Qualifying a stability or climatic chamber

A stability chamber must hold its set condition everywhere product is stored, for months. Qualification therefore emphasises spatial uniformity and temporal stability of both temperature and humidity, and the behaviour of the chamber under disturbance.

Study design

• Calibrated sensors at corners, centre and shelf positions — humidity sensors alongside temperature at a subset.
• Each ICH set point in use, run until stable then for a defined period.
• Empty and loaded configurations.
• Door-opening test with recovery time measured.
• Alarm tests and power-interruption behaviour.

EURAMET cg-20 describes how to characterise chamber uniformity and stability metrologically — the same methods serve qualification when paired with acceptance criteria.

Acceptance criteria, and where they come from

Criteria are not invented for the protocol. They come from the label conditions of the products the chamber will hold, which in turn come from ICH Q1A(R2). A chamber running the long-term condition of 25 °C ± 2 °C / 60 % RH ± 5 % RH has to keep every storage position inside those bands, and the qualification has to demonstrate it with sensors whose own uncertainty is small enough to leave room for the demonstration.

Empty and loaded, and why both

An empty chamber is the easy case: unobstructed air, even circulation, and a uniformity figure that flatters the equipment. Loading changes it. Packed shelves block return air paths, cartons hold thermal mass that slows recovery, and the product itself may be at a different temperature when it goes in. The empty study characterises the chamber; the loaded study is the one that supports a storage decision. Run the empty study first so that if it fails you have not wasted a load.

Loading should reflect the worst realistic case rather than a tidy one. If shelves will be filled to the front edge in service, fill them to the front edge in the study. A qualification performed on a half-empty chamber places no limit on how full it may later be packed.

The door-opening test in practice

Open the door for the longest period routine work requires, with the chamber loaded, and record the recovery. Two numbers come out of it: how far the nearest storage position moved, and how long every position took to return inside the band. If a 30-second opening pushes a front-shelf position to 27.5 °C for four minutes in a 25 °C ± 2 °C chamber, that is a real excursion the monitoring system will alarm on daily, and the answer is either a procedural limit on door time or a different storage layout — not a wider acceptance criterion.

What usually goes wrong

• Humidity sensors placed only at the centre. RH varies more across a chamber than temperature does, and the centre is the best-behaved point in it.
• The study run at one set point when the chamber is used at several. Uniformity at 25 °C says little about uniformity at 40 °C, where the heaters work harder.
• No power-interruption test, so nobody knows whether the chamber restarts, holds its set point, or comes back with an alarm nobody acknowledged.
• Sensors calibrated before the study but not checked after it. A sensor that drifted during the run invalidates the data, and only a post-check will show it.
• Control sensor used as the qualification sensor. The chamber will always look like it is at set point if you ask the instrument that is holding it there.

Requalification is triggered by change, not only by the calendar: a new shelving layout, a refrigeration repair, a move, or a new product with tighter label conditions all reopen the question the original study answered.