Timing and Clocking Issues

ECE152B
Prepared by Roopa Chari
Based on the slides from Professor Tim Cheng
Timing Parameters for Combinational Logic
Physically implemented combinational circuits (NAND or NOR gates for example)
exhibit certain timing characteristics.
A 0 or 1 applied at the input to a combinational circuit does not result in an
instantaneous change at the output because of various electrical constraints. Input-tooutput delay in combinational circuits can be expressed with two parameters, propagation
delay, tpd, and contamination delay, tcd.
Propagation delay (tpd): The amount of time needed for a change in a logic input to
result in a permanent change at an output, that is, the combinational logic will not show
any further output changes in response to an input change after time tpd units.
Contamination delay (tcd): The amount of time needed for a change in a logic input to
result in an initial change at an output, that is, the combinational logic is guaranteed not
to show any output change in response to an input change before tcd time units have
passed.

Combinational propagation delays are additive and so the propagation delay of a larger
combinational circuit can be determined by adding the propagation delays of each of the
circuit components along the longest path. In contrast, finding the contamination delay
of the circuit requires identifying the shortest path of contamination delays from input to
output and adding the delay values along this path.
Timing Parameters for Sequential Logic
When sequential circuits are physically implemented they exhibit certain timing
characteristics that unlike combinational circuits, are specified in relation to the clock
input.
1

Latch vs. Flip-Flop

A latch is level-sensitive while a flip-flop is edge triggered. A latch stores when the clock
level is low and is transparent when the level is high. A flip-flop stores when the clock
rises and is mostly never transparent. Since flip-flops only change value in response to a
change in the clock value, timing parameters can be specified in relation to the rising (for
positive edge-triggered) or falling (for negative-edge triggered) clock edge.
The following parameters specify sequential circuit behavior. Note that these are all for
positive edge-triggered flip-flops unless otherwise specified, but are easily applied to
negative edge triggered flip-flops as well.
Propagation delay (tclkq): The amount of time needed for a change in the flip-flop clock
input D to result in a change at the flip-flop output Q. When the clock edge arrives, the D
input value is transferred to output Q. After time tclkq the output is guaranteed not to
change value again until another clock edge trigger arrives.
Contamination delay (tcd): This value indicates the amount of time needed for a change
in the flip-flop clock input to result in the initial change at the flip-flop output Q. The
output of the flip-flop maintains its initial value until time tcd has passed and is
guaranteed not to show any output change in response to an input change until after tcd
has passed.
Note: delays can be different for both rising and falling transitions.
Setup time (tsu): The amount of time before the clock edge that data input D must be
stable the rising clock edge arrives.
Hold time (thold): This indicates the amount of time after the clock edge arrives that data
input D must be held stable in order for the flip-flop to latch the correct value. Hold time
is always measured from the rising clock edge (for positive edge-triggered) to a point
after the clock edge.

Setup and hold times are restrictions that a flip-flop places on combinational or
sequential circuitry that drives a flip-flop D input. The circuit has to be designed so the D
input signal arrives at least tsu time units before the clock edge and does not change until
at least thold time units after the clock edge. If either of these restrictions is violated for
any of the flip-flops in the circuit, the circuit will not operate correctly. These restrictions
limit the maximum clock frequency at which the circuit can operate.
Determining the Maximum Clock Frequency for a Sequential Circuit
Most digital circuits contain both combinational components (gates, muxes, adders, etc.)
and sequential components (flip-flops). These components can be combined to form
sequential circuits that perform computation and store results. By using combinational
and sequential component parameters, it is possible to determine the maximum clock
frequency at which a circuit will operate and generate correct results. This analysis can
best be examined through use of an example.
Example 1: For the circuit shown below, assume the delay through the register (tclk-q) is
0.6ns and the delay through each logic block is indicated inside the box. Assume that the
positive edge-triggered registers have a set-up time tsu of 0.4ns. What is the minimum
clock period?

The first step is model the given circuit so that the datapath between flip-flops are characterized
by the longest and shortest combinational delays between the flip-flops.

direction of clock

Let Tclk be the clock period. Then,

Tclk tclk-q (register) + tpd logic (longest) + tsu (destination register)
Tclk 0.6 ns + 8 ns + 0.4 ns
Tclk 9 ns
Delay Modeling
Gate propagation delay (tPHL and tPLH): Gate propagation delay is measured from 50%
input to 50% of the output. tPHL is measured from 50% of the rising edge of the input
voltage to 50% of the rising edge of the output voltage. Similarly, tPLH is measured from
50% of the falling edge of the input voltage to 50% of the falling edge of the output
voltage.

Interconnect Delay: This is delay caused by wires. Interconnect introduces three types
of parasitic effects capacitive, resistive, and inductive all of which influence signal
integrity and degrade the performance of the circuit. After the signal frequencies of our
interests (up to several GHz), we can ignore the effects of inductance of interconnect.
When gate G1 drives gate G2, we can model the circuit as the one showing below in the
right-hand side figure where Rout is the output resistance of G1, Rin and Cin are the
interconnect resistance and capacitance, and CL is the input capacitance of G2.

This model can be further

simplified as the one shown in the
right. Now if Vin (i.e. output of G1)
switches from 0V to VDD, the
waveform at Vout (i.e. input of G2)
can be expressed as:
Vout = VDD (1- e-t/RC)

R= Rout+Rint
Vin
(output
of G1)

(input
of G2)

Vout
C=Cin+CL

Definition: Rise time (Fall time) - The time it takes for a waveform to rise from 10% to
90% (90% to 10%) of its steady state value as illustrated below.
90%

VOL

VOH

VOH

10%

90%
10%

VOL

Fall time

Rise time

The time for the waveform Vout = VDD (1- e-t/RC) to rise from 0.1VDD from 0.9VDD can be
easily calculated:
Rise Time ~= 2.2 * R * C
Note that the rise time is linearly proportional to the product of (Rout+Rin) and (Cin+CL).
So, if the interconnect is long, Rin and Cin will be large, and, in turn, the rise time will be
long. Similarly, if a gate drives a large number of gates, the rise time will be long too as
CL, which is proportional to the number of driven gates, is large.

Clock Trees
If the number of flip-flops driven by the clock line is large, the clock rise time (also
called slew rate) will be unacceptably long. The solution to this problem is to use a clock
power up tree which means adding buffers into the clock tree.
When designing a clock distribution network, the absolute delay from a central clock
source to the clock elements is irrelevant, only the relative phase between two clock
elements is important.

When designing clock trees in this way, first determine the number of levels your clock
tree can have. This will depend on the total number of flip-flops in the circuit and the
number of fan outs (to flip-flops or buffers) that you are limited to. See the examples in
the lecture slides.

Clock Skew and Clock Jitter

Clock Skew (): The spatial variation in arrival time of a clock transition is known as
clock skew. The clock skew between two points j and k is given by tj tk, where tj tk are
the rising edge of the clock with respect to the reference. Clock skew is constant from
cycle to cycle and does not cause clock period variation, but only phase shift.
Clock skew might cause the race problem illustrated in the lecture slides.
Clock Jitter: Clock jitter refers to the temporal variation of the clock period, that is, the
clock period can expand or reduce on a cycle-by-cycle basis.

Variation of the pulse width is important for level sensitive clocking.

Positive and Negative Clock Skew

Positive Skew: Clock and data flow in the same direction.

Minimum cycle time: Tclk + tclk-q + CriticalPathDelay+ tsu

This means that clock skew has the potential to improve the performance of the circuit
(minimum required clock period reduces!). However, increasing clock skew makes the
circuit more susceptible to race conditions.
Negative Skew: Clock and data flow in opposite directions

Note: Receiving edge arrives before the launching edge

On one hand, negative skew adversely impacts the performance (increase the clock
period). On the other hand, negative skew implies that the system never has the race
problem (since receiving edge happens before).

Summary for Clock Skew

Minimum Clock Period (Tclk)

Tclk + >= tclk-q + CriticalPathDelay+ tsu

Worst case is when receiving edge arrives earl, i.e. negative . )

Race Condition: Hold Time (Thold) constraint

< tclk-q + tpd logic (shortest path) - thold

To avoid the race condition, use this equation above to calculate the maximum allowable
positive or negative clock skew.
Worst case is when receiving edge arrives late, which results in race between data and
clock.
Example 2: Assume tclk-q is 0.6ns, tsu is 0.4ns, and thold is 0.5ns.

The first step once again is to re-model the circuit:

direction of clock
8

(a) Determine the minimum clock period assuming a positive clock skew: = (t t) = 1ns.
Redraw the circuit to reflect this positive skew. If the clock is in the same
direction as the data then (tpd logic) is the effective delay and if the clock is in
the opposite direction, then (tpd logic + ) is the effective delay.

With the new effective delay, we can now calculate the clock period in the way
as no skew exists. The critical path has now 7ns. Then,
Tclk tclk-q (source register) + tpd logic (longest) + tsu (destination register)
Tclk 0.6 ns + 7 ns + 0.4 ns
Tclk 8 ns
(b) Repeat part (a), factoring in a positive clock skew: = 3.

Here, the critical path is now 8ns. So,

Tclk tclk-q (source register) + tpd logic (longest) + tsu (destination register)
Tclk 0.6 ns + 8 ns + 0.4 ns
Tclk 9 ns
(c) Repeat part (a), factoring in a negative clock skew: = -2ns.

The critical path is now 10ns.

Tclk tclk-q (source register) + tpd logic (longest) + tsu (destination register)
Tclk 0.6 ns + 10 ns + 0.4 ns
Tclk 11 ns
(d) Derive the maximum positive clock skew (i.e. t > t) that can be tolerated
before the circuit fails.

direction of clock

The original circuit is shown above. To determine the maximum (or minimum)
clock skews, you need to use the hold time constraint so that race conditions do
not occur.
tclk-q + tpd logic (shortest) > thold + [tpd logic (shortest) in direction of clock]
< 0.6 ns + 4 ns 0.5ns
< 4.1ns
(e) Derive the maximum negative clock skew (i.e. t < t) that can be tolerated
before the circuit fails.
tclk-q + tpd logic (shortest) > thold + [tpd logic (shortest) in opposite clock direction ]
< 0.6 ns + 5 ns 0.5 ns
< 5.1 ns

References
J. M. Rabaey, A. Chandrakasan, B. Nikolic, Digital Integrated Circuits, A Design
Perspective (2nd Edition), Prentice Hall, Electronics and VLSI Series.

10