Hi, I've been thinking about a career change recently. I currently work in IT security but I've always had an interest in Aircraft/flight etc. I've been giving serious consideration to moving into aerodynamics but would like some pointers as where to start and what sort of skills/knowledge is required.

Many thanks,

Lee

The job oppotunities in CFD are not as many as in IT. You'd better think carefully if you really want change your career.

It might be a little tough, if you look on the jobs index a large amount of them require a PhD or at least a fair bit of experience in CFD. Its not as simple as you might think and isnt like FEA analysis where you dont need to know as much about material props, turbulence models, setup convergence etc etc. It might be an idea to find out about doing a Msc or some other course where you do research on CFD. Many such courses dont need much prior experience. I am currently doing a PhD and am finding it invaluable i want to go into CFD afterwards and i think i would have found it extremely hard with just a bachelors degree and limited experience. Remember a company isnt going to be very keen on having to spend a few thousand pounds on training a new employee up, especially if there is a risk of them just disapearing to another company after a while!!

Well, I have a PhD in CFD, a strong background in mechanical engineering as well as in mathematics. But it is atm. almost impossible to get a job, if you are not very familiar with the software used in industry. And I'm not... That sucks sooooo much.

P

Its tough i suppose, i am lucky im using star-cd for my PhD so it should hold put me in good shape, using a comercial code and all (i hope). I suppose the only option is to do postdoc work using a comercial code or if worst comes to worst pay for you own training!

I don't think having a PhD makes any difference at all in industrial CFD, unless perhaps you are hoping to move into code development.

The experience I gained in the first three years of my CFD career were much, much, greater than if I had spent the same amount of time researching a nichey topic for a PhD. In fact I can honestly say that I wouldn't exchange a single year of my career for a PhD.

When we interview here I frequently find that candidates straight out of University are as knowledgable and as enthusiastic as those with PhDs.

CFD is a mainstream engineering tool these days, you don't need to have a research background to work with it.

I sincerely hope they do not have this attitude at Boeing, could lead to some unfortunate results. Good luck saving some money though.

Why should researching some nichey topic for three years make you a better engineer? For Boeing or any other company... In my opinion you'd be much better working in industry for two years and then completing a Masters degree in CFD.

Training a graduate engineer from scratch is an expensive business, certainly more so than hiring someone with a PhD at a premium of a few thouands pounds. In my experience with three years of experience a graduate engineer will earn more than someone starting fresh of their PhD.

and since when did Boeing ever appoint only people with PhDs....?

I also got this response from a few companies that I applied to after I had finished my PhD (4 years ago). Companies which have this attitude are most often consultant companies - they need to hire a "Star-CD Engineer with x years Experience", because that is what the sell to their customers.

It is not a good idea to get your first job in this kind of company anyway. So sort away any company which main requirement is that you know the comercial code xxx.

As a fresh CFD PhD you don't want to work for these companies, you want to work for a company which has a throurough understanding of CFD and a long-term interest in your development as a CFD engineer. This is most often companies which already have a fairly large CFD group and which have products for which CFD plays an important role in the design process (aerospace, automotive, ...). Just hang in there, these companies do exist and they do hire fresh PhD's without any previous commercial code experience.

Btw, in relation to some of the other comments. I don't think that you need a PhD to be a good CFD engineer. However, you do need a thourough understanding of fluid mechanics, numerical methods, turbulence models etc. and you do need to be very interested in fluid-dynamics and CFD - a PhD is more likely to have these qualities since he decided to devote several years to them before going to industry.

Well, the point is that CFD application requires some rather thorough understanding of several fundamental aspects of engineering and applied mathematics, those are outlined in Jonas's message. You do not want to just use 'the industry's standard k-e model' by clicking buttons a, b and c on your colorful interface. As long as there is an established CFD team at your place, which should probably include CFD engrs. as well as mech. engrs., manufact. engrs., etc. you can hire and train anyone you feel like, but if you're starting your CFD from level zero you are probably better of by hiring someone who knows what he/she is doing.

Agreed, but no one (sensible) is going to throw any new engineer, graduate or Phd in at the deep end. If you need someone who knows what they're doing than your best bet would be to recruit someone from industry.

If you are going to make a career of CFD then you need to understand the fundamentals. I certainly learnt most of those fundamentals on the job, partly from my own experience and partly from the more experienced engineers around me.

In 10 years, my CFD career has taken me from consultancy, to industry, to a large code vendor. Particularly in the last role you need to know the fundamentals inside out. If I made a list of the best five CFD engineers I have worked with in that time, only one would have a Phd. Thats not to say that getting a Phd isn't a great start for a career in CFD, but it certainly isn't a necessity.

Agreed, good points.

hi fellows, Interesting points about having a career in CFD are discussed. Well in my opinion it all boils down, how you market yourself. Well obviously an industry prefers a guy who can do the job without any training. Well ideally speaking the so called "experience" gives us the rule of thumb, what to do and what not to do. This everyone gets through involving in multiple projects. you will get burned up several times and every time, you develop your own way to beat the problem. Normally a Phd guy out from school is not versatile in dealing with application problems where you have to optimize accuracy with time. And also after a certain extent of knowledge it boils down to how better and consistent your results are. Very importantly, infact truly no industry is looking for accurate results, all they look for is whether the particular design works or not, how better you can improve the design , and how fast you can do this. In other words today's CFD world is all about optimization. To be precise if u r fresh out of school, grab the basic concepts, get to an industry (well market yourself) and you can eventually be a good Engineer with CFD tool on your hand. And above all the one thing which makes all the difference is your feel for fluid problems and interest in CFD. It is really a wonderfull field guys, You will really enjoy it once you are into it. Sundar

If you want to develop CFD solvers, then you pretty much need a PhD or need to have had many years of experience developing in industry or in your own time as well, any of these would do the job. I think it just happens that most poeple that develop CFD solvers have spent all their time in school doing the necessary work to learn how. It actually makes me sad to see people using a commercial code in their PhD if the primary focus of their research is CFD oriented. I would never hire anyone who did this.

At the company I work for most of the people working direcly on solvers have a PhD, say 90%. Outside of core solver development the number of PhDs goes down significantly. It's probably on the order of 15% of the people. So, your chances of getting a job doing support/service/consulting or non core solver development without a PhD is quite high. Of course this does not rule out getting a job doing core solver development without one though.

Neale.

Joel, you don't have to listen to me, but if CFD research is the primary focus of your PhD, I think you are making a mistake. Pick a topic and do something completely original, you will benefit more in the end because you will learn more.

Trust me, if you want to do CFD development work in the future it does not matter if you know a commercial code or not. I work on the flow solver for one of the major vendors and before I got this job I had never seen or used commercial CFD software.

Neale.

I had just posted my own situation where I am looking for career advice. I have a BS. NO MS or PhD. I have dealt with many CFD engineers with PhD's, many of which I have learned from. Some have had the belief that an MS or PhD is necessary-Their now the same people who come to me to fix their problems! I learned by doing CFD for free at first (while in school). I continue to study and am very driven in solving my models so that I can influence the design.

I agree that CFD is a tool for engineering. I figure I've accomplished the equivalent of 2 or 3 masters degrees while applying cfd to industrial problems. I succeed because I spend the time to know the equipment- to know where I can use the cfd solution to identify the problem. Then I design the solution that saves money and labor with the constraints I'm given.

The big challenge is that very few people I work with understand what can and can't be done with CFD (at least not in the time I'm allowed).

I actually have the opposite problem. For five years I've been developing an airfoil design/analysis package and only have a Masters in Aero, not a PhD (all of my competition does!) For some customers, it doesn't matter that mine is more accurate than some and much easier to use than all of them.

However, I have nearly a decade of experience in wind tunnel testing and many of my customers appreciate that I incorporate that experience into the code interface.

Just be thankful that you have your PhD now; you won't have to go "back to school" someday when it's extremely inconvenient.

John Dreese

Maker of DesignFOIL airfoil analysis software.

www.dreesecode.com