Digital Integrated Circuits
Digital integrated circuits have three main advantages over circuits built out of discrete components: size, cost and performance. The IC’s components are printed together using photolithography on a single chip.
The operating speed of a logic circuit depends upon both the transition time and the propagation delay. The lower the transition and propagation delay times, the faster the IC.
Process
Digital ICs combine a variety of logic and memory components to perform linear or digital functions. They are also capable of carrying out arithmetic operations like addition, subtraction, integration and differentiation. The logic circuits in an IC work synchronously to a clock signal. This allows each operation to be performed in a single clock cycle, simplifying the circuits.
The design process begins with a micro-architectural specification that defines how the circuit will function. This is followed by a detailed circuit schematic using electronic design automation software that can run simulations to verify that the final product will meet its specifications.
Using the schematic, the circuit is transformed into a logic design that uses hardware description languages such as VHDL digital integrated circuits or Verilog. The logic design is then converted to a physical layout that specifies how the circuit will be constructed on a silicon wafer. This is a complex task that involves the creation of masks that specify the shapes and locations of various features on the chip. The wafer is then fabricated by depositing and etching various layers, doping and creating transistor structures.
During this stage, any problems that might affect the functionality of the finished IC can be predicted and corrected before the actual production begins. This includes issues such as added resistance from wiring, signal crosstalk and variations in the manufacturing process that might cause changes in signal voltage levels between p-type and n-type chips.
Components
The basic components of a digital circuit are logic gates. These circuits work with signals that have only two states, on and off (binary). They are designed to operate over a limited range of signal amplitudes and use the logic of Boolean algebra. ICs perform complex functions that cannot be achieved by single transistors alone, such as amplification of a voltage signal.
Standard logic ICs include flip-flops, registers, shift registers and memory units. They are the core components of a digital circuit and form the building blocks for more complex arithmetic-and-logic units. Sequential logic ICs incorporate memory elements and perform operations based on both the current input values and stored previous states. Memory ICs store data either temporarily or permanently. Random access memory, or RAM, is the most common type of memory IC. Read-only memory, or ROM, is also available in an IC.
Linear ICs, such as operational amplifiers, are used to amplify analog signals in applications like audio systems and analog ic medical equipment. Mixed-signal ICs combine digital and analog circuitry for applications that require both types of processing.
The CMOS technology used in modern digital ICs requires a regulated power supply of 5 volts. CMOS ICs are highly susceptible to static electricity, which can puncture the insulating barriers in MOSFET transistors and cause a failure of the chip. To minimize the risk of static discharge, ICs should be handled using anti-static foam for storage and transport and by frequently touching a grounded object.
Logic
Digital ICs deal with signals restricted to the extreme limits of zero and some full amount, which are interpreted as either “on” or “off.” This means they use binary logic, the electronic representation of Boolean algebra. Digital ICs are the backbone of digital electronics, as well as of digital computer processing. They are less susceptible to noise and degradation in quality than analog circuits, and they handle data faster.
The functions of digital ICs can be divided into two parts: the data path and control logic. The data path consists of components that process inputs and outputs and are synchronized to work together to complete the intended function at a given time. These components include digital memory chips, registers, counters, arithmetic units, logic gates, and flip flops.
The control logic handles the logical operations, such as “and” and “or.” It also performs error detection and correction. The control logic can be designed to operate on a single chip or on multiple chips. Multi-chip design allows designers to build a more complex circuit with millions or billions of individual components on a single die, increasing interconnection density. The resulting IC can be more robust, lower in cost, and easier to manufacture. Several different types of logic are available, including transistor-transistor logic (TTL), bipolar transistor logic (BTL), and complementary metal oxide semiconductor field-effect transistors (CMOS) logic.
Memory
A computer’s memory stores data in a form that can be accessed at any time. Each location in the memory is assigned a unique address, and by knowing that address one can retrieve its contents. While many of today’s microprocessors have relatively small amounts of memory, the human brain, in contrast, is a “memory-rich” structure. In fact, Sejnowski and others have found that even if each of the brain’s billions of synapses were to store information at only a rate equivalent to a single bit per synapse, the structure as a whole would command vast stores of memory. This suggests that the dual-store theory of memory may be useful in understanding brain function.
In this theory, short-term and long-term memory are both involved in responding to a stimulus.