Hi guys. I´m trying to simulate a fatigue analysis on a shaft. I got one example from shigley´s book.
I made a drawing of the shaft and I defined some restrictions and loads, which I don´t know if they are correct.
So, I have some doubts I would like you to help me with.
-I made an analysis defining the fixtures with the bearing fixture option, and made another with the cylindrical surface fixture option restraining the radial movement on one and the radial an axial movement on the other letting the rotation free(I guees). The results were diferent, and neither of them matched with the results of the book. With results, I mean the reactions...So, which is the difference between defining the bearing fixture option and the cylindrical surface fixture? Besides, in the bearing fixture there is a Stabilize shaft rotation option whitch also changes the results if is activated. What is this option for? When this options wasn´t activated the program couldn´t make the analysis because of the big displacements.
For the fatigue analysis: Should I make 4 diferent simulations(2radial forces and 2 tangencial forces? o just 1 with all of them since they are all alternating?
Thank you very much. I leave you the files if you want to see them.
SolidworksSimulation
https://forum.solidworks.com/thread/74624?tstart=0
i wonder if you guys are doing the same analysis?
same suggestions for what you're trying to do here. understand the assumptions the calcs and the analysis make and then go from there.
thanks Jared.
they are not the same analysis. the difference is that i`m trying to modelate a problem which aswer is known.
i will investigate more about these topics.
hi again. I read this in other post:
that helped me a lot to solve this example: (i got the correct answer with stabilization only on one bearing fixture)
But with this other example, the solution was only right without stabilization
with stabilization (no torque trought the right part of the shaft)
without stabilization
Can you explain me why is that?
thanks
i'm not sure i'm clear on whats wrong with the without stabilization one/
i was thinking of this:
in the first model, the remote loads produce moments which have opposite directions, but because of there are small differences in the magnitude of the loads or the distance, i need to enable the stabilization to get the correct answer.
solidworks model:
But in the other example, there is only one load defined, and if i use the stabilization on one bearing fixture, then, the program will apply and opposite load of the remote load so there will not be torque in the right part of the shaft.
Jared, what do you think of my hypothesis?
can you setup and run this problem without stabilize?
for example what about cutting this guy down the middle and applying symmetry. this would eliminate the need for stabilize. i suspect that is the root of your problem.
Hi Jared. Sorry for the delay, i was really busy.
I can´t run the problem without stabilize. Some solver error came up.
If i aplly symmetry without stabilize, the same error came up.
the only way to solve it was with the stabilize.
the numerical results were different of the foregoing post. i think that is because the loads are not symmetric.
What is the exact error?
Please setup and post a sample model.
Have you spoken with your var on this yet?
the error is:
"the solver has numerical problems
the model may not have the adequate retrains"
sorry, who is the var?
how is the torque being resisted?
based on the foregoing image, the symmetry should create an opposite foce. those forces creat opposite moments which make the problem static.
am i right?
VAR = value added reseller, the company that supports your use of solidworks.
insufficient restraints, this doesnt' surprise me if you're still using bearing restraints. as discussed above, they do not provide rotational restraint so any offbalance = insufficient restraints.
i think it would be best to post a sample model at this point to know where you are at on this.
what do you mean with a sample model?
sample model = model that you can post without concern for the intellectual property
also helps yo uconfirm the problem that you're running into isn't file and setup specific. setup simple geometry that describes your problem, setup the same analysis and see what happens.
ok, i will make a similar example and post it.