2
\$\begingroup\$

I came across the following circuit in my text book:

schematic image

Summary of the explanation given in my text book: When control input Q3 is low, Q1 is on and output is HI. When control input Q3 is high, Q2 is on and the output is forced LO. D1 is necessary to insure that Q1 is off, when Q2 is turned on. R1 is necessary to limit the transient current when changing states (because bipolar transistors turn on more quickly than they turn off).

I can't seem to figure out how the control input being low or high results in the output states it does. Would someone be able to walk me through the logic? Basically what would the voltages be at the intermediate points between input and output?

Thanks.

\$\endgroup\$
2
  • \$\begingroup\$ See what I wrote with some detail here and here. May be more than you need. Or not. But these may help all the same. \$\endgroup\$
    – jonk
    Sep 18, 2018 at 20:19
  • \$\begingroup\$ Note this is an amplifier. The usual input to this amplifier is a NPN transistor with the base pulled up to rail, and the collector tied to base of your Q3. The internal nodes are VERY SLOW and are controlled by very small currents. This results in very bad jitter, very bad phase noise. Beware. Do not use TTL for low phasenoise synthesizers. \$\endgroup\$ Sep 19, 2018 at 4:23

2 Answers 2

11
\$\begingroup\$

Note to New Contributors

Draw out the schematic using the existing schematic editor that you have access to when writing out your question. It's helpful to do so, despite having a "nice picture." If for no other reason, it allows others to quickly snap up and use your schematic as a starting point for adding additional notes.


Short Overview

Your schematic is basically this:

schematic

simulate this circuit – Schematic created using CircuitLab

I've added the typical resistor values for TTL here, as well. Just for reference.

The first thing to understand how the input (labeled \$IN\$ in the schematic) is actually driven. The circuit does not stand by itself. The drive will typically either be holding \$IN\$ close to ground (with output HI) so that \$Q_3\$ is OFF or else by sourcing a current of about \$700\:\mu\text{A}\$ into the base of \$Q_3\$ (with output LO.)

With that in mind, let's discuss the two cases.


Output HI

Here is the schematic in the case where the output is HI. I've added some details about how \$IN\$ is actually driven in a real circuit as well as a few short notes here and there. I've also included the equivalent resulting circuit on the right side.

schematic

simulate this circuit

First, look at the circuit on the left side and make sure that you agree with me about which BJTs are on and which are off. Then verify that you also agree with me about the equivalent circuit on the right.

In this state, the output cannot sink any current. You should also easily see that \$R_1\$ provides some high-side current limiting (in case the output is grounded, for example.)

With normally low sourcing currents, the output impedance (given \$\beta\ge 40\$) is about \$\frac{R_2}{\beta+1}\approx 40\:\Omega\$. However, when attempting to force higher sourcing currents by driving \$Q_1\$ into saturation (at sourcing currents exceeding close to \$3\:\text{mA}\$), the output impedance rapidly moves towards \$R_1\$'s value, or \$130\:\Omega\$. (That's actually a desirable behavior; having lower sourcing impedance when operating normally and having the source impedance increase if the circuit is being pushed harder towards its limits.)

(In this saturated case, the only requirement is that the saturated \$\beta\ge \frac{R_2}{R_1}\approx 12\$. Which is reasonable.)


Output LO

Here is the schematic in the case where the output is LO. I've added some details about how \$IN\$ is actually driven in a real circuit as well as a few short notes here and there, once again. I've also included the equivalent resulting circuit on the right side... once again:

schematic

simulate this circuit

First, look at the circuit on the left side and make sure that you agree with me about which BJTs are on and which are off. Then verify that you also agree with me about the equivalent circuit on the right.

In this state, the output cannot source any current. It can only sink current. And it can sink a fair amount of it, in fact. (Just look at the base current into \$Q_2\$.)


That's about it. If you want more discussion then please refer to the other two posts here I've made on the topic: (1) TTL Inverter and also (2) TTL AND.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ +1 belatedly. A beautiful answer. \$\endgroup\$
    – Russell McMahon
    Jul 16, 2019 at 13:26
  • \$\begingroup\$ Nice answer. Question: under your heading "Output LO", where does the 0.7mA into Q3 base come from? If the driving ckt was another identical totem pole that was in the HI state, then would that 0.7mA be much larger? \$\endgroup\$ May 7 at 0:07
  • \$\begingroup\$ - Just realised this was an old question, I have no idea why it came up in my feed. \$\endgroup\$ May 7 at 1:09
0
\$\begingroup\$

When control input is low, Q3 is off, so R3 pulls Q2 base low and turns off Q2.

With Q3 off, R2 drives Q1 base current, turning Q1 on and letting current flow thru R1 and D1 to the output. Output is then 5V - output current x R1 - Vce of Q1 - Vf of D1.

With Q3 on, Q1 is off, Q2 is on, and output is basically Vce of Q2.

If you assume Vbe of 0.5 to 0.7V, and same for Vce, then you can determine the intermediate voltages.

\$\endgroup\$
4
  • \$\begingroup\$ Thanks that helps a lot. I still don't understand the latter case, when Q3 is on. Why would that force Q1 to be off? Wouldn't a current still flow through R2 and into base current for Q1? \$\endgroup\$
    – user198450
    Sep 18, 2018 at 17:24
  • \$\begingroup\$ No, because the collector of Q3 is only about 1.0V above ground (Vbe of Q2 plus Vce of Q3). This puts the base voltage of Q1 at about the same as its emitter (Vce of Q2 plus diode drop), keeping it cut off. All of R2's current flows through Q3. \$\endgroup\$
    – Dave Tweed
    Sep 18, 2018 at 17:28
  • \$\begingroup\$ Makes sense. Last question, how come Q2 and Q3 don't have resistors to control their base current (like R2 with Q1)? Is that not needed here? \$\endgroup\$
    – user198450
    Sep 18, 2018 at 17:35
  • \$\begingroup\$ R2 also protects Q2, and R3 protects Q3. \$\endgroup\$ Sep 18, 2018 at 17:40

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.