Learning CFD

Discussion in 'Hydrodynamics and Aerodynamics' started by Peter Marcellus Epe, Jul 5, 2020.

  1. Peter Marcellus Epe
    Joined: May 2020
    Posts: 2
    Likes: 0, Points: 1
    Location: Philippines

    Peter Marcellus Epe New Member

    Hi guys, I am new in this site. I would like to ask question about the ideas of simulating a stepped hull.
    like what to do? or what software do i need to use?
    i have already a 6 months experiences in using CFD but i still have doubts in my skills.
    please could anyone help me or educate me.
    Thanks :)
     
  2. CocoonCruisers
    Joined: Dec 2015
    Posts: 70
    Likes: 19, Points: 8, Legacy Rep: 10
    Location: Marseille & BuenosAires

    CocoonCruisers Junior Member

    Welcome aboard Peter!

    That is a broad question ... it would be helpful to narrow it down a bit: size, speed, money or academic backing available for licenses, workstation or HPC cluster, finality of the project (ie do you need absolute or relative results, and do you need to be sure of them or is it more for learning or promotion), still water or wave action.

    IMHO there is no difference between a stepped hull sim and any other planing boat sim, except having two transoms. Just make sure you have as much refinement on the air channels as on the planing sole.

    Since you are already into CFD you have probably figured out that from the 30% of calculations which don't crash, 80% produce nice pictures but unphysical results.

    Step one is to make sure the software can function properly.
    - A correction agains numerical ventilation ("streaking") is needed. See what your software offers and how to use it.
    - Turbulence sim is needed, typically k-OmegaSST
    - Much of the quality is related to the meshing
    - figure out what conditions (yPlus, Courant Numbre) are needed for the turbulence model to function properly, mesh accordingly, set your time step accordingly, and check that the values come out correct in the calculation
    - mesh check tools in the software may help but most often you won't be able to eliminate all the warnings.
    - Other problems will only show once you start to solve, or get a feel for typical problem areas like wing trailing edges, junctions etc. Track freak cells giving high Courant numbers or other extremes and try to eliminate them. It's probably gonna be an iterative process for every case.

    Step two, verify
    - If you use local time stepping, check with a time-accurate solver once the flow is stabilised
    - do mesh independence studies

    Step three, validate
    - against regression formula , Savitsky etc for your own hull
    - against measurements of comparable boats if available
    - against model testing data if you can't find anything better, but beware the specifics of tank testing
    - for stepped hulls specifically, try to look at flying boats and seaplane floats. That's where the steps come from.

    Good luck !
     
    Last edited: Jul 6, 2020
  3. fredrosse
    Joined: Jan 2005
    Posts: 361
    Likes: 33, Points: 38, Legacy Rep: 56
    Location: Philadelphia PA

    fredrosse USACE Steam

    "Since you are already into CFD you have probably figured out that from the 30% of calculations which don't crash, 80% produce nice pictures but unphysical results."

    Amen to that! In the thermal/fluid representations I have seen far more garbage than usable results, with inexperienced modelers making nonsense conclusions because they often fail to recognize the obvious flaws in the analysis.
     
  4. jehardiman
    Joined: Aug 2004
    Posts: 2,763
    Likes: 357, Points: 83, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    Aeration is one of the big holes (pardon the pun) in CFD. You are well into chaos theory with that. Try sending a message to
    [​IMG]
    Leo Lazauskas
    This is up his billy-wack.
    Hummm... is Leo still on?
     
  5. DCockey
    Joined: Oct 2009
    Posts: 4,707
    Likes: 315, Points: 83, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Leo last posted here three years ago. His website www.cyberiad.net has appeared to be defunct for several years. Internet searches do not show any recent activity.
     
  6. Ad Hoc
    Joined: Oct 2008
    Posts: 6,610
    Likes: 617, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    I know his busy doing research and rarely uses the internet these days.
     
  7. Peter Marcellus Epe
    Joined: May 2020
    Posts: 2
    Likes: 0, Points: 1
    Location: Philippines

    Peter Marcellus Epe New Member

    guys does anyone of you can help me or send me a guide on to proper way use of IACS (rules and regulations of ships) like how to use or when to use... sometimes i still have doubt in my calculations especially in those ship design requirements like thickness and section modulus. ive tried onetime to obtain the design load through CFD. am i doing it right?
     
  8. DCockey
    Joined: Oct 2009
    Posts: 4,707
    Likes: 315, Points: 83, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Which rules from a member society of IACS are you trying to use? Perhaps you should concentrate on learning the standard methods of using the selected rules.

    What design loads are you trying to obtain through CFD? Hydrostatic loads do not require CFD. To determine slamming loads usind CFD requires specialized knowledge of several areas including what events/wave spectra to use as inputs, the details of how to use the particular code for slamming type calculations, and how to post-process the CFD results to obtain loads which are compatible with the particular set of rules/method which will be used for the structural calculations.
     
    jehardiman likes this.
  9. Ad Hoc
    Joined: Oct 2008
    Posts: 6,610
    Likes: 617, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    In addition to DC's comment.

    Choose a Class society, LR, DNV-GL .. any one of them.
    Then read and apply the rules and it provides you with the answers you seek.

    No need for CFD at all... that's just a red herring
     
    jehardiman likes this.

  10. jehardiman
    Joined: Aug 2004
    Posts: 2,763
    Likes: 357, Points: 83, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    I concur with David and Ad Hoc, Just use the societies codes...you are going to need to meet them anyway to get funding or insurance. You can use the search function in the upper right to search the forums for " CFD slamming". I know it will turn up several threads which include me and others discussing the relevance of CFD to single event loads.
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.