Full test report from TEC Edmonton and UA Solve

Executive Summary

Dynamic Composites Inc., through TECEdmonton’s UA Solve program, contracted impact testing of a prototype helmet retrofit, termed Dynamic Composites Retrofit (DCR), to the University of Alberta, Department of Mechanical Engineering. The retrofit comprises a foam layer that can be fixed to external helmet surfaces, at the approximate center of which is a collapsible carbon fibre insert that is designed to plastically fail, and therefore dissipate energy, in a rear helmet impact.

Drop tests onto a steel anvil were conducted using Bauer 4500 helmets, to compare posterior impacts with and without the DCR. Impacts at a nominal 4.8 m/s exhibited 70% lower peak accelerations than the same helmet without a retrofit. The maximum velocity change of the headform was approximately constant (7 m/s) for all helmets. The average time duration of the headform accelerations associated with the primary helmet impact were 8.3 ms for helmets without a DCR and 16.6 ms for helmets with a DCR.

Post impact visual inspection of the Bauer 4500 helmets indicated that the presence of the DCR did not result in obvious damage to the helmet’s exterior hard plastic shell.

Overall, this preliminary testing suggests that the prototype DCR meets the design objective of reducing peak headform accelerations in rear helmet impacts, while not interfering with or damaging the helmet outer shell.

1.0 Introduction
Dynamic Composites Inc. (1995) is a Canadian owned (sole proprietor) and operated company that develops and markets niche sport equipment based on composite materials (e.g. carbon fibre). Their primary revenue stream is from sporting goods manufacture and design with examples including composite helmet shells, bicycle frames, and cycling racing wheels that are both lighter and stronger than conventional alternatives. Recently, Dynamic Composites has directed resources to improving the impact attenuation performance of existing single-use (e.g. bicycle) and multi-use (e.g. hockey, ski, and snowboard) helmets subjected to impacts ranging in severity from mild (comparable to what would be considered sub- concussive) to severe (comparable to what would be considered as high risk for severe brain injury) through exterior helmet retrofits (e.g. Figure 1). Concussion rates in hockey are the highest of all contact sports [1]. Flik et. al. indicated that concussion as the single most commonly sustained injury in hockey [2].

Figure 1
Figure 1: Prototype Dynamic Composites Inc. Retrofit (DCR) installed at rear of Bauer 4500 Hockey Helmet. Helmet is medium size and is certified to meet ASTM F1045 and CAN-CSA Z262.1

The target impact scenario that Dynamic Composites is focusing on (currently) is the so-called backwards fall[3], [4] where a helmet wearer falls backward and suffers an impact to the back of the helmet at a location corresponding to the rearmost position in the helmet “test zone” or in some cases outside the test zone. The impetus for focusing on backwards falls and more broadly rear helmet impacts is clinical evidence from some sports (e.g. snowboarding [5]) indicating that 68% of severe head injuries were associated with backwards falls. In the context of Canada’s national sport, rear helmet impacts (from checking and falls onto ice) are the most common (closely followed by frontal impacts) according to an American study considering both female and male collegiate players [6][7]. In hockey, rear helmet impacts are the most severe in that they result in the greatest magnitude head/brain accelerations relative to other impact locations [7]. This latter point is worrisome given that head/brain injury risk scales with acceleration magnitudes (all other factors being equal, higher accelerations carry higher injury risk). Dynamic Composites has developed a prototype helmet retrofit (Figure 1) that comprises a composite sandwich designed to affix to the rear outside of commercially available helmets and when subjected to impacts of adequate severity, this retrofit predictably and permanently deforms and thereby attenuates the impact energy.

2.0 Objectives
Dynamic Composites intends to develop and commercialize their retrofit. As part of their ongoing development and commercialization strategy, they require preliminary testing and data to prove the concept underlying the retrofit, or more specifically preliminary data showing that the retrofit can significantly reduce head acceleration (the de-facto standard metric to quantify the severity of impact energy transfer to the head) in helmeted impacts.

Dynamic Composites Inc. has contracted with TECEdmonton to perform the preliminary testing. TECEdmonton has retained Dr. Chris Dennison (Mechanical Engineering, University of Alberta) to perform preliminary testing and data analysis. Through consultations, Mr. Al Beyer of Dynamic Composites Inc.; Mr. Chris Lerohl (TECEdmonton); and Dr. Chris Dennison (University of Alberta) have agreed on the following objectives for the preliminary testing:

1. Establish baseline performance of exemplar hockey helmet (Bauer 4500) in rear helmet impacts. Performance will be quantified by headform accelerations and velocities throughout impact.
2. Establish performance of exemplar hockey helmet (Bauer 4500) equipped with retrofits constructed and supplied by Dynamic Composites Inc.

3.0 Materials and Methods
3.1 Materials Received

Mr. Al Beyer has supplied ten (10) Bauer 4500, size medium (56-59cm head circumference), hockey helmets (Figure 2). Nine (9) helmets have been retrofitted with the DCR) (Figure 1). Mr. Beyer has conveyed that the construction of the nine retrofits is variable, but he has withheld these details. Each DCR was attached to each helmet using velcro-tape. One (1) helmet is without the DCR and will be the only helmet used to ascertain baseline helmet performance (per Mr. Beyer’s preference).

Pre-impact, the mass of each helmet was measured and recorded. Masses reported for helmets 2 through 10 are inclusive of the DCR.

3.2 Helmet Marking and Identification

Each helmet has been permanently marked using permanent ink, on the helmet interior. The baseline helmet is marked Baseline-helmet 1. All remaining helmets are marked as helmet 2 through 10. After testing, all helmets were returned to their original packaging which comprised a cardboard box from the manufacturer. Each box was also permanently marked with helmet number (eg. Helmet 1) and details of the impact that the given helmet has undergone (eg. Helmet 3, 1 impact at retrofit, 4.5 m/s).

3.3 DCR Marking and Identification

Each DCR was supplied pre-installed on a helmet, by Mr. Beyer. Each DCR was pre-marked by Mr. Beyer. All DCRs were left in place on the helmet, as installed, and remained in place after impact testing. The helmets, returned to Mr. Beyer in their original packaging, are equipped with the same DCR that was installed at the time of delivery.

3.4 Impact Attenuation Testing

General considerations:
The methods used for impact attenuation testing were based on those prescribed in CAN-CSA Z262.1-09 and ASTM F1045-07, the two helmet certification standards for which the exemplar helmets are certified [8], [9]. Impact testing was performed based on these standards, however helmets were not tested to all aspects of these outlined in the standards.

FIgure 2
Figure 2: Bauer 4500, size medium. 10 exemplars supplied by Dynamic Composites Inc. Each helmet shows both ASTM and CAN-CSA certification marks.

All helmets were impacted onto a flat steel anvil as opposed to a modular elastomer programmer (MEP) called for in CAN-CSA and ASTM (25.4 mm thickness, 0.09 m2 area, shore A durometer 60) certification. The decision to impact helmets onto a flat steel anvil was agreed upon by Mr. Beyer and Dr. Dennison in verbal communication during the planning phase of this work, and the rationale for this decision is that impacting onto steel is thought to be a worst-case impact.

Drop Tower and Instrumentation Specifications:
The University of Alberta drop tower (Figure 3) is a linear monorail that is roughly 14ft tall. It is equipped with a CADEX Inc. drop pistol and ball arm which are purpose built to accept the spherical impacter and subsequently ISO-CEN K1A Mg headform (size medium in this case, circumference 575 mm).

The drop tower MEP pad meets both ASTM and CAN-CSA Hockey Helmet Standard requirements (Shore A = 59.3, 1 inch thickness) and was supplied by CADEX Inc. The total mass of the drop assembly when equipped with the spherical impacter is 4,999 grams (ASTM and CAN-CSA compliant) and was supplied by CADEX Inc. The total mass of the drop assembly when equipped with the K1A Mg head was 4,699 grams (CAN-CSA and ASTM compliant for medium headform tests) and was supplied by CADEX Inc. The steel base mass is 140 kg.

At the center of mass of the Mg headform a single axis accelerometer meeting SAE j211 specification is rigidly affixed to the ball arm (Niell-Tech CAYZ147V-2-@KA, 2000 g range, with frequency characteristics exceeding j211 requirements). The accelerometer is powered and signal conditioned with purpose-built electronics that supply reference excitation and subsequently anti-alias filter (j211 compliant) the analog voltage from the accelerometer.

Figure 3: (left) Linear drop tower configured for pre-test system verification with spherical impacter and MEP; (right) drop tower configuration with medium Mg headform installed, in preparation for helmet testing.

Filtered analog voltages are sampled at 100 kHz using National Instruments© hardware (PXI62151 mDAQ) and software (LabView version 8). Pre-impact velocity is captured in the last 30 mm of headform (or impacter) drop using the same National Instruments© equipment. During impact tests, high speed video (1000 frames per second) were collected using a FASTEC monochrome high speed camera (TroubleShooter, TS1000ME). Following testing, all data were filtered per j211 Channel Frequency Class 1000 (low pass filter, 4th order, corner frequency 1,650 Hz) in MATLAB©.

General test procedures: pre-test system verification and post-test verification check.

Before helmet testing commenced, drop tower function and instrumentation were verified. The spherical impacter was dropped at nominal 5.4 m/s (range 5.38 m/s to 5.43 m/s, as measured in last 40 mm) three times, and the resulting peak linear acceleration was on average 389 g (range 388 g to 390 g). Post-helmet testing, the verification was repeated and conformance with ASTM was confirmed.

General test procedures: baseline helmet testing and DCR retrofit helmet testing.

All impact velocities were 4.5 m/s or greater. All helmets were fitted to the Mg headform and adjusted to provide snug fit. The Mg headform was adjusted to impact the DCR (Figure 4). Owing to the installation location of the DCR (as supplied), the impact site on the DCR when translated to the headform fell below the “rear” location specified in ASTM (intersection of median and reference planes). As a result, in baseline helmet performance testing, the baseline helmet 1 was impacted at a site corresponding to the DCR location (but with the DCR absent) twice, and then subsequently a third time at a location corresponding to the rear position on the headform (within the test zone, again with DCR absent, Figure 4). The single impact at the “rear” test location was performed so that the baseline performance of the helmet (within the test zone) was known, as well as the baseline performance below the test zone. Therefore, the baseline helmet 1 was impacted three times. All other helmets, 2 through 10, were impacted once at the location of DCR carbon fibre insert. Only one impact was performed for DCR helmets because the carbon fibre inserts in the DCR fracture as a result of the impact.

FIgure 4
Figure 4: DCR helmet adjusted pre-impact. Inset: red circle – rear impact location, yellow circle- approximate impact location covered by carbon fibre insert in DCR.

Data Processing and Results:

Figure 5 shows typical results compiled for each helmet drop. As shown, the acceleration versus time data was used to determine peak linear headform acceleration (peak g) and the timeframe of the primary acceleration corresponding to the helmet impact with anvil (delta t). Delta t was determined as the time difference between the first acceleration which exceeded 3 g and the final acceleration before the acceleration was smaller than 3 g. To determine headform velocity throughout the impact, acceleration versus time data was integrated using a simple forward integration algorithm (MATLAB©). From this velocity calculation, the maximum change in linear velocity of the headform was determined. High speed video data was not analyzed for kinematics. Images from high speed video are included to convey deformation of DCR exterior during impact.

FIgure 5
Figure 5: Exemplar acceleration and velocity versus time results (helmet 2). Top left: first acceleration peak corresponding to first impact of helmet with anvil; Top right: entire acceleration versus time profile of 1 second data acquisition window; Bottom left: linear headform velocity associated with timeframe of first impact; Bottom right: linear headform velocity for 1 second data acquisition window.

4.0 Results

4.1 System Verification

Tables 1 and 2 show results from system verification drops. Pre- and post-helmet testing indicates the drop tower and instrumentation meet standard specifications.

Table 1: Pre-helmet test results for system verification.
Test Condition             Drop height (m)          Impact velocity (m/s)      Peak acceleration (g) System verification            1.58                                       5.42                                           388
System verification            1.58                                       5.43                                           388
System verification            1.58                                       5.38                                           390

Table 2: Post-helmet test results for system verification.
Test Condition              Drop height (m)        Impact velocity (m/s)        Peak acceleration (g) System verification             1.58                                     5.47                                            391
System verification             1.58                                     5.49                                            392
System verification             1.58                                     5.48                                            389

4.2 Impact Attenuation of Baseline (Helmet 1) and DCR Helmets (2 through 10)

Table 3 shows a summary of helmet mass, acceleration, velocity, and delta t for all drops performed in this preliminary testing. Note that Baseline tests 1, 2, and 3 were all performed using Helmet 1.

Helmets with the DCR had on average 57 grams more mass than Helmet 1.

Table 3: Summary helmet mass and impact data for all drop tests.

Test Condition             met mass (grams)      Drop  ht (m)    Impact vel (m/s)    Peak accel (g)    ΔV (m/s)    Δt(ms)
Baseline test 1***             487.3                               1.22                  4.85                             520                   6.97         7.1
Baseline test 2**                                                         1.22                  4.83                             272                   7.11        10.2
Baseline test 3***                                                       1.22                  4.81*                           360                   7.14         7.5
Helmet 2                              545.5                                1.22                  4.77                             118                   7.01        16.8
Helmet 3                              546.1                                1.22                  4.81                             117                   6.98        16.1
Helmet 4                              549.2                                1.22                  4.67                             114                   6.91        16.8
Helmet 5                              549.0                                1.22                  4.73                             107                   6.71        17.1
Helmet 6                              551.0                                1.22                  4.80                              117                  7.04        16.7
Helmet 7                              545.0                                1.22                  4.68                              108                  6.77        16.9
Helmet 8                              535.1                                1.22                      –                                 116                  7.02       16.6
Helmet 9                              538.6                                1.22                  4.82                              118                  7.00        15.9
Helmet 10                            543.0                                1.22                 4.72                               118                  7.01       16.7

* Velocity gate did not trip on this test, impact velocity determined from integrated acceleration data alone.
** This impact at rear location of test zone, intersection of midsagittal and reference planes
*** This impact at rear of helmet at location on DCR carbon fibre insert (below test line)

Impact velocity over all tests ranged from 4.7 to 4.9 m/s over all tests, with the exception of the test for Helmet 8 had an instrumentation error.

Maximum headform delta V ranged from 6.7 to 7.1 m/s over all tests. For Helmet 1, average delta V was 7.1 m/s, while for DCR helmets it averaged 6.9 m/s.

Delta t for Helmet 1 averaged 8.3 ms, while for DCR helmets it averaged 16.6 ms.

Figure 6 shows the peak headform acceleration for Helmet 1 (baseline) and the group of DCR helmets. The average baseline (Helmet 1) peak linear headform acceleration was 384 g. The average DCR helmet peak acceleration was 115 g (70% lower than 384 g for Helmet 1, Table 3 and Figure 6).

Figure 6
Figure 6: Bar chart comparing peak headform acceleration for Baseline (helmet 1) and DCR helmets. Note that Baseline helmet was impacted three times, once at rear location (red circle), and twice at head rear but below test line (yellow circle).

Figure 7 shows acceleration versus time for Helmet 1(Baseline 2) and Helmet 5. To highlight the contrast in acceleration versus time data for these helmets, the data for both helmets has been shifted in time such that the onset of acceleration increase for both helmets is approximately coincident (roughly 5 ms in Figure 7). In all cases and relative to Baseline Helmet 1, DCR helmets had reduced peak acceleration and increased delta t.

Figure 7
Figure 7: Acceleration versus time for Baseline 2 (Helmet 1) and Helmet 5 (with DCR).

All impact telemetry for all helmeted drops is plotted in Appendix A for reference. All telemetry data is also provided on the media accompanying this report.

4.3 High speed video

Figure 8 shows typical images captured from high speed video and typical deformation of the external DCR structures during impact. The touchdown of the DCR on the anvil was estimated from visual inspection of the high speed video (DCR touch) and, as shown, 17 ms post touchdown the helmet and DCR are rebounding (onset of DCR losing contact with anvil). In all impacts, the high speed video indicated that the DCR remained intact (externally visual surfaces) and fixed at its pre-impact location on the helmet rear. No DCRs detached from the helmets in any of the drop tests performed. High speed videos for all drop tests can be found on the media accompanying this report.

Inspection of the carbon fibre inserts post impact indicated plastic deformation, in most cases it was evident that the carbon fibre insert had collapsed. Manual palpation of the insert revealed that the, initially rigid pre-impact, insert no longer had the structural integrity to support external loads. With modest palpation forces, the insert could be easily compressed. Visually, the impacted DCRs did not show any obvious signs of damage and looked identical to the DCR pre- impact. No obvious damage to the helmet outer shell was evident around or beneath the DCR.

Figure 8
Figure 8: Images captured from high speed video. Pre- denotes pre-DCR touchdown while post indicates post touchdown on steel anvil.

5.0 Discussion
Dynamic Composites Inc., through TECEdmonton, contracted impact testing of a prototype helmet retrofit (termed DCR) to the University of Alberta, Department of Mechanical Engineering. The retrofit comprises a foam layer that can be fixed to external helmet surfaces, at the approximate center of which is a collapsible carbon fibre insert that is designed to plastically fail, and therefore dissipate energy, throughout a rear helmet impact.

In general, the impact testing data indicates that the DCR reduces peak headform acceleration by 70% when compared to the baseline hockey helmet (without a DCR). It is controversial to assert any claims of improved head protection, or reductions in risk of head/brain injury based solely on peak acceleration data. However, all factors being equal, reduced peak head accelerations in impacts with timescales typical of helmet impacts are generally associated with lower risk of brain injury [10]. The range over which accelerations were decreased would suggest that in this preliminary work and for the specific linear impacts presented, the DCRs were associated with a reduced risk of brain injury.

DCR helmets effectively increased the primary acceleration pulse and decreased peak accelerations in the 1.22 m drops performed here which represent severe head impact. The 70% reductions in peak acceleration suggest that the DCR effectively manages the transfer of impact energy to the headform in the impact scenario of this preliminary work. Added thickness, equivalent to the DCR thickness, increases the distance over which helmet structures can deform (termed ride-down) and can be associated with reduced peak accelerations but longer delta t of the acceleration versus time history of the primary impact. This explanation is consistent with the measured data (Table 3) which for DCR helmets shows delta t approximately 100% greater than Baseline helmets but peak acceleration 70% lower than Baseline helmets.

Presence of the DCR did not result in obvious damage to the helmet’s exterior hard plastic shell and this fact could be viewed as promising in the event that Dynamic Composites Inc. continues toward commercialization because standards organizations typically require that any helmet retrofit not interfere with a helmet’s ability to remain intact (both shell and liner) during impact.

Overall, this preliminary testing suggests that the prototype DCR meets the design objective of reducing peak headform accelerations in rear helmet impact, while not interfering with or damaging the helmet outer shell.

6.0 Conclusions
The impact testing results indicate that the DCR is capable of reducing peak headform accelerations in rear helmet impacts and further that the presence of the DCR in an impact does not result in helmet damage. Headforms in Bauer 4500 helmets fitted with a DCR and subjected to rear helmet impact on a steel anvil at nominally 4.8 m/s exhibited 70% lower peak accelerations than the same helmet impact without a retrofit. The maximum velocity change of the headform was approximately constant (7 m/s) across all helmet tests. The average time duration of the headform accelerations associated with the primary helmet impact was 8.3 ms for helmets without a retrofit and 16.6 ms for helmets with a retrofit.

7.0 References
[1] D. H. Daneshvar, C. J. Nowinski, A. C. McKee, and R. C. Cantu, “The Epidemiology of
Sport-Related Concussion,” Clin. Sports Med., vol. 30, no. 1, pp. 1–17, Jan. 2011.
[2] K. Flik, S. Lyman, and R. G. Marx, “American collegiate men’s ice hockey: an analysis of injuries,” Am. J. Sports Med., vol. 33, no. 2, pp. 183–187, Feb. 2005.
[3] D. Richards, M. Carhart, I. Scher, R. Thomas, and N. Hurlen, “Head Kinematics During
Experimental Snowboard Falls: Implications for Snow Helmet Standards,” Ski. Trauma Saf. 17 Th Vol. 1510, 2009.
[4] I. Scher, D. Richards, and M. Carhart, “Head injury in snowboarding: evaluating the protective role of helmets.,” J. ASTM Int. JAI, vol. 3, no. 4, 2006.
[5] H. Nakaguchi and K. Tsutsumi, “Mechanisms of snowboarding-related severe head injury: shear strain induced by the opposite-edge phenomenon,” J. Neurosurg., vol. 97, no. 3, pp. 542–548, Sep. 2002.
[6] L. L. Brainard, J. G. Beckwith, J. J. Chu, J. J. Crisco, T. W. Mcallister, A.-C. Duhaime, A. C.
Maerlender, and R. M. Greenwald, “Gender Differences in Head Impacts Sustained by
Collegiate Ice Hockey Players,” Med. Sci. Sports Exerc., vol. 44, no. 2, pp. 297–304, Feb.
[7] J. T. Gwin, J. J. Chu, T. A. McAllister, and R. M. Greenwald, “In situ measures of head impact acceleration in NCAA Division I Men’s Ice Hockey: implications for ASTM F1045 and other ice hockey helmet standards,” J ASTM Int, vol. 6, no. 6, pp. 1–10, 2009.
[8] “CSA Z262.1-09: Standard for Ice Hockey Helmets.” CAN CSA, 01-May-2012.
[9] “ASTM F1045-07: Standard Performance Specification for Ice Hockey Helmets.” .
[10] L. Zhang, K. H. Yang, and A. I. King, “A Proposed Injury Threshold for Mild Traumatic
Brain Injury,” J. Biomech. Eng., vol. 126, no. 2, p. 226, 2004.