2. Dummy anatomy and sensor locations

A crash test dummy is not a mannequin with a few accelerometers bolted on. It is a segmented mechanical system whose mass distribution, joint stiffness, and sensor placement are all defined by the standard it is built to — most commonly the Hybrid III family for frontal impact, the THOR-M for advanced frontal work, and the EuroSID-2re or WorldSID 50th for side impact. Before you can position a dummy or route its cabling, you need a working mental model of how the body is divided into measurement regions and what each region is expected to report.

The segmentation follows human anatomy closely: head, neck, thorax (upper and lower), abdomen, pelvis, and the four limbs. Each segment is a rigid or semi-rigid assembly joined to its neighbours through a calibrated element — a neck spring pack, a lumbar spine, a ballasted pelvis block. Sensors are mounted on or inside these segments, and the mounting point matters as much as the sensor itself, because the measured quantity is defined relative to a specific location in the dummy's local coordinate system.

Coordinate convention used throughout this course

All positions in this chapter are expressed in the dummy's local axis system, consistent with SAE J211 and the dummy certification drawings: +X forward (toward the front of the vehicle), +Y to the dummy's left, +Z upward. Rotations follow the right-hand rule about these axes. When a sensor is described as "mounted at the head CG," that means the geometric centre of gravity of the head assembly as defined in the certification drawing, not the visual centre of the skull shell. Keep this convention in mind when reading the diagram below and when cross-checking against a build sheet.

Head and neck

The head assembly carries a triaxial accelerometer cluster at the head centre of gravity, and on many builds a second set of angular-rate sensors or a nine-accelerometer array for rotational kinematics. The mounting plate is machined into the skull shell and the sensor block is oriented so that its axes align with the dummy's local X/Y/Z. A common error is to install the block rotated 90° about Z; the data will look plausible but the sign of Ay will be inverted, which silently corrupts any HIC or rotational injury criterion computed downstream. Always verify orientation against the build sheet before the run, not after.

The neck is instrumented with an upper and a lower load cell that measure the six components of force and moment transmitted between head and thorax. These cells are calibrated as a pair, and the neck spring pack between them is a certified component with a defined stiffness. If you replace a neck, you must re-run the neck certification pendulum test and record the new serial numbers in the test log — the load-cell data is only meaningful when tied to a certified neck.

Thorax and abdomen

The thorax carries a triaxial accelerometer at the upper sternum and, on the Hybrid III, a chest deflection potentiometer or IR-TRACC that measures rib deflection at a defined lateral offset. The deflection sensor is the one most sensitive to setup: its zero position must be set with the dummy unloaded and the chest at rest, and the push-pull range must be checked so the sensor does not bottom out under belt loading. A deflection channel that flat-tops at its mechanical limit is a classic sign that the zero was set with the belt already tensioned.

The abdomen region on the Hybrid III is largely a ballasted segment with a lower-spine load cell measuring the force transmitted from thorax to pelvis. On the THOR and WorldSID, the abdomen carries its own deflection and force instrumentation, and the mounting points are defined per rib pair. Treat the abdomen as a load path, not a passive filler: the lower-spine cell is often the channel that reveals belt submarining, because a sharp drop in lumbar force combined with rising pelvis acceleration is the signature of the lap belt sliding off the iliac crest.

Pelvis and limbs

The pelvis block carries a triaxial accelerometer at the pelvis CG and, on belted configurations, left and right iliac crest load cells that measure the belt reaction force. These cells are mounted in the load path between the belt and the pelvis structure, so their mounting bolts must be torqued to the drawing value — an under-torqued iliac cell will show a force trace that lags the belt force and can be mistaken for belt slip. The femur and tibia carry knee and ankle load cells measuring axial and shear force plus bending moment, and the ankle cells in particular are fragile: they are the first thing to be damaged by a mis-set footrest or a pedal that intrudes into the footwell.

Typical sensor inventory and mounting reference

The table below summarises the channels you will most often see on a frontal 50th-percentile build. Treat the counts as typical, not mandatory — the actual channel list is defined by the test protocol and the injury criteria being evaluated.

With the anatomy and sensor map in hand, the next practical question is how these segments are physically assembled and aligned before the dummy goes into the sled or vehicle — which is where positioning fixtures and the certification checks come in.