2. Cable and connector standards for crash testing
Crash test instrumentation is only as trustworthy as the signal path that carries the measurement. A 50 g accelerometer with excellent linearity can still produce a corrupted pulse if its cable is routed across a deforming structure, if the connector backshell is not strain-relieved, or if the channel filter does not match the standard the data will be judged against. Three layers of requirements govern this signal path: the measurement-channel standards (SAE J211 and ISO 6487), the component-level performance of cables and connectors, and the OEM-specific specifications that add routing, marking, and validation rules on top. This chapter explains what each layer actually controls, where they overlap, and how to translate them into concrete hardware and routing decisions.
What SAE J211 and ISO 6487 Actually Govern
SAE J211 and ISO 6487 are frequently described as "sensor standards," but they are more precisely data-channel standards. They define the frequency response, phase behavior, and filtering of the entire measurement chain from transducer to recorded data, so that results from different laboratories are comparable. The key parameters are the channel class (CFC — Channel Frequency Class), the anti-alias and low-pass filter characteristics, and the tolerance bands on amplitude and phase within the passband.
The practical consequence is that the standard constrains the cable and connector indirectly but firmly. A channel declared as CFC 1000 must remain within its amplitude tolerance up to 1000 Hz and is typically filtered at a corner frequency near 1650 Hz. Any cabling that introduces excessive capacitance, intermittent contact resistance, or shield discontinuity will distort the channel response and can push it outside the tolerance band — even though the cable itself is not the object the standard names.
| CFC class | Typical use in crash testing | Filter corner (approx.) | Cabling implication |
|---|---|---|---|
| CFC 60 | Occupant chest/abdomen displacement, belt force | ≈100 Hz | Low bandwidth; long cable runs acceptable if shield intact |
| CFC 180 | Head acceleration, pelvis acceleration | ≈300 Hz | Moderate; twisted pair recommended |
| CFC 600 | Dummy chest acceleration, structural pulse | ≈1000 Hz | Short runs, continuous shield, low-capacitance cable |
| CFC 1000 | Sled/barrier acceleration, high-rate events | ≈1650 Hz | Critical: minimize length, avoid shield breaks, verify with shunt calibration |
Note that the corner frequencies above are the standard filter values associated with each class; always confirm the exact filter definition and tolerance in the revision of J211 or ISO 6487 your program cites, because revisions differ in detail. The table is a routing guide, not a substitute for the controlled document.
Component-Level Requirements: Cables and Connectors
Neither J211 nor ISO 6487 specifies a cable part number. That gap is filled by component standards and by OEM specifications. In practice, crash-test cabling is judged on four properties:
- Shielding and shield termination. A braided or foil shield must be continuous from transducer to data acquisition, with the shield bonded to chassis ground at the acquisition end (and, where the OEM requires, at the transducer end through a defined impedance). A shield that is "grounded somewhere" is not a shield; it is an antenna.
- Twisted-pair construction for differential channels. Charge-mode accelerometers and strain-gauge bridges are differential. Twisting preserves common-mode rejection and prevents loop area that couples magnetic fields during the crash pulse.
- Mechanical retention. Connectors must survive the acceleration environment without contact fretting. Positive-lock (bayonet, threaded, or push-pull with secondary latch) shells are mandatory; friction-fit connectors are not acceptable on body-mounted channels.
- Strain relief and bend radius. The backshell must carry the load, not the solder joints or crimp contacts. A minimum bend radius of roughly 10× cable diameter is a common OEM rule of thumb.
How the Three Layers Interact
The diagram below shows the signal path from transducer to recorded channel and maps which layer of requirement governs each segment. The key insight is that the standards define the end-to-end performance envelope, while OEM specifications define the physical implementation that must fit inside that envelope.
Translating OEM Specifications into Routing Decisions
OEM specifications are where the standards become physical. A typical OEM document will state, for example, that occupant-restraint channels must be routed with a minimum 25 mm clearance from pyrotechnic devices, that cable ties must be spaced no more than 150 mm apart on rigid structure, that no connector may be located within a zone expected to deform by more than a defined amount, and that every channel must carry a unique label at both ends. These are not arbitrary; they encode lessons from tests where a cable was severed by a pretensioner firing or a connector was pulled apart by floorpan deformation.
A practical routing workflow that respects all three layers looks like this:
- Classify each channel by CFC. Pull the channel list from the test request and assign CFC 60/180/600/1000. This immediately sets the maximum acceptable cable length and the shielding rigor.
- Draft the physical route on the body-in-white drawing. Mark every connector location, every tie point, and every crossing of a deformation zone. Flag any CFC 600/1000 channel whose run exceeds the OEM length limit.
- Select cable and connector from the approved list. Match the connector shell to the transducer interface (e.g., a 4-pin micro connector for a bridge channel) and confirm the cable's capacitance per meter against the channel's source impedance.
- Install with strain relief and shield bonding. Terminate the shield at the DAQ end, torque the connector to the OEM-specified value, and apply the secondary latch.
- Verify before the test. Run the wiggle test, a shunt calibration on each bridge channel, and a continuity check on every shield. Record the results on the channel sheet.
Documentation and Traceability
Because the standards judge the recorded channel, not the hardware, the evidence that a channel met its class must be documented. A minimal channel record for each measurement includes: the transducer serial number and calibration date, the cable part number and length, the connector type and torque value, the CFC class and filter setting, the shunt calibration result, and the shield continuity result. This record is what allows a laboratory to defend its data when a result is challenged, and it is what an OEM auditor will ask for first.
The three layers are not alternatives to choose between. SAE J211 and ISO 6487 tell you what the channel must achieve; component standards tell you what the hardware must be capable of; OEM specifications tell you how to install it in a specific vehicle. A channel that satisfies only two of the three will pass a bench check and fail on the sled.