At the previous report we have explained that the pulsating torsional fatigue deformation of the carbon steel may be prevented by the pretorsional stress.
In the present investigation for the carbon steel, it was found that by the heat-treatment called to be Stress-ageing in which cooled the test piece to room temperature in the furnace after heating during 10min. at 300°C under loading, the torsional fatigue deformation decreases more than the two; the cold-working and the heat-treatment called to be Strain-ageing which has the same conditions of heating and cooling under unloading after cold-working.
According to the above experiment the stress-ageing is the available treatment for determining of the setting stress which improves the fatigue deformation and the fatigue strength of helical spring for valve and etc.

View full abstract

It has been well known that the observed fatigue life for the constant repeated stress is scattered widely.
This experimental study was carried out in order to investigate the character of the frequency distribution of the fatigue life of steel.
The specimen was a flat bar with a circular hole in its center and not annealed. Its material was SS 41 steel.
The sort of repeated stress was the pulsating tension stress and the stress levels were 30kg/mm², 27kg/mm², and 24kg/mm², 20 specimens were tested each stress levels with Lcsenhausen Fatigue Testing Machine (U. H. S. type). Besides these, to 20 specimens the stress level 24kg/mm² was applied to failure after stress level 27kg/mm² were applied 50000 cycles, in order to investigate the scatter of fatigue life of the specimen to which two different stress levels were applied.
Experimental errors were analized statistically.
The conclusion from experiment was as follows:
1. The scatter of fatigue life was well interpreted by introducing the probability of fracture μ.
2. The method to estimate the probability of non-fracture was obtained.
3. The scatter observed in this experiment was considerably smaller than that reported by other investigators.
4. It was clarified that the scatter of fatigue life was not obtained from the experimental errcrs only.

View full abstract

In order to explain the effect of induction hardening on fatigue strength, we tried several fatigue teats by changing. the stress concentration factor α_K.
We used C=0.41% steel, and chose the three types, as the heat treatments.
The following facts were cleared up from the experimental results.
(i) At normalized specimens, α_K becomes greater than β_K, as we have known already.
(ii) At ordinary quenching (heating in electric furnace and quenching in water), α_K equals approximately, β_K.
(iii) At induction hardening, the fatigue limits is considerably high, and does not change largely by changing of α_K.
It was apparent that, within the range of the investigation, the greater α_K the larger became the effect of induction hardening on the fatigue limit.
Lastly we discussed the reason why the fatigue limits became high by induction hardening.

View full abstract

Though our air mill has many excellent points compared with other mills, we consider it should be impoved in some points. We regarded the flow in an air mill as that around a standing propeller. Along the axis of the air mill we set a hollow cylinder which has a hole at the middle of it to guide the flow in the mill.
We obtained the results by this alteration that the separating ability was improved, pulverizing efficiency was increased and the abrasion of the propellers in the mill was tolerably reduced.

View full abstract

Constructing an apparatus being similar to that of Ferry and his co-workers which was used for measuring the dynamic properties of some hypolymers, the changes in dynamic viscosity and dynamic rigidity during the casting of clay slips for refractory crusibles were measured.
It was found out that the casting properties were affected by
(1) the distance from the mold surface,
(2) the amount of defloculant, and
(3) the quality of the defloculating electrolyte.

View full abstract

In the creep of rubber vibration absorber of Tread type, we recognized experimentally the existence of virgin curve and found the creep strain was not proportional to the stress at a certain time after loading. These phenomena are explained reasonably by the dynamical model proposed by Sawaragi and Tokumaru for nonlinear viscoelastic bodies. In practice, it is not desirable to use these materials in the state in which virgin curve will be obtained, for then the permanent set from internal solid friction results. In order to make these materials useful in practice, we must get them to creep to a certain degree and to recover enough. Then we can get the stable state of these materials. In this report a suitable load and the duration time of load to give such a stable state are estimated.

View full abstract