Dear all,

I'm an engg. student pretty interested in CFD at the moment. I will probably pursue an MS in this area and maybe even a PhD.

I was just wondering about the kind of money a CFD specialist could expect after an MS or after a PhD? Can we compete with the IT industry in this regard. I hope my question doesnt sound too materialistic. Anyway....

You should be damn good at CFD if you get a little after your PhD. Otherwise none, I guess.

-Selina

Hi, thanx for quashing my aspirations!!!

If you were a carpenter you'd sell yourself more effectively to prospective employers as a maker of furniture rather than a user of hammers. The application is important, CFD is only a tool.

Fred.

Hi, thanx for quashing my aspirations!!!!

Hi

I appreciate and understand your reply, but you havent really answered my question. What could I hope to get and which applications would command more etc?

My answer is going to be general; one that I used to give my students (way back when) at the beginning of the semester:

If you're in engineering for the money, then you're in the wrong field and it's still not too late to change directions. Take up real estate, plumbing or open up a mom-and-pop grocery store and you'll make more money putting in fewer hours.

This holds true for CFD as well.

I will not charge you for the following free advice ) You should absolutely enjoy what you're doing, no matter what the field is and what the money is. If you don't enjoy your work it will show very quickly and you can easily lose your job (and the "good" money that came with it). If you enjoy your work, at the very least you'll be happy with what you'll be doing day in and day out. Usually, that translates into eventually becoming an expert in what you do (and in this case the money will follow)

I know someone who became a dentist for the money but he was miserable. Having spent years studying and even establishing a good business, he finally quit the field and went back to what he always loved to do: painting! He is a struggling artist now, but a happy one.

Usually, these examples don't mean much to students but you ask anyone who's been working for more than 5 years whether money is number one in their list of what's important in a job and the answer is invariably "no". A job that challenges you on a daily basis is the most important factor.

Adrin Gharakhani

T,

You concerns are very legitimate given where you are and the current status of the economy. Depending on your luck (and many other factors like the field, adviser etc.), Ph.D can take a long time. After that, you should have a sense of accomplishment and not regrets. Not having lot of options after Ph.D does not help even if you are interesting in CFD more than anything. The fact that you are asking these questions means you are not too materialistic but are not a idealist either.

I do not have an answer for you but here are some observations/suggestions that might help you.

1) CFD has evolved rapidly over the last 20 years. By the time you finish your Ph.D (which takes about 5-8 years after BS), it can change more than one might expect. In the 80s, CFD was predominantly an aerospace field with lots of research funds from the government. There was some commercial interest mostly in the automotive/mechanical engg. in areas like heat transfer, phase change etc. Research was focussed lot more on scheme and code development. The work on scheme development has come down a lot. Most of the present interest seems to be in application of the scheme and physical modeling (i.e., sub-models for combustion, turbulence (DNS/LES), radiation, two-phase flow effects etc.). Also, the commercial CFD companies have grown tremendously in last 10 years. The people who joined these companies early make good money (so I heard).

2) CFD is not as mature as some other modeling fields (like FEM for heat transfer, structural mechanics etc.) and so the research in CFD would remain for a long time. But as I said, most funding is in applications.

3) CFD will become a rapid prototyping (design) tool in many fields. It already has in some areas. I have heard of CFD use in manufacturing, electronic cooling, biology, chemical industry and some unexpected processes like making potato chips, ice-cream etc. However, you may not like running commercial software for livelihood which is what most of the industry requires.

4) IT industry fluctuates a lot. The latest I have heard is that there are still a lot of jobs in the field and the field would still be growing for a long time. But, apparently, the astronomical salaries that some people used to make are long gone. Still, I think they make more than a CFDier at entry level and the next 5 years. In the long term, I am not fully sure of the kind of job security in the IT sector. 5-10 years beyond Ph.D, you will be making good money and engineering jobs in general offer a better balance between money, job security and satisfaction (that is if you are still interested in the field).

5) you need to keep up with latest stuff (software, trends) in the IT industry. People in CFD do move from one area to another but in a more gradual way. After all, research is hard and you are not expected to become an expert overnight by reading books/papers/monographs. CFD is not exact science. There are lot of tricks, well known concepts that are rarely seen in print in the field. You have to develop some wisdom in the field before you claim an expertise. Most IT skills are easy to pick up but there are so many. My brother works in a software consultancy and is constantly learning new software or languages. I suppose it is not very difficult, but can get annoying.

6) You can make money in any field if you are good and are a good salesman. If you can not sell yourself, partner with somebody who can. Some CFD professors in private universities, CFD manager in industry, partners in CFD consultancy/research firms (which rely on SBIR grants) do make good money if they get there.

7) If you have a tough time making a choice, take some time to talk to people in all the areas you might be interested in. You can do this while doing an MS. Once you are a couple of years into Ph.D, I am afraid you may not not have much exposure to anything other than your lab. In my case, I did not have a clue what sort of work guys in the next lab were doing let alone people in other engineering disciplines. If you interested in research, remember, CFD is not the only field. They are fields (especially in elec. engineering like networks, DSP etc.) which are as interesting and mathematical but are also more commercially oriented.

(1). On the cover of the current issue of the Industry Standard. it prints "Boeing drops a Bomb" (2). On the page 40, it says: The aviation giant is going from bending metal to building networks. (3). The CEO is trying to meet the investor's expectation of 17% a year(EPS).

As far as I see this I experienced the following: The question is not whether you can make money in CFD but in which line of bussiness you're in. If you work in academia or in industry where cost of labour is a mayor factor than wages are relatively low. f.i. in a consultancy almost all costs are wages. In bulk industry (chemicals, oil, mass production, aircraft?) wages are only a small percentage of the total costs for a company so wages are normally higher. If you want to make some money while doing the thinks you like ("All I ever wanted to be is an engineer", Scotty/Startrek) go work for these industries. This does not include the factor whether you are any good. Anybody agrees/disagrees?

Hello T the answer is so simple. I don't understand why every one has missed it. Go for "IT". With a BS/MS degree in IT, if you are good, you can work for many different industries including the CFD industry. The salaries are of coarse higher for IT than they are for CFD. IT industry is the back bone of CFD. If there is a need for CFD, then there will always be a bigger need for IT. We can hire (today), CFD specialist with a PHD @55-65k/yr. We cannot hire a good MS "IT" with two years of experience with less than 80-85k/yr in the same city.These are the facts.

(1). This is because the market is very big for "IT" people. (2). If you translate the CFD into a more appropriate name "Numerical Analysis and Mathematical Modelling in Fluid Dynamics", then right away you know that it is a very specialized field. (3). Therefore, it is incorrect to think that CFD is in anyway related to "IT" field. People in "IT" speak in "C++", while researchers in numerical analysis are still speaking in FORTRAN.

(1). You don't need a PhD to make more money. (Bill Gates is a good example) (2). On the other hand, you can't buy a PhD with money. (3). So, if you like to have a PhD and a lot of money, it is a good idea to get a PhD first. Then you can forget about what you have learned in PhD, and concentrate in making more money. (4). If a company making small computer CUP fans can get in the CFD field and selling commercial codes, then I guess, you can start selling watches in the flea market and then move on to sell CFD codes. The market is there, because there are millions of engineers out there who have not used any CFD code yet. (5). So, there is no correlation between a PhD and the amount of money he is making. But in general, a person with a PhD should be getting more in salary. (6). And you can't ask a person how much he is making, unless he is in money making business or service. In that case, the amount of money he is making represents his capability to provide better professional service. (7). CFD itself does not deal with the issue of money making directly. Even if you can win a contract, does not mean that you will be able to provide a CFD solution. And even if you can provide a good CFD solution, it does not mean that the aircrafts will be delivered on time, without strike or parts delay. (8). So, if you still do not have an answer, take a bike ride and come back here so that your brain can have an opportunity to exercise.

> (1). You don't need a PhD to make more money. (Bill Gates is a good example) (2). On the other hand, you can't buy a PhD with money.

Agree with point 1, and disagree with point 2! Quite often very famous universities give out honorary PhDs to "famous" people. Now, they may be famous due to politics, art, science, and _even_ because they are rich and have donated huge sums to the university (bought their PhDs)

))

Adrin Gharakhani

(1). So, there is another way to have PhD and a lot of money, and with the money you can buy a PhD. I think, it is a practical approach. (2). Then, after that you can still come back here and ask a CFD question. (3). Perhaps, this is a better approach. It is easier to make money, but harder to ask the right cfd question. "Numerical Analysis and Mathematical Modelling in Fluid Dynamics" is reserved only for very talented and hard working people.

That is the whole point. "CFD" (or what ever that you want to call it) is a very specialized filed. AND a new engineer thinking long term (as T is) will be a lot better off getting a degree in a broader field where his/her knowledge can be applied to a greater number of applications (Including CFD).

BTW, I don't know where you are getting your information, BUT CFD people today ARE thinking and doing C++. REMEMBER CFD is not just a solver (and hence the name numerical analysis). There is a lot more to it than just a solver. AND yes, some one with good knowledge of C++ is a great asset to the CFD industry as well.

(1). I think you are right. And a school can not survive by creating a CFD department. As a matter of factor, there are very few aerospace or aeronautical engineering Departments. (2). At the MS level, one normally receive only the basic training in advanced calculus, partial differential equations, fluid dynamics and boundary layer theory, heat transfer (conduction, convection and radiation), and thermodynamics and statistical thermodynamics. And these are rather basic stuff. You can use these to solve a real problem directly. (3). At the PhD level, you will get additional training especially in reading the technical papers. At this level, you will have the needed background to read and follow the material presented in the paper, which is the first step to learn how to eventually solve your problem in your research project. (4). In the "IT" field, there is no need to have any advanced degrees in order to provide good services. This is the other extreme of CFD. In CFD, the current commercial trend is to create a commercial code. On the other hand, in "IT", the trend has been to create a language, so that people can use it. This create more applications of the new language and more jobs as well. The increased activities promote the health of the "IT" field. (5). So, it seems to me that the commercial cfd code development is following the old mainframe computer approach, where the goal is to create a better code or a better computer. That approach to the computer hardware and related software industries are gone. And This is the same reason why MS is having problems with the courts. (6). So, a healthy industry is one which provide common technical platform to many people, and not to implement a common solution approach. There is no such thing as "which is the best C++". (7). If you say that you have ten years of experience with a IBM mainframe computer, then I think, your job is probably not very secure. On the other hand, if you say that you have five years of experience in C++, you can probably find many high pay jobs today. (8). CFD problems are always specialized problems. If people are still trying to solve a simple flow over airfoil, then it is clear that one can not generalize the solution of it to other flow problems. (9). Most commercial cfd vendors have at one time worked for large companies in the past. Since that home base has long gone since late 80's, they are trying to stay alive in a different form. That's not their fault. It is because of the changing time. It is not possible to follow their footsteps, by repeating the history. (10). So, I like the idea of the "flea market" for the cfd forum. You simply do not ask the owner about the product he is selling ( a black box). If you like the code, just pay the cash and you have it.(no warranty provided, and the price is not fixed either. it is cheaper if you are a student.) (11). Once you have this overall picture, and still have the time left over, then come here and relax.

Here is my experience:

I'm just finishing up my MS in chemical engineering. It took me 2 years; in that time, I learned to use Fluent and CFX. Once a few inventors in my city heard about my skills they came begging for a simulation of their 'machine'. Each one came with $5000+ for a few days of work. Not bad for someone with no experience and little education (it paid for my MS degree and I'm living well).

I don't know if there is money for you, but everyone around here is begging for my time.

Looks good so far.

When I trained on CFX (several years ago), a full day of consulting by a CFX 'expert analyst', including code use and exclusive use of a high end work station, cost $1,500 at the CFX training center in Pittsburgh.

What did you have left of your $5,000 after you paid for the computer time and the license use? If they were 'free', you and your clients got a good deal.

Would you be able to work this way over a long period of time?