Product Development

Developing an electronic product is rarely just a case of designing a schematic, laying out a PCB and sending it for manufacture.

In practice, there are many decisions made early in the project that can have a big impact later on: component selection, power architecture, PCB layout, EMC, testability, manufacturability and even how easy the product will be to troubleshoot once prototypes arrive.

From my experience, the earlier these things are considered, the easier the rest of the development tends to be.

Start With the System, Not the PCB

One of the easiest mistakes to make is to jump into the schematic too early.

Before getting into detailed circuit design, it is worth defining how the complete system is expected to work. What are the main power rails? What interfaces are needed? Which signals are sensitive? Are there noisy loads, motors, relays or switching regulators? What environmental conditions will the product operate in?

These questions affect the design from the start.

A clear system architecture also makes it easier to separate analogue, digital, power and communication sections and to identify potential problems before they are built into the PCB.

Schematic Reviews Are Worth the Time

A schematic review is not just about checking whether the circuit is electrically correct.

I normally look at things such as:

• power supply margins;

• component ratings;

• protection against abnormal conditions;

• unused inputs and floating signals;

• pull-ups and pull-downs;

• filtering;

• connector interfaces;

• programming and debugging access;

• component availability;

• and whether the circuit will behave sensibly during power-up and power-down.

Many prototype issues come from small details rather than major design errors.

Finding those details during a review is much easier than trying to understand them later with an oscilloscope on a populated PCB.

PCB Layout Is Part of the Circuit

A schematic can be completely correct and the PCB can still perform badly.

This is especially true when the design contains switching power supplies, fast digital signals, sensitive analogue measurements, communications interfaces or high-current loads.

Track length, return paths, grounding, decoupling and component placement all affect how the circuit behaves in the real world.

I have seen cases where the circuit itself was fine, but a relatively small layout decision created noise, EMC or signal quality problems.

For that reason, I see PCB layout as part of the electrical design rather than simply the stage that comes after the schematic.

Think About EMC Early

EMC is much easier to design for than to fix later.

A product can work perfectly during development and then fail when it reaches formal EMC testing. At that point, the team is normally under pressure because the PCB has already been manufactured and mechanical parts may already be fixed.

Good EMC performance starts with basic design decisions: PCB stack-up, grounding, return paths, filtering, shielding, cable interfaces and the placement of noisy circuits.

My experience working with EMC testing has made me much more cautious about these details during the design stage.

It is always better to ask, “Where will this current return?” or “How could this noise couple into another part of the circuit?” before the PCB is manufactured rather than after the first failed test.

Design for Testing and Debugging

Something else that is often underestimated is how the board will actually be tested.

During development, having access to the right signals can save a huge amount of time.

Simple things such as test points, accessible programming connectors, status LEDs or the ability to isolate parts of the circuit can make debugging much easier.

If a board is completely closed off and the only way to inspect a signal is to solder a wire onto a small component, development becomes unnecessarily difficult.

I normally try to think about what I would need if I had to troubleshoot the board six months later without remembering every detail of the design.

Component Selection Is More Than the Datasheet

Choosing a component because it meets the electrical specification is only part of the decision.

Availability, lifecycle, lead time, package size and alternative parts can become just as important.

This is particularly relevant when moving from prototype quantities into production.

A component that works perfectly but becomes unavailable can force a redesign at the worst possible point in a project.

Where possible, it is worth checking sourcing risks early and avoiding unnecessary dependency on difficult-to-replace components.

Prototype Testing Should Try to Break the Design

Once the first prototypes arrive, I do not think testing should simply confirm that the product works.

It should try to find where it stops working.

That means looking at operating limits, unusual power conditions, communication faults, temperature effects, noisy environments and other situations that may not happen during a normal bench test.

Real products are rarely operated under perfect laboratory conditions.

Finding weaknesses during development gives you the opportunity to fix them before the customer finds them.

Documentation Matters More Than It Seems

Engineering documentation is often treated as something to complete at the end of the project.

In reality, keeping design decisions, calculations, test results and known issues documented during development makes the project much easier to manage.

It is particularly valuable when a design needs to be modified later or when another engineer has to understand why a particular decision was made.

Good documentation does not need to be excessive. It just needs to contain the information that will matter when someone comes back to the design months or years later.

Final Thoughts

There is no single step that guarantees a successful electronic product.

Good results normally come from paying attention to lots of smaller engineering decisions throughout the development process.

For me, some of the most important areas are:

• getting the system architecture right before starting detailed design;

• reviewing schematics and PCB layouts before manufacture;

• considering EMC from the beginning;

• designing the product so it can actually be tested and debugged;

• and testing prototypes beyond normal operating conditions.

A few extra hours spent thinking about these things early in the project can often save much more time later.

That is also the approach I try to bring to projects through ElectronPath: practical engineering support focused on finding problems early and helping move a design towards a reliable, manufacturable product.