Sumo Workshop

Find out what Sumo has to offer!

SUMO WORKSHOP TRAINING MANUAL

TABLE OF CONTENTS

Page #

1. Introduction ......................................................................................................................................... 1

2. Purpose of this course ......................................................................................................................... 1

3. Methodology ........................................................................................................................................ 1

4. Answers................................................................................................................................................ 1

4.1 No supports ................................................................................................................................ 1

4.2 No rotational supports............................................................................................................... 1

4.3 Unstable structure ..................................................................................................................... 1

4.4 Pinned column, pinned support ............................................................................................... 1

4.5 No deflection in beam ................................................................................................................ 2

4.6 Too many pins............................................................................................................................ 2

4.7 Unconnected models .................................................................................................................. 2

4.8 Node is not connected to the model .......................................................................................... 2

4.9 Buckling analysis manual calculations .................................................................................... 3

4.10 Second-order unsuccessful........................................................................................................ 3

4.11 Area support lifting ................................................................................................................... 3

4.12 Drop objects tool to modify geometry...................................................................................... 3

4.13 Meshing problems ..................................................................................................................... 4

4.14 Monolithic beam results ............................................................................................................ 4

4.15 Convergence of shear in shells.................................................................................................. 4

5. Class modelling example..................................................................................................................... 4

Sumo Workshop Training Manual

1. Introduction The Sumo Workshop is aimed at deepening your understanding in the use of Sumo. Although there is a default workflow to this course, your own examples are encouraged.

Note: The original Sumo training is a prerequisite for this course.

2. Purpose of this course The support staff and trainers at PROKON have worked together to compile a list of topics that users regularly ask about. This course has been compiled to address most of these regular issues that users face. Sumo is PROKON’s flagship analysis program a nd contains a lot of functionality. Although the Sumo training covers Sumo’s functionality, some tools need more attention than other . This course doesn’t address anything new. Rather, we aim to strengthen your knowledge and application within Sumo. 3. Methodology The course is set up so that there is active participation by attendees. A list of predefined Sumo models will be used to point out certain topics that must be addressed. In some cases, the analysis isn’t successful until errors are addressed. In other cases, the tools in Sumo are used to show important information about the analysis.

Although this document contains the answers to all the provided Sumo models, please attempt to solve the problems first.

4. Answers

4.1 No supports Single column with loading, no support.

A structure must have supports before the analysis is attempted.

4.2 No rotational supports Single column with loading and Pinned support.

A model must have supports to restrain all 6 degrees of freedom, even if no results are expected for a certain support condition.

4.3 Unstable structure Beam pinned at both ends, columns have pinned supports.

The beam element’s end conditions have torsional fixities, and there is no rotational support about the X-axis. The structure is numerically unstable.

4.4 Pinned column, pinned support Column start fixity is pinned; connected to pinned support.

A pinned support in conjunction with a pinned end condition causes a numerical instability. There are, essentially, two pins connected to each other.

Sumo Workshop Training Manual

Page 1

4.5 No deflection in beam Single beam loaded with UDL – No deflection.

All calculations are done at node points. If a beam element doesn’t have any internal nodes, no calculations can be done within the span of the beam. Therefore, no deflections can be calculated along the length of the beam.

Solution: define more segments for the beam.

4.6 Too many pins Top chords and vertical members are all pinned at the apex.

4.7 Unconnected models The column is 1 mm too short.

One of the unique properties of a stiffness matrix in both matrix structural analysis and the finite element method is that the entries on the main diagonal are positive and non- zero, or positive definite. A ‘Stiffness matrix not positive definite’ error is caused when an analysis encounters a negative or zero number on the diagonal of the stiffness matrix.

This is usually caused by one of the following:

• Instability, e.g. neglecting to restrain the rotational degree of freedom at the fixed end of a cantilever. • Parts of the model are not connected. • Numerical instability, e.g. connecting a 5m deep beam to a 5mm deep beam (resulting in an ill- conditioned matrix).

For more information see the following articles on our website:

• Unconnected Models Article: https://support.prokon.com/kb/articles/sumo-unconnected-models- message. • Unconnected Models YouTube video: https://youtu.be/VYgMk6BYSZg

4.8 Node is not connected to the model Floating Line Load stops short of a column: a node is generated over the punched-out area of the slab.

Sumo automatically handles the node generation and numbering. This means that the model is completely dependent on the user input. This example shows how a node can be generated that is unconnected from the structure.

To resolve this, we can extend the line load to the position of the column.

Sumo Workshop Training Manual

Page 2

4.9 Buckling analysis manual calculations Single column, pinned top and bottom. Calculate buckling analysis by hand. The supplied Calcsheet contains an equation to calculate the critical buckling load.

A manual Calculation can be done to calculate the buckling capacity of the single column.

Buckling Analysis with 1 Segment vs. Buckling analysis with 10 segments.

Buckling factor is too high compared to manual calculation – only one segment in the column; the column is numerically too stiff.

4.10 Second-order unsuccessful Bottom chords of trusses are unbraced, buckling analysis will show this.

For a buckling analysis, values between 0 and 1 indicate that an element (or elements) in the structure will buckle. Buckling factors should be greater than 1.

4.11 Area support lifting Area support doesn’t allow tension in springs. Uplift load case has no restraint against complete uplift.

4.12 Drop objects tool to modify geometry In the example, one of the purlins are only connected at one end. The other end is higher than the rafter.

Use the Drop Objects to connect the higher end to the rafter.

Sumo Workshop Training Manual

Page 3

4.13 Meshing problems Three examples are given:

1. A circular slab. 2. Rectangular slab with small indentations (architectural features). 3. Rectangular slab resting on a rectangular column.

Here are possible solutions to the three slabs:

1. What if we want a node in the centre of the slab (in this example, the maximum deflection will be in the centre)? Without modifying the mesh size, we can force Sumo to place a node there by using Constraint Lines. 2. Small architectural features will complicate the mesh and inevitably introduce triangular shell elements. We can modify the mesh size (e.g. 0.2m) but this may be costly in terms of time and computing power if the slab forms part of a larger structure. Furthermore, a 100mm indentation in the slab’s edge won’t affect the analysis considering that the slab’s thickness is 300mm. We can have a more accurate analysis with a mesh with better quality if we remove the indentations. 3. There are several ways to improve on the quality of the mesh. Here are three options: i) We can always decrease the mesh size. ii) Use Constraint Lines to force the mesh around the column’s footprint. iii) Most importantly, we should always improve on the quality of the modelling. Although this example looks nice in a 3D view, the analytical model is what’s important for analysis. It would be wise to simply move the column’s analytical li ne to the edge of the slab to simplify the mesh. 4.14 Monolithic beam results Interpreting monolithic beam results. For a detailed discussion, download the following document on our website: Monolithic Beam White Paper. 4.15 Convergence of shear in shells Compare the shear force results in a slab with varying mesh sizes to results in a beam. Shear force is usually the last component in the convergence chain. That means that a finer mesh will produce more accurate results. 5. Modelling exercise The final part of this course will offer students the opportunity to apply the skills learned throughout this course by building a Sumo model from 2D drawings.

The attached drawings contain all the information needed to create the model. You need to run a successful linear analysis and see if your results correlate to those of your classmates and colleagues.

Sumo Workshop Training Manual

Page 4

6000

1300

1100

1205

1100

2500

300X600 MONOLITHIC BEAM

2000

250

775775

3050

250 775775

2000

3050

130

1590

1720

1590

130

4015

250 775 775 775

3050

Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11