HOLDING SHACKLES. 



131 



was transmitted to the test portion of the bar. So long as 

 the holes were drilled in the true axis of the bar, this 

 formed a good method of holding the specimens, but it 

 was only applicable to flat bars cut from plates, and involved 

 a waste of metal in order to form the enlarged ends. 



FIG. 61. 



Many test bars are turned from the solid, and are pro- 

 vided with screwed ends. The typical way of holding 

 these is shown on Fig. 61. Here the specimen is at S, the 

 shackle is marked C. A taper bush A rests in the conical 



hole in the shackle, and serves to support a nut B, which 

 is screwed on to the end of the specimen. The under part of 

 the nut has a spherical form, and rests on a correspondingly 

 spherical seating on the bush. By this arrangement 

 the ends of the specimen are free to rotate to a small extent, 

 and the bar can thus adjust itself so that the pull is along 

 the axis, and there is no tendency to bending produced. 

 This is especially important in the case of cast-iron 



