Characterising a dry-block calibrator

A dry-block calibrator is a heated (or cooled) metal block with wells for the reference and the units under test. Its display is not the calibration reference; an external reference thermometer in a well is. The block's job is to hold a stable, uniform temperature — and its performance must be measured, not assumed.

Tests from EURAMET cg-13

• Axial homogeneity: reference at full immersion vs withdrawn by 20, 40, 60 mm.
• Radial homogeneity: two references swapped between wells.
• Stability: readings over a period at least as long as a calibration.
• Loading effect: gradient with all wells occupied vs one.
• Heat conduction (stem effect): with a probe of different diameter or a thermally conductive probe.
• Hysteresis of the block's control after approaching from above and below.

Typical results for a good block at 100 °C: axial gradient 0.05–0.2 °C over 40 mm, radial 0.02–0.1 °C. These numbers dominate the budget for loggers with fine resolution.

The five tests, and how to run each

EURAMET cg-13 sets out the characterisation. Each test isolates one contribution, and each produces a number that goes into the budget rather than a pass or a fail.

The swap technique in the radial test is the one worth learning. Whatever difference remains after averaging the two orientations belongs to the wells, not to the thermometers, so the measurement does not need two identically calibrated references.

The display is not the reference

A dry block shows a number, and the number is the temperature of the control sensor buried in the block — not the temperature in the well where your unit under test sits, and not the temperature at the depth its element occupies. The last row of the table is often the largest single figure in the whole characterisation. Every calibration performed with a dry block uses an external reference thermometer in a well; the display is for setting up.

Why these numbers dominate the budget

• For a 0.1 °C logger, resolution contributes 0.029 °C and the block's axial term may contribute 0.06 °C or more. The block is then the largest item, and the reference thermometer is nearly irrelevant.
• Match the well to the probe. The air gap in an oversized well is the biggest single contributor to both the radial and axial terms, and an insert costs very little.
• Insulate the block mouth and the exposed stem. This attacks the axial gradient and stem conduction together.
• Re-characterise after any insert change, and annually. Inserts wear, and a worn one reopens the gap.
• Record results per block, per temperature range, per loading. A single figure carried forward from when the block was new is not a characterisation.

If the characterisation shows the block cannot support the uncertainty you need, no amount of care with the reference will close the gap. That is the point at which a stirred liquid bath stops being an extravagance — it reduces all three terms by roughly an order of magnitude.