SolidWorks Tip: Projectile Motion of Ball

[BA SW]

This is to show teams how to use motion study in SolidWorks to check if their ball will reach the target or not.

Watch: https://www.youtube.com/watch?v=wWAakaA3wAA
Download:projectile.zip

I have attached a zip file that contains the assembly I used in the video. Once you have extracted everything from the zip file, you can open the assembly and run the motion study with your own numbers.

You can change:
- position of the target (edit the sketch and modify the x and y values)
- launch angle (edit the sketch of the 3 lines and modify the angle dimension)
- force exerted on the ball and for how long (in motion study, change the force, or add your own force. Also, you can set how long the ball is in contact with the device you are using to launch it or add multiple forces according to time. This is shown in the video)
- change the mass of the ball (by overriding the default mass properties values of the ball)

I will try to get a video on optimizing this as well.

[Ether]

What are you using for the differential equations of motion once the ball leaves the launcher?

[BA SW]

I am not inputting any equations, the software is running equations that were programmed into it. The video just shows how to run the simulation in SolidWorks. So I am just setting up the variables in the video.
It just runs the equations for projectile motion.
The x component is constant and can be calculated by trigonometry. Just split up the x and y components of the initial force. Then it knows the y component and acceleration due to gravity.
You could equally solve the same thing that is being simulated by hand.

This simulation, however, shows you a visual representation and could be faster if you need to keep making iterations. The simulation could be made more accurate by accounting factors like drag by applying an additional force into the simulation. I am also trying to put together a video of how the software can optimize the values of the force so the ball hits the target. This is like asking the software to work backwards and solve the inputs for what you want it to do.

I hope I have answered your question. Please let me know if you want me to clarify a few things.

[Ether]

Originally Posted by BA SW

It just runs the equations for projectile motion...
I hope I have answered your question. Please let me know if you want me to clarify a few things.

"Projectile motion", as in "no air drag, just gravity"?

[BA SW]

Originally Posted by Ether

"Projectile motion", as in "no air drag, just gravity"?

Yes, the simulation does not account for drag. However, it would be easy to add drag. Additional forces can be added in the simulation and then by running the same simulation you would get results with the drag included. I just did not put that in the video.

There are also flow simulations in SolidWorks if you are concerned with aerodynamics and drag.

[Ether]

Originally Posted by BA SW

it would be easy to add drag. Additional forces can be added in the simulation

I couldn't see what you were doing on the screen in the video. Can you add a force that varies with the direction and magnitude of the ball velocity?

Air drag (and possibly backspin/topspin for some shooters) can have a nontrivial effect on the flight path of this year's game piece (because of its large size and light weight).

[BA SW]

Originally Posted by Ether

Can you add a force that varies with the direction and magnitude of the ball velocity?

Yes. There are different forces you can add in the simulation, including equation driven forces where the forces vary according to time for example. You would probably want an equation driven function for the forces in your example.

I tried to keep the video simple so I used an easy step function for the force. If you have data, you can even import data of these forces and run the simulation to see how the drag data you collected influences the ball.

The types of forces that you can use in this simulation are:
- Constant
- Step (choose how much force at a specified moment in time and how long this force lasts)
- Harmonic (oscillations of forces)
- Segments (how much force to apply over what range)
- Data points (with data you collect)
- Expression (an equation that varies the force on the object as a function of time)

I also want to point out that in the video, I went with the default: Action only force, and you can pick action & reaction instead.

Originally Posted by Ether

Air drag (and possibly backspin/topspin for some shooters) can have a nontrivial effect on the flight path

backspin/topspin are harder to simulate. The Pumkin Chunkin competition could also be a good place to look for more accurate SolidWorks simulations.

[Ether]

Originally Posted by BA SW

The types of forces that you can use in this simulation are:
- Constant
- Step (choose how much force at a specified moment in time and how long this force lasts)
- Harmonic (oscillations of forces)
- Segments (how much force to apply over what range)
- Data points (with data you collect)
- Expression (an equation that varies the force on the object as a function of time)

Just so I understand: Is the above a comprehensive list of what is supported by SolidWorks, or a subset that is supported in the demo you made?

I don't see anything in that list that allows you to add a force which is a function of the ball's velocity (direction and magnitude).

[BA SW]

Originally Posted by Ether

Is the above a comprehensive list of what is supported by SolidWorks

Yes, these are the list of forces you could apply on the motion study. I have been using the word simulation interchangeably with motion study and i probably should not have done that. If you want to know more about the properties of the ball, like aerodynamics, you could run a simulation, which is different than the motion study video i posted above. the simulation could give you results like drag that you could plug into the motion study.

Originally Posted by Ether

I don't see anything in that list that allows you to add a force which is a function of the ball's velocity (direction and magnitude).

Correct, but there is a way around this. If you split up that function into x and y components, then you could apply a force as a function of the balls velocity. For example, when you are applying the additional forces in the motion study, set the direction of the force and use an equation that determines the magnitude of the force per unit time in that direction. So you would have a force that represents the magnitude per time in a given direction. Do this in both the x and y directions and you should get what you want. (you can also do this in 3D and include the z direction)

Also, you have the option to plot graphs of the motion study. These graphs can be about a wide range of values from linear or angular velocities and accelerations, pitch/yaw/roll, to angular momentum and kinetic energy. There is a fairly big list of motion study results that can be graphed.

[Ether]

Originally Posted by BA SW

use an equation that determines the magnitude of the force per unit time in that direction

You do not have a priori knowledge of that equation (drag force as a function of time). That's an output of the simulation, not an input to it.

Also, the drag force in the X direction depends in both the X and Y components of the ball velocity (ditto for the drag force in the Y direction).

[BA SW]

Below is a link that explains how to run a flow simulation in SolidWorks:
http://personal.stevens.edu/~ffisher...oefficient.pdf

You need to input parameters that will influence the drag of the part to run this simulation, or just use the default values and simulate based on the shape of the part.

This simulation can give you an output as a drag force in a direction or a drag coefficient (among other variables that you ask it to solve for).

The data that you obtain from this simulation can be applied as a force in the motion study. The motion study would be separate than the flow simulation and you would need to first run the flow simulation before running the motion study. The results you obtain from the flow simulation is what you could use as an input to the motion study.

At this point this becomes a complicated and time consuming series of simulations and motion studies. I admit that the process would be easier if you could input data as a function of velocity (or at least i am not aware of how to do this for a motion study). Maybe there is a way by adding an equation driven linear motor in the motion study. However, I think it would be a better use of time to optimize the simple projectile motion rather than go through multiple simulations to accurately account for drag.

[Ether]

FWIW, Team 4265 says they have a model based on a linear approximation for drag vs velocity which they have validated with test data:

http://www.chiefdelphi.com/forums/sh....php?p=1339667

Still waiting for them to post the "k" value they determined for this year's game piece to make the model agree with their test data.