Note that the input threshold voltages are symmetrically spaced about the midpoint of the signal swing ( -IER1/2).The input hysteresis is given by v.21cT () l+A III 1 -Vinz = AI. RI -yln~. By inserting values of the output voltage signal swing (IERI) into this equation, it can be verified that the hysteresis increases as the output voltage signal swing increases.Note that when IERI = 41cT/q = 104 mV, A = O, and the circuit exhibits zero hysteresis. For values of the output voltage signal swing less that 104 mV, the output-input voltage gain is always less than unity, and the regenerative action of the positive feedback is lost.In summary, it has been shown that the modified FCS circuit described by Weiss and Jen exhibits reduced hysteresis because of a reduction of the equivalent load resistance, which results in a smaller output voltage signal swing. In addition, the delta-star transformation of the load resistor network shows that the crosscoupliig of R* fails to introduce a negative feedback path between the collectors of Q1 and Q2, thus rendering nugatory the claims of improved dc stability against variations in components and temperature. J. E. PRICE Research & Development Lab. Fairchild Semiconductor Palo Alto, Calii. 94304Author's Reply% J. E. Price, in attempting to simpliiy the problem, has missed the point of our paper. In any switching circuit there are two thresholds to be concerned with, i.e., turn-off and turn-on. A solution based only on dc analysis fails to emphasize the time dependence of these thresholds to the speed of the circuit. We felt that a pictorial allagram (the VO -10 characteristics) of the two threshold conditions wouId attempt to show such a relationship and give a much better tilght into the operation of the circuit during switching, although it does not give the intermediate output points. Through dc analysis the quiescent conditions can be determined; but the relative delays between the input transition and the internal dynamic response of the circuit components will determine the actual path of the operating point during switching.The delta-wye transformation for analysis of the load line is equally satisfactory, and we agree with the determination of resistors rl, rz, and r,,. However, by simple application of Thevenin's theorem for the wye, the equivalent fixed load resistor at the node v, is R,. = T,. + r, connected to a voltage source of V,, = -?-.. (1.-1,). Ifsubstitutions are made back to the original then R,(R, + R..) R,, = R, + R, + R,. and circuit parameters, -R,R2 Vth = & + R, + Roe (I, -IO). The slope of the load line remains constant and equal to Rth; but the equivalent voltage level of Vt. is a function of the term (~. -Io). 2 Manuscript received April 3, 1967.AC VO gets more positive, transistor Q2 conducts more current. That portion of the Q2 current flowing through the resistive path R,. and RI will tend to make point VO more negative, thereby compensating for the original change in VO. Thus, the insertion of Rti has introduced negative feedback, and, in ef...