1042 THE BELL SYSTEM TECHNICAL JOURNAL, NOVEMBER 1952 



50 



E 15 



0.1 1 10 20 30 40 50 60 70 80 90 99 99.9 



PER CENT OF CONTACTS WITH WELD STRENGTHS ^ VALUE SHOWN 



Fig. 15 — Tj'pical weld strength distributions for the stationary and twin con- 

 tacts. The horizontal scale is graduated so that normal distributions will plot as 

 straight lines. 



2. Clamping the relay pile-ups by means of a simple spring clamp 

 instead of the more conventional method using screws. 



3. A single, easily mounted, operating card. 



Less obvious, but equally important is the basic philosophy whereby a 

 large variety of different relay codes are obtained by assembling parts 

 which for assembly purposes are essentially identical for each code. As 

 previously described, the spring combination for each relay is controlled 

 by selection of the proper code card, twin T\qre assemblies with wires in 

 the proper positions for that combination, and a stationary wire assembly 

 with the right kind of contacts welded to the proper Avires. At the present 

 time six different card varieties, fifty twin wire assemblies and seventy- 

 five stationary wire assemblies have been standardized. The twin uire 

 assemblies are provided with any number from one to twelve pairs of 

 wires in various positions while the stationary wire assemblies have 

 from one to twelve contacts in matching positions, with the added 

 variable that each contact may have precious metal on either or both 

 sides as needed. With these it is possible to obtain more than 300 differ- 

 ent contact combinations, although only about 100 of these are now 

 needed. Yet, with a few exceptions, each relay code is assembled from 



