Jieve
Mechanical
- Jul 16, 2011
- 131
I am simulating a frame out of 3" x 2" x 1/8" rectangular steel tube and am trying to verify the beam bending stress results that I get from the simulation.
Hand-calc methodology: Using the bending moment value at an element node taken from the "list beam forces" table, the cross-section inertias calculated by the software and the total moment vector direction, I calculate the neutral axis angle. Once I locate the maximum stress location, I calculate the bending stress at that point due to both moment components about the principal axes.
When I compare my calculations to the "Bending Dir1" and "Bending Dir2" values listed under the "list beam stresses" table, they are usually off by a significant amount, sometimes 2-3x. I've tried doing hand calcs using both the principal axes and the part axes using the product of inertia (this is the equation listed under solidworks help), and while both of these methods give very similar answers, they deviate significantly from the simulation result.
According to the help page, the equation I used is the same used by the software. I initially tried verification using a complex beam cross-section and couldn't get a matching result, so switched to the rectangular tube to make sure my hand calcs were right. Still can't get a matching answer. What am I missing here?
Thanks for any input.
UPDATE: So it seems when I select the option "render beam cross-section", I get the correct extreme fiber stresses in the cross section (hand calcs pretty much dead on with the simulation). What I'm apparently not understanding are the "upper bound bending stress in DIR1 (-Ms/Ss)" and "upper bound bending stress in DIR2 (-Mt/St)". The help suggests that they are the extreme cross-section fiber stresses, exactly like I calculated above, but the values don't resemble anything that I've calculated.
Hand-calc methodology: Using the bending moment value at an element node taken from the "list beam forces" table, the cross-section inertias calculated by the software and the total moment vector direction, I calculate the neutral axis angle. Once I locate the maximum stress location, I calculate the bending stress at that point due to both moment components about the principal axes.
When I compare my calculations to the "Bending Dir1" and "Bending Dir2" values listed under the "list beam stresses" table, they are usually off by a significant amount, sometimes 2-3x. I've tried doing hand calcs using both the principal axes and the part axes using the product of inertia (this is the equation listed under solidworks help), and while both of these methods give very similar answers, they deviate significantly from the simulation result.
According to the help page, the equation I used is the same used by the software. I initially tried verification using a complex beam cross-section and couldn't get a matching result, so switched to the rectangular tube to make sure my hand calcs were right. Still can't get a matching answer. What am I missing here?
Thanks for any input.
UPDATE: So it seems when I select the option "render beam cross-section", I get the correct extreme fiber stresses in the cross section (hand calcs pretty much dead on with the simulation). What I'm apparently not understanding are the "upper bound bending stress in DIR1 (-Ms/Ss)" and "upper bound bending stress in DIR2 (-Mt/St)". The help suggests that they are the extreme cross-section fiber stresses, exactly like I calculated above, but the values don't resemble anything that I've calculated.