Tuesday, May 1, 2012

Best Practices for Stent Analysis with Abaqus (1)

Element Selection:


As stent is loaded dominantly in bending,
 the reduced-integration (C3D8R) and 
incompatible mode element (C3D8I) are more appropriate for the stent analysis. For C3D8R, it is computational inexpensive as only one integration point per element. However, due to hourglassing issue, at least four elements through the thickness is required to mitigate the potential problems. There is no particular limitation for C3D8I. But this type of element is more expensive than C3D8. Since t
he maximum stress/strain of a stent usually lies on the stent surface due to bending-dominated loading conditions and material calculations are carried out at the integration points, it is better to overlay membrane elements on the free surface of a reduced-integration element model. This will add integration points on the surface and reduce the error caused by extrapolation. As pointed by a study conducted at Dassualt Systemes, 

"models with C3D8R elements alone (i.e., without overlaid membranes) underestimate the peak maximum principal stress unless a very fine mesh is used, which could result in overestimation of the fatigue safety factor and fatigue life. With membrane overlay, the peak stress is overestimated with the coarsest mesh studied. The overestimation is reduced significantly with a finer mesh"

Reference: Abaqus Answer 4516

Wednesday, June 29, 2011

Duplicate node output when using fieldoutput

When using fieldoutput, if a node is used more than once (e.g. a node contacts with more than one master surface), multiple field output will be created for the same node. It will cause that created XY data has a duplicate X value and fail any XY data operation with different XY data.

Thursday, May 26, 2011

Stent design parameters

1. Flexibility: Apply angular displacement on both ends and measure the reaction moment. The unit for this study is Moment(Nmm) vs. Angle/length (rad/mm)
2. Radial recoil: (R_load-R_unload)/R_load
3. Foreshortening: (L-L_load)/L
4. Radial Strength: force per unit of length when stent suddenly collapse
5. Fatigue Safety Factor (FSF).The FSF was estimated by using the modified Goodman relationship (Shigley, 2001)

FSF ≡ Fatigue Safety Factor response
• Sa ≡ Alternating Stress response
• Sm ≡ Mean Stress response
• Se ≡ Stent Material’s Endurance Limit (10 year fatigue life equivalent of blood pressure
loading)
• Suts ≡ Stent Material’s Ultimate Tensile Stress (defined in terms of true stress)

Reference: Baillargeon et al SCC 2011

Stent expansion methods

Three approaches:
(i) ignoring the balloon and applying an increasing uniform pressure directly on the stent inner surface.
(ii) accounting for balloon-stent interaction and enforcing a radial
displacement-driven process on a cylindrical balloon (using rigid body surface RSURFU).
(iii) accounting for balloon-stent interaction and applying an increasing uniform pressure inside of balloon (using hydrostatic element or surface).

Friday, May 20, 2011

Using membrane elements to better capture surface strains

There has always been some concern with finite element analyses using solid elements that the most exact characterization of field quantities occurs at the integration points, which are located in the interior of the part being modeled, whereas often (particularly where significant bending or torsion is involved) the largest strains occur at the surface, which are traditionally available only by extrapolation from the integration points to the nodal points using the element’s shape functions. The M3D4R membrane elements in this study is to “trick” the solver into providing surface values that are more exact and are not interpolated. (DeHerrera 2009)