A product’s electronics work correctly during bench testing. The firmware performs as expected. The battery, sensors, mechanical parts, and connectivity features all pass their individual checks.
Then the team assembles everything inside the final enclosure, switches the product on, and discovers problems that did not exist before.
This is a common but often underestimated stage of new product development. Once separate components and subsystems are brought together, they create a new operating environment. Heat accumulates, airflow is restricted, electromagnetic interference may appear, mechanical clearances tighten, and access to important test points can disappear.
In this episode of China Manufacturing Decoded, Adrian and Paul Adams examine why apparently successful components can fail when integrated into a complete product. They also explain how staged integration, early testing, accessible debugging points, and realistic user testing can prevent these problems from derailing a project.
Listen to the audio here or on Apple Podcasts · Spotify · Amazon Podcasts · Deezer · iHeartRADIO · TuneIn.
Episode Summary
- 00:00:21 – Why individually working parts can fail together
Electronics, firmware, power systems, wiring, mechanical parts, and the enclosure may all work independently. Combining them creates new interactions and operating conditions. - 00:02:28 – Why everything seems to work on the test bench
Bench testing gives engineers plenty of space, controlled power, monitoring equipment, test probes, and easy access to the product’s electronics. - 00:07:40 – What changes when product integration begins
Integrating systems too early, before each one has been properly validated, can make it extremely difficult to determine which subsystem is causing a fault. - 00:08:39 – Products must be designed so engineers can debug them
PCBs need accessible test points and connections that allow engineers to monitor signals, update firmware, and investigate faults after components have been installed. - 00:10:16 – Integrate one subsystem at a time
Adding several PCBs, sensors, motors, or other systems simultaneously makes troubleshooting unnecessarily difficult. A staged build allows the team to identify exactly when a problem appears. - 00:13:45 – A technically compliant product can still fail
A product may meet its written specification and applicable safety limits but still provide a poor or even alarming user experience. - 00:15:00 – How heat buildup creates hidden product problems
A kitchen appliance discussed in the episode remained within its assumed technical limits, but its surface became too hot for users to handle comfortably. This illustrates why regulatory or specification compliance should not be treated as the only measure of success. - 00:20:19 – The first assembled product is only a mini milestone
Building the first complete, functioning unit is an important achievement. However, it should be treated as the beginning of integrated system testing rather than proof that the design is finished. - 00:21:39 – Solving an enclosure airflow problem
Paul describes a recent project where heat had nowhere to escape from the enclosure. Adding an airflow path and a very small fan reduced the temperature substantially. This issue would not have appeared while the heating subsystem was being tested by itself on the bench. - 00:23:53 – A five-step product integration playbook
Paul concludes with five practical rules for managing integration:
1. Integrate the product in stages.
2. Test important risks early.
3. Keep debugging access available.
4. Test the product as customers will actually use it.
5. Treat the enclosure as an active part of the complete system.
The Main Lesson
Testing components separately is essential, but it does not prove that the finished product will work correctly.
The enclosure affects airflow, temperature, signal behaviour, component positioning, accessibility, and the way users interact with the product. It should therefore be considered part of the system rather than simply a cosmetic shell added at the end.
Teams should also avoid treating the first complete build as a finished product. It is the first opportunity to observe how all of the subsystems interact under realistic conditions.
Problems found at this stage may still require changes to the electronics, firmware, component selection, internal layout, materials, ventilation, or enclosure design. Discovering and correcting those issues before tooling and production is far less damaging than finding them after products reach customers.
Further Reading
- How Many Prototypes Are Needed Before We Get ‘Perfection?’
- Transitioning to Manufacturing from Product Development | 2 Options
- Prototype Success Is NOT Production Readiness
- Why a Good Prototype Can Still Fail in Production
- Final Prototype of Your New Hardware Product: When Will You Get There?
- How Many Prototype Iterations and Safety and Reliability Tests Do We Need?
- Handover to Manufacturing: What Not to Do and Best Practices
- Test to Failure: Why You Need This Reliability Test
