Saturation Region
The saturation region is the operating condition of a bipolar junction transistor in which both the base-emitter junction and the base-collector junction are forward biased, so the collector can no longer pull in current in proportion to the base current and the voltage across the device collapses to a small residual value, typically about 0.1 to 0.3 volt between collector and emitter (written VCE(sat)). In that state the transistor behaves like a nearly closed switch: the load current is set almost entirely by the external supply and the load resistance rather than by the transistor, which is exactly the property digital logic, relay drivers and power switches rely on when they need a device that is firmly "on."
Saturation happens because the base is driven harder than the collector circuit can use. In the active region the collector current is roughly the base current multiplied by the current gain β, but a load resistor limits how much collector current can actually flow: as the collector current approaches the supply voltage divided by the load resistance, the collector voltage falls until it drops below the base voltage and the base-collector junction starts to conduct too. Past that point, extra base current no longer raises the collector current appreciably; it just stores more charge in the base. The ratio of collector current to base current in this regime, sometimes called the forced beta, is lower than the datasheet β, and a common switch-design rule is to choose a forced beta of about 10 so that the transistor is safely saturated across temperature and part-to-part spread in gain.
Designing a saturated switch is therefore a matter of choosing the base resistor generously, not precisely. For an NPN device switching a 10 milliampere load from a 5 volt logic output, a forced beta of 10 calls for about 1 milliampere of base current; with roughly 0.7 volt dropped across a silicon base-emitter junction, a base resistor near (5 - 0.7) volts divided by 1 milliampere, or about 4.3 kΩ, is the target, and the next lower standard value is a sound pick. The payoff is low dissipation, because the power wasted in the transistor is only VCE(sat) × IC, a few milliwatts here, whereas the same current passed in the active region at a large collector-emitter voltage would waste far more. The price is speed and a little drive current, which brings in the next effect.
The main speed penalty of saturation is storage time. While both junctions conduct, excess minority carriers pile up in the base and collector regions, and they must be swept out or must recombine before the collector voltage can start rising when the drive is removed. This delay, the storage time, can be much longer than the rise and fall times of the same transistor and is the reason a deeply saturated switch turns off slowly. Engineers shorten it by driving less hard, by pulling base charge out with a negative turn-off current or a speed-up capacitor, or by preventing deep saturation altogether: a Baker clamp, a diode path that diverts excess base drive to the collector, and Schottky-clamped TTL logic, whose Schottky diode across the base-collector junction conducts at a lower voltage than the junction itself, are classic ways to keep a transistor at the edge of saturation.
The word "saturation" means nearly the opposite thing for a MOSFET, and this is the most common naming trap in electronics. In a bipolar transistor, saturation is the fully-on, switch-closed region with a small voltage drop. In a field-effect transistor, saturation is the region where the drain voltage is high enough that the channel pinches off near the drain and the drain current becomes nearly independent of drain voltage, controlled by the gate voltage: that is the amplifying region, the counterpart of the bipolar active region. The MOSFET's switch-closed region, where it acts like a voltage-controlled resistor with a small drain-source voltage, is called the triode or ohmic region. A designer who reads "saturation" in a MOSFET datasheet while thinking of a bipolar switch will predict the wrong operating state, so it pays to check which device family a text or simulator is describing.
Saturation remains relevant in modern practice even though most digital chips now use MOSFETs. Bipolar saturation still governs discrete switching transistors, relay and LED drivers, the output stages of many analog parts and the behaviour of older TTL families, and VCE(sat) is a headline specification for power bipolar devices and IGBTs, whose on-state voltage drop sets conduction loss. A frequent pitfall is assuming that a transistor shown with a large base current is simply "amplifying more": once saturated, the output stops following the input, signals clip, and recovery is slower than the datasheet gain suggests. Another is forgetting that VCE(sat) rises with collector current and with a weakly driven base, so a switch that sits at 0.1 volt at 1 milliampere may sit at several tenths of a volt near its maximum rating.
Read saturation region as the bipolar transistor's closed-switch state, entered when both junctions conduct and the load rather than the device sets the current, bought at the cost of stored charge and slower turn-off, and never to be confused with the MOSFET's identically named but opposite, amplifier-like region.