Chapter 7 Designing Sequential Logic Circuits Rev 1.0: 05/11/03 1.1: 5/23/03 1.2: 5/30/03

Sequential Logic Finite State Machine (FSM) Pipelined System 2 storage mechanisms: Positive feedback (SRAM) Charge-based (DRAM)

Naming Conventions In our textbook: a latch is Level-sensitive flip-flop a register is Edge-triggered flip-flop There are many different naming conventions For instance, many books call Edge-triggered elements flip-flops (asynchronous JK, SR)  This leads to confusion

Latch v.s. Register Latch stores data when clock is low (or high) stores data when clock rises (on edges) D Q D Q Clk Clk Clk Clk D D Q Q

Latches transparent hold hold hold

Latch-Based Design N latch is transparent when f = 0; hold when f = 1 P latch is transparent when f = 1; hold when f = 0 f f N P Logic Latch Latch Logic

Timing Definitions t CLK D c - q hold su Q DATA STABLE Register CLK D Q (a) Setup time (T_su): the time before the clock edge that the D input has to be stable (b) Hold time (T_hold): the time after tue clock edge that the D input has to main stable (c) Clock-to-Q delay (Tc-q): the delay from the positive clock input to the new value of the Q output.

Characterizing Timing D -Q D Q D Q Clk Clk t t C -Q C -Q Register Latch

Maximum Clock Frequency T CLK tclk-Q + tp,comb + tsetup Also: tcdreg + tcdlogic >= thold tcd: Contamination Delay = Minimum delay tclk-Q + tp,comb + tsetup <= T

Mux-Based Latches Negative latch (transparent when CLK= 0) Positive latch (transparent when CLK= 1) CLK 1 D Q 1 D Q CLK

Mux-Based Latch

Mux-Based Latch NMOS only Non-overlapping clocks CLK Q CLK Q CLK CLK M

Writing into a Static Latch Use the clock as a decoupling signal, that distinguishes between the transparent and opaque states D CLK Forcing the state (can implement as NMOS-only) Converting into a MUX

Master-Slave (Edge-Triggered) Register Positive Latch Negative Latch Master Slave Two opposite latches trigger on edge Also called “master-slave latch Pair “

Master-Slave Register Multiplexer-based latch pair Q M D CLK T 2 I 1 3 4 5 6

Setup Time of MS-Register I2-T2 : I2 output to T2 Check input of T2 and output of T2 are the same

Clk-Q Delay Volts 2.5 CLK 1.5 D t t Q 0.5 2 0.5 0.5 1 1.5 2 2.5 - q(lh) t Volts c - q(hl) Q 0.5 2 0.5 0.5 1 1.5 2 2.5 time, nsec

Reduced Clock Load Master-Slave Register c.f: 8 Clock loads in Mater-Slave Register Design

Avoiding Clock Overlap X CLK CLK Q A D B CLK CLK (a) Schematic diagram CLK CLK (b) Overlapping clock pairs

SR Flip-Flop: Cross-Coupled Pairs Cross-coupled NORs S R Q NOR-based Set-Reset Flop-Flop S Q R Forbidden State Q R S 1 Cross-coupled NANDs

Clocked NOR-based SR Flip-Flop Q R Cross-coupled NORs Added Clock Control This asynchronous SR FF is NOT used in datapaths any more, but is a basic building memory cell

Output voltage dependence on transistor width Sizing Issues Output voltage dependence on transistor width Transient response

Storage Mechanisms Static Dynamic (charge-based) D CLK Q

Clock Overlap T0-0: T1 and T2 on  Race Condition

Making a Dynamic Latch Pseudo-Static Adding a weak feedback inverter

Clocked CMOS (C2MOS) Clock 1 Master Evaluate Hold Slave High-impedance: Hold Output Previous value stored in CL2 New Value of CL1 “Keepers” can be added to make circuit pseudo-static

Insensitive to Clock-Overlap DD DD DD DD M M M M 2 6 2 6 M M 4 8 X X D Q D Q 1 M 1 M 3 7 M M M M 1 5 1 5 (a) (0-0) overlap (b) (1-1) overlap

True Single-Phase Clocked Register (TSPC) Positive latch (transparent when CLK= 1) Negative latch (transparent when CLK= 0) A register can be constructed by cascading Positive and Negative Latches  12 transistors are used!

Including Logic in TSPC Example: logic inside the latch AND latch

Pipelined TSPC CMOS System Compared with NORA CMOS, we need two extra transistors per stage. But we can operate at a true singe-phase clock signal. Very attractive from system-design point of view.

Positive Edge-triggered Register in TSPC

TSPC-based Positive Edge-Triggered DFF From Referenced Textbooks: [1] “CMOS Integrated Circuits: Analysis and Design,” 3rd Ed., by Sung-Mo Kang and Yusuf Leblebici, McGraw-Hill, 2003.

Pipelined Systems using Dynamic CMOS Circuits

Pipelining Reference Pipelined T_(non-pipe) = 3 x T_(pipeline)

Pipelining At the expense of “Latency (input-to-output delay)”  Not good for interactive communicaitons

Latch-Based Pipeline Be careful of Race! Hold G Hold F

Review of NP-Domino Logic

NP-Domino Logic Example

NORA CMOS Evaluation at Phi=1 Evaluation at Phi=0 Pipelined NORA CMOS system

Latch-based Pipeline using C2MOS Race-free as long as function F (implemented by static logic) between the Latches are Non-inverting!

Potential Race Condition during 0-0 (if F is inverting)

Example of NORA-CMOS (I)

Example of NORA-CMOS (II) NOR2 + INV = OR2 (Dynamic + Static Stages)

Summary Sequential circuits need good latches and registers for speed performance. Dynamic circuits can realize the pipelined system in a very efficient and compact way. But it should be designed with extreme care. Current trend is NOT to use dynamic CMOS for normal-speed operations  good for design, maintain, and verification.