In short: An ATEX mobile phone solves the ignition problem and leaves the coverage problem untouched. Zone 1 handsets are certified, expensive and increasingly capable, but on most UK hazardous-area sites they lose signal in exactly the places they were bought for: inside the bund, between the columns, under the pipe rack. The fix is a network planned around the zoned area, with radios outside the zone doing most of the work.
Key Takeaways
- Certification is not coverage — A Zone 1 handset from ecom or i.safe MOBILE costs well over £1,000 and is useless in a tank farm with no signal; the network is the half of the purchase most sites skip.
- Ex-rated WiFi does not scale across a process unit — Every access point inside the zone needs an Ex enclosure, a certified installer and a permit, so sites cover the control room and give up on the plant.
- Put the radios in the safe area — Private 5G cells sited outside the zone reach across it at licensed power, cutting the count of certified radios in the hazardous area to a handful or none.
In a nutshell
Why does an ATEX mobile phone need more than certification?
Because a certified handset with no signal is a certified camera and torch, and most UK hazardous-area sites have bought the handset without buying the coverage.
The logic behind the purchase is sound. Field operators, maintenance technicians and contractors on a refinery, fuel terminal, gas plant or chemical works need to carry a device into the zoned area, and a consumer phone is not permitted there. So the site buys ATEX-certified smartphones, and the plan is that operator rounds move from clipboard to app, permits get signed at the equipment, photos travel with the work order, gas detectors pair to the handset, and lone workers check in from wherever they are. Every one of those uses depends on a live connection at the point of work, and that is where the plan usually meets the site.
Walk a UK tank farm with a certified handset and the signal map is depressing. Coverage on the office side is fine, because that is where the public network was planned for. Coverage on the plant side falls away between the storage tanks, drops to nothing inside a bund, disappears under the pipe rack and comes and goes between the columns of a process unit. The handset is certified for all of those places. It has nothing to talk to. We see the outcome on site after site: a fleet of expensive devices used mainly for photographs, with the data entered later at a desk, and the digital operator round quietly reverting to paper.
What do the zones and Ex markings mean for a handset?
They mean the device cannot be a source of ignition in a place where flammable gas or dust may be present, and the classification of that place decides how strict the design has to be.
Under the Dangerous Substances and Explosive Atmospheres Regulations 2002 (DSEAR), a site operator must classify areas where an explosive atmosphere may occur. For gases and vapours, Zone 0 is where the atmosphere is present continuously or for long periods, Zone 1 where it is likely in normal operation, and Zone 2 where it is not likely and would only persist briefly. Dusts use Zones 20, 21 and 22 on the same pattern. The zone map, usually drawn on the site plan in coloured shading, is the document that decides which devices can go where.
Equipment for those zones is certified under the ATEX equipment directive 2014/34/EU in the EU, under the parallel UK regulations with UKCA marking since January 2021, and under the IECEx scheme internationally. Most Zone 1 handsets are certified as intrinsically safe, marked Ex ib, which means the electrical energy in the device is limited below what could ignite the gas group it is rated for even under a fault. Zone 0 devices carry the stricter Ex ia marking. The engineering that achieves this in a smartphone, with a battery, a camera flash and multiple radios, is why the current Zone 1 devices from ecom (part of Pepperl+Fuchs), i.safe MOBILE and Bartec cost well over £1,000 each and why they lag consumer handsets by a generation or two.
The marking on the handset says nothing about the radio network it uses. The certification covers the device as a potential ignition source. Whether there is a 4G or 5G signal for it to use is somebody else's problem, and on a hazardous-area site that somebody is usually nobody.
Why does coverage collapse inside a process unit or tank farm?
Because the plant is built of steel, spread over a large area, and located where the public mobile operators have the least commercial reason to build.
Firstly, the geography. UK oil, gas and chemical sites sit where the pipelines and jetties are, which is generally the coast or an estuary: Grain and South Hook for LNG, Bacton and St Fergus for gas, Grangemouth, Stanlow and Fawley for refining, Teesside and the Humber for chemicals. Public macro cells serve the nearby villages and the A road. The process area is inside a security fence, on the wrong side of the site for the mast, and often below the level of surrounding land or behind a bund wall. Signal that is adequate at the gatehouse is marginal by the time it reaches the far end of the site.
Secondly, the plant itself. A floating-roof storage tank is a steel cylinder 40 or 50 metres across with an earth bund around it; stand inside the bund and you are in a radio shadow with a wall of steel on one side and earth on the other. A distillation column, a compressor house, a pipe rack carrying dozens of steel lines at head height, all of them attenuate and reflect signal. The signal is patchy rather than weak: full bars on one side of a column, nothing on the other. Handsets react to that patchwork badly, hunting between weak cells, draining battery and dropping any application that expects a stable connection.
Thirdly, the site cannot fix this the way an office would. The obvious response to poor indoor coverage elsewhere is to add WiFi or a small-cell repeater. In a Zone 1 area every piece of powered equipment is itself an ignition risk and needs its own certification.
Why is Ex-rated WiFi so expensive to extend?
Because every access point inside the zone has to be treated as hazardous-area electrical equipment, and the cost is in the enclosure, the installation and the paperwork rather than the radio.
An Ex-certified WiFi access point is a standard radio built into a certified enclosure, typically flameproof (Ex d) or increased safety (Ex e), with certified cable glands and barriers on every connection. Suppliers such as Extronics and Pepperl+Fuchs make good ones. The problem is arithmetic. WiFi has short range, poor penetration and a strong dislike of steel, so covering a process unit needs many access points, and each one in the zone multiplies the cost several times over its office equivalent before anyone has run a cable. Installation is hot-work-controlled, needs certified installers and an inspection regime under BS EN 60079-17, and every subsequent change to the network is a change to hazardous-area equipment.
The result is predictable. Sites cover the control building, the workshop and the canteen with ordinary WiFi, put one or two Ex access points at the edge of the unit they most care about, and stop. The zoned area, where the handsets were bought to be used, stays uncovered. We think this is why so many digital operator-round and digital permit-to-work projects on UK sites stall after the pilot: the software works, the handsets are certified, and the network covers a tenth of the site.
How does private 5G cover a hazardous area with fewer radios in the zone?
By putting the transmitters in the safe area and letting licensed cellular power do the reaching, so that the zone is covered from outside.
A private 5G cell in the UK operates on Ofcom's shared access licence in the 3.8 to 4.2 GHz band, at power levels an order of magnitude above WiFi, from radios mounted on a mast, a building or a pipe bridge in a non-hazardous area. A small number of well-sited cells, planned against the zone map and the plant model, can cover a tank farm and process area that would need dozens of WiFi access points. Where a particular corner is genuinely shadowed, a single Ex-enclosed remote radio can fill it, but the count of certified radios in the zone drops from tens to a few, and often to none.
This matters for reasons beyond cost. Fewer devices in the zone means fewer items on the Ex inspection register, fewer permits for every network change, and a network the site can extend when the plant changes, which on a refinery it does at every turnaround. Cellular also handles the patchwork better than WiFi: a handset holds one connection to the cell rather than hopping between access points, and 5G's uplink is sized for the video, photographs and sensor traffic that hazardous-area applications generate.
The handset side has caught up. The newest Zone 1 devices ship with 4G and, in the latest models, 5G modems, and they take an ordinary SIM or eSIM, so they join a private network the same way a consumer phone joins a public one. The point to check, before buying a fleet, is that the handset supports band n77, which is where UK private networks operate. Some certified devices are built for North American or Asian band plans and will never see a UK shared-access cell.
We should be honest about the trade-offs. A private network is a capital project with radio planning, a licence application and a core to run, and for a small depot with one zoned area a handful of Ex access points may still be the sensible answer. Public 4G improved by an operator-funded small cell is sometimes on offer near large sites, though the operator's commercial interest in a fenced process area is limited. Our argument is narrower than "cellular beats WiFi": where the zoned area is large, steel-dense and central to the site's operation, covering it from the safe area is the only approach that scales.
What should a site check before buying handsets?
Whether there is a signal where the handsets will be used, which is a question most procurement exercises never ask.
A practical sequence looks like this. Take the DSEAR zone map and mark on it every task the handsets are meant to support: operator rounds, permit sign-off, gas-detector pairing, lone-worker check-in, contractor access. Then walk those routes with a survey device and record what the public networks and any existing WiFi deliver at each point. On most sites we have surveyed, the overlap between "where the work is" and "where the signal is" is uncomfortably small, and the survey settles the argument about whether the network is a nice-to-have.
Then check the handset specification for band support, SIM or eSIM flexibility, and the certification the zone map requires: a Zone 2 device is cheaper and adequate for a good deal of a typical site, and it is worth buying to the zone rather than to the worst case everywhere. Finally, look at the existing digital projects that have stalled. The digital permit-to-work system that was piloted and shelved, the operator-round app that reverted to paper, the lone-worker alarms that only work near the gatehouse. Each of them is a business case for the network waiting to be re-opened, and each becomes cheaper to justify once the same coverage serves all of them.
The handset fleet most UK hazardous-area sites already own is the expensive part of the purchase. The network that would make it useful is, in our experience, the cheaper part, and the part still missing.