(1). What about the turbomachinery?

(2). And I guess you have never heard of compressor blade failure, combustor structure failure, and turbine blade failure. (3). They are all structure failure, caused by the thermo-fluid loading in terms of pressure, oscillation, temperature non-uniformity,etc...

Everybody is going far beyond the initial topic so it is better to stop talking irrelevant.

John,

If you look at the different types of PE licenses that you can get (all you gotta do is pass the test), there are ~10 different types. The ones I remember are: Mechanical, Electrical, Civil, Nuclear, Chemical, and Structural.

Someone who is a "Professional Structural Engineer" designs bridges and buildings, etc.

If I were looking for a PE to design turbine components, I would go looking for a "Professional Mechanical Engineer".

BTW: I worked with a guy a few years ago who had his office in a tempoarary building (i.e.: shack) set up on the turbine deck in a power plant. He once had a blade land on the roof of his office. Apparently it was a highly unpleasant experience.

Alton

(1).Jet engines power commercial aircrafts as well as military aircrafts. Tanks are also powered by turbine engines. (2). Power plants are powered by turgine engines, gas, steam or nuclear. (3). In these areas, many problems are not usually published because companies involved and the government officials (any countries) do not want the public to know. (4). It is reaching the point that, if you can't get it right, it will be wrong. I think, the technical failure is mainly the source of the market problem. (in addition to the competition in the market) (5). If two companies with two hundred years of experience in making car and tires can't pinpoint the SUV design problem, then I guess, the high tech is not reliable at all. (or they don't know how to apply the high tech properly.)

(1)."how much the going rate for such CFD analysis" involves: CFD analysis, specific area of CFD analysis, the going rate for such service, which is a contract for delivering the service. (2). such CFD analysis service must provide useful results and must be validated and so that it can be accepted to get paid at certain rate. (3). So, the key issue is still the useful and acceptable service which can be inspected and validated by the client. Without that, even if there is a going rate, the contract will fail. For example, can the person who provide the CFD analysis service provide a money back warranty, or 90-day warranty? so that if the client is not satisfied, he can get his money back?

(1)."how much the going rate for such CFD analysis" (2). So, the issue is really the definition of "such CFD analysis" and the going rate for it. (3). I think, once we have the proper definition for such CFD analysis, then the going rate should be easier to establish. (4). If it turns out that "such CFD analysis" requires something like a three-year PhD level research, then the rate would be high. And I think, the probability is rather high for such CFD analysis. (5). If such CFD analysis is just a non- converged, non-mesh independent solution, then I would say that there will be never-ending discussions here. (6). Well, if the cancer patient has only three months to live, then obviously he is not going to wait for the converged solution. In that case, the going rate will be anything that the cancer patient is willing to pay.

Thanks for the input...I hadn't intended to stir up a hornet's nest. Rest assured that I am aware of our ethical obligation as engineers, and will make every effort to be conscientious...

(1). Your question will always be used to get discussions useful to the rest of the readers.

(1). As our readers decide to go public and ask questions here in the public CFD forum, the questions and answers are no longer personal. (2). The answers will be related to only the questions asked, not to the person(s) who posted the question(s). (3). We hope that this is a common understanding. (otherwise, we will be saying only "nice things" here. even occasional somewhat incorrect answers are good opportunities for our readers to exercise their brain power and to stimulate their thinking process and decision.)

This is quite funny.

You seem quite happy to take engineers rules of thumb and scale-up, but not CFD solutions.

In general, there is a place for both - they do complement each other and should be used as such.

I think companies used to not use CFD to save money. Perhaps now, they try not to use phyiscal models to save money.

As has been mentioned, it boils down to the CEO wanting to make more profit to make the're rich shareholders happy. I would have thought that we bear responsiblity, by demanding cheaper goods and services.

And even more so to our Amercian cousins, who saw fit to vote an extremely right-wing president who's only mandate is to make his rich backer richer and forget any other people.

But I digress.......

This is quite funny.

You seem quite happy to take engineers rules of thumb and scale-up, but not CFD solutions.

In general, there is a place for both - they do complement each other and should be used as such.

I think companies used to not use CFD to save money. Perhaps now, they try not to use phyiscal models to save money.

As has been mentioned, it boils down to the CEO wanting to make more profit to make the're rich shareholders happy. I would have thought that we bear responsiblity, by demanding cheaper goods and services.

And even more so to our Amercian cousins, who saw fit to vote an extremely right-wing president who's only mandate is to make his rich backer richer and forget any other people.

But I digress.......

(1). The drop in stock price says that merger is more difficult now to cut workers and to make stockholders happy. It is getting harder to buy another company after 10 years of trend. (2). At the same time, because the share holders were happy due to merger instead of new products, the technology in making good product is way behind, because experienced workers have long left the company for cost cutting purpose. (3). So, the only way to make money now is to use mass production to lower the unit cost. So, in the last few years, companies have been investing in production lines and also moving production lines to locations with cheaper labor cost. (4). With the products mass produced, the supply is now over the market demand. More companies are looking for limited market share. This has been the case in dot.com, memory IC, PC areas. (5). The over-supply of services and products with lower quality will make the market situation even worse. The old rule applies," garbage in and garbage out". (6). A few not-so-good CFD solutions is not a problem. But if everyone can generate not-so-good solution with parallel network, it will become the problem like dot.com and IC chips, not to mention the aerospace industries. (7). Quality product and service at market rate is the way to go. The current stock market problem only begin to reflect the fact that real technology development has been ignored for a long time. (8). In CFD, if you have been using a commercial for the last five years, then you have not done any CFD home work for the same period of time. And it will take at least ten years to get back the track to compete against the person who has been working on the CFD technology for the last five years. (9). It is a very simple fact. Merger, cut costs and make money now, after that the hard time come and the company dies. The hard time comes from the competition, over-supply of cheap products, and shrinking market.

John,

Although you are cleary experienced and knowledgeable, I think your 'all commercial codes are rubbish and of no use' attitude is short-sighted at best and ignorant at worst.

There are many very-well validated and strong physics in cfx4. For example, many multi-phase models implemented in cfx4 mainly be Simon Lo are very good and probably better implemented than many user-written code, purely because they were written by a true expert in that field.

I'm not saying that your codes are equally strong, becasue I haven't seen them, but to suggest that everybody writes their own models is ludicrous. There is not the time to understand, implement and then test/validate them each time you face a different challenge.

Also, to say that work using commerical codes is always wrong or has no worth is purely farcical. I appreciate this is a bug-bear of yours, but lets keep it within reality.

Sounds to me like Jon might be working for CFX?

He does, but the above comments still stand.

(1). How do you make improvements or creating new products, if you insist that your product is the best? (2). And it is known that the highly secret and expensive stealth fighter can now be detected by relatively cheap mobil phone netwrok. (3). If you count the total number of questions posted in each forum, how many satisfied users can you find who are saying that they are happy with the commercial codes. That's the fact. (4). I am not saying that a solution from a commercial code is useless. I am saying that there is an over-supply of not-so-good solutions because these not-so-good cfd solutions can be easily created now. A good model, a good algorithm in a general code really can not ensure that user's solution will be accurate. (this is because the original author of the model and code is no longer with the users. You can't repeat a professor's process an win the Nobel prize again.)

OK,

You seem to like analogies and the use of stories and what ifs? in you correspondance, so I'll follow your lead.

If you are knowledgeable about the mechanics of cars and know how they work, then I suppose you could build one.

It would be easiest to build one that was a car, but only had one main use, e.g. only had one gear, as you only needed that gear for driving from your house upm the hill and back again. This car could by very efficient to do this and if you had the skills to design and build this, this would be good for you.

However, if you wish for this car to drive down the motorway (highway), for example, then if will not be suitable. You will not have time on your own to be able to design and build a car that will do all you wish as a car you could buy.

Also, you skills could be in the engine department, but you are lacking the welding, painting and road-testing department.

Further, there are a lot of people who car drive, bt who cannot build/design cars and therefore they will buy one model or another.

Dont get me wrong, I think people doing research into CFD solvers/models and meshing is extremely important. I just believe that saying that commercial CFd codes are all wrong, all bad and not worth a jot, is as bad as saying that there they are the be-all and end-all.

Also, I notice on another question, you question why CFD software does not come with a warantee about the results.

Contining the analogy, car manufacturers cannot guarantee that cars will not crash. They will just make them safe to drive, if driven responsibly. They cannot guarantee what the driver or other drivers will do.

(1). I like your "car" example very much. (2). I think, commercial CFD codes vendors should use car manufactures as good examples, and try to make commercial CFD codes the same way a car is required to met... (3). This is especially important, because I heard that car manufactures are using commercial CFD codes in their design of the car. (4). I hope that pretty soon, everyone will treat the commercial CFD codes like a car. (5). Your example is an excellant example. I agree with you 100% in this example or analogy. Good work. (6). For the racing car design, the irony is ,based on a recent journal article, I don't think the racing car designers trust the results of commercial CFD codes at all. (7). I think, it is a good idea to have a "Department of CFD" in the government.