r 



28 



BULLETIN 704, U, S, DEPARTMENT OF AGRICULTURE. 



to the axis of the cyliuder as it is pos- 

 sible to make it. After drying and cool- 

 ing, a specimen to be tested is placed on 

 the anvil (W) of a Page impact ma- 

 chine, shown in figure 2, taking extreme 

 care that when in position the specimen 

 rests with a perfect bearing on the anvil 

 and that the spherical end of the plunger 

 is tangent to the specimen exactly in the 

 center of its upper surface. The small 

 thumbscrew used for separating the 

 clips which hold the specimen in place 

 should be so adjusted that the clips hold 

 the test piece firmly without any undue 

 pressure on the spring. Attention to 

 these details is very important if accu- 

 rate results are to be obtained. When 

 the specimen is in position the current 

 is turned on the magnet and the upper 

 cross-head is lowered until the rod pro- 

 jecting from it just comes in contact 

 with the rod projecting from the lower 

 cross-head (M). This occurs at the in- 

 stant the pilot lamp, in series with the 

 magnet circuit, flashes out. The dial 

 (Q) is then adjusted to O, the upper 

 cross-head raised 1 centimeter, and the 

 motor started. After each blow the fall 

 of the hammer is increased 1 centi- 

 meter until failure of the test piece 

 occurs. 



The height in centimeters of the blow 

 at failure is recorded as the toughness of 

 the material, using for the final value 

 the average of three determinations. In 

 the case of foliated rock the average of 

 three determinations in each direction is 

 taken as the toughness of the sample. 



"Weight per Cubic Foot of Broken 

 Slag. — The weight per cubic foot should 

 be determined by shaking the dried slag 

 to refusal in a cubic foot cylindrical 

 measure. About i of the total amount 

 of slag necessai'y to fill the measure is 

 first introduced in such a manner as to 

 avoid any separation of sizes. This ma- 

 terial is then shaken down by rocking 

 the measure from side to side until no 

 further settlement takes place. This 

 process is repeated until the measure 

 has been filled to overflowing, after 

 which it is struck off level with the top 

 with a straight edge and weighed. 



