Basic Guide To Transistor Selection
Step 1 – Technology
Unlike passive components, where you usually start by checking voltage, tolerance, and power rating, transistor selection almost always begins with technology. The reason for this is that different transistor technologies behave fundamentally differently, with those differences dominating how the device can be used in a real circuit (for example, BJTs are current-controlled devices while FETs are voltage-controlled). If you pick the wrong technology, no amount of clever biasing will save you.
Thus, we arrive at the first question to ask is which is what kind of control you need. Are you controlling current, or are you controlling voltage? That single decision immediately narrows the field.
Bipolar Junction Transistors (BJTs)
Bipolar junction transistors are current-controlled devices, and one of the earliest technologies (similar to the very first point-contact transistors). In BJTs, a small base current controls a much larger collector current, which makes BJTs excellent in current amplifiers and predictable linear applications (when biased correctly). Despite their age, they are still widely used in analog signal paths, current mirrors, differential pairs, and low-noise amplification stages. Their downside, however, is that they always consume drive current, making them power hungry, and their behavior varies noticeably with temperature.
Field-Effect Transistors (FETs)
Field-effect transistors, by contrast, are voltage-controlled devices. The gate voltage controls the channel conductivity, ideally with virtually no steady-state gate current. This makes FETs attractive wherever input loading matters, or where the control signal is generated by logic rather than an analog current source. If your design has control signals in terms of voltages rather than currents, you are already in FET territory.
Transistor Polarity
Once you have chosen between BJT and FET, the next decision is polarity, and this is where things can often get confusing. N-type versus P-type determines the direction of current flow and which rail the device naturally prefers to sit on. For example, N-type devices generally perform better compared to their P-type counterparts, switching faster, having lower losses, and cheaper at a given rating. P-type devices, on the other hand, exist mostly for convenience in high-side switching or complementary designs, where N-Type variations may struggle to operate correctly.
Switching Mode
During polarity selection, it is also important to separate polarity from operating mode. "Normally on" versus "normally off" is not determined by N-type or P-type, but by whether the device is an enhancement or depletion-mode variation. Most BJTs and MOSFETs you will encounter are enhancement-mode, meaning they are off with zero drive. Depletion-mode devices exist, particularly in JFETs, but they are typically the exception, not the rule. Confusing these concepts quickly leads to incorrect assumptions that tend to surface at the worst possible time.
Device Subtypes
BJTs come in two main variations, NPN and PNP, and while each of these operate with opposing polarity (NPNs require a positive base-emitter current to work, while PNPs require a negative base-emitter current), they are extremely similar in characteristics. However, when it comes to FETs, the numerous variations have massive impact on circuit function.
JFETs and MOSFETs
JFETs are voltage-controlled but behave more like analog components as opposed to digital switching devices. They offer low noise, predictable transconductance, and graceful overload behavior, which is why they still appear in sensitive front ends and audio circuits. MOSFETs dominate switching applications because they scale well, interface cleanly with logic, and can handle large currents efficiently. Of all MOSFET variations, enhancement-mode MOSFETs are generally the default choice for digital and power switching.
Insulating Gate Bipolar Junction Transistors (IGBTs)
IGBTs occupy a strange but useful middle ground, as they combine the voltage-controlled gate of a MOSFET with the current-handling capability of a BJT. The resulting device is well suited to handle high-voltage / high-current switching where absolute speed is less important than robustness and efficiency. IGBTs almost always appear in motor drives, inverters, and industrial power electronics as well as EV systems where high DC voltages are found.
Wide-Bandgap Technologies (SiC and GaN)
Finally, there are wide-bandgap technologies such as silicon carbide and gallium nitride. These devices are NOT drop-in replacements for silicon devices in low voltage / current applications, as as such will never be used for processors and digital signal systems. Instead, they exist because silicon has real physical limits when dealing with high voltages and currents. SiC and GaN tolerate higher voltages, higher temperatures, and can switch much faster, and these factors make them perfect for power supplies and power control.
However, because they are so different to silicon-based transistors, they demand careful layout and gate drive design. This need for specialist gate driving circuits has made some engineers hesitant to use the technologies, but when seeing that SiC and GaN can see massive reductions in magnetics and higher power conversion efficiencies, the pain in using them is universally well worth it.
Transistor technology choice sets the boundaries of what is possible in a circuit. As a rough rule, use BJTs for current amplification and linear analog work, MOSFETs for voltage-controlled switching and logic interfacing, JFETs for sensitive low-level signals, IGBTs for heavy power switching, and wide-bandgap devices when silicon devices have run out of headroom. Everything else is detail, but this step is not optional.
