1. Limited time only! Sign up for a free 30min personal 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!

Speed of Sound and its relation to weight?

  1. Jul 16, 2015 #1
    Good day (or night). I am new here, so I hope my question doesn't bother many.

    If (forgetting other laws of nature) a shoebox where to hit the sound barrier and an F-35 were to do the same. Would the sound waves be equal. Would it sound the same to a ground observer?

    I always assume that mass matters (no pun intended).

    Many Thanks- CuriousMonkey
     
  2. jcsd
  3. Jul 16, 2015 #2

    phinds

    User Avatar
    Gold Member
    2016 Award

    SIZE matters because a really large moving object creates a more forceful shock wave than a really small one.

    If a bullet were fired out of a high-up balloon, would you expect it to be perceived on the ground the same as the shockwave from a huge supersonic aircraft?
     
  4. Jul 16, 2015 #3

    Drakkith

    User Avatar

    Staff: Mentor

    From here: https://en.wikipedia.org/wiki/Sonic_boom#Causes

    The power, or volume, of the shock wave is dependent on the quantity of air that is being accelerated, and thus the size and shape of the aircraft. As the aircraft increases speed the shock cone gets tighter around the craft and becomes weaker to the point that at very high speeds and altitudes no boom is heard. The "length" of the boom from front to back is dependent on the length of the aircraft to a power of 3/2. Longer aircraft therefore "spread out" their booms more than smaller ones, which leads to a less powerful boom.

    Since a shoebox would accelerate a much smaller amount of air than most aircraft, the sonic boom would from the shoebox would be much weaker than the one from the aircraft.
     
  5. Jul 17, 2015 #4
    Thank you for the concise answers.
     
  6. Jul 17, 2015 #5
    There are 600 and 1000 yard shooting matches where target scorers sit behind a bullet proof berm and pull targets down to indicate the location of the hit and the score with markers visible way back at the firing line.

    In this location, the sound of the guns going off is not audible, but the sounds of the supersonic bullets passing overhead is audible, because each one has a Mach cone and the shock is clear and evident.

    The sound of a 30 caliber bullet (7.62mm diameter) is much louder than the sound of a 22 caliber bullet (5.56 mm), and the sound of bullets passing over at Mach 2 (say at 600 yards) is much louder than the sound of bullets passing over at Mach 1.2 (at 1000 yards.)

    These Mach cones made by bullets can be used by acoustic systems not only to determine where a bullet hits on a target, but also to trace the entire bullet trajectory back to the shooter. See: http://www.raytheon.com/capabilities/products/boomerang/

    The weight is not the key parameter. The key parameters are cross sectional area and coefficient of drag.
     
  7. Jul 17, 2015 #6

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    I'd argue that isn't entirely correct. You could almost certainly correlate the volume of the sound with the drag coefficient, but this is because the drag on a supersonic object is dominated by wave drag. Really, then, the drag coefficient depends on the shock strength and the amount of air being displaced just like the volume does. I'd contend that the drag coefficient is a derived quantity here and that the real key parameters are going to be the size, shape, and speed (but still not weight).
     
  8. Jul 17, 2015 #7
    Sure, but if you actually have to predict something measurable, it is much easier to start with speed, cross sectional area, and drag coefficient than size, shape, and speed.

    If you have an accurate drag coefficient, you are well ahead of the game compared with someone starting with size and shape. Theoriests _think_ they can compute accurate drag coefficients from size and shape, but those of us who can measure drag coefficients to 1% embarrass them again and again.
     
  9. Jul 17, 2015 #8

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    No theorist worth his salt will make that claim. If they do then they don't properly understand what goes into drag. Supersonic drag coefficients have always seemed even sketchier to me since they aren't constant with speed.

    So I guess in my mind the most "predictable" quantities are going to be cross-section (or cross-sectional profile a la the Whitcomb area rule), the shape of the front (oblique vs normal vs bow shock), and speed. I suppose if you have a measured drag coefficient that works too.
     
  10. Jul 17, 2015 #9
    The Army's big honcho in external ballistics for decades, Rob McCoy, did, in fact, claim to have written a computer program called "MCDRAG" capable of using shape to predict drag coefficients to "within 3% error at supersonic speeds, 11% error at transonic speeds, and 6% error at subsonic speeds." Since he wrote many influential papers on external ballistics as well as the book, Modern Exterior Ballistics, his work was very well received, and his program for computing supersonic drag coefficients was widely used for many years.

    See:

    https://scholar.google.com/scholar?hl=en&q=Robert+McCoy+ballistics&btnG=&as_sdt=1,11&as_sdtp=
     
  11. Jul 17, 2015 #10

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    MCDRAG sounds like something you buy at McDonald's. In fairness, drag would be easier to predict on smaller objects like bullets or artillery shells where viscous drag would be substantially lower.
     
  12. Jul 17, 2015 #11
    Many of the drag coefficients for computed by MCDRAG end up being 10-15% too low compared to measured values.

    Free flight measurement of drag coefficients has gotten easy to do for projectiles in flight.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook