## Assn 10

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

### Re: Assn 10

ME751James wrote:For number five, are we supposed to derive the coefficients for order 4 BDF?  Or is the table in the notes accurate enough?  I'm having problems with my BDF and I'm not sure if I'm supposed to do all of the work of solving the order 4 coefficients.

I used the coefficients from the table and was able to get it to work.  Problems I ran into were not setting the first value of the exact answer to 1, and not aligning the time points up correctly.

ME751 Student

Posts: 13

Joined: Tue Mar 02, 2010 10:20 pm

### Re: Assn 10

Code:
I used the coefficients from the table and was able to get it to work.  Problems I ran into were not setting the first value of the exact answer to 1, and not aligning the time points up correctly.

Yeah, mine was a similar indexing problem.  I'd be nice if reality was wrong once or twice though...

Global Moderator

Posts: 833

Joined: Wed Sep 03, 2008 12:24 pm

### Re: Assn 10

ME751Chris wrote:is F the forces in the global reference frame? so [0 0 -9.81*mass]'

Jbar is simply a diagonal matrix filled with the value from 1/12*mass*4^2 right?

Force is in the global reference frame.
you do not have torque produced by the gravity since it acts at the center of mass.  for the torque, there is a term like \omega-bar-tilde J-bar omega-bar that should be accounted for.
Jbar is diagonal, with Jxx, Jyy, Jzz.

On a different note, something is wrong with my browser i guess, i still can only see two plots...

checking out.
dan

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

### Re: Assn 10

Here is a nice website to help calculate Jbar.
http://www.roymech.co.uk/Useful_Tables/ ... ertia.html
courtesy of Tyler

ME751 Student

Posts: 4

Joined: Tue Mar 02, 2010 10:19 pm

### Re: Assn 10

convergence plot
Attachments

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

### Re: Assn 10

anyone try to change the order in which their constraints are arranged in their mdl file?  I changed mine around and it affects my inverse dynamic problem, but not the kinematic problem (differences are only observed in the terms in which they should be zero, otherwise everything else is the same).  Since it doesn't change my solution to the kinematic problem, I feel like I have my matrices arranged correctly.  What could be causing this?  This is how I'm solving my inverse dynamic problem:
tau = -wbarsquig*Jbar*wbar;
lambda = [phi_r pibar]\-[M*rdotdot-F; Jbar*wbardot-tau];
Torque(:,crntOutputPoint) = -pibar(3,:)'*lambda(3);

Where the 3rd row is my DP1 driving constraint.
Last edited by ME751Chris on Thu Apr 08, 2010 1:59 am, edited 1 time in total.

ME751 Student

Posts: 25

Joined: Tue Mar 02, 2010 10:19 pm

### Re: Assn 10

When calculating the reaction torque.  should we use the entire matricies or just the rows corresponding to the driving constraint?

so is it

n = phisubr' * lambda

or

n = phisubr(6,:)' * lambda(6)

(When I use the whole matricies the maximum and minimum is about 1/3 greater than when I use just the driving constraint)
Last edited by ME451Tyler on Thu Apr 08, 2010 2:01 am, edited 1 time in total.

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

### Re: Assn 10

Dan Negrut wrote:
ME751Chris wrote:is F the forces in the global reference frame? so [0 0 -9.81*mass]'

Jbar is simply a diagonal matrix filled with the value from 1/12*mass*4^2 right?

Force is in the global reference frame.
you do not have torque produced by the gravity since it acts at the center of mass.  for the torque, there is a term like \omega-bar-tilde J-bar omega-bar that should be accounted for.
Jbar is diagonal, with Jxx, Jyy, Jzz.

On a different note, something is wrong with my browser i guess, i still can only see two plots...

checking out.
dan

I'm using mozilla firefox and it seems to work well.

ME751 Student

Posts: 55

Joined: Tue Mar 02, 2010 10:15 pm

### Re: Assn 10

ME751Chris wrote:anyone try to change the order in which their constraints are arranged in their mdl file?  I changed mine around and it affects my inverse dynamic problem, but not the kinematic problem (differences are only observed in the terms in which they should be zero, otherwise everything else is the same).  Since it doesn't change my solution to the kinematic problem, I feel like I have my matrices arranged correctly.  What could be causing this?  This is how I'm solving my inverse dynamic problem:
tau = -wbarsquig*Jbar*wbar;
lambda = [phi_r pibar]\-[M*rdotdot-F; Jbar*wbardot-tau];
Torque(:,crntOutputPoint) = -pibar(3,:)'*lambda(3);

Where the 3rd row is my DP1 driving constraint.

Figured out the problem.  I hadn't transposed the values of phi_r or pibar,
lambda = [phi_r'; pibar']\-[M*rdotdot-F; Jbar*wbardot-tau];

ME751 Student

Posts: 13

Joined: Tue Mar 02, 2010 10:20 pm

### Re: Assn 10

My convergence plot

Attachments

ME751 Student

Posts: 5

Joined: Tue Mar 02, 2010 10:17 pm

### Re: Assn 10

My convergence plot.
Attachments
Previous