Mechanism Dynamics reaction forces

[solidworm]

lets say we have this assembly:
View attachment 4688
we have defined the loads on top of the cylinder, and we want to calculate reaction forces and moments during motion on the crack shaft bearings. to calculate those forces and moments,(1)it matters where the pin joint is defined along the crank shaft.which is not important when doing kinematic analysis using MDX. also (2) it should have two bearings to support the shaft from the force analysis point of view. this means two pin joints should be defined which one of them is redundant from the assembly point of view. Does MDO takes these facts into account when it calculates reaction forces?

Edited by: solidworm

 

[solidworm]

thanks alot for your reply.
for this problem, to get bending moment reactions on the shaft, it could be modeled with 2 pin joints, which over constrains the shaft but still works or a pin and a cylinder joint which also works (without over constraining DOFs)
generally speaking, should i take caution not to over constrain DOFs?
(EDIT: i guess i found my answer. generally speaking, they shouldn't be over constrained. for example a shaft with it's axis normal to ground with gravity force applied, would have different set of axial reaction forces when modeled using pin-pin or pin-cyl joints.)



Edited by: solidworm

 

[Kaz Z06]

Using a pin joint for either bearing support will give incorrect results. This is because the joint with a pin will provide all the constraints required to provide the motion of the crankshaft all by itself. There are a couple ways to get around this issue. I have attached a file that gives good background on the topic - the help files should also provide this same info:

2010-10-05_080314_Redundancies.zip

another sure-fire way to do it is to use 6DOFs. I've modeled an entire 4 cylinder engine using 6DOFs for every bearing support with exact results. Simply limit/constrain the directions of each 6DOF with a servo motor or stiff spring and you will get the desired motion and reactions in all the directions you have constrained. If forces are unbalanced across bearing supports, you will get the proper result. Remember to make a measure that provides "Redundancies" and "DOF" so you can check your model.

 

[solidworm]

thanks paul and kaz. actually i tried a very simple case. a vertical rotating shaft under gravity load, constrained with two pin joints. each joint got half of the shaft's weight as axial force.
help files say that there shouldn't be any redundancy in the model,which makes sense.
i'll study the document and perform an analysis on some other already solved problem to verify solutions done by MDO in each case.

Edited by: solidworm

 

[Kaz Z06]

Solidworm,

Don't be fooled into a false sense of security - Mechanisms will try to divide the load among multiple, or redundant, constraints. I've seen this happen many times - two redundant pin joints can have 1/2 the load resolved to each, but that is just because it is a simple system - it will not work for more complicated supports - 3 or more. It may not provide correct results for your simple piston model of there is an imbalance of forces - it may, but you cannot count on it to provide accurate values consistently.

 

[solidworm]

thank you very much kaz, i'll definitely try your approach and learn from your experience.
another question is how to constrain 6DOF joint with servo motors? should i do it by setting the velocity of it's axes to zero? this way i will lose dynamic drag.


Edited by: solidworm

 

[Kaz Z06]

You can simply apply a "position" servo motor and set it to "zero", or whatever position represents the location you need to hold the joint axis. If you know the approximate stiffness of the support in each direction you need to constrain, apply this through a spring for that DOF axis and you will get more accurate behavior.

 

[solidworm]

i have some difficulty understanding 6DOF constraint. lets say i have a shaft like this under gravity (normal to the plane shown)

i have constrained all 3 translational degrees of freedom of the 6DOF constraint on the right with servo motors. when i run a dynamic analysis, the shaft falls and the DOF measure shows 6 degrees of freedom. i expect it to have a pendulum like motion and the degree of freedom to be 3. (only 3 rotations) because i have fixed the position of one point of the shaft

EDIT: oops! somehow the three servo motors were excluded from the motors tab in analysis definition window. it's now working as expected.


Edited by: solidworm

 

[solidworm]

i'm stumped a little bit. to make joints with 6DOF constraints, you need two coordinate systems.one on part A at the location of the joint and one on part B located locally at the joint position. what other rules should i follow?

 

[Kaz Z06]

That should be all you need to do, just make sure you place the co-ords in each part oriented so that they will line-up properly when you assemble the parts together. If you do nothing to each DOF your assembly will fall apart, so you must apply motors or forces on the DOF that must be constrained to make your assembly behave as you wish.