OpenFOAM

tahanson43206 · 2023-10-07 08:27:34

OpenFOAM is a software package (with related supporting software) that provides Computational Fluid Dynamics simulations at no charge.

There are major players in the CFD industry. FLOW-3D is a company I've been following for some time.

FLOW-3D
www.flow3d.com
FLOW-3D is an accurate, fast, proven CFD software that solves the toughest free-surface flow problems. A pioneer in the CFD industry, and a trusted leader, FLOW ...
FLOW-3D HYDRO
CFD Software
Additive Manufacturing

This topic is offered for collecting information about OpenFOAM and it's use.

According to ChatGPT4, OpenFOAM has improved and developed significantly since I first encountered the program several years ago.

The only book available at the time was written in German, and I was unable to use it effectively, so I donated it in hopes someone would find it useful.

OpenFOAM would be useful for analysis of many problems under discussion in the NewMars forum. As just one example, it might be possible to model Ballistic Delivery of non-perishable supplies to Mars. The hypothesis in play is that costs of preparing delivery packages and related support services would be less than the cost of gentle delivery of those supplies using chemical rockets.

(th)

tahanson43206 · 2023-10-07 08:29:19

This post is reserved for an index to posts that may be contributed by NewMars members over time.

Index:
Post #4 Transcript of conversation with ChatGPT4o regarding possible use of OpenFOAM to model a heat pipe for Optical Plane space vessel.

(th)

tahanson43206 · 2025-04-16 09:30:02

This post is a duplicate of the orignal Post #2

tahanson43206 wrote:

The transcript at the link below covers discussion of a stretched centrifugal pump, Blender and OpenFOAM.
https://docs.google.com/document/d/1hIs … sp=sharing
I had not realized OpenFOAM has progressed so far since I first investigated it some years ago.
This new topic is available for anyone to provide updates on the package, and examples of it's use.
(th)

tahanson43206 · 2025-04-16 09:31:02

This post is about how to model the heat pipe and nozzle systems for the Optical Plane space vessel.
Attempts to model this complex system using Python yield approximations, but Python is limited in what it can show.
There are a number of CFD (Computational Fluid Dynamics) programs in use in 2025. OpenFOAM is one I'm familiar with.

The transcript below begins after the idea of considering OpenFOAM had been considered.

Based upon my experience working with you over the past year and more, I am moderately optimistic that we might be able to pull this off. I am about 85% confident we could replicate work that has been done previously. That would not be easy, but If all the files are available we should be able to replicate the output. To go beyond that to model our specific heat chamber design would require stretching beyond what has been done before. Please prepare a BBCode summary of this discussion so I can share it with our team, and with our wider audience.
***
ChatGPT said:
Absolutely! Here's a BBCode-formatted summary for you to share with your team and broader audience:
Exploring OpenFOAM for Modeling a Solar-Heated Rocket Engine
Background: In previous phases of the project, we used Python to simulate a plugged heat pipe scenario in order to estimate time-to-temperature rise to 3000 K. These experiments indicated that the resulting pressures would be too high for practical use, leading us to abandon the plugged design in favor of an open-nozzle configuration.
Engineering Challenge: Our team seeks to model the performance of a solar-heated engine with an open nozzle, using realistic constraints such as a 240 meter heat pipe assembled from 10-meter segments. The pipe and joints must withstand high temperatures and internal pressure while enabling continuous hydrogen flow.
Current Concern: While Python provided useful insights during earlier development, it may not deliver the level of fidelity required to convince our aerospace collaborators of the system's viability—particularly in modeling compressible flow, heat transfer, and pressure gradients within the heating chamber.
Potential Solution – OpenFOAM:
OpenFOAM is a professional-grade open-source CFD toolkit capable of simulating compressible gas dynamics, thermal conduction, turbulent flows, and conjugate heat transfer.
It has been successfully used to model chemical rocket combustion chambers, nozzles, and exhaust plumes in both academic and industrial contexts.
Example applications include modeling the RD-107 rocket engine, simulations of hybrid rocket chambers under the HYPROB-HYBRID program, and URANS plume studies of the VEGA launch vehicle.
Team Outlook: Based on over a year of collaboration and successful technical outcomes, we are moderately optimistic about using OpenFOAM for our purposes. We estimate:
85% confidence in our ability to replicate existing OpenFOAM simulations if complete project files are available.
Cautious optimism about stretching beyond prior work to model our unique solar-heated chamber design, which differs from conventional combustion chambers in that it receives heat externally from focused solar energy.
Next Steps: Begin with a simplified, validated OpenFOAM case—possibly a 2D or axisymmetric heated pipe with an open nozzle—to build experience and confidence. If successful, scale up to model the 240-meter heat pipe and its distributed heat input structure.
Conclusion: Our collaboration continues to move into deeper technical waters. OpenFOAM appears to be a suitable platform for advancing our modeling goals and may help bridge the gap between conceptual feasibility and engineering validation for our solar propulsion system.
Let me know if you'd like to tailor this for a different audience or expand any section!

(th)

tahanson43206 · 2025-04-16 11:15:39

Here is the main page for the OpenFOAM foundation:

OpenFOAM
The OpenFOAM Foundation
Download
Development
Resources
Funding
Contact Us
OpenFOAM v12 for Linux, Windows and macOS
Install the latest version of OpenFOAM
Ubuntu Linux
Packaged installation for Ubuntu Linux. A simple installation option for OpenFOAM.
Install on Ubuntu
Other Linux
Compile the OpenFOAM source code to run on other Linux distributions.
Install on Other Linux
Windows
Run on Windows using WSL. A convenient installation using the Ubuntu package.
Install on Windows
macOS
Run on macOS using Multipass. A convenient installation using the Ubuntu package.
Install on macOS

While I recognize NewMars does not have active members who can install OpenFOAM, we have registered members who probably could.

This forum is open for registration of new members, so if a reader would like to assist with the initiative to model the heat engine for the Optical Plane space vessel, we check the email every morning.

(th)

tahanson43206 · 2025-04-16 11:20:57

And this is what the download page looks like:

9th July 2024
Download v12 | Ubuntu
Ubuntu Versions
OpenFOAM 12 is a major new release of OpenFOAM provided by the openfoam12 pack. It is accompanied by ParaView using the standard paraview package with Ubuntu versions 24.04 and 22.04 and a customised paraviewopenfoam510 package for Ubuntu 20.04. OpenFOAM 12 is available for the following versions of Ubuntu:
20.04 LTS, codename focal
22.04 LTS, codename jammy
24.04 LTS, codename noble
Installation
OpenFOAM and ParaView can be simply installed for the first time using the apt package management tool. The user will need to provide superuser password authentication when executing the following commands with sudo
Copy and paste the following in a terminal prompt (Applications → Accessories → Terminal) to add dl.openfoam.org to the list of software repositories for apt to search, and to add the public key (gpg.key) for the repository to enable package signatures to be verified.
Note: use secure https:// for the public key to ensure secure transfer, but usehttp:// for the repository, since https:// may not be supported and is not required since the key provides secure authentication of the package files.
sudo sh -c "wget -O - https://dl.openfoam.org/gpg.key > /etc/apt/trusted.gpg.d/openfoam.asc"
sudo add-apt-repository http://dl.openfoam.org/ubuntu
**Note: This only needs to be done once for a given system
Update the apt package list to account for the new download repository location
sudo apt update
Install OpenFOAM 12 which also installs paraview (or paraviewopenfoam510 on Ubuntu 20.04) as a dependency if it is not already installed.
sudo apt -y install openfoam12
OpenFOAM 12 and ParaView are now installed in the /opt directory. If any problems are encountered during the installation, see the Installation Problems section below.
Patching OpenFOAM 12
OpenFOAM-12 is periodically recompiled into a new openfoam12 pack. The package will be upgraded automatically whenever repository package files are updated and the software is upgraded, e.g. by typing
sudo apt update
sudo apt upgrade
Users can also specifically upgrade openfoam12 only by the following option:
sudo apt update
sudo apt install --only-upgrade openfoam12
User Configuration
In order to use the installed OpenFOAM package, complete the following:
Open the .bashrc file in the user’s home directory in an editor, e.g. by typing in a terminal window (note the dot):
gedit ~/.bashrc
At the bottom of that file, add the following line (see Note 1 below) and save the file (note the leading dot):
. /opt/openfoam12/etc/bashrc
Open a new terminal window (see Note 2 below) and test that the foamRun application, from the OpenFOAM package, is working by typing
foamRun -help
A “Usage” message should appear. Your installation and user configuration is complete.
Note 1: If a similar line has already been added to the user’s .bashrc file, e.g. for a previous version of OpenFOAM, then the line should be deleted or, alternatively, commented out by inserting a # at the beginning of the line.
Note 2: If a user wishes to execute foamRun imme the same terminal window, they must first register the change to the .bashrc file by typing at the terminal prompt (note the dots): . $HOME/.bashrc
Getting Started
Create a project directory within the $HOME/OpenFOAM directory named <USER>-12 (e.g. chris-12 for user chris and OpenFOAM version 12) and create a directory named run within it, e.g. by typing:
mkdir -p $FOAM_RUN
Copy across the steady-state backward facing step example, generate the mesh with blockMesh and run foamRun with the incompressibleFluid solver module
cd $FOAM_RUN
cp -r $FOAM_TUTORIALS/incompressibleFluid/pitzDailySteady .
cd pitzDailySteady
blockMesh
foamRun
paraFoam
Refer to the OpenFOAM User Guide to get started.
Installation Problems
1. Networking must be enabled in order to be able to install software packages in Ubuntu. Installation failure due to networking problems usually results in an error message that reports failures in fetching archives and ends with a line like the following:
E: Unable to fetch some archives...
2. A few users have encountered an error message similar to the following during installation (step 3):
Some packages could not be installed. This may mean that you
have requested an impossible situation...
The following information may help to resolve the situation:
The following packages have unmet dependencies: openfoam12...
The most likely cause of this error message is that the universe repositories are not enabled. To enable them, type the following in a terminal window:
sudo apt-add-repository universe
sudo apt update
3. If MPICH is installed on the system, then during user configuration (below), the following error message might occur when sourcing the etc/bashrc file in the OpenFOAM installation:
gcc: error: unrecognized command line option '--showme:link'
The user can check whether MPICH is installed by typing the following:
sudo update-alternatives --list mpi
which returns /usr/mpich/include if MPICH is installed. This error is caused when the default mpicc is MPICH, rather than OpenMPI. The simplest way to fix the problem is to set mpicc to OpenMPI is by running the following command in a terminal prompt:
sudo update-alternatives --set mpi /usr/lib/openmpi/include
Reporting Bugs in OpenFOAM
We appreciate that bugs in OpenFOAM are reported so we can fix them. Please refer to the OpenFOAM Issue Tracking to report bugs.
9th July 2024 Chris Greenshields News

(th)

tahanson43206 · 2025-04-16 11:34:01

And this is what the installed OpenFOAM help display looks like:

$ foamRun -help
Usage: foamRun [OPTIONS]
options:
-case <dir> specify alternate case directory, default is the cwd
-fileHandler <handler>
override the fileHandler
-hostRoots <((host1 dir1) .. (hostN dirN))>
slave root directories (per host) for distributed running
-libs '("lib1.so" ... "libN.so")'
pre-load libraries
-noFunctionObjects
do not execute functionObjects
-parallel run in parallel
-roots <(dir1 .. dirN)>
slave root directories for distributed running
-solver <name> Solver name
-srcDoc display source code in browser
-doc display application documentation in browser
-help print the usage
Using: OpenFOAM-12 (see https://openfoam.org)
Build: 12-86e126a7bc4d

(th)

tahanson43206 · 2025-04-16 20:12:03

The series to be recorded in this topic is intended to find a solution to the challenge posed by the Optical Plane Solar Powered vessel.

Hello Again! Let's begin a gradual build up of our experience with OpenFOAM. I am planning to share our adventure with the forum audience as we go. Yesterday we confirmed that OpenFOAM V12 is installed in our Ubuntu 22.04 system and it displayed the -help message. I think we are ready to begin. Our goal is to model the Heat Pipe and Nozzle system for our Solar Powered rocket engine.
***
ChatGPT said:
Updated saved memory
Welcome back! That sounds like an exciting project and a great way to introduce OpenFOAM to your forum audience. Since we're starting fresh with OpenFOAM and aiming to model the Heat Pipe and Nozzle system for your solar-powered rocket engine, here’s a suggested roadmap we can follow together:
OpenFOAM Learning Roadmap for Solar-Powered Rocket Engine Modeling
Step 1: Run a Simple Existing Tutorial Case
Start with a built-in OpenFOAM tutorial (e.g., incompressible flow in a pipe or nozzle). This will help confirm that your environment is fully operational and give us a feel for the directory structure and file types.
Step 2: Modify Geometry and Boundary Conditions
Use the simple tutorial as a base and begin modifying the geometry and boundary conditions to resemble a simplified version of your heat pipe or nozzle.
Step 3: Transition to Compressible Flow
Move to a compressible solver (like rhoCentralFoam or sonicFoam) appropriate for simulating high-speed gas dynamics in the nozzle.
Step 4: Introduce Heat Transfer
Explore solvers like chtMultiRegionFoam for conjugate heat transfer to represent thermal interaction between hydrogen and the pipe/nozzle walls.
Step 5: Customize for Solar Heating
Begin implementing or approximating the solar heating process by introducing internal heat sources or temperature boundary conditions.
Step 6: Post-processing and Visualization
Use ParaView to visualize velocity, pressure, and temperature fields in the pipe and nozzle region.
Step 7: Validation and Tuning
Check results against your existing Python model and engineering expectations. Tweak mesh resolution, solver settings, and physical properties.
Step 8: Document for Forum Presentation
Prepare BBCode summaries, screenshots, and explanation of each step to share with your audience.
Shall we begin by selecting and running a simple tutorial case? I recommend starting with icoFoam or rhoCentralFoam, depending on whether you want to start incompressible or jump right to compressible gas dynamics.
Let me know your preference, and we’ll set up the first run!

(th)

tahanson43206 · 2025-04-16 21:15:16

ChatGPT4o took me through a number of steps to get the first tutorial running.
The display starts out as a beige square, but when animation is requested there is activity at the top of the square and colors appear.
This is enough progress for one day. My guess is that at some point the entire engine will show up in that Workspace,
I am imagining hydrogen entering at the left while the heat pipe glows white hot, and the hydrogen gathers energy and pushes against the walls and escapes through the nozzle throat. That day is a ways off, I suspect.

(th)

tahanson43206 · 2025-04-17 19:51:57

Post #9 shows the simplest possible Tutorial. It is good for learning how to set up a run, and how to use ParaView to animate the computations.

ChatGPT4o and I moved on to a different tutorial that is a step closer to where we want to end up. This one features a small flame that grows over a period of time. In this case the time involved is 4 seconds. A program (a "solver") has to be run before there is anything for ParaView to show.

Here is a view of the flame before animation:

And here is the flame after four seconds. It grew towards the far right and then came back to the left:

Tomorrow I'll attempt to learn how to step through the animation, and perhaps even create a video of the animation.

Ultimately we want to animate for operation of the propulsion system for the Optical Plane space vessel. We are a long way from ** that **.

(th)

tahanson43206 · 2025-04-18 10:44:21

In this session, ChatGPT4o and I covered how to create a video of the animation of the scenario in focus. In this scenario, we allow a simple flame to grow in a chamber characterized by a step which influences the behavior of the gases. The animation will be saved in Dropbox, and you can run it on your system by clicking on the Dropbox link.

Thank you (again) for your assistance with the forwardStep tutorial for OpenFOAM. We left off yesterday with the animation running. I took screen shots of the before and after states and published them to show our progress. Today I would like to explore animation a bit, with a view to possibly creating a video that might be shared with the forum. I see in the File menu two or three options that appear to have to do with saving animation. Rather than experiment, it seems reasonable to me to explore the options with you, to save time. The File menu goes away when I change focus from the ParaView window to yours. Let me approach the question this way: Is there a way to create a video of the animation that I can share, or do I need another tool?
***
ChatGPT said:
You're asking exactly the right question—and the great news is: yes, ParaView includes a built-in tool to create high-quality videos directly from your animation—no extra software required.
Let’s walk through it together to ensure your output looks polished and is easy to share with your forum audience.
How to Save an Animation as a Video in ParaView
Follow these steps in ParaView 5.11.2 to render and export your animation:
1. Make Sure Your Animation is Set Up
Load your case (you’ve done this)
Click Play to confirm the animation works as expected
Use the Time slider or Animation View (if needed) to adjust the time range
2. Open the Save Animation Dialog
Go to File > Save Animation…
(This opens a dialog with many options for resolution, file type, etc.)
3. Choose Output Format
In the dialog that appears:
Select Format: Choose AVI or OGV for a video file
(OGV is more compact and web-friendly; AVI is more universally playable)
You can also export as an image sequence (e.g., PNGs), but that's optional for advanced use
4. Set Resolution and Frame Rate
You can keep the default resolution or set your own (e.g., 1280×720 or 1920×1080)
Adjust Frame Rate if needed (24 fps or 30 fps are common)
5. Choose Output Filename
Click the folder icon to pick a save location and filename
Example: forwardStep_animation.avi
6. Click OK to Start Rendering
ParaView will now process each frame and compile the video.
This may take a few minutes depending on resolution and frame count.
✅ Result: Shareable Video File
The output .avi or .ogv file can be shared on your forum, uploaded to Dropbox, Google Drive, or attached in emails.
If the forum supports embedding videos (via BBCode or otherwise), you could link to it.
Optional: Tools for Further Editing (if needed)
If you later want to:
Add annotations or titles
Splice multiple clips
Convert between formats
You could use free tools like:
Shotcut (cross-platform video editor)
HandBrake (format converter)
ffmpeg (command-line, powerful)
But in most cases, ParaView’s built-in exporter is sufficient for clean, standalone animations.
Would you like to proceed now with rendering your animation to .avi or .ogv? I can help you pick optimal settings for web sharing if you’d like!

(th)

tahanson43206 · 2025-04-18 10:55:04

The link below will bring up a video of the Flame Tutorial from OpenFOAM.

https://www.dropbox.com/scl/fi/ks90snxh … z2emw&dl=0

This is a preview of what I hope we can achieve in coming weeks. The model of kbd512's update of the Solar Powered space vessel is far more complicated that the simple flame shown in the video, but it seems possible that OpenFOAM can model the engine, since it is already said to be successfully modeling other engines that are in service today in early 2025.

Update: When I tested the Dropbox link the animation appeared in a finished state. There should be a way to restart the video.

In the case of the specific browser I used, I found that Back Step 10 would reveal the starting flame.

Update after testing on a different browser on a different machine and a different operating system:

The display seems the same. Using the Back Step icon revealed the starting flame.

It is possible to slow down the video slightly, apparently, as well as to speed it up.

(th)

tahanson43206 · 2025-04-19 10:14:49

Here is an image of a first cut at designing a model for the engine system for the Optical Plane space vessel:

(th)

tahanson43206 · 2025-04-19 17:56:11

In an effort to have something to show to the Google Meeting tomorrow evening, I've been working most of the day with ChatGPT4o to see if we could persuade OpenFOAM to run hydrogen through the little model we show in Post #13. Unfortunately, even something as (apparently) simple as that little model appears to be a really hard problem. We got ** so ** close, but the program designed to handle flowing gases threw in the towel and generated a floating point error. The system offered to send a core dump to the developers so I approved that.

Then I reported the unexpected outcome to ChatGPT4o, and it expressed sympathy, which I thought was a nice touch. Then it offered to review the command files I'd been working on under it's watchful "eye". Unfortunately, since I was doing the typing and cutting and pasting, I could easily have made a mistake.

So! Here's the diagnosis:

smoothSolver: Solving for h, Initial residual = 1, Final residual = 0.0111863, No Iterations 1
#0 Foam::error::printStack(Foam::Ostream&) at ??:?
#1 Foam::sigFpe::sigHandler(int) at ??:?
#2 ? in "/usr/lib/x86_64-linux-gnu/libc.so.6"
#3 ? in "/usr/lib/x86_64-linux-gnu/libm.so.6"
#4 Foam::PsiThermo<...>::calculate() at ??:?
#5 virtual thunk to Foam::PsiThermo<...>::correct() at ??:?
#6 Foam::solvers::fluid::thermophysicalPredictor() at ??:?

and:

This suggests that during the thermophysical property calculations, a division by zero or another invalid operation occurs, possibly due to uninitialized or non-physical values.

Now ** that's ** a surprise! Early on I learned to test for 0 before doing a division, but obviously since a floating point error occurred, there must be code somewhere in the system that does not check for that condition.

It's been a long day. We (ChatGPT4o) and I are trying to create a model that is different from all the existing tutorials, so it seems to me hitting potholes along the way is inevitable.

Update: Before closing down for the day, ChatGPT4o and I added the foundation of a heater to surround the heat pipe. I observed that the hydrogen in the pipe was blue, and ChatGPT4o confirmed that is because the hydrogen is so cold it is given the deepest possible blue color.

That led to discussion of how to heat the hydrogen, and it turns out OpenFOAM has modules just for that, so we started to create a heater for the heat pipe. The coordinates for the heat pipe heater are entered into the system, and tomorrow (if all goes well) I'll find out how to heat the pipe.

I also realized we need a propellant tank for this system, so I'll ask about adding one tomorrow.

(th)

tahanson43206 · 2025-04-20 06:57:21

This post is about calculation of a term for enthalpy (heating of a substance).

I've tried to make clear from the outset that I am in this way over my head. If ChatGPT4o were not available I wouldn't even attempt to use OpenFOAM.

However, OpenFOAM appears capable of creating a model of the engine for the Optical Plane space vessel, so i am forging ahead despite formidable obstacles.

It is highly likely the NewMars membership includes persons who could handle this better, but no one has spoken up so I just keep going in hopes of achieving something that might be useful.

Today, ChatGPT4o and I struggled with calculation of a term (Su) that is needed at model run time to show heating of hydrogen in our heat pipe. The default value suggested by OpenFOAM was 5000, and ChatGPT4o calculated 28860 for that field. This was based upon my specifications of a heat pipe 240 meters long, 10 centimeters in diameter, and with hydrogen flow rate of 1 kg/s. The hydrogen is assumed to enter the tube at 300 Kelvin and to exit at 3000 Kelvin.

This configuration applies a heat source that increases the enthalpy of hydrogen by approximately 28,660J/kg·s, which corresponds to the desired temperature rise from 300K to 3000K at a mass flow rate of 1kg/s.
Please adjust the Su value based on more precise measurements of density and specific heat capacity if available.
If you need further assistance or have more specific data, feel free to provide it, and I can help refine these calculations.
Sources
You said:
Please compare the expression kg-s to the expression kg/s
ChatGPT said:
Certainly! Let's clarify the distinction between "kg·s" and "kg/s":
? kg/s (kilograms per second)
Meaning: This denotes a mass flow rate—specifically, the amount of mass (in kilograms) passing through a given point per second.
Usage: Commonly used in engineering and physics to describe how quickly mass moves through a system, such as the flow of hydrogen in a heat pipe.
Example: If hydrogen is entering a chamber at a rate of 1kg/s, it means that every second, 1 kilogram of hydrogen is flowing into the chamber.
? kg·s (kilogram-second)
Meaning: This represents the product of mass and time. It's not a standard unit in the International System of Units (SI) and doesn't have a commonly recognized physical interpretation.
Usage: Rarely used; it might appear in specific theoretical contexts or derivations but isn't standard for expressing physical quantities.
Example: Multiplying mass by time without a clear physical context doesn't yield a commonly understood quantity.
✅ Summary
Use kg/s when referring to mass flow rates, such as the rate at which hydrogen flows through a system.
kg·s is not standard and should be avoided unless a specific context requires it.
Wikipedia
Informa TechTarget
If you have any more questions or need further clarification, feel free to ask!

This is just great! We are forced to use a unit that "does not yield a commonly understood quantity"

I interpret this to mean that the mathematicians who built OpenFOAM over many years have had to invent a term for a number they need for calculations.

The term may not have meaning in the Real Universe, but it may (must) have value in calculation of heating of hydrogen in a heating tube.

Thus, the term may be similar to the "imaginary number" that is at the heart of mathematics, the square root of -1

I asked ChatGPT4o to compare the two terms:

See next post...

(th)

tahanson43206 · 2025-04-20 07:07:18

I asked ChatGPT4o to compare kg-s to the imaginary number (i) that is the square root of minus 1

Please compare the kg-s term to the "imaginary" number that is the square root of minus 1
***
ChatGPT said:
The expressions kg·s (kilogram-second) and √−1 (the imaginary unit i) both represent concepts that extend beyond direct physical measurement, but they serve different purposes in science and engineering.
? kg·s (Kilogram-Second)
Definition: This is the product of mass (kilograms) and time (seconds).
Usage: It's not a standard unit in the International System of Units (SI) and doesn't correspond to a commonly used physical quantity. However, it might appear in theoretical contexts or as an intermediate result in dimensional analysis.
Interpretation: While not commonly used, kg·s could, in specific scenarios, represent a quantity where mass and time are multiplicatively related, but such interpretations are rare and not standardized.
? √−1 (Imaginary Unit i)
Definition: The imaginary unit i is defined by the property that i² = −1.
Wikipedia
Mathematics Stack Exchange
Math Is Fun
Usage: Imaginary numbers are fundamental in complex number theory and are widely used in engineering, physics, and mathematics. They enable the solution of equations that have no real solutions and are essential in fields like electrical engineering (e.g., analyzing AC circuits), quantum mechanics, and control theory.
Interpretation: Although termed "imaginary," these numbers have real-world applications and are integral to modeling and understanding various physical phenomena.
? Comparison
Aspect kg·s (Kilogram-Second) √−1 (Imaginary Unit i)
Nature Product of mass and time Mathematical construct extending real numbers
Standard Usage Rarely used; not a standard SI unit Widely used in complex number theory and applied sciences
Physical Meaning Lacks a direct physical interpretation in most contexts Represents a rotation in the complex plane; essential in modeling oscillatory and wave phenomena
Applications Limited; possibly in theoretical or dimensional analysis Extensive; crucial in engineering, physics, and mathematics
In summary, while kg·s is a seldom-used unit with limited application, the imaginary unit √−1 (i) is a fundamental concept with broad and significant applications across various scientific and engineering disciplines.

I deduce that if a NewMars member can understand the square root of minus one and apply it with success, then that individual would be capable of dealing with kg-s.

(th)

tahanson43206 · 2025-04-20 08:57:03

Because the attempt to create new model for a rocket engine that does not exist, starting from scratch, has caused core dumps, I have decided to attempt to run a tutorial of the Merlin engine. According to the information that ChatGPT4o found, the model is available on github, and apparently it is associated with a group that created the model. I'm hoping there may be support available if it is needed, but I'm not counting on it.

I'll be exploring this opportunity later today.

The model should include many if not all of the OpenFOAM features we need for our much simpler concept.

Update after weekly Google meeting:

The attempt to run the Merlin OpenFOAM model was not successful. Despite strenuous efforts, success did not occur.

The basic theme is the passage of time. The Merlin model was created using a solver called sonicFoam and I am sure it was working when it was published.

However, OpenFOAM has retired sonicFoam, and implemented a replacement called rhoCentralFoam.

There are differences between the two solvers, and ChatGPT4o and I struggled to find all the differences and correct them.

In the end, we ran out of time. The model is partially working, in the sense that the outline of the Merlin engine is visible in the render panel, but no animation is possible because rhoCentralFoam creates a core dump when asked to perform the mathematical operations needed to create animation.

ChatGPT4o thinks a possible cause of the core dumps is an attempt to take square root of -1 (or some negative number)

Going forward, the new plan is to look for working tutorials that combine:

1) Use of rhoCentralFoam and
2) Focus on rocket engine physics

ChatGPT4o has proposed a work plan that sets out to reach the goal using these new constraints, and I expect to begin work on Monday.

Looking forward, I hope to be able to adapt a working tutorial for the unique features of the Optical Plane space vessel.

For those who might be reading this without having followed developments in other topics, the engine for the Optical Plane application is distinctive because:

1) It uses external heating instead of internal heating as done by all other rockets ever built and flown (*)
2) It has a heating tube length of 240 meters, due to the architecture of the solar energy collecting wings

Other than those two details, everything else about the nozzle would (presumably) be identical to existing designs.

(*) Exceptions to the rule:
1) The laser boosted device designed by Leik Myrabo at Rensselaer Polytechnic Institute was actually tested (it does use external heating)
2) Theoretical studies were published considering the use of lasers to heat hydrogen to propel a vehicle. That concept is very similar.

Reminder: the proposed engine design under consideration here has never been built and it may not work.

Examples of systems that DO work ALL use internal heating.

If a reader is aware of an exception please let us know via the Recruiting email address.

(th)

tahanson43206 · Yesterday 20:34:34

ChatGPT4o and I searched for any tutorials provided by OpenFOAM that use rhoCentralFoam, and found none. rhoCentralFoam is the replacement for sonicFoam, which was used by the Merlin tutorial we found. I asked ChatGPT4o to look out on the Internet to see if there might be a tutorial that uses rhoCentralFoam and also has something to do with rockets. To my great surprise, it found that the organization that created the tutorial we were working with also created an update using rhoCentralFoam. I'm planning to download that tutorial tomorrow. hopefully it will compile and run without further hiccups.
(th)

tahanson43206 · Today 08:19:10

Update after repairing connection to github...

GitHub has implemented much stricter security since I last used the service. It took a bit of back and forth to resolve that issue, and now I have a brand new Personal Access Token that I can use to download repositories.

Unfortunately, the repository that ChatGPT4o found yesterday (no doubt by searching the Internet) was NOT FOUND by github today when I asked for it. That repository may well have been moved. All the content in the owner's directory is dated four years ago.

Not to Worry, says ChatGPT4o! It found ** another ** repository by a different author.

git clone https://github.com/AussieSean/openfoamS … Nozzle.git

I'll try to pull that one shortly. It does not appear to be specific to a particular engine, but it may be adaptable for our project.

I'm not sure what the term "supersonic" means in this context. According to GW Johnson, the velocity of gas at the throat is equal to the speed of sound of the gas in use, but perhaps the "super" aspect comes into play as the gas leaves the throat and fills the expansion bell.

(th)

tahanson43206 · Today 14:02:21

This post is to report modest progress in learning how to use OpenFOAM.

The image above shows that the Merlin Engine model was sucessfully ported to a new framework where rhoCentralFoam was able to create one full time cycle. The quantity of data for just one time cycle is immense, so I'll be investing in external storage. The viewer shows the model as it will appear later on, except that it is solid blue because it is registering the temperature of liquid Hydrogen. In future updates we will add heating to the heat pipe.

And that reminds me ... the heat pipe is not defined yet, and the propellant tank is not defined. Both are needed to reflect the Optical Plane vessel requirements.

Work immediately ahead includes finding out why the simulation stopped after one time cycle.

That may take a while

It took most of today to rebuild the model using only the mesh from the previous (2022) Merlin Engine study.

As reported earlier, OpenFOAM is written in C++ to achieve maximum performance so NO machine cycles are wasted on checking for divide by zero or divide negative numbers. That said, I must admit that the programmers of rhoCentralFoam (a solver) spent a ** LOT ** of time on helping users to set up the resource files correctly. As just one example of many, ChatGPT4o concentrated on the content of source files, and omitted the pesky header sections. I was able to add those as rhoCentralFoam complained they were missing.

(th)

New Mars Forums

Announcement

#1 2023-10-07 08:27:34

OpenFOAM

#2 2023-10-07 08:29:19

Re: OpenFOAM

#3 2025-04-16 09:30:02

Re: OpenFOAM

#4 2025-04-16 09:31:02

Re: OpenFOAM

Exploring OpenFOAM for Modeling a Solar-Heated Rocket Engine

#5 2025-04-16 11:15:39

Re: OpenFOAM

#6 2025-04-16 11:20:57

Re: OpenFOAM

#7 2025-04-16 11:34:01

Re: OpenFOAM

#8 2025-04-16 20:12:03

Re: OpenFOAM

OpenFOAM Learning Roadmap for Solar-Powered Rocket Engine Modeling

#9 2025-04-16 21:15:16

Re: OpenFOAM

#10 2025-04-17 19:51:57

Re: OpenFOAM

#11 2025-04-18 10:44:21

Re: OpenFOAM

How to Save an Animation as a Video in ParaView

✅ Result: Shareable Video File

Optional: Tools for Further Editing (if needed)

#12 2025-04-18 10:55:04

Re: OpenFOAM

#13 2025-04-19 10:14:49

Re: OpenFOAM

#14 2025-04-19 17:56:11

Re: OpenFOAM

#15 2025-04-20 06:57:21

Re: OpenFOAM

#16 2025-04-20 07:07:18

Re: OpenFOAM

#17 2025-04-20 08:57:03

Re: OpenFOAM

#18 Yesterday 20:34:34

Re: OpenFOAM

#19 Today 08:19:10

Re: OpenFOAM

#20 Today 14:02:21

Re: OpenFOAM

Board footer