[MBDyn-users] Consistency in the initial assembly

MBDyn users list mbdyn-users at mbdyn.org
Sun Jan 31 16:44:24 CET 2021


Could you please post the entire model?  Otherwise, it is nearly 
impossible to investigate your issue.

Sincerely, p.

Il 31/12/20 10:04, MBDyn users list ha scritto:
> Dear MBDyn users,
>
> I am a little confused with the consistency in the initial assembly.
>
> I used a clamp joint, an axial rotation joint and some beam elements to simulate a blade rotating around axis-x with a constant speed omega_init = 1.2671.
> The nodes, clamp and axial rotation joint are set as follows:
>
> structural: 1, dynamic,
> 	position, reference, global, 0.0, 0.0, 0.0,
> 	orientation, reference, global, eye,
> 	velocity, reference, global, null,
> 	angular velocity, reference, global, null;
>
> structural: 11, dynamic,
> 	position, reference, global, 0.0, 0.0, 0.0,
> 	orientation, reference, global, eye,
> 	velocity, reference, global, null,
> 	angular velocity, reference, global, null;
>
>
> structural: 101, dynamic,
> 	position, reference, global,
> 	0.00000e+00,0.00000e+00,1.00e+00,
> 	orientation, reference, global,
> 		1, 1., 0., 0., 3, 0., 0., 1.,
> 	velocity, reference, global, null,
> 	angular velocity, reference, global, null;
>
> structural: 1001, dynamic,
> 	position, reference, global,
> 	0.00000e+00,0.00000e+00,2.00e+00,
> 	orientation, reference, global,
> 		1, 1., 0., 0., 3, 0., 0., 1.,
> 	velocity, reference, global, null,
> 	angular velocity, reference, global, null;
>
>
>
> joint: 1, clamp,
> 	1, node, node;
>
> joint: 11, axial rotation,
> 	1, position, reference, node, null,
> 	orientation, reference, global, 3, 1.,0.,0.,2, 0., 1., 0.,
> 	11, position, reference, node, null,
> 	orientation, reference, global, 3, 1.,0.,0.,2, 0., 1., 0.,
> 	const, omega_init;
>
>
> beam: 100001,
> 11, reference, node, null,
> 101, reference, node, null,
> 1001, reference, node, null,
> ...
>
>
> And the residual of the axial rotation joint in initial assembly is as follows:
> Eq      1261:                    0 AxialRotationJoint(11): position constraint Px
> Eq      1262:                    0 AxialRotationJoint(11): position constraint Py
> Eq      1263:                    0 AxialRotationJoint(11): position constraint Pz
> Eq      1264:                    0 AxialRotationJoint(11): orientation constraint gx
> Eq      1265:                    0 AxialRotationJoint(11): orientation constraint gy
> Eq      1266:                    0 AxialRotationJoint(11): angular velocity constraint wz
> Eq      1267:                    0 AxialRotationJoint(11): position constraint derivative vx
> Eq      1268:                    0 AxialRotationJoint(11): position constraint derivative vy
> Eq      1269:                    0 AxialRotationJoint(11): position constraint derivative vz
> Eq      1270:                    0 AxialRotationJoint(11): orientation constraint derivative wx
> Eq      1271:               1.2671 AxialRotationJoint(11): orientation constraint derivative wy
>
> Then it iterates for once and terminated normally and the output of node 11 is as follows:
> 11 0.000000e+00 0.000000e+00 0.000000e+00 -0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 1.267100e+00 0.000000e+00 0.000000e+00
> 11 0.000000e+00 0.000000e+00 0.000000e+00 7.259948e-03 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 1.267100e+00 -6.148059e-14 -3.870283e-18
>
> First, is this residual due to the velocity of nodes are not consistent with the joint? I'm not sure why the residual 1.2671 appears in the orientation constraint derivative wy rather than wx = 1.2671, either. I noticed the residual of the axial rotation joint in the initial assembly is zero when I set the initial angular velocity of the joint zero and make it increasing gradually.

It seems that you initialize your nodes with zero velocity and angular 
velocity with respect to the global reference frame, but then prescribe, 
through the axial rotation joint, that a non-zero angular velocity is 
present between nodes 1 and 11.  Yes, this would make the model 
inconsistent (initial conditions violate the constraints).  The solver 
will try to recover a compatible initial solution, which will be a 
trade-off between remaining as close as possible to the given initial 
conditions while making it comply with the constraints (at both position 
and velocity level).


>
> Do I still need to add the same angular velocity on the structural nodes even if it can run normally in the first situation?
>
> Does the angular velocity of a node in the input file only represent its initial velocity while the angular velocity in an axial rotation joint represents a permanent velocity. But the torque of the joint can change with time to keep the velocity constant, right?

Sorry, this statement makes little sense to me.  Initial velocity is 
(part of) the initial conditions in a Cauchy problem.  It has nothing to 
do with torque, with time or so.

Sincerely, p.

>
> Best regards,
>
> Jun
>
>
>
> _______________________________________________
> MBDyn-users mailing list
> MBDyn-users at mbdyn.org
> https://mail.mbdyn.org/cgi-bin/mailman/listinfo/mbdyn-users


-- 
Pierangelo Masarati
Professore Ordinario di Costruzioni e Strutture Aerospaziali
Dipartimento di Scienze e Tecnologie Aerospaziali
Politecnico di Milano



More information about the MBDyn-users mailing list