YURUNOX / IC fundamentals and selection

What Is an Integrated Circuit? Types, Uses, and Selection Basics

An integrated circuit (IC) is an electronic circuit whose transistors and interconnections are fabricated together on semiconductor material, usually silicon. The die performs the function; the package protects it and provides connections; the PCB links it with the rest of the system. To choose an IC, identify the function, signal type, operating limits, package, and complete orderable part number.

These distinctions prevent common mistakes: a familiar package does not prove compatibility, a product-family name does not identify an exact device, and a logic pin does not automatically have enough current to drive a load.

By YURUNOX · For electronics learners, engineers, and component buyers
Sources reviewed:

Exposed integrated circuit die with fine bonding wires connecting it to the surrounding package
The die contains the integrated circuit. Fine bond wires connect this die to its package; other devices use different interconnection methods. Photo: Mister rf, Wikimedia Commons, CC BY-SA 4.0. Uncropped.

What Is Inside an IC—and How Is It Different from a Package or PCB?

The integrated circuit is fabricated on the semiconductor die. The package protects that die and provides usable electrical and mechanical connections. The printed circuit board connects the packaged IC with power, connectors, discrete parts, and other ICs.

Separate the concept before selecting a part
QuestionWhat it identifiesWhat it does not prove
Is it an IC?A circuit integrated on semiconductor material.Its function, electrical limits, or suitability.
What signal type?Analog, digital, or mixed-signal behavior.The product family or exact application.
What function?Amplifier, logic, processor, memory, power, interface, sensor, or driver role.Pin, package, voltage, timing, or firmware compatibility.
Which orderable device?The full manufacturer part number, suffix, package, grade, and packing option.That an offered lot is approved until source and lot evidence are reviewed.

On a narrow screen, scroll the table sideways to compare all columns.

How Do the Die, Package, and PCB Fit Together?

The die is the small piece of semiconductor carrying the fabricated circuit. Transistors, other integrated elements and patterned interconnections form its working structure. These are not separately packaged miniature components glued together.

Die
The semiconductor circuit itself: devices and interconnections fabricated in and on the material.
Package
The protective and connection structure around the die, with leads, pads or balls for mounting and electrical access.
PCB
The printed circuit board that connects packaged ICs, discrete components, connectors and other parts.

A package also contributes to heat flow and mechanical reliability. A black rectangular part on a board may therefore look simple while hiding a complex circuit inside.

One package does not always mean one die. Modern packages can combine multiple dies or chiplets, each contributing a function. Intel describes this approach in its advanced packaging overview. The familiar single-die IC is the starting point, not a rule for every product.

Language note: “chip” and “microchip” often mean an IC in everyday use. For drawings, inspections, and purchase records, distinguish the bare die, packaged device, and complete PCB assembly.

02 / Inside the circuit

How Does an Integrated Circuit Work?

An IC uses interconnected electronic devices to control voltages and currents. Transistors are central to many designs: depending on the circuit, they can act as switches or control a signal continuously. Integrated resistors, capacitors and other structures support the intended function.

In a digital circuit, specified voltage ranges represent logical states such as 0 and 1. There is no universal voltage that means “1” for every chip. In an analog circuit, the changing value of a voltage or current carries information—for example, the output of a sensor.

Microscope view of a ULN2003 transistor-array die with repeated circuit structures and metal interconnections
A ULN2003 transistor-array die under a microscope. The repeated structures illustrate integration; this is a real die image, not a diagram of a processor. Image: Zeptobars.ru, Wikimedia Commons, CC BY 3.0. Uncropped.

Power and information are separate concerns. A chip needs a suitable supply to operate, while its input and output pins carry signals or deliver a specified load current. Connecting power does not guarantee correct operation: reset, enable, clock, configuration and external circuitry may also matter.

This is why a useful IC still needs a suitable circuit around it. Decoupling capacitors support the supply; other designs need external resistors, inductors, crystals, protection devices or power transistors. “Integrated” does not mean “works alone.”

03 / Signal categories

What Are the Main Types of Integrated Circuits?

Analog, digital and mixed-signal describe the kinds of signals a circuit handles. They are a different classification from job names such as microcontroller or voltage regulator.

Start with the signal, then identify the function
TypeWhat it handlesTypical jobsQuestions to ask
Analog ICContinuously varying voltages or currentsAmplification, filtering, voltage regulationWhat input range, noise, accuracy and bandwidth are needed?
Digital ICDefined logical states and digital dataLogic, computation, timing and storageAre the logic levels, timing and interface compatible?
Mixed-signal ICBoth analog and digital signalsData conversion, sensor readout, audio conversionWhat analog performance and digital interface are required?

On a narrow screen, scroll the table sideways to compare all columns.

Analog example: the LM358B operational amplifier

An operational amplifier, or op amp, can amplify or condition a signal using an appropriate feedback circuit. Texas Instruments specifies the LM358B as a dual op amp with a 3–36 V supply range. That number does not mean every input or output voltage can reach both supply rails.

The suffix matters: TI lists a different supply limit for older LM358 versions in the same family document. A buyer should check the exact version; a designer should also check input common-mode range, output swing and the conditions behind each specification. See the TI LM358 family datasheet.

Digital example: the SN74HC00 NAND gate

The SN74HC00 contains four two-input NAND gates and has a recommended supply range of 2–6 V. For each gate, the output is low only when both inputs are high; otherwise it is high. No user firmware is needed to perform this logic function.

However, a “high” from another device must meet the gate’s input threshold. A shared connector or nominal supply label is not enough to establish compatibility. The TI SN74HC00 datasheet defines the relevant limits and timing conditions.

Mixed-signal example: converting a sensor voltage into data

An analog-to-digital converter (ADC) represents an analog input with a digital code. A digital-to-analog converter (DAC) performs the opposite type of conversion. Both connect analog behavior with digital information, as described in Analog Devices’ mixed-signal definition.

A microcontroller containing an ADC includes mixed-signal circuitry even though a distributor lists it under “microcontrollers.” Product categories are useful labels, not mutually exclusive descriptions of everything inside a chip.

04 / Functions and applications

Which IC Family Fits Each Electronic Function?

A circuit designer usually starts with a task: read a sensor, hold data, control a motor or communicate with another board. Functional families help turn that task into a search for suitable parts.

Match the job to the family
IC familyTypical roleWhat needs attention
Microcontrollers and processorsRun instructions, make decisions and coordinate a system.Processing needs, memory, peripherals, firmware tools and power use.
MemoryStore working data, program code or retained settings.Capacity, interface, retention and, where relevant, write endurance.
Power-management ICsRegulate, convert, sequence or monitor electrical power.Input range, load current, efficiency, thermal conditions and external parts.
Interface and communication ICsTranslate levels or provide an electrical link between devices.Protocol role, voltage levels, termination, protection and isolation needs.
Sensor and signal-chain ICsSense a physical quantity or condition and convert a measurement.Accuracy, bandwidth, noise, calibration and operating environment.
Driver ICsProvide controlled drive for LEDs, motors or power-switch gates.Load behavior, voltage, current, protection and heat dissipation.

On a narrow screen, scroll the table sideways to compare all columns.

How Do MCUs, MPUs, GPUs, Memory, and Drivers Differ?

A microcontroller (MCU) combines a processor core, memory and peripherals for embedded control. For example, ST’s current STM32F103C8 product page lists an Arm Cortex-M3 core, a 72 MHz maximum frequency, on-chip memory, converters, timers and communication interfaces. Those facts apply to the named product page, not every STM32 device. Check the STM32F103C8 product page and its orderable variants before selection.

A microprocessor (MPU) focuses on processing and often works with external memory and system-support ICs. A graphics processing unit (GPU) is designed for highly parallel work. These are functional distinctions, not a simple ranking in which one category is always better.

Memory technologies are not interchangeable. Static RAM (SRAM) and dynamic RAM (DRAM) are volatile working memories; flash and electrically erasable programmable read-only memory (EEPROM) retain data without normal operating power. Capacity alone does not establish compatibility: the interface, operating voltage and usage limits also need to match.

A controller and a driver do different jobs. A controller decides what should happen; a driver provides the electrical drive needed to make it happen. Likewise, a CAN or RS-485 transceiver is not automatically a complete protocol controller. Buying the right category is only the beginning.

Not every IC needs user programming. Basic gates, amplifiers, and many regulators work from their circuit conditions. MCUs normally execute firmware, FPGAs require hardware configuration, and some devices only need register settings. Confirm the actual startup and configuration requirements.

Documented product example

Arduino UNO R3: one board, several different IC jobs

The UNO R3 is a useful public example because Arduino publishes its board documentation. Its ATmega328P microcontroller runs the application. A separate ATmega16U2 handles the USB connection, while a regulator supplies a regulated voltage rail. The board also needs capacitors, an oscillator, connectors and copper traces.

The lesson is practical: the board is a system; no single chip is the whole system. Replacing the application MCU would not, by itself, fix every USB or power problem. This is an explanation of the published design, not a YURUNOX test or customer project. Sources: Arduino UNO R3 overview and board datasheet.

Arduino UNO R3 printed circuit board carrying a microcontroller, supporting integrated circuits, connectors and discrete components
An Arduino UNO R3 board makes the distinction visible: packaged ICs are components mounted on a larger PCB assembly. Photo: Arduino.cc, Wikimedia Commons, CC BY-SA 4.0. Uncropped.

Where Do ASICs, FPGAs, and SoCs Fit?

These names describe purpose or architecture. An ASIC is an application-specific integrated circuit: it is designed for a particular application and may include analog, digital or mixed-signal functions.

A field-programmable gate array (FPGA) contains configurable logic and interconnections. Configuring that hardware is different from simply loading a program into a conventional MCU. See Analog Devices’ FPGA definition.

A system-on-chip (SoC) integrates several system functions, such as processing and interfaces. It can still require external memory, power management and other support components. These labels overlap: describing a chip as an SoC does not tell you its package, supply voltage or complete external circuit.

05 / A worked system example

How Do Several ICs Work Together in a Real System?

Consider a temperature-controlled fan. This illustrative architecture shows how IC functions fit together; it is not a tested reference design or a customer case study.

An analog sensor produces a changing signal. Conditioning circuitry may filter or scale it, an ADC converts it into a code, and an MCU uses that code to decide how the fan should run. A suitable driver or power stage handles the load.

Information path: measurement → decision → action
  1. SensorTemperature becomes an analog signal
  2. Conditioning + ADCSignal becomes a digital code
  3. MCUControl logic decides the response
  4. Driver / power stageElectrical drive is applied
  5. FanThe physical load responds

Separate power path: the power supply and regulators feed the required rails. Power delivery is not another step in the measurement signal chain.

Functional blocks, not a wiring schematic. Some blocks may share one IC; others require several components.

The implementation can change. A digital temperature-sensor IC may integrate the sensing and conversion stages, while an MCU may already contain the ADC. A fan may include its own motor-drive electronics. In every case, check the actual electrical interface instead of assuming the blocks must be separate chips.

A logic pin is not automatically a load-power output. Do not connect a fan motor directly to an MCU pin simply because the control diagram shows an arrow between them.

A 12-bit ADC example: resolution is not accuracy

Suppose an ideal ADC maps a 0–3.3 V input span into 12-bit codes. Twelve bits provide 212, or 4,096, possible codes. The nominal code width is the input span divided by the number of codes.

Nominal code width = input span ÷ 2N

3.3 V ÷ 4,096 ≈ 0.806 mV per code

Idealized example—not a specification for the STM32 device above. Use the chosen ADC’s full-scale definition and transfer function.

That result does not promise ±0.806 mV measurement accuracy. Sensor error, reference error, noise, analog conditioning, converter errors and calibration can all affect the answer. More bits make the code steps finer; they do not automatically remove those errors. For the underlying conversion concepts, see Analog Devices’ ADC and DAC chapter.

How Does the Same Signal Chain Change Across Applications?

In an industrial sensor, signal conditioning, conversion and communication must work together. In an appliance, the control logic must work with power supplies and load drivers. In test equipment, precision analog stages and references can matter as much as digital processing. Phones and computers add much more computation, memory and communication, but they still depend on power and interface circuits.

The application changes which constraints dominate. An automotive or industrial design may need specific temperature ranges, qualification evidence and lifecycle support; a part is not suitable merely because its basic function sounds right.

06 / From wafer to packaged part

How Are Integrated Circuits Manufactured and Packaged?

Manufacturers create many circuits across a semiconductor wafer. Repeated processing steps form device regions, insulating layers and interconnections. Lithography transfers patterns; other steps deposit, remove or modify material.

Round silicon wafer, the substrate used to fabricate many semiconductor dies
A silicon wafer is a fabrication platform, not a finished board-level component. Photo: Inductiveload, Wikimedia Commons. Public domain. Uncropped.

What Happens Between Wafer Processing and Final Test?

  1. Form the circuits through repeated patterned processing steps.
  2. Test on the wafer to assess the fabricated devices.
  3. Separate and package dies to create usable component formats.
  4. Test the packaged devices against the product’s specified requirements.

Exact sequences depend on the technology and package. ASML’s chip-manufacturing explanation provides a more detailed process overview.

For a buyer, the finished ordering code matters more than the wafer’s appearance. Devices from a broad family may differ in package, memory, temperature grade or other options. The manufacturing story explains where chips come from; the datasheet and ordering table identify what to buy.

07 / Integration and its limits

When Should a Design Use ICs Instead of Discrete Components?

A discrete component is an individual device, such as a separately packaged transistor, diode or resistor. An IC brings multiple circuit elements together. That can reduce board area, assembly connections and the effort required to implement a complex function.

Integration can also provide useful internal matching and short interconnections. But it is not a blanket guarantee of lower cost, greater speed or better performance. A highly integrated device may add software work, concentrate heat or make a future substitution harder.

Discrete parts remain valuable when the design needs particular power handling, component values, protection arrangements or flexibility. Large energy-storage components and some power stages stay outside the IC. Most practical boards combine both approaches.

There is a serviceability trade-off, too. A failed internal element is generally not repaired individually; the IC is replaced. If that part becomes unavailable, changing the circuit may involve electrical, mechanical and firmware work. Evaluate integration at the whole-system level, including lifecycle and supply continuity.

08 / From understanding to selection

How Should Engineers and Buyers Select the Exact IC?

Begin with the required function and the circuit’s operating conditions—not a package photograph or an abbreviated marking. Two ICs can look nearly identical and still have different pinouts, voltage limits, timing or firmware requirements.

Several surface-mount integrated circuit packages with different lead arrangements, shapes and sizes
Package appearance helps identify the mounting format. It does not establish electrical function or interchangeability. Photo: NobbiP, Wikimedia Commons, CC BY-SA 3.0. Uncropped.

Which Datasheet Sections Should You Check First?

  1. Confirm the function and block diagram.Check what is integrated, what is configurable and which external components remain necessary.
  2. Match pins and the package drawing.Verify dimensions, pitch, exposed pads, pin assignments and the PCB footprint—not just the pin count.
  3. Use recommended operating conditions.Check supply, input and temperature limits for normal use. Absolute maximum ratings are stress limits, not operating targets.
  4. Read electrical limits with their conditions.Distinguish minimum, typical and maximum values. Check the supply, load and temperature at which each applies.
  5. Review implementation requirements.Include decoupling, layout, clocks, startup, thermal design and software. Ambient temperature and junction temperature are not the same thing.
  6. Resolve the full ordering code.Confirm package, grade, memory or feature variant, packing format and lifecycle status before approving the purchase.

TI explains the stress-limit distinction in “Understanding the absolute maximum ratings of an MCU.”

When Is an IC Replacement Actually Interchangeable?

A replacement needs more than the same headline function. Check form, fit and function, including startup behavior, logic thresholds, analog ranges, timing and software compatibility. A new part number should not enter a production order as an unreviewed “equivalent.”

The earlier examples show why: an op-amp family suffix can change a supply limit; an MCU variant can change available memory. Even when the package fits, the circuit may need a different operating point, configuration or qualification step.

What Must a Component Sourcing Request Specify?

An actionable request includes the manufacturer’s full part number, quantity, delivery target, permitted alternatives and any agreed packaging, date-code or traceability requirements. Separate engineering approval of the device from the commercial decision about a particular offered lot.

For sourcing discussions, review YURUNOX’s quality-assurance information and purchasing process. Delivery destinations and handling requirements can be discussed alongside shipment arrangements; they do not replace technical part approval.

What Should You Check Before Blaming the IC?

Confirm the exact part and orientation, inspect solder joints, then measure supply and ground conditions. Next check enable, reset and clock requirements, followed by inputs, outputs, loads and external components. Review configuration or firmware where applicable.

An amplifier clipping near a rail may be outside its usable signal range; an apparently silent MCU may still be held in reset. These are diagnostic possibilities, not proof that the IC is good or bad. Record measurements, conditions and lot identity before drawing a conclusion or requesting a quality investigation.

What Information Should Buyers Send Before Requesting an IC Quote?

Know the job, check the operating conditions and confirm the complete ordering code. When you are ready to discuss sourcing, share your part number or BOM, quantity, delivery target and approved-alternative requirements with YURUNOX.

Where Can Readers Verify the IC Specifications and Examples?

Manufacturer specifications apply to the named devices and their stated test conditions. These references were checked on September 5, 2026; use the latest datasheet for the exact ordering code before design release or purchase.

  1. ASML — How microchips are madeSemiconductor fabrication, patterning and the manufacturing sequence.
  2. Intel — Packaging and test researchMulti-die integration and advanced package architecture.
  3. Texas Instruments — LM358 family datasheetDual operational amplifiers and variant-specific specifications.
  4. Texas Instruments — SN74HC00 datasheetNAND logic, recommended operating conditions and timing.
  5. STMicroelectronics — STM32F103C8Processor core, maximum CPU frequency, integrated memory, peripherals and orderable variants.
  6. Arduino — UNO R3 documentation and board datasheetThe public board example and its supporting circuitry.
  7. Analog Devices — Mixed-signal definition and analog-to-digital and digital-to-analog conversionSignal categories, quantization and conversion concepts.
  8. Analog Devices — FPGA definitionField-programmable logic architecture.
  9. Texas Instruments — Understanding absolute maximum ratingsStress limits versus recommended operating conditions.

Image credits and licensing links appear with each photograph. Product names identify public technical examples and do not imply manufacturer affiliation or endorsement.

Cart (0 items)