Functional safety is only as strong as its weakest component

Blake Shields of Tapeswitch explores why “basic” safety parts can undermine great machine design and how to avoid it.

Machine builders have come a long way in how they think about functional safety. It’s no longer viewed as a last-minute compliance hurdle. Done well, it supports uptime, usability, and customer trust, and can even become a competitive advantage.

But, the best system architecture in the world can be fatally undermined by the wrong “small” component. Meanwhile, the components most likely to be treated as interchangeable are often the ones that sit right on the interface between people and moving machinery, such as pressure-sensitive safeguarding and sensing devices.

Safety edges, mats, bumpers and pressure-sensitive switches can look deceptively simple. They’re often sourced late, chosen from a catalogue, and bolted on as if they were generic. Yet in the field, they operate in the harshest, most unpredictable part of the machine’s ecosystem, where moisture, contamination, abuse, vibration, poor routing, and user behaviour combine to expose weak assumptions.

This article looks at why “good enough” safety components frequently aren’t, and how consulting with a reputable component supplier at the earliest stage can prevent expensive re-engineering later – and help avoid the most serious outcome of all: a safety failure.

The hidden failure mode: when the design is correct but the component is wrong

In practice, “wrong” rarely means the component doesn’t work at all. More commonly, it means:

• it works initially, but fails prematurely in real conditions
• it triggers nuisance stops that drive operators to bypass it
• it degrades performance over time (sealing, actuation characteristics, wiring fatigue)
• it creates integration problems (mounting, cable exit, connector positioning) that only appear once the machine is in service

That’s why taking the time to specify components is about preventing future problems, not satisfying a quick requirement now. It’s not just “a part”, it’s a vital safety function that must behave safely in the long term.

Pressure-sensitive devices: the three questions that matter

Whether you’re designing a robotic cell, an automated conveyor interface, a mobile platform, or a guarded machine enclosure, pressure-sensitive devices tend to succeed or fail on three fundamentals:

1) What will the environment do to it?

Water ingress, UV exposure, oils, swarf, cleaning chemicals, temperature swings and corrosion don’t show up on a purchase order. But they define the lifespan of a component.

A pressure-sensitive edge used outdoors (for example on access equipment or external automated systems) lives and dies by sealing quality, cable exit design, and protection of connectors. Once moisture gets in, the “safety device” can turn into a reliability problem that creates downtime, service costs, and customer frustration. This is the exact opposite of functional safety as a competitive advantage.

2) How will people behave around it?

If the component nuisance-trips, gets damaged easily, or looks like an obstacle, you’ve just designed an incentive for operators to bypass it.

This is common on mobile and semi-autonomous equipment: a device that initially meets the requirement but doesn’t survive knocks, impacts, or interference can lead to repeated faults – and repeated faults change human behaviour. The moment a safety function becomes “the thing that stops us working”, it becomes a target.

3) How is it being integrated mechanically and electrically?

This is where many “catalogue buys” go wrong. The device might be fine. The integration is not. Cable routing, strain relief, connector position, mounting channels, tolerances, corner treatments, overtravel space, and actuation behaviour under load determine whether the safety function works as intended – not just on day one, but after thousands (or millions) of cycles.

Pressure-sensitive switching systems are often designed to be fail-safe in combination with suitable control units. But that outcome is a system property, not a component property. And these details matter.

Sensing edge sense

It’s easy to think of a sensing edge as “a rubber strip that stops motion.” For machine builders, the engineering reality is broader:

• Actuation force and actuation distance: does it trigger early enough for your stopping time and hazard?
• Overtravel: is there enough mechanical movement available to guarantee activation without damage?
• Mounting: does the edge mount cleanly, including corners, joins, and repeatable tolerances?
• Ingress protection (IP rating): is it genuinely suitable for washdown/outdoor exposure, including cable exits?
• Maintenance practicality: can it be replaced without dismantling half the machine?

The key shift is this: specify the sensing edge as part of the motion system, not as an accessory. Get the supplier involved early enough to design around your geometry and environment rather than retrofitting a generic profile.

Safety mats matter

Safety mats are often chosen late because they feel straightforward. But they can become a persistent source of faults if the installation ends up fragmented, over-wired, or poorly matched to contamination and traffic.

A common pattern is a machine being supplied with multiple small mats wired in series. It works on paper, but in practice it creates:

• more joints and more failure points
• more complexity for commissioning and troubleshooting
• awkward coverage that encourages unsafe workarounds

In many cases, a better approach is fewer, better-fitted mats, designed around the operator workflow and the physical zone. The right mat specification should be driven by the reality of the floor: oils, swarf, washdown, heavy footfall, and any fixtures/anchors that require cut-outs or protected edges.

Bumper performance

Where equipment moves with enough mass or speed that contact is likely (e.g. AGVs/AMRs, transfer vehicles, moving gantries, automated handling systems) safety bumpers are often a more realistic solution than slimmer edge profiles.

Here, “basic” is especially dangerous, because the component has two jobs at once:

• absorb impact physically
• detect contact reliably and consistently

A bumper that deforms incorrectly, has the wrong actuation force, or can’t survive repeated knocks may still “work” in tests, but fail in the fleet.

Corner cutting on safety components is rarely saved money. It’s deferred cost. Functional safety can absolutely be a competitive advantage, but only if the last metre of the system (where humans meet moving machinery) is engineered with the same seriousness as the overall design.

This article was first published in Design Products & Applications in June 2026