The OpenSees Community

Thanks again for pointing out the bugs.
The bug related to the geometric stiffness calculation has been fixed.
The joint offsets are still not being considered yet. I will fix this at a later time.

Please update your source code using SVN until Frank posts a new release.

Thanks a lot for finding these bugs.
I will have a look at this next weekend and will fix the source code.
Best,
Andreas

I just checked and the input field for E seems to work fine.
It is the second input field called "Modulus of Elasticity (E) :"

Thanks for finding this bug. We have posted a new version 2.5.7 (64-bit) that should fix the problem. Please download the new OSN version and make sure to use the correct MCR (2016a) with it. Let us know if you discover any further issues.

The maximum scale factor that will still work for this example is ~1.7.

The reason is that at a scale factor of 1.7 you already get a horizontal displacement of 15.7 in. Given that the radius of the sliding surface for this example is 34.68 in. you are already nearly on the vertical portion of the sliding surface (to illustrate this I suggest you draw a circle with radius of 34.68 in. and then see where you are on the circle at a horizontal displacement of 15.7 in.).

The other issue is that the vertical motion is getting scaled by the same factor and you start having a lot of uplift. At a scale factor of 1.7 you will get nearly 2 in. of uplift. The impact following this uplift will cause numerical convergence issues. I recommend assigning a vertical material with some small amount of viscous damping to help with convergence. I updated the example files on the Wiki to illustrate how to do this. So please look at the updated examples.

Please update to the latest version, since I have fixed a few bugs.

First you can use the Assign->Element->Local Axes to define the local x-axis of the zeroLength element.
Second you can use the Assign->Element->Rotations to rotate the element around the local x-axis which will allow you to define the local y-axis as up.

Finally, the Edit->Element->Add/Delete ZeroLength Element option allows you to automatically add or delete zeroLength elements at the ends (nodeI or nodeJ) of frame elements, such as beamColumn elements. It will not do anything if you try to use this on other types of elements such as zeroLength elements, quads, or bricks.

There has been a bug in version 2.5.5 which has been fixed in build 514. The new 64-bit version has been uploaded (32-bit version will follow shortly). Please download the latest build and try again. I tested it out using the basic truss example and the 2d steel frame example and everything should work fine now. I also uploaded the two examples files that correspond to the videos here: http://openseesnavigator.berkeley.edu/d ... on/videos/
In addition, I put all the OpenSees primer examples on the website which might be helpful to some of you: http://openseesnavigator.berkeley.edu/d ... /examples/

Please let me know if you encounter any other issues.

First of all your periods seems very small. The first mode period is 0.0135 seconds. Please check your mass assignments and make sure they are correct.

Secondly, you are simulating the experimental beam column element using the SimUniaxialMaterials control in the shear direction. This means that the experimental column at the tip where you attach the actuator provides shear resistance but no moment resistance (you are not controlling the rotational DOF so it will return zero moment). So your column acts as a cantilever column. I am assuming that this was not your intention?

If you like to simulate both, the moment and shear resistance, please use the SimDomain experimental control. Please look at the LShapedColumn example in OpenFresco to see how to use the SimDomain exp control.

Finally, I like to suggest that you do not divide by integers when calculating parameters that should be floating point numbers. This is an easy source of errors because Tcl will return integer values unless one of the values in the formula is a float.

The initial stiffness matrix for the experimental beam column element in the cantilever basic system (see OpenFresco command language manual on the OpenFresco website) is given by:
kinit = [EA/L, 0, 0; 0, 12EI/L^3, -6EI/L^2; 0, -6EI/L^2, 4EI/L]

The columns in the PortalFrame example have a length of L = 50 in. Now when we simulate the columns which behave like cantilevers that are fixed at the top and free to rotate at the bottom using a uniaxial material the stiffness of the cantilever is given by 3EI/L^3 which we determined from tests to be 2.8 kip/in. This is why the uniaxial material has its stiffness set to 2.8 kip/in.

Given the above we can calculate EI = 2.8 L^3/3 = 116'667. From that we get 12EI/L^3 = 11.2, -6EI/L^2 = 280 and 4EI/L = 9333

It looks like you are using the new consistent mass matrix option for the truss element. Please make sure that you use the latest version of OpenSees 2.4.5 because the consistent mass option is not available in earlier versions.

You can use any set of units in OpenSees Navigator, as long as you are consistent with defining ALL your properties in the units you selected. It is the same as with OpenSees.

This command should work fine in the latest OpenSees v2.4.5.

Please try to run one of the examples that are on the OpenSees Wiki which use the VelNormalFrcDep friction model.

Several examples can be found here:
http://opensees.berkeley.edu/wiki/index ... um_Element

Try this example first:
http://opensees.berkeley.edu/wiki/image ... iction.tcl

Are you using the self-executable or the p-code of OpenSeesNavigator when this error occurs? What versions of OpenSees Navigator and OpenSees are you using?
The latest OpenSeesNavigator v2.5.3 should work fine with the latest OpenSees v2.4.5.
Please provide some more information so that I can try to help.

Are you using the self-executable or the p-code of OpenSeesNavigator when this error occurs? What versions of OpenSees Navigator and OpenSees are you using?
The latest OpenSeesNavigator v2.5.3 should work fine with the latest OpenSees v2.4.5.
Please provide some more information so that I can try to help.

Did you apply gravity load to your model? If there is no axial load on the element there will be no shear resistance and the element will not converge.