THE L3 SYSTEM DESIGN 



797 



ct 



— UJ 



o5 



I UJ 

 -I Q 

 < 



zz 



"J Z 



^ -80 



>- 0.1 0.2 0.3 0.4 0.6 0.8 1.0 2 3 4 5 



VIDEO FREQUENCY IN MEGACYCLES PER SECOND 



Fig. 11 — Threshold values for bar patterns. 



No. 2 is —16 db and that of mastergroup No. 3 is —11 db. As a con- 

 sequence of setting levels in this way, the random noise in the message 

 channels is approximately 2 db higher on the average than modulation 

 noise. It can be shown that with both second and third order modulation 

 products contributing, and with third order somewhat predominant, this 

 relation between random noise and modulation noise produces optimum 

 signal-to-noise performance. With these levels, the 1,860 channel tele- 

 phone system has approximately 6 db margin against repeater overload 

 which, for L3 purposes, has been defined as the point at which the re- 

 peater modulation coefficients just depart from their constant small- 

 signal values. The signal-to-noise objective of +29 dba at the — 9 db 

 level is met with about 2 db margin. 



When the system is used to transmit television and message signals 

 simultaneously, the level of the telephone channels in mastergroup 

 No. 1 at the repeater output is the same as that of mastergroup No. 1 

 in the all -telephone application, —21 db. The most convenient measure 

 of the television signal is the power of the unmodulated carrier at the 

 output of a repeater. Its value is +6 dbm. Due to the inter-relations 



o 



X 



in in 

 oi J 

 a: UJ 

 xm 



3 4 5 6 8 10 20 30 40 



FIELD BRIGHTNESS IN FOOT LAMBERTS 



60 80 100 



Fig. 12 — Assumed deviation from Weber-Fechner law. 



