Available online at www.sciencedirect.com Procedia Engineering Procedia Engineering 2 (2010) 002605 2611 (2009) 000 000 Procedia Engineering www.elsevier.com/locate/procedia 8 th Conference of the International Sports Engineering Association (ISEA) Experimental installation for evaluation of snowboard simulated onsnow dynamic performance Aleksandar Subic a *, Patrick Clifton a, Joe Kovacs a, Francois Mylonas a and Adrien Rochefort a a RMIT University, Plenty Road, Bundoora, VIC 3083, Australia Received 31 January 2010; revised 7 March 2010; accepted 21 March 2010 Abstract The paper presents an original dynamic experimental installation that has been designed and developed at RMIT University for the evaluation of snowboard simulated on-snow performance. The dynamic rig comprises two independent pneumatic cylinders to replicate rider foot movement, as well as three independent adjustable pressure airbags used to simulate the snowboarding terrain by varying their pressure according to the change in the terrain configuration. A custom designed, PLC control system allows for the simulation of common snowboard manoeuvres, from turns of varying board inclination, to nose or tail presses, board slides, and finally the performance of jumps. The PLC programs can be modified or re-written to simulate a variety of terrain features of interest to the investigation. Stress-strain characteristics of the snowboard tested under load are determined using a comprehensive strain gauge system. The paper presents in detail the design of the dynamic experimental installation, rig calibration and testing protocols. c 2009 2010 Published by Elsevier Ltd. Open access under CC BY-NC-ND license. Keywords: Snowboard; dynamic performance. 1. Introduction The on-snow dynamic performance of modern snowboards has proven thus far to be difficult to measure accurately and objectively. The replication of on-snow conditions in the laboratory or with an analytical model, including snow forces and board vibrations, rider inputs and ambient temperature is a difficult proposition. Several analytical models to date have addressed the effect of such issues on overall snowboard dynamic performance [1, 2, 3, 4]. However, only Foss and Glenne [5, 6], and Sutton [7] have attempted to reproduce on-snow loading conditions in the laboratory for dynamic snowboard tests, whilst only Clifton et al. [8] have considered the effects of on-snow temperature on static snowboard properties. * Corresponding author. Tel.: +613-9925-6000; fax: +613-9925-6108. E-mail address: aleksandar.subic@rmit.edu.au. 1877-7058 c 2010 Published by Elsevier Ltd. doi:10.1016/j.proeng.2010.04.039 Open access under CC BY-NC-ND license.
2606 A. Subic et al. / Procedia Engineering 2 (2010) 2605 2611 A.Subic et al. / Procedia Engineering 00 (2010) 000 000 2 Furthermore, obtaining objective measurements on-snow is also problematic due to the difficulties associated with dynamic data collection setups (such as signal noise and system portability), fluctuating weather conditions and the accurate reproduction of snowboard manoeuvres by the rider. Brennan [9] validated a speed prediction mathematical model using a multi-camera/gps system, to obtain position, velocity and roll angle data for a standard slalom course. However the system setup was complex and laborious, whilst the rider was required to carry approximately 15 kg of instrumentation during the tests. Buckingham and Blackford [10] used piezoelectric sensors mounted on a snowboard to compare dry-slope vibrational outputs of different riders during basic turns. The data logging system was reported as highly portable, allowing minimal interference to the rider during manoeuvres. Whilst frequency spectra outputs were able to be used as a rider coaching tool, the system would be less suitable for comparisons of dynamic properties between snowboard models due to the aforementioned test repeatability issues. Finally, several prior studies have conducted subjective analyses of dynamic snowboard performance as an alternative to the problematic objective tests. Subic et al. [11, 12] characterised the on-snow performance of modern snowboards using nine subjective parameters, then correlated these parameters to objective design attributes to identify the key aspects of designs across the major riding styles. Darques et al. [13] also described a similar method for the design of skis and snowboards, but only applied the process to modern skis. The aim of the experimental installation described in this paper was to replicate on-snow snowboard loading conditions to the maximum extent possible, allowing reproducible, accurate and objective dynamic tests on different snowboard models to be undertaken. Such tests would facilitate the determination of objective on-snow dynamic properties (strain or vibration/damping) to aid in the design and analysis of modern snowboards, and secondly provide a means for conducting performance comparisons of current snowboards on the market. Whilst similar in aims to the rig manufactured by Bucknell University (described by Sutton [7]), their installation did not have terrain control, only actuators mimicking rider leg and ankle movements. The paper describes in detail the design of the installation, along with the calibration process and testing protocols. 2. Experimental installation design The experimental installation design is shown in Figure 1(a) and (b). Rider foot movement was replicated using two double-acting independent pneumatic cylinders (Pneutec model CXP0500060 [14]), capable of imparting snowboard edging angles up to ±17 (angles manually set using threaded stoppers). The foot actuation system contains two cylindrical joints and associated degrees of freedom; one proximal joint in the Y direction, one distal in the X direction and the Z direction stroke (shown in Figure 1(a)). The installation also possesses three independent dynamic pressure controlled airbags (Firestone Airide Springs, model 22D [15, 16]) used to replicate a varying snow surface. Each airbag is capable of applying a variable force up to approximately 500 N (in the Z direction), and the locations of the airbags are fully adjustable (laterally and longitudinally) to ensure simulation freedom. Thus any desired loading conditions can be applied to the snowboard tested to simulate any on-snow manoeuvre. The full dynamic system is enclosed by a structural frame and plexi-glass to ensure the safety of the operator. A custom designed, PLC control system allows the simulation of common snowboard manoeuvres, from turns of varying board inclination, to nose or tail presses, board slides and finally the performance of jumps. The PLC output can be modified or re-written to simulate a variety of terrain features of interest to the investigation. The full electropneumatic system is illustrated by the schematic in Figure 2. To consider firstly the pneumatic system, the compressed air supply (85 psi) feeds air through a regulator into both the three airbags and two double-acting pneumatic cylinders (feet actuators), via MAC PPC93A proportional pressure controllers [17]. The controllers are reported as fast responding and accurate, due to constant feedback from downstream pressure sensors. Thus both the simulated feet and snow surface force inputs on the snowboard tested can be rapidly varied as desired, to replicate any series of on-snow manoeuvres. Conversely, the electrical system links the simulation program inputs to a processor (Fatek FBs-24MC [18]) which provides the input signals to the five respective pressure controllers. A subsystem also links the strain gauges fitted to the test snowboard to a DataTaker DT800 [19] for strain measurements via a standard Wheatstone bridge completion box.
A. Subic et al. / Procedia Engineering 2 (2010) 2605 2611 2607 A.Subic et al. / Procedia Engineering 00 (2010) 000 000 3 Figure 1: (a) CAD model (b) Manufactured installation Figure 2: Electro-pneumatic system diagram
2608 A. Subic et al. / Procedia Engineering 2 (2010) 2605 2611 A.Subic et al. / Procedia Engineering 00 (2010) 000 000 4 3. Calibration In order to fully calibrate the installation, the exact force output response of each airbag was required. A force measurement system consisting of a load cell on a rigid stand was implemented within the rig, between the upper structural frame member and each airbag in turn. The PLC input for each airbag was then varied incrementally and the force output measured. Three force measurements per input were taken and the results were averaged. The load cell system and calibration setup are shown in Figures 3(a)-(b), whilst the averaged results of the calibration are shown in Figure 4. Dynamic properties (natural frequency and damping ratio) of the airbags were available from manufacturer data sheets [15, 16], and hence were not examined in the validation process. Figure 3: (a) Force measurement system (b) Calibration setup Figure 4: Airbag calibration It was noted from the calibration results that the non-linear airbag responses were of similar nature, however there were notable differences in magnitude of the force outputs. The discrepancies were rationalised as a result of different piping lengths and hence pressure drops within the system, though without further pressure measurements
A. Subic et al. / Procedia Engineering 2 (2010) 2605 2611 2609 A.Subic et al. / Procedia Engineering 00 (2010) 000 000 5 taken within the pneumatic system, this cannot be confirmed. As a result of the different response profiles, the PLC inputs for any simulation conducted were scaled to ensure the same forces from each airbag were applied to the snowboard tested. The average standard deviation in force measurements for each airbag was very low however, between 1.2 and 2.9 N. Thus, despite the different airbag responses, the overall force control was highly accurate. 4. Testing protocols To fully evaluate the dynamic performance of each snowboard tested in the experimental installation, a comprehensive set of simulations were designed. They aimed to replicate standard snowboarding manoeuvres performed on-snow by the rider, who was assumed to weight 75 kg. Four basic manoeuvres were modelled, which are listed and described in Table 1. These manoeuvres were then combined to form a general snowboarding track, which is shown in Figure 5. The track design parameters of turn radii (R 1 R 4 ), jump height (H), slope angle (θ), rail length (L) and initial rider velocity (V 1 ) were all varied throughout the dynamic tests. Table 1. Manoeuvre descriptions Manoeuvre Turn Jump Press Slide Description The board is put on either the toe or heel edge by the rider and pressed into the snow, causing it to have a circular path of defined radius. For initiation of the turn, the rider transfers weight to the nose of the board on the relevant edge, whilst for turn exit, the opposite weight transfer occurs. The rider weight is transferred towards the tail of the snowboard as the nose and tail are lifted off the snow in succession. For landing, the entire board impacts the snow at the same time. The entire weight of the rider is placed on the nose or tail, where the opposite end of the board is raised off the snow. The weight of the rider is applied only to the body section of the snowboard (between the legs) as it travels across a thin structure (beam/rail). Figure 5: General track schematic Forces on the snowboard during the individual manoeuvres were determined using analytical dynamic force balances. For the turn simulations, arcs of various radius and speed were modelled, utilising the analysis conducted by Michaud and Duncumb [20] to calculate the total snow reaction force and snowboard inclination angle. The total reaction force was then split over the three airbags, with respective magnitudes dependent on the selected weight
2610 A. Subic et al. / Procedia Engineering 2 (2010) 2605 2611 A.Subic et al. / Procedia Engineering 00 (2010) 000 000 6 distribution. The nominal distribution utilised for the turns was an even division of reaction forces across the three airbags. For turn initiation however, the turning reaction forces are progressively applied from the nose to the tail (and the opposite for turn exit) using 0.1 second time lags between airbag inflations. The jump simulation was broken into two separate modelling phases, to characterise the jump itself and the associated landing. To model the board lifting off the snow for any jump, the distributed weight of the rider is progressively removed from the airbags (from nose to tail), again using 0.1 second time intervals. Basic energy and momentum theory (using a default impact time of 0.1 seconds) was then utilised to determine the landing impact force, which is equally distributed across the three airbags. The press and slide simulation models aimed to address the freestyle (terrain park) aspects of modern snowboarding. For the simulated press manoeuvre, the entire weight of the rider is applied to either the nose or tail of the snowboard tested using the shortest time scale possible for airbag inflation. It is noted that this simulated manoeuvre neglects the slight impact force present as the rider completes the press on any structure. Similarly, for the slide manoeuvre (performed on thin rigid structures), an identical simulation approach was utilised, except in this case only the centre airbag is inflated, applying the rider s entire weight to the body section of the snowboard tested. 5. Conclusion This paper described an original dynamic experimental installation that has been designed and developed at RMIT University for evaluation of snowboard simulated on-snow performance. The dynamic rig comprises two independent pneumatic cylinders to replicate rider foot movement, as well as three independent adjustable pressure airbags used to simulate the snowboarding terrain by varying their pressure according to the change in the terrain configuration. A custom designed, PLC control system allows for the simulation of common snowboard manoeuvres, from turns of varying board inclination, to nose or tail presses, board slides, and finally the performance of jumps. The experimental installation described allows reproducible, accurate and objective dynamic tests on different snowboard models for different terrains and ride conditions to be undertaken in a controlled laboratory environment. Such tests would facilitate the determination of objective on-snow dynamic properties to aid in the design and development of modern snowboards, and secondly provide a means for conducting performance comparisons and benchmarking of current snowboards on the market. References [1] Borsellino C, Calabrese L, Passari R, Valenza A. (2005) Study of Snowboard Sandwich Structures. In O. T. Thomson, E. Bozhevolnaya, & A. Lyckegaard (Ed.), Sandwich Structures 7: Advancing with Sandwich Structures and Materials. Aalborg. [2] Kawai S, Otani H, Sakata T. (2001) Dynamic response of a snowboard due to snowboarder's snowboard control. Proceedings of the 7th Japan International SAMPE Symposium on Information and Innovation in Composites Technology, (pp. 459-462). Tokyo, Japan. [3] Sakata T, Kawai S. (1995) Dynamic bending analysis of snowboards. Transactions of the Japan Society of Mechanical Engineers, Part C, 61 (584), 1469-1475. [4] Sakata T, Kawai S, Kawada F. (1996) Free Vibrations of a Snowboard. International Journal of Mechanical Sciences; 38 (6), 579-588. [5] Foss G, Glenne B. (2007) Reducing On-Snow Vibrations of Skis and Snowboards. Sound and Vibration; 41 (12), 22-27. [6] Glenne, B., DeRocco, A., & Foss, G. (1999). Ski and snowboard vibration. Sound and Vibration; 33 (1), 30-33. [7] Sutton E. (2000) Better Snowboards by Design. 2000 ASME International Mechanical Engineering, Orlando, Florida, November 5-10. [8] Clifton P, Subic A, Sato Y and Mouritz A (2009) Effects of temperature change on snowboard stiffness and camber properties. Sports Technology; in-press. [9] Brennan S. (2003) Modeling the Mechanical Characteristics and On-Snow Performance of Snowboards. http://www.brennansnowboards.com. [10] Buckingham M, Blackford J. (2004) Analysing snowboard mechanics. Snow Engineering V - Proceedings of the Fifth International Conference on Snow Engineering, (pp. 203-208). Davos, Switzerland.
A. Subic et al. / Procedia Engineering 2 (2010) 2605 2611 2611 A.Subic et al. / Procedia Engineering 00 (2010) 000 000 7 [11] Subic A, Clifton P, Beneyto-Ferre J. (2008) Identification of innovation opportunities for snowboard design through benchmarking. Sports Technology; 1(1): 65 75. [12] Subic A, Clifton P, Beneyto-Ferre J, LeFlohic A, Sato Y, Pichon V. (2009) Investigation of snowboard stiffness and camber characteristics for different riding styles. Sports Engineering; 11(2): 93 101. [13] Darques J, Carreira R, de la Mettrie A, Bruyant D. (2004) Quality Function Deployment A means for developing adequate skis and snowboards. The Engineering of Sport 5 - Conference Proceedings of The Engineering of Sport, (pp. 428-434). Davis, USA. [14] Pneutec. http://www.pneutech.com.au. [15] Firestone Industrial Products Co. Airide Springs model 22D datasheet. http://www.fsip.com/pdfs/industrial/datasheets/metric_pdf/22.pdf [16] Firestone Industrial Products Co. Airstroke/Airmount brochure. http://www.firestoneindustrial.com/pdfs/industrial/airstroke_airmount/masam_203_e.pdf. [17] MAC Valves. Pressure controller model PPC93A. http://www.macvalves.com/products/ppc/ppc93a.pdf [18] Fatek Automation Corp. PLC model FBs-24MC. http://www.fatek.com/download%20page/english/fatek2007_cat_enu.pdf [19] DataTaker. DT800 data logger. http://www.datataker.com/library/product_data_sheets_ts/dt800_ts0039c1.pdf [20] Michaud J, Duncumb I. (1999) Physics of a Snowboard Carved Turn http://www.bomberonline.com/bomber_files/articles/jackm8/jackm8.htm.