Low power consumption and high-density ICs are required to rapidly develop portable systems such as laptops, PDAs, digital wristwatches, mobile phones, etc. As a result, there has been an explosion of inventive research in low-power devices and design methodologies. In most situations, low power consumption needs must be balanced against the equally demanding goals of high chip density and high throughput circuits. As a result, low-power digital design and digital integrated circuits are very active domains of research and development. Reduced power dissipation in CMOS circuits is a vital challenge in this cutting-edge technology. It is most important when the size of transistors is scaled down to enhance transistor density over the silicon chip. Power dissipation reduction is another essential goal in the designs.
In a CMOS circuit, overall power dissipation may be stated as the sum of three major elements:
- Static power dissipation (due to leakage current when the circuit is idle)
- Dynamic power dissipation
2.1 Switching power dissipation
- Short-circuit power dissipation during transistor switching
Sources of Power dissipation in CMOS Circuits
A VLSI circuit’s power consumption may be divided into two categories: static power dissipation and dynamic power dissipation.
1. Static Power
is the power consumed when there is no circuit activity, often known as a quiescent mode. Even if we remove the clocks and do not modify the circuit’s inputs in the presence of a supply voltage, the circuit will still consume some power, which is referred to as static power consumption. It is mostly caused by leakage currents that flow when the transistor is off-state. When the junction diodes within the transistors are reverse biased, reverse bias leakage current occurs. When VGS Vth [Vth is the transistor’s threshold voltage], sub-threshold leakage current flows from drain to source across the channel. Leakage power dissipation in a transistor is typically inversely proportional to its threshold voltage.
2. Dynamic Power
The power consumed when the circuit is in operation, which implies we have supplied a supply voltage, a clock, and changed the inputs, is referred to as dynamic power. It is primarily due to dynamic currents such as capacitance currents (switching power) and short-circuit currents (short-circuit power).
2.1. Switching power dissipation
Switching power dissipation in CMOS circuits is a significant factor influencing overall energy consumption. It primarily occurs due to the charging and discharging of capacitive loads during transistor switching events. When a CMOS transistor switches from one state to another, there is a brief period during which both the NMOS and PMOS transistors are simultaneously conducting, resulting in a direct path between the power supply and ground, leading to high power dissipation. The amount of power dissipated during switching depends on factors such as the operating frequency, the capacitance of the load, and the voltage swing across the transistors. Minimizing switching power dissipation is crucial for improving energy efficiency in CMOS circuits, and various techniques such as voltage scaling, clock gating, and power gating are employed to mitigate this issue while maintaining performance.
2.2 Short-circuit power dissipation
As the input progressively changes, there will be a period when some of the transistors in the pull-up and pull-down networks are switched ‘ON’ at the same time, providing a short-circuit path from Vdd to GND.
So far, we have covered a variety of ways to lower power dissipation in CMOS circuits. Good design strategies usually necessitate deliberate trade-offs between many criteria:
Pt = Pstatic + Pdynamic + Pshort = total power dissipation
How to Cut Off the Power Dissipation in CMOS Circuits?
Energy-delay products can be a suitable approximation for making trade-offs like speed, area, and design time. This enables a designer to uncover improvements that yield the greatest decrease in energy for the lowest change in performance. The VTCMOS configuration can minimize static or standby power at the expense of additional bias control circuitry and substrate supply voltage- a strict need for laptops, mobile phones, and PDAs that are maintained in standby mode for a few hours.
To decrease dynamic power dissipation, voltage scaling might be used. While low operating voltage appears to be specifically appealing for low-power operation, it is extremely susceptible to manufacturing inconsistencies and operating point changes. To ensure that the circuit meets the specified speed and power requirements, the benefit of adopting technologies with very low threshold voltage vanishes.
The short circuit power is mostly determined by the rise and fall period of the driving signal. It enters the picture only if it is less than a few nanoseconds.
CMOS circuits must use adaptive control on both the threshold voltage and the supply to decrease the effective fluctuations in device characteristics to achieve high potential benefits of operating at low voltages. Until energy-efficient approaches are found, supply and threshold voltage scaling will most likely be minimal.
Conclusion
So, if you are interested in knowing more about such technical aspects of a VLSI design, Chipedge is the right place for you. Being the best VLSI training institute in Bangalore, Chipedge offers a wide range of VLSI design courses including RTL design courses, DFT courses, and much more. You can enroll yourself here if you want to make a career in VLSI or if you are looking for the best VLSI job-oriented courses in Bangalore. Get in touch with us today!
