A prototype that works reliably on a bench, powered from a lab supply, checked daily by the person who built it, is a genuinely different engineering problem from a unit that has to survive six months unattended outdoors. We see this gap catch people who are otherwise strong engineers more often than any single component failure, because it's a set of decisions that simply never comes up until a device actually leaves the workshop.
The enclosure is a system component, not an afterthought
On the bench, an enclosure is often whatever project box was on hand, if there's one at all. In the field, the enclosure has to actively manage heat, moisture, dust, and physical protection, and it interacts with every other design decision. A board that ran cool in an open-air bench test can overheat inside a sealed enclosure sitting in direct sun, because the enclosure itself now traps the heat the open bench setup let escape freely.
This means enclosure selection has to happen early in the design process, not at the end: sizing and material chosen with thermal behaviour, IP rating, UV resistance (direct equatorial sun degrades many plastics faster than people expect), and mounting method all considered together, not bolted on as a last step around a board that was designed without them in mind.
Connectors and cable entry are where field units actually fail
In our experience, the single most common point of field failure isn't the main board. It's connectors and cable entry points: a cable gland that wasn't fully tightened, a connector not rated for outdoor use that corrodes, or strain on a cable at its entry point that eventually works a wire loose. A bench prototype often uses whatever connector was convenient, with cables that never move and are checked constantly. A field unit's connectors need to be rated for the actual environment, properly strain-relieved, and, where possible, chosen so that a technician can service them without needing to fully disassemble the enclosure.
Thermal management without active cooling
Fans and other active cooling are rarely practical in a sealed outdoor enclosure exposed to dust and moisture, which means thermal management has to be handled passively: enough thermal mass and surface area for heat to dissipate through the enclosure itself, components and layout chosen to minimise heat generation in the first place, and enclosure colour and material chosen to reduce solar heat gain, not just for aesthetics. This has to be validated against real conditions, not a lab bench at room temperature; a unit that runs cool indoors can run hot enough in direct sun to shorten component life or cause thermal throttling that a bench test would never reveal.
Designing for maintenance, not just for the first install
A bench prototype gets opened and modified constantly during development. A field unit, once deployed, should be designed on the assumption that maintenance visits are costly (travel time, weather, access) and should therefore be as rare and as fast as possible when they do happen. That means: components most likely to need replacement (batteries, certain connectors) accessible without needing to fully open or disturb the rest of the enclosure, clear labelling inside the enclosure so a technician other than the original designer can service it correctly, and a design that fails in an obvious, diagnosable way rather than a silent one, so a maintenance visit can be triaged before the technician even leaves the workshop.
Testing under the conditions it will actually face
The last, and most commonly skipped, step is validating the finished unit under conditions that actually resemble deployment: real thermal cycling (not just a single temperature check), humidity exposure over time rather than a brief spot check, and if possible, a pilot deployment in a real but low-stakes location before committing to a wider rollout. A unit that has only ever been tested in a controlled workshop environment has not actually been tested against the failure modes that matter in the field.
The gap is engineering discipline, not exotic technology
None of this requires unusual components or techniques. It requires treating "will this survive six months outdoors, unattended" as a design requirement from the start, with the same seriousness as the sensing or communication requirements, rather than as a packaging decision made after the "real" engineering is finished.
If you have a working prototype and are trying to work out what it takes to turn it into something you can actually deploy and leave in the field, get in touch and we'll help you think through the gap. You can also see how we take hardware from prototype to the field.
