ICE PRESSURE MEASURED DURING JOIA INDENTATION TESTS

Transcription

1 ICE PRESSURE MEASURED DURING JOIA INDENTATION TESTS Devinder S. Sodhi 1, T. Takeuchi 2, N. Nakazawa 3, S. Akagawa 4, H. Saeki 4 1 Retired from CRREL, Hanover, NH, USA 2 Hachinohe Institute of Technology, Hachinohe, Japan 3 Forest Works, Tokyo, Japan 4 Hokkaido University, Sapporo, Japan ABSTRACT The interfacial pressure was measured during medium-scale indentation tests by the use of tactile sensors. This paper describes the procedure of acquiring the tactile sensor data, and the organization of the data in the DVD containing the data. Ice force records are given to compare the force measured by load cells to those obtained by summing the interfacial pressure measured by tactile sensor. KEY WORDS Ice Pressure; Ice forces; Interfacial Pressure; Tactile Sensor; Ductile Crushing; Brittle Crushing. INTRODUCTION As part of a five-year program sponsored by the Japan Ocean Industries Association (JOIA), medium-scale indentation tests were conducted with sea ice in the harbor of Lake Notoro, Hokkaido, from 1996 to Sodhi et al. (1998) describe these tests, and Sodhi et al. (2001) present a few results from the measurements of interfacial pressure using tactile (TEKSCAN) sensors. JOIA has developed a DVD to make the tactile sensor data available to researchers and other users. This DVD contains most of the data from tactile sensors that were collected in 1998, 1999 and This paper describes the experimental setup to acquire the tactile sensor data and the organization of the data in different folders and files of the data DVD. Besides two README (MSWord DOC and PDF) files, there are three folders and one EXCEL file in this DVD. The Excel File has been copied from the JOIA CD_ROM, and contains the data -199-

2 on ice conditions and other relevant information for all tests conducted from 1996 to The data from tests, during which tactile sensor data were acquired, are in three folders marked by the year: 1998, 1999 and There is a separate folder for each test. Some folders do not have tactile sensor data, which were either not in the JOIA_CD_ROM for Tactile Sensor Data or the files could not be read by the ISCAN software.separate folder for each test. The folder for each test has 7-8 files, including data from load cells, tactile sensors, and processed data with plots. The procedure to calibrate the tactile sensor data is to match the force measured by load cells with the raw sum from tactile sensor at a particular time during a test (Sodhi, et al. 2001). Because a tactile sensor covered more than one segment of an indentor, a proportional load from load cells is computed. For example, a 480-mm-wide tactile sensor was installed on the middle five segments of the 15-segment, 1500-mm-wide indentor during all tests in The tactile sensor covered 100% of the 7 th, 8 th and 9 th segment, and 90% of the 6 th and 10 th segment. Therefore, the force on the tactile sensor is assumed to be sum of 100% of the force measured by load cell 7, 8, and 9, and 90% of the force measured by load cell 6 and 10. In 1999, the indentor width was 600 mm for all tests, except during Test (After attending the REIFS Workshop in Mombetsu, a group of international visitors came to the test site and observed Test being performed.) For the 600-mm-wide indentor width, a simple 80% of the total load was assigned to be the load on 480-mm-wide tactile sensor. For indentation tests in 1998, Figure 1 shows the installation of four tactile sensors on a 1500-mm-wide indentor, and Table 1 gives the percentage of width covered by four 238-mm-wide tactile sensors (Sodhi et al. 1998). For Test , Figure 2 shows the installation of three tactile sensors, and Table 2 gives the percentage of width covered by three 480-mm-wide tactile sensors. The tests numbers for tests conducted in 1998 and 1999 begin with 98 and 99, respectively, whereas the tests number for tests conducted in 2000 begin with 01 and 02 corresponding to the month, January and February. Thus the tests numbers take the following form: 1998 Tests - 98mmdd-Test# 1999 Tests - 99mmdd-Test# 2000 Tests - mmddxxyy, where mm = 01 or 02 stands for January or February, dd is for day of the month on which that particular test was performed, Test# is the test number for that day, xx is the stiffness of the spring used for the test (MN/m), and yy is the indentation speed (mm/s). A folder for a test contains the following data: (a) forces measured by load cells on each 100-mm-wide segment (kn) either in text or EXCEL format, (b) contact pressure measured by tactile sensor(s) (MPa), (c) files having cal, fbx and fsx extensions can only be read by ISCAN software installed on computers with Windows 98 operating system, and (d) comparison of forces measured by load cells and tactile sensor(s) after calibration at one point in time (EXCEL spreadsheets and plots). Large text files contain contact pressure measured by tactile sensor(s). The first 21 lines contain the header information of that file and give a description of the measurement made with one or more tactile sensor. The contact pressure is arranged in rows and columns given by the 5 th and 6 th lines of the header information, and this constitutes a frame of data on contact pressure in MPa. The number of frames is given in the 19 th line of the header information. The row spacing, column spacing and the area of each pressure-sensing grid is given in the 7 th, 8 th and 9 th line of the header information

3 Figure 1. Installation of four tactile sensors during indentation tests in 1998 Table 1. Percentage coverage of 100-mm-wide segments of indentor by tactile sensors in Segment Column in Excel sheet for load cell Percentage Coverage of Segment by Tactile Sensor data 1 A B C D 1 D 2 E 3 F 26% 4 G 100% 5 H 100% 6 I 12% 88% 7 J 100% 8 K 50% 50% 9 L 100% 10 M 88% 12% 11 N 100% 12 O 100% 13 P 26% 14 Q 15 R 1 Information in this column is not relevant to the discussion here. Notes Each segment is 100 mm wide. Each tactile sensor is 238 mm wide and 238 mm high. A coverage of 26% of Segment 3 by Tactile Sensor A means that the tactile sensor covers a width of 26 mm of Segment

4 Figure 2 Installation of tactile sensors during Test Table 2. Percentage coverage of segments by tactile sensor during Test Segment Column in Excel sheet Percentage Coverage of Segment by Tactile Sensor Notes for load cell A B C D data 1 1 C 90% 2 D 100% 3 E 100% 4 F 100% 5 G 90% 6 J 70% 7 N 100% 8 R 100% 9 V 100% 10 Z 100% 11 AB 10% 90% 12 AC 100% 13 AD 100% 14 AE 100% 15 AF 90% 1 Information in this column is not relevant to the discussion here. Each segment is 100 mm wide. Each tactile sensor is 480 mm wide and 440 mm high. A coverage of 90% of Segment 1 by Tactile Sensor A means that the tactile sensor covers a width of 90 mm of Segment 1. 30% of segment 6 and 10% of segments 1, 5, and 15 were not covered by tactile sensors

5 In 1998, four tactile sensors, each having 44 rows and 44 columns of grid points, were attached to the face of an indentor. However, the data were acquired over only one tactile sensor in some of the tests. If only one tactile sensor was used for a test, the data in each row is the contact pressure across the width of the indentor, and the data in each column is the contact pressure on points vertically up and down. If more than one tactile sensor was used for a test (as indicated by number of rows being more than number of column), the data in each row are for points vertically up and down on the indentor, and the data arranged in each column are for points across the width of the indentor. During all tests conducted in 1999 (except ) and 2000, only one tactile sensor having 48 columns and 44 rows was used to measure contact pressure. Thus each row of data in a frame gives contact pressure across the width of the indentor, and each column vertically up and down. During Test , the contact pressure was measured using four tactile sensors, each having 48 columns and 44 rows. Of the four tactile sensors, only three were installed on the 1500-mm-wide indentor, and the fourth sensor was attached to the data acquisition system, which required four sensors to be installed. Figure 2 shows the installation of three sensors during Test It should be noted that sensor A was installed in reverse position to avoid damage to lead wires and connecting handle. The data on contact pressure measured during Test are arranged in 44 columns and 192 (=4x48) rows for each frame of 1374 frames. The data in each row are the contact pressure on grid points spaced vertically up and down, whereas the data in each column is for grid points spaced horizontally across the indentor. Therefore, the first row contains the data for the first column of points of the first tactile sensor, starting at a distance of 485 mm from the left side of the indentor while viewing it from the front (Figure 2), and the 48 th row is for points at a distance of 15 mm from the left edge. Progressing towards the right, the 49 th and 96 th rows contain the data for the first and 48th column of the second sensor, and the 97 th and the 144 th rows contain the data for the first and 48 th column of grid points on the third sensor (see notes under Figure 2). The remaining rows ( ) do not contain any meaningful data, because the fourth sensor was not installed on the indentor but simply attached to the data acquisition system. RESULTS AND CONCLUSION Sodhi et al. (1998, 2001) have presented some results from the measurements of interfacial pressure by tactile sensors. The calibration of tactile sensors is done by summing the pressure (or the force at each grid point) measured by the tactile sensor and matching it with the forces measured by load cells at one point in time. Figures 3-5 show plots of the comparison of ice force records as measured by the load cells and the tactile sensor for the full duration of three indentation tests ( , , ), conducted against a 287-mm-thick ice sheet on 21 January 1998 at speeds of 3, 30 and 0.3 mm s -1, respectively. They also show plots of the actual contact area and the contact area ratio, which is defined as the ratio of contact area divided by the nominal contact area (ice thickness x sensor width). During Test , interfacial pressure was measured using four 238-mm-wide tactile sensors, whereas only one 238-mm-wide tactile sensor was active to measure the pressure during tests and In test , a first-peak-force event takes place because of the high (60%) contact area ratio (Fig. 3). During test , the high indentation speed of 30 mm s -1 results in brittle crushing, as indicated by non-simultaneous, line-like contact in the pressure plots (not given here) and low (<20%) contact area ratio (Fig. 4). In test , the ice contact area expanded vertically during the low (0.3 mm s -1 ) indentation speed test, resulting in ductile crushing of ice, and the contact area ratio -203-

6 increased to a value of 55%, which is close to the value of 60% during the first-peak-force event. Furthermore, the maximum force during the tests and are in the range of kn. The maximum force during brittle crushing in test is in the range of kn. It appears that both the maximum force and the contact area ratio decrease during brittle crushing by a factor of about 3 to that for ductile crushing. Figure 3. Time-history plots of the ice force records and the contact area ratio

7 Figure 4. Time-history plots of the ice force records and the contact area ratio

8 REFERENCES Figure 5. Time-history plots of the ice force records and the contact area ratio. Sodhi, D. S., Takeuchi, T., Nakazawa, N., Akagawa, S., Saeki, H. (1998) - Medium-scale indentation tests on sea ice at various speeds. Cold Regions Science and Technology, 28: D. S. Sodhi, D. S., T. Takeuchi, M. Kawamura, N. Nakazawa, and S. Akagawa (2001) Measurement of ice forces and interfacial pressure during medium-scale indentation tests. Proceedings, 16th International Conference on Port and Ocean Engineering under Arctic Conditions, Ottawa, Canada, Vol. 2, pp