Understanding the Relationship Between MOSFETs and CMOS
MOSFET stands for Metal-Oxide-Semiconductor Field-Effect Transistor. CMOS stands for Complementary Metal-Oxide-Semiconductor. They share a name because CMOS is actually built out of MOSFETs. That is the entire answer to the question most people are trying to ask. No, they are not the same thing, but they are deeply related. A MOSFET is a single electronic component — a transistor that works by using an electric field to control the flow of current through a channel. CMOS is a fabrication technology and circuit design methodology that uses pairs of MOSFETs, one P-channel and one N-channel, wired together so that one handles the high side and the other handles the low side of a signal path. I spent years designing power supply circuits and digital logic boards, and the confusion here is completely normal. People see the similar acronyms and assume they refer to the same category of thing. They don't. One is a component. The other is a way of building integrated circuits out of those components.
When someone says "CMOS" in the context of a computer motherboard, they are usually talking about the small battery-powered section that retains BIOS settings. That is a separate thing entirely from both MOSFETs and the CMOS logic family. It just happens to use the same acronym by historical accident. The battery-backed memory on your motherboard is a tiny static RAM chip. It has nothing to do with the power MOSFETs that regulate voltage to your CPU. In digital logic, a CMOS gate like a 74HC04 inverter is made by connecting a P-channel MOSFET on top and an N-channel MOSFET on the bottom. When the input is high, the P-channel turns off and the N-channel turns on, pulling the output low. When the input is low, the P-channel turns on and the N-channel turns off, pulling the output high. The key advantage is that in a steady state, one transistor is always off, so static power dissipation is nearly zero. You only draw significant current during the brief switching transition between states. This is why CMOS dominated the semiconductor industry. It replaced TTL and other logic families because it consumed a fraction of the power at comparable speeds. Modern processors contain billions of these CMOS transistor pairs. A high-end CPU can have over a hundred billion transistors now, all built using CMOS process technology.
There is a practical issue that comes up when you are working with both concepts simultaneously. I once designed a motor controller using discrete power MOSFETs and ran into a problem where the MOSFETs were switching fine but the control circuitry was resetting unpredictably. The issue traced back to voltage spikes from the motor inductance coupling into the CMOS gate inputs of my microcontroller. The spikes exceeded the absolute maximum gate-source voltage rating of the CMOS pins, which is typically only 5.5V for a 5V device. I solved it by adding a Zener clamp and a series resistor on each gate pin, which limited the transient current and clamped the voltage to a safe level. Another counter-intuitive point that beginners miss is that CMOS technology is not inherently better than other transistor technologies in every situation. For high-frequency analog applications, bipolar junction transistors still outperform CMOS in terms of noise figure and linearity. RF front ends in radios frequently use GaAs or InP devices instead of CMOS because the electron mobility in those materials is significantly higher. CMOS also has issues with latch-up, a condition where parasitic thyristor structures within the silicon can turn on and create a short circuit between power and ground. This is why CMOS processes include guard rings and substrate contacts to minimize the risk. The tradeoffs are worth understanding. CMOS circuits have very high input impedance, which means they are sensitive to electrostatic discharge. A human touching a CMOS IC without grounding can easily deliver thousands of volts through their fingertip, destroying the gate oxide in nanoseconds. That is why proper ESD handling procedures exist for a reason. Power MOSFETs are similarly vulnerable on their gates, though they can usually handle a bit more charge before failure.
Get the Full Details
When you see "MOSFET" on a component datasheet, you are looking at a switch or amplifier. When you see "CMOS" on a chip package, you are looking at a complete logic circuit built from many MOSFETs arranged in complementary pairs. The relationship is similar to asking whether a resistor and a radio are the same thing. One is a fundamental component. The other is a system built from those components and others. For most practical purposes, if you are replacing a failed component in a power supply, you need a MOSFET. If you are designing a digital circuit or reading about how modern microprocessors are manufactured, you are dealing with CMOS technology. Both rely on the same underlying physics — a voltage-controlled channel in a semiconductor with an insulating oxide layer — but they operate at completely different levels of abstraction and complexity.
