A gauge that is not being used is not necessarily safe. Between tests, a lamp cap gauge sits in a drawer, a toolbox or a shelf, and that is where a surprising amount of its accuracy is lost. Five failure modes account for most of it.
A threaded gauge has crests that are sharp by design. Dropped into a drawer with screwdrivers, the crests pick up dents that are invisible at arm's length but measurable at the pitch diameter. A dent of 0.02 mm on a thread crest changes the effective size at that point.
The fix is a fitted insert with one pocket per gauge. Cost: less than one replacement gauge.
Gauge steel is hardened and often not stainless. In a workshop at 70 % relative humidity, a gauge left unprotected develops a surface film within weeks. The film is 1 to 3 µm thick and changes the effective diameter long before it is visible.
| Storage fault | Typical effect | Fix |
|---|---|---|
| loose in a drawer | crest dents, 0.02 mm | fitted insert |
| open shelf, humid air | surface film, 1 to 3 µm | VCI strip in a closed case |
| mixed sizes in one pocket | thread wear from rubbing | one gauge per pocket |
| near a heat source | slow dimensional change | store at room temperature |
| gauge left on the bench | dropped, dented, dirty | case closed between uses |
Two threaded gauges in the same pocket rub against each other every time the case is moved. Neither is being used, but both are wearing. Over a year, the crests that contact each other show a polished band and the pitch diameter grows by a few micrometres.
The fix is obvious and often ignored: one pocket per gauge, and the case does not get shaken.
A gauge stored on a shelf above a heater, or in a cabinet next to an oven, sits at 40 °C or more for months. Steel does not permanently change at that temperature, but a gauge that is 20 °C when calibrated and 40 °C when used measures differently — about 2 µm on a 50 mm gauge. That is a thermal effect, not storage damage, but it produces the same symptom.
The fix is to let a gauge come to room temperature before use, and to store it away from heat sources. A gauge taken from a cold car in winter needs an hour on the bench before it means anything.
The fifth failure is not dimensional. A gauge with no marking, or with a marking that has worn off, cannot be traced to its certificate. It becomes an anonymous piece of steel that may be the right size and may not be.
The fix is engraving on a non-working face (see the separate note on gauge marking) plus a case label that matches the certificate number.
A closed case, a fitted insert, one gauge per pocket, a vapour-phase inhibitor strip, room temperature, and a label on the case lid listing every gauge inside with its certificate number. Total cost is a few tens of yuan per gauge and it prevents the failure mode that calibration cannot detect.
The larger point: storage is part of the measurement system. A gauge that is calibrated annually and stored badly spends most of the year drifting away from the certificate, and nobody notices until a batch is rejected.