4. Head injury channel assessment

The head of a Q-Series child dummy is instrumented to capture the translational acceleration that the head experiences during an impact. In NCAP testing, this single measurement channel drives the most heavily weighted injury metric in the entire assessment: the Head Injury Criterion (HIC). Getting the acceleration channel right — its mounting, its filtering, its time-zero alignment, and its post-processing — determines whether the reported HIC value is defensible. This section walks through the measurement chain, the HIC computation, the head impact criteria used in Q-Series protocols, and the practical data-processing steps that turn a raw acceleration trace into a certified result.

The measurement chain inside the Q-Series head

The Q-Series head is a rigid, instrumented assembly. Depending on the dummy variant (Q0, Q1, Q1.5, Q3, Q6, Q10), the head contains one or more accelerometers, typically a triaxial accelerometer block mounted at or near the head centre of gravity. The accelerometer axes are aligned to the dummy's anatomical coordinate system: the X axis points forward (anterior), the Z axis points upward (superior), and the Y axis completes the right-handed set pointing to the dummy's left. All acceleration signs follow this convention throughout the test and the analysis; a forward head impact produces a positive X acceleration, and the resultant is computed from all three axes.

Signal conditioning is equally important. The accelerometer output passes through an amplifier and anti-alias filter before digitisation. For NCAP head channels, the standard acquisition chain uses a low-pass filter with a CFC 1000 characteristic (cut-off frequency 1650 Hz) for the raw channel, and the HIC computation is performed on a channel filtered to CFC 180 (cut-off 300 Hz) or, in some protocols, on the CFC 1000 data with a separate software filter applied. The exact filter class must match the protocol being followed; mixing filter classes between the raw and processed channels is one of the most common sources of HIC discrepancies between laboratories.

Computing HIC from the acceleration trace

HIC is defined over a time window bounded by two instants, t₁ and t₂, within the impact event. The formula is:

HIC = max over (t1, t2) of [ (1/(t2 - t1)) * ∫ a(t) dt from t1 to t2 ]^2.5 * (t2 - t1)

where:
  a(t)  = resultant translational acceleration at the head CG, in g
  t1,t2 = window boundaries, with t2 - t1 ≤ 36 ms (15 ms for some protocols)
  the integral is the velocity change over the window

In practice the resultant is computed first from the three filtered axes:

a_res(t) = sqrt( ax(t)^2 + ay(t)^2 + az(t)^2 )

Then the HIC search is performed numerically. A robust implementation samples the acceleration at the acquisition rate (typically 10 kHz or 20 kHz), computes the cumulative integral once, and then evaluates every admissible (t₁, t₂) pair. For a 36 ms window at 20 kHz, that is roughly 720 samples per window and on the order of 10⁵ candidate pairs — trivial for modern analysis software but worth understanding because the window length and the search resolution directly affect the reported value.

Two practical points dominate the result. First, the time-zero reference: HIC windows are anchored to the first contact or to a defined trigger, and a shift of even 1 ms can move the optimum window and change HIC by several percent. Second, the window duration: the 36 ms window is the classical limit, but many NCAP protocols for child dummies specify a 15 ms window for certain assessments. The analysis must state which window was used.

Head impact criteria in the Q-Series context

NCAP protocols for child occupants typically apply a HIC limit in the range of 1000 for the 36 ms window, with lower thresholds (often 700–800) applied in some rating schemes or for specific impact conditions. The Q-Series head channel is also used to derive the 3 ms clip — the maximum resultant acceleration sustained for at least 3 ms — which provides a complementary check on peak severity that HIC can under-represent in short, sharp impacts.

These limits are illustrative of common NCAP practice and must be confirmed against the specific protocol revision in force for the test programme. The important engineering point is that HIC and the 3 ms clip are computed from the same underlying acceleration trace but with different filters and windows, so both must be derived from a validated, time-aligned channel.

Data processing workflow

The processing sequence from raw acquisition to certified HIC follows a fixed order, and deviating from it introduces error. The steps are:

1. Channel verification. Confirm the accelerometer serial number, sensitivity (mV/g), and calibration date against the test sheet. Check the zero-offset before impact; a drifting baseline indicates a loose connector or a failing sensor.
2. Time-zero alignment. Align the head channel to the common trigger (first contact or a defined electrical trigger). Verify alignment against a reference channel such as the chest or vehicle acceleration.
3. Filtering. Apply the protocol-specified filter class. For HIC, apply CFC 180 (or the protocol equivalent) to the resultant; for the 3 ms clip, retain CFC 1000.
4. Resultant computation. Compute the vector resultant from the three filtered axes.
5. HIC search. Run the sliding-window maximisation over the admissible window length and record t₁, t₂, the window duration, and the HIC value.
6. Cross-checks. Verify that the peak resultant, the 3 ms clip, and the HIC window are mutually consistent — the HIC window should contain the peak acceleration unless the impact has a double-lobe profile.

Worked example: from trace to HIC

Consider an illustrative frontal impact where the head channel, filtered to CFC 180, shows a resultant acceleration that rises from zero at t = 0 ms to a peak of 120 g at t = 8 ms, then decays to near zero by t = 25 ms. The velocity change over the 36 ms window is obtained by integrating the resultant; for this simplified profile, the integral evaluates to approximately 6.5 m/s (expressed in g·s, about 0.66 g·s). The mean acceleration over the window is therefore about 18.3 g, and HIC = (18.3)^2.5 × 0.036 ≈ 1450. This exceeds the 1000 limit, indicating that the head channel would fail the criterion for this event. The example is constructed for illustration; actual values depend on the measured trace, the filter class, and the exact window boundaries.

The diagnostic value of the calculation lies in the window: if the HIC-optimal window is much shorter than 36 ms, the impact is dominated by a short, sharp pulse, and the 3 ms clip should be checked as the governing metric. If the optimal window spans nearly the full 36 ms, the event is a sustained loading and HIC is the appropriate discriminator.

Common failure modes and troubleshooting

When a head channel result looks implausible, the fault usually sits in one of four places. A baseline offset — the pre-impact mean deviating from zero by more than a few tenths of a g — shifts the integral and inflates or deflates HIC; re-zero the channel against the pre-trigger window. A time-zero misalignment between the head channel and the trigger channel moves the optimal window; cross-correlate the head trace against the chest or vehicle channel to confirm alignment. A filter mismatch between the raw and processed channels produces a HIC that cannot be reproduced; re-derive from the archived raw data with the documented CFC class. Finally, a saturated or clipped channel — visible as a flat-topped peak — invalidates the result entirely; the accelerometer range must be checked against the expected peak before the test, and a saturated channel must be reported as invalid rather than processed.

Verification of the final result follows a simple consistency check: the HIC-optimal window should contain the peak resultant, the 3 ms clip should be consistent with the peak magnitude and pulse width, and the velocity change over the window should match the integral of the trace. When all three agree, the head channel result is traceable and ready for the NCAP assessment.