A temperature calibration bath is a stirred liquid reservoir used to maintain a uniform isothermal zone for comparison calibration of PRTs, thermocouples, thermistors, and process probes, with silicone-oil-based units typically spanning -45 °C to +300 °C depending on design [S1].
Selection is governed by four published specifications: temperature range, stability, uniformity, and tank working volume, and any one of them silently failing collapses calibration uncertainty, so this buying guide walks through the checks a process or metrology engineer should run before signing a PO [S3].
Temperature range and the fluid-switching problem
Manufacturers advertise a single temperature range per bath, but no single heat transfer fluid covers the full envelope; a -45 °C to +150 °C bath, for example, takes ethanol or methanol below 0 °C and silicone oil above, which forces either a fluid change or two parallel baths in the lab [S3]. Below 0 °C, the candidate fluid set is ethanol, methanol, ethylene glycol, halocarbon, HFE, Dynalene, and certain low-viscosity silicone oils, while above 0 °C the practical choices narrow to water, mineral oil, and several silicone oil grades, with molten salt preferred above +300 °C [S3]. WIKA's published application range for silicone-oil calibration baths sits between -45 °C and +300 °C, demonstrating that the same bath hardware can be re-tasked across a wide envelope as long as the fluid and heater/cooler capacity are re-matched [S1].
Stability: the spec most often mis-quoted
Stability is the bath's short-term temperature variation over time at a given setpoint, and it is the single number most often published without context, so engineers should ask for the fluid, the setpoint temperature, and the measurement interval behind any quoted figure [S4]. Industry guidance calls for stability figures that hold across the full advertised range, not just at 20 °C in water, because fluid viscosity and heater duty cycle change with setpoint [S3]. liquid bath calibrators with automated ramp control typically hold stability inside the ±0.01 °C class over a 30 °C to 200 °C sweep, a level sufficient for secondary SPRT and high-grade PRT comparison work [S5]. If a vendor publishes only a single ambient-temperature stability value with no fluid or setpoint declared, treat the figure as marketing rather than a specification [S3].
Uniformity and the working volume trap
Uniformity is the spatial temperature spread inside the working volume, and it is where the most expensive misreading happens: some baths are specified at the tank geometric centre, while working labs put probes near the walls, the lid, and the bottom, so the published number can overstate real probe-to-probe agreement [S3]. The heat-port design, where the cooling coil and heater are clamped to the outside of a stainless tank, is one approach that delivers uniformity across the entire working area rather than just at a single sweet spot [S8]. temperature monitoring of multiple probe positions during a uniformity test at the lowest, mid, and highest setpoint is the practical way to confirm the published figure; if only one point is quoted, the data is incomplete [S3]. For multi-probe batch calibrations, models that accept up to six thermometers in a single bath load cut throughput bottlenecks and let one standard track several units under test at once [S7].
Tank size, immersion depth, and sensor fit
Tank size governs both how many probes can run in one batch and how deeply each probe can be immersed, and deep immersion is the whole point of a liquid bath versus a dry block, because it eliminates stem-conduction error on long probes [S2]. Typical laboratory calibration baths offer working volumes in the 185 mm × 140 mm × 300 mm class for high-volume comparison work, with a 30 °C to 200 °C operating window and ±0.01 °C comparison accuracy [S5]. Tank access is sized by the probe length and any well or sheath hardware, so the engineer should lay out the largest probe plus any required handle hardware before selecting a tank footprint [S3]. Standard accessory lids with replaceable sleeves for 3 mm, 4 mm, and 6 mm probe diameters are a common feature, allowing the same bath to handle mixed-diameter loads without improvised foam plugs [S7].
Heat transfer fluid selection rules of thumb
Fluid choice is governed by thermal conductivity, low viscosity at the operating setpoint, chemical inertness, low vapour pressure, and thermal stability across the working range, and silicone oils dominate the mid-range because they score well on all five [S1]. The single most useful rule is to keep fluid viscosity at 50 cSt or less at the control temperature, since thicker fluids strain the stirrer, create gradients, and inflate calibration uncertainty [S3]. Above +300 °C the fluid question becomes a hardware question: salt baths take over because organic fluids decompose or smoke at those temperatures [S3]. For the sub-zero end, ethanol and methanol are the common picks because they stay fluid below 0 °C where water and most oils turn to slush or solid, but both are flammable, so fume hoods and grounding for static are part of the spec, not optional extras [S3].
Refrigerated vs heat-only, and throughput math
Heat-only baths cover roughly ambient+10 °C to +300 °C and need a separate chiller or cooling coil to reach sub-ambient, while refrigerated/heated models integrate a compressor and cover -45 °C to +200 °C in one box, eliminating fluid-transfer steps and warm-up waits [S7]. Throughput scales with working volume, stirrer horsepower, and heater/refrigeration capacity: a large-volume 30 °C to 200 °C bath at ±0.01 °C can calibrate a full set of sensors in a single load rather than queuing them [S5]. Automated ramp control with software integration reduces operator variability between setpoints and is worth the spend on regulated lines where audit trails matter [S2]. For a broader view of how calibration equipment sits inside a plant-wide instrument program, the top measuring instruments companies 2026 spec-driven market map gives useful context on vendor capability spread, while the loop calibrator price and cost guide 2026 covers the electrical side of the same calibration workflow.
Acceptance criteria and when to walk away
Acceptance criteria for a new bath should be: (1) stability ±0.01 °C or better verified at three setpoints across the range with the named fluid, (2) uniformity ±0.02 °C or better mapped over the full working volume, and (3) documented fluid compatibility and viscosity at every setpoint used [S3][S5]. If a vendor cannot produce all three, the bath is not qualified for ISO/IEC 17025-style work and the engineer should walk away or budget for an in-house qualification campaign [S3]. Replace, do not repair, any bath showing tank corrosion, heater insulation breakdown, or stirrer bearing noise, because each of those failure modes degrades uniformity faster than the published stability number suggests [S8]. Track the next two signals going forward: vendor moves to publish full-range rather than single-point stability data, and the uptake of automated ramp+datalogger integration as standard rather than optional add-ons [S2][S3].
The underlying component specifications are covered under temperature calibration bath.