Loading…
On flow-enhanced crystallization in fiber spinning: Asymptotically justified boundary conditions for numerics of a stiff viscoelastic two-phase model
For flow-enhanced crystallization in fiber spinning, the viscoelastic two-phase fiber models by Doufas et al. (2000) and Shrikhande et al. (2006) are state of the art. However, the boundary conditions associated to the onset of crystallization are still under discussion, as their choice might cause...
Saved in:
Published in: | Journal of non-Newtonian fluid mechanics 2021-10, Vol.296, p.104636, Article 104636 |
---|---|
Main Authors: | , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | For flow-enhanced crystallization in fiber spinning, the viscoelastic two-phase fiber models by Doufas et al. (2000) and Shrikhande et al. (2006) are state of the art. However, the boundary conditions associated to the onset of crystallization are still under discussion, as their choice might cause artificial boundary layers and numerical difficulties. In this paper we show that the model class of ordinary differential equations is singularly perturbed in a small parameter belonging to the semi-crystalline relaxation time and derive asymptotically justified boundary conditions. Their effect on the overall solution behavior is restricted to a small region near the onset of crystallization. But their impact on the performance of the numerical solvers is huge, since artificial layering, ambiguities and parameter tunings are avoided. The numerics becomes fast and robust and opens the field for simulation-based process design and material optimization.
•Asymptotic analysis of a viscoelastic two-phase fiber model for melt spinning.•Derivation of asymptotically justified boundary conditions.•The underlying ordinary differential equations become regularly perturbed.•Avoidance of artificial boundary layering, ambiguities and parameter tunings. |
---|---|
ISSN: | 0377-0257 1873-2631 |
DOI: | 10.1016/j.jnnfm.2021.104636 |