Hi, I have a question regarding average salaries and the trend of salaries within the CFD community. My impression is that considering the computing skills requried the average salary is very low £<25k. When you compare this to the average in the IT or Engineering fields (£35k) there's a difference. Does anybody know why and is there any benefit on going contracting. Upto now my interest in CFD has been purely academic and I'd like to make a career of it but are the financial benefits too limiting!?!

If you think about the actual worth of a CFD engineer I would place it about £40k - £60k per year. This I do agree is slighty high but considering the education and computer skills requried it would put it at the correct level. I've been to one or two of the City bank presentations and they're offereing very large salaries (£>80k) for qualified individuals.

I think your probably looking at £25k for graduates specialised in CFD with a years experience raising to about £30-35k for about five years experience, raising to about £40-50K for very specialised PhD persons. Unless of course, you land a job as the director of a CFD company.

Contracting would be good rates, however if you want to by your own system the software and hardware requires a substantial investment and there does not seem to be that much contracting work around for individuals.

Teaching CFD is an option, good rates of pay, status? and long summer breaks where you could do a bit of consultant work.

If you want to be rich, spend five years creating your own software, sell it for £alot on an annual basis, seems to work for others.

(1). I agree with you. Your observation is correct. (2). Before commercial cfd code era, say before 80's, it was normally required to have a PhD degree in CFD and real experience of writing a full CFD code, to get an engineer job in CFD. Most of the time, the job was in the advanced engineering or research department. In those days, that training was minimum. Without that, there is no other ways. It is between you and the super-computer. (3). In 90, two things have evolved. One is the so-call commercial CFD codes, and the other is the Internet webpages. In the Internet webpages area, people got the impression that they can learn just a few off-the-shelf codes and be able to create impressive webpages. Every company loves to put their company on Internet, so the market is created, because the webpages have to be created by IT programmer. There is no other automatic way yet. This is labor intensive field. So, even this field is moving to the 3rd world countires, like China, India. (4). This fast-food approach in Internet webpage business, coupled with the available commercial cfd codes, contaminate the real nature of the CFD research. (ask the big boss of the commercial cfd code vendors about that. then you will get the answer like, business is business. commercial is just commercial. It depends on how one look at it, from your end or their end.) (5). This 90's wave of commercial cfd code approach is failing many companies without experienced and well trained CFD expert (PhD level). It does not just happen at the small company level, it is also having very serious impact on world level big companies, with 50 to a couple hundred years combined histories. It is a CFD war. It is there already. (if 99% of cfd code users have not even obtain the mesh independent solution of a turbulent flow through a fully-developed channel, then it is a joke to talk about getting a 3-D solution.) (6). To be born again, the industries in general have to die first. No pains, no gain. My suggestion to you is: life is not a straight line, you have to get a job to keep alive first. If you are really in love with cfd like me, 20 years from now, you will still be doing cfd. Money is not everything. Bill gates can easily create several world class cfd research center and keep scientists/engineers in cfd very busy for the rest of their life, or he can just give away his money to non-profit organizations. (7). I would say, from now on, this forum is going to change the trend of the world in CFD. And it begins to change. But it took two years to get people to think "mesh independent solution"... not yet doing mesh independent solution. When all of these cfd requirements become clear to the general public, they will be forced to demand the quality of cfd expert. At that time, you will be responsible to lead the field in a company. Will you be there?

Noel,

I'm not nearly as familiar with Pounds as I am with $s (I can't even find the Pound symbol is on my keyboard), but I know a little bit about pay rates in the U.S. for engineers.

Most large companies base their pay scales more on level of experience than on area of expertise. My first job was at a large, defense oriented shipyard. Engineers were grouped into broad catagories (Associate Engineer, Engineer, Senior Engineer), the pay within those catagories fell within a fairly narrow range. Engineers who specialized in design were paid roughly the same as those who specialized in FEA and CFD.

When I left a large company and went to a much smaller consulting company I found that impact on the bottom line was reflected in compensation. I worked with an engineer who had served in the Navy, and worked on his Batchlor's degree, but never quite got around to finishing it. Despite that, he had a natural flair for engineering. He would go to a client's plant, look at whatever was causing them difficulty, and know how to improve the design. He was very well paid for his abilities.

I think that the point is that different types of companies want different types of employees. Many large companies want a person to fill a space on an organizational chart. They tend to have very regimented compensation practices. If you have an MS in mechanical engineering and 2 years of experience, then you are an associate engineer, and your salary falls within some range. If you are good at what you do, you'll be at the top of the range, but still within the range.

Smaller companies tend to look for people who can make a difference. If you can get a product to market faster, or bring in more business, that makes you more valuable than someone more complacent about his job.

Regards, Alton

(1). In old days, the CFD expert(PhD) develop and control the code. The engineering aids run the job deck for him to get the output, and do the plotting as well. (2). in 90's, The engineering aid run the code directly from the computer system, obtain the results and plot the data, all by himself. Naturally, the salary is at the much lower level. (3). The company figured out that if the code is supported by a company, then there is no need to invest in the cfd expert in-house. And if the engineering aid fail to get the right answer, the company can easily find another engineering aid at a lower rate. This will go on for a while until the company finally founded out it can not stay in business this way. But then it is too late. I have been seeing this trend for the last 5 years.

Noel and others,

For once , I think Chien is really to the point. If one can just click and find his way to flow solutions using commercial CFD codes, I see no reason why companies should worry about paying hefty sums to 'experts'. I am new to commercial CFD codes and am still sceptical about their ability to do so when handled by CFD and flow novices, but perhaps I am wrong.

If the e-doctor can merely ask you a few questions and you can click and find your prescriptions, I am sure the salary of the pathologist at least will drop heavily. Of course the real surgeon cannot be challenged by the e-surgeon, but can we imagine ourselves to be CFD surgeons?

Salaries anyway are just determined by the market demand/supply. If the latter is favorable to the IT chaps, few can whine.

ravi

I think this is true for post-doc and research associate opportunities not for industry.

The salary scale in the commercial CFD companies (as I work in one of the top ones) depends on whether you are working in the development, support or sales.

For CFD development engineers:

New graduate PhD: from 25K pounds in UK. 60k dollars in US.

PhD with a few years experience: from 32k Pounds; $70k.

Specialised in one particular (expert/good CFD developers): upto 55k pounds; upto $100k.

If you think you are good CFD developer, why don't you try out.

(1). I think, a good CFD developer is hard to quantify. (2). a CFD developer means that a person is developing a marketable cfd code, which is not every bodies's goal of doing CFD. (design, analysis, teaching, research,etc...) (3). So, I would say that a good CFD developer is actually two-jobs-in-one. That is, he is a person with math background in CFD, and also a highly skillful programmer who can transform the algorithms into a working code at the commercial level. (4). I would say that the training of a CFD expert(PhD) is not to turn him into a programmer, but to solve thermo-fluid dynamics problems through research, design and analysis. (the codes he developed may not be easily marketable, but are sufficient to solve the problem.) (5). Sure, if you can come up with a product, there is a better chance to be rewarded.

I support the views put forward by John Chien. Nowadays, so-called cfd experts have a feeling that whatever colourful picture you get out of cfd package is right. Yes, it is right as long as it is converged solution. But does it tell you the story of fluid dynamics? At least as much as one would expect from the chosen physical modeling. My observation is, a good number of 'cfd experts' are really too naive to understand what is wrong in it. This kind of cfd analysis has no credibility and designers in serious industries probably do not believe the results either. The reason for the low salary of cfd people is this. If any one can produce anything, that's got to be cheap.

In reality, establishing a cfd result needs systematic and careful validation. Unless the people in wider community realises this, cfd will remain a software for producing just pictures not science.

(1). Two years ago, when the forum just started, I posted a series of messages under CFD War Principles. (2). In one of the messages, I cited the example of a nurse. To work as a practical nurse, you need a practical nurse license. Naturally, you have to pass a series of exam. To work as a registered nurse, you have to get some kind of college degree and pass a series of exam also. And even if you are not working, you still have to keep taking refreshing courses to have the licence active. (3). I don't know whether it is absolutely necessary to get a license to become a nurse, But I am sure that the quality of service will be maintained if every registered nurse has a license. (4). If you don't care, nobody will. (5). This is an old issue, but I happen to be highly visible here. And I hope that the CFD industries will follow the same trend, that is to become a high-quality oriented service, rather than a cheap pirate copy of a CFD code.

At a recent group meeting from one of the larger commercial codes, two types of cfd engineer were defined. Those who work with the code to get it to do something that the code doesn't currently do(experts) and those who use the code as a design tool(users). I think there are a few more bands of cfd people: i) those who can get a code to run something and produce pretty pictures to keep marketing dept happy ii) those who understand something about what they are running to produce those pictures iii) those who can relate what they are doing with CFD to the actual problem that they are trying to solve iv) those who, knowing the problems with the CFD code, can adjust and validate it to actually solve their problem and can do it in a timely manner v) those who, so concentrated with the fine detail, lose sight of the bigger picture, solve the problem but take a year to do it vi) those who would rather someone else do the above and use a consultant.

Obviously salary is dependent upon personal capability and experience.

To the point, Star CD recently advertised posts thus: CFD developers: just qualified phd's £25k more experienced developers £32k-50k .Higher salaries for 'leaders in key areas'