FAQ  •  Login

Exam 2 Take Home Component

<<

F11Carrie

Newbie
Newbie

Posts: 23

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Mon Dec 12, 2011 6:40 pm

Re: Exam 2 Take Home Component

F11Christopher wrote:Is see the exact same results as Ryan. I used the slider in the far left position as my Initial state. That means my x3 was 1.5 and phi1=pi. The rest of the positions follow from this. I set all velocities and accelerations equal to zero except for phi_dot_1 which I set to 30 as the book suggested. I'm not really sure about all the other accelerations but I'm pretty confident in all the positions and accelerations I chose.

My frequency is much lower than the books solution and the angular velocity of the crank seems to diminish over time.

Again, I'm with Ryan. Any other suggestions?



Body 2 should have non-zero ydot and phidot based on the way the revolute joint between bodies 1 and 2 moves.
<<

F11Christopher

Newbie
Newbie

Posts: 22

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Tue Dec 13, 2011 12:45 am

Re: Exam 2 Take Home Component

F11Carrie wrote:
F11Christopher wrote:Is see the exact same results as Ryan. I used the slider in the far left position as my Initial state. That means my x3 was 1.5 and phi1=pi. The rest of the positions follow from this. I set all velocities and accelerations equal to zero except for phi_dot_1 which I set to 30 as the book suggested. I'm not really sure about all the other accelerations but I'm pretty confident in all the positions and accelerations I chose.

My frequency is much lower than the books solution and the angular velocity of the crank seems to diminish over time.

Again, I'm with Ryan. Any other suggestions?



Body 2 should have non-zero ydot and phidot based on the way the revolute joint between bodies 1 and 2 moves.


Got it!, Thanks Carrie. Also, my lambda's are also negative what the book shows for the reaction force. This is correct because the equation for the reaction force is
-Phiq *lambda. The negative out front corrects the sign of the lambda and the solutions match.
<<

F11Christopher

Newbie
Newbie

Posts: 22

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Tue Dec 13, 2011 1:40 am

Re: Exam 2 Take Home Component

Here are my results.
Attachments
Plots.png
Plots.png (15.81 KiB) Viewed 7510 times
<<

F11Graham

Newbie
Newbie

Posts: 11

Joined: Wed Sep 14, 2011 9:56 am

Unread post Tue Dec 13, 2011 2:32 pm

Re: Exam 2 Take Home Component

Is anyone else having problems with their system speeding up? I think this has something to do with my input files but I've checked them over a bunch and I can't find anything wrong with them. Here are my plots. Also, is it necessary to specify initial velocities for the bodies? I haven't been doing that and my system can converge. In fact when I do specify initial velocities it screws up my x reaction force plot.
Attachments
takeHomeResults.jpg
takeHomeResults
takeHomeResults.jpg (40.96 KiB) Viewed 7501 times
<<

F11Sarah

Newbie
Newbie

Posts: 7

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Tue Dec 13, 2011 3:57 pm

Re: Exam 2 Take Home Component

I'm having the same problem with my plots speeding up. Can't figure out where the error is. Let me know if you figure it out.
<<

Dan Negrut

Global Moderator
Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

Unread post Tue Dec 13, 2011 6:14 pm

Re: Exam 2 Take Home Component

F11Graham wrote:Is anyone else having problems with their system speeding up? I think this has something to do with my input files but I've checked them over a bunch and I can't find anything wrong with them. Here are my plots. Also, is it necessary to specify initial velocities for the bodies? I haven't been doing that and my system can converge. In fact when I do specify initial velocities it screws up my x reaction force plot.


Graham - you always need to specify initial velocities. Otherwise, how would you use Newmark method? Remember that it needs an old velocity when computing the new velocity.
If your choice of initial velocities is not correct then you'll see spikes in the value of the Lagrange Multipliers. Make sure that \Phi_q * qdot = -\Phi_t, that is, the velocity are consistent with your kinematic constraints.

For the results speeding up: do you have the gravity acting in the right direction? Also, you might have the value of the force be too large. I don't have the book with me, but if i recall correctly there was some damping applied. Do you factor that in?

Finally, plot the value of q, qdot, qddot and lambda after *one* integration time step (at time t1). Specify your integration time step, and ask one colleague who finished the assignment to plot his/her corresponding values. You'll then get an idea where things are not right.
dan
<<

F11Sarah

Newbie
Newbie

Posts: 7

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Tue Dec 13, 2011 8:19 pm

Re: Exam 2 Take Home Component

I adjusted my Q matrix to include the compression force on body 3, and now when x.3.dot drops below zero the system will not converge. Any suggestions?
<<

F11Wendy

Newbie
Newbie

Posts: 26

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Tue Dec 13, 2011 8:36 pm

Re: Exam 2 Take Home Component

F11Christopher wrote:Is see the exact same results as Ryan. I used the slider in the far left position as my Initial state. That means my x3 was 1.5 and phi1=pi. The rest of the positions follow from this. I set all velocities and accelerations equal to zero except for phi_dot_1 which I set to 30 as the book suggested. I'm not really sure about all the other accelerations but I'm pretty confident in all the positions and accelerations I chose.

My frequency is much lower than the books solution and the angular velocity of the crank seems to diminish over time.

Again, I'm with Ryan. Any other suggestions?



For this initial configuration, if your phi_dot_1 is non-zero, then the connecting rod must have non-zero y_dot2 and phi_dot2. Make use of the relation between tangential linear velocity and angular velocity/radius for circular motion. Initial accelerations should not all be zero, but these you solve for at the inital time step given your ICs for positions and velocities.
<<

F11Graham

Newbie
Newbie

Posts: 11

Joined: Wed Sep 14, 2011 9:56 am

Unread post Tue Dec 13, 2011 9:34 pm

Re: Exam 2 Take Home Component

I fixed my program and now I get the results in the text.

Sarah - I ended up finding a problem in my pistonForceDefinition function. It was pretty simple actually, I forgot to add the -5.25 in the second equation. Not sure if this helps.

Here are my new plots.
Attachments
takeHomeResults.jpg
takeHomeResults.jpg (67.06 KiB) Viewed 7474 times
<<

F11Gratz

Newbie
Newbie

Posts: 17

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Wed Dec 14, 2011 12:47 am

Re: Exam 2 Take Home Component

Here are my plots. I had that wobble at the beginning, but a smaller step size fixed it.

Also, I suppose it's kind of late in the year to be asking this, but could someone tell me what the code is to put all of the plots in one window instead of taking up the entire page...?
Attachments
Flywheel Angular Velocity.png
Flywheel Angular Velocity.png (5.03 KiB) Viewed 7471 times
Slider Velocity.png
Slider Velocity.png (4.65 KiB) Viewed 7471 times
Slider Position.png
Slider Position.png (4.65 KiB) Viewed 7471 times
Last edited by F11Gratz on Wed Dec 14, 2011 12:51 am, edited 1 time in total.
<<

F11Gratz

Newbie
Newbie

Posts: 17

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Wed Dec 14, 2011 12:51 am

Re: Exam 2 Take Home Component

One more:
Attachments
Revolute Joint Reaction Force.png
Revolute Joint Reaction Force.png (4.67 KiB) Viewed 7471 times
<<

F11Christopher

Newbie
Newbie

Posts: 22

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Wed Dec 14, 2011 3:25 am

Re: Exam 2 Take Home Component

@Gratz

Its called subplot....instead of saying figure say subplot(m,n,p). m is the number of rows and n is the number of columns. P is the position you want that particular plot to be in. I think for a 2x2 subplots the positions are as follows:

12
34

You can use help subplot if your still confused.
<<

F11Ryan

Newbie
Newbie

Posts: 9

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Wed Dec 14, 2011 4:55 pm

Re: Exam 2 Take Home Component

Thanks for the suggestions Carrie and Wendy, I was able to figure it out with your help.
Attachments
Graphs.JPG
Graphs.JPG (165.39 KiB) Viewed 7441 times
<<

F11Rebecca

Newbie
Newbie

Posts: 10

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Thu Dec 15, 2011 12:11 am

Re: Exam 2 Take Home Component

Here are my plots:
Velocity&PositionPlots.png
Velocity & Position Plots
Velocity&PositionPlots.png (9.25 KiB) Viewed 7435 times

ReactionForce.png
Reaction Force
ReactionForce.png (4.35 KiB) Viewed 7435 times

SliderForce.png
Slider Force
SliderForce.png (4.17 KiB) Viewed 7435 times
<<

F11Kwang

Newbie
Newbie

Posts: 12

Joined: Mon Sep 12, 2011 7:58 pm

Unread post Thu Dec 15, 2011 11:18 am

Re: Exam 2 Take Home Component

I had some trouble to get the graph that is consistent with the book. I tried my best, but I couldn't get it well.
Anyway, I submitted my work with little ugly graphs.
But, I still can't figure out where my program had some problem.
Could you provide me any solution for this matlab coding or an example coding for similar problem?
So, I can study the example and figure out where did I make a mistake in my coding so that I can prepare the final exam.
Is that possible?
PreviousNext

Return to ME451 Fall 2011: Kinematics and Dynamics of Machine Systems

Who is online

Users browsing this forum: No registered users and 1 guest

cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software.