149 



possible unequal strain from the line wire could be considered a cause 

 for breakage. Oialy four whole insulators were available for this test, 

 two of which had marked internal stresses, while in the other two the 

 stresses were almost negligible. The insulators were broken, by placing 

 them upon an iron pin as in service and by exerting a strain upon them 

 in the direction of the line wire. One insulator which was poorly an- 

 nealed broke at 960 pounds, while the others failed at 1,890, 1,675 and 

 2,220 pounds respectively, the latter being one which was also poorly an- 

 nealed. While this test did not show very conclusively that the poorly 

 annealed insulators were weak mechanically, it is believed that if the 

 pull in the latter case, could have been in such a direction as to cause the 

 insulator to break along strained internal planes as was probably the case 

 in the first test, the latter insulator as well would have been found to 

 have been weak mechanically. For conclusive evidence of this fact, how- 

 ever, a much larger number of tests should be available. 



It will be seen from the foregoing, therefore, that a very practical 

 use has been made of the phenomenon which has so long been only an in- 

 teresting physical experiment. With the aid of the polariscope it is not 

 only possible to determine some of the causes for the unsatisfactory service 

 given by certain glass insulators, but it is also possible to make preliminary 

 acceptance tests upon new insulators and to eliminate all of those which 

 show signs of improper annealing and which for this reason would be un- 

 desirable for installation where they must be subjected, not only to severe 

 electrical and mechanical strains, but also to vibration and sudden tem- 

 perature changes. Although porcelain is rapidly supplanting glass for 

 high tension insulators, it is expected that this method of test will be used 

 in the future to advantage and that it will prove of equal, if not greater 

 value, than it has in this particular instance. 



[11—21368] 



