https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/Computational Fluid Dynamics (CFD)en-USTelligent Community 13https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/flow/66309/fidelity-cfd-automating-geometry-simplification-and-cad-preparationFri, 04 Sep 2026 05:40:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:941e53d2-22b4-4d0b-bb63-b396386d950cGauravFidelity CFD: Automating Geometry Simplification and CAD Preparation One of the biggest challenges in CFD projects is the time spent preparing CAD geometry for meshing and simulation. In many engineering workflows, up to 80% of the total effort can be consumed by manual geometry cleanup, defeaturing, repair, and meshing activities . Fidelity CFD addresses this bottleneck with an automated geometry-preparation workflow that reduces preprocessing time from days to hours. At the core of this capability is Fidelity CFD AutoSeal ™ , an automated geometry cleanup technology designed to handle imperfect or "dirty" CAD models commonly received from design teams. AutoSeal uses a series of successive octree refinements to analyze the geometry and identify regions that are open, incomplete, or unsuitable for CFD mesh generation. By automatically detecting gaps, missing surfaces, and other geometric inconsistencies below a user-defined threshold, AutoSeal progressively seals these openings and creates a watertight geometry suitable for simulation. Key Benefits of AutoSeal Automatically repairs and seals defective CAD geometry. Identifies internal and external cells within complex models. Detects and closes small gaps and missing surfaces. Produces CFD-ready geometry without extensive manual intervention. Reduces geometry preparation and meshing effort from days to hours. Improves consistency and repeatability across simulation projects. Automation Through Python Scripting Beyond automatic geometry sealing, Fidelity CFD provides a powerful Python scripting environment that enables users to automate the entire CAD preparation workflow. Engineers can develop scripts to perform repetitive preprocessing tasks such as: Geometry import and cleanup. Defeaturing and simplification. Surface and volume meshing. Boundary condition assignment. Batch model preparation for multiple design variants. These automation workflows can be executed on local engineering workstations or scaled across High-Performance Computing (HPC) clusters , enabling efficient processing of large numbers of models and significantly accelerating simulation turnaround time. Why It Matters By combining AutoSeal's automated geometry repair capabilities with Python-based workflow automation , Fidelity CFD allows engineers to spend less time on model preparation and more time evaluating design performance, exploring design alternatives, and making engineering decisions. The result is a faster, more scalable, and highly efficient CFD workflow that supports both individual users and enterprise-level simulation programs.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/automesh/66176/flip-a-gemotry-in-autogrid/1408976Thu, 23 Jul 2026 11:36:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:08ddccde-28e6-4b26-8af9-691ec3469a36GauravAt this time, mirroring geometry is not available in Autogrid. However, users can easily achieve this by opening the same geometry in IGG, where the mirroring feature is accessible.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/automesh/66176/flip-a-gemotry-in-autogrid/1408964Wed, 22 Jul 2026 13:50:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:ffdf27e1-839f-4fe2-9297-7b06f3aeafaasima101fHi, Thank you very much for your response. Could you please tell that in Autogrid enviornment? unfortunately I not fimiliar with Fidelity. Thank you Sid TU Dresdenhttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/automesh/66176/flip-a-gemotry-in-autogrid/1408948Wed, 22 Jul 2026 07:13:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:d22f6ec2-b4b7-44b1-973f-a3279ed30dd0GauravHello, you can locate the Flip option in the geometry section of the software. This feature allows users to switch the face orientation of geometry in Fidelity. Steps: Click on the Rendering options widget on the toolbar. Change the Draw mode option to Orientation. Exit the tool. On the 3D view, select the entities to change the face orientation. Right-click on the selected entities. Select the Flip orientation option. Limitation If there are different face orientations in the view, it is not possible to set all front or back faces in a single click using this feature.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/automesh/66176/flip-a-gemotry-in-autogridMon, 20 Jul 2026 12:57:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:95c95090-2a31-4ac8-bdb0-ae10293aac8esima101fHI , i would like to know how to flip a geomtery in Autogrid. I currently have gemturbofile and want to flip it so that after flipping pressure side becomes suction and vice versa ? can someone please guide Thanks Sid TU DresdenIGGmeshAutomeshhttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/flow/66173/when-should-you-use-les-instead-of-ransSun, 19 Jul 2026 06:48:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:0884dab9-8fa7-4568-ab84-ad53a6049670GauravFor CFD applications, consider using LES (Large Eddy Simulation) when unsteady turbulent structures play a significant role in the engineering problem, rather than focusing solely on the time-averaged flow. LES is effective because it captures the behavior of large, energy-containing eddies while modeling the smaller scales, thereby providing a more comprehensive understanding of turbulence. In contrast, RANS (Reynolds-Averaged Navier-Stokes) is better suited for scenarios where modeling all turbulent fluctuations and calculating the mean flow is sufficient. By choosing the appropriate simulation method, you can enhance the accuracy and relevance of your analysis. Use RANS when: Consider using RANS in the following scenarios to enhance your design process: - When you require predictions for steady-state performance to ensure reliability in your system's operation. - If the mean quantities, such as average temperatures, average pressure drops, average airflow distribution, and overall cooling capacity, are adequate for your analysis. - When computational resources or turnaround time are constraints that you need to manage effectively. RANS is particularly effective for: - Conducting airflow studies in data centers - Sizing air conditioning units (ACU) - Predicting average rack inlet temperatures - Calculating pump and fan performance By leveraging RANS, you can take advantage of its efficiency and widespread use in the industry, as it allows for faster solutions with lower computational costs, ultimately facilitating more effective decision-making in your projects. Use LES when: Consider using Large Eddy Simulation (LES) when: Large-scale turbulent structures have a significant impact on results, including vortex shedding behind obstacles, recirculation areas, mixing layers, and jet interactions, where understanding these dynamics is crucial. Strong transient behavior is a key factor: For example, scenarios involving time-varying hot spots in data centers, pulsating flows, flow-induced vibrations, and transient thermal plumes can greatly benefit from LES analysis. Accurate mixing predictions are essential: LES is particularly valuable in applications such as the mixing of cold and hot air, electronics cooling, combustion processes, and HVAC airflow studies, where precision in mixing behavior is important. Aeroacoustics or noise prediction is required: In cases where noise is generated by turbulent fluctuations—something RANS may struggle to resolve—LES proves useful. It is commonly used to predict fans, turbomachinery, and aircraft noise. RANS encounters significant challenges, including massive flow separation, highly curved and swirling flows, stall prediction, and complex wake flows. Research indicates that hybrid RANS-LES approaches can effectively address these challenges, particularly in predicting stalls and separated flows.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/66125/unstable-connection-to-active-glyph-server/1408889Thu, 09 Jul 2026 13:13:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:ef41d6f6-7d9b-488a-8e83-ebc83db9799cClaudio M PitaHi Roelof, Thank you very much for your quick response. I am glad to hear that the provided information proved useful and you were able to resolve the problem. Best regards, Claudio M. Pitahttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/66125/unstable-connection-to-active-glyph-server/1408888Thu, 09 Jul 2026 13:11:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:e03d04b5-2ebf-402f-8341-d005f184d268RF20260309933Hello Claudio, Thank you for your answer. I have resolved the issue now. The problem was that I did not perform step 5 in your list, and instead left the Glyph Server tab open. If I open any other tab, the server will be busy (as you described). Kind regards, Roelof Fennemahttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/66125/unstable-connection-to-active-glyph-server/1408847Mon, 06 Jul 2026 18:33:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:51b28bdb-d9df-4ec5-b9dc-db96263de4d2Claudio M PitaHi Roelof, Thank you very much for submitting your question. The error you received usually means that Fidelity Pointwise is busy with another task when a script is trying to be run. Note that I was able to run your test script multiple times without issues after opening the GUI and setting the Glyph Server to Active. Now, if I try to use any other task in the GUI and then I attempt to run the script, I receive the exact message you posted (which is expected). As a test, please do the following: Launch Fidelity Pointwise Script, Glyph Server Listen Mode: Active Network Port: 2807 Close Run the test Python script and it should run as expected. Back in the Pointwise GUI, Create, Notes (or open any other panel for that matter) Try to run the test Python script once again and note that you will receive the posted error. I hope this clarifies things, Best regards, Claudio M. Pitahttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/66125/unstable-connection-to-active-glyph-serverWed, 01 Jul 2026 12:05:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:71455c87-a6f1-4238-9b1d-60bf73a03ac6RF20260309933Hi All, For a project I am looking to implement an automated meshing pipeline using the python API. I am running into issues with the stability of the connection to my localhost. The Pointwise GUI is running as instructed (active glyph server with default port 2807). The license is also correctly reserved as I can use the GUI as intended. The issue I am encountering is that my test scripts connect to Pointwise unreliably. For example, I ran the provide backstep python example program as provided on github ( https://github.com/pointwise/GlyphClientPython/blob/master/examples/BackstepTutorial.py ). It worked as expected last evening, but this morning I get the attached error message in my console: Both the test code and the provided Backstep Tutorial have ran without issue on my machine. In quick succession as well and multiple times. When the error occurs, I get the response that the server is busy. Note that I have tried restarting my PC, it does not help. Have any of you encountered this problem before, and how was it resolved? Is there a way to check and terminate active server connections in the pointwise GUI? Any other tips or tricks? See also the exact code I use attached. The extension is .txt, as .py is not allowed on this forum. Note that the test program is less then 10 lines of code. Thanks in advance for your assistence, Kind regards, Roelof Fennema community.cadence.com/.../PW_5F00_Test.txt community.cadence.com/.../BackstepTutorial.txtPythonGlyphPointwiseServer Connectionhttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/turbo/66045/analysis-of-flow-physic-within-tip-clearance-gap-of-an-unshrouded-high-pressure-turbine-blade/1408609Tue, 09 Jun 2026 11:44:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:25cc56ef-2467-4717-bc71-f5cdec7d9d21GauravWhen flow enters a tip gap with a sharp pressure-side corner, it undergoes a sharp contraction, resulting in a flow restriction known as a vena contracta. This phenomenon causes a separation bubble to form immediately after the corner. Whether the flow reattaches to the blade surface depends largely on the ratio of the blade tip thickness (t) to the gap height (h). There are two distinct scenarios based on this ratio: For thin blades, where the thickness is less than four times the gap height (t 4h), the physical surface area is sufficient for the flow to mix, recover, and reattach to the blade tip before exiting into the suction side. Specifically, given the blade's thickness being five times the gap height, reattachment of the flow is assured. In the context of unshrouded turbomachinery blades, the primary driving force behind leakage is the aerodynamic load on the blade, specifically the pressure difference between the high-pressure (pressure side) and low-pressure (suction side) surfaces. This pressure gradient effectively forces fluid through the clearance gap. The resulting over-tip leakage flow has two significant consequences: 1. Aerodynamic losses: These losses account for approximately one-third of the total losses in a turbine stage. Even a relatively small gap height, equivalent to just 1% of the total blade span, can result in a stage efficiency penalty of 1 to 3 percent or more. 2. Thermal losses: The high-velocity over-tip flow significantly enhances convective heat transfer coefficients, subjecting the blade tip region to extreme thermal loads.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/turbo/66045/analysis-of-flow-physic-within-tip-clearance-gap-of-an-unshrouded-high-pressure-turbine-bladeFri, 05 Jun 2026 02:10:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:2f3866dd-06b3-45d9-877a-69e996e979c2FA20260604939Hello everyone, I'm currently using Fidelity Turbo to generate a mesh for an unshrouded high-pressure turbine blade. This particular blade features a sharp pressure-side gap corner and a tip thickness that is five times the gap height (t > 4h). Before I run the simulation, I would appreciate it if someone could provide an explanation of the following question: What is the most accurate description of the expected flow physics within the tip clearance gap, and what is the primary driving mechanism behind the resulting efficiency losses? The options are as follows: A) The flow will separate at the sharp corner and remain entirely detached across the gap, with the main loss being driven by centrifugal forces mixing at the hub. B) A separation bubble will form at the sharp corner, but the flow will reattach before exiting, with the primary loss being driven by the pressure differential moving flow from the pressure side to the suction side. C) The flow will remain perfectly attached due to the high thickness-to-gap ratio (t/h = 5), and the primary loss will be strictly due to over-tip thermal heat transfer penalties. D) A vena contracta will form, causing immediate supersonic choking, with the primary loss being driven by shockwave-boundary layer interaction near the stationary casing.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/65872/question-about-reading-mesh-files-in-pointwise-v18-4/1408124Fri, 27 Mar 2026 13:06:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:b36c0613-33bd-4ae9-8953-caa330e0e33cClaudio M PitaHi Akshay, Thank you very much for the update on this matter. I am glad to hear that you were able to open the grid as desired. Have a great weekend! Best regards, Claudio M. Pitahttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/65872/question-about-reading-mesh-files-in-pointwise-v18-4/1408114Thu, 26 Mar 2026 23:48:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:382e5657-078f-4d14-bc2a-595817225374AN202505136027Hey Claudio, Thanks, I was successfully able to read the mesh created in Fidelity Pointwise 2023.2.1 in Pointwise V18.4. Thanks for the quick reply. Regards, Akshayhttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/65872/question-about-reading-mesh-files-in-pointwise-v18-4/1408109Thu, 26 Mar 2026 13:33:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:c67bfcfe-b7b1-4ca3-a148-3f0893e161f9Claudio M PitaHi Akshay, Thank you very much for your question. A potential workaround would be to export the database entities and the grid entities separately in Fidelity Pointwise V2023.2.1 and then import those files separately in Pointwise V18. I would recommend to export the database entities in IGES format and the grid entities in CGNS format. Furthermore, when exporting the grid via File, Export, CAE, make sure you set the current CAE solver to CGNS, and set the CGNS export version 3.3.1 via CAE, Set Solver Attributes (latest CGNS format version won't be readable in V18.4). Last but not least, you may want to make sure all the domains in your grid are assigned to CAE boundary conditions to ensure that the domains in Pointwise V18 look the same as the domains in Fidelity Pointwise 2023.2.1. If you have any other questions, please submit a support ticket at your earliest convenience (www.support.cadence.com). Best regards, Claudio M. Pitahttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/pointwise/65872/question-about-reading-mesh-files-in-pointwise-v18-4Wed, 25 Mar 2026 23:14:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:01006fd1-2f76-47d2-9cdb-e5b1ccb3db02AN202505136027Hi All, I would like to check whether there is backward compatibility, or any available workaround, for reading a mesh created in Fidelity Pointwise 2023.2.1 in Pointwise v18.4. Kind regards, Akshayhttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/turbo/65839/turbulence-model-comparison-for-compressors-sst-vs-k--vs-rsmMon, 16 Mar 2026 04:46:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:b306d46b-fe3c-4559-8192-b7f8aacdcde4GauravTurbulence modeling plays a vital role in compressor simulations by enabling accurate prediction of complex flow phenomena. These phenomena, including strong pressure gradients, boundary-layer separation, and rotational effects, pose significant challenges for numerical simulations. To address these complexities, various turbulence models have been developed, each offering unique advantages and limitations. The choice of turbulence model, such as the Shear Stress Transport (SST) K-omega, Standard K-omega, and Reynolds Stress Model (RSM), significantly affects the accuracy, numerical stability, and computational cost of the simulation. Selecting the most suitable model is crucial to achieving reliable results while minimizing computational resources. The SST K-Omega turbulence model, also known as the Shear Stress Transport model, is widely regarded as an industry standard for turbomachinery and compressors. This model effectively combines the near-wall accuracy of the K-Omega model with the free-stream independence of the K-Epsilon model. The resulting hybrid approach enables excellent predictions of flow separation under adverse pressure gradients. As a result, the SST K-Omega model is well-suited for addressing complex compressor flows. Its reliability in predicting compressor performance and capturing separation in diffusers has made it a preferred choice in industry applications. However, its accuracy can be limited in highly curved flows unless curvature corrections are applied. The Standard K-Omega turbulence model is widely used in computational fluid dynamics for its ability to accurately capture near-wall boundary layer flow. This model is notable for its improved performance in the viscous sublayer, yielding more accurate results than the K-Epsilon model without the need for complex damping functions. The Standard K-Omega model is particularly well-suited for capturing fluid flow behavior in regions with significant near-wall interactions. However, its application is often limited by extreme sensitivity to inlet boundary conditions, which can lead to instability and inaccuracy in complex or rapidly changing flows. In compressor simulations, the Standard K-Omega model has largely been superseded by the SST K-Omega model due to stability concerns. Despite its limitations, the Standard K-Omega model remains a widely utilized tool in turbulence modeling. The Reynolds Stress Model (RSM) is a turbulence modeling approach that delivers high-fidelity simulations of complex, highly swirling, or rotating flows, making it an attractive choice for applications that require detailed flow-field analysis. By directly solving the Reynolds stress transport equations, RSM relaxes the isotropic eddy-viscosity assumption commonly used in K-Omega models, allowing for a more accurate capture of turbulence anisotropy in rotating impellers. This, in turn, enables the model to effectively simulate the intricate flow dynamics present in such applications. However, the increased accuracy of RSM comes at a high computational cost, requiring up to 7 equations to be solved, which can lead to convergence issues and reduced numerical stability.https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/installation/65820/resolving-the-rwboundserr-error-in-fidelity-fine-turbo-2025-2-a-pvm-related-issueTue, 10 Mar 2026 07:05:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:d7df4356-1676-4a1f-a8ea-f74be7b51099GauravThis article discusses a common issue faced by users of Fidelity Fine Turbo 2025.2, where an 'RwBoundsErr' error message appears when attempting to run the solver. The article identifies the root cause as a PVM issue and provides a step-by-step solution to resolve the error. Resolving the 'RwBoundsErr' error in Fidelity Fine Turbo 2025.2: A PVM-related issuehttps://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/turbo/65799/understanding-gpu-acceleration-in-fidelity-fine-turbo-solverTue, 03 Mar 2026 03:17:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:db0018ae-5d68-41ff-bc52-3f8083a4f7f3GauravThis article explains key considerations for using GPU acceleration in Fidelity Fine Turbo Solver, including the maximum number of queues, former RAM limitations, and how to submit a GPU job. By understanding these concepts, users can optimize their solver configurations and take full advantage of GPU acceleration. Troubleshooting GPU simulation setup in Fidelity Fine Turbo 2025.2https://community.cadence.com/cadence_technology_forums/computational-fluid-dynamics/f/installation/65774/how-to-prevent-fidelity-25-2-gui-crashes-caused-by-nvidia-driver-bugsWed, 25 Feb 2026 06:31:00 GMT75bcbcf9-38a3-4e2e-b84b-26c8c46a9500:d174ba51-d9de-4c89-bf8f-83a5380aa698GauravThis article describes a few crash issues with the Fidelity version 25.2 GUI caused by NVIDIA driver bugs. It recommends updating the NVIDIA graphics driver to version 581.80 or newer and changing key NVIDIA Control Center settings (GPU selection, power mode, threaded optimization, V‑Sync, and antialiasing) to fix the problem. https://support.cadence.com/apex/ArticleAttachmentPortal?id=a1OPP000002P90X2AS&pageName=ArticleContent