Technical guide · Cold chain & low-temperature operations

Cold Work Below Zero

Where cold work actually happens, what low temperature breaks inside a PDA or tablet, why the battery decides everything, and how to read a −20 °C claim without being misled.

JNXTECH Engineering · 12 min read · Updated October 2026

Most cold-chain device failures are not failures of the device. They are failures of the specification — a single number on a datasheet that was never qualified against the way the equipment is actually used. A tablet rated “−20 °C” may run at −20 °C for twenty minutes and still fail a four-hour shift at the same temperature, because the rating describes the moment the device is switched on, not the work it has to do.

This guide separates the problem into its real parts: the environment, the components, the operator, and the cell chemistry that ultimately sets the limit.

1. Where cold work actually happens

“Cold” is not one environment. It is at least three different problems that demand different devices, and conflating them is the single most common specification error we see.

1.1 Geography and season

Outdoor cold is driven by latitude and continental climate. In Russia, mean January temperature runs from about −6 °C in Moscow to roughly −16 °C in Novosibirsk and near −38 °C in Yakutsk. Northern Scandinavia, Canada, Kazakhstan and Mongolia sit in the same band. But equipment is not specified against the monthly mean — it is specified against the cold snap, the two weeks a year when the mean is exceeded by fifteen or twenty degrees. A device chosen for the average will fail on the coldest week, which is exactly when a delivery cannot be postponed.

Seasonality matters just as much as latitude. Loading docks, rail yards, ports, utility inspection, agriculture, construction and last-mile delivery all continue through winter. In these settings the device is not stored in the cold — it is carried outdoors, brought into a heated cab, taken out again, and expected to keep working across that cycle several times an hour.

1.2 Purpose-built cold rooms

Indoor cold is a different problem because it is constant, controlled, and often humid. The standard bands are well established across the cold-storage industry:

Note where the real risk sits. It is rarely in the middle of the room. It is at the boundaries — the door, the dock, the ante-room, the transfer between a freezer and a warm humid dispatch area. That is where condensation forms, and condensation is the leading cause of cold-chain device failure.

1.3 Three exposure patterns, not one

Zebra, which has published more on this than most, draws the distinction precisely in its cold-chain application brief. The three categories it defines are worth adopting regardless of whose hardware you buy:

Cold-rated devices are designed for areas where temperature shock and condensation are not an issue. These workers spend their day primarily in a single temperature zone and are not constantly moving between cold and warm environments.

Condensation-resistant devices are designed to withstand moderate temperature shock and condensation, ideal for workers who travel constantly between refrigerators/coolers and temperature-controlled, less humid areas.

Condensation-free devices are designed to withstand the severe condensation that occurs when workers are constantly traveling between the freezer and warm areas with high humidity.

— Zebra Technologies, Optimizing Operational Efficiencies with Mobile Solutions Designed for Cold Chain Environments

This taxonomy matters because it maps to three different engineering answers. A device that lives inside one freezer needs cold capacity, and little else. A device that crosses a freezer door forty times a shift needs sealing, thermal mass control and internal heating — a fundamentally harder problem. Buying for the wrong pattern is how projects end up with equipment that meets its datasheet and still fails the shift.

2. Why the work cannot simply move somewhere warmer

A reasonable question from anyone outside the industry: why not scan the pallet in the warm ante-room? There are four answers, and each of them is a hard constraint.

The chain must not break. Cold chain is regulated as a continuous process. Food safety frameworks (HACCP-based systems, and in the US the Sanitary Transportation and FSMA rules), pharmaceutical GDP requirements, and retail audit schemes all require records generated at the point of handling, with a timestamp and a location. A temperature reading taken after the pallet has left the freezer is not evidence; it is an estimate.

Cold space is expensive. Refrigerated volume costs several times ambient volume to build and to run. Nobody dedicates a heated work station inside a freezer, and nobody builds a warm annex beside every frozen chamber. The work has to happen where the goods are.

The workflow is physical. Receiving, put-away, picking, cycle counting, dispatch and proof of delivery are all actions performed against the goods. Moving the data entry away from the goods adds a transcription step, and transcription is where errors and shrinkage appear.

Some cold work is outdoors by nature. Utility meter reading, railway inspection, port operations, agricultural traceability and winter construction do not have an indoor alternative. When the temperature drops, the work continues and the device has to follow.

3. What the cold breaks, component by component

Low temperature is not a single stress. It attacks different subsystems through different mechanisms, at different thresholds. Below is what actually happens, and why.

Two of these deserve emphasis because they are widely underestimated.

Condensation is a bigger killer than cold. Water does not need to be liquid to cause damage; it needs to be present when the device warms up. A device taken from −20 °C into a +20 °C room at 60 % relative humidity will collect liquid water on and inside every cold surface within minutes, because the dew point of that room air is far above the device's surface temperature. Repeated cycling drives moisture past seals that were never tested for that direction of transport.

Plastic embrittlement turns a survivable drop into a dead unit. A polycarbonate blend that absorbs impact energy at +20 °C can shatter at −25 °C. This is why a drop rating measured at room temperature tells you nothing about a drop in a freezer aisle — a distinction that some manufacturers state openly and others quietly omit.

4. The operator side: gloves, fogging and frozen fingers

Even a device that survives the cold has to be usable by a person wearing PPE in it. This is the part most datasheets ignore entirely.

The practical conclusion is that cold-zone interface design has to be scan-first, touch-second. Every tap you remove from a workflow is worth more in a freezer than any screen specification, because the constraint is not the display — it is the hand inside the glove.

5. What a standard PDA or tablet actually does at −20 °C

Here is the sequence that field engineers see over and over with consumer-grade or “office-rugged” devices.

There is also a failure mode that has nothing to do with the battery: moisture ingress after repeated thermal cycling. A device can pass an IP65 or IP68 immersion test in a lab and still take on water in a freezer over a season, because the test measures static water pressure, not the pump action of an enclosure breathing through its seals every time it crosses a temperature boundary.

6. The battery is the bottleneck

Every other component in the chain — display, radio, scanner, CPU — is powered from the same cell. So the practical low-temperature limit of the whole device is set by the cell, and specifically by two numbers that move in opposite directions as temperature falls: available capacity goes down, internal resistance goes up.

Figures are indicative ranges for standard Li-ion chemistries, not guarantees. The exact curve depends on cell design, electrolyte, discharge rate and state of charge. Always obtain the supplier's measured curve at your intended current.

Two mechanisms explain why the resistance number is the more dangerous of the two.

Voltage sag, not energy depletion, causes the shutdown. The BMS cuts off at a fixed minimum voltage. At high internal resistance, a normal load current produces a large IR drop, so the terminal voltage crosses the cutoff while the cell still holds usable energy. This is why a device can die at 40 % and revive at 40 % — the pack was never empty.

Charging has a hard floor at 0 °C. This is not a cost decision or a vendor limitation; it is chemistry. Below freezing, lithium ions intercalate into graphite more slowly than they arrive, so metallic lithium deposits on the anode surface instead of entering the lattice. That lithium is lost permanently, reduces capacity, and in the worst case forms dendrites that pierce the separator. Every serious manufacturer therefore disables charging below 0 °C and derates it between 0 and 10 °C. A device that appears to charge happily in a freezer is a device whose protection is wrong, not one whose chemistry is better.

7. What determines a battery's cold performance

Cold performance is not one property. It is the outcome of five layers, and only the first two are inside the cell.

Layer 5 is worth dwelling on, because it is free. A worker carrying one spare pack inside a jacket has a warm battery available at every swap — and a warm cell delivers roughly twice the usable energy of a cold one. In many cold-store deployments the cheapest single improvement is a second battery and a pocket, not a new device.

Layer 4 is where the industry splits into two camps, and the split is dictated by one thing: whether external power is available.

“True frost and condensation free operation with touchscreen, connectors, board and battery heaters … Smart temperature sensors that automatically control heater and heating speed … User-replaceable heated battery keeps device stable under all conditions.”

— Zebra, describing its vehicle-mount VC8300 / VC80X computers (−30 °C to +50 °C), in the same document

That is a vehicle-mounted computer, permanently wired to truck power. Heating is affordable there because the energy does not come from the battery. Now look at the same manufacturer's tablets:

The pattern is clear and it is not an accident. Heaters appear on the products with external power. Battery swapping appears on the products that run on their own cell. The tablets reach −20 °C without a heater, and they say so by marking themselves “non-condensing” — which is an honest way of saying they are cold-rated, not condensation-free.

The same logic applies to the current-generation tablet, the ET401. Its published specification lists a −20 °C to +50 °C operating range, a −40 °C to +70 °C storage range, the same range for thermal shock, IP68, a rechargeable lithium-ion polymer battery of 6100 mAh (8-inch), an optional hot-swappable 3400 mAh second battery, and battery-free operation from USB-C or the bottom/rear I/O. It lists no heater. What it does list is the means to keep the pack out of the energy budget: swap it, or remove it from the equation entirely at fixed stations.

8. Low-temperature electrolyte: what actually changes

This is the part that decides whether −20 °C is reachable at all, so it is worth explaining properly.

A lithium-ion electrolyte is a mixture of three things: a lithium salt (commonly LiPF₆) dissolved in an organic solvent blend (typically ethylene carbonate plus one or more linear carbonates), with small amounts of additives. Lithium ions move through this liquid during charge and discharge. When it gets cold, the liquid becomes viscous, ion mobility falls, and the energy required for a solvated ion to shed its solvent shell at the electrode interface — the desolvation barrier — rises. Both effects increase resistance.

Ethylene carbonate is the specific problem. It is essential for forming a stable SEI layer on the graphite anode, but its melting point is around 36 °C, so at room temperature it is already solid on its own and only stays liquid because it is blended. At low temperature the blend approaches its freezing range and conductivity collapses.

Three chemical levers, and one structural one

Note the direction of the trade-off. A low-temperature electrolyte is not a free upgrade. The low-melting solvents and extra additives occupy volume that would otherwise hold active material, so capacity falls for the same cell format. Cost rises, and high-temperature cycle life can degrade.

This is the honest explanation for a detail that confuses buyers: Zebra's freezer battery for its MC9300 handheld is rated 5000 mAh against 7000 mAh for the standard pack, and the company says of it that the lower-capacity pack “is capable of providing more power at typical freezer temperature than the standard battery.” The capacity went down because the chemistry was changed to work in the cold. It is not a cheaper battery; it is a differently formulated one, and the lost capacity is the price.

9. How to read a −20 °C claim

Given all of the above, a single “−20 °C” in a specification is close to meaningless. Here is how to test one.

9.1 Ask for three temperatures, not one

A specification that merges these three into one “industrial temperature range” is not more capable; it is less precisely described.

9.2 Match the test method to the claim

MIL-STD-810H Method 502.7 (Cold) defines three procedures, and the distinction is decisive: Procedure I is storage, Procedure II is operation, and Procedure III is manipulation while wearing thick gloves. Only Procedure II supports a claim that a device operates at a given temperature. Worth noting too: MIL-STD-810 is a test method standard. It issues no certificate. The correct phrasing is “tested per MIL-STD-810H Method 502.7, Procedure II” — not “MIL-STD certified”.

9.3 Five statements to treat as red flags

1. A single −20 °C with no power source stated.
2. Storage temperature presented as operating temperature.
3. “Battery heater” claimed on a battery-powered handheld with no heater in the BOM.
4. Low-temperature improvement claimed with unchanged capacity — physically inconsistent.
5. “Works at −20 °C” used to imply a full shift at −20 °C, with no runtime figure published. Reputable vendors publish operating range and stay silent on cold runtime, because the honest answer depends on the workload.

10. A field checklist before you commit

Before accepting any low-temperature device into a cold operation, run these eight checks. They take one day and they will save a deployment.

1. Soak test, not a demo. Leave the device in the actual target temperature for at least four hours, powered on, before touching it. A cold start after ten minutes proves nothing.
2. Load test at temperature. Scan continuously, transmit over Wi-Fi and take photos while cold. Peak current is what triggers undervoltage shutdown, not idle.
3. Weigh the unit as configured. Battery, strap, case, scanner grip. Quoted weight is usually the bare tablet, and the configured weight decides whether a worker will carry it for eight hours.
4. Test with the real gloves. Not a thin liner — the actual PPE grade used on the floor. Count mis-taps per hundred operations.
5. Cross the boundary ten times. Freezer to warm humid room and back, with the device powered. Then inspect the scan window, camera lens and display for condensation and frost.
6. Try charging in the cold zone. If the workflow depends on it, confirm the behaviour rather than assuming it. Most devices will refuse, correctly.
7. Time a battery swap with gloves on. Multiply by the number of swaps per shift. If it exceeds a minute, the swap will not happen in practice.
8. Ask for the runtime curve, not the runtime number. At what current, at what temperature, at what state of health. A single “10 hours” figure is marketing; a curve is engineering.

What this means for specification writing

The recurring theme in every section above is that cold capability is a chain, and a chain is only as strong as its most optimistic claim. The practical discipline is simple, and it costs nothing:

State the three temperatures separately. State the power source. State the test procedure. State the cost of the low-temperature option, including the capacity you give up.

A specification written that way is shorter than a marketing datasheet and far more useful, because it lets a buyer see exactly where the boundary is and plan the workflow around it. It also survives a cold-room trial, which is the only test that matters.

Product specifications referenced in this article are quoted from manufacturer published documentation as of October 2026 and are used for technical comparison only. All trademarks belong to their respective owners. Figures for capacity and internal resistance versus temperature are indicative industry ranges, not guarantees for any specific cell; obtain measured data for the exact part you intend to deploy.