Spurious deformation of B32 and B32R ends with *CLOAD or *BOUNDARY.

[victorkemp]

With a point load at the end of a quadratic beam, there's spurious deformation at the end.

*NODE
1,0,0,0
2,0.0025,0,0
3,0.001666666666667,0,0
4,0.003333333333333,0,0
5,0.005,0,0
6,0.004166666666667,0,0
7,0.0008333333333333,0,0
*ELEMENT,TYPE=B32R,ELSET=Default
1,4,6,5
2,1,7,3
3,3,2,4
*MATERIAL,NAME=material
*ELASTIC,TYPE=ISOTROPIC
100000000000,0
*BEAM SECTION,ELSET=Default,MATERIAL=material,SECTION=RECT
0.001,0.001
0,1,0
*BOUNDARY
1,1,6,0
*STEP
*STATIC
*CLOAD
5,1,20
*NODE FILE
U
*EL FILE
S
*END STEP

With a nonhomogeneous boundary condition, there is also spurious deformation as well as wrong stress which propagates through the whole model.

*NODE
1,0,0,0
2,0,0,0.1
3,0,0,0.2
4,0,0,0.15
5,0,0,0.05
6,0,0,0.075
7,0,0,0.025
8,0,0,0.175
9,0,0,0.125
*ELEMENT,TYPE=B32R,ELSET=Default
1,3,8,4
2,4,9,2
3,2,6,5
4,5,7,1
*MATERIAL,NAME=Material
*ELASTIC,TYPE=ISOTROPIC
10000000,0
*BEAM SECTION,ELSET=Default,MATERIAL=Material,SECTION=RECT
0.1,0.1
0,1,0
*BOUNDARY
1,1,6,0
*STEP
*STATIC
*BOUNDARY
3,3,,-0.003
*NODE FILE
U
*EL FILE
S
*END STEP

It looks like this is caused by CCX using only the 4 corner nodes of beams ends without knots according to the manual equation (753)

u1 + u2 + u3 + u4 − 4u0 = 0

I think it would be better to use all 8 nodes on quadratic beam ends or make a knot there.

[lucas-bueno777]

The bug persists in v2.21.