Abstract of Special Research Report (RR-22)

National Institute of Occupational Safety and Health, Japan

Flame Quenching of Hydrogen-Air Mixtures under Low and Reduced Pressures --Estimation of Maximum Safe Gaps for Diametral Clearances--

RR-22-1

Toshihiro HAYASHI and Heizaburo TSURUMI

: Flame quenching in narrow gaps is not only interesting as one of the fundamental aspects of combustion but is an important technical background in considering safety practices such as flame arresters or flameproof enclosures of electrical apparatus for use in hazardous areas.
    This report describes about the flame quenching abilities of various sizes of spigot joints. Experiments were carried out with stoichiometric hydrogen-air mixtures for three cylindrical enclosures. Each enclosure was consisted of ignition chamber and protected chamber, and these two chambers were connected through the gap of spigot joint. The internal volume of the ignition chamber was 100, 450 and 1920 cm³ respectively. The gap width (diametral clearance) had a range from 0.1 to 0.4 and the gap length (length of flame path) were ranged between 0.5 and 25 mm.
    After setting up the test enclosure for a specified combination of the gap width and length the whole assembly was evacuated and then filled with premixed gas mixture up to a test pressure. The mixture was ignited in the center of the ignition chamber through a high voltage electrical discharge and observation was made if any explosion occurred in the protected chamber. The highest initial gas pressure which gave at least five successive quenchings was defined as the Limiting Safe Pressure (L.S.P.) and used to compare the relative quenching ability of the gap.
    Main results are as follows ;
  1) For a constant gap width W, the longer the gap length L is, the higher the L.S.P. became, and the rate of increase of the L.S.P. was larger for smaller W. The relation between the L.S.P. and L was computed to be nearly parabolic in so far as this study concerned.
  2) For the gap width wider than 0.85 mm, it was estimated that any gap length could no longer quench the explosion flame if initial pressure exceeded 100 mmHg (abs.). This result could be compared with the reported values of quenching distance ( 0.5 - 0.64 mm) under atmospheric pressure.
  3) For a constant L, the L.S.P. increased with the decrease of the gap width, and this trend was more distinctly observed for the width less than 0.2 mm.
  4) The gap width which corresponded to the L.S.P. of 760 mmHg was considered to be Safe Gap. For 100 and 1920 cm³ flameproof enclosures, Safe Gap estimated for L = 25 mm was 0.23 and 0.18 mm respectively.
5) Safe Gaps increased linearly with the length of flame path in the range of this study, but this relation might be extrapolated up to L = 50 mm on safe side.

Decaying of Gaseous Detonation by Expansion (Second Report) --Experiments on Some Fuel-Oxygen Detonations in Circular Tubes--

RR-22-2

Hidenori MATSUI

: When a flammable mixture is ignited within a run of pipeline, the flame is often developed into a detonation, causing a very high pressure resulting in serious damages.
    If the detonation once occurs, it is difficult to interrupt it only by the use of conventional flame arresters. However, there is a fact that the detonation can be decayed into the normal combustion wave by enlarging the pipeline dimension on the way.
    From this, the author reported, in the first series of this experiments, the decaying properties of the detonation in the two-dimensional flat tubes for oxy-acetylene mixtures. There had still remained some questions to be solved. The first is that how the properties would be if the detonation proceeds and expands in the three dimensions. The second is whether the detonation decaying would depend on the expansion ratio of pipeline dimension or the I.D. of the driving pipeline.
    In the present report, results of the experiments on gaseous detonations are introduced with C₂H₂ - O₂, C₂H₄ - O₂, C₃H₈ - O₂, H₂ - O₂ and Town gas-O₂ mixtures with certain initial pressures. The detonations were drove in a run of steel pipeline which is varied from 3/8 to 2 inches in nominal diameter. The flange-jointed expanded steel pipeline is 4 inches, however 2 or 8 inches ones are also used for comparison. The wave velocities were measured by ion-probe before and after the enlargement of pipelines, to determine the effect of initial pressure on detonation decaying. Piezo electric pressure transducers were used for measuring the pressures of the wave front in some experiments.
    Main conclusions obtained are as follows ;
  (1) When the detonation wave was expanded into a larger volume, the pressure of the wave front and the velocity of it suddenly were decreased and the detonation was decayed into a combustion. The difficulty of decaying for the stoichiometric oxy-fuel mixtures was in the order of 2C₂H₂ + 5O₂, C₂H₄ + 3O₂, C₃H₈ + 5O₂, 2H₂ + O₂, T.G + O₂.
  (2) The limit of pressure of detonation decaying (P_decay) depends on the gas mixture and the diameter of the driving pipeline (d), and is given by the following equation.
      P_decay = K·d ^-0.9 , where K is constant for each gas mixture.
  (3) To obtain a complete decaying of detonation, it is required to enlarge the diameter of the pipeline at least by four times on the way. The larger the diameter of the expanded pipeline is,the slower the acceleration of the flame after decaying becomes. So the decaying distance glows longer.

Test-Manifucture of Artifitial Fingers (1st Report) --Characeristics of Electro-Fluid as a Control Element--

RR-22-3

Taiji KONDO and Noboru SUGIMOTO

: Winslow effect is defined as an essentially instantaneous reversible change in apparent viscosity when, a fluid is subjected to an externally applied electric field.
    In this papers, we report the result of our investigation which will be devoted primarily to rheological and electrical properties of Electro-Fluid, and the laws, and mechanisms which govern their behavior.
    Winslow effect had been so far evaluated by the increase of apparent viscosity, but here E.V. Induced ; shear stress τ_E was introduced to take place of it. τ_E is expressed as the difference between whole shear stress and the shear stress which is induced in the absence of the electric field.
    The property of Winslow effect of Crystalline cellulose-Diphenyl chloride suspension was typical of most Electro-Fluids and was found to be a function of many parameters such as composition, shear rate, field strength, frequency, temperature and moisture.
    When only one parameter is varied, E.V. Induced shear stresses increase with increasing volume fraction (it must be approximately equal to weight fraction) of disperse phase, field strength, and temperature, but decrease with increasing frequency.
    The moisture (water content) is a parameter that is difficult to be evaluated quantitatively and universally as well. So, here the dielectric constant was introduced in place of the moisture.
    Thus, E.V. Induced shear stress τ_E by measurement is :
      τ_E = C·E ₀² (ε_S - ε_d)² / p_g
where E ₀ is the external electric field, p_g is the weight fraction of disperse phase, ε_S is the dielectric constant of the Electro-Fluid, ε_d is the dielectric constant of the Electro-Fluid no moisture in the disperse phase, and C is a constant coefficient.
    Here we can find the weight fraction in a denominator because of interaction between p_g and ε_S - ε_d.
    On the mechanism of Winslow effect, at first we adopted the following model.
    "Each particle in flowing fluid repeats to pass the other particles, and the electric field applying to particles results induced inter facial polarization, thus induced attraction among particles, hence dissipation heat rises".
    Consequently, by connecting the attraction with E.V. Induced shear stress theoretically we get :

where ε^* is the space permitivity, f _a is the energy barrier coefficient, a is the mean radius of particles, ε₀ is the dielectric constant of dispersion medium.
    In view of the fact that ε₀ is approximately equal to ε_d, and the energy barrier coefficient is inversely
proportionate to √p_g at high weight fraction, this equation has the analogous form to the form previously expressed experimentally.
    As for the mechanism of the polarization of particle, the dielectric dispersion of Electro-Fluid was useful to study it quantitively.
    When electro-fluid has little moisture in its disperse phase, the characteristic frequency is very sub and is approximately equal to the theoretical value which ionic mobility on the surface of starch particle expresses as the characteristic frequency.
    Thus we get the conclusion that the mechanism of Winslow effect is able to be explained sufficiently by the electro double layer hypothesis quantitatively.
    On the base of this theory, we developed new Electro-Fluid and more typical one at that, which had ionic exchange resin as its disperse phase. It will justify this theory.

JNIOSH

Abstract of Special Research Report (RR-22)

National Institute of Occupational Safety and Health, Japan

Flame Quenching of Hydrogen-Air Mixtures under Low and Reduced Pressures --Estimation of Maximum Safe Gaps for Diametral Clearances--

Decaying of Gaseous Detonation by Expansion (Second Report) --Experiments on Some Fuel-Oxygen Detonations in Circular Tubes--

Test-Manifucture of Artifitial Fingers (1st Report) --Characeristics of Electro-Fluid as a Control Element--

Other Publications