General question,

just obtained MEng in Aerospace Engineering and thought about broadening my options for future job applications therfore I am thinking about doing a PhD in the CFD domain.

any CFD engineers or experts acn provide me with advices and recommendation from their experiences will be very much apprecited.

Waiting for your replies.

Cheeers.

Steve.

In my opinion, where you go from here depends largely on what you want to do in the field of CFD. And, you need to decide if you're willing to push through 4-5 more years of school.

I don't feel a Ph.D. is necessary to work in CFD. For many companies, your current degree (and of course, at least some CFD coursework/experience) would be sufficient.

However if your interest lies in methods development or research, a Ph.D. (or a masters with several years of experience, perhaps 5+) becomes a requirement.

Either way, I personally feel CFD is a field rich in opportunities and being either an applications specialist or a researcher/developer makes you very marketable.

Get out and get a job - nothing like experience with real applications rather than some namby pamby degree.

Of course, finding a good job that gives you some degree of responsibility to start with is the trick. Look long and hard (and not just in the interview).

A PhD should give you some understanding and feeling for the methods behind CFD, because you are more likely to develop something rather than just apply a commercial software package. One thing I noticed about people with purely industrial CFD experience is that they tend to trust the software a lot more than somebody with a little insight. People would take a non converged and first order accurate solution and base engineering decisions on that. In my opinion academical experience might give you the necessary distrust in CFD. Don't get me wrong, CFD is a great tool, but you should know the limits!

O.

I would beg to differ with O. Using CFD implies knowing the techniques behind it and also knowing the phsyics that is involved. Having a PhD doesnot help any better. It's understanding the numbers that is more important and not the degree. Moreover having done MS I am sure you (or me or anyone) are well prepared to deal with it. amol

The degree does not make any difference - I completely agree with you. I personally know people who have a PhD and whom I would not trust to give any real world CFD/fluid dynamics task! What I meant is that having done some research in the field of CFD helps a lot. People who have no code development (research) experience have a tendency to not question the results. Where and how you get that experience doesn't really matter. What I was trying to say is that if you go to industry with just the theoretical background from CFD classes, you might have to work on applications and you might not get the time to aquire a 'feel' for numerical fluid dynamics by writing your own stuff. But yes, the degree itself doesn't mean a thing!

Doing a PhD can be fun, though ... and it leaves the door open to academia!

O.

Experience is the key, whether that's through a PhD (where you're likely to be developing a CFD code using Fortran 90 or C++) or through an industrial position using a commerical CFD code (where you're likely to input the mesh and initial conditions using a GUI into the "magic box" and recieve the results out of the other end).

These days though, you're not restricted to just academia OR industry. One idea could be to do a part-time PhD over a couple (5+) years in association with your industrial employer and a suitable university while still working for the company (most decent companies will encourage this as they will reap the benefits, and universities are always looking for industrial partners!)

An alternative may lie in a Knowledge Transfer Partnership KTP program (formerly known as Teaching Company Scheme), which (I think) you can undertake as part of a PhD program. In the KTP program you work with a company for a period, working on a pre-defined project. You have one industrial tutor at the firm, and another academic tutor at an associated university. This gives you the best of both worlds. One example, I guess, might be that you'd be developing (part of) an in-house CFD code for the company using the CFD knowledge at the university as a basis.

Personally, I would say that working with a commercial code for a year or two (on a proper salary) before a PhD would be invaluable anyway when you get around to starting your PhD. It gives you a sense of what industry is actually looking for in a CFD code!

CFD can be very rewarding and looks quite cool but it can also be VERY frustrating waiting for the code to converge, especially when developing the code (little coding bugs that only appear after a couple of days of CFD convergence!)

The message, I guess, is to get as much hands-on experience of different practices and disciplines within CFD as you can before, during and after any potential PhD.

anyway, good luck with whatever you decide to do!

Well I am not an experst in CFD but if you are interested in research, then do try finding out more about physical aspects of fluid dynamics. You may choose from some of the following branches: .astrophysical fluid dynamics .geophysical fluid dynamics .micro/nano fluids .transition to turbulence .meteorology . .

Moreover, CFD jobs in the industry are more about knowing how to use particular code or package.

Amol, you are saying this because you dont have a Phd.

I said that because I don't want to do PhD at this point. I am still in MS and both options (job, phd) are open for me. I said based on my research experience in using a commercial CFD software since last 1.5 years (i know thats not much). amol