Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

What is the best way to mesh this part?

Status
Not open for further replies.

drennon236

Civil/Environmental
Mar 27, 2020
102
My assembly consists of three parts; the inner part (first picture - 45 m long) is of steel and has shear keys (circular steel extrusion) to increase friction so the parts dont slide. In the middle (2nd picture - 17.3 m long) is made of concrete, and the third picture (26 m) is also a steel part with inner shear keys. I was wondering how to best mesh the parts to reduce computational time. I need most information for the area around the shear keys and grout, I was thinking of having small element size around the shear key area, and then biased seeding away from the shear keys. Is there a better way to do this? Plated steel is meshed with one 2nd order, reduced integration element though the thickness. For the grout body a minimum of three 2nd order, reduced integration elements through the thickness is recommended.

Capture_scpbh2.png
Capture2_tmypwt.png
Capture1_c6chr3.png
 
Replies continue below

Recommended for you

You can create a partition to separate the region with shear keys and assing smaller element size to it (via local seeds). The rest should be easy to mesh since these are just regular cylinders (very thin though).
 
And biased seeding up from the partition above the shear keys, and then biased seeding down from the partition below the shear keys? - because need for detail decreases as we get further away from shear keys. And can I assign all these mesh controls on one part?


 
Sure, you can use bias to coarsen your mesh as it gets further away from the region of interest. Thanks to partitions you should be able to assign these mesh controls to the part. You can use vertical edges for local seeding with bias in areas without shear keys.
 
Hey FEA way, what I am trying to do is to use as few elements as possible and get convergence. I did use mesh bias and I think it helped. But I only get convergence at very high amount of elements (150 000+), I believe I am doing something wrong but not sure what. I am doing dynamic, implicit step and quasi-static application. Below is the contact surface for inner steel part (MP - first picture) and middle concrete part just to show you how I am doing my contact. Do you have any ideas how I could improve my model to get convergence faster? Please let me know if I can provide any more information.

Capture1_lve2fd.png

Capture_n7vyxd.png
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor