984 



THE BELL SYSTEM TECHNICAL JOURNAL, SEPTEMBER 1952 



4 6 8 10 12 14 16 18 



NUMBER OF 95% RELATIVE HUMIDITY CYCLES 



Fig. 

 coils. 



4 — Corrosion comparison of impregnated and non-impregnated relay 



winding washers. Likewise, one group was impregnated while the other 

 was not. Fifth and sixth groups of coils having cellulose acetate insulation 

 throughout with and without impregnation were exposed and there were 

 no failures at the end of the test. This shows that the corrosive effects 

 of impure materials can be retarded, but not overcome by resorting to 

 impregnation. In general, impregnation of relay coils is not desirable 

 because of the risk of contaminating the vital working surfaces of the 

 relay. 



In the normal operation of a relay when the circuit through its winding 

 is opened, a transient voltage, which may reach hundreds of volts is 

 generated across the winding terminals by the collapsing flux. If the 

 insulation between the lead-out wires or between the lead-out wires and 

 the end turns of the winding is not adecjuate, electrical breakdown causes 

 arcing and repeated operation of the relay may cause ultimate disintegra- 

 tion of the wire and consequent failure of the rela3^ It is important, there- 

 fore, to design the coil so that lead-out wires under all conditions are 

 properly spaced, and to provide adequate insulation between those por- 

 tions of the winding where high voltages can exist. A test has been de- 



