+ Reply to Thread
Results 1 to 9 of 9

Thread: Torsional Rigidity Test Survey

  1. #1
    Hey guys!

    I'll be focusing a good deal of attention for my Senior Design Project next semester on the testing, analysis, and theory of torsional rigidity for a racecar. I'd like to gather some info on the test setup/procedures different teams use and compare each one to determine/verify the most accurate setup.

    So if your team is willing to give me some basic info I'd love to hear from you. This is strictly optional (obviously) and I don't need actual numbers, just how/when/if you guys run your TR tests. You can either post here on this thread or PM in the forums if you want to keep your info private. I can also be reached at sampsonrm (at) comcast (dot) net.

    My hope at the end of this is to compile and publish a paper on the relative merits/accuracy of different test procedures for measuring the torsional stiffness of a racecar chassis. Obviously no personal team/university details will be included, but if you would like your info to be excluded from the data in the paper, please mention as such.

    Here's the info I'd like to get:

    (Including team name/university name is optional)

    1. What type of chassis do you run? (spaceframe, monocoque, hybrid, ...)

    2. Does your team run a TR test?

    3. When during the competition cycle (expected and actual) do you run your TR test? (# months prior to competition)

    4. Do you run FEA simulations during the design process? If yes, what simulation do you run? (frame only, through modeled suspension, static/dynamic/both, ...)

    5. What stiffness parameter(s) do you measure? (Frame only, installed stiffness w/ suspension, both, stiffness distribution, ...)

    6. How do you set up your physical test? (fixtures, loads, gauges, ...)

    7. Lastly, how do you determine a target stiffness value in the design phase? (compare to rigid case, chassis efficiency, 10x difference in roll stiffness, LLTD approach, ...) (This is a very loaded question and extremely optional, as some consider it a "trade secret")

    You can be as detailed or as vague as you'd like. Again, the goal for this is strictly to create a data pool with which to run statistical analysis.

    I look forward to hearing from you all!
    San Jose State University

    FSAE Chassis and Ergonomics Lead '12-'13
    FSAE Chief Engineer '11-'12
    FSAE Chassis Technical Lead '10-'11
    Formula Hybrid Chassis Grunt '09-'10

    "A designer knows he has achieved perfection not when there is nothing le

  2. #2
    I answer on the basis of when I was involved

    1. Tubular steel spaceframe with composite stress panels as side impact structure and floor

    2. yes

    3. after the car is put together (expected, actual varies)

    4. yes, both frame with virtual suspension (force vectors) and modeled suspension (later in design phase). Static, dynamic and modal.

    5. frame only, stiffness distribution in frame. Suspensions guys take concern on installed stiffness etc

    6. forced displacement of front wheels with the car rolling on corner scales (i.e. only -z constrained at each wheel). Wheel displacements are known as they're changed, chassis stiffness is measured with two inclinometers attached to the frame: one between front front a-arms and the other between rear rear a-arms. Pros: no disassembly required, fast to perform, low friction, "realistic loads", both corner and frame deflections measured, possibility to measure stiffness distribution

    7. trade secret

    Take care.
    "...when this baby hits 88 miles per hour... you're gonna see some serious shit" - Dr. Brown

  3. #3
    1. Spaceframe

    2. Yes

    3. -(4-6) (When the car is retiring after comp)

    4. Yes, static with frame only. FEA values fromt last year car validated/corrected with it's TR test for next design cycle.

    5. Installed stiffness with suspension (rigid springs and no powertrain). Deflection is measured at 10 or more points at some load steps.

    6. Rigidly mounted on the rear hubs. Front hubs loaded in torsion with a bar on a pivot. Angle and deflection measured with dial gauge in and out of measurements point (Front hub, different place on frame, rear hubs (Setup compliance)).

    7. Mostly LLTD based approach.

    Good luck
    :::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::
    2007-2012 - Suspension, chassis, and stuff (mostly stuff)
    Université de Sherbrooke

  4. #4
    Answers for 2004 UQ Racing Team, Australia
    We race in December.

    1. spaceframe

    2. Yes Torsion tested

    3. Torsion tested in October, 2 months before competition.

    4. FEA in January, 11 months before competition
    FEA matches physical testing

    5. Hub to Hub, with dumbshocks that have rod end bearings.

    6.
    Try as best possible to ensure translation in for-aft and left-right is free at all 4 corners.
    Corner A - Translation Up fixed
    Corner B - Translation Up fixed
    Construct a rotating beam
    Rotating beam ensures Corner C Translation Up is equal to Corner D Translation Down
    Apply torque to rotating beam.
    Use Laser pointed to measure change in angle across hubs


    7.

    see this thread

    Design for 3,000 Nm/deg in FEA
    Expect for 2,000 Nm/deg in physical

    The testing was performed by Jon Blessing.

    FEA by Mark Fenning and Frank Evans.




  5. #5
    Show me yours!

  6. #6
    I'm surprised at the lack of actual chassis torsion tests here - everyone seems to be doing full compliance tests. While it's good to know how much (or little) your corner assemblies add to overall rigidity, an industry standard torsion test does not include any suspension components. This is why you see new and updated sports cars come out bragging about "50% stiffer chassis" not "50% stiffer chassis and hubs."

    Frank seems to pretty much have the system down, though there is one critical portion regarding the calculation/data processing that nearly everyone forgets. Unfortunately it's also the reason that my bosses get paid the big bucks, and I get paid to stay quiet about it so I can't share.

    Of course the secret calculation only applies if you are measuring chassis deflection at multiple points, rather than just between opposite corners as I expect most FSAE teams are doing.

  7. #7
    Senior Member
    Join Date
    Sep 2002
    Location
    Perth, Western Australia
    Posts
    717
    Sorry no pics ... not out of secrecy, just out of the fact that I keep far too few pics from working on these cars.

    Pretty much the same setup as Frank, however you can use the weight of the car as your torsional load by simple use of a screw jack at the corner where the weights are applied. Without the addition of aero it will represent the maximum moment that can be applied to the car through normal handling motions. Also like to have a set of scales under each corner.

    Have tested cars chassis only, full car bolted to a rig that could apply ridiculous loads, and with the compliance test as per Franks pics. In all cases have had times where measurements were taken along a few places to determine the linearity of stiffness.

    After all the tests I don't see too much point in just testing the chassis. It is hard to load anywhere close to how the suspension will feed loads, as opposed to applying the loads through the suspension. Furthermore it is all the compliance that really determines what is happening in the car. That being said every team I have been involved with focused on building stiff chassis structures, and not the low chassis member lightweight noodles. Switching priorities may lead to this sort of testing offering more insight.

    The test that I am generally most concerned with is around the light load areas and reversal of load. You tend to see a fair amount of hysteresis in the measurement of the chassis deflection when reversing load. This is very noticeable at the sorts of loads the car sees on track.

    In a design spec sheet you might note a hub-to-hub stiffness of X nm/deg but the reality at low loads and at reversal it will be a LOT lower. Lots of bolts and bearings means a build-up of slop. For this reason the physical stiffness of the chassis is not that important. It is amazing how much better the suspension of a car can be made by taking a lot of care with bolt and bearing tolerances, not to mention wheel and upright stiffness.

    Kev

  8. #8

  9. #9
    +1 to what Kevin said

    If you are using scales and jacks, note your scale accuracy, and significant digits.

    Also, make sure you dont yield the frame of your FSAE car. If you are not sure, I'd try and keep the twist under 1 degree.

+ Reply to Thread

Posting Permissions

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