enter image description here

First, I drew the cascaded small signal equivalent circuit (unilateral hybrid pi transformation) for Figure 3 as: enter image description here

After drawing the cascaded small signal equivalent circuit (unilateral hybrid pi transformation) of Figure 3, I determined that the output resistance \$R_o=R_{C2}\$ and the input resistance \$R_i=R_{B1}+r_{\pi}\$. Meanwhile, to find the voltage gain \$A_v=\frac{V_o}{V_i}=-G_mR_o\$, I first had to find an expression for the transconductance \$G_m\$. I began by writing an expression for the output current \$i_o=-g_mV_{EB2}\$. I needed to express \$V_{EB2}\$ in terms of \$V_i\$ and I did so using the following circuit analysis equations:

1) \$g_mV_{BE1}-\frac{V_{EB2}}{r_{\pi}}+\frac{V_{C1}}{R_{C1}}=0\$

2) \$\frac{r_{\pi}}{R_{B2}+r_{\pi}}V_{C1}=V_{EB2}\$

3) \$V_{BE1}=\frac{r_{\pi}}{r_{\pi}+R_{B1}}V_i\$

Eventually, \$V_{EB2}=-g_m(\frac{r_{\pi}}{r_{\pi}+R_{B1}}V_i)(\frac{r_{\pi}R_{C1}}{R_{B2}+r_{\pi}-R_{C1}})\$.

Plugging this into \$i_o=-g_mV_{EB2}\$ and solving for \$\frac{i_o}{v_i}=G_m\$, \$G_m={g_m}^2\frac{{r_{\pi}}^2R_{C1}}{(r_{pi}+R_{B1})(R_{B2}+r_{\pi}-R_{C1})}\$ so that \$A_V=-G_mR_o=-\frac{{g_m}^2{r_{\pi}}^2R_{C1}R_{C2}}{(r_{pi}+R_{B1})(R_{B2}+r_{\pi}-R_{C1})}\$ where \$r_{\pi}=\frac{\beta}{g_m}=\frac{100}{0.01{\Omega}^{-1}}=10000\Omega\$. My problem is that the denominator factor \$(R_{B2}+r_{\pi}-R_{C1})=30000\Omega+10000\Omega-40000\Omega=0\Omega\$ so that the gain \$A_v\$ becomes negative infinity. Is there something wrong in my process? Any help please?

  • \$\begingroup\$ First, please show us the small signal equivalent circuit. Also, why don't you try another approach. 1 - find the Q1 gain and Q2 gain. Av1 = gm1*Rc1||(Rb2+rpi2) and Av2 = gm2*Rc2. And finally, I can include rpi1 and rpi2 effect on the gain. Av3 = rpi1/(Rb1+rpi1), A4 =rpi2/(Rb2+rpi2). Av = Av1*Av2*Av3*Av4 \$\endgroup\$
    – G36
    Feb 19, 2017 at 14:11
  • \$\begingroup\$ @G36 Edited. My small signal equivalent circuit can now be seen above. \$\endgroup\$
    – John Smith
    Feb 19, 2017 at 15:11
  • \$\begingroup\$ Recognize that Rc1 = (Rb2 + Rpi). Current source gmVbe1 splits equally between Rc1 and (Rb2 + Rpi). So your second-stage input voltage Veb2 is gmVbe1/2. Have you run afoul of signs? \$\endgroup\$
    – glen_geek
    Feb 19, 2017 at 15:38
  • \$\begingroup\$ @glen_geek Yes, but I don't get why the "second-stage input voltage Veb2 is gmVbe1/2." What do you mean? \$\endgroup\$
    – John Smith
    Feb 19, 2017 at 15:45
  • 1
    \$\begingroup\$ As glen_geek point out the gm1*vbe1 current will splits equally between Rc1 and (Rb2 + Rpi). This means that Vbe2 = 0.5*(gm1*Vbe1)*rpi2. Or from current divider rule we have (gm1*Vbe1)*(Rc1)/(Rc1+Rb2+rpi2) = I_rpi2 current. I ignore the minus sign, and Vbe2 = I_rpi2*rpi2. \$\endgroup\$
    – G36
    Feb 19, 2017 at 16:54

1 Answer 1


Givem gm is 0.01 mho [or 100 ohms emitter 'reac'], knowing gm = 1A/V at 26mA, we know your Ic is 0.26mA in each transistor. With beta=100, the rin of each transistor is (100=B * 100=reac) = 10,000 ohms.

The load on collector of Q1 is 40Kohm || (30Kohm + 10Kohm) = 20Kohm. Gain of stage1 is gm * effectiveRC (ignoring Vearly effects), thus 0.01 * 20K == 200x. Scaled by voltage division of Rb1 and beta*reac = 4:1, thus only 50X.

The load on collector of Q2 is 4Kohm. Av is gm*Rc = 0.01 * 4K = 40X. But again we have that input 4:1 divider, so only 10X.

Total gain is 50 * 10 = 500x.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.