Remote point

[PedroCarneiro22]

Hi,

I am trying to impose a moment in my structure through a remote point.
I am testing a simple case of a C shape composed by beam188 elements.
If I impose a moment Mz=1 Nm in the remote point, Ansys induce a force FX in top and bottom nodes of 500 N.
Can anyone explain why the reason behind this? Why the magnitude of the load is so high comparing with the moment impose in the remote point?

Find in annex a simple image with the implemented structure and coordinate systems.

Regards

 

[NRP99]

Force and moment distribution to the slave nodes from the master node (remote point) depends on the type of behavior selected -rigid or deformable or constraint equation coefficients used. Rigid behavior distributes forces or moments equally to the slave nodes. Deformable behavior distributes based on the distance of master node from the slave nodes. (weighted residual). Basically it depends on RBE3 (deformable or in ANSYS APDL called CERIG) or RBE2 (Rigid) connection technique.

Check Ansys help documentation for how the forces and moments are distributed to slave nodes using remote point.

 

[PedroCarneiro22]

Thanks for the answer and clarifications.

I am using APDL commands to run my test.
I am creating the remote point using targe170 and the two nodes that will receive the loads I am using conta175.

I am using the following commands:

ET,2,targe170
keyopt,2,2,1
tshaple,pilo

et,3,conta175
keyopt,3,12,5
keyopt,3,2,2
keyopt,3,4,1

How this is related with RB3 or RB2?
I notice that if I change keyopt,3,4,1 to keyopt,3,4,0 (which according to Ansys Workbench represents the modification from RBE3 to CERIG (RBE2) the results change significantly.
Can you explain what is the difference?

 

[NRP99]

Remote point in the context of Ansys workbench (and in context of RBE3 or CERIG) is what I explained. Is there any specific reason to transfer the forces and moments using the TARGE170 and CONTA175? There is no relation of using TARGE170 and CONTA175 for transferring forces and moments to RBE3/RBE2 connection. Generally TARGE170 and CONTA175 are used as overlay on other elements to represent contact between point and surface which is I think automatically assigned once we define node to surface contact.

The preferred method to transfer the forces and moments is using the rigid coupling (RBE3 or CERIG) if the objective is to transfer the forces and moments on the slave nodes.

 

[PedroCarneiro22]

Hello again,

There are no specific reason for that. I just use it because it is the code write automatically by Ansys when I create a remote point and the scoping region in Ansys workbench.
What I did was write the input file from Ansys workbench and see the corresponding apdl code. And I am now using Ansys Mechanical.

My concern is manly related with the way Ansys is doing the loads and moments transfer, for example in the case I show above of the C structure where I apply a MZ of 1 N.m in the remote point one of the constraint equations created says that the ROTZ of remote point is equal to -500 UX of node 1 plus - 500 UX of node 2. How are these coefficients calculated?

Thanks for your time.

 

[NRP99]

Considering equal distribution of moment to two nodes then, Moment/(2*distance from master to slave)= Forces on nodes.

Constraint Equations are of type Constant=coefficients*DOF. Check help documentation for more information.

 

[PedroCarneiro22]

Ok, so considering that the applied moment is equal to 1 N.m and the distance between the master and the slaves is equal to 1 meters the applied force in the nodes shall be 0.5 N.
But when I plot the nodal force is not given the correct value. The same happens with the constraint equation.
Please find in annex the txt file with the code.
I will be very grateful you can look at it.

 

[PedroCarneiro22]

Code

 

[NRP99]

Preliminary check gives me that you are using bonded contact with pilot node applied moment. No constraint equations are modeled in your model. So constrain equation distribution doesn't apply. Here internally created multi-point constraint (MPC) equations for contact applies.

Check multi point constraint for node to surface contact in the Ansys contact technology guide for more information on how it works.

 

[PedroCarneiro22]

Thanks I will look at it.
Meanwhile, i am reading about RBE3. In this case, I need to have at least more two nodes because if i only select the nodes in the end of the beam188 elements it says that the slaves nodes are collinear and RBE3 command is ignored. This is right correct?

One more question is related with the fact that when you create a remote point and connect to two nodes in Ansys workbench the commands associated with this operation are the creation of targe170 and conta175 elements. Why doesn’t Ansys Workbench create RBE3 and create this elements?

 

[NRP99]

Its long time since I used Ansys so yes this is surprising to me now.

Indeed the remote point application uses the TARGE170 and CONTA175 elements in the Workbench. You can see the above two elements used in solution output window. So yes CERIG/RBE2 or RBE3 is not used in the Workbench for remote point application. These are specific to Ansys APDL I think. In workbench MPC constraint are created using above two elements for force and moment transfer.

I tested simple example -30mm ID and 50mm OD hollow circular surface and applying the 1000Nmm torsional moment at remote point at the centre scoped to 30mm dia edge with rigid behavior. Fixed the 50mm edge. I tried to see constraint equations by issuing CELIST command but there are no constraint equations displayed in post processor which proved Ansys is creating MPC's using above elements. Then I extracted the nodal forces using the ENFOVECTORS-user defined results at the 30mm dia edge. Exported to excel file and summed for each node and I got the exact same output sum of 1000Nmm. Each node has two force components, finding the resultant gives me range of forces on nodes=2.073N to 2.086N (I have no clue why there are different forces on each node at this stage).
Node count on inner edge=32
Moment applied = 1000Nmm
Radius = 15mm
Force on each node should be =Moment/(node*radius)= 1000Nmm/(32*15mm)=2.083N on each node. This proves behavior is kind of ok and approx matches with the RBE2/CERIG application in APDL.

Check your workbench model using the above method.

And yes for RBE3 command this is what help says-
"RBE3 creates constraint equations such that the motion of the master is the average of the slaves. For the rotations, a least-squares approach is used to define the "average rotation" at the master from the translations of the slaves. If the slave nodes are colinear, then one of the master rotations that is parallel to the colinear direction can not be determined in terms of the translations of the slave nodes. Therefore, the associated moment component on the master node in that direction can not be transmitted. When this case occurs, a warning message is issued and the constraint equations created by RBE3 are ignored."

Edit-You can insert constraint equations e.g. in static structural =insert constraint equation.

 

[PedroCarneiro22]

Thanks for the answer and support.
In fact, it was the first time I have tried to use elements to distribute loads.
I also become surprised why Ansys in workbench environment uses targe and conta elements but it seems the standard process.
In Ansys mechanical APDL i think also that RBE2/Cerig or RBE3 are the most appropriated choices.
One thing that happens, and for me at least it seems not ok, is that if you use the targe and conta elements in Ansys Mechanical APDL, similarly to approach used in Ansys workbench, you can have collinear slaves nodes. However the load distribution is somehow strange… maybe I am missing something on this topic.

 

[NRP99]

MPC constraint uses shape functions to distribute the forces and moments. Check contact technology guide for more details. This is what happening in the MPC constraint force moment transfer. And yes MPC is standard way of force and moment transfer in workbench and this is ok to me by doing simple exercise like in my earlier post. In APDL we have choice of going to RBE2/CERIG and RBE3. You can insert constraint equation in workbench to achieve RBE2/CERIG and RBE3 like behavior or however behavior you want.