(F-207) Characterizing Natural Frequencies of the Hybrid III and NOCSAE Headforms

Introduction:: Anthropomorphic test device (ATD) headforms are used to study different head impact scenarios. Two popular ATD headforms are the Hybrid III and the National Operating Committee on Standards for Athletic Equipment (NOCSAE). Despite structural differences, both headforms have biofidelic biomechanical responses and perform similarly when used to study sports-related head impacts.

While these headforms are used to understand the behavior of the human head in different impact events, little research has been done on fully understanding their dynamic properties, especially the natural frequencies. All systems have natural frequencies at which they vibrate in the absence of any driving forces. If a time-varying external load is applied to a system, and its frequency is equal to any of the system’s natural frequencies, resonance will occur. This causes the system to oscillate at high amplitudes, causing abnormally large deflections of the system. Understanding the natural frequencies of headforms used in impact testing is important because the head will behave differently depending on the characteristics of the impact. This is especially true for shorter-duration impacts (t < 5 ms); the broader frequency spectrum can cause resonant frequency excitation. Therefore, in impact testing, it is important to consider headform frequency response. This study’s objective was to identify the natural frequencies of the Hybrid III and NOCSAE headforms to gain insight into how they respond to different types of impacts.


Materials and Methods:: Experimental modal analysis uses excitation and response signals to identify natural frequencies of a structure, so modal testing methods were implemented. A 50^th percentile male Hybrid III headform and a medium NOCSAE headform were suspended with wire from a metal frame to avoid any damping effects that could occur from attaching to a rigid neck. To collect linear acceleration data, which served as the response signals, the headforms were instrumented with three linear accelerometers at their respective centers of gravity. An impulse force hammer (PCB Piezotronics, Model 086C04), hereafter referred to as an impact hammer, with a stainless-steel tip, was used to excite the headforms with a bandwidth of approximately 1000 Hz. A “roving hammer” test approach was used, where the accelerometers remained fixed and the headforms were struck at different locations (Figure 1). A total of 360 tests were conducted, with nine locations on each headform struck 20 times. Multiple locations were selected to excite various possible modes in the headforms, as natural frequencies can vary with location.

The data from each location were averaged but not filtered to avoid attenuating any important frequencies. The output contained more noise than the input, so the H1-estimator was used to determine the Frequency Response Function (FRF); this is the cross-power of the excitation and response signals divided by the auto-power of the excitation signal. When the FRF is plotted, the frequencies corresponding with each local maximum are the natural frequencies. The coherence was also calculated, evaluating the quality of the measurements.


Results, Conclusions, and Discussions:: In the Hybrid III headform, there were no natural frequencies observed below 1000 Hz. The average value for the first natural frequency of the NOCSAE headform was 848 Hz, with values ranging from approximately 733-1073 Hz. Coherence describes what portion of the response is attributed to the excitation, essentially assessing the measurement quality and determining the reliability of the modal parameters estimated. The coherence across the frequency bandwidth was found to be greater than 0.995 for each signal of interest, allowing for confidence in the accuracy of the natural frequency values. The first natural frequency of the Hybrid III headform is likely above 1000 Hz, and since the bandwidth of the impact hammer prevented us from identifying frequencies greater than this value, we were unable to determine specific natural frequencies for this headform.

Figure 2 shows FRF plots for oblique hits for each headform, as this location was most representative of the frontal and temporal impacts. Peaks are seen much earlier in the frequency range for the NOCSAE headform when compared with the Hybrid III. Previous studies have shown the human head to exhibit lower natural frequencies ranging from approximately 500-800 Hz, so the NOCSAE headform might actually be a more accurate representation of the human head.

When performing impact testing, it is important to be aware of these dynamic properties, because when resonant frequencies occur due to the head experiencing a loading at or near one of its natural frequencies, the responses will be affected accordingly. The same impact on two distinct headforms that perform similarly in blunt impacts can be quite different in short-duration loading environments due to their unique vibrational responses. Therefore, it is important to consider this when comparing data across headforms. By understanding the intrinsic characteristic of two popular ATD headforms, informed decisions can be made in test environments to determine which one will best capture the frequency response of the head depending on the impact scenarios.

Acknowledgements (Optional): :

References (Optional): :