Selecting the correct thermocouple is the single most decisive factor in obtaining a valid temperature uniformity survey (TUS) for a horizontal vacuum furnace, because sensor accuracy, drift behavior, and vacuum compatibility directly determine whether the measured uniformity band is trustworthy. This article explains how to match thermocouple type, accuracy class, sheath construction, and calibration requirements to furnace class and process temperature in accordance with AMS 2750, IEC 60584, ASTM E230, and GB/T 9452. It provides practical guidance for accredited third-party laboratories, heat-treat suppliers, and quality engineers performing vacuum furnace qualification and periodic re-surveys.
Technical Overview
A horizontal vacuum furnace is a batch-type thermal processing system in which the workload is loaded horizontally into a sealed, evacuated chamber containing a graphite or all-metal (molybdenum) hot zone. These furnaces are widely used for vacuum brazing, hardening, tempering, annealing, sintering, and solution treatment of aerospace, medical, and precision components. Because many of these processes are governed by strict metallurgical tolerances, the furnace must be periodically surveyed to demonstrate that every point within the qualified working zone holds temperature within a specified band, typically ±3°C to ±10°C depending on furnace class.
In a temperature uniformity survey (TUS), a defined grid of calibrated test thermocouples is placed inside the working zone, the furnace is run to the survey temperature, and readings are recorded after stabilization. The measured spread between the hottest and coldest sensors, corrected for sensor error, defines the uniformity. Since the thermocouple is the only element of the measurement chain physically inside the hot zone, its selection governs the credibility of the entire survey. A sensor with the wrong calibration class, excessive drift, or non-vacuum-rated insulation can turn a genuinely uniform furnace into a failed survey, or mask a real non-uniformity.
Governing Standards
- AMS 2750 (pyrometry standard for thermal processing in aerospace): defines furnace classes, instrumentation types, TUS sensor calibration, usage limits, and survey frequency.
- IEC 60584 / ASTM E230: classify thermocouple tolerances (Class 1/2 or Special/Standard Limits of Error) and voltage-temperature relationships.
- GB/T 9452: Chinese national standard for determining the effective heating zone of heat treatment furnaces, including detection thermocouple requirements.
- GB/T 16839: Chinese thermocouple standard harmonized with IEC 60584.
- ISO/IEC 17025: competence requirements for the accredited laboratories performing sensor calibration.
Selection Criteria at a Glance
- Temperature range and process temperature relative to sensor limits.
- Accuracy class sufficient for the furnace uniformity tolerance (tolerance should be a small fraction of the allowed band).
- Base-metal versus noble-metal construction, which controls drift and reuse rules.
- Sheath and insulation materials compatible with high vacuum (low outgassing, low vapor pressure).
- Calibration traceability, correction factor application, and usage counting.
- Mechanical form factor: probe diameter, flexibility, routing through vacuum feedthroughs, and response time.
Testing Samples
Type K and Type N Mineral-Insulated (MI) Base-Metal Probes
Sheathed, magnesium-oxide-insulated thermocouples in Inconel 600 or 310 stainless sheaths, typically 1.5 mm to 6 mm diameter, are the default choice for surveys up to approximately 1100°C to 1200°C. Type K (NiCr–NiAl) offers Class 1 accuracy of ±1.5°C up to 375°C and ±0.004·|t| above, while Type N (NiCrSi–NiSi) provides measurably better drift resistance at sustained high temperature due to its silicon-bearing alloys. For a ±5°C or wider furnace class at moderate temperatures, special-limits Type K or Class 1 Type N probes are usually the most economical option.
Type S, R, and B Noble-Metal Probes
For surveys at 1200°C to 1600°C, or for tight furnace classes (±3°C) where the sensor tolerance must consume as little of the band as possible, platinum-rhodium thermocouples are selected. Type S (Pt-10%Rh vs Pt) and Type R (Pt-13%Rh vs Pt) hold Class 1 accuracy of ±1°C up to 1100°C and approximately ±1 to ±2°C at higher temperatures. Type B (Pt-30%Rh vs Pt-6%Rh) extends coverage toward 1700°C and exhibits very low drift, though it is unsuitable below roughly 400°C. Noble-metal probes use high-purity alumina insulators rather than MI construction, because metallic magnesium oxide insulation would contaminate platinum sensors at high temperature.
Survey Fixtures and Dummy Loads
Sensors are rarely floated in free space. They are mounted on a survey rack or attached to a dummy load of representative mass and thermal properties, with the measuring junction located at the exact grid point defined by the survey plan. The fixture material (typically stainless steel or heat-resistant alloy, graphite-free unless the process requires it) must not outgas, spall, or react with the hot zone under vacuum.
Testing Items
Before any survey, the selected thermocouples are verified and characterized through the following test items, each of which feeds the measurement uncertainty budget of the final uniformity result.
- Calibration verification against a traceable reference (Type S/R working standard or fixed-point cell) at points bracketing the survey temperature, with correction factors documented.
- Limits-of-error confirmation to the claimed class (IEC 60584 Class 1 or ASTM E230 Special Limits of Error).
- Insulation resistance testing of MI probes (typically ≥100 MΩ at room temperature, verified after any bending or exposure) to detect sheath damage or moisture ingress.
- Thermoelectric homogeneity screening to reveal junction-region chemical or metallurgical inhomogeneity that causes spurious offsets in thermal gradients.
- Vacuum compatibility assessment: outgassing rate, insulation material integrity, and vapor pressure of sheath alloys at survey temperature.
- Cold-junction reference verification of ice baths, isothermal terminals, or electronic compensation in the data acquisition system.
- Post-exposure recalibration of noble-metal sensors to quantify drift accumulated during the survey cycle.
Testing Methods
Sensor Placement Geometry
The qualified working zone of a horizontal vacuum furnace is bounded by the hot zone ends and the hearth rails. For most working zones a minimum nine-point grid is applied: eight sensors at the corners of the bounding box and one at the geometric center. Larger zones or longer chambers require additional intermediate sensors along the axis. Every sensor position is dimensionally documented so the survey can be reproduced, and sensor junctions are fixed to the fixture or dummy load with high-temperature wire or ceramic caps to guarantee positional stability during pump-down and heating.
Vacuum Feedthrough and Routing
Probe leads are routed out of the chamber through hermetic vacuum feedthroughs or compression fittings at a chamber port. Inside the chamber, cable routing must avoid contact with heating elements, avoid sharp bends that mechanically damage MI sheaths, and keep base-metal extension wire fully inside isothermal or low-gradient regions where possible. Only vacuum-rated, inorganic insulation (MgO, alumina) may be used inside the chamber; organic insulation releases gas and contaminates the load.
Survey Execution Sequence
- Install the survey fixture with calibrated sensors at the documented grid positions and complete vacuum leak-back verification.
- Pump down to the specified vacuum level and heat at the production ramp rate to the survey set temperature.
- Observe stabilization: wait until all sensors settle and the control thermocouple holds setpoint, then begin recording.
- Record all sensors at fixed intervals (commonly every one to five minutes) for the specified number of readings, or the specified soak duration.
- Apply the calibration correction factors to each sensor, compute the corrected spread (maximum minus minimum), and compare against the furnace class tolerance.
- Repeat at additional temperature setpoints if the qualified range spans multiple zones; low, mid, and high setpoints are typical.
- Perform an empty-chamber survey and, where required by the process specification, a loaded survey using production-representative dummy loads.
Usage Limits and Reuse Rules
Under AMS 2750, base-metal TUS thermocouples are generally restricted to a single survey at temperature, because each excursion permanently alters their calibration; noble-metal sensors may be reused for a defined number of cycles subject to recalibration and recalculated correction factors. GB/T 9452 similarly requires detection thermocouples to carry valid calibration certificates covering the survey temperature. Records of sensor serial numbers, usage counts, and correction factors are retained with the survey report.
Applications
- Vacuum brazing of aerospace heat exchangers, honeycomb structures, and turbine components requiring Class 2 or tighter uniformity.
- Vacuum hardening and tempering of high-speed steels, tool steels, and bearing steels where ±5°C control of the austenitizing temperature governs hardness and distortion.
- Solution treatment and aging of nickel-base superalloys and titanium alloys, often surveyed at multiple setpoints across a wide range.
- Vacuum annealing of stainless steel and specialty alloys for bright, oxide-free finishes with tight metallurgical tolerances.
- Sintering and degassing of powder metallurgy parts in graphite or all-metal hot zones.
- Medical device heat treatment, where process validation depends on documented uniformity within the qualified working zone.
- Initial furnace qualification, annual or semi-annual re-surveys after hot-zone repairs, and surveys following relocation or control system changes.
FAQ
Which thermocouple type should be used for a 1300°C horizontal vacuum furnace survey?
At 1300°C, base-metal Type K or Type N probes are at or beyond their practical stability limit, so Type S or Type R noble-metal thermocouples with alumina insulation are recommended. They retain Class 1 tolerance, drift far less at sustained high temperature, and remain reusable subject to recalibration. If the furnace also operates near 1600°C to 1700°C, Type B should be evaluated for the upper setpoints.
How accurate do the survey sensors need to be?
The sensor tolerance, including calibration uncertainty, should consume only a small fraction of the allowed uniformity band, commonly one-third or less. For a Class 2 furnace (±6°C), Class 1 or special-limits sensors with a total expanded uncertainty near or below ±1°C to ±2°C are appropriate. Using standard-limits base-metal probes for a ±3°C furnace class risks rejecting a compliant furnace or, worse, accepting a non-compliant one.
How many thermocouples are required for the survey?
The minimum count depends on the working-zone volume and the governing specification. A nine-point arrangement (eight corners plus center) is the baseline for typical furnace chambers; small zones may permit fewer sensors while large or long horizontal zones require intermediate points. AMS 2750 tables and GB/T 9452 both define the minimum sensor count by zone dimensions.
Can survey thermocouples be reused?
Base-metal (Type K, N) TUS thermocouples are normally single-use per survey because thermal cycling permanently shifts their calibration. Noble-metal (Type S, R, B) thermocouples may be reused for a limited number of cycles provided they are recalibrated and new correction factors are applied, and any visibly contaminated or embrittled section is removed.
Why is ordinary insulated thermocouple wire unsuitable inside a vacuum chamber?
Organic insulation such as PVC or PTFE outgasses under vacuum, degrades the vacuum level, and can deposit contamination on the hot zone and workload. Vacuum-rated surveys use mineral-insulated metal-sheathed probes or ceramic (alumina) insulated noble-metal assemblies, both of which are stable and low-outgassing at process temperature.
How often must a horizontal vacuum furnace be re-surveyed?
Survey frequency is set by the governing pyrometry specification and the furnace class and instrumentation type, with typical intervals ranging from monthly to annually. Additional surveys are required after hot-zone replacement or repair, relocation, controller changes, or any event that could alter the thermal field, such as heating element failure or significant insulation degradation.