Learning Forum

[Eleonora Vezzani]

Hello!

I have a geometry model of a structure made of beam elements(BEAM181) with few lumped masses (MASS21) and 1D springs-damper(COMBIN14) acting on different DOF with both linear and nonlinear damping. I ran this geometry first in a modal analysis encountering no errors. Now I am trying with the transient analysis that would be final goal for the research but it gave the following errorin the at LSSOLVE command: node I and J for element [COMBIN14 element number] are not coincident. I tried to fix it by putting the second node of the element at the same location but I still have problems, and I did not find any clear requirement in the Element Reference that the nodes have to be coincident. Do they have to be? Does anyone know how to fix this code?

Here are the MAPDL lines where I define this elements:

lumped mass element -> floaters (3d with rotary inertia + coordinate system  parallel to node)
et,2,mass21,0,0,0

!elements -> damping & stiffness of floaters
et,3,combin14,,1 !spring-damper ux (mooring stiffness)
et,5,combin14,,2 !spring-damper uy (mooring stiffness)
et,4,combin14,,3 !spring-damper uz (mooring stiffness+ heave c33)
et,6,combin14,,4 !spring-damper rotx (pitch c44)
et,7,combin14,,5 !spring-damper roty (roll c55)

!connectors
et,9,combin14,1,1
et,10,combin14,1,2
et,11,combin14,1,3
et,12,combin14,1,4
et,13,combin14,1,5
et,14,combin14,1,6

!! define nodes for lumped masses
    !floaters masses COG
    !left down
    n,1,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),h_fl_up/2
    n,2,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_down/2
    n,3,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_down+T_fl/2 !added mass
    !right down
    n,4,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),h_fl_up/2
    n,5,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_down/2
    n,6,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_down+T_fl/2 !added mass
    !centre up
    n,7,L_tot/2,(L_t-L_obl)*sin(pi/3),h_fl_up/2
    n,8,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_down/2
    n,9,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_down+T_fl/2 !added mass
    
    
    !! define nodes for hd spring-damper second nodes
    !left down
    n,10,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+T_fl/12
    n,11,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+2*T_fl/12
    n,12,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+3*T_fl/12
    n,13,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+4*T_fl/12
    n,14,-L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+5*T_fl/12
    !right down
    n,15,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+T_fl/12
    n,16,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+2*T_fl/12
    n,17,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+3*T_fl/12
    n,18,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+4*T_fl/12
    n,19,L_tot+L_obl/2*cos(pi/3),L_obl/2*sin(pi/3),-h_fl_tot+5*T_fl/12
    !centre up
    n,20,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_tot+T_fl/12
    n,21,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_tot+2*T_fl/12
    n,22,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_tot+3*T_fl/12
    n,23,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_tot+4*T_fl/12
    n,24,L_tot/2,(L_t-L_obl)*sin(pi/3),-h_fl_tot+5*T_fl/12
    
    !define nodes for mass connectors
    n,25,-L_obl/2*cos(pi/3)-dist_conn*cos(angle_conn),L_obl/2*sin(pi/3)+dist_conn*sin(angle_conn),h_fl_up/2
    n,26,L_tot+L_obl/2*cos(pi/3)+dist_conn*cos(angle_conn),L_obl/2*sin(pi/3)+dist_conn*sin(angle_conn),h_fl_up/2
    n,27,L_tot/2-dist_conn*cos(angle_conn),(L_t-L_obl)*sin(pi/3)+dist_conn*sin(angle_conn),h_fl_up/2
    n,28,L_tot/2+dist_conn*cos(angle_conn),(L_t-L_obl)*sin(pi/3)+dist_conn*sin(angle_conn),h_fl_up/2
    
    !second nodes spings mass connector
    n,29,-L_obl+L_obl/2*cos(a),L_obl/2*sin(a),0
    n,30,L_tot+L_obl-L_obl/2*cos(a),L_obl/2*sin(2*a),0
    n,31,L_tot*cos(a)-L_obl/2*cos(a),L_obl*sin(2*a)+L_tot*sin(a)+L_obl/2*sin(a),0
    n,32,L_tot-L_tot*cos(a)+L_obl/2*cos(a),L_obl*sin(2*a)+L_tot*sin(a)+L_obl/2*sin(a),0

!lumped masses for floaters
type,2 
!up
real,1
en,1,1
en,2,4
en,3,7
r,1,m_fl_up,m_fl_up,m_fl_up,I_fl_xx_up,I_fl_yy_up,I_fl_zz_up

!down
real,2
en,4,2
en,5,5
en,6,8
r,2,m_fl_down,m_fl_down,m_fl_down,I_fl_xx_down,I_fl_yy_down,I_fl_zz_down

!added mass
real,3
en,7,3
en,8,6
en,9,9
r,3,m_xx,m_yy,m_zz,I_xx,I_yy,I_zz 

!spring-damper element for floaters
type,3 !spring-damper ux (mooring stiffness)
real,4
en,10,3,10
en,11,6,15
en,12,9,20
r,4,c_x,0

type,4 !spring-damper uy (mooring stiffness)
real,5
en,13,3,11
en,14,6,16
en,15,9,21
r,5,c_y,0

type,5 !spring-damper uz (mooring stiffness+ heave c33)
real,6
en,16,3,12
en,17,6,17
en,18,9,22
r,6,c_z+c_33,0

type,6 !spring-damper rotx (pitch c44)
real,7
en,19,3,13
en,20,6,18
en,21,9,23
r,7,c_44,0

type,7 !spring-damper roty (roll c55)
real,8
en,22,3,14
en,23,6,19
en,24,9,24
r,8,c_55,0

!lumped masses connectors
type,2
real,9
en,25,25
en,26,26
en,27,27
en,28,28
r,9,m_tot_ccs/2

!springs connectors
!ux
type,9
real,10
en,29,25,29
en,30,26,30
en,31,27,31
en,32,28,32
r,10,k_conn,b1_conn,b2_conn

!uy
type,10
real,11
en,33,25,29
en,34,26,30
en,35,27,31
en,36,28,32
r,11,k_conn,b1_conn,b2_conn

!uz
type,11
real,12
en,37,25,29
en,38,26,30
en,39,27,31
en,40,28,32
r,12,k_conn,b1_conn,b2_conn

!rotx
type,12
real,13
en,41,25,29
en,42,26,30
en,43,27,31
en,44,28,32
r,13,k_conn,b1_conn,b2_conn

!roty
type,13
real,14
en,45,25,29
en,46,26,30
en,47,27,31
en,48,28,32
r,14,k_conn,b1_conn,b2_conn

!rotz
type,14
real,15
en,49,25,29
en,50,26,30
en,51,27,31
en,52,28,32
r,15,k_conn,b1_conn,b2_conn

!----
!linking floaters lumped masses to structure
!---
!left down
ksel,s,kp,,1
ksel,a,kp,,23
ksel,a,kp,,11
ksel,a,kp,,53
nslk,s
rbe3,1,all,all
rbe3,2,all,all
rbe3,3,all,all

!right down
ksel,s,kp,,5
ksel,a,kp,,14
ksel,a,kp,,6
ksel,a,kp,,37
nslk,s
rbe3,4,all,all
rbe3,5,all,all
rbe3,6,all,all

!centre up
ksel,s,kp,,18,19,1
ksel,a,kp,,50,51,1
nslk,s
rbe3,7,all,all
rbe3,8,all,all
rbe3,9,all,all

!----
!linking connectors lumped masses to structure
!---

!left down
ksel,s,kp,,23
ksel,a,kp,,33
ksel,a,kp,,53
nslk,s
rbe3,25,all,all

!right down
ksel,s,kp,,14
ksel,a,kp,,36
ksel,a,kp,,37
nslk,s
rbe3,26,all,all

!centre up left
ksel,s,kp,,18
ksel,a,kp,,51,52,1
nslk,s
rbe3,27,all,all

!centre up right
ksel,s,kp,,19
ksel,a,kp,,50,52,2
nslk,s
rbe3,28,all,all

!----------
!constraints springs for floaters hydro properties
!----------

!constraints on second node
nsel,s,node,,10,20,5
d,all,ux,0

nsel,s,node,,11,21,5
d,all,uy,0

nsel,s,node,,12,22,5
d,all,uz,0

nsel,s,node,,13,23,5
d,all,rotx,0

nsel,s,node,,14,24,5
d,all,roty,0

!----------
!constraints springs for connectors
!----------

!constraints on second node
nsel,s,node,,25,28,1
d,all,ux,0
d,all,uy,0
d,all,uz,0
d,all,rotx,0
d,all,roty,0
d,all,rotz,0

[dlooman]

It shouldn't be an error if the nodes aren't coincident.  The COMBI14 Assumptions and Restrictions has the bullet item:

• For noncoincident nodes and KEYOPT(2) = 1, 2, or 3, no moment effects are included. That is, if the nodes are offset from the line of action, moment equilibrium may not be satisfied.  

You are using somewhat archaic commands, EN and LSSOLVE.  What version are you running?  RBE3 (force distribution constraint equations) is an unusual way to connect nodes.  CERIG would be a fixed connection.

[Eleonora Vezzani]

Hi Dave!

Thank you for the reply! I am actually running the latest version my uni has the licence for (ANSYS APDL 2023R2), but to build this code I referred to old documentation. What are more recent commands for EN and LSSOLVE? I was using RBE3 to connect the lumped masses to the main structure. These point masses are located in the cog of other structures that I am not including in the model, but I need to represent their weight in the model. Would CERIG serve this purpose?

Thank you again!

[dlooman]

[Eleonora Vezzani]

Hi Dave!

Thanks for the help, but your last reply to my message appears to me as a blank message. Can you please reply again or edit the previous message?

Thank you again!

[dlooman]

Sorry, I tried to reply from my phone, but apparently that's not possible.  What I said was that the E command is much more common than EN.  The element number isn't specified with the E command.  Just issuing solve instead of lswrite and then lssolve is recommended.  RBE3 is a good way to attach a mass to the model.  It's thought of as a force (mass) distribution construct and that's perfect for your application.  I was skimming your input and thrown off by your description of RBE3 as linking nodes.  Because of your use of older commands I was thinking possibly the error you are getting is due to using an old version.  What is the current error message?