Boolean Algebra, Logic Gates & Semiconductor Logic Families
Mathematical foundations and solid-state physical implementation of digital logic: George Boole's algebraic axioms and Huntington's postulates; duality principle; De Morgan's laws and algebraic multi-variable reduction; canonical digital logic gates (AND, OR, NOT, NAND, NOR, XOR, XNOR); functional completeness and universal gate synthesis (NAND-only and NOR-only networks); bipolar Transistor-Transistor Logic (TTL Totem-Pole) and complementary MOS (CMOS) inverter circuit electronics; propagation delay, fan-out, power dissipation, and high/low noise margins.
ยง2.1 Huntington's Postulates, Algebraic Axioms & The Duality Principle
1. Formal Mathematical Axioms of Boolean Algebra
In 1904, Edward V. Huntington formalized George Boole's algebraic logic as a deductive mathematical system defined on a set of elements $B$ with two binary operators: logical OR ($+$) and logical AND ($\cdot$), satisfying six fundamental postulates:
- Closure: For every $x, y \in B$:
$$x + y \in B \quad \text{and} \quad x \cdot y \in B$$
- Identity Elements: There exist unique elements $0, 1 \in B$ such that for every $x \in B$:
$$x + 0 = x \quad \text{and} \quad x \cdot 1 = x$$
- Commutative Laws: For every $x, y \in B$:
$$x + y = y + x \quad \text{and} \quad x \cdot y = y \cdot x$$
- Distributive Laws: Each operator distributes over the other:
$$x \cdot (y + z) = (x \cdot y) + (x \cdot z)$$$$x + (y \cdot z) = (x + y) \cdot (x + z) \quad (\text{Crucial rule with no ordinary algebra analog!})$$
- Complement: For every $x \in B$, there exists a unique complement $\bar{x} \in B$ such that:
$$x + \bar{x} = 1 \quad \text{and} \quad x \cdot \bar{x} = 0$$
- Distinct Elements: There exist at least two elements $x, y \in B$ such that $x \ne y$.
2. Fundamental Theorems of Boolean Algebra
From Huntington's postulates, the standard operational theorems are deduced:
- Idempotence: $x + x = x$ and $x \cdot x = x$.
- Null Elements (Dominance): $x + 1 = 1$ and $x \cdot 0 = 0$.
- Involution (Double Negation): $\overline{\bar{x}} = x$.
- Absorption Laws:
$$x + (x \cdot y) = x \quad \text{and} \quad x \cdot (x + y) = x$$$$x + (\bar{x} \cdot y) = x + y \quad \text{and} \quad x \cdot (\bar{x} + y) = x \cdot y$$
- Associative Laws:
$$x + (y + z) = (x + y) + z \quad \text{and} \quad x \cdot (y \cdot z) = (x \cdot y) \cdot z$$
3. The Principle of Duality
The Principle of Duality states: Any true Boolean algebraic identity remains strictly valid if the operators $+$ and $\cdot$ are interchanged, and the identity elements $0$ and $1$ are simultaneously interchanged.
For example, the dual of the distributive law $x \cdot (y + z) = (x \cdot y) + (x \cdot z)$ is directly obtained by swapping $\cdot \leftrightarrow +$:
Duality halves the labor of Boolean mathematical proofs: proving one theorem automatically validates its dual.
ยง2.2 De Morgan's Theorems & Multi-Variable Function Reduction
1. Augustus De Morgan's Laws
Augustus De Morgan (1847) established the two most celebrated theorems in digital circuit design, defining the rigorous relationship between conjunction, disjunction, and complementation:
In digital circuit topology, De Morgan's laws state that a NOR gate (OR followed by inversion) is functionally identical to an AND gate with inverted inputs (negative-AND), and a NAND gate is identical to a negative-OR gate:
2. Generalized Generalized Multi-Variable Formulation
By mathematical induction, De Morgan's laws extend to an arbitrary number of $n$ variables:
To complement any arbitrary Boolean function $F(A, B, C, \dots, +, \cdot)$, one simultaneously applies De Morgan's rule across all levels: interchange all $+$ and $\cdot$ operators, and complement every individual literal ($A \to \bar{A}$ and $\bar{A} \to A$).
3. The Consensus Theorem
The Consensus Theorem provides a powerful shortcut for eliminating redundant terms in multi-variable equations without tedious K-map expansion:
where $B C$ is the consensus term formed from the conjunction of the two literals associated with the complemented pair $A$ and $\bar{A}$. In its dual form:
ยง2.3 Canonical Logic Gates: AND, OR, NOT, NAND, NOR, XOR & XNOR
1. Basic Logic Gates and Operational Truth Tables
Digital logic gates are physical electronic circuits that perform elementary Boolean switching operations on binary voltage signals ($V_{LOW} \leftrightarrow 0$, $V_{HIGH} \leftrightarrow 1$):
- NOT Gate (Inverter): Implements single-input complementation: $Y = \bar{A}$. If $A=0$, $Y=1$; if $A=1$, $Y=0$.
- AND Gate: Output is HIGH if and only if all inputs are HIGH: $Y = A \cdot B$.
- OR Gate: Output is HIGH if at least one input is HIGH: $Y = A + B$.
- NAND Gate: Negated AND operation: $Y = \overline{A \cdot B}$. Output is LOW if and only if all inputs are HIGH.
- NOR Gate: Negated OR operation: $Y = \overline{A + B}$. Output is HIGH if and only if all inputs are LOW.
2. Exclusive-OR (XOR) & Exclusive-NOR (XNOR) Gates
The XOR gate ($\oplus$), or modulo-2 adder, yields a HIGH output when the inputs are different:
Properties of XOR:
- $A \oplus 0 = A$, \quad $A \oplus 1 = \bar{A}$ (programmable inverter).
- $A \oplus A = 0$, \quad $A \oplus \bar{A} = 1$.
- Commutative: $A \oplus B = B \oplus A$; Associative: $(A \oplus B) \oplus C = A \oplus (B \oplus C)$.
- For $n$ inputs, an XOR gate acts as an odd parity detector: output is 1 if and only if an odd number of inputs are 1.
The XNOR gate ($\odot$), or equivalence detector, produces a HIGH output when the inputs are identical:
ยง2.4 Universal Gate Synthesis: NAND-Only & NOR-Only Networks
1. Functional Completeness in Digital Logic
A set of Boolean operators is defined as functionally complete if every arbitrary Boolean function can be expressed solely using operators from that set. The standard set $\{\text{AND}, \text{OR}, \text{NOT}\}$ is functionally complete. However, fabricating multiple distinct gate types on an integrated circuit increases silicon area and manufacturing complexity.
A Universal Gate is a single gate type capable of synthesizing all elementary logic functions (NOT, AND, OR, XOR) without requiring any other components. There exist precisely two universal logic gates in digital electronics: NAND and NOR.
2. NAND-Only Gate Realizations
- NOT using NAND: Tie both inputs together:
$$\overline{A \cdot A} = \bar{A}$$
- AND using NAND: Feed the output of a NAND gate into a NAND-inverter:
$$\overline{\overline{A \cdot B}} = A \cdot B \quad (\text{2 NAND gates})$$
- OR using NAND: Invert each input with a NAND gate, then feed into a third NAND gate (De Morgan's law):
$$\overline{\bar{A} \cdot \bar{B}} = \overline{\bar{A}} + \overline{\bar{B}} = A + B \quad (\text{3 NAND gates})$$
- NOR using NAND: Invert the output of the NAND-synthesized OR gate:
$$\overline{A + B} \quad (\text{4 NAND gates})$$
- XOR using NAND: Synthesize $A \bar{B} + \bar{A} B$ with minimum four 2-input NAND gates:
$$X = \overline{A B}, \quad Y = \overline{A \cdot X} \cdot \overline{B \cdot X} = A \oplus B \quad (\text{4 NAND gates})$$
3. NOR-Only Gate Realizations
- NOT using NOR: Tie both inputs together: $\overline{A + A} = \bar{A}$ (1 NOR gate).
- OR using NOR: Invert the NOR output: $\overline{\overline{A + B}} = A + B$ (2 NOR gates).
- AND using NOR: Invert each input, then combine in a NOR gate: $\overline{\bar{A} + \bar{B}} = A \cdot B$ (3 NOR gates).
- NAND using NOR: Invert the output of the NOR-synthesized AND gate (4 NOR gates).
- XOR using NOR: Synthesized with five 2-input NOR gates.
ยง2.5 Semiconductor Logic Families: TTL Totem-Pole vs CMOS Inverters
1. Transistor-Transistor Logic (TTL) & The Totem-Pole Output
Standard BJT Transistor-Transistor Logic (7400 series) operates from a single $+5\text{ V}$ power supply ($V_{CC}$). The canonical TTL NAND gate consists of three stages:
- Input Stage: Multi-emitter NPN transistor $Q_1$. If any input is LOW ($0.2\text{ V}$), $Q_1$ conducts base current to the LOW input, pulling $Q_1$'s collector voltage low and cutting off phase-splitter $Q_2$.
- Phase-Splitter Stage: Transistor $Q_2$ generates complementary out-of-phase drive voltages at its collector and emitter.
- Totem-Pole Output Stage: Consists of pull-up transistor $Q_4$, diode $D$, and pull-down transistor $Q_3$:
- Output LOW State ($Y = 0$): $Q_2$ and $Q_3$ are saturated. $Q_3$ pulls the output to $V_{OL} \approx V_{CE,\text{sat}} \approx 0.2\text{ V}$. Meanwhile, $Q_4$ is completely off because diode $D$ drops $0.7\text{ V}$, ensuring base-emitter voltage $V_{BE4}$ is insufficient to turn $Q_4$ on.
- Output HIGH State ($Y = 1$): $Q_2$ and $Q_3$ are cut off. $Q_4$ acts as an active pull-up emitter follower, charging the capacitive load rapidly to $V_{OH} = V_{CC} - V_{BE4} - V_D \approx 5.0 - 0.7 - 0.7 \approx 3.6\text{ V}$.
The totem-pole active pull-up provides low output impedance in both states, dramatically accelerating capacitive line charging compared to passive resistor pull-ups.
2. Complementary MOS (CMOS) Inverter
CMOS technology pairs an enhancement-mode pMOS pull-up transistor with an nMOS pull-down transistor in a symmetric, push-pull configuration:
- Input LOW ($V_{\text{in}} = 0\text{ V}$): $V_{GS,n} = 0 < V_{tn} \implies$ nMOS is OFF. $V_{GS,p} = -V_{DD} < V_{tp} \implies$ pMOS is saturated/linear, pulling $V_{\text{out}}$ to exactly $V_{DD}$ with zero static current.
- Input HIGH ($V_{\text{in}} = V_{DD}$): $V_{GS,n} = V_{DD} > V_{tn} \implies$ nMOS is ON. $V_{GS,p} = 0 \implies$ pMOS is OFF. nMOS pulls $V_{\text{out}}$ to exactly $0\text{ V}$ with zero static current.
- Power Dissipation: Because one transistor is always cut off in the steady state, static power dissipation is virtually zero ($P_{\text{static}} \sim\text{nW}$). Power is consumed only during switching transitions as dynamic power:
$$P_{\text{dynamic}} = C_L V_{DD}^2 f$$where $C_L$ is load capacitance and $f$ is clock switching frequency.
3. Key Logic Family Performance Metrics
| Metric | TTL (Standard 74xx) | CMOS (74HCxx / Modern) | Physical Significance |
|---|---|---|---|
| Supply Voltage ($V_{CC}/V_{DD}$) | $5\text{ V} \pm 5\%$ | $2\text{ V} - 6\text{ V}$ (Core: $0.8 - 1.8\text{ V}$) | Power rail tolerance |
| $V_{IH,\text{min}} / V_{IL,\text{max}}$ | $2.0\text{ V} \ / \ 0.8\text{ V}$ | $0.7 V_{DD} \ / \ 0.3 V_{DD}$ | Input threshold boundaries |
| $V_{OH,\text{min}} / V_{OL,\text{max}}$ | $2.4\text{ V} \ / \ 0.4\text{ V}$ | $V_{DD} - 0.1\text{ V} \ / \ 0.1\text{ V}$ | Output drive levels |
| High Noise Margin ($NM_H$) | $V_{OH} - V_{IH} = 2.4 - 2.0 = 0.4\text{ V}$ | $V_{DD} - 0.7V_{DD} = 0.3 V_{DD} \ (1.5\text{ V})$ | Immunity against positive spikes |
| Low Noise Margin ($NM_L$) | $V_{IL} - V_{OL} = 0.8 - 0.4 = 0.4\text{ V}$ | $0.3 V_{DD} - 0.1 = 0.3 V_{DD} \ (1.5\text{ V})$ | Immunity against ground bounce |
| Propagation Delay ($t_{pd}$) | $10\text{ ns}$ | $8\text{ ns}$ (Advanced CMOS: $< 0.1\text{ ns}$) | Maximum operational clock speed |
| Fan-Out | $10$ standard loads | $> 50$ (limited only by capacitive delay) | Number of parallel gate inputs driven |
Using only Huntington's postulates and fundamental Boolean theorems: (a) Prove algebraically that $A + A B = A$ (Absorption law). (b) Simplify the multi-variable expression $F(A, B, C) = A B + \bar{A} C + B C$ to its irreducible two-product form using the Consensus theorem. (c) Determine the complement of the function $G(W, X, Y, Z) = W X + \bar{Y}(Z + \bar{W})$ using De Morgan's laws.
Apply identity (A = A*1), distributive law, null element (1 + B = 1), and identity element.
Expand consensus term BC with (A + A_bar) = 1, absorb ABC into AB, and absorb A_barBC into A_barC.
Apply De Morgan's laws step-by-step: invert sums to products, invert products to sums.
A + AB = A \quad (\text{Proved}); \quad F = A B + \bar{A} C; \quad \bar{G} = (\bar{W} + \bar{X})(Y + W \bar{Z})
Given the Exclusive-OR logic expression $Y = A \oplus B = A \bar{B} + \bar{A} B$: (a) Derive an algebraic transformation expressing $Y$ strictly in terms of NAND operations with only 4 two-input NAND gates. (b) Draw the gate connection equations and verify with a truth table for all four input combinations $(0,0), (0,1), (1,0), (1,1)$.
Add null terms AA_bar = 0 and BB_bar = 0, then factor out (A_bar + B_bar) = (AB)_bar.
Apply De Morgan's law to convert the outer sum into an inverted product (NAND).
Gate 1 computes (AB)'. Gate 2 computes (A G1)'. Gate 3 computes (B G1)'. Gate 4 computes (G2 * G3)'.
For A=1, B=1: output is 0. For (0,1): G1=1, G2=1, G3=0 => Y=1. For (1,0): G1=1, G2=0, G3=1 => Y=1. Exactly matches XOR.
Y = \overline{ \overline{A \cdot \overline{AB}} \cdot \overline{B \cdot \overline{AB}} } \quad (\text{Exactly 4 two-input NAND gates})
A standard 74-series TTL gate has the following guaranteed datasheet electrical parameters: $V_{OH} = 2.4\text{ V}$, $V_{OL} = 0.4\text{ V}$, $V_{IH} = 2.0\text{ V}$, $V_{IL} = 0.8\text{ V}$, $I_{OH} = -400\text{ \mu A}$, $I_{OL} = 16\text{ mA}$, $I_{IH} = 40\text{ \mu A}$, and $I_{IL} = -1.6\text{ mA}$. (a) Calculate the High and Low noise margins ($NM_H, NM_L$) of this TTL gate. (b) Calculate the maximum DC fan-out of the TTL driver. (c) A designer attempts to drive a standard CMOS gate ($V_{IH} = 3.5\text{ V}$) directly from this TTL gate. Determine if direct driving is reliable and specify the required pull-up resistor solution.
Evaluate noise margins in Volts.
Both states yield Fan-out = 10 standard 74-series TTL loads.
The maximum guaranteed HIGH output of TTL (2.4 V) falls far below the minimum required HIGH input of 5V CMOS (3.5 V = 0.7 V_DD), stranding the CMOS input in the forbidden linear region.
Connecting a 2.2k to 4.7k pull-up resistor from the TTL output to +5V pulls V_OH all the way to 5.0 V, guaranteeing reliable switching.
NM_H = 0.4\text{ V}, \ NM_L = 0.4\text{ V}; \quad \text{Fan-Out} = 10; \quad \text{Direct TTL-to-CMOS unreliable } (2.4\text{V} < 3.5\text{V}) \implies \text{Requires } 2.2\text{ k}\Omega \text{ pull-up resistor}
Solved University Examination Problems
Step-by-step mathematical solutions to classic university honors examination questions.