How Do PCB Designers Choose Components?

Component selection is one of the earliest engineering tasks in any PCB project. The parts chosen for a design influence firmware options, power architecture, layout constraints, manufacturability, sourcing, and the amount of technical risk in the project. Engineers sometimes treat component selection as a catalog search problem, but in practice it is a system design problem first.

The right way to approach component selection is to begin with functionality and performance requirements. In some projects, those main components are dictated by a client or by an internal stakeholder. In other projects, the engineer has freedom to choose them. In either case, the process becomes much easier when it is approached systematically.

Start With System Requirements, Not Distributor Listings

The first input into component selection is a set of design or performance requirements: these are the high level requirements that define what the product is supposed to do electrically and functionally.

At a minimum, these requirements should answer questions such as:

• What is the product supposed to do?
• What signals does it need to generate, process, or transmit?
• What interfaces are required?
• What voltage, current, bandwidth, speed, or accuracy targets must be met?
• What external systems must it connect to?

Once those questions are answered, the engineer can start building a block diagram. That block diagram becomes the bridge between product requirements and actual part selection. It identifies the major functional sections in the design and the interfaces between them, which immediately narrows the categories of components that need to be evaluated.

For example, a block diagram might show a processor connected to DDR memory, USB, Ethernet, UART, I2C peripherals, and a power management section. Even if no part numbers are assigned, the diagram already tells you what kinds of devices the design will need and what electrical standards those devices must support.

Identify the Main Value-Creating Component(s)

Most systems have one component, or a small group of components, that enable the main functionality a customer is actually paying for. This is the best place to begin the selection process. In an embedded system or IoT device, that may be the SoC or MCU. In a data acquisition system, it may be a high-speed multi-channel ADC. In a communications product, it could be an FPGA or network processor.

I often think of this as the main value-creating component. It is the component that makes the product what it is. Once that part is identified, many of the other selections become constrained by its requirements.

The main component in this Mini PC example project in Altium Designer is an FPGA, and it is the main driver for selecting all the other components in this design.

A processor may dictate memory type, power rails, oscillator requirements, boot flash, and communication peripherals. An FPGA may require multiple supply rails, configuration memory, clocking resources, DDR memory, and PHY devices. A high speed ADC may require a specific analog front end, low noise power architecture, a JESD204 interface, and a downstream FPGA or processor.

This is why component selection is usually not a flat process where every part is chosen independently. It is hierarchical. The core device drives many of the downstream selections.

Sometimes the Component is Chosen For You

In consulting work and in many in-house engineering teams, there is often much less freedom in component selection than new designers expect. The customer, program manager, firmware team, or system architect may have already decided what the main component needs to be.

There are several common reasons for this:

When this happens, the first task is not to question the selection unless there is a clear technical problem. The real task is to understand how to build the rest of the system around that part.

This is where reference material becomes valuable. If a client specifies a particular ADC, FPGA, SoC, or RF transceiver, then the next step is to determine what that device needs in order to operate correctly and meet the system requirements. That information usually comes from datasheets, application notes, evaluation boards, and reference designs.

Use Reference Designs and Eval Boards to Fill in the Blanks

For many parts like processors and ASICs, the component is never used alone. It usually requires some additional regulators, clocks, memory, passives, filters, or interface devices. One of the fastest ways to identify these supporting parts is to study reference designs and dev boards.

These resources do two useful things. First, they show what other components are needed just to make the main device boot up and run. Second, they reduce technical risk by showing a working implementation that the component vendor has already validated.

For example, if you are given a high speed ADC with a JESD204 interface, the reference design might immediately reveal:

• Required power rails and sequencing
• Recommended input filter or amplifier stages
• External reference clock or oscillator
• Interface connections to downstream logic
• Example layout practices

The reference design for this 10.4 Gbps ADC from Texas Instruments provides excellent guidance for circuit design with ADC32RF45. [Source: TRF1208-ADC12DJ5200RFEVM]

Likewise, a development board for an FPGA often shows a practical set of peripherals required to support the device in a real system. You can quickly find the complete list of DDR memory connections, high-speed interfaces, and power regulators needed to implement a complex application without ever opening the FPGA datasheet. Even if you do not copy that exact implementation, it gives you a low risk starting point for your own design.

For programmable devices, development boards also matter for another reason: they are often needed anyway for early firmware or application development. That makes them doubly useful as both a development platform and a reference design showing peripheral selection.

When Nothing is Specified, Start From Electrical Requirements

Not every project begins with a dictated part number. In some cases, the engineer has broad freedom to choose components. When that happens, the selection process should still begin with requirements, but now those requirements must be translated into searchable specifications.

Suppose a block diagram indicates that a power management IC is needed to generate several rails for an FPGA or ASIC. That high level need can then be translated into a more specific search problem:

• Number of outputs/channels
• Input voltage range
• Output current capability
• Sequencing requirement
• Buck, boost, or buck-boost topology
• Switching frequency range
• Package style
• Temperature range

At that point, some kind of component search tools (e.g., Octopart) become useful because they help narrow a large range of parts down to a manageable candidate list. The key is to search by electrical need, not by brand preference or random familiarity.

Apply Supply Chain Filters Early

A part that meets the electrical requirements but cannot be procured at the required volume is not a good part for a production design. This is why supply chain considerations need to enter the selection process early, not after the schematic is complete. Narrow candidates using sourcing filters such as (not in any specific order):

• Approved distributors
• Active lifecycle status
• In stock or acceptable lead time
• Manufacturer preference
• Packaging options suitable for assembly flow

If your organization uses an approved vendor list, then part searches should be restricted to those distributors first. In many companies, this is not optional. Procurement may already have credit terms, preferred relationships, or qualification requirements that limit where components can be sourced.

Lifecycle state is equally important. If a part is obsolete, not recommended for new designs (NRND), or already moving toward end of life (EOL), it should usually be removed from consideration early. A very simple way to do this is in Octopart, either in the main search results or in the BOM Tool: lifecycle is one of the standard filters in both areas of the platform and it allows you to immediately exclude an outdated component.

Octopart allows you to narrow down to parts that are production thanks to a dedicated lifecycle filter.

Reuse Proven Parts When Possible

One of the best ways to reduce design risk is to reuse components that have already worked in previous designs. If a part has already been validated in a similar application, there is often little reason to replace it unless there is a performance, cost, or sourcing problem.

Reusing known parts offers several advantages:

• Existing schematic circuits can often be reused directly
• Layout practices are already known
• Simulation models may already be available
• Firmware support may already exist
• Prior project experience reduces integration risk

This is especially useful in custom design environments where schedule and first pass success matter. A proven regulator, PHY, memory device, or interface IC can remove a large amount of uncertainty from a new design.

The challenge is keeping track of what has been used before and where it was used. A component management system or library with category tagging and where-used visibility is very helpful here. If you can search your design database and quickly find prior projects that used a candidate part, you can often retrieve known good circuitry and layout examples without starting from scratch.

A Practical Workflow For Component Selection

To summarize everything I’ve presented in this guide, a good component selection process can be summarized in a practical sequence:

1. Define the product’s functional and electrical requirements.
2. Translate those requirements into a block diagram.
3. Identify the main value-creating component or components.
4. Determine whether those components are specified or need to be selected.
5. Use datasheets, development boards, and reference designs to identify required peripherals.
6. Search for unspecified parts using electrical requirements and filters.
7. Apply sourcing, lifecycle, and AVL constraints.
8. Narrow to a short list and compare implementation details.
9. Prefer proven parts from earlier projects when that reduces risk.

Learn more in this video on Altium Academy.

This workflow scales well from small embedded systems to much more complex digital and mixed signal designs. The right process starts with requirements, moves through the block diagram, identifies the core devices that enable product functionality, and then uses reference material and structured filtering to select the supporting parts. That is ultimately what makes component selection an engineering discipline rather than a catalog exercise.

Whether you need to build reliable power electronics or advanced digital systems, use Altium’s complete set of PCB design features and world-class CAD tools. Altium provides the world’s premier electronic product development platform, complete with the industry’s best PCB design tools and cross-disciplinary collaboration features for advanced design teams. Contact an expert at Altium today!