reaction point, cylinder connection

[rework]

I have found that mechanism dynamics is giving me the wrong force at a connection --- that connection being a cylinder. It appears there is no way to change the dimension to the point where the connection force acts. One can move the "cylinder" icon, but the force comes back the same. Am I doing something wrong, or is this a glitch in the software? WF 4.0

Thanks in advance for any insight into this matter.

Rework

 

[Kaz Z06]

Try creating a point on the body (in "Standard" mode) where you want the force to act. Then apply your force to that point in MDO - this should give you the behavior you're looking for. you can check by verifying a hand calculation for the moment about the cylindrical joint center caused by your point force.

Hope this helps,

Kaz

 

[rework]

Thanks Kas. I did create points on the bodies where I wanted the force to act, and the cylinder icon moves to that point. However, the force number stays the same.

How do you "apply your force to that point"? From what I can tell, one is working blind here as to where the force is acting. This is a centripetal force problem, so I can calculate by moments what the reaction forces should be. The force is off by about 100#, or 1 inch.

Thanks again for your interest.

Rework

 

[rework]

One more clarification as to my procedure. I use "Translation" to choose the points on the bodies. Is that necessary and/or correct?

Rework

 

[Kaz Z06]

rework,

It would be helpful to review the model you're working with, if this is possible. If not, a simple representation would help me understand more thoroughly the details of what you are describing.

Cheers,

Kaz

 

[rework]

Kaz ----Thanks so much for your help. The main issue for me is where the MDO places the line of action of the connection force. For instance, I am involving a U-joint, which as you know, has two bushings in line. When one sets this up as a "Pin" connection, you align the axes, and then set the "Translation". The icon (and the force arrow) align with the Translation point. But that doesn't appear to mean that is where the force actually acts. Because if you shift the Translation point, the force arrow follows, but the force number is the same. So it appears to me by trial & error, that the point of action of the forces can't be changed. I would like to be proved wrong!

Rework

 

[Kaz Z06]

Try changing the coordinate system that the force is created from - if you select the co-ord of the body, it should follow the body and remain oriented WRT to this body. If you select the global, or "ground', co-ord, it should remain in the same direction throughout the analysis.

Kaz

 

[rework]

I guess I don't know how to do that. When one sets-up a measure, you only have the choice of which body the force acts upon (direction of the arrow). When one makes the connection, I don't see any options for co-ord system?

By the way, I make sure the redundancies are zero.

Rework

 

[Kaz Z06]

It is more difficult to provide suggestions unless I could see/work with the model on my desktop. If you can share it, or a watered-down version, that would make a big difference.

Cheers,

Kaz

 

[rework]

Kaz --- In a bit of a pickle, since I have a V4.0 Student Edition. I can send you a JPEG of the problem, see attached. The swinging mass has a known centripetal force, which is hidden from us in MDO (correct?). But the reactions R1 & R2 add up correctly to equal that centripetal force. The discrepancy is that the numbers are wrong by about 100#. If R2 could be moved to the right by an inch or so, the numbers would be correct. This simple example doesn't give me much confidence in the program.

Of course, I am looking at the components of R1 & R2 in the proper direction.

Rework

 

[rework]

2010-11-03_190236_Swinging_weight.zip

I didn't have the JPEG zipped, so it didn't get on my last post. Here it is.

Rework