I’ve been practicing CFD for the past 15 years. I started practicing at my second year of undergraduate studies at Tel-Aviv University, as the stars aligned for me, and a childhood very close friend of my dad, who was also a faculty senior Professor by the name of Moshe Rosenfeld, head of the faculty’s CFD LAB, and a very experienced Fluent practitioner (that was before ANSYS acquiring Fluent), took me under his wing and offered me to start working at my spare time at the CFD LAB (I guess the incentive was my dear dad, he’s close childhood and still one of his best friends. Certainly not my beautiful eyes…).
I was also lucky enough for the Professor to not only give me little chores, and although my knowledge in fluid dynamics as a second year undergraduate student at the early stages of even taking some fluids 101 courses, was for every meaningful purpose close to nonexistent, he always seem to take the time explaining carefully what it is I’m actually doing. He also gave me a library card allowing a free access to a top-notch of technology machine called “a copying machine”, hence after taking the time to fully explain what we are aiming for I would run to the library (just over the bridge), copying the relevant references he was sure to mention, and read…. of course I had so much basic knowledge missing, I couldn’t really make sense of what I’m reading (although his office explanations of the phenomenological CFD application were handling was wholeheartedly sincere and elaborated), I quickly took note that I simply miss the basic knowledge… But then there’s the university engineering library, with all the books in hard copy fashion, so I simply decided to start from the beginning: fluids 101, discretisation schemes, boundary layers and turbulence (schlichting-pope-tennekes lumley-Chorin’s random walks…Ahhhh….), Oh, transition – got me hooked on KTH’s Dan Henningson team, then pure CFD (structured pde methods, unstructured methodologies, pde solving ) – got me accidentally to a set of lecture notes by a mathematician named J.M. McDonough, who seemed to want to cover many of the topics I was just on the need for: incompressible flows (methods, got acquainted with this professor Patankar guy… Solving PDEs: Jacobi, SOR, ADI, ILU, Multigrid, marching methods, Domain decomposition, Newton-Kantorovich, d&g, Cell Reynolds/Peclet…). He also wrote about Turbulence, funny to say as somewhat of an opposer to statistical methods such as RANS, it is turbulence from a mathematician standpoint (get all of his lectures here. I mean really, download all of them now)
In the lab getting wonderful cfd explanations from the Moshe, than doing my job which was mostly enclosing a vortex, get the value for the circulation and c&p to an excel worksheet, then repeat…
But something happened on this weird road, not really a CFD practitioner yet, reading every book I could lay my hand on, and while doing that for a long time I new already what’s “the good stuff”, I knew it! I shall be practicing CFD for a long long time now!!
Time came for some more software experience, it was wonderful. I kind of felt well acquainted with the GUI from the circulation calculations, but till then didn’t do much else, so I took a CFD course, yet something in me was different while doing the tutorials… everything was knew in perception, first time I did what the tutorial told me, the nice colourful pictures where there, but something else was happening also dazzling me… there were my library friends: types of meshes, Boussinesq hypotheses, the law of the wall, the “viscous button” also had some other friends I knew all about: Here’s Spalart and Almaras, I know exactly what they are there for, here’s Wilcox’s k-w, there’s the BSL, it combines smoothly two models, here are the k-epsilon’s family, I know why there’s a wall treatment window added to it.McDonough and Menter explained it perfectly. Then there’s the kr-w SST,I remember this one from Menter’s 92 NASA publication, it got me reminded of Bradshaw, that beyond his hypothesis that makes the model somewhat improved over the BSL for “regular” aerodynamics, had quite interesting contributions.
I felt a little weird following the instructor, a graduate MSc soon, because I knew our boundary layer for the simulation was not covering the entire log layer. I’ve checked the Eddy viscosity ratio. But he said that’s the tutorial, the mesh was already prepared, and anyway, he doesn’t know what I mean by log layer and how can I decide that just from looking at the mesh, that he’s colleague, made a mistake. So modesty I said, that I’m sorry, that maybe that’s what they want us to see in the simulation but still as it happens, if the choice is to compare the SST with the standard k-e for such a low Reynolds flow and with the use of standard wall functions will perform badly, it’s a shame that the SST can’t give the good numbers for the drag force just because the boundary layer thickness larger what the prismatic cells cover, especially as he gave an experimental result to be compared to … but then, I can repeat the simulation while taking a break from circling a vortex and c&p the resulting circulation to excel and repeat the CFD and also get some more experience at calculating boundary layers, doing that eddy viscosity ratio checkup and see whether I get a better fit. Moshe wouldn’t mind…
As far as I know, many undergraduate “CFD” courses are still awful (in comparison to the ANSYS learning hub let’s say…). And CFD is a lot more then getting a converged solution…
I get many, and I mean many guidance requests. I do like to help, I can spend a lot of rime helping students of any kind (undergraduates, graduates, doctorate, lecturers). I’m not a saint, I enjoy it. I seem to get in a state of Flow (the Mihaly Csikszentmihaly kind of Flow).
But time is not an inexhaustible resource. Many questions I would have liked to spend hours on explaining to a person eager and passionate to learn are lost. Notwithstanding the fact that most of those who actually can help would not give it even the slightest time. For them there is nothing gained. That’s sad, but true:
Now I finally get to the point 😉:
My “Recipe” for learning how to become a CFD self sustaining practitioner, Monday, 22.6.20, 19:30-21:00 (IDT).
A clear road map explaining each step to finally becoming a self sustaining real CFD practitioner:
- Beginning steps: what should be the first thing to start your FLOW from (exact resources and priorities are going to be given)
- The optimal road to continue: which subject should be dealt with before others, the exact pinpointed resources to go for (this is critical, it’s a prioritization issue, but also having the material communicated as ideally is paramount) – those who know me know that you will not be directed to a cave in a savana jungel to get the resources, staying part of my network and reaching out should do to any resource mentioned.
- Experience and Practice:
Hopefully all of us know that the C in CFD carries much importance and practice is paramount. Here, I will do something quite rare and just explain: when, where, with whom, what is to be highly valued and what should be counted as crap. This part shall include also the strategy to perform an actual simulation from the minute you open your computer to the when you have the few specific steppingstones you shouldn’t miss, then end when you had finished explaining to your superior the outcome. This is called the Strategic part of optimizing the CFD cycle. Nothing would work if you ignore it, no matter how smart you think you are, and after a few times it’s a second nature and a win-win for everybody (I owe most of my strategic view to a friend colleague, and mentor by the name of Maciej Ginalski, to our long conversation on the road and beyond. The best engineering mind, with a managerial clear view and a CFD artist I had the chance to work with).
We are all set with technical issues, the zoom has no time constrain. And it’s going to start on 18:30 (IDT) sharp.
I wish to see as many of you there.
To be able to keep count and avoid disregarding from late requests for the resources (each and every resource mentioned shall be available), registration is required.
Email for registration: Tom_Avraham@AllAboutCFD-TomersBlog.com
My “Recipe” for learning how to become a CFD self sustaining practitioner:
- Professor Lorena A. Barba (Boston university) courses (free open learning):
My advice: do not begin your CFD Voyage without going trough those carefully
- Complete undergraduates studies – The prize you get for that is walking on the wall-to-wall carpet.
- Learning Hub/StarCCM+ Self-Paced courses.
- Almost every practitioner can find someone to mentor him (and should, no matter how experienced he feels himself to be). If someone worthwhile grants you the chance for mentoring always arrive to a mentoring session ready with a set of issues the mentor can directly approach. Never waste your mentor’s time.
- The CFD Bare Minimum – a frequently updating sets of high quality courses
- Get all the book recommendations from the library and start writing notes:
- Pure Aerodynamics: Doug Mclean: Understanding Aerodynamics
- David Wilcox – Turbulence Modeling (out of print)
- J.M. Mcdonough
- Arkady Tsinober – The essence of Turbulence – This one is for those who are able and wish to get very deep into the phenomenology.
- Pope – Turbulent Flows
- Sagaut LES – Compressible/Incompressible
- Turbulence – Nieuwstadt
- Tennekes and Lumley – a first course on turbulence – a breakthrough in turbulence writing.
- Publications – be connected with the future. A good publication is based on good references which lead to good references which… well, quite understood 😉
- “All About CFD” blog – we have breadth, while still keeping our posts inclusive and communicated for different levels of CFD practice nature. Moreover, we are networking souls, we love when our readers reach and communicate, we are indeed Flowing while communicating CFD 💕
- Stability and Transition in Shear Flows – Dan S. Henningson, J.Shmid
- LLoyd Trephthen – Spectra and PseudoSpectra (my interpretation of the theory as far as the impact on transition is concerned)
- Gad-El-Hak – Flow Control, Doug Mclean: Understanding Aerodynamics
- Follow Stanford Center for Turbulence Research
ZOOM RECORDING (34 participants at most during the zoom):
My “Recipe” for learning how to become a CFD self sustaining practitioner PART I:
My “Recipe” for learning how to become a CFD self sustaining practitioner PART II: