Industry 5.0

For flammable gases the safety chain is dimensioned around the lower explosive limit. Alarm thresholds there sit in the per-cent range — for hydrogen that means several thousand ppm. A sensor resolving ppb is not more sensitive than needed for that task; it is orders of magnitude too sensitive. So the question is a different one: what do you do with the range below?

What the datasheet warrants — and what it does not

We put this first because it determines how much weight every further statement can carry.

Among the gases relevant here, the datasheet warrants a value for CO only. For the others, measured values and test conditions have to be established per project — that is part of the feasibility assessment, not a detail afterwards.

The arithmetic of the explosive limit

Hydrogen is ignitable in air from 4 volume per cent. Gas detection systems typically alarm at 10 to 20 per cent of that value, so that time remains to intervene.

8,000 ppm is 0.8 volume per cent. A sensor that distinguishes substances in the ppb range works four orders of magnitude below that. For the safety function this brings nothing — there it does not matter how early you see something, but that the report is dependable, tested and approved.

The benefit lies elsewhere: in the time before. A leak does not begin at 0.8 volume per cent. It begins small and grows. Watching the plant far below the alarm threshold shows a leak as a trend before it becomes a safety question — and lets it be fixed in planned maintenance rather than in a shutdown.

The real problem is cross-sensitivity

Common sensors for flammable gases — catalytic bead devices as well as simple MOx elements — respond to anything combustible. They measure that something flammable is present, not what. In a plant where hydrogen, methane, carbon monoxide and solvent vapours can occur at the same time, that produces exactly the two errors nobody wants:

• False alarm. A cleaning routine, a solvent, a vehicle in the hall — and the system reports a leak that does not exist. Repeat that and the report gets ignored.
• Mistaken identity. A methane release is reported as hydrogen or the other way round. Yet the countermeasure is not the same: hydrogen rises immediately and collects under the ceiling; heavier gases behave differently.

This is where NZGS 2 comes in. Four sensing elements on a die of 1.15 × 1.15 mm, two different layers, different operating temperatures — one gas event produces four signal traces, and the substance is determined from the ratio between them rather than from the height of a single deflection. The method is set out on the technology page.

Carbon monoxide: the case with a documented figure

Among the industrially relevant gases, CO is the only one with a warranted limit of detection: 900 ppb, roughly 0.9 ppm. It is detected via the tungsten trioxide layer.

That is a usable order of magnitude for the question of whether combustion is running cleanly. A boiler, burner or furnace receiving too little air shows it in the CO content of the flue gas long before it becomes a safety matter. As a channel for condition monitoring and predictive maintenance, this is the most dependable part of the application.

Explosion protection is a design question

One point belongs settled early: NZGS 2 works with two heated elements at heater voltages of 1.8 to 2.2 V and around 70 mW. A heated component in a potentially explosive atmosphere is not an accessory topic.

Not a safety function

This is the second boundary, and it is just as important. Fixed gas detection systems for flammable gases fall under their own standards, among them the EN 60079-29 series. They require self-monitoring, defined response times, regular functional testing with test gas, and approval.

NZGS 2 does not meet those requirements and is not intended to. It delivers a measurement. Anyone who needs a safety function uses an approved system — and can run the sensor alongside it to watch the range below. Under our terms, use in safety-critical applications requires a separate written agreement in any case.

Where the sensor actually fits

• Early detection of creeping leaks at fittings, valves and seals — far below the alarm threshold, as a trend across days and weeks.
• Condition monitoring of combustion processes via CO content, with a documented limit of detection.
• Telling which gas is escaping where several flammable gases occur in the same plant.
• Distributed measurement at many points, where installation space and power consumption decide feasibility: QFN-8 at 2.8 × 2.8 × 1.6 mm, around 70 mW.

What the sensor does not do

• It does not replace an approved gas detection system or a safety function under EN 60079-29.
• It brings no warranted limit of detection for hydrogen. That figure only emerges from a measurement campaign with your mixture.
• It is not Ex certified. The type of protection is established at device level, not at the component.
• It ages and it is vulnerable. Baseline drift is compensable if temperature and humidity are measured alongside. Silicone compounds poison the tin dioxide layer irreversibly — in industrial environments a realistic risk that must be considered in housing, seals and adhesives.

How a project begins

Because no warranted values exist for most gases of this application, a measurement stands necessarily at the beginning. The evaluation kit with read-out electronics and software records the four signal traces, extracts the features and shows whether your target gases separate from the interferents of your plant in feature space.

What is needed is reference air, the target gases at known concentrations, and the interferents that realistically occur at the installation point — in industrial environments typically solvents, exhaust and cleaning agents. Plus temperature and humidity across the expected range. All specifications of the component are in the datasheet extract.

Explosive limits after generally recognised safety characteristics; sensor specifications after NANOZ NZGS 2, datasheet rev. 3.3.0. References to standards serve orientation and do not replace a safety assessment of the specific device.