Flownex – PADT https://www.padtinc.com Website Fri, 05 Jul 2024 17:16:22 +0000 en-US hourly 1 https://wordpress.org/?v=6.7.1 https://www.padtinc.com/wp-content/uploads/2022/02/PADT-Main-Favicon.png Flownex – PADT https://www.padtinc.com 32 32 234120705 An Independence Day Brisket Analysis https://www.padtinc.com/2024/07/05/an-independence-day-brisket-analysis/ Fri, 05 Jul 2024 17:16:18 +0000 https://www.padtinc.com/?p=45598 The smell of fireworks and freedom is in the air! This year, my contribution to the family’s Independence day festivities is a smoked brisket. Unfortunately, I have learned from experience that cooking at elevation (~7000ft) is not the same as it is in the valley (~1000ft). To avoid potential embarrassment it only makes sense to run a few simulations before lighting the smoker.

Why?

Smoking a brisket can be tricky, some would say it’s more of an art than a science but at PADT science is our art so we’re going to lean into the science hard. It can take anywhere from 6 to 24 hours to smoke a brisket depending on a variety of factors. Let’s identify what those factors are;

  • Weight of the Brisket
  • The Moisture Content of the Brisket
  • Air Flow
  • Ambient Air Conditions

The influence of the weight (or mass) of the meat is the main influencer of overall smoke time. The bigger our cut, the more time it’s going to take to smoke. Air flow is going to affect the oxygen available for combustion, effecting the energy available for cooking the brisket, as well as having an effect on heat transfer from the hot air to the brisket. The ambient air conditions will have an effect on energy requirements, heat transfer, and the boiling point of water.

Brisket Stall

The moisture content is going to be our trickiest variable to monitor. We want our brisket to come out of the smoker juicy but as we heat the beef we will evaporate some of the water. This is where the term brisket stall comes from. We eggheads may be more familiar with the scientific description evaporative or adiabatic cooling.

At a certain temperature and pressure water boils and becomes a vapor. This phase transition takes energy, which we are providing by combusting wood chips or pellets, in my case, bourbon barrel oak. Unfortunately, when we hit this critical point we may see all of the energy going to converting water to vapor. Losing too much moisture to vaporization can ruin our brisket. We might increase combustion to add more heat and end up overcooking the meat. We could wait it out and end up drying out the brisket. Worst case scenario is a combination of both!

To avoid this issue many backyard barbecuers recommend the Texas crutch. The brisket can be wrapped in foil or butcher paper. This helps us overcome the stall point by putting a physical barrier in place to reduce the ability for the water vapor to escape the meat.

The Birth of a Brisket Digital Twin

Okay, so not exactly a digital twin. Maybe next time I will incorporate sensor data into the model via OPC server but for now we’re keeping everything digital. I’m using a thermal-fluid modeling tool called Flownex to create a 1D simulation of our brisket and smoker.

The brisket is being modeled using a compound component. I’ve iterated a few times on how exactly I wanted to model it. For the first iteration I kept it simple and used a solid node with an assigned mass and custom solid material characteristics for beef across a temperature range [Baghe-Kandan et, al].

Flownex Compound Component

Unfortunately, although our beef’s thermal characteristics (conductivity and thermal capacitance), is temperature dependent, we do not observe the characteristic stall associated with the adiabatic cooling near the boiling point of water. This is due to not modeling any water evaporating, duh.

First test of 1D smoker model and brisket in transient.

Beef Brisket V2 was also a failure. I tried to model the solid portion of the brisket separate from the water with heat transfer between the two. It kind of worked, but was needlessly complex, and I didn’t see any effect of a change in elevation on the cooking behavior.

Second beta test of smoker plus brisket 1d simulation in transient.

For Beef Brisket V3 I went back to basics. Forget about the smoker, lets just model the solid and liquid that makes up the meat with an applied heat flux and see if we see the stall happen in transient.

Animated transient analysis for the 3rd iteration of the brisket model

The beta testing for V3 was successful! As seen in the above recording, we can see a huge drop in temperature of our solid as soon as our temperature reaches the boiling point of water. We also can see the difference between a sea level analysis and one representing a system operating at 7000 feet above sea level. For the final version of our brisket model I swapped out the two-phase tank with a open container component (to let our water evaporate to atmosphere), and added script to do some basic calculations.

Flownex compound component for the final version of our 1D brisket model.

Transient Simulation and Conclusions

Let’s run the transient simulation in parallel for a smoker at sea level and a smoker at 7000 feet elevation and see what happens!

Transient side-by-side comparison of smoking at brisket at sea level and at 7000 feet.

As we can see in the above transient analysis both briskets experience “stall” once the water begins to evaporate. The brisket at elevation hits its stall at around 199°F, at around the 6 hour mark, the brisket at sea level stalls at 212°F near the 8 hour mark. What this means is that if we are fighting an impossible battle if we try to hit the typically recommended 205°F pull temperature when cooking at altitude! We’d dry our brisket out and burn the meat before we’d hit that mark. I think I’ll be pulling this brisket at ~192°F

Brisket in Smoker

Happy Independence Day!

]]>
45598
Flownex Version Control with GIT Client https://www.padtinc.com/2024/05/31/flownex-version-control-with-git-client/ Fri, 31 May 2024 17:23:24 +0000 https://www.padtinc.com/?p=44671 Flownex 2024 has added version control in the form of a Git Client. In Git, all versions of a project and related files are stored in either a local or a remote repository (aka “repo”). This benefits the user as we are able to revert to older versions of a project merge changes made by multiple users, create branches within a design, and capture all changes in a log. The repository can also function as a form of backup.

Accessing the Git Client

There are two ways to access Flownex’ Git client. We can either navigate directly to the executable: C:\Program Files\Flownex SE X.X.X.XXXX\GitClient.exe, or we can open the Git client from the Flownex GUI itself.

To open the GIT client we will want to navigate to the configuration tab in the Flownex GUI and click on “Open Git Client”.

01 Git Client

Prerequisites for Setting Up a Local Git Repository

Before we can begin storing versions of our project we want to create a repository. The first thing we must do is specify the Local Repository Credentials found in the settings tab.

02 git settings

The Flownex Git utility automatically records modifications and changes made to a network model. Through the Command History settings, we can omit certain user actions from the recording. Typical exclusions include:

  • “Open page”
  • “Deselect all”
  • “Select”
  • “Close page”

There is further reading on the settings customizations in section 24.1 of the Flownex General User Manual.

Creating a Git Repository

We will want to define our repository directory location in the Path or Repository field found near the top of the Git Client GUI. In this example I use the location E:\Tech Tips\5.31.2024.

03 git repo

If a valid Git repo does not already exist at this location we will click “Create” to make one. A new repo with a master branch will be created and an initial commit will also be performed when we click create.

Committing Changes to Git Repository

Once we’ve defined our GIT repo location we can create and commit the current version of our Flownex network. This can be done by clicking “Save and Commit” in the Flownex GUI in the Configuration Ribbon.

04 git save and commit

This will save our project and open the Git client. We should see the command history for any changes since the previous commit. At this point we can add any notes and commit our changes to the Git repository. Then we should click “Commit” to save.

git commit

Note that if we haven’t already defined our user credentials (found in the Settings tab of the Git Client GUI), we will get an error when trying to commit or create the repository.

Reverting to an Older Version

Through the Git client it is now quite simple to roll back to a previous version of a project.

To roll back to an older version of a project we should first ensure that the Flownex project in question is closed. Next, we will launch the standalone Flownex Git Client

We will need to navigate to our Git repo and then we can simply right-click on the version of the project which we’re reverting to and select “Revert to this Version”.

06 git revert

All changes after the selected version will be reverted without changing affecting the Git log. In this case, I can re-open the project and will see that the inlet pipe length has reverted to it’s original 100 m length.

Additional Operations

Other actions and operations available through the Git Client are described below. We can find more detailed information on these in section 24.2 and 24.3 of the Flownex General User Manual.

  • Export this Version
    • Used to export a copy of a selected version of a project.
  • Create and Switch to Branch
    • Used to create new version branches within a given master.
  • Switching between Branches
    • Users are able to switch between branches for a given project.
  • Merging Branches
    • This action will merge changes captured from separate branches into a specified “To” branch.
  • Add Remote and Push
    • This function will push existing local repo branches to a remote repo.
  • Clone
    • Used to copy a remote repo to a local location.
  • Fetch and Merge (Pull)
    • Retrieves changes from a remote repo after a clone or previous fetch.
  • Push
    • Updates all locally made changes to the remote repo.
  • Track Remote Branches as Local
    • This option allows for tracking a non-master branch from a cloned or fetched remote repo.

Thanks for reading through this blog post, we hope you found some information that was helpful. If you would like to learn more about Flownex version control you can do so on our website here.

Have other questions or looking for a quote? Contact us or call (480)813-4884 to get in touch with one of our engineering experts today.

]]>
44671
What’s New in Flownex SE (2023 – Update1)? https://www.padtinc.com/2023/12/22/whats-new-in-flownex-se-2023-update1/ Fri, 22 Dec 2023 12:00:00 +0000 https://www.padtinc.com/?p=43793 The 2023 release of Flownex is adding even more benefits to close out 2023. Flownex® SE 8.15.1 a has improved GIS importing capabilities, transient node results, and flexible graph axis labels. The development team has also updated manuals, tutorials, and verification content. Additionally, the new release offers several improvements to robustness and stability. Below is a list of featured updates in ® SE 8.15.1.

ENHANCEMENT LIST

The complete Release Notes containing detailed descriptions of the enhancements in the latest Flownex® SE 8.15.1 release are available under the Help ribbon in the latest Flownex® SE release.

image1

GIS IMPORTER

  • The latest update has improved the GIS Importer feature to facilitate attribute mapping from shapefiles to component inputs, enhancing customization and adaptability. Additionally, the importer now supports the import of shape length, enabling the creation of single pipes for a shape. This powerful tool streamlines workflow efficiency and offers enhanced flexibility for accurate system modelling.
image1

GRAPH AXIS LABELS

  • Users now have increased flexibility with graph customization. This update introduces the ability to modify the font and font size of axis labels (the numerical markers) on graphs. Experience enhanced graph customization to better suit your visualization needs 
image2

SOLID NODE RESULTS

  • Introducing a new result parameter to solid nodes – the ‘Heat accumulation rate.’ This parameter depicts the rate of change in total thermal energy within a solid node throughout transient simulations. Users can now obtain a more comprehensive understanding of thermal behaviour within solid components during dynamic scenarios. 
image3

ZEHNER-SCHLÜNDER MODEL IN REACTOR BUILDER SCRIPT

  • Users can now leverage the Zehner-Schlünder model for effective conductivity using the Reactor Builder Script. The model can be used by selecting it on the appropriate zones on the reactor chart. Activating this model enables precise conductivity adjustments, optimizing the accuracy of thermal behaviour simulations for superior performance assessments. 

ADDITIONAL ENHANCEMENTS

  • To ensure updated guidance and best practices, various manuals, tutorials, and the Validation and Verification pack have been updated. These updates aim to provide users with refined resources, enabling them to maximize their simulation capabilities. For more detailed information on these enhancements, please refer to the comprehensive Release Notes available on Flownex® SE 8.15.1. 
image1
]]>
43793
Flownex Student Rocketry Highlights https://www.padtinc.com/2023/12/15/flownex-student-rocketry-highlights/ Fri, 15 Dec 2023 13:00:00 +0000 https://www.padtinc.com/?p=43558
Captain James T Kirk of the Starship Enterprise

Space: The final frontier. These are the stories of the next generation of starship engineers.

PADT is excited to share highlights from our partnerships with student-teams taking part in the 2023 Launch Canada (LCRA) and the Spaceport America Cup (SACUP) competitions.

Rocketry Competition Overview

image 6

US and Canadian university rocketry teams competing in the Launch Canada and SACUP are developing student researched and designed (SRAD) rockets using Flownex to explore cycle design, component sizing, pump/compressor selection, and more!

Below are some of the key features make Flownex an ideal tool for engine fuel cycle design:

  • Implicit solver – fast and slow transients.
  • Two-phase fluid and multi-fluid mixture modelling
  • CEA Adiabatic Flame modelling
  • Joule Thompson effect and choking predictions – based on fundamental principal calculations.
  • Exit Thrust modeling
  • Integrated controls modeling
  • Native ANSYS Coupling (Fluent, CFX, Mechanical, and Workbench)

Competition Highlights

The Flownex partnership with Launch Canada began in 2022. As we close out 2023 there have been a total of 16 Canadian teams taking advantage of Flownex Simulation Environment for design and analysis. This years competition saw very impressive simulation implementation from several teams which we’d like to highlight.

The effort by the Metropolitan Aerospace & Combustion Hub (MACH) team, was especially notable. A special thanks to Shivesh Maraj and Umar Shabir (Former Team Director), for helping the team get access to the software.

We were fortunate to be able to talk with Ben Kubica (Technical Director) and Shivesh, who had recently interned at Pratt and Whitney Canada, to learn how the team was able to use Flownex in this competition. The MACH team had used Flownex to develop a full system model of the liquid run lines to evaluate system performance, and size components for their liquid bipropellant engine.

MACH team-members saw opportunities to implement Flownex to design their fuel delivery system. Members used Flownex to predict the onset of cavitation and resultant flowrates of venturis to control fuel flow. The team leveraged these predictions to design venturis to achieve target flows and evaluate diffusor efficiency.

Injector and Venturi
Injector and cavitating venturi design

The MACH team was able to conduct successful inert fluid cold flow tests on their completed components. This is an integral step in their component design validation. The test resulted in successful flow and pressure data collection, which was processed for later analysis.

flowrate
Ethanol Mass Flowrate
pressure
Injector inlet pressure (P20), venturi back pressure(P22), tank pressure (P21) vs time

The Mach team compared fuel side data empirical data to their simulation results. They reported that they were able to predict the fuel choked flowrate in the fuel lines within <15% of initial test results and Injector pressure drops deviated <3% from test results from their early design phase simulations.

fnx data comparion

The MACH team was also able to perform their first hot fire attempt at the Launch Canada competition. While they experienced an ignition failure, they were able to collect valuable dual cold flow data during the competition.

20230827 162921 1
Photograph of the MACH team after the 2023 Launch Canada Competition
MACH full system 1
MACH full fluid system with test stand

We were also able to ask Ben how the use of Flownex impacted his project timeline. Ben Commented that Flownex allowed projected timelines to be cut in half with verification and manufacturing reduced from 6 months to 3 months, advancing the GAR-E Engine development process. Ben Concluded:

“With Flownex in our toolbelt, we were able to shorten our test campaign by months, allowing us to nail our optimistic deadlines.”

Ben Kubica – MACH Team Director

Ben and Shivesh informed us that the team planned to characterize the oxidizer system and preform another hot fire attempt this coming year. We are thrilled to see the MACH team’s progress and look forward to their future developments!

New Team Highlights

Flownex is also gaining momentum with US teams as well. Flownex does not have an official affiliation with the Spaceport America Cup, but we have formed academic partnerships with several US teams to launch 2024 projects. We are excited to see how these teams use Flownex to propel their design process.

Ben Kamer, the Vice Director of Boston University Rocket Propulsion Group (BURPG) reached out to PADT to acquire Flownex for the BURPG fluids team. Although Ben’s team has only had access to Flownex for a few months, they have used it to model flows in their Icharus flight vehicle.

The BURPG team has also started to use Flownex to verify hand calculations for their initial pressure drop calculations. Ben stated that the tool was quickly utilized by members of the fluids team. The rapid adoption and reliability of the tool has expedited the BURPG ten’s design process. Ben commented on how Flownex is helping move their team’s testing forward:

“Flownex lets me verify my pressure drop calculations so much faster than having 3 other people check them. It’s been wonderful, verifying my pressure drop calculations with only 0.2% error”

Eric Zhao – BURPG

“Using Flownex for my tank draining simulations saved me so much time, I didn’t have to make a whole MATLAB program. It probably saved me 3 days of work.”

Benjamin Kamer – BURPG

Ben added that the BURPG team plans to eventually compare data to a future cold flow attempt to Flownex results.

New Horizons

We are excited to provide the next generation of thermohydraulic systems engineers with resources for designing safe, robust, and efficient propulsion systems. We hope to provide these students with modelling skills and experience that will propel their team projects forward and prepare them to engineer the next Starship Enterprise.

If you are interested in using Flownex to launch your design process, please contact us at flownex@padtinc.com

]]>
43558
Borrowing a Flownex License https://www.padtinc.com/2023/12/08/borrowing-a-license/ Fri, 08 Dec 2023 13:00:00 +0000 https://www.padtinc.com/?p=43652 Today we will walk through how to obtain and maintain a license by connecting to a server and checking out a license. Flownex requires that users to have a license to run, so checking out a license may be useful if you are planning to travel, but do not have a virtual environment setup. Prior to checking out a license, you should check with your management and/or server admin to ensure that sufficient licenses are available for other users. For this Example, we will use Flownex version 8.15.0.

Introduction

We will discuss prerequisites for borrowing a license, how to borrow a license and finally how to return a license. Flownex uses Reprise License Manager (RLM), which is relatively easy to interface with from the client side. Once the license is checked out, you may use Flownex for up to 30 days after disconnecting from the server. Whenever reconnect to the server, your checkout will be renewed for the requested number of days (up to the expiration date). This article covers license checkout for Flownex 8.15.0.

Considerations

You should ensure that you meet the following requirements prior to checking out a license:

  • Adequate licenses are available for other users
  • You are using a locally installed version of Flownex
  • You can access the server to borrow/return a license

Opening Flownex

After you connect to the license server network, Open the version of Flownex that you would like to borrow a license for.

Startup

Connecting to the Server

After you have successfully opened Flownex, Click “Help” then Click “Borrow License from Server”. You will be prompted to specify the number of days that you would like to borrow a license and Click “Ok”.

License Checkout

You will receive a prompt that you have successfully borrowed a license! Click “OK” to dismiss the prompt. You are now ready to disconnect from the network until the borrowed license expires.

Verifying Checkout

Flownex will check for the new license after opening. Open a new instance of Flownex to verify that a license is checked out. You should see “- Borrowed License -” in grey text at the top of the newly opened Flownex window. Flownex will also indicate the date that the checked-out license will be returned without renewal.

Verification

The license will automatically renew each time that you reconnect to the network. This is a good way to check the expiration of your borrowed license from the Flownex client.

Returning a License

To return a license, make sure that you are connected to the network, then click the “Return License to Server” button.

Return Licence

You should receive a prompt that the license has successfully been returned to the server.

CheckIn

Similar to the checkout, Flownex will not reflect that you have you have returned a license until you open a new project.

Closing Thoughts

That is all you need to know to checkout and check in licenses. This is a great way to use Flownex if you are travelling and your company has not setup a VM, or if you plan to use Flownex in an area without internet access. Make sure your company is aware that you are checking out licenses, and that you return them as needed. Thanks for and stay tuned for more Flownex Tech tips.

]]>
43652
Using Gauge Pressure in Flownex https://www.padtinc.com/2023/12/01/using-gage-pressure-in-flownex/ Fri, 01 Dec 2023 13:00:00 +0000 https://www.padtinc.com/?p=43506 Pressure formatting varies with the target industry. Many solvers, including Flownex use absolute pressure to resolve the mass, momentum and energy equations. However, it is often convenient to view and present pressure data in a way that aligns with standards or external specifications. Previous releases of Flownex have allowed users to vary the ambient pressure, but the 8.15.0 Flownex release allows users to specify and read gage pressures in (standard metric and imperial units) with respect to ambient. This blog will discuss how to modify the ambient pressure, and specify gauge pressure.

Gauge Pressure Boundary Condition

We can quickly create a gauge pressure boundary condition by changing the units on an existing pressure boundary condition.

Setting Gage Pressure 1

Varying Ambient Pressure

In environments where ambient pressure changes, such as deep-sea exploration, or spaceflight, you may wish to modify the ambient pressure. You can easily modify the ambient pressure from the Flow Solver.

Changing Ambient

Flownex 8.15.0 will, by default, store a total pressure and report a gauge pressure for a boundary condition. In this case, the boundary condition will be specified as 110 kPa, and Flownex automatically reevaluates the gauge pressure input to maintain the same total pressure. If you would like to maintain a constant gauge pressure with respect to ambient, you can use a data transfer link with assigned.

Additionally, if you are interested in transiently varying the atmospheric pressure, this setting will not save in a snap because flow solver settings are not saved to snaps by default. You will need to set the atmospheric pressure as a global variable to capture it in a snap.

Closing Thoughts

Now we are familiar using gauge pressure boundary conditions in Flownex. Gauge boundary conditions can increase the readability of a network or define pressure based on ambient conditions. Thanks for reading and stay tuned for more Flownex tech tips!

]]>
43506
Embedding a Designer in a Parametric Study https://www.padtinc.com/2023/11/22/embedding-a-designer-in-a-parametric-study/ Wed, 22 Nov 2023 15:26:27 +0000 https://www.padtinc.com/?p=43367 Many Flownex users are familiar with using the Designer to size components to meet design requirements or using parameter tables to sweep through various operating conditions. Well, what if you need to do both? You may want to know operational inputs (without building out controls) for various operating conditions. Flownex currently offers component level (“Change Design”), precompiled steady state controller scripts, or running the designer through a script. Flownex version 8.15.0 now allows users to run the Designer during a Parameter Table run through the GUI.

Introduction

Today we will walk through how to embed a designer setup inside of a Parameter Table to vary the oil inlet pressure into a gearbox cooling compartment. We will use the designer to adjust the opening of the air bleed opening, to ensure that our container interface (top/bottom) maintains force balance across an oil lubrication system at each design point.

Network Description

For our network, we will be modelling a simplified turbine lubrication system. Air is introduced through labyrinth seals, modelled as a boundary condition, and exits through a restriction at the top of the compartment. Oil is introduced to the container and partially entrained with air (through a simple fixed percentage) before exiting the compartment. Below is a diagram of the system.

New Bitmap Image

The container interface top/bottom components allow us to specify the steady state level of oil in the bearing compartment. We will need to enforce force balance across the air/oil interface to ensure that oil will not be flooding or purging. In this network, we have used the designer to balance interfacial forces by adjusting the area of the outlet air restriction.

Designer Setup

We will look at how the temperature of the compartment varies with oil inlet pressure. However, we will need to vary the pressure drop of the air outlet to maintain force balance across the air/oil interface for each inlet pressure scenario. Rather than model air outlet pressure regulation, we will use the Designer to set the opening of the restrictor to achieve force balance.

Designer 2

Parameter Table Setup

After creating a Designer setup to maintain force balance, we can create our Parameter Table (covered in a previous post). Our parametric study will vary the oil inlet pressures and log the resultant mass flow rate and temperature of the oil air mixture exiting the system.

Parameter Table

Nested Parametric Study

Once we have created our parameter table and Designer setup, we can run a parametric study by clicking “Run Parameter Table” from the Home Menu. The parametric study will require us to choose a Parameter Table, then select additional run configuration options. We will select our Designer name from the dropdown and leave all of the other options as default, then select “OK”.

Nested Designer

With a few clicks we have nested our Designer inside of the parametric study – no scripts required.

]]>
43367
Setting Steady State Boundary Conditions https://www.padtinc.com/2023/10/19/setting-steady-state-boundary-conditions/ Thu, 19 Oct 2023 21:32:05 +0000 https://www.padtinc.com/?p=43110 The newest release of Flownex offers many intuitive features to expedite your workflow. One of my personal favorites is the steady state boundary condition. This is a powerful tool that can reduce clutter transient action setups or scripts. This feature allows users to specify an initial boundary condition, such as a property inside of a loop or a reservoir without enforcing the condition in transient. For this post, I will use version 8.15.0.

New Bitmap Image 4

Network Setup

We will be looking at a simplified tank blowdown to demonstrate the steady state boundary condition. This network will use a reservoir and a resistor to represent a tank and valve, respectively. The boundary conditions can be set in the same manner as a steady state simulation. Once our boundary conditions are setup, we will need to check the “Steady state only” option.

Untitled Project

Note that the temperature and pressure indicators for the tank boundary condition -“P(S)” and “T(S)”- now have a steady state suffix “(S)” to indicate that the boundary condition is only valid for steady state.

Running The Simulation

After setting up the boundary conditions, we can solve our steady state solution, setup a graph of mass flow, and run our simulation.

Run and Graph

We can see that the “Steady state only” option enables the boundary condition in steady state but does not enforce any condition in transient. You can also quickly check the initial conditions during a transient run by looking at the boundary condition properties.

Closing Thoughts

We have learned how to implement a steady state boundary condition in our simplified tank blowdown. Setting steady state boundary conditions is an easy way to simplify transient setups and increase the visibility of initial conditions in a network. Thanks for reading and stay tuned for more posts.

]]>
43110
Adjusting Timesteps in Flownex https://www.padtinc.com/2023/09/29/adjusting-timesteps-in-flownex/ Fri, 29 Sep 2023 12:00:00 +0000 https://www.padtinc.com/?p=42724 Striking the right balance between fidelity and solve speed can be difficult. Flownex has several methods of handling timesteps. We will cover some basics of the adaptive time step as well as how to change the timestep during a simulation using actions. For this post, we will use Flownex version 8.15.0.5222.

Adaptive Timestep Background

We reached out to the developer to gain some additional insight on the adaptive timestep. We learned that the primary purpose of this feature is to maintain the fidelity of the solution rather than reduce the solution time. Below is a summary of the developer’s response:

The adaptive timestep is recommended for specific simulations where rigorous timestep independence studies are required, which will then rather aid the user in saving time to determine when smaller timesteps are typically required and what time step size to use. We would therefore recommend the following to consider when to use the adaptive timestep algorithm:

  • The goal of adaptive timesteps:
    • Automatically calculate the required timestep size to ensure that the truncation errors that result from the timestep integration are less than the specified tolerances in the scheduler settings. 
    • This does not speed up the time of a single simulation as the adaptive timestep algorithm repeats timesteps to calculate the truncation error at every timestep (this means, at minimum, the solver will need to solve 3x as many timesteps).
    • Instead, this speeds up the complete workflow for timestep independence studies (for example, water hammer simulation in a pipe network or uncertainty analysis for a nuclear system simulation) by automating the reduction in timestep and calculating the truncation error at each timestep.
  • How it works:
    • The theory is simply to halve the timestep and check that the truncation error is less than the tolerances that the user specified. If the truncation error is larger than the tolerance then the timestep is reduced and the process is repeated until the truncation error is less than the tolerances. The truncation error is found by “brute-force”: finding the difference in (results after one ΔT) and (results after two ΔT/2). 
  • Using this feature:
    • The truncation error is the error that arises by using the derivatives and a non-zero timestep to approximate the function (see image for extra information) for the non-iterative solver. The iterative solver is less sensitive to truncation errors but not entirely immune to it.
    • The first step is to determine an appropriate truncation error tolerance. This is not simple as choosing the right tolerance is similar to choosing the right relaxation parameter. There aren’t general rules that work all the time for all simulations and it requires experience to choose good values (or at least we can’t explain these rules in a simple way that is easy to generalize).
    • If you are performing a workflow that requires a rigorous timestep independence study then we suggest you speak to us so we can explain what values we would choose for that application. The hope is that with time you get a feel for how we choose these numbers.
ATSA

In the image the red line is the numerical solution and the blue line is the exact solution. At point A_0 the 2 solutions are the same as set by the initial conditions. The value at point A_1 is calculated using the gradient(s) at point A_0 and a specific timestep size. If the timestep is larger the difference between the numerical solution and the exact solution gets larger. This difference is referred to as the truncation error.

Using Adaptive Timestep

Changing the timestep using the adaptive timestep setting is relatively straightforward. Timesteps may be changed like any input property in Flownex. Users can access this property by viewing the Scheduler properties. This can be accessed through the solvers pane by clicking on the scheduler. You can also pop out the property window by clicking on the “Time Step Settings” in the Home Ribbon.

Timestep settings

Modifying the Time step control property to “Adaptive time step” will allow you to set how quickly the timestep increases, high and low thresholds, in addition to appropriate tolerances.

Timestep Settings

Once you have selected appropriate timestep and tolerance settings, you can run your simulation. We will run a simple network where the mass flow on a pipe is ramped down to zero through an action. The video below shows our network running with the adaptive timestep.

Adaptive Timestep

Modifying the Constant Timestep

You may also choose to modify the constant timestep during the simulation. This is most commonly implemented using actions, but users can also change this property through a script or parametric study.

If we know the timing of fast transient events during a long transient, it may be a good option to use an action. We will start with a simple pipe network, where a mass flow is reduced to zero through a transient action. We want to reduce the timestep quickly before the valve closes.

We will set the solver to initially run at 10ms prior to the transient event, and ramp down to capture the water hammer. First, we will need to open up the Actions Setup to view the running actions. We can drag and drop the timestep from the Scheduler properties into the into the actions pane to modulate our timestep.

Actions Setup 2

This action will ramp the flow solver from 100ms to 1ms. Now that we have our timestep setup, we can run our simulation.

Action Timestep

The simulation runs faster until the moment before mass flow is reduced. We are than able to capture pressure oscillations with a higher resolution by ramping our timestep down.

Closing Thoughts

Now we have learned a bit more about how the adaptive timestep operates and how we can modify the timestep property during a simulation. Timesteps may be adjusted in Flownex to reduce simulation time, or to ensure that reasonable truncation errors are achieved in a simulation. These tools are helpful for efficiently running transient simulations with multiple timescales.

]]>
42724
What’s New in Flownex SE 2023 https://www.padtinc.com/2023/09/15/whats-new-in-flownex-se-2023/ Fri, 15 Sep 2023 13:00:00 +0000 https://www.padtinc.com/?p=42788 The anticipated 2023 version of Flownex (8.15.0.5222) has just been released under 8.15.0. This release comes with several major enhancements. This release focuses on improving workflow productivity, enhancing user interaction, and expanding accuracy of simulations. Major highlights include parallelized sensitivity analysis setup, improved accuracy for various rotating components, enhanced visualization features, and much more. Below is a list of enhancement highlights and we invite you to explore the comprehensive release notes for an in-depth understanding of the enhancements and features that await you in our 2023 release.

ENHANCEMENT LIST

The complete Release Notes containing detailed descriptions of the enhancements in the latest Flownex® SE 8.15.0 release are available under the Help ribbon in the latest Flownex® SE release.

image1

BATCH RUNS

  • Streamline Data Management for batch runs such as Monte Carlo and Parametric Studies with Flownex 2023’s Batch Run Graph feature. Simplify the handling and visualization of extensive datasets, empowering in-depth analysis and insights from your simulations.
image3

PARALLEL MONTE CARLO RUNS

  • Experience faster results for complex simulations as you leverage multiple instances of Flownex in parallel to perform Monte Carlo Analysis and Parametric Study Runs. Now, you have the flexibility to define a designer configuration, scenario action, and snap to initialize each parallel run, streamlining your analysis. You can also configure and execute your analysis seamlessly from the same dialog, enhancing your simulation efficiency.
image3 2

PARAMETER TABLES SELECTION

  • Now, it’s easier than ever to identify which Parameter Table is set to run. This enhancement enhances the clarity of your simulation setup and contributes to a more intuitive user experience.
image13
image14

  • Experience a more intuitive and consistent approach to managing Flow Solver Component links in Flownex’s latest update. We’ve aligned the functionality of flow solver component links with regular links, streamlining your workflow for enhanced usability. Now, you can effortlessly select, format, and draw links across pages, making the interaction more cohesive and efficient.
image16 1

COMPOUND COMPONENT EXPOSER ENHANCEMENTS

  • Compound exposer enhancements were added to facilitate the changes to the links for the Flow Solver components. When you connect a link Exposer within the compound template, it becomes accessible outside the compound as well. This link is automatically generated along with the compound, simplifying your workflow and providing seamless access to essential components.
image23
Figure 22: Link Exposer Tool.

ENHANCED SNAP MANAGEMENT

  • Flownex 2023 brings an improved Snap Window, making your simulation workflow even more efficient. Organize your Snaps with names, nested folders, and comments, allowing you to categorize and retrieve them effortlessly. Plus, the added search function ensures that you can quickly locate specific Snaps, streamlining your simulation data management.
image8 1

ENHANCED 3D VIEWER

  • Flownex’s 3D Viewer has received significant improvements in the 2023 version, enhancing your simulation experience. Navigating in 3D space is now more responsive and seamless, providing you with a smoother interaction. The addition of a dedicated ribbon tab makes edits within the 3D space a breeze, allowing you to enjoy a more streamlined and efficient workflow.
image9 1

  • Experience a more streamlined and automated setup process with the improved Ansys Mechanical link in Flownex. Now, the link is enhanced to effortlessly retrieve the named selection list and surface areas from Ansys Mechanical, enhancing your workflow efficiency. Say goodbye to manual setup hassles and welcome a more automated and efficient interaction between Flownex and Ansys Mechanical.
image10 1
Previous Method
image49 1

ENHANCEMENTS TO 3D GRAPHS AND 3D GRAPHS FOR NUCLEAR REACTORS

  • Experience enhanced visualizations with the upgraded 3D graphs in Flownex. These graphs now offer improved responsiveness and utilize actual lengths and distances as coordinates, providing a more accurate representation of your data in a three-dimensional space.Additionally, for nuclear reactor simulations, a new component has been introduced. This component facilitates the translation of text results from reactor simulations into a format compatible with the 3D graphs. This integration allows for seamless visualization of reactor data in the enhanced 3D graphs, providing valuable insights into nuclear reactor behavior.
image14 1

INTRODUCING OPC UA INTEGRATION

  • Flownex 2023 now offers enhanced connectivity with the addition of an OPC UA client component. Seamlessly exchange data between Flownex and OPC UA servers using this powerful new feature. Whether you’re working with real-time data or process information, the OPC UA client component enables efficient and reliable data communication, further expanding Flownex’s capabilities to integrate with industrial systems and processes.
image55

STEADY STATE BOUNDARY CONDITION

  • Flownex 2023 introduces a significant enhancement to Boundary Conditions by introducing a “Steady State Only” option. Now, you have the power to exclusively apply a Boundary Condition during steady-state simulations. This innovative addition streamlines your simulation setup, eliminating the need to modify Boundary Conditions for transient analyses.
image16

GAUGE PRESSURE OPTION

  • With this advancement, you can effortlessly set pressure fields as gauge pressure, eliminating the need for tedious conversions. Define atmospheric pressure in the Flow Solver for your entire network and specify gauge pressure at designated boundary points.
image17 1

MATERIAL SCALING FACTORS

  • Now at your fingertips are several scaling factors meticulously incorporated into pure fluid and solid materials. These factors are your key to unlocking sensitivity analysis, empowering you to explore uncertainties tied to material properties.
image18 1

COOLING TOWER ENHANCEMENTS

  • Elevate your cooling tower simulations with the introduction of additional terms to the Merkel function, adding a new layer of sophistication to your analyses.
image19 1

ROTOR-ROTOR AND ROTOR-STATOR ENHANCEMENTS

  • Revamp your rotor dynamics analysis with the new option to use a linearly interpolated function for core swirl ratio when connecting subsequent radial increments. By allowing the core swirl ratio to be interpolated linearly between subsequent increments, radial changes in temperature and pressure can be accounted for more accurately.
image20 1

CUSTOM VORTEX ENHANCEMENTS

  • In addition to the existing exponential relationship, users can now define a linear change in swirl ratio along the vortex path, from inlet to outlet. This option empowers you to model user-specified swirl variations, offering unparalleled realism for scenarios such as secondary flow reintegration into the main flow path.
image21 1

ROTATING NOZZLE ENHANCEMENTS

  • In the latest update of Flownex® SE, the Rotating Nozzle feature gains a new level of adaptability. Now, users have the option to treat discontinuities between forward and reverse flow as warnings, rather than error messages.
image22 1

TWO-PHASE FLOW PROPERTY ENHANCEMENTS

  • Flownex 2023 introduces enhanced Two-Phase Flow Properties for more accurate and stable simulations. The updates include consistent data distribution for files produced by the fluid importer. Refined interpolation routines for subcooled to superheated transitions and lower temperature limits in the superheated and subcooled region using REFPROP methods as well as improved consistency in forward and reverse interpolation. And improved property derivation for azeotropic fluids at low qualities.
image23 1

CONVECTION COMPONENT TEMPERATURE DIFFERENCING SCHEME

  • A user-selectable option is now available to use the difference between the solid wall temperature and the average of the bulk fluid temperature that heat is transferred to via the convection transfer mechanism.  This allows the use of the Convection element with a two-phase fluid undergoing phase change.
image83

SCRIPTING IMPROVED INTELLISENSE

  • The intellisense for Scripts have been improved to work with local variables and member variables that has not been compiled yet. It also now functions when operators like equals are used between variables without using spaces

ERROR AND WARNING MANAGEMENT

  • Critical Warnings will now be separated into their own category to ensure the user can easily identify critical information. The capabilities to limit the warnings shown and log errors and warning to a file have been added for easy error management.
image25 1

ZEHNER-SCHLÜNDER MODEL FOR EFFECTIVE CONDUCTIVITY

  • The Zehner-Schlünder model for effective conductivity is now implemented in the conduction heat transfer component and can be used in the Reactor Builder Script. The model now uses the Poisson Ratio and Young’s Modulus specified for the Effective Conductivity Material instead of the previously hardcoded values in the Reactor component.
image26 1

PYTHON INTEGRATION IN SCRIPTS

  • Functions in Python can now be called directly in Flownex scripts.
image27 1

INSULATED PIPE

  • The Temperature differencing scheme input option has been added as an input field for the Insulated Pipe.
image83

ENHANCEMENTS

  • For detailed descriptions of each update and how to use it, please refer to the complete Release Notes available on Flownex 8.15
image1 2
]]>
42788