Academic

BLOG | 14 December 2021

FLOWNEX AS A TEACHING TOOL

“Through simulation and experimentation, students are exposed to real systems and can gain a better understanding of fundamental concepts and relationships.”

Mastering your course

One of the great challenges of education, for both student and lecturer, is to ensure the knowledge gained is also retained past the last exam paper. This is especially difficult in the Engineering environment as class schedules are fully packed and the time to reflect and internalise key concepts become limited. Lecturers are faced with the challenge to convey complex subject matter in a short timeframe, while simultaneously ensuring students do not simply memorize but understand the material they are presented.  Often these fundamental concepts are both novel and confusing to students at first, and further subject matter only builds on this foundation.  If it is not firm and unyielding, students may find themselves at the end of a semester not having fully mastered their course.

A practical teaching element is usually incorporated in courses to facilitate discovery and understanding with subsequent tests and assignments to stimulate long-term memory recall of the material. Through simulation and experimentation students are exposed to real systems and can gain a better understanding of fundamental concepts and relationships. By incorporating simulation models, students can assess and experiment on simple and complex networks they would not have had access to otherwise.

“Flownex makes this experimentation possible without neglecting the mathematical grounds and fundamental principles that it is based upon.”

In a classroom environment, components are usually evaluated separately and in idealised conditions, while in practice this is almost never the case. The simulation of thermal flow components and networks facilitates the practical and realistic application of conditions to a 2-D representation.

Components are interconnected and dependent on the response of its neighbouring connections as well as the overall network. By giving students the ability to actively influence the conditions of an example, there exists a greater chance of them understanding the principles. Changing the pressure and seeing the system respond in real time solidifies the relationships drawn by the theoretical knowledge much more than simply being told they exist.

Using Flownex as an advantage

With material covering Thermodynamics, Fluid mechanics and Heat transfer, Flownex facilitates the understanding and experimentation of fundamental concepts. Lecturers can use this to their advantage by demonstrating different concepts and conditions whilst enabling students to easily learn on their own through experimentation. As Flownex is used in more than 40 countries around the world, students gain important experience in thermal flow simulation and evaluation.

Rankine cycle example

An essential cycle to understand in practice is the Rankine cycle, as most steam-based power plants are based on this closed cycle. The Rankine cycle, usually utilising water as a fluid, starts by pressurizing the fluid with a pump. The pressurized fluid is then heated beyond its boiling point in the boiler to produce steam which is expanded through a steam turbine that extracts mechanical energy from the system through the shaft. Finally, the steam is cooled back to a liquid state in the condenser and returned to the pump to start the cycle again.

To boil the fluid any heat source of adequate temperature can be used. Historically the combustion of fossil fuels was used, but more sustainable heat sources like nuclear and solar radiation are frequently used and becoming more prevalent.

The network and efficiency of the Rankine cycle is highly influenced by the temperature and pressure of the cycle and can be increased by:

  • Lowering the condenser pressure,
  • Increasing the pressure during heat addition, and
  • Superheating the steam.

Cycle variations of the basic Rankine like the reheat and regenerative cycles offer improved efficiency at the cost of simplicity.

In Thermodynamics the Temperature – Entropy diagram (T-s diagram) is most frequently used to analyse energy transfer, as the work done by or on, and the heat added or removed from the system can easily be visualised. Of course, the system response can be calculated from the first principles, but the real-time response of the system allows faster analysis with less complexity.

What becomes apparent in the simulation of such a system is the inherent energy losses of real thermodynamic cycles due to inefficiencies in the components. Power cycles are often evaluated on their thermal efficiency, as the ratio of the mechanical output to the thermal input gives a better sense of the real-world cycle performance. The track bars can be adjusted in the Rankine example to demonstrate the system response and the increase/decrease inefficiency.

An intuition of system and cycle response usually formed through years of experience can easily be developed by interacting with simulated examples. The visual and transient response presented by interacting with the simulated example gives students the opportunity to better understand the complex and sometimes abstract theoretical material.

“The visual and transient response presented by interacting with the simulated example gives students the opportunity to better understand the complex and sometimes abstract theoretical material.”

Creating a stimulating environment and equipping the students with the right tools to reach a point of deeper understanding and experience is the true calling of a lecturer. We encourage you to consider the engineers you send out into the world and how you can prepare them to be the best they can be.

Let Flownex help you in that journey!

Riani Wagner

Riani Wagner

Riani is the technical sales engineer for Flownex direct and has a passion for promoting education.

Hydrogen

BLOG | 29 October 2021

THERMAL AND WATER MANAGEMENT OF THE PROTON-EXCHANGE MEMBRANE FUEL CELL

“Flownex gives you several options and capabilities for modelling the system components and fuel cell stack of the PEM fuel cell.”

HYDROGEN AS A PATHWAY TO DECARBONISATION

As climate change has become a rising concern, more and more countries have dedicated themselves to a zero-emissions goal. Many have now joined the Paris Agreement on climate change and are aiming for net-zero by 2050.

“When using hydrogen within fuel cells, no CO2, or other hazardous emissions such as SOx’s or NOx’s are produced.”

Green hydrogen (hydrogen produced from renewable energy) has been one of the main topics of discussion as the world is shifting efforts to a zero-emissions solution. The combustion of hydrogen with pure oxygen will release only heat and water as by-products – without any direct emissions. Combustion of hydrogen in air at high temperatures will, however, result in NOx emissions: a dangerous group of gasses that can cause respiratory problems, headaches, eye irritation, and other impacts on human and animal life. When using hydrogen within fuel cells, no CO2, or other dangerous emissions such as NOx are produced. Due to this, fuel cell technology has been getting a lot of attention in the past few years.

THE PROTON-EXCHANGE MEMBRANE FUEL CELL

Fuel cells can be used in a wide variety of applications, from transportation to electrical systems on a space craft. Fuel cells also have several advantages when compared to conventional combustion-based technologies. One being that it has a much higher efficiency, even above 60%. Another, as mentioned, no harmful emissions are produced during the operation of the fuel cell. 

“This highlights the reason why so much focus is put on fuel cells: it is an energy source with the only emission being water and heat.”

The Proton-exchange membrane (PEM) fuel cell has the advantage that it operates at lower temperatures (at around 80oC) when compared to other fuel cells. This means that it has a quick start up time because less warmup time is needed. It also has a high power density and is low in weight. The fuel cells used in electric vehicles are most commonly the PEM fuel cell.

  • Hydrogen is supplied from a high-pressure tank into the anode side of the PEM fuel cell where it comes into contact with the electrolyte and splits up into Hydrogen ions (protons) and electrons.
  • The electrons cannot move through the membrane and is forced move through a conductor.
  • Air is suppled at the cathode side using a compressor.
  • When the electrons arrive at the cathode side through the conductor, the oxygen reacts with the hydrogen ions to form water and heat.

This highlights the reason why so much focus is put on fuel cells: it is an energy source with the only emission being water and heat. The fuel cell can be connected to a load, such as an electric motor in a hydrogen vehicle, and can be used in conjunction with a battery to accommodate for certain high demands.

PEM FUEL CELL SYSTEM COMPONENTS

Even though the PEM fuel cell has several advantages, compared to other conventional fuel cells. There are some technical challenges.

“So, to ensure that high performance is achieved, proper thermal and water management is required for the PEM fuel cell.”

  • The membrane used as an electrolyser must be kept at a specific humidity to allow for adequate hydrogen ion transfer.
  • If the membrane dries out, performance will be decreased, whereas when there is too much water content in the air, flooding will occur blocking the channels and preventing proton transfer.
  • To keep the humidity within a certain range has proven to be quite challenging. A humidifier is needed on the air supply line to increase the humidity as required.
  • A thermal management system is also added to extract the heat generated due to the chemical process at the cathode side.
  • If this heat is not removed, the fuel cell will heat up and damage the membrane, reducing the performance significantly.

So, to ensure that high performance is achieved, proper thermal and water management is required for the PEM fuel cell.

CAPABILITIES INCLUDED IN FLOWNEX

Flownex gives you several options and capabilities in modelling the system components and fuel cell stack of the PEM fuel cell.

Fluid capabilities

Flownex includes compressible gasses, two phase fluids, mixtures, etc. Custom fluids can be created from properties defined in literature. Higher incremented fluids or properties at different operating conditions can also be imported from NIST. The mass or mole fraction of mixtures can also specified and Flownex allows for changes in these fractions throughout the network. More important for the air side: Flownex allows for property calculation of humid air with detailed psychrometric charts available.

Chemical reaction models for the fuel cell stack 

The complex physics associated with the fuel cell stack can be included in the Flownex network. This allows for the analysis of how the entire fuel cell will behave at different conditions. Custom stack models can be implemented using scripting languages such as C#, Phyton and EES. Flownex can also be integrated with software packages such as Cantera and Matlab to include external stack models into Flownex. Reduced order models (ROMS) can also be imported using the Flownex FMI capabilities.   

With the ability of coupling Flownex to Ansys Fluent, the advanced fuel cell models in Fluent can be directly coupled to a network in Flownex. Ansys Fluent has specific addon modules to include the complex physics of the proton exchange membrane fuel cell, the solid oxide fuel cell and electrolysis. This means that the geometry of a fuel cell can be created, then imported into Ansys Fluent and the performance will then be calculated with these advanced fuel cell modules. The performance can then be coupled to Flownex using the Ansys Fluent coupling components or by importing a ROM exported from Fluent.

Pump and compressor models

Flownex includes a large library of turbos and pumps – including centrifugal and positive displacement compressors, centrifugal and positive displacement pumps, turbines, etc. Detailed compressor maps and pump charts can be included in these components.

Discretised heat exchange models

Flownex includes different heat exchange models, such as common geometries – finned tube heat exchangers, plate heat exchangers and so on. Flownex also allows for custom correlations for heat transfer to be included. Flownex also gives you the capability of coupling to Ansys Mechanical, allowing the user to integrate complex conduction problems into Flownex.

Typical pressure drop models

Pressure drop models are available such as pipes, valves, etc. The pressure drop models ranges from basic components to more detailed components to replicate real life components.

Steady state and transient solver

Flownex also allows for steady state as well as transient solving. Including an implicit transient solver which allows for very large timesteps for long transients.

Control models for transient simulations

Lastly, Flownex includes a large control library with analogue and digital controls such as PIDs, filters, switches, etc.

MODELLING THE PEM FUEL CELL

Pre-modelling setup

Flownex has a comprehensive list of standard components found in typical applications. Flownex also gives you the ability to create and use custom components.

Before modelling a network, it is custom to apply a background image on which the components can be placed. This will simplify the network building process and give the user a more structured approach.

Modelling the network

Flownex uses a drag and drop approach in the design of a thermal fluid systems. This user-friendly approach makes using Flownex easy for beginners.

Custom components

Custom components can be designed and used in Flownex. This allows the user to include non-standard components into the network.

One of the components that can be created using a custom component is the fuel cell stack. As mentioned previously, the physics of the fuel cell stack can be included using several methods. For this instance, a scripting component utilizing C# will be used in this example.

The script in the above fuel cell stack is used to calculate the hydrogen ion transfer rate from the anode (left) to the cathode (right), the stack current and voltage, heat generation, etc. The CEA Gibbs reactor calculates the chemical reaction of the hydrogen passed through the membrane and the oxygen in the air at the cathode side.

Anode and cathode side components

Custom components can be created for the humidifier as well as the condenser. The anode and cathode side components can then be dragged and dropped into the window.

The humidifier will ensure that the water content of the air stream is sufficient to prevent dry out of the membrane. The condenser is used to extract some of the water out of the stream which can then be introduced at the humidifier.

Adding thermal management

The heat being generated within the fuel cell can be extracted using a thermal management system. This can be quickly implemented in Flownex to ensure that this system is properly designed, preventing any damages to the membrane.

Adding control

A control system can be added to the network using components such as the PID controller. These components can be added to

  • Keep the fuel cell stack temperature consistent by changing the speed of the thermal management system’s fan.
  • Ensure the correct hydrogen flow rate according to the required power by changing the valve fraction opening.
  • Keeping the correct air to fuel ratio by changing the compressor speed at the cathode side.

In summary, Flownex is a tool that can be used by engineers to design, optimise, and evaluate thermal fluid systems. In the case of the PEM fuel cell where the thermal and water management of the stack is extremely important, Flownex provides a way of understanding the behaviour of these systems.

“It can be used to simulate the chemical reaction within a PEM fuel cell using different methods and allows for coupling with external software.”

Flownex gives you the ability to complete a full system design. It can handle any complex fluids and gives you the ability to include custom fluids. It can be used to simulate the chemical reaction within a PEM fuel cell using different methods and allows for coupling with external software. It includes a large library of components which ranges from basic to more complex models. It also gives you the ability to create your own custom components. Flownex also has numerous heat exchange models from basic to more advanced. It also includes various pressure drop models such as pipes, valves and so on. Flownex allows for steady state and transient solving allowing you to understand the behaviour of the system in its entirety.

Flownex is a useful tool for engineers in the designing and analysis process. As more emphasis is put on a cleaner zero emissions world, the design of greener energy solutions has become imperative. Using a systems simulation software such as Flownex will allow for rapid design of such systems, saving on cost and time.

Leander Kleyn

Leander Kleyn

Leander is a simulation design engineer at Flownex who specialises in Propulsion and Energy systems.

FLOWNEX SE 8.12.8 (2021 – UPDATE 1)

FLOWNEX SE 8.12.8 (2021 – UPDATE 1)
The latest Flownex® 2021 update brings you new components, more flexibility and increased user-friendliness as well as bug fixes.
 
Enhancements include a new Velocity PID component added to the Distributed Control System (DCS) library and improved steady-state solving. Components can now be managed with ease thanks to new features allowing search, multi-select description editing and link display. The Reactor Builder Script and associated Runtime Neutronic Script functionality have also been expanded to include TRISO particle modelling and the effect of Xenon poisoning during reactor transients.

LIST OF ENHANCEMENTS

MAJOR ENHANCEMENTS

Distributed Control System (DCS)

  • Added a Velocity based PID controller. The goal of the Velocity PID controller is to have a PID that calculates the change of output variables instead of the output itself. This way different controllers can easily be used to control the same variable and switching between them does not cause spikes in the output.
Fig. 1 - Velocity PID Controller
  • Added steady state operation for DCS components. The DCS components can propagate their inputs before or after steady state. The Execute before steady state option ensures that the control system values are initialised and passed to the flow components before the steady state solution is calculated. This results in a smooth change in controlled values when the transient solution starts. The DCS components can also be executed during steady state. In this case, the DCS components will execute as if a transient is being executed using the time step specified in the Scheduler (or Time Step settings). The Steady State behaviour settings for the DCS can be found on the DCS Solver, as seen in Figure 2. By default, the Execute before steady state option will be enabled for new projects and it will be disabled for older projects.
Fig. 2 - DCS Solver Steady State Behaviour Settings.

MINOR ENHANCEMENTS

Component Inputs

  • The handling of structural only rigids with the CAESAR II importer has been improved. They are now ignored and not imported at all.
Fig. 3 - Multi-Editing Component Descriptions and Using the Description to Search for Components using the Find Dialog.
  • A Search feature on the Component Selection dialog has been enabled. Input fields where users select a certain component as input, now shows a dialog with more filters to enable easier location of components in a large network. This works similar to the Find dialog.
Fig. 4 - New Search Feature for the Component Selection Dialog.Figure 4: New Search Feature for the Component Selection Dialog.

Drawing

  • The component connectivity can be shown by right clicking on any component and choosing “Display Links/Connections” from the context menu. This dialog has been updated to make it easier to understand and now also shows the port and hub names in the connections. The displayed port, fiber and hub names are useful when a user wants to determine which names to use when making connections to specific ports, fibers or hubs using the API.
Fig. 5 - Links and/or Connections Dialog.

API

  • The ability to specify port names when linking components using the network builder has been added. This is useful for linking components that contain many ports like the DCS controller components.
  • The ability to specify a hub name when connecting components using the API was added. This is useful for linking components that contain several hubs. Details of the new function can be found in the API Help manual.
  • Communication through the API with existing open Flownex instances has been improved. The same functions can now be used to control instances that have been launched by the API or other existing instances. This makes the code simpler, for instance the same Python functions can now be used with both.

Flow Solver

  • Added humid air fluid functions (psychrometric fluid functions) to Material Scripts that enables the calculation of mixture mass fractions from the following three input combinations:
    • Pressure, dry bulb temperature and relative humidity.
    • Pressure, dry bulb temperature and wet bulb temperature.
    • Humidity ratio.
  • Added an error for non-physical user-specified constant convection coefficients. The solver will not run when a negative heat transfer coefficient is specified.
  • Added an option to select between Effective area and Conduction shape factor specification for the Composite Conductivity element.
Fig. 6 - Effective area and Conduction shape factor options.

Scripts

Compiler warnings were not shown when compiling Scripts. For Script components, the warnings are now shown when using the Debug button. This provides users the opportunity to address any warnings that the Scripts may issue during compilation.

Installation Verification Tool

The installation verification tool has been updated to allow it to run several Flownex instances in parallel. This aids in both increased stability and speed. The user can select the number of parallel instances to use on the user interface. By default, it is set to half the number of available logical processors.

Nuclear Scripts

Reactor Builder Script Changes

  • Cavity flow paths have been updated to:
    • Allow for adjacent vertical cavities to generate separate vertical flow paths.
    • Allow for adjacent horizontal cavities to generate separate horizontal flow paths.
    • Allow the user to specify that adjacent single cavities must remain distinct. This feature can be used in conjunction with the above options, to allow separate connections to flow elements outside of the generated reactor network.
  • Implemented the associations between Ports and Node connection as a list to facilitate removing the previously hardcoded limit of 20 ports.
  • Changed the folder structure and location of feedback files from the project folder to a sub-folder \point_kinetics_inputs\<Reactor drawing page>. The point kinetic input files for a network shown in a page specified in the Reactor drawing page, are now located in the same sub-folder.
    • When the network is generated the first time the default point kinetics input files are copied to this subfolder. A new file, nominal_conditions.txt will be added to the existing list of input files. It will have the nominal power (in kW) as input with the rest of the file consisting of comments lines for the user to specify the nominal conditions associated with the set of point kinetic inputs contained in this sub-folder.
  • Removed the discretization options for the sphere; the option is now always equally spaced. This was done since internal grid independence studies showed superior grid independence for the equally spaced option.
  • A Tri-structural Isotropic (TRISO) model option was added, consisting of one node per fuel kernel, one node per composite coating, and a node on the interface between the fuel and composite coating (for a total of three nodes per TRISO particle). The discretization for this feature is described in the Reactor Builder Script and Runtime Neutronics Script User Manual in Appendix B, Section 5.2.
  • Feedback groups:
    • Removed spurious warning messages when generating a network with general fuel zone specified, that nodes have already been added to a feedback group.
    • Allow for numbered (arbitrary) feedback groups in addition to the named feedback groups.
    • Allow for an unlimited number of materials to be specified per feedback group.

Reactor Result Script Changes

  • Added a feature to specify a period for results file generation during a transient simulation.
  • Added tabular output for newly added TRISO particle results.
  • Added Pebble Wall Node results to Secondary Node result tables.
  • Addressed error in the flow element indexing that resulted in partial table printouts.

Runtime Neutronics Script Changes

  • Runtime Neutronics Script was updated to accommodate the generic/numbered reactivity feedback groups. When the numbered feedback group option is selected then:
    • The number of groups that are active needs to be specified by the user,
    • The feedback group that is associated with the moderator needs to be selected by the user.
  • Updated and properly documented comment lines in all sample point kinetics input files:
    • Added nominal_condition.txt file.
    • Updated comment in heat_distribution.txt to clearly indicate that the first two lines refer to the boundary of the zones and not the zone increments.
  • Added Xenon modelling capability—the model is described in the Reactor Builder Script and Runtime Neutronics Script User Manual in Section 3.2.
    • Xenon parameters are provided in a xenon_transient.txt input file.
    • The reactivity associated with the Xenon along with the normalized Xenon and Iodine concentrations is provided as transient outputs.
  • Nominal power is now read from the nominal_condition.txt input file and shown on the property page for ease of reference after the script has executed once.
  • The “update reference temperature” option has been changed to “Calculation mode” with two options:
    • Initialize/Nominal conditions: This option will update the reference temperature for the feedback groups and set the control rod reactivity to zero.
    • Off-nominal conditions: When this option is selected, the steady state power will become available as input and will be used in the steady state run. The reference temperatures will remain unchanged, and the control rod depth will be adjusted to offset the fuel/moderator/reflector reactivity that resulted from the non-nominal operation.
  • Moved displayed modelling parameters that are read from the point kinetics input files to a separate category (“Read from files in point_kinetic_input folder”) to clearly differentiate them from inputs that should be provided on the property page.

Select Flow Circuit

Added the ability to select all components on the drawing canvas that are part of a specific Flow Circuit (components that are connected and have the same assigned fluid).
This feature can be used by right-clicking on any component in the Flow Circuit and selecting “Select Flow Circuit”. All components in that Flow Circuit will then be selected and their common properties are displayed on the property grid.

Fig. 7 - Select Flow Circuit.

BUG FIXES

Compound Components

  • Fixed the problem where lists could not be exposed in compounds.
  • Fixed the problem where the result layers were not displayed for some compound components. These included components such as the Gradual Pipe Transition, Orifice in Pipe Transition, Abrupt Contraction and Expansion and Long Orifice. 
  • Fixed the problem where any fiber could connect to an exposed hub from a compound component. This caused a problem where invalid component connections could be made to the Angle Valve, Gate Valve, Globe Valve, Sharp Edge Diaphragm Valve, Sluice Valve, Smooth Diaphragm Valve and Y-Valve.
  • Fixed the problem where a large number of items in the Reactor Chart got duplicated when a user pressed Cancel and the chart was reloaded. This caused the chart file to increase in size every time a user pressed Cancel and kept many unused items in the lists.

Flow Solver Components

  • Removed the incorrect reporting of negative values being calculated for specific heat. This happened when the specific heat polynomial option was used for defining the enthalpy of the fluid, the latter which may be negative.
  • Added a warning when the static pressure of a two-phase fluid at inlet to an ANSI standard based Valve exceeds the critical pressure of the fluid. Under these conditions Flownex® applies the gas-phase equations, not the liquid phase equations.
  • Fixed the reporting of incremental areas on the Rotor disk of the Rotor-Stator component that erroneously reported the outer radius instead of the area.
  • Limited the participating gas partial pressure emissivity correction in the Leckner correlation for gas to surface radiation to zero in order to prevent calculating undefined numerical values.
  • A relaxation parameter was applied to the friction factor for very small non-condensable mass fractions that resulted in non-physical friction factors being calculated. The problem was addressed by removing the relaxation calculation.
  • The incorrect vapor mass fraction approximation was used for determining the specific heat for a two-phase fluid with non-condensable gas during the critical mass flow and static pressure calculation.
  • Addressed issue with Composite Heat Transfer (CHT) convection script indexing for CHT connected to flow nodes by adding an index indicating which index in the parallel path a convection element is associated with. This allows for correct indexing of the associated flow nodes in CHT element convection scripts.
  • When a CHT element is connected to a Node on one side and a sub-divided Pipe on the other, two convective elements were created per increment on the side connected to the Node. This resulted in the subdivided results being displayed incorrectly.
  • Fixed the problem where the Adiabatic Flame element added combustion product species that were not available in the gas mixture list to first fluid in list and not into background fluid.
  • Addressed the incorrect reporting of the pressure difference excluding elevation change result for the Primary side of the Shell and Tube Heat Exchanger. This pressure difference was incorrectly divided by the number of gas passes.
  • Removed the limitation that the CHT element can only be connected between two subdivided elements. CHT elements may now also be connected between a sub-divided element and a single node. The convection heat transfer to a flow Node and the conduction heat transfer to a solid Node is now correctly accounted for. Previously a warning was issued that this topology was not permitted, but the simulation was not stopped and non-physical results were reported.
  • Addressed the incorrect treatment of loss coefficients for sub-divided reactor zones. The user specified loss coefficients were assigned to each element associated with a sub-divided zone, whereas the inlet and outlet loss coefficients are now only applied to the first and last element in a sub-divided string of elements.
  • Fixed problem where two-phase tank level was temporarily set to zero when loading a snap.
  • Fixed precursor equation in solid and liquid fuel nuclear example scripts.

Excel Input Sheets

  • Fixed the problem where data was lost when a user performed a save-as operation on a project after editing Excel input sheets.

Licensing

  • Fixed the problem where some example Script components were available for use in the Basic Thermal Fluid module.

GIS Importing

  • When importing GIS files, the generation of background maps using some online sources did not function correctly anymore. The importing was updated to remove sources that do not work anymore and the speed of importing was significantly improved.

Initialize Steady State from Snap Files

  • The problem was fixed where the fixed mass flow value was used from the snap file and not from the user interface specification when initializing a steady state from a snap file.The problem was fixed where the fixed mass flow value was used from the snap file and not from the user interface specification when initializing a steady state from a snap file.

Spaceflight

BLOG | 30 August 2021

AN INTEGRATED SYSTEM MODEL OF THE RS-25 LIQUID-FUEL CRYOGENIC ROCKET

“Capable of fast transient and steady state solving times, Flownex affords users the ability of rapid system design iterations.”

Share on linkedin
Share on facebook

Space exploration and development in the 21st century

With the commercial interest in space exploration, from space tourism and mineral exploration to the need for space colonisation, it has become evident that the development of space craft and propulsion systems need to be done at a faster pace than ever before. Several rocket engine designs have been attempted with different types of fuels in the last decade with the main goals being:

  • increased efficiency,
  • reduced cost,
  • low operational risk,
  • and re-use.

“Flownex is the industry leader in 1D CFD software, allowing for the systems simulation of large interconnected thermal fluid systems.”

It has become Increasingly important to evaluate different cycles as fast as possible, however, when all the complexities of the rocket engine designs are taken into consideration, 3D CFD analysis on a system scale is not a viable option. Simulating the system components of the rocket engine in its entirety is important to understand how all these components will interact with each other and how the system will perform at different operating conditions. Parametric studies need to be performed to find optimal design conditions, but these can be tedious and time consuming.

This is where Flownex has gained popularity in the aerospace industry. Flownex is the industry leader in 1D CFD software, allowing for the systems simulation of large interconnected thermal fluid systems. Capable of fast transient and steady state solving times and introducing the ability for rapid system cycle design iterations, Flownex has been used extensively for R&D purposes where 3D CFD just cannot compete. Flownex uses fundamental physics and theory from leading research to solve some of the world’s most complex thermal system problems.

Modelling the RS-25 rocket engine

Today we look at the system simulation of one of the more complex rocket engines available, the Aerojet Rocketdyne RS-25. Used exclusively in the past as the 1st stage propulsion system for NASA’s Space Shuttle and now the Space Launch System (SLS), this rocket uses liquid cryogenic hydrogen and oxygen as the fuel and oxidiser. Rated at around 491,000 pounds of thrust at its nominal power level – it packs a punch.

“It is crucial then, to simulate the entire system as one to capture the interdependencies between components.”

Figure 1: RS-25 rocket engine (NASA: Sept 13, 2013).

Engine specifications

  • The RS-25 engine is a closed-loop engine utilising fuel-rich combustion in two pre-burners driving the turbopumps used to pump the hydrogen and oxygen sides.
  • Liquid cryogenic hydrogen is pumped through a low-pressure and high-pressure pump, then distributed through pipe networks in the nozzle and main combustion chamber.
  • The branches flowing around the nozzle removes heat from the nozzle material and is then directed to the pre-burners.
  • Another branch passing beside the main combustion chamber is also used for cooling and to then drive the low-pressure turbopump.
  • A portion of the liquid oxygen is branched to the pre-burners and the rest to the main combustion chamber where it mixes and combusts with the fuel rich products from the pre-burners.
  • The thrust is controlled using the two pre-burner oxygen valves. Adjusting these valves will alter the oxygen flow rate, changing the state of the combustion products, altering the power of the turbopumps which then changes the pumping rate of the fuel and oxidiser from the tanks and ultimately changes in thrust.

It is crucial then, to simulate the entire system as one to capture the interdependences between components.

Figure 2: RS-25 diagram (Wikipedia: RS-25).

Modelling the engine from scratch

To start modelling a thermal hydraulic system, it is best to work from a diagram or schematic of the network. This will structure the modelling process and ensure that the entire system and the necessary physics processes are covered. In Flownex, a background image can be imported onto the drawing page by simply right clicking on the clean canvas and selecting “edit”. From there, the “Style” can be selected using the three dotted button next to it and following the instructions:

Once the background image is selected, components can be dragged and dropped in from the components library.

The components library contains a large selection of components which can be used to simulate:

  • The piping system – realising pressure drops from primary and secondary losses,
  • turbopumps – with power matching capabilities,
  • convection and radiation heat transfer in the combustion chamber, nozzle, etc.

It is important to work from a point (such as the inlet of hydrogen) and model the components step-by-step while solving, as the model is progressed. This will ensure that any errors introduced during the modelling phase are picked up and rectified before carrying on.

Flownex also gives you the capability to import fluid properties for non-standard applications (such as cryogenic states) directly from the NIST database, REFPROP, with a level of detail chosen by the user. This is very useful especially if the fluid’s being used within the network are operating in ranges outside of those included in the standard fluids library.

The entire Flownex network of the RS-25 rocket engine is shown in the following image as overlay on the system diagram:

Some of the important Flownex library components used in the RS-25 model are explained here:

Turbopumps

These are modelled using the basic centrifugal pump in combination with the simple turbine. The shaft component can be used to apply power matching between the pump and turbine – matching the power output of the turbine with the input power of the pump.

Heat transfer

The heat transfer from the combustion process in the nozzle and main combustion chamber can be realised using the composite heat transfer element. This element is capable to model the combined radiation, convection, and conduction heat transfer from the combustion process.

Combustion process

Flownex includes a CEA adiabatic flame component, which allows for combustion calculations and adiabatic flame temperatures to be obtained using NASA’s Chemical Equilibrium with Applications library. The combustion process in the pre-burners is fuel-rich and the excess hydrogen is burnt in the main combustion chamber after the addition of more oxygen. The exit thrust nozzle is used to calculate the thrust produced by the rocket, while the PID controller component controls the correct ratios in the branches.

“Flownex opens a world of opportunities for system level design, saving time and development cost.”

The RS-25 rocket engine is highly complex and the performance of different components influences each other. For this reason, such a system should be simulated as an integrated system of interdependent components.

Component optimisation

Flownex can be used to optimise individual system components and their interaction with the entire system.

Transient simulations

Flownex allows for transient simulations, giving the user the ability to analyse changes in the system over time including thermal inertia. Using the transient capabilities of Flownex, this network can be used to simulate the performance of the RS-25 during start-up, lift off and flight through the earth’s atmosphere.

Flight mission simulation

This model can be coupled to a full kinematic model to realise the drag force during flight (with changes to the drag as the density of air changes with elevation) and the mass change of the spacecraft as the fuel and oxidiser are consumed. The acceleration and velocity of the rocket can be calculated, and a full flight mission can be simulated for different scenarios.

The above example shows how the transient simulation capabilities of Flownex are used to analyse the system response when introducing changes in the system. For the above demonstration, the oxygen pre-burner valves are adjusted as well as the exit pressure. In this way, the engine performance can be monitored for different throttle percentages and different elevations (nozzle exit pressures).

Flownex opens a world of opportunities for system level design, saving time and development cost. A 1D CFD code can be used to solve complex thermal hydraulic systems beyond the reach of what can be done in a similar timeframe with 3D CFD, which is more appropriate for design of individual components and will compliment a 1D workflow. Flownex is used extensively in the aerospace industry to design and optimise systems and to gain a deeper understanding of the overall performance at different operating conditions.

Leander Kleyn

Leander Kleyn

Leander is a simulation design engineer at Flownex who specialises in Propulsion and Energy systems.

Share:

Share on linkedin
Share on facebook

FLOWNEX SE 8.12.7 (2021)

FLOWNEX SE 8.12.7 (2021)

The latest release of Flownex® for 2021 comes with much anticipated enhancements and fixes to ensure that you can do even more with real world systems simulation.

Some of these include the option to initialise the steady state solver from a previously saved snap, specifying transient actions through tables, python link functionality. With the support of Adobe Flash having come to an end we have updated our video tutorials to now work with HTML5.

LIST OF ENHANCEMENTS

MAJOR ENHANCEMENTS

Start Steady State from Previous Conditions

A steady state solve can now be initialised from conditions from any saved snap. The user can use this to initialise a steady state solution from a previous steady state solution or from a transient run.

The setting can be activated on the Flow Solver Properties Inputs window, where the related snap can be selected, as seen in Figure 1.

Fig. 1 - Initialise Steady State from Snap Option in Flow Solver Inputs

Action Specifications via Tables

A table can now be used to specify the value of an input at a specified time in an action. The table allows the user to copy and paste data from Excel. Values will be extrapolated if the time values fall outside the range specified in the table.

Fig. 2 - Table Input Option in Actions

Video Tutorials

The video tutorials in the Flownex® startup page that used Adobe Flash has been replaced by HTML5 videos. This was necessitated by Adobe not supporting Flash since 1 January 2021.
Fig. 3 - Flownex Video Tutorials

MINOR ENHANCEMENTS

Caesar II Link

  • The handling of structural only rigids with the CAESAR II importer has been improved. They are now ignored and not imported at all.

Global Parameters

  • The way global parameters are handled with snaps has been improved. The values will be read in from a snap. The number of parameters and the number of associated properties (as well as which properties are associated) will not be changed when reading in a snap.

Python Link

  • The Python API files now contains a function that allows the user to run the Flownex® Designer using Python.

Finned Tube Heat Exchanger

  • Added an improved default Finned Tube Heat Transfer calculation example script for the Finned Tube Heat Exchanger – Fin Side.
  • The description of the input “Tube pass length” has been changed to clarify that this input represents the total primary flow path length.

License Borrowing

  • The maximum number of days a license can be borrowed, as specified in the server license file, is now indicated on the Borrow License dialog, rather than the default 28 days.
Fig. 4 - Number of Days to Borrow License from Server

Energy Solver Convergence Enhancement

  • The warnings associated with node temperature changes were enhanced to facilitate the identification of the non-converged node.
  • An option was added to the Flow Solver settings to exclude nodes with temperature changes as a result of very small mass flows in its associated flow element.
Fig. 5 - Apply Mass Flow Rate Threshold to Transient Temperature Change Residual

BUG FIXES

Excel Reporting

  • Fixed the problem where the option to “open in Excel” sometimes opened multiple copies of the Excel template.

Parameter Display

  • Fixed the problem where the parameter display did not work correctly during transient, which sometimes caused the application to terminate.

Heat Transfer Correlations

  • Fixed the problem where a warning was shown when the Gnielinski correlation was used that stated that the roughness used in the correlation is different from the Pipe roughness, even when the user specified the option that the roughness from the Pipe should be used.

Isentropic Head Compressor

  • Updated the maximum number of constant speed curves (61 to 101), data elements in a row (61 to 101) and constant blade angle maps (31 to 51) for the Isentropic Head Compressor Chart.

Variable Speed Pump

  • Fixed the pump shut down speed option that did not execute correctly.
  • Fixed the unit problem with pump shut down speed.

Slurry Flow

  • Fixed the vertical slurry frictional pressure drop calculation, which used the mixture density instead of fines density.

Bend

  • Changed the two-phase Bend gross secondary loss factor to display results without multiplication factors.

Flow Path Graph

  • Fixed the bug where the unit of a plotted line item is not removed when the Property is changed to a unitless property.

FLOWNEX SE 8.12.6 (2020 – UPDATE 2)

FLOWNEX SE 8.12.6 (2020 – UPDATE 2)

The second update of Flownex® SE for 2020 further expands the possibilities of simulating real-world systems.

Exciting features for this update are a reworked deep solver coupling with Ansys Mechanical for explicit transient co-simulation and automated Nuclear Reactor building script along with many minor enhancements described below.

LIST OF ENHANCEMENTS

MAJOR ENHANCEMENTS

Ansys Mechanical Coupling

The Ansys Mechanical Flow Solver Coupling component has been improved to allow the simulation of complex 3D conduction and stress in Ansys Mechanical coupled to a flow and heat transfer simulation in Flownex®. This enhancement includes the addition of a deep solver coupling between Flownex® and Ansys Mechanical, allowing data exchange between iterations, full transient co-simulation functionality and unit integration.
Figure 1: Ansys Mechanical Coupling

Nuclear Reactor Builder Scripts

Script Generated Reactor Results

Results in tabular form corresponding to the grid layout of the generated reactor network are written to a text file in the ScriptResults subfolder of the project folder. The tabular results are reported at the end of a steady state run as well as during user specified times and at the end of transient runs. A feature is provided to specify output variables that are reported at every time step. These results are written to a comma separated values (csv) file as a time series residing in the ScriptResults subfolder.

Reactor Generation Script

Various improvements were made to the geometric parameter calculations. Network topology generation for the transitions between different types of reactor zones was extended.

Figure 2: Reactor Generation Script Enhancements

MINOR ENHANCEMENTS

FMI

The “Description”, “Minimum” and “Maximum” values have been added to the imported FMU variables. The description will be displayed if it is not blank. Minimum and maximum values will be displayed if they are specified in the FMU and the “Display variable information” property is turned on. A warning is given if the minimum or maximum values for inputs are exceeded.

Figure 3: Description, Minimum and Maximum Values added to Imported FMU’s

Licensing

The License modules selection have been updated so that it can work for users that do not have write access to their ProgramData folders.

Scripting

  • Functions to retrieve minimum and maximum temperatures and pressures, as well as critical temperatures and pressures from two-phase fluids via Scripts have been added.
  • Added the ability for Scripts to properly work with lists and Snaps – functions were added that get called in the Script before and after loading and saving Snaps. This gives users flexibility to implement their own Snap saving and loading code.

Composite Heat Transfer Component

The “Area multiplication factor” input for the Composite Heat Transfer (CHT) component has been moved to a separate category to indicate that the input is applied to all surfaces.

Figure 4: Area Multiplication Factor Moved to Separate Category

BUG FIXES

Licensing

Fixed the problem where many of the modules were missing when using a borrowed license from the server.

Find Dialog

Fixed the problem where the Find dialog did not open the property page for components on closed pages (it appeared that multi edit did not work).

Graphs

Line Graph: Removed trailing list separator from the ‘Save As CSV’ file rows.

Scripts

  • Fixed the Solid Properties script, as the script did not compile correctly due to the enthalpy function that was removed.
  • Fixed the problem where properties for two-phase non condensable mixtures returned zero values as results in the Mixed Fluid Properties Script.

Solver Results

Fixed the problem where the Fluid volume and Fluid mass results in the Flow Solver Results window did not update when it was linked to an Action or displayed on the canvas.

Composite Heat Transfer Component

  • Fixed the problem where an error was issued when connecting the Composite Heat Transfer element to non-pipe flow elements because the roughness could not be determined when the Dittus-Boelter option was specified.
  • Fixed the Composite Heat Transfer element StPr chart option not allowing the user to specify the flow area.
  • Fixed the Reynolds number result on the Composite Heat Transfer element for StPr chart input option not displaying correctly.

FLOWNEX SE 8.12.5 (2020 – UPDATE 1)

FLOWNEX SE 8.12.5 (2020 – UPDATE 1)
The new Update to Flownex® SE for 2020 expands the possibilities of simulating real-world systems. Some exciting features included with this release include: built-in functionality to generate system resistance curves within seconds, force calculations for pipe sections, interfacing with CAESAR II for detailed pipe stress analyses, a co-simulation link to the latest version of 6SigmaDCX, along with many other improvements that are listed below.

LIST OF ENHANCEMENTS

MAJOR ENHANCEMENTS

System Resistance Graphs

The capability has been added to easily plot system resistance graphs. At the click of a button, a parametric run is automatically configured and executed providing the user with an accurate system resistance graph in seconds, as seen in Figure 4. The system resistance graph can be exported as a CSV file in a few clicks and sent to a manufacturer for pump selection. The system resistance graph can also be plotted on a pump chart allowing the user to quickly determine operating points at different pumps speeds.
Figure 3: System Resistance Graph added to Operating Point Plots.
Figure 4- System Resistance Curve and Pump Curve Plot

Force Calculations for Piping Sections

It is not possible to perform structural pipe analysis for water hammer scenarios in most real world systems using hand calculations due to the complex nature of the pressure wave reflections. Flownex® already provides the capabilities to simulate fast transients such as water hammer in these complex systems. In this release, enhancements have been made to the axial pipe force calculations to make them valid for all steady state and transient simulations. This allows users to easily simulate the pipe forces in Flownex® and export the results to structural codes such as ROHR 2 and CAESAR II. Pipe sections for net force calculations can also easily be defined in the “Force Calculation Piping Sections” dialog that is available under the Results menu.
Figure 5. Calculated Pipe Forces in Flownex
The enhanced force calculations are applicable to Pipes, Bends, Valves, the British Standard Orifice, Secondary Loss and the General Empirical Relationship components.

CAESAR II Integration

Flownex® provides a very easy to use interface to work with CAESAR II. By using this interface Flownex® can calculate dynamic loads for pipe stress simulations for water hammer cases or pressure waves. The interface allows a user to import the geometry for a piping system from CAESAR II directly into Flownex®. This saves a user time and also eliminates possible errors that could occur when users need to manually duplicate piping systems in Flownex®. Flownex® also provides an intuitive way to define the piping sections for which users wants the net forces to be calculated. These calculated forces can be automatically exported in a time series that is easily imported into CAESAR II.
Figure 5- Imported Pipeline and Calculated Pipe Forces in Flownex.
Figure 6- Forces Imported to CAESAR II.

6Sigma Link

6Sigma is a world leading tool for data centre simulation and a link to 6Sigma has been added. The link allows users to quickly setup combined simulations and allows both steady state and full transient simulations. Typically, a detailed model of the inside of a data centre can be connected to a complete external cooling model in Flownex®, where all the cooling towers, pumps, heat exchangers etc. is modelled in detail. This allows a complete system simulation that is not available in other software and opens up a whole new spectrum of possible efficiency improvements in both design and operation mythology.
Figure 7- 6Sigma Link Example.

Nuclear Reactor Building Scripts

The ability has been added to build a nuclear reactor using a script. This script reads information from a reactor geometry chart and builds a corresponding reactor. The reactor is built on a separate page in the user interface. Users have access to all of the elements and nodes that defines the reactor.

There are several advantages using this capability. The first advantage is that all the inputs to the elements and nodes that build the reactor can be verified. Furthermore, all internal results are available and finally, the internal connectivity and structure can be manually modified by users if needed.

 Example scripts are available on request. Some components have been added to aid in the reactor building, which include a Porous Flow Element and a Composite Conduction component. They are found in the Reactor Building Blocks category in the Nuclear library, as seen in Figure 9.

Figure 8- Generated Reactor using the Nuclear Reactor Building Scripts.
Figure 9- Composite Conductivity and Porous Flow Element added to Nuclear Library.

Relap Component

The Relap component has been updated to work with newer versions of Relap. Newly updated examples are available on request.

Angled T-Junctions and Y-Junctions

The existing junction functionality has been enhanced to include the ability to model angled T-junctions and Y-junctions, as seen in Figure 1.

Many of the junction losses as defined in “Internal Flow Systems, 2nd Edition” by D. S. Miller have been implemented, thereby extending the previous functionality from the limited perpendicular junction options. These junctions are available for selection from pre-defined junction types, as seen in Figure 2.

Figure 1: T-Junction and Y-Junction Components in Flownex®.
Figure 2: Junction Branch Angle Options for Converging T-Junctions.

Mathcad Component

A new Mathcad link has been added to work with Mathcad Prime version 4.0, 5.0 and 6.0.
Figure 10- Mathcad Prime Link

Graph Improvements

Graphs that have been disabled, now shows a red cross on them to easily identify disabled graphs, as seen in Figure 11. The “Save As CSV” option has been added to all graph types and is available on the context menu of a graph, as seen in Figure 12.
Figure 11-Disabled Graph in Flownex®.
Figure 12- “Save As CSV” Option added to Graphs.

Licensing System

The licensing system has been updated to Version 14. Subsequently all users using server licenses will need to install Version 14 of the license server. An installer that does the upgrade is available to download from our website or from Support. There are several fixes in the newer version of the license server. More information about the fixes is available on the RLM website.

Trace Elements

The trace element modelling capability has been significantly expanded. Previous versions only allowed for homogeneous mixing of trace elements on nodes, where users now have the following additional capabilities at their disposal:

  • The ability to filter trace elements from the network.
  • The ability to specify selective throughflow of trace elements between nodes, thus resulting in non-homogenous mixing of trace elements within nodes.
  • The ability to specify trace element sources and sinks on nodes without the precondition that they enter or leave the system via mass sources or sinks defined for the carrier fluid.
  • Modelling trace element decay during transients using the trace element decay constant.

Filtering and selective throughflow of trace elements are specified on flow element components and sources or sinks and decay are specified on node components, as seen in Figure 13.

Figure 13- New Trace Element Input Options on Nodes.

Isentropic Head Compressor

A new compressor type has been added specifically for modelling compressors operating near the critical point of the fluid with significant changes in fluid properties such as the specific heat, making the use of conventional gas flow dimensionless parameters less accurate. The Isentropic Head Compressor uses isentropic head vs. volume flow data at different speeds. Given the volume flow and speed, the isentropic head is interpolated from the characteristic curve, after which real gas entropy tables are used to find the corresponding pressure.
Figure 14- Isentropic Head Compressor in the Turbos and Pumps Library.

MINOR ENHANCEMENTS

Video Tutorials

The video tutorials can now be played with a built-in player in Flownex® eliminating problems of browser compatibility. Adobe Flash player needs to be installed for the player to work.

Result Layers

For gradient result layers, components were not coloured when they had properties smaller than the minimum value or larger than the maximum value. A new option has been added namely: “Gradient <-[MinValue, MaxValue]->”, as seen in Figure 15. With this option the components with properties lower than the minimum value is painted the minimum colour and components with properties higher than the maximum is painted the maximum colour. This option is now set as the default option.
Figure 15- Gradient Option in Result Layers.

Property Grid

Disabled input properties did not allow users to change their units. This has been changed since these fields display valuable information which users may want to view in different units, as seen in Figure 16.
Figure 16- Units Can be Changed for Disabled Properties.
Figure 17- Toggle Button for Properties.

Screenshot Preview

The screenshot preview that displays in Windows Explorer when the preview pane view is activated, has been updated. In the past, the entire Flownex® window area was captured and sometimes it included overlapping windows. Now only the main item inside Flownex® is captured in order to show the most relevant view of the project, as seen in Figure 18. 

Figure 18- Screenshot Preview.

Psychrometric Boundary Condition

The “Not specified” option as a Boundary condition type, has been added to the Psychrometric Boundary Condition. This option has been added so that the specified condition can be unfixed during a transient simulation.
Figure 19- Not Specified Option for the Psychrometric Boundary Condition.

Heat Transfer

The following enhancements has been made to the heat transfer components:

  • Implemented the option to calculate conduction area from the circumference of the connected pipe.
  • Made Kugeler-Schulten correlation available for the Convection element.
  • Implemented second order convection calculation for a Convection element connected to a Node. This allows convection calculations to be done using the mass weighed average upstream temperature of the flow elements connected to the node that the convection calculation is performed on. This functionality should result in faster grid independence for subdivided flow fields such as those used in a reactor geometry.
  • Implemented dispersion for porous flow elements to account for the enhance heat transfer resulting from the disruption of flow in porous media. The dispersion is modelled as increased diffusion heat transfer within the fluid.
  • Implemented length over diameter warning for Dittus-Boelter and two-phase flow applications.
  • Added the Overall convection heat transfer coefficient and Convection coefficient as results for Convection elements.

Nuclear

The following enhancements has been made to the Nuclear components:

  • Second order convection can be specified in the reactor chart.
  • Dispersion can be turned on in the reactor chart.
  • An error will be given when not all reactor ports that are defined in the chart are connected.
  • Added gamma_f0, gamma_m0 and gamma_x0 to the fuel reactivity, moderator reactivity, and Xenon reactivity equations to allow the user to specify a constant offset independent of the prevailing temperatures. 
  • Changed “Cross section x neutron flux” input in die neutronics chart to “Xenon cross section x neutron flux” to clarify the use of the input.
  • Implemented warning that normalized control rod insertion depth when upper or lower limit has been reached.
  • Updated “Reactivity” result description to “Control rod reactivity” to clarify its meaning.
  • Added warning if neutronics parameters such as Normalized Power, Normalized concentrations of neutron precursor isotopes, Normalized concentrations of decay-heat producing isotopes, Iodine concentration or Xenon concentration go negative and are limited to zero.
  • Changed point kinetics error condition to be issued if fission power equal to zero and reactivity greater than the decay neutron fraction, Beta.
  • Updated Kugeler-Schulten calculations to interpolate between Nusselt numbers calculated at Reynolds number = 100 and Nusselt numbers of 4 (at Reynolds number = 0) for Reynolds numbers smaller than 100 to correctly reflect the documented range of applicability of the Kugeler-Schulten correlation.

Nodes

Implemented the ability to specify a solid material volume for “Solid Nodes”, as seen in Figure 16. This provides the ability to account for the thermal mass of solid nodes in an all solid heat transfer network such as those generated by the nuclear reactor model generating script.
Figure 20- Specify Node Solid Volume

Materials

Exposed fluid mixture component count so that it can be used in a script.

Heat Exchanger Component

The Effectiveness input on the Heat Exchanger Primary component, as seen in Figure 21, is now a dynamic input and can be changed during transient simulations. An option has been added to change the two-phase region error that is given when the heat exchanger operates in the two-phase region to a warning, instead of an error. This allows users to use the heat exchanger in the two-phase region if required. The new option can be seen in Figure 22.
Figure 21- Effectiveness Input on Heat Exchanger Primary Component.
Figure 22- Treat Heat-Exchanger in Two-Phase Errors as Warnings Option in Flow Solver.

Turbine

Turbine chart scaling factors are now a value of 1.0 by default on new charts, as seen in Figure 23.
Figure 23- Turbine Chart Scaling Factors for New Charts.

Mixtures

The following enhancements has been made to fluid mixtures:

  • The capability has been added to specify a mass sink with mass fractions specified to nodes that are internal to a network. This allows for the selective removal of fluid components from a fluid stream to model the action of a filter or membrane. This functionality is not available on boundary or edge nodes as the transfer of mass is an advection problem and boundary conditions cannot be propagated in the opposite direction to the flow.
  • Psychrometric results are calculated for all two-phase non-condensable mixtures featuring Water as the two-phase fluid.

Scripting

The amount of iterations during steady state that is solved before iterative scripts or data transfers start being executed can now be specified in the Flow Solver settings, as seen in Figure 24. The default value is 6 – meaning that the scripts and iterative items will start executing at iteration 7 of the pressure solver (main iterations).
Figure 24- Iterative Scripts Calculations and Data Transfer Settings in the Flow Solver.

API

The API now provides users with the ability to create copies of components and links. Several functions have been added to the NetworkBuilder interface in order to facilitate this. These functions are documented in the API help file. Examples of how to use these functions have been added to the “NetworkBuilderScripts” demo project located under Demo Networks on the Flownex® Start Page. The Python example “3. Simple Network Builder” has also been updated to show how to use these functions. This example is available in the Help menu under Python Link.

Ducting

Added the capability to specify an angle for the Louver parallel and opposed 3V blades damper types.

BUG FIXES

General

Fixed problems where Flownex® crashed when a user deleted a project and its sub directories in Windows Explorer, after the project was closed but Flownex® still open.

Container Interface

Fixed a bug where flow still occurred even when the level of the Container Interface was at zero. 

Heat Transfer

Limit the emissivity on the Surface Radiation component to values between 0 and 1. 

Iterative Scripts

Fixed the problem where Iterative Scripts did not start running until a certain convergence has been reached.

Nodes

Fixed the problem where the Temperature guess value specified on a mass sink results in a flow solver error.

Pipe

Fixed the bug where the K Calculation dialog sometimes showed inputs from a different pipe.

Reporting

The Steam/Water trap was not added to a default report, this has been fixed.

Results Layers

Fixed the problem with painting Result Layers when a user has disconnected links, as parts of the network was not shown at all.

Shell and Tube Heat Exchanger and Finned Tube Heat Exchanger

Fixed the Pressure Drop Excluding Elevation Result, as the result changed with number of tubes specified in parallel.

Download full release notes here.

FLOWNEX SE 8.10.0 (2019)

FLOWNEX SE 8.10.0 (2019)

Flownex® SE 2019 is pushing the boundaries of thermal fluid system simulation.

We’ve added a few new enhancements and additions, including improvements to our heat exchangers, a custom vortex as well as a whole new appearance and functionalities to our graphs. We’ve also updated a few of our components to allow liquid-gas mixture fluid types.
A NEW student version 2019 is also available to download. 

LIST OF ENHANCEMENTS

MAJOR ENHANCEMENTS

Adaptive Timestep

Fig. 1 - Adaptive Time Step in Scheduler
An adaptive timestep functionality has been added to the Flownex® solver that automatically refines the timestep size through a transient simulation. This results in small timesteps when fast transients (such as pressure pulses during water hammer) are occurring to accurately predict the solution and larger timesteps when possible to effect shorter solving times. This feature monitors the pressure, energy, mass flow and density of all the components and will automatically reduce the timestep to ensure that the solution remains within the specified accuracy criteria. This allows the user to accurately predict fast transients such as pressure pulses without having to perform a temporal convergence study first. For more information about how the adaptive timestep algorithm is implemented, please refer to the Scheduling chapter of the General User Manual. A button has been added to the toolbar that gives the user quick access to the time step settings. It is located next to the Reset Time button in the Simulation Control section, as seen in Figure 2.
Fig. 2 - Time Step Settings

Cavity Editor

The inputs of the Rotor-Rotor and Rotor-Stator components have been significantly enhanced in order to allow a user to easily specify a complex geometry for the cavity.

The complex geometry can be specified by using the Cavity Editor, which opens when double clicking on a Rotor-Rotor or Rotor-Stator component. The Cavity Editor allows the user to import a background picture for the cavity. The geometry and dimensions can then be defined on the picture in the Cavity Editor, as seen in Figure 2.

Fig. 3 - Rotor-Stator Cavity Editor - Reference Measurements

After a picture has been imported, the user can define the dimensions of the cavity by specifying two points at any location on the drawing. Thereafter, the rotor and stator surface geometries are easily drawn on top of the picture.

Fig. 4 - Rotor-Stator Cavity Editor - Rotor Surface Geometry

Other geometric items like the position of bolts, gap and shroud width, as well as defining the discretization is also done easily using this Cavity Editor.

Fig. 5 - Rotor-Stator Cavity Editor - Discretisation

Custom Vortex

Fig. 6 - Customer Vortex Component
A Custom Vortex component has been added to the Rotating Components Library. The custom vortex is a vortex model commonly used in gas turbine cavity modelling. The tangential velocity is specified to produce a velocity profile between that of a forced vortex and free vortex. The radial velocity profile is specified in the following format Cθ= s⋅rn. A custom vortex is characterised by a swirl constant, s, and a vortex weighing factor, n. The custom vortex model provides a simplified cavity model that allows the user to adjust the swirl constant and vortex weighing factor to match the swirl pressure rise seen in empirical measurements.

Heat Exchanger Improvements

The Heat Exchanger components in Flownex® has been updated. These components are now easier to use and a few essential features have been added. The heat exchangers that has been updated is the Shell and Tube Heat Exchanger, Finned Tube Heat Exchanger and the Recuperator, which has been renamed to a Plate Heat Exchanger. The changes make using the heat exchangers for a general application like radiators etc. simpler. Furthermore, fouling factors and fin efficiencies have been added to the heat exchangers where relevant. These can be used to model degradation over time and changes in the condition of the heat exchangers.

Shell and Tube Heat Exchanger

Cosmetic Changes

New icons have been added for the Shell and Tube heat exchanger and the names now clearly indicate the shell side and the tube side.

Fig. 7 - Shell & Tube Heat Exchanger
Input Changes
Shell Side Primary Loss Calculations

The shell side supports specification of the friction factor through a constant value, using a script or using a Fanning friction chart. By default, the Fanning friction factor chart is used. The user can however easily use a correlation from another source in the script defined friction factor specification. These options can be seen in Figure 8.

Fig. 8 - Shell Side Primary Loss Options
Shell Side Heat Transfer Coefficient Calculation

The shell side now supports built in correlations for the heat transfer coefficient calculation, as well as script defined heat transfer coefficient calculation, constant value specification and Stanton Prandtl charts. By default, the Shell Side heat transfer coefficient calculation correlation is used. The user can however easily use a correlation from another source in the script defined heat transfer coefficient calculation, as seen in Figure 9.

Fig. 9 - Shell Side Convection Coefficient Options
Tube Side Primary Loss Calculation

The tube side supports specification of the friction factor through a constant value, Darcy Weisbach correlations, using a script or using a Fanning friction factor chart. By default, the Darcy Weisbach correlation is used, as seen in Figure 10.

Fig. 10 - Tube Side Primary Loss Options
Tube Side Heat Transfer Coefficient Calculation

The tube side supports built in correlations for the heat transfer coefficient calculation, as well as script defined heat transfer coefficient calculation, constant value specification and Stanton Prandtl charts, as seen in Figure 11. By default, the Gnielinski correlation is used for the tube side heat transfer coefficient calculation.

Fig. 11 - Tube Side Convection Coefficient Options
Finned Tube Heat Exchanger
Cosmetic Changes

New icons have been added for the Finned Tube heat exchanger and the names now clearly indicate the fin side and the tube side.

Fig. 12 - Finned Tube Heat Exchanger
Input Changes

Fin Side Geometry

A simplified set of inputs has been added to specify the geometry of a rectangular finned tube heat exchanger with round fins. This is the default option now, as seen in Figure 13.

Fig. 13 - Rectangle HX with Round Fins Inputs

The user now specifies more readily available geometric parameters like the heat exchanger width height and length as well as tube and fin diameters. The older more generic specification is still available.

Fin Side Primary Loss Calculation

The fin side supports specification of the friction factor through a constant value, using a script or using a Fanning friction factor chart. By default, the Fanning friction chart is used. The user can however easily use a correlation from another source in the script defined friction factor specification.

Fin Side Heat Transfer Coefficient Calculation

The shell side now supports script defined heat transfer coefficient calculation, constant value specification and Stanton Prandtl charts. By default, Stanton Prandtl chart is used. The user can however easily use a correlation from another source in the script defined heat transfer coefficient calculation.

Tube Side Primary Loss Calculation

The tube side supports specification of the friction factor through a constant value, Darcy Weisbach correlations, using a script or using a Fanning friction factor chart. By default, the Darcy Weisbach correlation is used.

Tube Side Heat Transfer Coefficient Calculation

The tube side supports built in correlations for the heat transfer coefficient calculation, as well as script defined heat transfer coefficient calculation, constant value specification and Stanton Prandtl charts. By default, the Gnielinski correlation is used for the tube side heat transfer coefficient calculation.

Plate Heat Exchanger
Cosmetic Changes

The Recuperator heat exchanger has been renamed to the Plate Heat Exchanger, which describes the functionality of the heat exchanger better. New icons have been added for this heat exchanger too.

Fig. 14 - Plate Heat Exchanger
Input Changes

Primary & Secondary Side Primary Loss Calculations

Both sides support specification of the friction factor through a constant value, Darcy Weisbach correlations, using a script or using a Fanning friction factor chart. By default, the Darcy Weisbach correlation is used with the addition of friction factor multipliers that can be used in the laminar and turbulent ranges to adjust the friction factor.

Primary & Secondary Side Heat Transfer Coefficient Calculation

Both sides support built in correlations for the heat transfer coefficient calculation, as well as script defined heat transfer coefficient calculation, constant value specification and Stanton Prandtl charts. By default, the Gnielinski correlation is used for the tube side heat transfer coefficient calculation.

Graph Improvements
The appearance of the graphs in Flownex® has been updated and the new graphs can be seen in Figure 14.
Fig. 15 - Comparison Between the Old and New Graphs in Flownex

The graphs inputs have been modified such that only basic graph properties are shown when creating a new graph to ease formatting/styling. Additional formatting properties are available when checking the Advanced Formatting properties, as seen in Figure 15.

Fig. 16 - Graph Properties

New graph functionalities include:

  • By default, the Y-Axis will zoom and pan automatically for easy navigation.
  • The X-Axis will auto scale.
  • Line types can be changed to Step, Spline, Scatter Line, Area, Step Area or Spline area.
  • Line graph plotting data can be saved to a CSV file by simply right clicking on the graph.
  • A Crosshair cursor showing all Y-Axis values for a specific X-Axis value has been added.
  • Graphs formatting can be changed without having to solve the network again.
Rotating Components

The Daily and Nece correlation for calculating moment coefficients on disk surfaces was added as an option for the Rotor-Stator and Rotor-Rotor cavities as seen in Figure 16.

Fig. 17 - Friction Coefficient Correlation Options

In the case of the Rotor-Stator cavity, the Daily and Nece correlation allows for four different regimes, including fully interfering boundary layers within very small gap widths. The possibility to modify the Haaser et. al. correlation to be dependent on the gap width to disk diameter ratio was also added.

The Rotor-Stator and Rotor-Rotor cavities were upgraded to allow specifying an inner radius and outer radius for each disk individually, this is specifically useful when modelling cavities with axial inflows/outflows.

The Rotor-Stator and Rotor-Rotor cavities were upgraded to allow a disk surface profile specification that is not strictly rising with radius.

An option was added to all elements connected to vortices and cavities to specify the radius at the connection rather than the radius fraction. This allows the user to easily link the connection radius input to a measurement on a scaled drawing.

An increment result for windage power was added to the Rotor-Stator and Rotor-Rotor cavities. The windage power calculation of the Rotating Channel, Rotating Nozzle, Labyrinth Seal and Rotating Annular Gap was modified to automatically account for windage addition/removal to the element on account of upstream node swirl speeds not equal to that inside the element. Previously the windage power added to these elements was solely attributed to that required to maintain the swirl speed inside the element. This modification may lead to modified results since windage affects gas density.

The windage power calculation on the Forced Vortex component was modified to account for changes in kinetic energy of incoming flow streams that must increase/decrease in order to be equal to the swirl speed of the vortex at the particular connection radius. This modification may lead to modified results since windage affects gas density.

Liquid Gas Mixtures

The following components were extended to allow liquid-gas mixture fluid types, thereby allowing coupling of the secondary air system with the lubrication system:

  • Rotating Channel
  • Rotating Nozzle
  • Nozzle
  • Rotor-Stator Cavity
  • Rotor-Rotor Cavity
  • Forced vortex
  • Free Vortex
Chemical Reactions

The combustion category has been renamed to Chemical Reactions. The existing Adiabatic Flame model is a chemical reaction where the end temperature and composition of the end product of the reaction is determined by the CEA calculations. Another component has been added to the Chemical Reactions library where the user can specify the end temperature of the chemical reaction, namely the Gibbs Free Energy Reactor.

Gibbs Free Energy Reactor
Fig. 18 - Gibbs Free Energy Reactor
The component uses the NASA CEA program to predict the reaction products at the specified end temperature. This component will then calculate the change in Gibbs free energy and enthalpy during the reaction. As part of this enhancement, a Gibbs free energy result has been added on all flow nodes. The first application of this reactor is to model fuel cells and use Flownex® to optimise the surrounding systems. There are however many other possible applications.

MINOR ENHANCEMENTS

Scale Drawings

The setup of measurements in scale drawings has been simplified. The user can now drag and drop properties from components onto measurement points and measurement lines.

If it is a measurement point, the user will be asked to which part of the coordinate (X,Y, or Z) it should be assigned.

Fig. 19 - Coordinates

If the user drags and drops on a line, the property will automatically be associated with the length of the line.

CFX Interface

Errors when using a comma as decimal separator were fixed. The interface is able to handle both a point or a comma as decimal separator.

The option was added to deactivate the simulation when the CFX Generic Interface encounters an error. A user can continue the simulation from the point where the error (can include CFX solver crash) occurred, saving the time it took to reach that point.

Fig. 20 - Deactivate Simulation when Error Occurs Option

FMI

The Flownex® Co-Simulation FMU capability has been enhanced. Flownex® now supports state serialization of the FMU. State serialization is however a slow process and therefore it is not recommended unless it is crucial. There is a new option in the FMU configuration to turn state serialization on and off. The option is false by default. An option has also been added to the FMU configuration to hide Flownex® during FMU execution.
Fig. 21 - FMU Configurations
An example of using a Flownex® FMU with CFX has been added as a tutorial (Tutorial 44).

AFT Importer

In the past, the AFT importer always searched for a scenario named Base Scenario or Base Case to import. This has been improved and the first scenario listed in the Scenario Manager section of the file will now be imported. This enhancement fixes issues were scenarios did not have the default Base names.

API

Functions were added to the network builder to set the page size for any page.

Ansys Mechanical Link

The Ansys mechanical link now allows a user to specify the name of boundary conditions to transfer load data. Specifying the same names for matching boundary conditions and named selections allows the user to use named selections in the link setup.

Fig. 22 - Use Boundary Condition Names

Units

The unit g/mol was added for molar mass.

The unit kN was added for force.

FlownexSEConsole

Errors and warnings were displayed in the console version of Flownex®, but were not recorded. The errors and warnings are now recorded to files named FlownexSEConsoleWarnings.txt and FlownexSEConsoleErrors.txt respectively. These files are located inside the project folder.

Convection Coefficient Correlations

Correlations for calculating convection heat transfer to ambient was added to the Composite Heat Transfer component, as well as the Insulated Pipe component.

Correlations for three different convection mechanisms were added, namely free convection over a horizontal cylinder, free convection over a vertical cylinder, or forced convection over a cylinder. The user can select the mechanism that should be used, as seen in Figure 22.

Fig. 23 - Convection Coefficient Correlation Selections

Pipe

Added Wall Shear Stress result for Pipe elements with Newtonian Fluids.

Material Warnings

Added warnings for low and high limits when interpolating from two-phase tables. Added warning when two-phase critical mass flux cannot be calculated due to low total pressure.

Nuclear Reactor

Improved the checking and issuing of errors for materials in the Advanced Reactor – materials that did not exist issued warnings even when options were active where they were not used.

Several improvements were made to the text based Nuclear results. This includes correcting the generated heat results for solid nodes and solid node volume results. Also, units were added and corrected for heat in several places.

Node Results

Added an energy source result on solid nodes, as well as transient energy source calculation on all node types. A Gibbs free energy result was also added to flow nodes.

Container Interface Components

The Container Interface Top and Container Interface Bottom components were enhanced to allow the specification of height or height fractions on adjacent elements. Previously, elements could only be connected at the bottom or top.

Fig. 24 - Connection Specification Options

BUG FIXES

Nuclear Reactor

Fixed a problem where the volume on flow nodes were too large when using Darcy Weisbach pressure drop in a nuclear reactor.

Scripts

Prevent snaps from loading new script code. This could cause the user to lose the current script and it would be reverted to the script that was used when the snap was saved.

Heat Transfer Component

When multiple Composite Heat Transfer elements were connected to a component, the downstream temperature exceeded the wall temperature. This has been fixed.

User Specified Pressure Drop

Fixed the problem with networks using multiple User Specified Pressure Drop Components that did not converge under some conditions.

Pipe

Fixed incremented pipes giving different results than non-incremented pipes when used with non-Newtonian fluid types.

Positive Displacement Compressor

Fixed the problem where the Polytropic coefficient value was limited between 0 and 1 for the Positive Displacement Compressor.

Charts & Data References

The full name and path of the chart were saved in the chart and then Chinese characters in the path caused a problem with the saving. This was removed and the charts are now better compatible with Chinese characters in file names and folders.

Fixed problems when turbo machinery charts have too many points. This caused losses of data and abnormal program termination. The errors that are issued when there are too many points are now handled gracefully.

Scale Drawings/GIS

Elevation retrieval for nodes on imported GIS networks did not work due to changes to the Google elevation API. The code has been updated to work with the new changes.

There was a problem on scale drawings where IO boxes displaying measurement positions or lengths did not update when measurement items were moved.

Fixed a problem where using groups on scale drawings caused duplicate views of components on the scale drawings.

Excel Input Sheet

Saving of snaps caused Excel Input Sheets to stop functioning. This has been fixed.

Excel Reports

There was a problem where the last time step during a run was not reported to the table with time steps. This has been fixed.

The cyclic option on reports did not always sample data at the correct time – it was sometimes off by a couple of timesteps.

Global Parameters

Fixed the problem where loading snaps created duplicate global parameters. This caused the global parameter list to become very large and caused serious slowdowns in the user interface. A script has been added to remove the duplicate items.

Licensing

Fixed a problem where checking licenses back in if the roam (borrowing) period is longer than the server license expiration date caused an exception.

Download full release notes here.