4. Sled system architecture

A crash sled is not a single machine but a tightly coupled system of four subsystems that must be designed, tuned, and validated as one. The propulsion subsystem generates the kinetic energy; the track subsystem guides and constrains the moving mass; the braking subsystem removes the remaining energy in a controlled way; and the data acquisition (DAQ) subsystem captures the physical response of the test article. A weakness in any one of them degrades the fidelity of the entire pulse. This section explains what each subsystem does, how the four interact, and how to reason about their key parameters during commissioning and troubleshooting.

Propulsion Subsystem

The propulsion subsystem accelerates the sled carriage to a target velocity, V0, at the moment of impact or pulse initiation. Three architectures dominate modern facilities:

• Pneumatic (air) propulsion — a pressure vessel charges to a set pressure, and a fast-acting valve releases compressed gas into a cylinder or directly behind the carriage. Thrust is high but decays as the gas expands, so the acceleration profile is inherently front-loaded.
• Hydraulic propulsion — a hydraulic accumulator drives a piston or winch. It offers smoother, more controllable acceleration and is common where long, low-g acceleration runs are required.
• Bungee / elastic propulsion — stretched elastomer cords store energy and release it as the carriage is launched. Simple and repeatable for low-speed tests, but energy storage is limited and the release is difficult to modulate.

The governing relationship is energy balance. For a carriage of mass m accelerated to V0, the required kinetic energy is E = ½·m·V0². In a pneumatic system, the stored energy in the pressure vessel is approximately E_stored = (P·V / (γ−1))·[1 − (P_atm/P)^((γ−1)/γ)] for an ideal gas, where P is the charge pressure, V is the vessel volume, and γ is the specific heat ratio. In practice, engineers work from a calibration curve: charge pressure versus achieved velocity, established empirically for each carriage mass. A typical commissioning table might read:

These numbers are illustrative. The point is that velocity is a function of both charge energy and carriage mass, and the relationship is not perfectly linear because friction and valve losses grow with speed. During a real commissioning, you would sweep pressure at two or three masses and fit a curve before trusting any single shot.

Track Subsystem

The track provides a straight, level, low-friction path and reacts the vertical and lateral loads generated during acceleration and braking. Key components are the rails, the carriage bogies or linear bearings, and the alignment system. Track straightness directly affects yaw and pitch of the test article; a misaligned rail can inject a rotational component into an otherwise axial crash pulse.

Two parameters matter most during setup:

• Rail parallelism and levelness — measured with a laser tracker or dial indicators. A common tolerance is ±0.5 mm over the full run length for a high-fidelity sled.
• Bogie preload and rolling resistance — too much preload increases friction and reduces repeatability; too little allows the carriage to lift or rattle. Rolling resistance is typically characterized by a coast-down test: accelerate to a known speed, release propulsion, and measure deceleration. A coast-down deceleration of 0.02–0.05 g is typical for a well-maintained track.

The track also carries the reaction structure for the braking system and the mounting points for the DAQ cable management. Cable management deserves early attention: signal cables and power lines must be routed so they do not snag, stretch, or induce noise as the carriage travels.

Braking Subsystem

After the test pulse, the carriage still carries significant kinetic energy. The braking subsystem must absorb it without damaging the carriage, the track, or the test article, and without generating a deceleration that contaminates the post-pulse data. Common braking methods include:

• Hydraulic dashpot / damper — a piston forces oil through an orifice; braking force rises with speed, giving a smooth deceleration.
• Friction brake — calipers or brake pads clamp a rail or drum. Simple and strong, but can be grabby and generate high-frequency noise.
• Energy absorber (crush tube or honeycomb) — a sacrificial element deforms plastically. Excellent for controlled deceleration but requires replacement after each shot.
• Water brake — a paddle wheel or rotor churns water; braking torque is controlled by water level. Common on high-speed tracks.

The design target is usually expressed as a maximum deceleration, for example a_brake ≤ 2 g, and a stopping distance d = V0² / (2·a_brake). For V0 = 13.6 m/s and a_brake = 2 g (19.6 m/s²), the stopping distance is about 4.7 m. The track must be long enough to accommodate this plus a safety margin. If the brake is too aggressive, the carriage may bounce or the test article may experience a secondary shock that corrupts the post-test inspection.

Data Acquisition Subsystem

The DAQ subsystem converts physical quantities — acceleration, force, displacement, strain, temperature — into digital records synchronized to a common time base. Its architecture has four layers: sensors, signal conditioning, digitization, and recording.

Sensors. Piezoresistive accelerometers are standard for crash pulses because they respond to DC (constant acceleration), unlike piezoelectric types which cannot measure sustained acceleration. Force links measure the interface load between the test article and the carriage. String potentiometers or laser displacement sensors capture deformation.

Signal conditioning. Bridge completion, amplification, and anti-alias filtering happen here. A typical accelerometer channel uses a gain of 100–1000 and a low-pass filter set to 1/2.5 of the sample rate to prevent aliasing.

Digitization. The sample rate must be high enough to resolve the pulse. A common rule is f_s ≥ 10 × f_max, where f_max is the highest frequency of interest. For a crash pulse with meaningful content up to 1 kHz, a sample rate of 10–20 kS/s per channel is typical. Resolution is usually 16-bit, giving a dynamic range of about 96 dB.

Recording and synchronization. All channels must share a common trigger and time base. A hardware trigger derived from the propulsion release or a break-beam at a known track position is preferred over software triggering. The time-zero reference must be documented: it is often the moment the carriage passes a specific track marker, not the moment the valve opens.

The following SVG shows how the four subsystems connect and where the key signals and energy flows occur.

Integration and Commissioning Checks

Before a sled is trusted for production testing, the four subsystems must be verified together. A practical commissioning sequence looks like this:

1. Static checks. Verify track levelness and parallelism, brake stroke, sensor mounting torque, and cable routing. Confirm that the DAQ records a known calibration signal (e.g., a shunt calibration on a strain channel or a known 1 g static offset on an accelerometer).
2. Low-speed dry runs. Fire the propulsion at the lowest practical pressure with no test article. Confirm that the carriage travels smoothly, the brake engages, and the DAQ triggers at the expected track position. Compare measured V0 against the calibration curve.
3. Pulse verification. With a dummy test article of known mass, run a representative pulse. Check that the measured acceleration pulse matches the target within tolerance — typically ±5% on peak g and ±10% on pulse duration.
4. Repeatability. Repeat the same shot three times. The coefficient of variation on peak acceleration should be below 3% for a well-controlled system. If it is higher, investigate track friction, valve timing, or brake interference.

Common Failure Modes and Troubleshooting

Each subsystem has its own failure signature, but the symptoms often appear in the data of another subsystem. A yaw problem shows up as an unexpected lateral acceleration channel; a brake problem shows up as a post-pulse oscillation in the longitudinal channel. Reading the data across all channels — not just the primary accelerometer — is the fastest way to localize the root cause.

In summary, the sled system architecture is a chain of energy conversion and measurement. Propulsion converts stored energy into motion; the track constrains that motion; braking removes the residual energy; and the DAQ subsystem records what the test article experienced. Designing and commissioning each subsystem with its own parameters and tolerances, then validating them together through dry runs and repeatability checks, is what separates a reliable crash sled from one that produces unusable data.