1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Parachute deployment (non-instantaneous)

  1. Sep 9, 2006 #1
    Here's the problem...

    A car of mass 2000 kg, drag coefficient .3, frontal area 1 sq. meter releases parachute of area Pi sq. meters (diameter 2 m). The inital velocity of the car is 100 m/s. Calculate distance and velocity after 10 seconds.

    I know how to work this problem for instantaneous deployment. However, if the deployment is non-instantaneous, I need to make a few assumptions. The first is that the area of the parachute increases linearly with time: A(t) = A0 + kt, where A0 is the initial area of the undeployed parachute and k is the time rate of area increase.
    ***What should I assume for A0???
    ***Is it reasonable to assume that the parachute can be deployed in 3-5 seconds? Is this too great a time interval? Too small? From this I can determine the approximate value of k (area increase per second).
    ***I was told I can assume the coefficient of drag remains constant.

    Does this sound like a reasonable model for non-instantaneous deployment? If so, I know how to find the desired values for distance and velocity. Am I on the right track???

    Thanks for any input!
     
  2. jcsd
  3. Sep 10, 2006 #2
    3-5 secs seems ok.

    I happen to know that at terminal velocity on a parachute jump i have pulled at 5500ft, on a student canopy (large area), and been under a fully inflated canopy in secs.

    They train you to count the inflation before making malfunction checks, and its about 5 sec countdown.

    With modern chutes the slider has a big effect on the deployment time. It catches the air and slides up along the risers slowing deployment and reducing Jerk.
     
    Last edited: Sep 10, 2006
  4. Sep 10, 2006 #3
    For area at t0 i would take the vehicles area only ( i.e. the drag that has allways been on the vehicle ).

    The drag will be effected as the vehicle slows, but i think thats taken care of in the drag coefficient. Im not familiar with the equation for the coefficient of drag.

    Seems like a good start to me, same as i would do, though i would prob try and define a curve for the deployment :P
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Parachute deployment (non-instantaneous)
  1. Instantaneous Voltage (Replies: 2)

Loading...