2. Navigating the D3PLOT interface

D3PLOT is the post-processor in the Oasys LS-DYNA environment, and almost every productive session begins with the same four interface areas: the model tree that organizes what is loaded, the view controls that determine how geometry is projected on screen, the selection tools that let you pick entities precisely, and the result visualization controls that map state variables onto the model. Learning to move between these areas fluently is what separates a slow, menu-hunting session from a fast diagnostic workflow. This chapter walks through each area with concrete operations and the observable outcomes you should expect.

Understanding the model tree

The model tree is the hierarchical index of everything currently in the session. When you open a .d3plot state database, D3PLOT creates a tree whose top-level nodes typically include the model title, the part list, and the state/result data. Expanding a part node reveals its element types and, where applicable, its material and section assignments. The tree is not merely a list: it is an active selection surface. Clicking a part name highlights that part in the graphics window; Ctrl-clicking adds further parts to the current selection; right-clicking opens a context menu with visibility, colour, and result-scoping options.

A practical habit is to use the tree to build named groups before you start interrogating results. For example, in a frontal-impact model you might group the B-Pillar_Inner, B-Pillar_Outer, and Rocker parts into a group called occupant_cell. Groups persist for the session and can be recalled instantly, which matters when you need to compare the same region across dozens of states. The tree also exposes the state list: each state corresponds to a saved time step, and selecting a state in the tree is equivalent to scrubbing the animation to that frame.

View controls and the projection model

D3PLOT uses a right-handed Cartesian coordinate system inherited from the LS-DYNA input deck. The default view is an isometric projection looking from the positive octant toward the origin, with +X to the right, +Y upward, and +Z toward the viewer. Every view operation is defined relative to this frame, so it is worth confirming the axis triad in the corner of the graphics window before interpreting any deformation plot.

The view controls fall into three groups. Standard views snap the camera to the principal axes: front (+Z looking toward −Z), back, left, right, top, and bottom. Free rotation is handled by dragging with the middle mouse button, while panning uses the right mouse button and zooming uses the scroll wheel or a box-zoom drag. The fit command (F by default) frames the entire visible model, which is the fastest recovery when you have zoomed into empty space.

Two view settings deserve explicit attention because they change what you are actually looking at. Perspective versus orthographic projection: perspective mimics a camera and is useful for presentation, but orthographic is preferable for measuring and comparing displacements because parallel lines remain parallel. Clipping planes let you cut through the model to inspect internal structure; a clip plane defined at Y = 0 with the normal pointing in +Y will hide everything above the mid-plane, exposing the section behind it. Always record the clip plane position when you capture an image, otherwise the view is not reproducible.

Selection tools

Selection in D3PLOT operates on a pick hierarchy: parts, nodes, elements, and (where present) sets. The pick tool activates a mode in which clicking in the graphics window selects the entity under the cursor; the status bar reports what was picked, for example Element 148223 (Part 17). The box pick and polygon pick tools select all entities whose centroids fall inside the drawn region, which is far quicker than clicking individual elements on a dense mesh.

Selection is almost always used to scope a result rather than to edit geometry. If you select a set of elements and then choose a contour plot, D3PLOT plots only the selected entities and leaves the rest in wireframe. This is the standard way to isolate a weld line, a flange, or a single shell patch for detailed inspection. The invert selection command is useful when it is easier to pick the small region you want to exclude than the large region you want to keep.

Basic result visualization

Result visualization in D3PLOT is driven by the contour panel. The essential inputs are the result quantity (for example Effective Stress (v-m) or Resultant Displacement), the component (X, Y, Z, or magnitude), and the colour scale. The default scale is auto-ranged to the current state, which is convenient for a first look but misleading for comparison: a fringe that looks severe at state 40 may simply reflect a narrow auto-range. For any quantitative comparison, switch to a fixed range and enter explicit minimum and maximum values, for example 0 to 350 MPa for a steel component with a 350 MPa yield strength.

The legend in the corner of the window reports the current range and the number of contour bands. Increasing the band count from the default to 12–16 improves the readability of gradients without creating visual noise. The displacement scale factor is a separate control: it multiplies the nodal displacement before plotting, so a factor of 10 makes a 2 mm deflection look like 20 mm. This is a visualization aid only — it does not change the underlying data — but it must be stated whenever a deformed plot is used as evidence.

A reproducible first-look sequence for any new state database is: fit the model, set an orthographic view along the loading axis, apply a fixed contour range based on material allowables, set the displacement scale to 1.0, and step through the state list while watching the legend maximum. If the legend maximum saturates at the top of your fixed range early in the event, you have identified the onset of yield in the region of interest and can then zoom in with the selection tools to find the specific elements driving it.

Putting the four areas together

The diagram above shows how the four interface areas connect in a single workflow. The model tree supplies the entity scope (parts, groups, states); the selection tools refine that scope to specific elements; the contour panel defines what quantity is mapped and over what range; and the view controls determine the projection, clipping, and scale factor under which the result is rendered. A change in any one of these propagates to the graphics window, which is why a disciplined sequence — scope, quantity, range, view — produces reproducible images and defensible conclusions.

As a concrete check, load a state database, select the occupant_cell group, set the contour quantity to Effective Stress (v-m) with a fixed range of 0–350 MPa, switch to an orthographic front view, and step from state 0 to state 40. If the legend maximum reaches 350 MPa before state 40, note the state number at which saturation first occurs; that state marks the onset of yield in the selected region. Then use box pick to isolate the saturated elements and read their IDs from the status bar. This short sequence exercises every interface area covered in this chapter and produces a result you can defend in a review.