Accurate temperature measurement of both the furnace hot zone and the workload is critical for a successful heat-treat process cycle. Various types of thermocouples are used to record and control the furnace temperature and monitor and record the workload temperature during the heat-treat process.


Most vacuum heat-treat operators today follow the requirements set forth in AMS 2750E to ensure that all temperature measurements remain consistent and reproducible.

The Thermocouple

Thermocouples (TC) are temperature-measuring sensors composed of two dissimilar metals joined together to form a thermoelectric circuit that applies the Seebeck effect.[1] The change in temperature of the wire correlates to a change in voltage in the thermoelectric circuit. TC construction includes exposed, ungrounded sheath and grounded sheath designs. Exposed or bare-wire construction (Fig. 1) leaves the wires open to the atmosphere and is best suited for nonreactive atmospheres and the need for rapid response time ideal for sudden temperature changes.

Typical sheathing material may be Refrasil® or stainless steel (SS) mesh. Ungrounded and grounded junctions are placed in a protective sheath. The ungrounded junction has the wires electrically isolated from the sheath, whereas the wires are welded directly to the sheath in the grounded junction. The use of a grounded or ungrounded junction is a topic for deeper discussion than presented here.

Sheath material such as Inconel® 600, 304 SS or 316 SS are ideal for corrosion resistance at high temperatures up to 1148°C (2100°F). Sheath and wire dimension selection is also important for accurate results. A smaller-diameter, sheathed TC with smaller wires provides faster response times but is more fragile. Larger-diameter wires and sheaths are more rugged with longer life expectancy when used in the high-temperature range.

The ANSI-type symbol (Table 1) indicates wire material combinations in the thermocouple circuit. The chemistry of each wire and the metallurgical response to heat and atmospheric conditions determine the use and limitations of each thermocouple type.

Type S, used mainly for the furnace over-temperature control, operates well up to 2800°F. For higher-temperature applications, Type B or Type W3 would be required. A rigid and fragile alumina (Mullite) sheath protects the furnace thermocouple wires (Fig. 2a). Proper placement of the furnace hot-zone thermocouples is critical to ensure accurate measurement of the chamber temperature, not the heating element output. The TC junction must be a minimum of 2.5-3.0 inches beyond the heating elements (Fig. 2b).

Temperature uniformity surveys (TUS) are required by ASM 2750E. A universal TUS measurement structure shown in Figure 3 allows more flexibility for owners of various vacuum furnace designs.[2]

Types K and N are used to measure temperature of the work parts and can be bare wire or sheathed design (Fig. 1). In oxidizing, reducing, inert or vacuum environments, they are reliable and accurate up to 1148°C (2100°F). Type K is quickly losing favor in the industry to Type N due to the improved accuracy and extended lifetime of Type N (Table 1). In aerospace and medical applications, a one-use limit for Type K thermocouples is strictly enforced.

Work thermocouples (Fig. 1) measure heat transfer from the heating elements to the “work.” Factors affecting the rate of heat transfer include mass, surface area, geometry, thermal conductivity, absorptivity/emissivity, chemical composition of the work and position in the furnace.

Proper placement of work thermocouples is necessary to achieve correct temperature profiles of the work (part) and ensures proper metallurgical properties upon completion of the heat-treat cycle. Thermocouples placed deep into the center of the work in areas most shielded from radiant heat, such as the core or thickest part of the cross section, will be the slowest to reach equilibrium. Therefore, the exchange of heat between the furnace elements and the work is not linear (Fig. 4). Whenever possible, place the TC in existing holes or crevices such that the TC tip (hot junction) is in direct contact with the work.

Figures 5a and 5b illustrate how surface condition and surface chemistry influence the heat-transfer rate for a carbon-steel block to reach temperature. Each block had a TC inserted into a hole drilled into the center. In Figure 5b, one can see that the nickel-plated and polished surfaces took the longest time to reach equilibrium. These two surfaces had the lowest absorptivity of heat owing to lower emissivity factors. A lower emissivity factor indicates a greater propensity to reflect radiant energy back to the source.[3]

For many workloads processed in a vacuum furnace, the work thermocouple cannot be placed into or in direct contact with the part. In these cases, a dummy block works as a mimic of the actual part in the furnace. The dummy block has holes drilled deeply into the center and must be sized to match the largest cross section of work in the furnace.

Matching the surface condition, chemical composition, emissivity and thermal conductivity are also important. For example, titanium is less thermally conductive than stainless steel. If one were to use a stainless steel dummy block of similar size to a titanium part, the dummy block would prematurely indicate that the work has reached temperature. The actual temperature of the parts, however, would be lower, leading to an inadequate heat profile and possibly poor mechanical, physical and microstructural quality after heat treatment.

It is often not practical to make a dummy thermocouple block that is an exact duplicate of the workload parts. In Figure 6, the dummy block is one-third the actual work height. However, the critical cross-section dimensions are able to approach the true heating rate of the work.

Other Temperature-Monitoring Devices

Occasionally, a vacuum heat-treat process requires a temperature level that is not conducive to the use of Type K or Type N thermocouples. Further, use of the very expensive Type S, B or W3 thermocouples would be prohibitive, especially if reactive chemicals in the process damage the TC during use. Ceramic disks, such as the Orton TempTAB®, which sinter at controlled rates over a range of temperatures, are useful for such situations. The ceramic tab is measured before and after the process. From the shrinkage data, a temperature relationship can be generated (Figs. 7 and 8) using the Orton Trakker® software.


Accurate temperature measurement in a vacuum furnace is complex. This discussion has only touched upon key factors in this important process. These factors include:

  • Selection of appropriate thermocouple type and thermocouple sheath
  • Proper placement of control and work thermocouples
  • Furnace temperature uniformity surveys as required by ASM 2750E
  • Attention to workload cross section, mass, emissivity, surface condition and material chemistry
  • Correct use of dummy blocks having the same characteristics as the work
  • TempTABs to measure peak furnace temperature when work TCs are not feasible


For more information:  Contact Patricia Niederhaus, executive technical administrator, Solar Manufacturing, 1983 Clearview Road, Souderton, PA 18964; tel: 215-721-1502 x1240; e-mail:; web: Contributors to this article were Virginia Osterman, senior scientist, and William R. Jones, CEO, both of Solar Atmospheres Inc.


  1. Wilson, J. A. Sam, “Industrial Electronics and Control,” Science Research Associates, Inc., Chicago,  p. 115
  2. Fradette, Real and Jones, William R., “Optimizing Procedures for Temperature Uniformity Surveying of Vacuum Furnaces,” Solar Atmospheres and Solar Manufacturing Inc. Booklet #2
  3. Fradette, Real and Jones, Trevor, “Considering Emissivity Factors of a Workload When Projecting Heating Rates in a Vacuum Furnace,” Proceedings of the 27th Heat Treating Conference, 2013, ASM International, Materials Park, OH
  4., March 2015