Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Aerospace Aeronautical engineering requires a formulaic process of thought!

  1. May 10, 2012 #1
    Hi,

    I'm a PhD student working on future airframe and propulsion systems and I'm starting to feel that everything in the aeronautical sector seems to have a pre-planned methodology. An example is to design an aircraft .. you can pick up various books and use very old formulas to design it and the science has been reused endlessly. Also the changes in design are ultimately tweaks of a current configuration. Even the blended wing body has many similarities to T&W's and the industry I work with just treat it the same virtually in many ways!

    Gas turbines too seem to have a similar design process i.e. non dimensional maps, whilst it's components are the same; comp. comb. turb. and nozzle, where PDE's are an exception.

    I personally spend most my time tweaking numbers on programs to re optimise to create 'new designs', however I feel, as an engineer, that I should be more innovative in my methods. I understand it's a conservative industry and I'm fairly pessimistic, but does anyone else understand where I'm coming from? Is this the right industry for me? My feeling is that it could be fundamentally the problem with specialising in an engineering field...

    Feel free to disagree with me and compare what I've said with other industries...

    Thanks,
     
  2. jcsd
  3. May 10, 2012 #2

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    "Industry" is always the last to change. The quest for profits leads to an enormous institutional inertial whereby it takes a lot of proof and potential for money to be made before industry will change away from a tried and true method. The exception would be within most advanced development groups within a given company. This happens in basically every industry other than startup industries.

    Academia, however, usually works more on the cutting edge, even if the investigation of the cutting edge is done with old techniques, you have to use something to come up with the new stuff before you can use it instead. Still, in my own experience in an aerospace engineering PhD program, we do a lot with really cutting edge ideas. We are still constrained by many tried and true processes such as machining, but that doesn't stop us from pushing the boundaries. In fact, this is the reason I went to graduate school. Working in industry really bored me since it seemed a lot like taking old ideas and applying them to new problems.

    Then again, I work on fundamental aerodynamics, not airframe design, so my particular area of research lends itself more easily to thinking outside the box since it is mostly fundamental research. Perhaps the design of airframes is just too subject to government regulation and other similar factors to easy go through paradigm shifts in the way the design process is done.
     
  4. May 11, 2012 #3
    Thanks for the speedy reply. I certainly agree that aerodynamics links much more closely to science than concept design, do you use CFD within this sector mostly or is there much mathematical/analytical modelling (and if so what sort of maths)? Also would an aerodynamicist be interested in the design of a product or is it usual to be concerned with the flow physics alone? My ideal job seems to occupy many roles, which I feel may be unrealistic in such a complex engineering discipline. Any info is much appreciated.
     
  5. May 11, 2012 #4
    Remember that the only reason you are designing airframes and airfoils is so that they can go on a craft which will be sold to someone for a profit. Those who pay the designers' salaries want them to be innovative enough that their designs perform better than their competitors, and cost less to make; they generally don't want the risk involved in brand new designs which likely wont provide suitable gains over existing ones to justify the risk and the cost involved in producing them.

    Then again, there are companies which deal almost exclusively in airframe/foil innovation and thrive on that very risk.

    There's a nice saying that I've heard. "Project managers hate innovation."
    Innovation means unknowns. Unknowns in schedule, in cost, in performance, etc. So while, yes, the field of engineering is built upon finding good solutions to tough problems, generally the people who are running these projects (and have to worry about the financial side of things) don't want to be innovative, they want to be cheap and effective.
    (again, this is just the general state-of-the-industry. There are plenty of companies which have innovation as one of their core principles; sadly, many of these companies will fail because of that goal)
     
  6. May 11, 2012 #5
    I've been in aerospace since 1979. True innovations are rare, but they do happen. Then they will be optimized in small increments for decades. When they do happen they are closely guarded proprietory secrets and you won't see them in the universities until they become old news. Much is happening right now that may turn out to be major game changers, but the developmental work might take another ten years before you see it at school.
     
  7. May 11, 2012 #6

    AlephZero

    User Avatar
    Science Advisor
    Homework Helper

    That's a bit like saying car designs are the same as they were was 50 years ago, because cars still have the same basic layout and the same major components.

    Look in more detail and you will see what the real rate of technical innovation is. But most of that is commercially sensitive, and doesn't find its way into textbooks till it is well past its sell-by date.
     
  8. May 11, 2012 #7

    AlephZero

    User Avatar
    Science Advisor
    Homework Helper

    The trick is to run "innovation" AS a project, the same way as any other project. (It sounds nicer if you call it something like "new technology acquisition"). What project managers really dislike is innovation done in panic mode to fix unexpected problems!
     
  9. May 11, 2012 #8

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    Plenty of people use CFD, but I am an experimentalist (mostly). Of course I use some CFD to some degree in designing experiments, but it is far from my primary job function. There is plenty of math and modeling in my field. Precisely what sort of math depends somewhat on what phenomenon one studies, but there are tons of partial differential equations/perturbation methods, Fourier analysis and other signals processing techniques, statistical correlation, etc. Then again, someone working on a different branch of fluid mechanics will probably use a different set of tools.
     
  10. May 11, 2012 #9
    AlephZero,

    What I'm trying to say is ultimately, for say a fan, the blade has been optimised so much that any other novel method of energising a fluid becomes less efficient during its infancy and therefore less attractive to financiers. Changes to fans exist (mainly due to CFD advances) but it is still just a fan blade and the design process (non-dimensionalisation, stage loading) remains relatively unchanged.

    Pkruse,

    The university I'm at does research for designs which are for a number of decades in advance, yet I still feel there is an underlying limit to the number of ways in which you can fly and propel aircraft and they are slowly being exhausted...

    I agree that the devil is in the detail and maybe that is where I should be heading, rather than the general performance side.

    Thanks for all your replies.
     
  11. May 11, 2012 #10
    You still have to obey physical laws using existing materials and production methods...
     
  12. May 11, 2012 #11
    Pkruse,

    I noticed on another thread that your familiar with the mechnical engineering side of the gas turbine. How varied are the designs that are conceptualised i.e. always a turbine disc roughyl speaking? Do you use kinematics? Do hand calcs and quick designs occur often as a designer or analyser? Are you inolved with demonstrators/prototypes/mockups? Sorry for the exccessive questioning. Thanks
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Aeronautical engineering requires a formulaic process of thought!
Loading...