Aircraft Design and Design Testing

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by Aegis616, Sep 4, 2019.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Sep 7, 2019 #21

    Aegis616

    Aegis616

    Aegis616

    Member

    Joined:
    Sep 3, 2019
    Messages:
    5
    Likes Received:
    0
    I should clarify that I am looking for software, whether standalone or a suite, that can do the following functions: structural modeling with load testing, aerodynamics simulation and component selection for avionics and powerplant solutions. It doesn't have to be a single piece of software but I need it to be compatible with what every software people like to use for the concept art to see if these designs are feasible.
     
  2. Sep 7, 2019 #22

    Chris In Marshfield

    Chris In Marshfield

    Chris In Marshfield

    Well-Known Member HBA Supporter

    Joined:
    Jan 29, 2008
    Messages:
    1,203
    Likes Received:
    996
    Location:
    Germantown, WI USA
    DAR Corporation has a configuration application that looks pretty comprehensive. When I talked to them about i it OSH one year, they said it was meant to be used in a professional environment (read: $$$$). They haven’t created a version suitable for consumer/amateur designer use.

    https://www.darcorp.com/advanced-aircraft-analysis-software/

    They do, however, carry Roskam’s design series. Not sure the software is based on his work, but it does have a lot of interesting configuration stuff in it.

    https://shop.darcorp.com/index.php?route=product/product&product_id=59

    I bought my set from them directly at OSH a number of years back.
     
  3. Sep 7, 2019 #23

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,680
    Likes Received:
    1,760
    Location:
    North Carolina
    Typo, I'll fix. You'll notice that i said that I used spreadsheets from before Excel...
     
  4. Sep 7, 2019 #24

    Lendo

    Lendo

    Lendo

    Well-Known Member

    Joined:
    Feb 7, 2013
    Messages:
    309
    Likes Received:
    47
    Location:
    Brisbane
    I love Excel, there are a world of quality formula for preliminary aircraft design, some formula can be quite complex looking, but broken down to their elements their not in any way difficult. I have been asked to help someone and decided to set the basics out in Excel. I have over the years set-up a lot of Excel files, for my own interest and design of my own Tandem aircraft. It may never be built but the math is all there. Some of the Formula is by Roncz and several other Text books and knowledgeable people on HBA. I don't mind spending the money on a Text book provided I just find one 'Golden Nugget' of information.

    I think or what's important for me is understanding what the formula is doing and why. Using the Spread Sheet (SS) to quantify known aircraft, is sometimes very interesting.

    A SS is only as good as the data entered 2D Lift Coefficient of the Airfoil selected and conversion to 3D, formula for Incidence of the wing, I have never cracked the Neutral Point formula 'Stick Free/ Stick Fixed'. Additional areas needed for lifting surfaces because of Sweep, Mach and AR. Sounds hard, look a little complex - but not really.

    Some reference material is required and guesstimates are only good until you find the real numbers - and always be conservative.

    So estimates on weight for Wing, Ht and VT are easy, if their radical apply the Sweep Mach and AR formula - simple!
    George
     
  5. Sep 8, 2019 #25

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    6,680
    Likes Received:
    1,760
    Location:
    North Carolina
    I use pencil, paper and calculator to check basic feasibility. The design devil is in the details, which is just a lot of plane tedious work. There is no magic software to do it for you, yet. If I'm expecting to tweak a complex part like a truss, I'll set up a spreadsheet. When (not if), I have to change it due to ripples from another change, it's then a simple matter of keying in a few revised values and it spits out the new numbers in a fraction of a second, that can be days or weeks faster than I could do it by hand. Sometimes, you may see no change in the design as the tube sizes that worked best for Mk1 also work out best for Mk2.
    I'm not sure that there is any software that will do all the functions for you outside specialist stuff. You absolutely must know how to use that and the theory behind it, if you could afford it. You see a lot of garbage results from people blindly using software that they don't understand. That is a terrible idea for aircraft design. There's an attempt at a high end canard design on YouTube that features this approach in spades. It's like watching a slow motion train wreck.
     
    gtae07 likes this.
  6. Sep 8, 2019 #26

    christos

    christos

    christos

    Active Member

    Joined:
    Nov 9, 2018
    Messages:
    39
    Likes Received:
    12
    Location:
    Greece
    You will not find software with so many abilities.
    Also, it will be not "automatic".
    ANSYS is one of the most completed analysis software.
    There are also a lot of cost-effective and open-source software out there like OpenFOAM(C++ knowledge is necessary), Lisa fea (low cost and user friendly fea software), Code_ASTER etc
    But, depending on your goals, if your pc could solve your problem.

    Personally (for fluid dynamics and non-linear problems), i am buying computing core hours.
    You can design a plane without spending money on computing hours if you go with a low order solver and avoid impact analysis. Off course some stages of your analysis (behaviour during the stall, flutter analysis, structural analysis etc) will be not accurate enough because of your solver choice (vortex lattice method in case of a fluid problem and element grid in case of a non-linear problem). Results will not improve enough if you go with a CFD software like ANSYS CFX or OpenFOAM etc, because of low elements, low boundary layers and solver tolerance. Your pc will run out of memory if you try to improve the grid, add boundary layers, or try a better solver tolerance.
     
  7. Sep 8, 2019 #27

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    11,523
    Likes Received:
    3,225
    Location:
    Memphis, TN
    If you follow people trying to design an airplane, really design not copy, it takes them five to ten years of part time work using anything that’s available at working mans cost. What your asking for is only at Boeing or Lockheed or SpaceX. That DAR program looks interesting probably the closest to what you are asking without spending $10M. That is why all the spread sheets, what you want does not exist. You actually have to design an airplane. Even at the big companies.
     
    Rockiedog2 and mcrae0104 like this.
  8. Sep 8, 2019 #28

    Dan Thomas

    Dan Thomas

    Dan Thomas

    Well-Known Member

    Joined:
    Sep 18, 2008
    Messages:
    4,824
    Likes Received:
    2,041
    "New" airplane designs are almost always an evolution of an existing design. That's why so many homebuilts resemble factory-made airplanes. One takes what is known to work well and make it your own by making relatively minor changes to it without sacrificing strength or useful load. Many of the older homebuilts, like my Jodel, were production airplanes in Europe before they were homebuilts here. Chris Heintz, of Zenair fame, was an engineer with Avions Robin in France before he came to Canada and built his original CH100 Zenith, which strongly resembled the airplanes Robin had been building.

    There aren't many gifted guys like Steve Wittman or Burt Rutan, guys that seemed to have the magic touch that made their airplanes go so well. Most of us are stuck with either plagiarizing or crunching an awful lot of numbers. Plagiary is easier:)
     
    Charles_says likes this.
  9. Sep 11, 2019 #29

    wktaylor

    wktaylor

    wktaylor

    Well-Known Member

    Joined:
    Sep 5, 2003
    Messages:
    109
    Likes Received:
    54
    Location:
    Midwest USA
    Always take the warnings expressed by Captain A. G. Lamplugh [1930s], Igor Sikorsky, Max Stanley [test pilot] and others, about aviation, VERY seriously!
     
  10. Sep 11, 2019 #30

    Jay Kempf

    Jay Kempf

    Jay Kempf

    Well-Known Member

    Joined:
    Apr 13, 2009
    Messages:
    3,635
    Likes Received:
    913
    Location:
    Warren, VT USA
    Never said MS invented the spreadsheet. But they took the concept and honed to intuitive, useful, powerful and plays well with all the other ways to present and interact with other software that engineers use all the time. Same with Solidworks. Just plain useful and invisible technology that doesn't get in your way when you are trying to be creative. If you have an analysis tool, a 3D modeler, and a spreadsheet, there isn't much in engineering you can't do short of actually machining the parts.

    I remember trying to run AutoCad, Lotus 123 and Aldus Pagemaker with runtime Windows 1.0 all under DesQview on a 386 with a math coprocessor WOOOHOOOO! and a 20meg hard drive doubled with compression software. Those were the days :) I think I had 2 megs of RAM and I thought I was somebody. What we have now is just ridiculous. Churning through first fiber to fail FEA in 3D on a home computer. Sheesh! We only dreamed of that stuff back in the day.

    Back to the OP: A good spreadsheet an open source 3D modeler and open foam XFLR5, XFOIL etc.. for aero. LISA for some FEA if you need to go there. Have you looked at Vehicle Sketch Pad VSP?
     
  11. Sep 11, 2019 #31

    Aerowerx

    Aerowerx

    Aerowerx

    Well-Known Member

    Joined:
    Dec 1, 2011
    Messages:
    4,894
    Likes Received:
    1,314
    Location:
    Marion, Ohio
    Someone else on here did.

    I can beat that! 1988. Working on my Master's project. I had a PDP-11/40 all to myself! Ran RSX-11 on dual 10 Mb hard drives (remember the "cake platter" drives?), Fortran 77, with 64k words (IIRC) memory. Had to split up the program. Ran data collection one day, and number crunching the next. Surprised me that my idea actually worked. Still have a hard copy of the program around here somewhere.
     
  12. Sep 11, 2019 #32

    christos

    christos

    christos

    Active Member

    Joined:
    Nov 9, 2018
    Messages:
    39
    Likes Received:
    12
    Location:
    Greece

    I told "fluid dynamics and non-linear problems".
    You will not need to use a workstation or to rent one to solve a linear fea problem. About computational fluid dynamics, it is common to use 2 million elements for a 2d airfoil analysis. It is hard to solve a 3d turbulence unsteady-state model. Also, non-linear's analysis requirements are far away from a linear model.
    It depends on your accuracy goals if you go with a real work station or a home pc. But, i am a young Master integrated Civil Engineer and i want the most accurate.
    Also, depends on "home pc". It will be fine if "home pc" is as strong as a work-station.
     
  13. Sep 12, 2019 #33

    rdj

    rdj

    rdj

    Well-Known Member

    Joined:
    Sep 10, 2009
    Messages:
    288
    Likes Received:
    185
    Location:
    Northern California
    Kids these days. 1978, summer job between high school / college was to sysadmin a PDP-11/70 and PDP-11/60. Four 5M and 10M "cake platters", RSX-11M on the 60, RSTS/E on the 70. Loaded the boot code in octal using the front panel switches, which loaded the paper tape reader/writer second stage OS to boot the timesharing OS off disk. Took 10 minutes to boot while writing the startup log to the DECwriter terminal. Dangerous; as such early exposure eventually led to hanging out in homebuilt aviation forums with other nerds later in life.

    BTW, if anyone needs some serious CFD compute horsepower these days, it's easiest to just buy some time on an Amazon AWS instance. Use the cloud, Luke. Unfortunately, no amount of compute power will speed up the process of de-burring 10,000 #30 holes, sanding fiberglass for months on end, or all the other skilled tasks that will determine the successful outcome of your design far more than your CFD coding prowess.
     
    Bill-Higdon and Mad MAC like this.
  14. Sep 12, 2019 #34

    wktaylor

    wktaylor

    wktaylor

    Well-Known Member

    Joined:
    Sep 5, 2003
    Messages:
    109
    Likes Received:
    54
    Location:
    Midwest USA
  15. Sep 18, 2019 #35

    PTAirco

    PTAirco

    PTAirco

    Well-Known Member

    Joined:
    Sep 21, 2003
    Messages:
    3,475
    Likes Received:
    994
    Location:
    Corona CA
    I have always been able to somehow find a free copy of Nastran somewhere online. I mostly work with frame type structures rather than monocoque or stressed skin/composite designs. Nastran works very well for my needs. I freely admit I only use about 20% of its capabilities but it sure takes the tedium out of finding loads in a framework. Being partial to biplane designs, this is a huge help.
     
  16. Sep 18, 2019 #36

    christos

    christos

    christos

    Active Member

    Joined:
    Nov 9, 2018
    Messages:
    39
    Likes Received:
    12
    Location:
    Greece
    Lisa is also excellent fea software. It is not expensive and has a lot of capabilities.
    In case you need a non-online fea software.
     
  17. Sep 29, 2019 #37

    Aegis616

    Aegis616

    Aegis616

    Member

    Joined:
    Sep 3, 2019
    Messages:
    5
    Likes Received:
    0
    I have the concept but I want to make sure that it works. What are some of these evaluation programs?
     
  18. Sep 29, 2019 #38

    Aegis616

    Aegis616

    Aegis616

    Member

    Joined:
    Sep 3, 2019
    Messages:
    5
    Likes Received:
    0
    Is there a modern interpretation of this idea? I can budget a little bit if this will be a parts thing.
     
  19. Sep 29, 2019 #39

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,304
    Likes Received:
    2,247
    Location:
    Rocky Mountains
    X-Plane and XFLR5/6 are good to see if the idea is even close to being practical.

    Start with Excel sheets for the basics. This means that you need to know the math formulas and the why behind them.

    The modern trend is FEA. There are many of those programs from free open source to 'there is no way I can afford that'. They have a steep learning curve to give good results. They are also pretty much not worth the time unless you are really trying to optimize the design. There are plenty of good planes that were built using nothing but the basic tools and a healthy portion of experience.

    No matter which path you take - it will take time and effort.
     
    pictsidhe likes this.
  20. Sep 29, 2019 #40

    Aerowerx

    Aerowerx

    Aerowerx

    Well-Known Member

    Joined:
    Dec 1, 2011
    Messages:
    4,894
    Likes Received:
    1,314
    Location:
    Marion, Ohio
    I am not sure what you are referring to. Can you elaborate?
     

Share This Page

arrow_white