Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Counterweight and Cable Model with singular stiffness matrix issues

Status
Not open for further replies.

cbat

Structural
Aug 20, 2012
6
Hi guys,

im using strand7 non linear transient analysis to examine the effect of a dynamic force on a counterweight. This countweight is suspended from a beam that is being impacted. I have modelled the cables using beam elements and selecting the beam element to act as a cable. The counterweight is modelled from plate elements. I have given the beam a free length equal to the actual length to try and start the analysis with the cable at full stiffness and in tension. However I get an error that says that the global stiffness matrix is singular at the node where the cable and plate meet. It says "Global Stiffness Matrix Singular at Equation 316: Pivot = -7.339492E-22 (Node 180 RX)".

Any ideas why this is or what I can do to fix this? I have made sure that all the beams and plates have the correct thickness, density and material properties.

Any help would be appreciated

Cheers
 
Replies continue below

Recommended for you

It has been quite a while since I did a transient analysis on Strand7, and due to a change of employment circumstances I no longer have access to the software.

That message you quote is quite clear: node 160 is unrestrained in the X translational direction.

Did the linear static solver give you a sensible solution?
Did the natural frequency solver give you sensible natural frequencies?

Maybe the cable element is "slack" at the start of the transient solution, leading to that singular global stiffness matrix. If so, consider using the results from a linear static analysis (or a non-linear static analysis) as the starting stress state for the transient solution.

You seem to imply, without categorically stating it, that the cable is hanging vertically. If is expected to continue to hang vertically right through the analysis, then you could use a beam element to model it rather than a cable element. That might remove some problems. (The beam's bending stiffnesses could be reduced to close to zero, if that helped.)
 
i don't know Strand&, but "global stiffness" suggests rigid body motion. have you restrained all 6 dof ?
 
RX suggests a rotational dof to me, perhpas your beam terminates in an ugly fashion at the counterweight.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Greg is right. In Strand7 terminology RX is a rotational DOF. DX would be a translational DOF.
My apologies for my misleading post: rust on brain.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor