📚 Course ContentsChapter 4 Sled test facility and equipment

4. Sled test facility and equipment

A sled test facility is a controlled linear accelerator that reproduces the crash pulse of a real vehicle impact without destroying a full vehicle body. The test article — a body-in-white, a seat with a dummy, a battery pack, or a restraint module — is mounted on a rigid sled carriage that runs on rails or a guided track. A propulsion system accelerates the carriage to a target velocity, and a deceleration or acceleration device shapes the pulse that the test article experiences. The quality of the resulting data depends on three subsystems working together: the sled and its pulse generator, the data acquisition system (DAS), and the instrumentation package. This section examines each in turn and shows how they interact during a typical run.

Sled Types and Propulsion Architectures

Sled facilities fall into two broad categories based on how the pulse is generated.

Deceleration sleds accelerate the carriage to the target velocity and then bring it to rest through a programmable energy-absorbing device. The test article is oriented so that the deceleration direction corresponds to the crash direction of interest. Common deceleration mechanisms include:

Hydraulic servo-brake sleds — a computer-controlled hydraulic actuator applies a closed-loop braking force. Pulse shape is programmed as a force-versus-time or deceleration-versus-time profile, and the controller adjusts valve position in real time. These sleds offer excellent pulse repeatability and are well suited to restraint calibration where small pulse variations change occupant response.
Metal-tube or crush-tube sleds — the carriage drives a mandrel through a ductile metal tube, plastically deforming it. The pulse is determined by tube material, wall thickness, and mandrel geometry. This is a single-use, low-cost approach with moderate repeatability.
Wire-bending and strap-bending sleds — the carriage pulls wire or strap through a bending die. Pulse shape is tuned by die radius, wire diameter, and number of passes. These are common for high-velocity, short-duration pulses.

Acceleration sleds (also called reverse sleds or accelerating sleds) start at rest and are accelerated by a propulsion system — typically a pneumatic piston, a linear motor, or a bungee/catapult arrangement. The test article is mounted facing backward relative to the acceleration direction, so the inertial loading on the article mimics a frontal deceleration. Acceleration sleds are often used when the required velocity change is large or when the pulse must be sustained over a long duration.

The choice between deceleration and acceleration depends on the required velocity change (Δv), pulse duration, peak deceleration, and the mass of the test article. A hydraulic deceleration sled can typically achieve Δv up to about 25–30 m/s with peak decelerations of 20–60 g, while a pneumatic acceleration sled may reach higher Δv but with less precise pulse shaping.

Acceleration Profiles and Pulse Shaping

The acceleration profile is the single most important input to a sled test because it defines the inertial environment that the restraint system must manage. Profiles are specified as acceleration versus time, and the sign convention must be fixed before any data is interpreted. In this course, positive acceleration is defined in the direction of vehicle travel; a frontal crash therefore produces a negative acceleration (deceleration) pulse. All plots and calibration targets follow this convention unless stated otherwise.

Typical pulse shapes include:

Trapezoidal — a linear rise to peak, a constant plateau, and a linear decay. Easy to program and widely used for component-level restraint calibration.
Half-sine — a smooth bell-shaped pulse. Common for occupant restraint tuning because it approximates many real barrier impacts.
Haversine and multi-peak — used when the real vehicle pulse has a complex shape that must be reproduced faithfully.

Pulse shaping is achieved by adjusting the energy absorber. On a hydraulic sled, the controller receives a target profile and iterates the valve command until the measured carriage acceleration matches the target within a tolerance band — typically ±3 g or ±5% of peak, whichever is greater. On a crush-tube sled, shaping is done by selecting tube geometry and material; a trial run is usually required to characterize the actual pulse before the real test.

Practical note: Always verify the pulse with a pre-test calibration run using an instrumented calibration carriage of similar mass. A pulse that looks correct on the controller display may differ at the test article mounting points because of structural flexibility in the carriage and fixture.

Data Acquisition Systems

The DAS captures all transducer signals during the run and must satisfy three requirements simultaneously: sufficient bandwidth, sufficient resolution, and reliable triggering. For automotive crash and sled testing, the de facto standard is SAE J211, which defines channel classes by frequency content. Most sled tests use:

Channel Class 1000 — anti-alias filter at 1650 Hz, sample rate typically 10–20 kHz. Used for accelerometers on the sled carriage and structural members.
Channel Class 600 — filter at 1000 Hz. Used for dummy head and chest accelerometers.
Channel Class 180 — filter at 300 Hz. Used for seat belt load cells and displacement sensors.

A modern DAS consists of a ruggedized on-board unit mounted on the carriage, signal conditioning modules, and a host computer connected by a tether or telemetry link. On-board recording is preferred because it eliminates cable-induced noise and tether drag. The system is armed before the run and triggered by a signal from the propulsion controller — typically a break-wire or an optical gate at a known track position. Pre-trigger data (50–100 ms before the trigger) is always recorded so that the initial conditions can be verified.

Key DAS configuration parameters for a sled test are summarized below.

Parameter	Typical Setting	Rationale
Sample rate	10–20 kHz per channel	At least 5× the anti-alias filter corner frequency
Resolution	16-bit minimum, 24-bit preferred	Resolves small signals from low-g accelerometers
Pre-trigger	100 ms	Captures zero reference and arming transient
Record length	500 ms–2 s	Covers pulse plus post-impact rebound
Trigger source	Optical gate or break-wire	Deterministic time zero for all channels

Instrumentation Package

The instrumentation package converts physical quantities into electrical signals. For a restraint calibration sled test, the minimum set typically includes:

Carriage accelerometers — one or more piezoresistive accelerometers mounted on the sled structure near the test article attachment points. These define the actual pulse delivered to the article.
Dummy instrumentation — head, chest, and pelvis accelerometers; chest deflection potentiometer or IR-TRACC; upper and lower neck load cells. These follow the dummy manufacturer's mounting specifications.
Belt load cells — mounted in line with the seat belt webbing to measure restraint force. A 5 kN capacity is typical for a single occupant.
Displacement sensors — string potentiometers or laser displacement sensors to track seat track travel, belt payout, or knee bolster intrusion.
High-speed video — synchronized with the DAS trigger, typically 1000–2000 fps, to correlate mechanical events with signal features.

All transducers must be calibrated within their specified tolerance before the test. Accelerometers are calibrated on a shaker or a centrifuge; load cells with a dead-weight or hydraulic calibrator. Calibration certificates are recorded in the test file and the calibration coefficients are entered into the DAS channel configuration.

Putting It Together: A Representative Run Sequence

Consider a frontal restraint calibration test with a target pulse of 40 g peak, 100 ms duration, and Δv = 15 m/s. The sequence proceeds as follows.

Pre-test setup. The test article is mounted on the carriage and torqued to specification. Accelerometers are bonded and their axes verified against the vehicle coordinate system. Belt load cells are zeroed with the belt routed but unloaded. The DAS is configured with the correct channel classes and calibration coefficients, and a shunt calibration is performed to verify the signal path.
Arming and trigger check. The DAS is armed. A dry-run trigger is issued to confirm that all channels record and that time zero is consistent across the system. The optical gate is aligned and its output verified.
Run. The propulsion system accelerates the carriage to 15 m/s. The pulse shaper engages and decelerates the carriage according to the programmed profile. The DAS records all channels from 100 ms before the trigger to 1 s after.
Post-test review. The measured carriage acceleration is compared with the target. If the peak is within ±3 g and the duration within ±5 ms, the pulse is accepted. Dummy responses are then examined: head resultant acceleration, chest deflection, and belt forces are checked against the calibration targets. High-speed video is reviewed frame-by-frame to confirm that the dummy kinematics match the expected restraint behavior.

Common failure modes: A pulse that is too high in peak usually indicates a pulse shaper that is too stiff or a controller gain that is too aggressive. A pulse with excessive ringing suggests insufficient structural damping in the carriage or a loose fixture. Missing or noisy channels are most often caused by a poor transducer bond, a damaged cable, or a ground loop between the on-board DAS and the host computer. Always check the pre-trigger baseline — a non-zero or drifting baseline invalidates the entire channel.

The sled facility is only as good as its calibration discipline. A pulse that is 2 g high or a load cell that is 3% out of calibration can shift the apparent restraint performance enough to send an engineer down the wrong design path. For this reason, every run should be preceded by a documented calibration check and followed by a pulse verification before the data is used for restraint model calibration.