) *EIGEN,MatK,MatC,,EiV,MatPhi ! Convert Eigenvalues for Frequency ! to Thermal time Constants ! *do,i,1,EiV_rowDim Eiv(i)=1/(2*PI*Eiv(i))**2 *enddo ! Done !
- Results from the eigenvalues are stored in a vector (EiV) and a matrix (MatPhi), which need not be declared but are created when executing the *EIGEN command (no *DIM required).
- For each APDL Math entity, ANSYS automatically maintains variables named Param_rowDim and Param_colDim, hence removing the burden to keep track of dimensions.
But where on Earth is my eye candy?Now that we have some procedure and results, we would like to be able to show this to the outside world (and to be honest, some graphical results would also help getting confidence in results). The additional task to do so is really minimal. What we need to do is simply to put back those numerical results into the ANSYS database so that we can use all the conventional post-processing capabilities. This can be made using the appropriate POST1 commands, essentially: DNSOL. And, while we are at it, why not do a hardcopy to an image file? Here is the corresponding input.
… User should place all nodes with non-prescribed temperatures in a component named MyNodeComponent … First, convert Eigenvectors from solver to BCS ordering ! Conversion needed *SMAT,Nod2Bcs,D,IMPORT,FULL,Jobname.full,NOD2BCS *MULT,Nod2Bcs,TRAN,MatPhi,,MatPhi ! Then, read in mapping vector to convert to user ordering *VEC,MapForward,I,IMPORT,FULL,Jobname.full,FORWARD ! Put the results in ANSYS database /POST1 *do,ind_mode,1,NRM cmsel,s,MyNodeComponent curr_node=0 *do,i,1,ndinqr(0,13) curr_node=ndnext(curr_node) curr_temp=MatPhi(MapForward(curr_node),ind_mode) dnsol,curr_node,TEMP,,curr_temp *enddo Tau=1/(2*3.14*EiV(ind_mode))**2 To=NINT(Tau*10)/10 ! compress to 1 digit after comma /title,Mode #%ind_mode% - Tau=%To%s plnsol,temp ! Hardcopy to BMP file /image,SAVE,JobName_Mode%ind_mode%,bmp *enddoThis way, modes can be displayed, or even written to a conventional .RTH file (using RAPPND), and used as any regular ANSYS solver result.
Nice, but an actual example wouldn’t hurt, would it?Now you may wonder what the results look like in reality. To remain within the field of precision engineering, let’s use a support structure typically designed for high-stability positioning. From a structural point of view, it must have a high dynamic stiffness and a low total mass so that a Delta shaped bracket is appropriate. Since we want the system to rapidly evacuate any heat load, we choose aluminum as candidate material. We do know from first principles that any applied disturbance will exponentially vanish and the system will go back to equilibrium state. Now, what will be the time constants of this decay? For the sake of simplicity we restrict the analysis to a highly simplified, 2D model of such a support. PLANE55 elements are used to model the structural part while the heat sink is accounted for using SURF151. Boundary conditions are enforced using an extra node. After applying boundary conditions, we execute the modal solution to obtain say - the first 8 modes.
|Index||Time Constant [s]||Comment|
|1||535.9||Quasi-uniform temperature field (i.e. “rigid body” mode)|
|2||32.1||1st order (one wavelength along perimeter)|
|3||23.8||1st order (one wavelength along perimeter)|
|4||8.1||2nd order (two wavelengths along perimeter)|
|5||6.8||2nd order (two wavelengths along perimeter)|
|6||3.5||3rd order (three wavelengths along perimeter)|
|7||3.1||3rd order (three wavelengths along perimeter)|
|8||2.2||4th order (four wavelengths along perimeter)|
Allocate a  Vector : EIV Allocate a  Dense Matrix : MatPhiPlease note that the solution has 227 DOFs whereas the entire problem has 228 DOFs. This is the consequence of having introduced the boundary conditions as an enforced temperature on a node, which DOF is therefore removed from the DOF set to be obtained by the solver. Also, we might want to use the modal shapes information to decide which locations are best suited to capture the entire temperature field on the structure. Without knowledge of the excitation source, one straightforward way to do so is to retain for each mode the node that has the largest amplitude. This is made even easier in this situation, since we have normalized each mode to have unit maximum amplitude we just need to select nodes having modal amplitude equal to 1 (or -1). On the figure below, each temperature sensor location is marked with a ‘TSm’ label where m is the mode index. Doing so, we reach a pretty satisfactory distribution for the sensors locations, completely consistent with intuition. In numerical terms, we can also check that the modal matrix [Φ]_sensors, i.e. the original full matrix restricted to the selected DOF, has an excellent condition number. But there are many other things we could do starting from this. For example, with additional information, such as the location and the frequency content of the temperature fluctuations, one could further restrict the set of needed temperature sensors by running a dummy transient analysis and choosing locations where the correlation between sensors readings is as low as possible (using *MOPER,,,CORR). Even better, one can estimate the thermally induced displacements and select locations best suited to build an empirical model (typically using AR or ARMA), allowing one to predict structural displacements induced by temperature fluctuations using just a couple of sensors. This in turn can be used to select control strategies, check modal controllability… all within ANSYS.
ConclusionAPDL Math was presented as an alternate route for users who need to include specialized steps in an otherwise standard FE process, and in my opinion it does just that. The benefits can be immense and the learning curve is steep but short. As long as the user knows what he/she is doing, there is little possibility to get lost: after all, APDL Math only comprises 18 additional commands. What hindered me so far was the necessity to account for internal, BCS and user ordering, but it really is not a big deal, as seen from the above example. What is more, the possibility to store the created results in the Mechanical APDL database (DNSOL and RAPPND are your friends!) provides every means to control your results and finally to build confidence in your developments. And for those of us who prefer to stay within Workbench environment, there is nothing preventing from including APDL Math procedures into Workbench command snippets. This was just an introductory example, since many other applications could be found, to name a few just in the fields of precision engineering and/or opto-mechanics:
- Speed up transient thermal mechanical analyses
- Perform harmonic analysis of thermal models
- Virtual testing of physical setup, including real-time control systems (model based)
- Modal testing, error localization, automated model updating
Posted on September 11, 2017, by: Ted HarrisBy now you’ve probably heard that ANSYS versions 18.0, 18.1, and 18.2 have all been released in 2017. While 18.0 was the ‘point’ release in January, it should be noted that 18.1 and 18.2 are not ‘patches’ or service packs, but are full releases each with significant enhancements to the code. We’ll present some significant and useful enhancements for each.
18.0Number 1: First and foremost – info on the new features is more readily accessible with the Mechanical Highlights list. The first time you launch Mechanical, you’ll see a hyperlinked list of new release highlights. One you actually do something in Mechanical, though, that list goes away. There is a simple way to get it back: Click on the Project branch in the Mechanical tree, then click on the Worksheet button in the menu near the top of the window. Clicking on the hyperlinks in the list or simply scrolling down gives us more information on each of the listed enhancements. Keep in mind the list is only highlights and by no means has all of the new features listed. A more detailed list can be found in the ANSYS Help, in the Release Notes. Number 2: A major new feature that became available in 18.0 is Topology Optimization. We’ve written more about Topology Optimization here Number 3: Another really useful enhancement in 18.0 is the ability to define a beam connection as a pretensioned bolt. This means we no longer need to have a geometry representation of a bolt if we want a simpler model. We can simply insert a beam connection between the two sides of the bolted geometry, and define the pretension on that resulting beam. Beam connections are inserted in the Connections branch in Mechanical. Once the beam is fully defined, it can have a bolt pretension load applied to it, just like as if the beam geometry was defined as a solid or beam in your geometry tool. Here you can see a beam connection used for bolt pretension on the left, with a traditional geometric representation of a pretensioned bolt on the right:
18.1Number 4: A very nice capability added in version 18.1 is drag and drop contact regions for contact sizing in the Mesh branch. Contact elements work best when the element sizes on both sizes of the interface are similar, especially for nonlinear contact. ANSYS Mechanical has had Contact Sizing available as a mesh control for a long time. Contact Sizing allows us to specify an element size or relevance level once, for both sides of one or more contact regions. What’s new in 18.1 is the ability to drag and drop selected contacts from the Connections branch into the Mesh branch. Just select the desired contact regions with the mouse, then drag that selection into the Mesh branch. Then specify the desired mesh sizing controls for contact. This is what the dragging and dropping looks like: After dropping into the Mesh branch, we can specify the element size for the contact regions: This shows the effect of the contact sizing specification on the mesh:
18.2Number 5: An awesome new feature in 18.2 is element face selection, and what you can do with it. There is a new selection filter just for element face selection, shown here in the red box: Once the element face select button is clicked, element faces can be individually selected, box selected, or paint selected simply by holding down the left mouse button and dragging. The green element faces on the near side have been selected this way: The selected faces can then be converted to a Named Selection, or items such as results plots can be scoped to the face selection: Number 6: Finally, to finish up, some new hotkeys were added in 18.2. Two really handy ones are:
- Z = zoom fit or zoom to the current selection of entities
- <Ctrl> K = activate element face selection
- F11 = make the graphics window full screen!
- Click F11 again to toggle back to normal size
Posted on September 8, 2017, by: Eric MillerYesterday ANSYS, Inc. did a webinar about a technology that was going to "Change the way simulation is done." If you have been around the world of FEA and CFD for the 30+ years I have you have heard that statement before. And rarely does the actual product change match the hype. Not true for ANSYS Discovery Live. If anything, I think they are holding back. This is disruptive, this is a tool that will change how people do simulation. In this post I'll share my thoughts on what it is and why I think it is so transformative, and then in the second half (go ahead, if you don't want to listen to me go on and on about how much I like this tool, skip ahead) there are some tips on how to get your hands on it to see for yourself.
What is ANSYS Discovery Live?ANSYS Discovery Live is a new multiple physics simulation platform that combines several key ingredients to produce a software tool that engineers can use to do almost instantaneous virtual prototypes of the behavior of their designs directly from their solid models. The developers at ANSYS, Inc. have combined their knowledge of advanced solver technology, making solvers parallel for Graphical Processor Units (GPUs, high-end graphics cards), direct solid modeling (SpaceClaim), and some advanced stuff on the discretization side I don't think I can talk about. All of those things embedded inside SpaceClaim make ANSYS Discovery Live. Once you have a solid model in the tool, you simply define what physics you want to solve and some boundary conditions, then it solves. In almost real time. Right there in front of you. The equivalent steps of meshing, building the model, solving it, extracting results, and displaying the results are done automatically. It may iterate a few times to converge on a solution, but in a few seconds, you will have a good enough answer to give you insight into your design. And that is the key point. This is not a replacement for ANSYS Mechanical, FLUENT, or HFSS. It is a tool for exploring your designs and gaining insight into their behavior. It allows the design engineer, with very little training or expertise, to exercise their design and see what happens. The product lives inside ANSYS Spaceclaim and can be installed on its own. It runs on Windows and requires a NVidia graphics card with a newer GPU (see below for more on that). Right now the product is in pre-release mode and anyone, yes anyone, can go to www.ansys.com/discovery and download it and try it out. And please, share your feedback. Expect the product to be released in the first quarter of 2018. Pricing and bundling have not been firmed up yet, but from what we have seen the plans are reasonable and make sense.
Why is it Unique in the Industry?Some of the first comments I saw on social media about ANSYS Discovery Live after the webinar were that it is not a unique tool. There are other GPU based solvers out there. That is true. But even though those tools are super fast at solving, they have not been widely adopted. The ANSYS product is unique because it: 1) combines GPU based solvers for multiple physics and 2) is built into a fully functioning solid modeling tools. A third might be that it is also an ANSYS product, which means it will be backed technically and supported well.
Why I think that the Simple Fact that it Exists is Important?During an interview for a magazine article about innovation in product development this week I was asked what is keeping innovation from happening more often. My answer was that most companies with the resources, both money and people, to innovate are choosing to acquire rather than innovate internally. They let others raise money, take all the risk, work out all the problems, deal with all the issues of trying to make something new. And then when they succeed, they buy them. There is nothing morally wrong with that approach, it is just inefficient and inaccurate. Every innovation has to not only survive its technical challenges, it has to survive being a startup. What ANSYS, Inc. has done is the opposite. They could have purchased a GPU based solver startup and checked the box. But instead, they took people from different business units, several that were acquired, and put them together and said: "innovate... but make it something very useful." And they did. The fact that they executed on the logistics of a new product that used new and old technology across physics and across software development realms, is fantastic. It makes me feel good about ANSYS, Inc's true dedication to improving their products.
How will it Change Simulation?In my career, I have had the same conversation dozens of times "Let me go out to the lab and tinker with it, I'll figure out what is going on." That is the way you had to explore your product to get a "feel" for what is going on. Simulation took too long and you became so wrapped up in the process of building and running a model that you could not really explore the behavior of your product. Now we can. ANSYS Discovery Live is called Discovery Live not because anyone at ANSYS is a marketing genius (sorry guys...) but because that is what it lets you do. Discover the behavior of your product live. You simply play with it and see what happens. And this will change simulation because we know can move from verification or optimization to simply experimenting and gaining a deeper understanding, early in the design process. We will still do what is now I guess called traditional simulation. We will need more accuracy, more complex physics, loads, and behavior. But early on we can learn so much by virtually experimenting.
Is it the Perfect Tool Right out of the Box?This is not a perfect-does-everything tool. First off, it is a pre-release. The basic functionality to make it useful is there. More than I thought would be available in a first release. But there are limitations because it is new, or because of the approach. It is not as accurate as more traditional approaches. The way it works takes some shortcuts on geometry and can't include some behaviors. This should improve over time but it will never be accurate as more time-consuming approaches that simply have more functionality. Over the next two to three years we will see it mature and add functionality and accuracy. The GPU's the tool depends on will offer more performance for less money as well. This is a journey, but right now everyone I have talked to who has actually played with the pre-release is very happy with the functionality and accuracy that is there now. Because it is sufficient to do the experimentation and exploration it was designed to allow.
- To get started, go to www.ansys.com/discovery and look around. The videos are awesome! When you are ready to try it out, click on Download Now. Fill out the form. Don't complain. Yes you will get a few emails and a salesperson (gasp!) may call you. It's worth some emails and maybe a phone call.
- Set yourself up there. There is a verification code step and once you put that in and create your login, you have to click on some legal agreements, including export controls. Save your login info, you will need it to get back in.
- After that either start the download or the Cloud Trial Option. The cloud trial didn't work for me, read below how I got to that function.
- If you chose download it will download a big Zip File, over 1 GB. It is a full solid modeler and CFD/Structural/Thermal solver... so it is big.
- Once it is there, unzip, and run Setup.exe. follow the steps and you will be there.
- If you don't have a graphics card that will run this, then use the cloud demo. Like I said above, the button didn't work for me. If you have that problem or you want to use it after your first login, go to:
- https://discoveryforum.ansys.com/ and click on "Getting Started."
- Scroll down a bit and find the "Cloud Trial" post. That one takes you to the page where you can find a server near you to try things out on. It's pretty slick.
- If you need to get back here, use https://discoveryforum.ansys.com/ and log in with the email and password you gave at registration,
Hardware RequirementsThe only sticky bit about this whole thing is that it run a subset of Nvidia graphics cards. So you have to have one of those cards. According to the information in the forum:
ANSYS Discovery Live relies on the latest GPU technology to provide its computation and visual experience. To run the software, you will require:
- A dedicated NVIDIA GPU card based on the Kepler, Maxwell or Pascal architecture. Most dedicated NVIDIA GPU cards produced in 2013 or later will be based on one of these architectures. - At least 4GB of video RAM (8GB preferred) on the GPU.
Also, please ensure you have the latest driver for your graphics card, available from NVIDIA Driver Downloads. You can also refer to the post on Graphics Performance Benchmarks. Performance of Discovery Live is less dependent on machine CPU and RAM. A recent generation 64-bit CPU running Windows, and at least 4GB of RAM will be sufficient. If you do not have a graphics card that meets these specifications, the software will not run. However, you can try ANSYS Discovery Live through an online cloud-based trial, which requires only an internet browser and a reasonably fast internet connection.I didn't know if my GPU on my laptop would work, so I went to https://www.techpowerup.com and put in my card model (nvidia m500m) and it told me it was Maxwell technology.
Go Forth and Discover, and ShareDon't hesitate, download this and try it out. Even if you are a high-end combustion simulation expert that will never need it, if you are interested in Simulation you should still try it out. Use the forum to share your thoughts and questions. The gallery is already filling up with some fantastic real world examples.
Posted on August 24, 2017, by: Joe WoodwardSo this is just a quick post to point out a handy feature in ANSYS Workbench, the ACT Console. There are times when you want some functionality in Mechanical that just is not yet there. In this example, a customer wanted the ability to get a text list of all the Named Selections in his model. A quick Python script does just that. So to use a piece of Python code, like this, we use the ACT Console in Mechanical. To access the ACT Console in Mechanical 17.0, or later, just hit this icon in the toolbar. The Console allows you to type, or paste, text directly into the black command line at the bottom. But if we are going to reuse this code, then the use of Snippets is the way to go. In R17.0 they were called ‘Bookmarks’, but they worked the same way. When you add a Snippet, a new window allows you to name the snippet and type in, or paste in, your code. When you hit Apply, your named snippet is added to the list From then on, to use the snippet you just click on it, and hit ‘Enter’. The text is basically, repasted into the command window, so you can set any variables needed prior to hitting your snippet. The snippets are persistent and remain in the console, so they are available for all new projects. Using snippets is a great way to reduce time for repetitive tasks, without having to create a full blown ACT extension. Happy coding!
Posted on August 22, 2017, by: Ted HarrisHere in the Phoenix area, we weren’t treated to the full total eclipse that others in the USA got to see. Our maximum coverage of the sun was a bit over 60%. Still, there was an eclipse buzz in the PADT headquarters and although we had some rare clouds for a few minutes, the skies did part and we did get to view the partial eclipse from the parking lot. So, how did ANSYS help us view the eclipse? It was in an indirect way – via a pinhole camera I made from an old ANSYS installation software box. The software box, a hobby knife to cut out a viewing port, a couple of post-it notes to allow for a small hole and a clear projection area, and a thumb tack were all that was needed, along with a couple of minutes to modify the box. Here we can see the viewing port cut into the software box. On the opposite side is a pin hole to allow the sun’s light to enter the box. After heading out to the eclipsing grounds (the parking lot), we quickly lined up the pin hole and the projection screen and got our views of the partially obscured sun: Here is a close up of the sun’s image projected inside the box: Others viewing the eclipse here at PADT HQ had a range of filters, eclipse glasses, etc. With the projection method as shown above, though, we don’t have to worry about eye damage. So, in a way, ANSYS did help us view the eclipse safely, by providing a box that was easy to convert to a pinhole camera. While we enjoyed the partial eclipse here in Arizona, we did have a couple of PADT colleagues in the path of totality. Here is a picture from one of my coworkers who viewed the eclipse in South Carolina: We hope you enjoyed the eclipse as well, either in person or via images on the web. We’re looking forward to the next one! Finally, In case you missed an earlier astronomical rarity back in 2012, here is a photo of the planet Venus transiting in front of the sun’s disk (black dot on the left side). The next one of these won’t be until December, 2117.
Posted on August 4, 2017, by: Ted HarrisWe’ve discussed topological optimization in this space before, notably here: If you’re not familiar with topological or topology optimization, a simple description is that we are using the physics of the problem combined with the finite element computational method to decide what the optimal shape is for a given design space and set of loads and constraints. Typically our goal is to maximize stiffness while reducing weight. We may also be trying to keep maximum stress below a certain value. Frequencies can come into play as well by linking a modal analysis to a topology optimization. Why is topology optimization important? First, it produces shapes which may be more optimal than we could determine by engineering intuition coupled with trial and error. Second, with the rise of additive manufacturing, it is now much easier and more practical to produce the often complex and organic looking shapes which come out of a topological optimization. ANSYS, Inc. has really upped the game when it comes to utilizing topology optimization. Starting with version 18.0, topo opt is built in functionality within ANSYS. If you already know ANSYS Mechanical, you already know the tool that’s used. The ANSYS capability uses the proven ANSYS solvers, including HPC capability for efficient solves. Another huge plus is the fact that SpaceClaim is linked right in to the process, allowing us to much more easily make the optimized mesh shape produced by a topological optimization into a more CAD representation set for use in validation simulations, 3D printing, or traditional manufacturing. The intent of this blog is to show the current process in ANSYS version 18.1 using a simple example of an idealized motorcycle front fork bracket optimization. We don’t claim to be experts on motorcycle design, but we do want to showcase what the technology can do with a simple example. We start with a ‘blob’ or envelope for the geometry of our design space, then perform an optimization based on an assumed set of loads the system will experience. Next we convert the optimized mesh information into solid geometry using ANSYS SpaceClaim, and then perform a validation study on the optimized geometry. Here we show our starting point – an idealized motorcycle fork with a fairly large blob of geometry. The intent is to let ANSYS come up with an optimal shape for the bracket connecting the two sides of the fork. The first step of the simulation in this case is a traditional Static Structural simulation within ANSYS Workbench. The starting point for the geometry was ANSYS SpaceClaim, but the initial geometry could have come from any geometry source that ANSYS can read in, meaning most CAD systems as well as Parasolid, SAT, and STEP neutral file formats. A single set of loads can be used, or multiple load cases can be defined. That’s what we did here, to simulate various sets of loads that the fork assembly might experience during optimization. All or a portion of the load cases can be utilized in the topological optimization, and weighting factors can be used on each set of loads if needed. Here we see the workflow in the ANSYS Workbench Project Schematic: Block A is the standard static structural analysis on the original, starting geometry. This includes all load cases needed to describe the operating environment. Block B is the actual topological optimization. Block C is a validation study, performed on the optimized geometry. This step is needed to ensure that the optimized shape still meets our design intent. Within the topology optimization, we set our objective. He we choose minimizing compliance, which is a standard terminology in topology optimization and we can think of it as the inverse which is maximizing stiffness. In the static structural analysis, 7 load cases were used to describe different loading situations on the motorcycle fork, and here all have been used in the optimization. Further, we defined a response constraint, which in this example is to reduce mass (actually retain 15% of the mass): Another quantity that’s often useful to specify is a minimum member constraint. That will keep the topology optimization from making regions that are too small to 3D print or otherwise manufacture. Here we have specified a minimum member size of 0.3 inches: Since the topological optimization solution uses the same ANSYS solvers for the finite element solution as a normal solution, we can leverage high performance computing (distributed solvers, typically) to speed up the solution process. Multiple iterations are needed to converge on the topology optimization, so realize that the topo opt process is going to be more computationally expensive than a normal solution. Once the optimization is complete, we can view the shape the topo opt method has obtained: Notice that only a portion of the original model has been affected. ANSYS allows us to specify which regions of the model are to be considered for optimization, and which are to be excluded. Now that we have a shape that looks promising, we still need to perform a validation step, in which we rerun our static simulation with the loads and constraints we expect the fork assembly to experience. To do that, we really want a ‘CAD’ model of the optimized shape. The images shown above show the mesh information that results from the topo opt solution. What we need to do next is leverage the ANSYS SpaceClaim geometry tool to create a solid model from the optimized shape. A simple beauty in the ANSYS process is that with just a couple of clicks we proceed from Block B to Block C in the Workbench project schematic, and can then work with the optimized shape in SpaceClaim. As you can see in the above image, SpaceClaim automatically has the original geometry as well as the new, optimized shape. We can do as much or as little to the optimized shape as we need, from smoothing and simplification to adding manufacturing features such as holes, bosses, etc. In this case we simply shrink wrapped it as-is. Continuing with the validation step, the geometry from SpaceClaim automatically opens in the Mechanical window and we can then re-apply the needed loads and constraints and then solve to determine if the optimized shape truly meets our design objectives. If not, we can make some tweaks and run again. The above image shows a result plot from the validation step. The geometry efficiently comes through SpaceClaim from the optimization step to the validation step. The needed tools are all nicely contained within ANSYS. Hopefully this has given you an idea of what can be done with topology optimization in ANSYS as well as how it’s done. Again, if you already know ANSYS Mechanical, you already know the bulk of how to do this. If not, then perhaps what you have seen here will spark a craving to learn. We can’t wait to see what you create.
Posted on August 1, 2017, by: David MastelI recently had a chance to run a series of benchmarks on one of our latest CUBE numerical simulation workstations. I was amazed by the impressive benchmark numbers and wanted to share with you the details for the SP-5 benchmark using ANSYS 18.1. Hopefully this information will help you make the best decision the next time you need to upgrade your numerical simulation C Drive from whatever to now is the time to buy a Non-Volitile Memory Express drive. Total speedup using identical CUBE hardware, except for the INTEL DC P3700 NVMe drive @32 Cores is a 1.19x speedup!
- Time Spent Computing Solution ANSYS SP-5 Benchmark
- 161.7 seconds vs. 135.6 second
- ANSYS 17.1 & ANSYS 18.1 Benchmarks
CUBE w32iP Specifications (July 2017)
- CUBE Mid-Tower Super Quiet Chassis (900W PS)
- CPU: 32 INTEL Cores – 2 x INTEL e5-2697A V4 email@example.comGHz/3.6GHz Turbo
- OS: INTEL NVMe – 1 x 1.6TB INTEL Enterprise Class SSD
- Mid-Term Storage: – 1 x 10TB Enterprise Class SATA 6Gbp/s, 256M, Helium sealed
- RAM: 256GB DDR4-2400MHz LRDIMM RAM
- GRAPHICS: NVIDIA QUADRO P6000 (24GB GDDR5X RAM)
- MEDIA: DVD-RW/Audio 7.1 HD
- Windows 10 Professional
Just how much faster the INTEL NVME drive performs over previously run ANSYS Benchmarks?
Check out the data for yourself:
- ANSYS Benchmark Test Case Information.
- ANSYS HPC Licensing Packs required for this benchmark
- I used (2) HPC Packs to unlock all 32 cores.
- 1.19x Total Speedup!
- Please contact your local ANSYS Software Sales Representative for more information on purchasing ANSYS HPC Packs. You too may be able to speed up your solve times by unlocking additional compute power!
- What is a CUBE? For more information regarding our Numerical Simulation workstations and clusters please contact our CUBE Hardware Sales Representative at SALES@PADTINC.COM
- Designed, tested and configured within your budget. We are happy to help and to listen to your specific needs.
ANSYS SP-5 Benchmark Details
|Analysis Type||Static Nonlinear Structural|
|Number of Degrees of Freedom||6,000,000|
|July 2017||TIME SPENT COMPUTING SOLUTION||TOTAL CPU TIME FOR MAIN THREAD||ELAPSED TIME|
|CUBE w32iP||CUEB w32iP||CUBE w32iP|
|# of Cores||CUBE w32iP||CUBE w32iP||CUBE w32iP|
|WO/GPU Acceleration||WO/GPU Acceleration||WO/GPU Acceleration July 2017, drjm, PADT, Inc.|