I hate it-but I love it
So I studied CFD in grad school. Not a choice really, it was what my school specialized in and I was going part-time after work so I studied what they offered. That was four years ago and I took my degree and never looked back at CFD, until now. I remember learning that you could not solve the full average NSE for turbulent flow (too many unknowns and not enough equations) and thinking "this stuff don't work and I'm not going to waste any more time on it".
I interviewed for a job a few weeks ago and when I showed up they wanted me to interview for about 20 positions and over half were for CFD related work. Maybe this stuff is worth while so I go out my textbooks and notebooks from class and have been pouring over them for the past two weeks. I have been reading in the afternoon on my spare time and it is making sense this time around. It is down right cool!
No real point here. Just seeing if anyone else has similar experience with this.
I agree with you - CFD doesn't work in theory, but once you know a bit about it it is quite cool. My story comes from the opposite side though.
Many years ago, when I started at University, I did so with the very clear goal to study and eventually work with theoretical physics. However, after a few years and after having read several courses on quantum physics I slowly lost the passion for that side of science. At that level theoretical phsysics just becomes statistics and incomprehensible equations without any clear physical meaning.
After three years at University focusing on theoretical physics I was longing for a course with a clear physical interpretation. By chance I saw a notice about a grad-course in turbulence. Turbulence sounded cool and something more concrete than quantum physics - you just had to play with a milk drop in your tea or watch the flow in a river to see how vortices broke down and eventually created turbulence.... So my first ever course in fluid dynamics was a grad-course in turbulence ;-) It wasn't quite as concrete and down-to-eath as I had imagined.
However, the balance between a foundation that is very theoretical and not fully understood and applications that are very down-to-earth and easy to understand is what makes CFD so interesting I think. In CFD you need to be a good engineer, sometimes even an artist, to produce results that are any good. Yet you can sit with your 4-year old son and watch the flow in a creek and he will marvel and to some extent understand the vortex-dynamics that you try to explain to him.
Actually your career path sounds much like mine. My undergrad experience with fluid dynamics generally (undergrad physics degree) was 1 short course - taught, ironically, by a retired engineer - and there was very much an unspoken viewpoint that physics = quantum mechanics. Went off to do a PhD in Astrophysics, and I found myself writing programs to solve fluid flow problems (gas dynamics in galaxies) - and found it fascinating. In fact I discovered;
1. Fluid dynamics is fascinating
2. There are plenty of unsolved problems, particularly in turbulence
3. CFD was just starting to really take off at the time (some might argue but I think that is probably the case) and it was a good moment to get into the subject!!
I'm constitutionally a generalist - I like solving problems but am not good at concentrating on one small area - and CFD, where we need inputs from maths, physics, numerics, computing, engineering... etc ; suits me rather well.
Several decades ago, as an apprentice (do people still know what an apprentice is?) I returned from completing my undergraduate degree with the intention of performing a postgraduate study in numerical simulation. I first approached the stress office where I had worked earlier and asked about their supporting a PhD study. Although willing I would have to sort it out myself because they had not internally supported a PhD study for about 10 years. Linear stress analysis at this time was a substantial and active development exercise for high tech companies (the switch of the bulk of the work from internal to commercial codes was still a few years off) but was not a significant research topic to the extent of involving external universities. Almost all the research effort was focussed on nonlinear problems which had little industrial interest at the time because the computational requirements were too large.
The development of CFD was in the process of moving from methods adapted to specific geometries like aerofoils to more general complex geometries. In this area I did not have to sort things out myself.
Today most engineering companies would consider it crazy to employ groups of people to develop internal CFD methods but humps in demand for CFD are likely to continue as computation continues to become cheaper making CFD a viable option for progressively lower tech companies.
Jola and grtabor- similar story here!
I studied astrophysics at undergraduate and then started a phd on particle physics. It became clear that all I was doing was spending my days moving chunks of someone else's code around.
Stuck it out for a year and then moved to an engineering phd that has led me to CFD. Now I'm dying to get a job for Brawn GP designing semi-legal diffusers!
|All times are GMT -4. The time now is 17:56.|