+ Reply to Thread
Page 1 of 4 1 2 3 ... LastLast
Results 1 to 10 of 36

Thread: Suspension Geometry Analysis

  1. #1
    Junior Member
    Join Date
    Jan 2014
    Location
    Coimbatore,India
    Posts
    9

    Suspension Geometry Analysis

    Hello Everyone

    I am currently learning vehicle dynamics and in this regard I started doing calculations for geometry analysis of an open wheeler with Push rod mechanism and a mono shock arrangement with belleville stacks.

    The method I used to determine the movable points in the mechanism (during wheel travel ) was three point analysis (Ref : The Multibody systems approach to Vehicle dynamics,Damian Harty & Mike Blundell)

    You can find the attached sketches that I made prior to the calculation.

    Pitch.jpgRoll.jpg

    I would like to know if this calculation sequence is correct ?

    Representative input : Damper length shortened by 50 mm i.e, JI is shortened to JI'
    Motion : Pitch

    As we need three known points to commence the calculation, here we shall take points J,Q and P to determine the new position I'

    Once the point I' is determined then the remaining sequence is as follows

    Using I',Q,P new position of point M i.e, M' can be determined

    then

    A,C,M' -----> N'
    A,C,N' -----> B'
    F,D,B' -----> E'
    E',B',H -----> G'
    E',B',G' -----> K'

    I welcome all comments and suggestions

    Thank you
    Koushik
    Attached Images
    Last edited by koushik123; 03-09-2015 at 10:13 AM.

  2. #2
    And you are Koushik?

    If you would want to feel welcome in a group you can't jump in without telling a bit about yourself, which school and where you are from if you (nhave already some experience in FS / FSAE

    I think you have more chance to get answers or more chance to get more complete answers if you would introduce your self.

    Simple common sense of being courteous....
    Claude Rouelle
    OptimumG president
    Vehicle Dynamics & Race Car Engineering
    Training / Consulting / Simulation Software
    FS & FSAE design judge USA / Canada / UK / Germany / Spain / Italy / China / Brazil / Australia
    [url]www.optimumg.com[/u

  3. #3
    Junior Member
    Join Date
    Jan 2014
    Location
    Coimbatore,India
    Posts
    9
    Hello Everyone

    I am Koushik Sivasubramaniam, a recent automotive engineering graduate from India with keen interest in motorsport engineering.I graduated from Anna University,India in 2014.

    Experience :
    2015 - Freelance Vehicle Performance Engineer, Formula Renault 2.0 Northern European Cup
    2014 - Intern,Spitfire Performance, India
    2013 - Virtual Supra SAE INDIA ( Indian FS Pre qualifier round )
    2013 - Mahindra Baja SAEINDIA
    2012 - National Go Kart Championship,India

    And thank you Mr.Claude for pointing out my lack of common sense.

    Regards
    Koushik

  4. #4
    Koushik,

    Welcome to the forum.

    What is the point of your calculations? What are you looking for? What is the ultimate goal?
    Claude Rouelle
    OptimumG president
    Vehicle Dynamics & Race Car Engineering
    Training / Consulting / Simulation Software
    FS & FSAE design judge USA / Canada / UK / Germany / Spain / Italy / China / Brazil / Australia
    [url]www.optimumg.com[/u

  5. #5
    Junior Member
    Join Date
    Jan 2014
    Location
    Coimbatore,India
    Posts
    9
    Thank you Mr.Claude.

    Ultimately I would like to make a plot of Camber Vs Wheel Travel. Of course this can be done using any available suspension softwares. But I am trying to create my own kinematic solver in MATLAB so that it will help me in understanding the computations involved.

    Koushik

  6. #6
    Senior Member
    Join Date
    Mar 2008
    Location
    Brighton, MI
    Posts
    686

    Alternative Suggestion

    I suggest you jump to SimMechanics if you will be using Matlab. You can still see the solution process, no suspension mechanism is just a bunch of rigid links, and your employer will be hugely disappointed with an overly simplified analysis that does not accurately (enough) provide the right answers (and direction).

    You're not in school now, the real racing world demands sophisticated models that include elastic, nonlinear and deformable elements (plus the kitchen sink).

    That's my opinion and I'm unanimous with it!

  7. #7
    If you are a FS student I can only applaud you to create your own kinematics software but it could take so much time that it could delay your car design and manufacturing

    If you are a professional why do you want to reinvent the wheel when you can get it cheaper and quicker somewhere else? Buy a kinematics software then use you energy using the software to get the most or to improve of a given car suspension design.
    Claude Rouelle
    OptimumG president
    Vehicle Dynamics & Race Car Engineering
    Training / Consulting / Simulation Software
    FS & FSAE design judge USA / Canada / UK / Germany / Spain / Italy / China / Brazil / Australia
    [url]www.optimumg.com[/u

  8. #8
    Junior Member
    Join Date
    Jan 2014
    Location
    Coimbatore,India
    Posts
    9
    Hello Mr.Bill

    Thank you for your response.I will have a look at SimMechanics too and I understand your point.

    Regards
    Koushik

  9. #9
    Junior Member
    Join Date
    Jan 2014
    Location
    Coimbatore,India
    Posts
    9
    Mr.Claude

    I am just curious enough to understand how a kinematics software solves the above problem.

    I do not wish to re invent a wheel rather I wish to understand the wheel first before using it. Obviously while at work where quick and smart work is required I will not be using a paper and pen and a bunch of simultaneous equations doing algebra to get my results, I will ofcourse use a kinematics solver to get the work done.

    Regards
    Koushik

  10. #10
    Senior Member
    Join Date
    Sep 2002
    Location
    Perth, Western Australia
    Posts
    717
    Koushik,

    I think you are on the right track, and I mildly disagree with both Bill and Claude.

    I would advocate the use of both SimMechanics and OptimumK (or equivalent kinematics software). However complicated models have simple beginnings, and this world needs people who can build these models from scratch. Keep doing what you are doing.

    If you haven't already try and solve this as a 2d problem. While 2d is not a good representation of what is happening it will help you sort out the general calculation flow as well as helping you get your head around how you will setup inputs and outputs.

    Another point you may want to consider is that you have some great advantages over SimMechanics and OptimumK. First is that your solution method does not have to be computationally efficient. Second is that computation is cheap. Finally you do not have to account for a massive range of error checking. This allows you to keep your program lean and simple.

    When I was doing the 3d kinematics as a student my program moved a suspension system through its expected motion by moving the lower a-arm through its expected angle of motion. Everything solved as a four bar linkage. All solutions to the mechanism were stored in a matrix. This was also swept with steering for the front as well. Done fine enough any point inbetween your calculations can be interpolated with minimal error. Once you have this data you can select the state using whatever measure you wish. Simple calculations from here can convert these motions to body based movements. A few functions for drawing (i.e. cylinders) and you had some cool graphical output. Please note this method works well for pure kinematics but cannot cover what Bill is leading to.

    Memory storage is not ideal. Calculation time is long for the difficulty of the problem. However in 2004 on a fairly normal pc solutions were calculated almost instantaneously.

    One advantage over OptimumK is that each of the solved points is a geometrical solution. OptimumK uses a Newton-Raphson solver which is much better for more general purpose solving. But it is good to keep in mind you don't have to solve a general purpose mechanism, all you need are calcs on one type of kinematic linkage. Harty and Blundell's book is dealing with a similar problem.

    I also noticed with the geometrical solution method that when you make a programming error you see it clearly in the output, as in it is almost always a big error. With the numerical solvers you can persist with something that looks right for a while without noticing the errors.

    All that aside I use a different approach to solving pure kinematics problems these days for a simpler design process. 3d sketches in solidworks (or other CAD) can be very powerful. The temptation is to define sketches on link lengths and positions. However For each position you draw you can leave an unknown undefined. For example instead of defining your swing arm length draw the suspension in two positions, one static, one compressed. Make sure there are restraints to make the links of each sketch equal to each other. You have to define a lot of things but make sure one of your measurements is on the camber angle. The SAL will then be defined by everything else. 3 positions allows you 2 unknowns, and so on. The basic idea is to start with the values you want and then force the mechanism to comply. There will be plenty of times you cant get a solution, or the potential solution makes the mechanism invalid. What is great about this is that it is kinematics at the design level. Programs like OptimumK are great for analysing what you have, but they are slow when it comes to design. Design is inherently about iteration and exploration within a given solution space.

    The ECU guys used this method to solve for the funky steering mechanism they used, which I will note cannot be modelled in any of the purpose made suspension kinematics software. Although SimMechanics and ADAMS are no problem, just a lot slower. Likewise the rear beam arrangement could also not be modelled. OptimumK's engine could easily handle it. All that would need to happen is the creation of appropriate input sheets, and a couple of changes to the calculation modules to build it in. One of the ways to make software accessible is to control and limit the input and output options. You will not need to do that with your own method. Just be careful that your method does not limit your design.

    Hope something in all of that is useful.

    Kev

+ Reply to Thread
Page 1 of 4 1 2 3 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts