Any engineers here? CFD??

Discussion in '2-Stroke Machines' started by 90kacoupe, Feb 8, 2017.

  1. 90kacoupe

    90kacoupe Novice seeking Help

    Has anybody used Ansys to do any modeling of air cooled 2 stroke. I am wanting to test the idea that the fins on my head and cylinder are too long. I am thinking about trying to scan the head (or do a basic model.) and vary fin height and air velocity to determine the reason for such long fins. This is more or less me just wanting to play around with the idea of trimming fins. It would be fun to see what ANSYS has to say about it.
     
  2. ck.mecha

    ck.mecha Well-Known Member

    We run ANSYS AIM at work mostly fluid/force interaction; I like it compared to other CFD I've used. It does takes a lot of computing power and the license isn't cheap if you're thinking about buying.

    Sent from my VS995 using Tapatalk
     
    peter leduc likes this.
  3. 90kacoupe

    90kacoupe Novice seeking Help

    I have access while in school and I also can get a 6 month student edition. I will probably start my 6 months right before I graduate in May.

    Sent from my SM-G930P using Tapatalk
     
  4. keypup

    keypup Active Member

    Understanding and properly applying the boundary conditions and environmental aspects of the operating conditions are key to properly analyzing the situation (whether through CFD or hand calc), so I would definitely spend more time understanding that before plugging numbers into software. As the old saying goes with any analysis, shit in = shit out, so it pays dividends to properly understand the situation before doing anything else.

    What engine are you looking at? I would think that the engineers who designed the cooling system spent some time deriving the proper fin geometry for what they considered to be the operating conditions - though that may be different than how you intend to use it.
     
  5. Phl218

    Phl218 8ƖᄅlɥԀ

    my professor always said: "calculate a shoe box first to see if you're on the right track" and: "colored pictures show nothing, except that you have a color printer" :D
     
  6. ck.mecha

    ck.mecha Well-Known Member

    Ill assume you're using AIM not one of the more focussed programs. They actually have a very good tutorial on heat transfer, study it closely. Look up your local ANSYS rep and become good friends with them, they are really open to helping. You will need to parameterize whatever measurement you're thinking of changing to keep from having to reimport and set up transfer surfaces for every dimension change. There is no good tutorial for this, and its imperative unless you want to spend endless hours repeating the same steps.

    Sent from my VS995 using Tapatalk
     
  7. 90kacoupe

    90kacoupe Novice seeking Help

    Yes I will spend some time setting up boundary conditions and running simulations on simple models. It is really going to be a two part problem. Since the air flow over the cylinder head will have to be understood before it can be used as a boundary condition for the heat transfer problem. I also need to determine if the model can be simplified by making some assumptions. I do have a professor that is our go to CFD guy. He said he would be glad to help in anyway.

    This is more a project for fun. Just to see what I can do and learn the software a little more.


    Sent from my SM-G930P using Tapatalk
     
  8. 90kacoupe

    90kacoupe Novice seeking Help

    Haha. Yeah I had professor that said the same regarding FEA structural analysis. If you printed the colored results. It better at least have element boundaries, or else it literally means nothing.

    Sent from my SM-G930P using Tapatalk
     
    Phl218 likes this.

Share This Page