[Shruti]

Hi,

I am a beginner to ICEMCFD. I have been trying to block a vertebrea in the cervical spine. I have attached images of my blocking structure with the negative elements. I have tried a lot of methods like moving vertices, splitting blocks, ogrids, mesh smoothing but I am not able to improve quality above around -0.1. I want to ask for any other suggestions I might try to improve the quality.
Pardon me if I haven't provided enough details to properly understand the question, I would be glad to provide any further information. If you want to take a better look at the actual model files, I can send them as a private message or through email.

2.jpg

3.jpg

4.jpg

5.jpg

6.jpg

Thanks!

[bluebase]

Huh, your blocking looks like you put a cumbersome amount of work into it!
For such a complex geometry you have to be meticulous about (all kinds of) associations.
I can't determine any explicit issues remotely via your screen shots (...other than that, your blocking contains some very distorted blocks, which likely return bad quality elements).

Assuming you could solve the negetiv vol. issues, achieving a good quality structured mesh will still be difficult.
Have you considered an unstructured mesh?

With regards,
Sebastian

[Shruti]

Hi Sebastian,

Thanks for your reply.
I am eventually going to convert this structured mesh into an unstructured mesh in order to run my analysis in Abaqus. Are you saying that this method may not provide me a good quality mesh?
My analysis is in non-linear domain and I have heard that hex-mesh is better and the analysis also runs faster for that, hence I went down this route. Frankly, we are still debating here whether blocking is really going to be helpful, but none of us know enough to give a proper reasoning and hence trying out seemed to be the only way to find out. The only reasoning in favor seemed to be that blocking will hopefully help us save a lot of analysis time in future and hence the time spent on creating a mesh now would justify it.

Thanks,
Shruti