Skip to content

Commit

Permalink
bouncing-balls: Analyze some drops from Freesound
Browse files Browse the repository at this point in the history
  • Loading branch information
jonnor committed Mar 2, 2024
1 parent de54eac commit c45fdaf
Show file tree
Hide file tree
Showing 2 changed files with 982 additions and 13 deletions.
26 changes: 13 additions & 13 deletions handson/bouncing-balls/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,16 +5,17 @@ Temporal distances

## Time difference of a bouncing ball

Max Clark asked on Stack Overflow in [Conversational Neural Network for Peak Detection and time difference of a bouncing ball](https://stackoverflow.com/questions/69426301/conversational-neural-network-for-peak-detection-and-time-difference-of-a-bounci).

Max Clark asked on Stack Overflow
Conversational Neural Network for Peak Detection and time difference of a bouncing ball
https://stackoverflow.com/questions/69426301/conversational-neural-network-for-peak-detection-and-time-difference-of-a-bounci
Goal is actually to get the heigth of the initial height.
There is information in that in the time difference between first and second bounce, but also in the distance between second and third, etc.

Target here is actually to get the heigth of the ball bounce.
There is information in that in the time difference between first and second bounce,
but also in the distance between second and third, etc.
So the neural network should maybe just
I think that to get the height from only two ground touches,
we would need to also to know the COR of the ball.
But if we have 3 bounces, one can estimate the COR.
Thus we should be able to get height, without knowing any further data?!

### Background
The bounceback amount (ratio?) is a way to measure how springy the ball and surface (combined) are.

https://demonstrations.wolfram.com/BounceTimeForABouncingBall/
Expand All @@ -34,19 +35,18 @@ Coefficient Of Restitution (COR) is mostly constant,
but some indications that it can change by up to 10% when height is increased 10x.

Sports balls have standardized COR values
Table tennis ball. Shall bounce up 24–26 cm when dropped from 30.5 cm on to a standard steel block. COR of 0.887 to 0.923.
Basketball. Shall rebound to between 960 and 1160 mm when dropped from a height of 1800 mm. COR between 0.53–0.64.
Tennis ball. ATP regulations. Bounce to a height of 135 cm and 147 cm when dropped from 254 cm on a hard surface. COR of between 0.728 and 0.759.
Golf ball. Roughly 0.70
Baseball prewar construction is about 0.45.

- Table tennis ball. Shall bounce up 24–26 cm when dropped from 30.5 cm on to a standard steel block. COR of 0.887 to 0.923.
- Tennis ball. ATP regulations. Bounce to a height of 135 cm and 147 cm when dropped from 254 cm on a hard surface. COR of between 0.728 and 0.759.
- Golf ball. Roughly 0.70
- Basketball. Shall rebound to between 960 and 1160 mm when dropped from a height of 1800 mm. COR between 0.53–0.64.
- Baseball prewar construction is about 0.45.

For drop from rest, ignoring air friction
C_R = \sqrt{\frac{h}{H}}, where
h {\displaystyle h} h is the bounce height
H {\displaystyle H} H is the drop height


https://www.millersville.edu/physics/experiments/045/
Bounce-to-bounce time is proportional to the square root of the bounce height.
Only dependent on the COR?
Expand Down
Loading

0 comments on commit c45fdaf

Please sign in to comment.