Abstract of Special Research Report (SRR-82-1)

National Institute of Occupational Safety and Health, Japan

Special Research of Integrated Preventive Countermeasures for Overturnning Accidents Transportation Machinery



: Woodworking equipment creates more accidents than every other machinery does. Especially, the accidents caused by the powered circular saw and hand-feed jointer account for 80 percents of all woodworking equipment accidents.
    These injuries come almost getting the hands and fingers against the saw-blade and jointer-head (70 percents), and another injuries are by kickback of stock (about 10 percents). Most of injured parts of body are fingers and hands : sometimes physically handicapped.
    Following studies are practiced to support the countermeasures against such injuries :
  (1) The factors producing kickback are pinpointed, then some devices for anti-kickback are developed and examined, and effect of the equipment are discussed.
  (2) Hood guards and other safety devices for preventing the worker from contacting with saw-blade are developed and application of the devices are discussed.
    Points of regard at which the worker is gazing during the operation are measured by means of video eyemark camera and, at the same time, worker's opinion are also tape-recorded as to what they pay attention during the operation. Workers' attention properties are analyzed from the side of human engineering.
  (3) Actually used safety devices are investigated regarding the structure and usage, and the actual unsafe condition and unsafe-act are investigated.
    Then, through systematical analysis of the accidents and troubles, subcauses and underlying causes such as fault in control or management are traced.
  (4) A large quantity of wood dust and shavings is produced from wood-working machines during the operation.
    Explosion experiments on a large scale dust collecting system were conducted, in order to prevent wood dust explosions and fires. Date and information were obtained on the explosibility dust concentration limits in the dust collecting pipe and the effectiveness of an explosion vent in the filter collecting unit.
    In relation to the ignition hazards of wood dust, the method to determine the minimum ignition energy of dust layers is reported with measurements of the values for some wood dust. The method described here for dust layers could be useful in estimating the ignition hazards of wood-dusts.
  (5) Moreover TG-DTA and the self-ignition behaviors of various kinds of saw dust were measured.
    Then the relationships between exothermic characteristics of each sort of saw dust and its environmental factors such as particle size, coexisting material and so on was studied.
  (6) Dust collecting system is necessary, not only to prevent the fire and explosion accidents, but also to improve upon the working environment for safety.
    So a dust collecting system for moulders was developed, and the relationships between the collecting effect and the shape, size and location of a few hoods were studied experimentally.
    On the other hand the process of systems was also investigated by computer-simulation method.
    Finally recommendations for models of collectors are given.

Systematical Analysis of Countermeasures against Hazards Involved in Operation of Woodworking Machinery by using MORT

Yoshinobu SATO, Kiyoshi FUKAYA and Yoshiyuki EGAWA

: In Japan, occupational accidents caused by the use of woodworking machinery has not decreased in the last several years, and the number of injuries by the accidents, for example, in 1980 accounted for 6.3 percent of all injuries resulting in more than four-days off from work in all sectors of industry. Thus, the prevention of such accidents at woodworking factories is given a more priority than other areas.
    In an attempt to find out the effective means of application of countermeasures to the workplace, an investigation was carried out of about 4300 accident cases collected throughout the country, fact-finding missions were undertaken, and relevant questionnaires were sent out.
    To analyse the various data collected as above on woodworking accidents, operations and machinery, the technique of MORT (Management Oversight and Risk Tree) was utilized. MORT is regarded as a useful tool to examine the existing elements of management oversight and hazards on workplaces systematically.
    As the results of analysis, it has been found that :
  (1) Approximately half of the total accidents occurred by circular saw bench and so on due to unsuitable design or construction of guards ;
  (2) There is a limitation of wider introduction of automatic and /or single-performance woodworking machines which will not require safety guards, because of the small or medium scale of the woodworking enterprises and of the working method adopted (in the case of the woodworking factories employing 20-49 workers, for example, production process comprises the assembly line operation (31 percent), individual operation (41 percent), single production (22 percent), and piecework operation (6 percent) ) ; and
  (3) There is a variety of operations in which no safety devices or apparatus are applicable. (These operations have been identified and classified in our hands, and the rate of the factories conducting such operations to the total number of woodworking factories has been estimated)
    Identification of the risk assumed to woodworking operations has been subsequently made, and the evidence was given that the risk would be accepted by management, not because the frequency and effect of accidents are too small, not to be ignored and not because the elimination of risk is technically impracticable though accidents would have great effects but simply because it costs too much for safety in comparison with the effects of accidents.
    It has been concluded that the development of safety guards be necessary which could meet the following requirements :
  i) not decreasing productivity,
  ii) easy to attach to, and/or detach from the body of machinery,
  iii) not lower the value of wood materials,
  iv) given a consideration of ergonomical design, and
  v) with reasonable cost in purchase and maintenance.

Study of the Functional Safety Measures for Multiuse Woodworking Equipments

Noboru SUGIMOTO, Yoshinobu SATO and Kiyoshi FUKAYA

: Also to woodworking operation, as to any industrial operation, the new technology of production has been introduced : the automation, robotalization, flexible automation system, which eliminate the necessity of the workers' operation.
    Generally speaking, the technology have not widely extended, in almost of the woodworking industries, workers' skills and technics have been counted on in the operation of the traditional powered woodworking machines/equipments: the power circular saws, ripsaws, band saws, planers, shapers and other multi-used machines.
    Especially the powered circular saws have produced frequent accidents, in addition, number of the accident has not been decreased in these few years. There are two principal types of accidents involving powered circular saw : (1) contact by the operator, helper, takeaway man with the blade during operation and (2) kickbacks. The powered circular saw as a object of the study is selected, safety measured are studied: fail-safe and fool-proof system in the operation.
    According to the safety and sanitation rule, regarding wood working operation, every machine should stop by means of shutting off the power within five seconds after pushing a "stop switch". Electro-magnetic braking should be provided with the powered circular saw to stop it immediately. Because long free running time after stop-switching not only creates another unnecessary hazard that the worker may produce the unsafe conduct during free running, but also gives a loss to the production efficiency.
    The braking system, capable of immediately stopping the saw, makes it possible to get the fool-proof system by electronically picking up unnecessary access to the saw blade and dangerous stock kickback. In this study, properly, executed the study on development of the finger sensor, kickback pickup, self-controllable hood guard and application of the equipments to the woodworking operation system.

Study on Prevention to Blade Injuries by Sawing Machine

Kiyoshi FUKAYA and Noboru SUGIMOTO

: According to the analysis of accidents caused by the use of woodworking machinery, approximately half of the total accidents occurred by circular saw bench and so on. And 79 % of accidents were blade injuries. Thus countermeasures against blade injuries involved in operation of circular saw bench were studied. There were three approaches ; measurement of human characteristic of response to the kickback, measurement of object of workers' attention during circular saw work, and evaluation of safety guards in the view point of ease of work.
    Human characteristic of response to the kickback were measured by using mockup circular saw bench on which kickback were simulated. The mockup consist of circular saw bench without blade, wood with sensors which moves along guide, driver which drives the wood backward in such manner as kickback, and controller. The output of sensors such as push-down force of man, displacement of wood, etc. were recorded by data recorder. In the experiments during repeat of simulated ripping work the kickback occurred and the responses were recorded. Experiment revealed following matters. According the method of holding wood, there were difference of push-down force. In other words by placing the finger on the back-edge of wood, push-down force in the time of kickback decreased. If there is sufficient holding width, man can stop the wood against kickback such as 15 kgf x 0.1 sec.
    During circular saw work object of workers' attention i.e. eye-point-of-regard were measured by using eye mark camera and VTR. Workers' opinion as to what to attend were also recorded.
    During almost all of cutting time, workers' attention were on the point of operation. In other words workers gave attention to the saw blade. But in other phases of work such as getting back of wood, it is on the object of task i.e. wood in this case, because workers' attention are close related to the work. And this phases of work, were 70-50 % of all work time. This means need of safety guard to prevent blade injuries.
    In the view point of ease of work, three type of safety guards were evaluated by two ways; one was motion study, and another was workers' opinion. Model work with safety guard were recorded by VTR, and time study were done. Questionnaire about safety devices including the guards were also used. Two type of guards were nearly same score in both scoring method. The last guard which got bad mark showed what is important for safety guards to be preferred by workers. The most important point for worker was whether it is easy to see the point of operation.

New Models of Wood-Shavings Collectors for the Molding Machine with High Speed Single Spindle


: Effective collector of wood-shavings or dusts has not yet been developed for the molding machines with high speed single-spindles. An obstacle is that works to chamfer and plane the curved surfaces of boards are formed with the machine. This paper describes two models of collecting wood shavings and dusts by purging jet-air and suction air.
    The model are shown Fig.5-1 (Type A) and Fig.5-7 (Type B).
    Evaluating experiments on efficiency of the collecting models were carried out on practical scale, and the computational simulations by the Monte-Carlo method, with aid of experimental data were found to give an estimate of movement of wood shavings on the molder and of efficiency of collecting model.
    From the results of study, concluded recommendations for the new models of collectors are as follows ; For the type A model,
  i) Dimensions of the outlet of jetting duct and the inlet of collector should be coincident with the scale values in Fig-1.
  ii) Average air velocities at the outlet and the inlet should be 10 and more than 15 m/s respectively.
  iii) The center of the outlet of jetting air should be located at L = 36 - 40 cm and H = 13 cm (see Fig.5-1).
  iv) Angle φ between the axis of purging air-jet and horizontal surface of the molder should be 30 degree,
and for the type B model,
  i) Dimension of width of the inlet at the table of the molding machine should be 6 cm.
  ii) Average air velocity at the inlet should be more than four tenth as fast as the revolution speed of plane cutter.
  iii) The center of the inlet should be positioned at L = 30 cm (see Fig.5-7).
  iv) The purging air jet may be unnecessary.

Dust Explosion avd Fire Hazards in a Wood-Dust collecting System

Toei MATSUDA and Michio NAITO

: In many woodworking shops, dust collecting systems have been installed for the purposes of environmental aspects, use of wood dusts as a fuel in the furnace and so on. The hazards of wood-dust explosions and fires in the systems, however, have not always been appreciated, partly due to the recognition of the wood dust as by-products or wastes among wood workers. If flow velocities in the duct are kept as high as practical, then the time of exposure of dust particles to an ignition source is minimal, and the possibility of ignition is minimal. On the other hand, the hazard of flame propagation and acceleration in the coal mine galleries under initial static conditions have been reported in many publications, but the effect of air currents on the dust explosibility is not well clear in the ventilating duct system. Wood-dust explosion propensity in pipeline was then examined in relation to the installation of protective measures in the collecting, or pneumatic transport systems.
    A full scale pneumatic transporting system was employed with steel piping of 3 in. and 4 in. diameters (Fig.6-1). The ignition sources for the dust flow were an explosion flame of acetylene-air mixture, near-stoichiometric composition, or a continuous induction spark. A unit bag filter with explosion vent was used as a receiving hopper in order to see whether safe explosion venting could be achieved in the case of the flame ejecting from the high speed flowing pipe.
    The result for an effect of conveying air velocity on explosibility limits of a wood-dust/air mixture is compared with that of a coal-dust/air mixture, and Essenhigh et al.3) (Fig.6-3). At an air speed of 15 m/s, the range of explosible concentrations agreed closely with the result of Essenhigh et al. in spite of different conditions (2" pipe ) and a different type of dust (coal-dust), but the flame was not extinguished at the higher conveying velocity of up to 40 m/s with the gas ignition source. From those comparison, flame extinction by velocity and concentration is estimated to be affected by the power strength of the igniting source : with a small source of ignition such as a spark, the critical air velocity, above which flame was no longer propagated at any dust concentration, could be given for the dusts used, but if a severe ignition source is present, then a wider range of explosibility limit would be obtained at conveying velocities higher than the critical one. The lower explosibility limit seems to be scarcely affected by increasing the air velocities.
    The extinction mechanism has not been well understood, but the higher mixing of cold unburned dust flow into the flame zone is expected to prevent the ignition and subsequent flame propagation with increasing turbulence. Above the critical air velocity, the dust flame formed in contact with the severe ignition source, would not have enough time to be extinguished by the higher conveying velocity and insufficient mixing of cold dust flow.
    Average flame speed were measured along the pipeline with photo-transistors (Figs.6-4 and 6-5). It was generally found that the flame was propagated initially in a more or less rapid acceleration and then gradually decelerated to a velocity such that the flame was conveyed by the dust flow into the filter unit.
We may suspect that the decreasing flame speed could be attributed to the condensed dust concentrations ahead of the flame zone, in the turbulence flow.
    The maximum pressure for vented explosions in the filter unit was as low as the static bursting pressure of the vent cover, 0.08 kg/cm2 . The capacity of the bag filter was 1.3 m3, and the vent ratio ( area of vent/ volume of vessel ) was 0.56-1, and relatively large compared to the value recommended for a bag filter by Bartknecht5). A fireball of 3.5 m maximum diameter was formed outside the dust filter unit on the explosion venting. All the combustible materials should be removed from the danger space relevant to the opening of explosion vents. A continuously burning flame was maintained at the mouth of the inlet pipe of the filter unit, unless the dust flow was cut off immediately after the explosion.
    A glowing cigarette-butt and grinding frictional sparks were introduced into the dust flow pipe as the practical igniting sources. The cigarette-butt, which came to pieces in the flow, caused smouldering fires in the filter unit, but the frictional sparks were mostly cold in the pipe.
    In general, wood-dust collecting system are operated under high flow speed and relatively low dust concentrations, and then it is concluded that they have the hazards of wood-dust explosions and fires, which could propagate instantaneously throughout the system by the presence of high flowing air, even in small duct systems.

Oxidative Self-Heating Behaviors of Fifteen Different Species of Wood Powder

Takashi KOTOYORI and Michio NAITO

: There exist many research reports on various aspects of pyrolysis and combustion phenomena of wood or woody materials. However, researches on oxidative self-heating behaviours of those materials under strict adiabatic condition are very few. For that matter works by Akita et al.2), by Robertson et al.3) and lately by Anthony et al.4) are quite precious.
    Thus we also tried to measure, at first, oxidative heating behaviors of some species of wood powder in air under atmospheric pressure and effects of coexistent substances on them using TG-DTA methods, and then, tried to measure the oxidative self-heating profiles of these materials using SIT (self-ignition testing apparatus, a sort of an adiabatic self-heating process recorder).
    Wood samples subjected to the measurement are fifteen different species, vis., Japanese cypress, Japanese red pine, Paulownia, Zelkova, Sawara cedar, Red lauan, White lauan, Western red cedar, Yellow cypress, Douglas fir, Port Orford cedar, Western hemlock, Spruce and Eli ayanskya. The wood samples were powdered in a mill. The particle size is 30 to 60 mesh.
    Standard experimental conditions are as follows. TG-DTA : sample amount, 5 mg ; sample cell, Al pan (2.5 mm high, 5 mm diameter) ; gas flow rate, 20 ml/min ; heating rate, 2.5 K/min. SI T: sample amount, 300 mg ; loading density, 0.12 to 0.14 g/ml; sample cell, silica tube (ca. 2 ml) ; initial (starting) temperature, 150 °C.
    Main conclusions gained are as follows :
  1) Adiabatic self-heating profiles of wood powder are linear in most cases and this fact implies that oxidative heating reaction of materials of this kind at temperatures near 150 °C is approximately zero-order. This finding fits well in with Anthony's description4).
  2) Relative self-heating rates of fifteen different species of wood powder are given in Table 7-2. Red lauan is easiest among them to heat oxidatively, and Spruce is hardest to do so.
  3) Wood powder, when heated in nitrogen atmosphere, decomposes exothermally. In that case the exothermal shift of DTA curve begins at about 180 °C (Fig.7-20,-21).
  4) When wood particle is pyrolyzed in oxidative atmosphere, decomposition products are pushed out of the interior of the particle and burn on or near the surface of the particle. In other words it is thought that oxygen does not enter into the wood particle or the charcoal during pyrolysis.
  5) Self-heating rate of the particle, the size of which is coarser than 30 mesh, is less than that of finer particles (Fig.7-25).
  6) Wood powder made from older lumber is easier to heat oxidatively than that made from newer lumber (Fig.7-32).
  7) Powder made from bark is easier to heat oxidatively than wood powder (Fig.7-35,-36).
  8) Powder made from knot is easier to heat oxidatively than sapwood powder or heartwood powder (Fig.7-38).
  9) Soot, which accumulates in the soot-collecting unit set up at the top of the sawdust incinerator, is easier to heat oxidatively than the original wood powder (Fig.7-40).
  10) Cu powder shows a remarkable catalytic effect on oxidative heating reaction of wood powder (Fig.7-52,-53).
  11) CuO catalyzes the oxidative decomposition reaction of the charcoal at temperatures over 200 °C. Although the effect of Pt is more remarkable, the similar effect can be said of Pt, too.
  12) Fe and Fe2O3 does not show any catalytic action on oxidative thermal decomposition reaction of wood powder.
  13) A mixture of wood powder and melamine resin (its main component is actually phenol-formaldehyde resin) powder shows a very remarkable oxidative self-heating behavior (Fig.7-58,-59).
  14) The oxidative heating reaction of wood powder is hindered by the added lubricating oil at temperature levels of 140 to 150 °C, at least. This effect is thought to be due to antioxidant that is usually added to the lubricating oil (Fig.7-61,-62).

An Investigation of Minimum Ignition Energies for the Layers of Some Wood Dusts

Toei MATSUDA and Michio NAITO

: In wood process industries, values of minimum ignition energies are required for safety standards in relation to possible explosion hazards of wood dust clouds. A mass of data on the ignition energy are available for dust clouds, but the data for the ignition energies of dusts as deposits, heaps or their layers and the accuracy of the values are limited.
    In this paper the method to determine the minimum ignition energy of dust layers is reported with measurements of the values for some wood dusts. Spark electrodes were located in tight contact with the surface of the dust layers and the stored energy in a condenser was discharged at the spark gap. The energy released in the spark was determined by computer-aided integration of voltage and current traces on a digitized synchroscope.
    On the release of sufficient energy at the gap, flaming of dust particles could be possible above the layer (Photo 8-1), but it would become difficult to recognize the ignition as criticality for ignition is approached (Photo 8-2). The photocell positioned above the electrode gap was then employed in order to facilitate the finding of ignition with naked eyes. An ignition criterion near the critical value was that glowing particles were discernible at the spark.
    Experiments have also been carried out to study the influence of electrode configuration, electrode separation, bulk density of the dust layer and particle size on the minimum ignition energy of lycopodium and wood dusts. The results show that particle size has a strong influence on the minimum ignition energy, and that optimum spark gap width is in similar order of values to those of gaseous or dust/air mixtures. The spark discharge induces dispersion of dust particles around the electrode gap (Photo 8-4), and subsequent ignition of the particles could be given if sufficient energy is dissipated. It seems likely that dispersability of dust particles due to the ignition spark may have an important effect in determination of minimum ignition energies of dust layers. The dispersion mechanism is likely that the dust is denuded, i.e. an earpickfull of dust is removed suddenly from the dust surface.
    The minimum ignition energies obtained for eleven different wood dusts with mean particle diameters ranging from 10 to 22 microns are summarized in Table 1. It will be of interest to compare these with results for dust clouds of the same samples and we expect that the method described here for dust layers can be useful in place of the testing method for dust/air mixtures.

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