JNIOSH

Abstract of Special Research Report (RR-2000)

National Institute of Occupational Safety and Health, Japan

Modeling and Servo System of the Compliant Actuator Controlled by Field-Sensitive Fluid

RR-2000-1
Noboru SUGIMOTO and Tsuyoshi SAITO

: In order to develop such a practical human-symbiotic robot that works in collaboration with human-workers for heavy work, physically helps the aged and the disabled person, and gives a patient rehabilitation training, the risk reduction concerning with the excessive force, which are applied to a part of human body by the contact of the robot with its users, is the most important problem. In accordance with the procedure of ISO/CD 12100, positively utilizing of inherently safe actuators capable of limiting their force output and moving speed individually should be considered firstly as a safety measure against this problem. The utilization of the safe actuators with such limiting function, however, has been hard to introduce practically due to their difficulty with motion control and their limited choices of types.
    In this report, as a generalized technique to add such limiting function to any types of conventional actuators, it is examined to combine a functional clutch mechanism by means of a field-sensitive fluid, which responds to an applied electric or magnetic field with a change in its yield stress, with an actuator. These materials are commonly referred to electro-rheological (ER) or inagnetorheological (MR) fluids. This report shows results of experimental comparison of control performance of antagonistic rotary actuators using ER and MR fluids, which are so constructed that the torque of two servomotors rotating in the different directions opposite to each other is transmitted to a freely rotatable output cylinder through the clutch mechanisms.
    Furthermore, as a control method for the safe actuators, such a new positioning method that makes an active use of their limiting function, is proposed. This method stably achieves an extremely high precision positioning without any complicated algorithms under such a condition that the maximum torque and the rotational speed of the actuator is limited to secure the safe contact between human and the robot. From experimental results, it is shown that the proposed method can be applied to both of actuators using ER and MR fluids. Consequently, it is proved that ER and MR fluids are effective in introducing the limiting function to human-symbiotic robots.

Application of Finite Element Analysis to Residual Stress Measurement with Hole Drilling Method

RR-2000-2
Takashi HONDA

: Residual stress is the stress which exists in components and structures before application of any service or external load. It is known that residual stresses significantly influence the structural stability of components and structures. Therefore an accurate quantitative estimation of residual stresses is very important and affordable techniques for measurement of residual stresses are needed.
    The hole drilling method involves measuring the localized strain relieved by the drilling of a shallow blind in stressed materials. This method is a popular and widely used technique for measuring residual stresses because of its ease to use in many different applications, its limited damage to the specimen, and its general reliability.
    The hole drilling method has been investigated over the past sixty years. Consequently, much of research has been incorporated into the ASTM standard E837-85. The procedure for making residual stress from the measured strains in this standard is based on the theoretical stress distribution in an infinite plate with a through hole in a plane stress field. Therefore this method has been used so far in situation where the residual stress field is assumed not to vary with depth below the surface. However the residual stress fields usually vary with depth. In such case, the error of estimated residual stress increases if the calibration factors provided in ASTM standard are used to calculate the residual stresses. In this study, the relieved surface strains around a blind hole location caused by drilling in the biaxial stress field, the shear stress field and the non-uniform stress field were calculated by finite element analysis in order to investigate the effect of the stress variation with depth, and the analytical results were compared with experimental values.
    The main results obtained in this study are as follows :
  (1) Calibration factors calculated from the results of finite element analysis are in good agreement with experimental values.
  (2) The accurate surface stresses of components under linearly varying stress field with depth can be obtained by using the relationship between the stress gradient and strains measured by rosette gage.
  (3) A procedure to estimate the varying stress field with depth is proposed, and validity of this method is confirmed.

Measurement of Autoignition Temperatures of Flammable Liquids Using 1 Liter Round Vessel

RR-2000-3
Haruhiko ITAGAKI

: Autoignition temperatures of flammable liquids have been measured by many researchers. Many data by the ASTM method have been reported especially. However, a fire sometimes happened due to the autoignition of the flammable liquid when the flammable liquid leaked out though it was dealt with under the lower temperature than this measured data by the ASTM method. Because, autoignition temperature is not the value specific to the material, but it varies with the measuring device and the method. Actually, the data reported by a literature is often different from the data reported by another literature. Therefore, the flammable liquid may sometimes ignite under lower temperature than the measured data by the ASTM method.
    So, the author paid attention to the size of the test vessel for the measurement of autoignition temperature among the factors which influenced autoignition temperature, and the test vessel was expanded to the 1 liter round bottom flask. The autoignition temperatures of pentane and decane were lower 16°C and 18°C respectively than the measurement data by the ASTM method using the 200 ml erlenmeyer flask, if the 1 liter round bottom flask is used.
    It is considered that the heat balance of the accumulation and the dissipation decides the occurrence of the autoignition of gas and vapor by the same reasons as for solid and liquid. In other words, as the scale of the vessel becomes large, the autoignition temperature of gas is thought to decrease by the same reasons as the phenomenon that the autoignition temperature of solid and liquid decrease. Therefore, if the vessel is large, there is a suspicion in the safety as for accepting directly the reported data which was measured by using the small vessel.
    So, the heat conductive model was introduced to estimate the autoignition temperature in the larger vessel. The model assumed that the heat conduction to the vessel wall from the gas was rate-determining step for the heat conduction to the outside from the whole gas. Next, the heat conductive parameter which mainly expressed the ratio of the volume and the surface area of the vessel was introduced. Then, it was shown that the autoignition temperature which depended on the scale of the vessel could be estimated by using this heat conductive parameter.

Decomposition Flame Propagation Properties of Ozone/Oxygen Mixtures

RR-2000-4
Takaaki MIZUTANI and Hidenori MATSUI

: Ozone is the strong oxidizing agent which has been used for water treatment, pulp bleaching and others. Recently, along with development of ozone generator, high concentration of ozone up to 20 vol.% in oxygen has been able to be obtained. It is well known that ozone is rapidly decomposed into oxygen with flame propagation by some ignition sources. But the decomposing explosion properties and the flame propagation behaviour of high concentration ozone in a closed vessel are not sufficiently clear. It is important to clarify the decomposing property for safe handling of ozone.
    In this study, explosion properties such as explosion pressure, flame propagation behavior and the flame velocity of ozone/oxygen mixtures in the closed vessel were investigated. A stainless steel cylinder (106 mm i.d., 114 mm in height, 1.0 L in volume) and a stainless steel tube (25 mm i.d., 5.0 m length, 2.5 L in volume) were used for explosion vessels. The concentration of ozone was detected by ozone meters placed on both sides ahead of the inlet and behind the outlet of the explosion vessel. The electric spark was used for ignition source. Schlieren system was introduced to visualize the flame propagation in the cylindrical vessel.
    The experimental results of ozone up to 14 vol.% in oxygen using the cylindrical vessel were as follows :
  (1) The maximum decomposition explosion pressures and the pressure rising rates increased in proportion to the initial pressures up to 1.6 MPa in this experiments.
  (2) The lower explosion limit was 10.5 vol.% in this vessel.
  (3) The ratio of the maximum decomposition explosion pressures to initial pressures was less than 3.5.
  (4) KG values were less than 1.
  (5) The location of ignition sources had less effect on the maximum explosion pressures, but had much effect on the lower flame propagation limits.
  (6) At the ozone concentration near the lower flame propagation limit, a partial flame propagation was observed. And the mixture below the ignition point was unburned.
  (7) The shape of upward flame was a sphere (under the higher concentration) or a mushroom like (near the lower flame propagation limit). The shape of downward flame was a horizontal plane.
  (8) The flame velocities were less than 0.5 m/s.
    The experimental results of ozone up to 20 vol.% in oxygen using the tube vessel were as follows :
  (1) Flame propagation took place above 14 vol.% ozone.
  (2) Except for just after the ignition, acceleration of the flame was not observed.
  (3) The steady flame propagation with a velocity of less than 1 m/s and the maximum explosion pressure of less than twice the initial pressure were obtained.
  (4) Deflagration to detonation transition (DDT) was impossible under the conditions for the experiments. However, with higher ozone concentrations, faster flame propagation is predicted and DDT is likely.

Propagation Brush Discharge Occurring on Insulating Inner-Bag Used with Anti-Static FIBC and Its Countermeasures

RR-2000-5
Mizuki YAMAGUMA and Tsutomu KODAMA

: An energetic form of discharge called the propagating brush discharge (PBD) was studied in view of a hazard assessment during the operation of an anti-static flexible intermediate bulk container (FIBC) with a highly insulating inner-bag. The conditions for producing PBDs were investigated using various kinds of materials of FIBD and insulating films made of polyethyleneterephthalate. The magnitude of PBD greatly depended on the structure and the resistivity of the FIBC materials. A material having fine conductive threads for dissipating static-charges, for example, was possible to produce extremely energetic PBDs due to an ionizing process on its threads. A less conductive, dissipating rubber type, however, produced rather weak discharges. In all conditions experimented, PBDs easily exceeded the critical level above which flammable gas or vapor can be ignited. This means that any level of PBD has to be avoided in the atmosphere where flammable gas/vapor exists. Two countermeasures to prevent the PBD were tried. Pinholes on an insulating sheet mitigated the charge accumulation, and when the interval of the pinholes was 4 cm or less, PBDs were no longer observed. Giving conductivity to the surface of an insulating film was also found to be effective for the prevention of the PBD if the level of static charge was relatively small.

Fire Spread over Metal Dust Layer

RR-2000-6
Masaaki YASHIMA and Toei MATSUDA

: Light metal powders such as aluminum (Al) and magnesium (Mg) have been known as combustible powder to cause dust explosions in industries. With development of material science and manufacturing process, other fine powders of metals (Ti, Ta, Zr, Fe, alloys, etc.) are produced mainly or secondarily, then they will fall on the ground or equipment. In this condition of the powder (dust) accumulation, dust explosions or fires may occur when an ignition source is given. It has a tendency for metal powder to form dust layer rather than dust cloud before ignition. The spreading of fire over a dust layer is the primary phenomenon in dust fires.
    An experimental study has been conducted in order to clarify the mechanisms of the combustion zone spreading over a metal dust layer. Nine sample holders are used to examine the effects of layer configuration and depth on the fire spread. Six of them have flat-shaped layer in which the combustion area is 10 mm wide, 160 mm long and various depths (0.5, 1, 2, 3, 4 and 8 mm, respectively). The 2 mm depth flat-shaped holder is mainly used in the current study. Ti, Zircaloy-2 (Zr alloy), Ta, Mg and Mg/Al alloy powders are used, whose median particle diameters, D v,50 are between 18 and 80μm. Furthermore, the powder of PMMA (polymethylmethacrylate), whose sheet has been usually an object material in basic studies for fire spread, is used to compare the combustion behavior to those of the metal powders. To achieve the purpose of this study, we take into consider some factors described below.
  (a) Layer configuration and sample holder.
  (b) Physical properties of the metal used and its powder state.
  (c) Inclination of the dust layer between -30 and +30 degree (0 degree is horizontal) and the direction of the fire spread.
  (d) Opposite air velocity up to 5 m/s.
  (e) Atmosphere in Ar, N2 and water mist for the fire extinction.
    Behavior of the fire spread is examined with a high-speed video camera and a digital high-precision still camera, and then the luminous zone (combustion zone) is analyzed by an image processing technique. A model analysis is made to understand the behavior of the fire spread over the metal dust layer.
    Results are as follows :
  (1) It is found that flat-shaped layer is more convenient to examine the characteristics of the fire spread than those of triangular prism layer and semi-cylindrical layer. Reducing heat loss from the burning area to the base is necessary by using an insulator plate to obtain proper experimental data.
  (2) The dependency of the layer depth on the spread rate of fire is shown quite small for the metal powder used in this study.
  (3) The spread rate of fire for zircalloy-2 powder ( D v,0.5 = 27μm) is 25.3 mm/s in the flatshaped layer (depth: 2 mm). This value is larger than any other metal powders in this study. The spread rate of fire for Ti (18μm); 6.9 mm/s, Ti (80μm); 2.4 mm/s, Ta (62μm); 2.5 mm/s, Mg (29μm); 1.4 mm/s and Mg/Al (31μm, Mg:Al = 60%:40% wt.); 0.38 mm/s. With increasing the Al component in Mg, the spread over the dust alloy can not continue.
  (4) It is found that the combustion occurs in the surface of the dust layer for Ti, Zr and Ta, while the combustion occurs in gas phase for Mg, Mg/Al and PMMA.
  (5) The thickness of luminous zone for Ti can be estimated between 0.4 and 0.8 mm by means of the image processing.
  (6) In the experiment of inclination of dust layer, the spread rate in downward direction increases with the angle, while the spread rate in upward direction is nearly constant.
  (7) The spread rate of fire increases with the opposite air velocity in all metal powder. These results are different from the results on smoldering combustion of organic materials such as cork, starch, paper, etc.
  (8) Oxidation process over the surface of the dust layer and the thermal conductivity through the layer (powder and air) would be important to control the spread rate of the metal dust fire.

Effect of Air-Bag Support for Prevention of Trench Collapse in Centrifuge Tests

RR-2000-7
Yasuo TOYOSAWA, Noriyuki HORII, Satoshi TAMATE, Tomohiro YAMADA and Hiroaki HAJYOUTA

: Accidents due to collapse sometimes occur in trench excavation site. The need for decreasing these accidents is currently one of major concern. Air-bag support can expect laborsaving of trench excavation. However, there are few research results about the air-bag for earth support, and its prevention mechanism for trench collapse is not fully understood.
    In this research, the centrifugal model experiment was carried out for the purpose of clarifying the mechanism of the trench collapse prevention by the air-bag support. In this paper, a total of fourteen centrifuge tests were illustrated. The trenches with a height of 125 mm and width of 50 mm were made in the right and left side of the model grounds. The air-bag with a height of 50 mm (25 mm and 100 mm for additional tests) and width of 50 mm (prototype size of the air-bag height is 1.0 m at centrifugal acceleration 20 G) was installed in the left-hand side of each trench. The model grounds were made by kaolin clay or the intermediate clay (mixtures of kaolin clay and Toyoura sand), which have different properties, strength, etc. While the centrifugal acceleration was raised gradually, the pressure of air-bag was controlled so that it would reach to 0.02 MPa at a centrifugal acceleration 20 G. The centrifugal acceleration was increased until the both trenches collapsed.
    Based on the results of these centrifuge model tests, the deformation and failure mechanisms were discussed. The conclusions are as follows :
  (1) When the air-bag was installed, compared with the trench without air-bag, the trench collapsed at the large centrifugal acceleration for both of clay and intermediated clay models. The air-bag was effective as an earth support for trench excavation.
  (2) The air-bag was effective in controlling the generation of cracks and the strain development in ground, etc. and prevented the trench wall from moving inward.
  (3) FME analysis using Mohr-Coulomb's failure criterion could follow the elastic behavior with sufficient accuracy. The acceleration generating the plastic strain and the ground displacement could be predicted correctly.
    This research was performed as the joint research with Nippon Steel Corporation who developed this air-bag support.

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