JNIOSH

Abstract of Special Research Report (RR-2004)

National Institute of Occupational Safety and Health, Japan

An Inquiry into Basic Safety Requirements of Field-bus Network for Safety Control

RR-2004-1
Tsuyoshi SAITO, Hideo NAKAMURA and Masaki MIURA

: In the bottom layer of a hierarchized computer network for manufacturing in a factory, serial communication networks, in which sensors, actuators and logic controllers are connected to a single wire (the bus) that has two endpoints, are collectively termed "Field Buses". Recently, application of the field bus technology to transmission of safety-related information from protective sensing devices for human safety to control parts of industrial machinery shows rapid progress. However, dangerous transmission errors on field buses and preventive measures against them have still not been investigated enough. The safety-related information transmission requires special considerations to ensure the promptness and inerrancy as well as the conventional hard-wired failsafe interlock system.
    This paper deals with the derivation of basic safety requirements of field bus network to transmit safety-related information (Safety Field Bus : SFB) and the experimental verification of safety performance of the SFB. From the perspective of the conventional fail-safe theories and requirements of international standards concerning safety of machinery, the four requirements are showed for the redundant architecture of processing interface controller, the cyclic signal transmission to confirm the normalcy of communication function and the double error detection coding of telegrams. In order to establish a concrete example of SFB system, a test model of safety control device capable of connected to SFB is experimentally produced by using a single board 16 bit microcomputer and a prepared SFB controller chip based on the controller area network (CAN) bus. Using the experimental SFB system including the test model, its response time to detect dummy hardware failures and to perform a designated safety function are measuared. From the results of the measurements, it is confirmed that the experimental SFB system has enough potential to achieve a higher safety integrity level than the level described on the associated international standards, however the importance of improvement in its slow response time was revealed.

Method for Characterizing Fracture Surface by using Two-Dimensional Local Hurst Exponent and its Application to the Quantative Evaluation of Stretched-Zone Width and Estimation of JIc

RR-2004-2
Kenta YAMAGIWA and Takashi HONDA

: Fractal analyses have been widely used to characterize the fracture surface. It has been recognized that the local Hurst exponent, which is based on the concept of self-affine fractal, is useful to detect the transition point of fracture surface. For the calculation of the local Hurst exponent, a high resolution profile is needed. To measure the profile, however, much time and effort are needed. Therefore, it is difficult to calculate all profiles of fracture surface and evaluate the feature of fracture surface in detail.
    In this study, a new method to calculate the two-dimensional local Hurst exponent is proposed.
It is realized to evaluate the feature of fracture surface using the local Hurst exponent. To investigate the validity of the two-dimensional local Hurst exponent, the calculation was applied to the grayscaled images in which the stretched zone was observed and the width of stretched zone (SZWc ) was measured. Consequently, SZWc calculated by the two-dimensional local Hurst exponent and detected by the human observation have the good agreement. Therefore, it was found that the twodimensional local Hurst exponent is useful to detect the transition point of fracture surface.

Study on the Mechanism of Generating Lateral Earth Pressures using Centrifuge with Movable Earth Support Equipmen

RR-2004-3
Yasuo TOYASAWA, Kazuya ITOH and Surendra B. TAMRAKAR

: To prevent the accidental collapse of earth supports in construction work, knowledge about collapse mechanisms, collapse precursors, and so forth are needed. However, the interaction between the generation of lateral earth pressures, deformation of soil, movement of retaining walls and the behavior that leads to collapse are very complicated. Therefore, it is difficult to predict accurately whether the observed movement at a construction site will result in a collapse, or whether the movement will stop. Further research is required to take account of wall movement, deformation in the soil, and the phenomena that lead to collapse. This research investigates the interaction between the generation of earth pressures and soil deformation due to wall movement. Centrifuge experiments were carried out using the newly developed movable earth support equipment, which controls the wall movements with high precision in a 50g centrifugal field. For the analysis of soil deformation, an image-processing system was utilized.
    The following conclusions were obtained.
  (1) The lateral earth pressure decreased with the wall movement on the active side in every test, however, the redistribution of earth pressure, displacement of the model ground, and development of strains varied depending on the mode of wall movement.
  (2) Until the movement of the wall reached X / L = 0.01, the activated area propagated behind the wall in a wedge shape. After the rupture surface appeared (analyzed by image processing), the deformation area was limited within the rupture surface and wall.
  (3) For rotation about the top, translation and center swelling (Crank mode) cases, the lateral earth pressure on the upper part of the wall was close to the line of earth pressure at rest and the lateral earth pressure on the lower part of the wall was smaller than Rankine's active earth pressure at around X / L = 0.02, so it is assumed that the arching action has a significant effect on the redistribution of earth pressure.
  (4) In the case of rotation about the base, it is supposed that the active state started from around X / L = 0.01.
  (5) The calculated earth pressures by using Rankine theory are almost same as the measured earth pressures.

Tensile Strength of Soil Measured using Newly Developed Tensile Strength Apparatus

RR-2004-4
Surendra B. TAMRAKAR, Yasuo TOYASAWA and Kazuya ITOH

: Recently, many soil-slopes get failed during the excavation or trimming process of slopes. In case of circular slope failure, it is generally thought that the failure is occurred due to shearing of soil layers. In such case, shearing surfaces are also seen at the failure sites. But in some failure excavation sites, such types of shearing surfaces are not seen. Instead soil layers seem to be detached from each other. The cause of such failure could be due to the development of tensile cracks. Also, many earth dams, embankments, pavements, etc. are failed due to the development of tensile cracks. In addition, location of the development of ice lens in freezing soils is also related with tensile strength. In order to understand the development of tensile crack, it is necessary to know the exact value of tensile strength that the particular soil consists of. Several methods of tensile testing of soils have been used in the past. But up to now, due to limitations of the existing test methods, the earlier tests were focused on to the more brittle and elastic materials (stiff, compacted and cement mixed soils) which have higher tensile strengths rather than for ductile (soft, saturated and clayey soils) materials which have lower tensile strength.
    This paper describes a newly developed tensile strength measuring apparatus which could be used for measuring the tensile strength of saturated, compacted and soft soils. At first, the tensile strength measurement procedure using this apparatus is explained which is very simple and quick. Then, two types of tensile molds are described and results obtained using them are compared. Statically compacted soilsGKanto loam, clay-sand mixture, clay-silt mixture and CFP silt-sand mixture, and one dimensionally consolidated NSF-clay are used as test samples. Unconfined compression test and suction test are also performed for Kanto loam to get their relationship with tensile strength.
  1) Maximum strengths for Kanto loam are obtained around 50 - 60% of water content for all the samples prepared at three different dry densities (ρd = 0.66, 0.68 and 0.7 g/cm3). In the dry side (water content lower than 60%), strengths are reduced with the decrease in water content whereas in the wet side (water content higher than 60%), strengths are decreased with the increase in water content. Difference in the strengths due to change in dry density at dry side is lower than that at wet side.
  2) For compacted Kanto loam, it is found that the average ratio of unconfined compression strength (q u) to tensile strength (q t) is around 12.5. But it varies differently with water content at dry and wet sides.
  3) The effect of the amount of finer particles and their size on tensile strength are also observed by changing the proportions in the mixtures in clay-silt, clay-sand and sand-silt. It could be observed that with the increase in the amount of finer particles, tensile strength increases. But with the increase in the size of finer particles, there is a reduction in tensile strength.
  4) In case of pre-consolidated saturated NSF clay, tensile and unconfined compression tests are performed. Saturated NSF clay samples are prepared under the pre-consolidation stresses of 100, 200 and 300kPa. The ratio of strength (q u / q t) obtained for this saturated NSF clay is around 6.

Experimental Study to Reduce the Damage in Both Quay Walls and River Dikes related to the Lateral Flow of the Ground

RR-2004-5
Satoshi TAMATE, Ikuo TOHATA and Tsuyoshi HONDA

: Liquefaction-induced flow causes severe damage to offshore structures and embankment of river dikes. It became important to assess safety of existing structures against a strong earthquake after Kobe earthquake 1995. Reinforcement of the ground is needed to ensure stability of the cranes on quay walls. Many researchers have studied mitigation techniques to reduce the damage. For example, installation of underground walls such as resistance piles, sheet piles and grouting was proposed to restrict ground flow behind quay walls and in subsoil of river dikes. This study aims to estimate the effects of underground walls on the reduction of ground movement. Centrifuge model tests of quay walls and river dikes were carried out. Additionally, laboratory shear tests and numerical analysis with distinct element method were conducted to investigate deformation mechanism.
    Two series of model test of quay walls were carried out by using two types of soil. In the first series, coarse sand with relatively high liquefaction resistance was used and resistance piles were adopted to mitigate ground flow behind quay walls. In the second series, Toyoura sand with the relative density of 40% was used and models with sheet piles or grouted soil were tested for the purpose of mitigating movement of quay walls. It was shown that resistance piles could restrict ground flow behind quay walls in the first series, and that the movement of quay walls could not be effectively restricted against a strong shaking in the second series.
    In the tests of river dikes, sandy silt which was collected from liquefied area in Western Tottori Earthquake (2002) was used as liquefiable material. The characteristics of this soil are high fines content and low permeability. Three models were tested. One is a model without mitigation measure. The others are models of sheet piles with drainage function or grouted soil walls. Test results showed that mitigation by sheet piles are most efficient in reducing the lateral flow. The results obtained from numerical analysis with distinct element method were similar to those of model tests. Subsidence of river dike was reduced in both of mitigation model with sheet piles and grouted soil walls.

Study on Earthquake Resistibility of Container Cranes and Jib Cranes

RR-2004-6
Seiji TAKANASHI and Yasumichi HINO

: The Hyogoken-nambu earthquake, which occurred on January 17, 1995, caused damage not only to buildings but also to a number of large cranes. The collapse of these large cranes did not cause human casualties because the earthquake occurred early in the morning. However, their collapse could have harmed many workers at the harbor. It was also pointed out that the problems with material-handling machinery caused delays in delivering relief goods such as medical products, foodstuffs, and water. In particular, transportation by ship was necessary because damage to roads was severe. This paper deals with the seismic response characteristics of container cranes and jib cranes. Damage observed in container cranes can be classified into two types. One type is damage due to opening of legs by the liquefaction of the ground. The other is buckling of crane legs due to rocking. We researched the damage caused by rocking of the crane with shaking table tests and numerical analysis simulations. We used a 1/15 crane model for the shaking table tests. We presumed that the marks of damage on the pier were due to the rocking of the cranes, and we confirmed that marks of damage were due to such behavior of the cranes by the shaking table tests. The crane response acceleration that can be used as an index for seismic intensity in the design was greatest with the maximum input acceleration of 8.5 m/s2. We found that the response acceleration remained constant even if the input acceleration increased further. The maximum axial force in the legs showed similar characteristics, but the maximum bending moment in the legs grew larger as the input acceleration grew. As we were sure that test results could be obtained through numerical analysis simulation, we carried out simulation analyses using seismic waves with various characteristics. We obtained the same results as from the shaking table tests regardless of seismic characteristics. Our test and analysis results show that the axial force and bending moment in the legs can be estimated without non-linear calculation. These stresses obtained by linear calculation give safer side results than the results obtained by non-linear calculation. In addition, we proved with numerical analysis simulations that seismic resistibility can be improved by lowering the container crane's center of gravity.
    The most serious damage to jib cranes was the falling of the part above into the roller path. We carried out shaking table tests and numerical analysis simulations in the same way as container cranes. We found through our tests and simulations that dynamic analyses are not necessary for jib cranes because of very high natural frequency. As a result, we concluded that the Construction Code for Cranes, which is simple and in use, is effective for jib crane design.

Development of Passive-Type Electrostatic Eliminator for Pneumatic Powder Transport

RR-2004-7
Tsutomu KODAMA, Mizuki YAMAGUMA, Teruo SUZUKI and Tomofumi MOGAMI

:In order to prevent a dust explosion caused by electrostatic discharges in a silo during loading of particulate products via a pneumatic transport system, we have developed a novel passivetype electrostatic charge eliminator to reduce the static charge on products before they enter a silo. The eliminator is installed at the end of the loading pipe inside the silo. It consists of a short length of plastic pipe and air-nozzle ionizers without a power supply mounted on it. Each ionizer is equipped with a grounded needle electrode within an insulated nozzle. Instead of a high-voltage power supply, the eliminator utilizes the electrostatic field strength resulting from the electrification of the plastic pipe as the energy source for the corona discharge at the needle electrode to produce air ions for reducing the static charge. The inside wall of the plastic pipe is electrified as particulate products pass through it. The charge elimination performance of the eliminators with plastic pipes of different materials was evaluated using a real-size experimental apparatus and 350-400 kg of polypropylene pellets. The eliminator used in the experiment consisted of a plastic pipe with an inner diameter of 10 cm and a length of 20 cm, as well as six air-nozzle ionizers arranged along the circumference of the plastic pipe. PTFE, PVC, PP, PE, and Nylon were used as the material for the plastic pipe. Two types of ionizers with different nozzles, i.e., insulated metal and PVC, were used. The air ions produced by corona discharges were blown inside the plastic pipe by compressed air at a pressure of 250-300 kPa to neutralize the static charge on the pellets within it. The experimental apparatus consisted of an SUS silo with a diameter of 1.5m and a capacity of 4.8m3, an SUS pipeline with a diameter of 10 cm and a length of approximately 20m, an air blower with an air volume of 10m3/min, and an air conditioning unit. Incendiary discharges, which occurred along the pellet-heap surface in the silo, were observed using a camera/CCD camera with an image intensifier set on the window of the silo roof. The charging tendencies of the pellets before and after elimination were evaluated using two air-blow-type electrostatic field sensors set on each sidewall of the silo and the pipe. It was found that effective elimination was achieved by electrifying the plastic pipe with the same polarity as the pellets and that the pipe was electrified due to the triboelectrification between the pipe and the pellets. The eliminator using PTFE for the pipe material was practically effective for reducing the static charge on the pellets.

An Inspection of Contact Failure of Connector on the Electric Circuit for Signal Use

RR-2004-8
Tatsuo MOTOYAMA, Hajime TOMITA and Kenji NAKATA

: One of the biggest weaknesses in an electric line is the connector. The connector for a signal transportation line use is particularly likely to lead to contact failure because of its low signal energy.
    On the other hand, the connector has the properties that the failure is apt to be recovered by direct applied testing signal, insert/withdraw action, etc.. Therefore, it is difficult to identify the cause of failure when recovery is accomplished by applying testing current, insert/withdraw action, etc. in spite of suspecting poor contacting at the connector.
    In this paper, a new method, which is a method for the indirect inspection of the contact resistance of a connector from the outside of an insulated wire, is reported. The principle of the method is to measure the impedance between a pair of electrodes formed by aluminum foil on insulated wires which are connected by a connector.
    In the method, capacity couplings are formed between the electrodes and the wire. And it is possible to show the capacity couplings by the equivalent circuit composed by R and C, approximately. And the possibility of the measurement by this method is shown by the analytic solution of the equivalent circuit.
    To begin with, the known value of resistive elements instead of the contact resistance of connector was measured by the method under the conditions of the 100kHz in applied signal frequency, 4V in applied voltage, and 5cm in width of electrode formed by aluminum foil on insulated wires. And the contact resistance of the aged connector was also measured. The results are as follows.
  (1) It is difficult to measure the contact resistance by the above method in case that the measured contact resistance is too small for the coupling impedance between the wire and the electrode formed by aluminum foil.
  (2) However, there is approximately the correlation between contact resistance value measured by the above method and the value by measuring instrument in the range from 1 kΩ to 1000 kΩ. And it seems that the inspection of the contact failure on connector is possible in above range.
    The main factors which the method depends are frequency, applied voltage in measuring signal and width of electrode formed by aluminum foil on insulated wires. So the measurement accuracy will be improved, if these factors are optimized. And, it is expected that the inspection range of the contact failure on connector enlarges more than above results.

Theoretical Evaluation of Electrostatic Properties of Thin Materials with Grounded Backing Conductor

RR-2004-9
Atsushi OHSAWA

:This paper derives analytical expressions for surface potential distributions on thin materials having a grounded backing conductor from using the model of a distributed resistor-capacitor network. In this paper, their steady-state and transient solutions for disk and rectangular thin materials are applied for determining the resistance to ground and charge relaxation, which characterize their antistatic properties.
    Thin materials having a grounded backing conductor are often found in industry, such as floors, sheets on desks or stocking shelves, liners and paints on metal surfaces, etc. The static charges of the materials themselves and an object on them (e.g., a human body on a floor) introduce electrostatic hazards and troubles in many fields of manufacturing industries. To prevent such hazards and troubles, many types of antistatic materials have been developed. However, there are sometimes problems that some of them are ineffective for reducing the charges. For example, materials, of which only surface resistance is as low as specified for the antistatic, cannot reduce the charge. To evaluate antistatic materials we usually measure their resistances and sometimes charge decay time is additionally measured in accordance with any standard method. The analytical solutions of resistance and charge decay, therefore, may be useful to discuss such measured results.
    Consequently, the analytial expressions demonstrate that both the resistance and charge relaxation as well as the surface potential distribution strongly depend on ρs / (ρvδ), that is, the ratio of surface resistance to volume resistance per unit square area, where δ : thickness of the material, ρs : surface resistivity, ρv : volume resistivity. Since the solutions are expressed by the functions of the surface and volume resistivities, permittivity and dimensions, the solutions are meaningful for evaluating and designing antistatic materials.

Measurement and Countermeasures of Induced Electromagnetic Field on a Rough-Terrain Crane due to Medium Frequency

RR-2004-10
Hajime TOMITA

: The sum length of the boom, jib, and wire of a large-scale crane may reach about 1/4 of the medium frequency wavelength. Such cranes may be effective medium frequency receiving antennas. Thus, the induced voltage at the hook of a crane located near a medium frequency broadcasting station may become high, which may result in electric shock to workers.
    To suppress hook voltage, a conductive wire is fixed along the boom of the crane and grounded via an inductor and a variable condenser, which change the resonance frequency. Alternatively, the height of the crane body may be changed to decrease its capacitance to alter the resonance characteristics. A feedback circuit to cancel the induced voltage at the hook has also been proposed.
    Recently, guidelines from ICNIRP and the Ministry of Internal Affairs and Communications have been proposed to protect the human body against the bioeffects of electromagnetic fields. To provide information to workers involved in working with cranes and to increase our understanding of the conditions in electromagnetic environments in typical crane work, the electromagnetic field around the hook of a rough-terrain crane in a high electric field of medium frequency was measured. The experiments indicated a field of more than 800 V/m in the vicinity of the hook. To mitigate this high electric field on the shop floor, a parallel resonance circuit was formed between the chassis of the crane and the ground to change the resonance characteristics, and the electric field in the vicinity of the hook decreased to 500 V/m.

Other Publications

Kiyose District Map

Umezono 1-4-6, Kiyose,
Tokyo 204-0024 Japan
TEL: +81-42-491-4512
FAX: +81-42-491-7846

Noborito District Map

Nagao 6-21-1, Tama-Ku,
Kawasaki 214-8585 Japan
TEL: +81-44-865-6111
FAX: +81-44-865-6124