Short answer: a wearable fan carries six claims that decide whether it is a good product, and every one of them can be measured before volume is committed: battery capacity, air output, runtime, sustained temperature rise, battery protection, and the motor plus its thermal path. A label states all six as nouns. A verification record states them as measurements with conditions. The difference between the two documents is the entire difference between a product that survives its first summer and one that does not.
Why this category gets specified badly
Wearable fans are small, cheap to tool and easy to source — which means they are frequently bought on a specification sheet rather than on a test report. The sheet is usually accurate in the narrow sense: it states things that are true at one moment, at one setting, on one sample. What it does not state is conditions.
The consequence is predictable. Returns in this category cluster around claims that were never false, only unqualified: a capacity figure measured at a different discharge rate, a runtime measured at a different setting, a noise number measured at a distance nobody wrote down.
The six claims, and what each one needs
| Claim | Why the label is not evidence | What we require instead |
|---|---|---|
| Battery capacity | A capacity number is a cell datasheet figure taken at a favourable discharge rate; it is not what the pack delivers in the product | Measured capacity on samples at a defined discharge rate and cut-off voltage, plus a cycle-life test on a smaller sample |
| Air output | A single maximum air speed describes one point at one setting, and says nothing about the air actually moved | Measured air speed at every setting, with the measurement distance stated, converted to air volume using the measured outlet area |
| Runtime | Quoted at whichever setting produces the largest number, and rarely stated at all for the settings people use | Runtime measured at every setting, on the same unit, with the ambient temperature recorded |
| Sustained temperature rise | Spot readings after a few minutes miss the part of the curve where heat accumulates | A continuous run held to thermal equilibrium, with temperature logged at the motor, the battery and the board, plus the housing surface |
| Battery protection | "Protected" names a category, not a topology, and protection only matters at the moment it is needed | The protection elements and their trip points named in writing, then verified on samples: over-charge, over-discharge, over-current and short circuit |
| Motor and thermal path | The motor is described by a speed figure; how heat escapes it is not described at all | The internal thermal path identified — where heat leaves the motor, what it passes through, and what temperature the surrounding plastics and wiring are rated for |
The three mechanical failures that decide the return rate
Electronics decide whether a wearable fan works. Mechanics decide whether it survives. From the complaint patterns across this category, three mechanical items account for a disproportionate share of returns, and none of them is expensive to verify:
- The belt clip or attachment point. Two distinct failures: the clip itself cracking, and the clip holding but losing clamping force over time so the unit works loose during normal movement. Verify by cycling the clip open and closed to a stated count, then measuring the retained clamping force — the second test is the one that catches the slow failure.
- Housing sealing and structural rigidity. Air that escapes through the housing is airflow the wearer paid for and did not get, and thin sections without reinforcing ribs crack under a drop. Verify by running the unit and checking for leakage paths, then dropping it from a stated height onto a defined surface and re-checking function, retention and electrical safety.
- The charging path. Intermittent or absent charging, slow charging that only reaches a full pack on some units, and connector wear from repeated insertion. Verify by timing a full charge from empty at a stated input on more than one sample, and by cycling the connector to a stated count.
Note what these three have in common: each is a mechanism that only appears after repetition, which is precisely why a spot check passes them all.
Why a continuous run, not a spot reading
Almost every figure in this category is easier to pass in the first five minutes than at minute forty. The difference is not marginal:
- Air output decays. Filters load, and an impeller that is warm and has run for an hour is not the impeller that was measured when new.
- Temperature climbs. Heat accumulates in the motor, the battery and the board until the design reaches equilibrium — and equilibrium is the state the product spends most of its life in.
- Protection can trip. A unit that shuts down after thirty minutes of normal use has a design fault that no five-minute test will reveal.
So the requirement is not a higher number. It is a longer measurement: run the unit at the setting a wearer would use, hold it until the readings stabilise, and record what they stabilise at.
What a supplier must be able to answer
- At what discharge rate was the capacity figure measured, and was it measured on the pack in the product or on a bare cell?
- At which setting was the runtime measured, and at what ambient temperature?
- At what distance was the noise figure taken, and at which setting?
- Which protection elements are on the board, and at what trip points?
- Where does heat leave the motor, and what temperature are the surrounding plastics and wiring rated for?
- How many cycles was the clip tested to, and was retained clamping force measured at the end?
A supplier who can answer all six is running a verification process. A supplier who can answer two is quoting a datasheet, and the difference will show up in the return rate rather than in the sample.
Frequently asked questions
Why is a battery capacity claim so often wrong?
Usually not wrong but unqualified. A cell datasheet rates capacity at a specific discharge rate and cut-off, and a product drawing current at a different rate will deliver less. Measuring on the assembled product at the real load is the only way to get the number a customer experiences.
How long should a sustained run be?
Long enough for the readings to stop moving — that is the definition of equilibrium, and it is what the product will do for most of its working life. In practice this is tens of minutes, not minutes, and the same run should capture air output, noise and temperature together.
Do I need to test every sample?
No. Verify the stated figures on a small sample at first article, then monitor consistency between consecutive units thereafter. A design that only exists on the golden sample is not a design.
What is the cheapest useful reliability test for this category?
The clip cycle test with a clamped-force measurement at the end, and a full charge timed from empty on more than one sample. Both are bench tests with no specialist equipment, and both catch failures that a visual inspection and a five-minute run will pass.
What should a verification report contain?
Four things: the test conditions, the instrument and measurement points, the pass threshold agreed in advance, and the result per sample. A report without conditions is a marketing document, whichever column of the table above it resembles.
If you are sourcing a wearable fan, these six claims are the core of the specification we write before sampling begins — the wider requirement set is in our OEM/ODM programme. For the measurement methods behind two of the six, see air speed is not air volume and why the top setting is the least useful number on the sheet. Request a quotation with your target market, volume and verification requirements.