In ideal digital logic theory, we often assume that signals propagate through gates instantaneously. However, in the physical world, every logic gate has a measurable response time, known as gate propagation delay. This delay is typically measured as the time it takes for an input change to result in an output change ( or ), often around 10ns per gate. When different paths in a circuit have unequal delays, the output may behave in ways that steady-state analysis cannot predict, leading to transient errors known as glitches.
What is a Hazard?
A timing hazard occurs when a circuit's output produces an unwanted short pulse (a glitch) at a time when the steady-state logic predicts the output should remain constant. While we usually design circuits based on their functional truth tables, the physical implementation—how the gates are actually connected—determines whether these hazards will appear.
Key Distinction: A hazard is not a functional error in the logic itself, but a timing error caused by the physical realization of that logic.
Types of Hazards
Hazards are generally categorized by how many times the output "bounces" or changes during a single input transition.
Static Hazards
A static hazard occurs when the output momentarily changes from its correct, steady state before returning to it. This involves a single glitch.
- Static-1 Hazard: The output is supposed to remain at a logic '1', but due to propagation delays, it momentarily drops to '0' and then returns to '1'. This often happens in Sum-of-Products (SoP) circuits when an input change causes one product term to turn off before another has turned on.
- Static-0 Hazard: The dual of the static-1 hazard. The output is supposed to remain at '0', but momentarily jumps to '1' before returning to '0'. This is commonly seen in Product-of-Sums (PoS) implementations.
Keep reading for free.
Create an account to finish this reading and make it part of your study routine.
Continue reading freeAlready have an account? Log in
Updated