Section 12
K6ROT
What’s New in MSC.Nastran 2004, July 2003
S-1
Defaults for SNORM and K6ROT in V2001
Shell normals provide higher accuracy
PARAM, SNORM, 20., is default in V2001
PARAM, K6ROT, 0. is the default for all linear solution
sequences in V2001
reduces a-size because drilling degree of freedom is constrained
by PARAM, AUTOSPC, YES
PARAM, K6ROT, 100. is the default in nonlinear solution
sequences.
What’s New in MSC.Nastran 2004, July 2003
S-2
Defaults for SNORM and K6ROT in V2001
(cont.)
The default combination in V2001 may cause problems in
cases where forces are transferred through the drilling
degree-of-freedom.
What’s New in MSC.Nastran 2004, July 2003
S-3
Effect of K6ROT – Flagpole Example
Intentional loading of the drilling degree-of-freedom in shells
may be discredited as bad modeling practice.
SNORM K6ROT No. of First Nonzero Mode Comment
Zero Modes
0. 0. 6 1.2842e+6 Default in V70.7, no
spurious mode, good first
flexible mode
20. 0. 7 2.4671e+6 Default in V2001, spurious
mode, bad first flexible
mode
20. 100. 6 1.2745e+6 Future default, no spurious,
good first flexible mode
What’s New in MSC.Nastran 2004, July 2003
S-4
Effect of K6ROT – Ring with Stiffener
Example
Unintentional loading of the drilling degree-of-freedom occurs in
some cases, for example, in stiffened shells modeled with
beams and offsets.
Z 10
R= 246
2
246
n
Θ 2
246
246
246
246
Ring Modeled with CQUAD4
Stiffener modeled with CBARS and offset
No I2 specified on PBAR
Offset direction and normal direction differ slightly
What’s New in MSC.Nastran 2004, July 2003
S-5
Effect of K6ROT – Ring with Stiffener
Example (cont.)
K6ROT SNORM Offset Disturbance Comments
1 rigid body mode
No spurious modes
0. 20. ε ≤ 1.8E − 4 AUTOSPC catches
singularity
2nd mode 9.667
1 rigid body mode
20.
ε ≥ 1.9 E − 4 5 spurious modes
0.
7th mode 9.734
1 rigid body mode
100. 20. ε ≥ 1.9 E − 4 No spurious modes
2nd mode 9.655
With PARAM, K6ROT, 100., we always get 6 rigid body modes for a
free system, here we get only 1 rigid body mode due to symmetry
conditions
What’s New in MSC.Nastran 2004, July 2003
S-6
Why is K6ROT= 100. a Good Value ?
The Membrane Problem is the test in which K6ROT has the
strongest stiffening effect
SPC =
123456
0.5
0.5
T= 1.e-4
Y
X 1.0
What’s New in MSC.Nastran 2004, July 2003
S-7
Why is K6ROT= 100. a Good Value ? (cont.)
Fine Meshes
1.0100 64x64
Normalized Tip Displacement
1.0000
32x32
0.9900
0.9800 16x16
0.9700
0.9600
0.9500
8x8
0.9400
0.9300
4x4
0.9200
0.9100
0. 10. 100. 1000. 1e +4 1e +5 1e +6
K6ROT
What’s New in MSC.Nastran 2004, July 2003
S-8
Why is K6ROT= 100. a Good Value ? (cont.)
Coarse Meshes
1.0000
4x4
Normalized Tip Displacement
0.9000
0.8000
2x2
0.7000
0.6000
0.5000
1
0.4000
0.3000
0.2000
0.1000
0. 1.e+1 1.e+2 1.e+3 1.e +4 1.e +5 1.e +6
K6ROT
What’s New in MSC.Nastran 2004, July 2003
S-9
K6ROT With and Without SNORM
Without shell normals-- HOOK
answer diverges with mesh
R = 14
refinement 60ο
t=2
With shell normals—answer 20
converges with mesh R = 46
refinement
150
Z 1N
What’s New in MSC.Nastran 2004, July 2003
S-10
K6ROT With and Without SNORM
•Shell normals reduce impact of K6ROT in Curved Shells
Effect of K6ROT in the Hook Example
Normalized Tip Displacement
5.0
4.0 QUAD4 Without Normals
3.0
2.0
QUAD4 With Normals
1.0
0.0
0 1.0 1.e+2 1.e+4 1.e+6
K6ROT
20 x 136 Mesh
MID3=1 with transverse shear flexibility
What’s New in MSC.Nastran 2004, July 2003
S-11
New Default for K6ROT
The membrane test shows that K6ROT=100. is the best
choice.
The MacNeal-Harder standard test problems and other
test problems confirm that K6ROT=100. does not
influence the results
stiffening effect is negligible as long as shell normals are used (PARAM,
SNORM, 20.)
When shell normals are turned off (PARAM, SNORM, 0.),
K6ROT has a significant stiffening effect in curved shell
type problems.
What’s New in MSC.Nastran 2004, July 2003
S-12
New Default for K6ROT (cont.)
K6ROT=100. has always been the default in nonlinear
SOLs 106 and 129
K6ROT=100. corresponds to stiffness values used by
other finite element codes
If V2001 default of K6ROT=0. is desired
set parameter defaults in the RC file
Defaults in v2004+ for all solution sequences are:
SNORM=20.
K6ROT=100.
PARAM, K6ROT, 100.
PARAM, SNORM, 20.
What’s New in MSC.Nastran 2004, July 2003
S-13
Remark for New K6ROT Default
For Plane Stress Problems
For plane stress problems (membrane only),
z it is common practice to constrain the 6th dof
z SPC 3456 if plane stress is in the x-y plane
No problem if K6ROT=0.
The structure is grounded if the 6th dof is constrained and K6ROT=100.
Do not constrain DOF 6 if K6ROT= 100. or set K6ROT=0. for plane stress
problems
It may take a bit longer to run for some problems due to
increase in the aset
What’s New in MSC.Nastran 2004, July 2003
S-14
