Rapid phase transformation and high thermal gradients that occur during quenching can cause distortion, which can lead to costly hard machining and rejection. Residual stress and deformation induced by quenching are a consequence of complex interaction among the parameters of phase transformation, heat transfer and strain in the steel (Fig. 1).

It is for this reason that researchers at the Center for Heat Treating Excellence (CHTE) at Worcester Polytechnic Institute (WPI) are working to identify heat-treatment process parameters to control distortion and residual stress. The team working on the research includes: Richard D. Sisson Jr., George F. Fuller professor of mechanical engineering at WPI and CHTE technical director; Mei Yang, WPI assistant research professor and associate CHTE technical director; and Haixuan Yu, a Ph.D. student in material science and engineering at WPI.

The overall goal of the research is to help provide the manufacturing industry with heat-treatment processing parameters that will help to significantly reduce costs.

This paper outlines the work that has been done to date in three key areas.

  1. The measurements of the heat-transfer coefficient (HTC) of alphaquench 5300 quenching oil[1] are conducted as a function of temperature with selected agitation and quench start temperature using CHTE quench probe system.[2]
  2. The quenching experiments with the Navy C-ring specimen, which is selected because it offers a range of cooling rates due to its varying cross sections, are described and discussed.
  3. Heat-treatment modeling, using commercial heat-treatment code DANTE, is presented. For clarification purposes, DANTE is heat-treatment analysis software that is a finite-element integration package that keeps track of the thermal-mechanical and phase transformations. The goal is to use this software to predict the distortion in the part as a function of the heat-treating parameters.

 

Measuring Heat-Transfer Coefficient Using CHTE Quench Probe

Quench probes are used to measure HTC at the interface between oil and material. Analysis of measurements provides a curve of HTC as a function of temperature. This curve is a required input for DANTE simulations.

The geometry of the CHTE quench probe designed and used for testing, along with the coupling, is shown in Fig. 2. The diameter of the probe is 3/8 inch, and the length of probe and coupling is 3 inches. An Allen set screw of 1/16 inch (#6-32) was used in the coupling to tighten thermocouple and probe. Resbond 989, a high-purity alumina ceramic, was applied to the interface between the probe and coupling to eliminate oil leakage during quenching. The fabricated CHTE quench probe made by a Haas ST-30Y CNC lathe is shown in Fig. 3a. The assembled CHTE quench probe is shown in Fig. 3b.

Figure 4 presents that HTC increases with the increase of agitation level, while the quench start temperature has no significant effect on HTC.

 

Quenching Experiments with Navy C-Ring Specimen

The Navy C-ring offers a range of cooling rates due to its varying cross sections. Its various rates of cooling and phase transformation result in residual stresses and geometric distortion.

The Navy C-ring is very convenient to capture these combined effects. Its sensitivity to these combined effects is aggregated at macroscopic level by changes of the width of its gap opening at the thinnest area. Therefore, the Navy C-ring allows affordable process quality control by placing some rings along with heat-treatment loads. The consistency of gap opening size, load after load, is an indication of a stable process (i.e., part quality). In the present work, it allows validating the model and understanding process sensitivity.

To investigate the correlation between processing parameters and quenching-induced distortion, Navy C-ring specimens were fabricated from AISI 4140 hot-rolled steel round bars using a Haas ST-30Y CNC lathe (Fig. 5). Ten Navy C-ring specimens were manufactured for quenching experiments, and the test plan is presented in Table 1.

Test parts and quench probe were tied to a quenching basket, heated in an AFC-Holcroft CJ-4718 atmospheric furnace at selected quench start temperatures for 40 minutes, and quenched in alphaquench 5300 quenching oil with various pump speeds and quenching orientations. The gap opening was measured by a Fowler electronic IP54 inside micrometer[3] after quenching experiments. The flow rate of the quenchant with selected pump speeds was measured by Swoffer current velocity meters.[4]

 

Heat-Treatment Modeling Using DANTE

The heat-treatment modeling of the Navy C-ring was carried out using finite element analysis (FEA) software ABAQUS coupled with heat-treatment analysis software DANTE. Figure 6 shows the three-dimensional finite-element (FE) model of the Navy C-ring for heat-treatment simulation. In order to have accurate simulation results, a finer mesh was applied near the surface.

Due to the symmetry of the specimen, only one-quarter of the part was simulated. The FE mesh consisted of 21,584 nodes and 18,900 quad elements. Node A was used to measure the distortion of the gap after heat-treatment simulation. Nodes B, C, D, E and F were used to track the temperature, cooling rate and phase fraction in quenching simulation.

The FE setup for the simulation is summarized in Table 2.
The air transfer time is the same as the time measured in the quenching experiment. Constant HTC as a function of temperature was applied to all surfaces of the part.

“We use this DANTE software to help us predict the effects of geometry and heat-treating process parameters and, in particular, the cooling ability of the media that we’re using for quenching,” Sisson said.

As is shown in Fig. 7 from both modeling and experimental results, higher pump speed results in larger distortion of Navy-C rings. Selected quench start temperatures do not have a significant effect on quenching distortion.

 

Key Findings and Conclusions

  • The simulation results have excellent agreement with the experimental results. Because of the geometry of the specimen, the difference of quenching response between tip and core is significant. The cooling rate in the tip region is much higher than that in the core region, and the distortion is also related to the phase transformation at various locations during quenching.
  • Higher agitation levels can produce higher cooling rates in the bottom section of the Navy C-ring and transform more austenite to martensite, instead of lower bainite, which enlarges the gap opening due to the relatively lower density (ρ) of martensite.
  • Selected quench start temperatures do not have a significant effect on the gap opening since they have no influence on HTC during quenching.
  • Design of experiment (DoE) combined with analysis of variance (ANOVA) was used to identify the quenchant type, part geometry, agitation speed, quenchant temperature and quenching orientation as the five most important processing parameters during quenching of the Navy C-ring specimen. These processing parameters have the most impact on Biot number; the greater the Biot number, the greater the distortion of the quenched part.
  • Heat-treatment modeling of the Navy C-ring using DANTE with selected steel grades was also conducted, as was the effect of carbon content on gap opening. Increasing carbon content corresponds to an increase in the gap opening. This trend continues up to carbon content of 0.5%, at which point the relationship is inverted due to the increase in retained austenite. DoE combined with ANOVA was used to identify the important features of the HTC curve during quenching of the Navy C-ring specimen.
  • The effect of parameters, in decreasing order of importance, are: Ms temperature, HTC at Ms, HTC max, HTC at convection, HTC at LF and the Tmax.

 

For more information:  Contact Richard D. Sisson Jr., director of the Center for Heat Treating Excellence, Worcester Polytechnic Institute, 100 Institute Road, Worcester, MA 01609; tel: 508-831-5335; e-mail: sisson@wpi.edu; web: www.wpi.edu/+chte.  

 

 

References 

  1. http://objects.eanixter.com/PD365571.PDF
  2. Maniruzzaman, M., Chaves, J., McGee, C., Ma, S., & Sisson Jr., R. (2002, July). CHTE quench probe system: a new quenchant characterization system. In 5th International Conference on Frontiers of Design and Manufacturing (ICFDM) (pp. 619-625)
  3. http://www.fowlerprecision.com/Products/Inside-Micrometers/1-2-Electronic-IP54-%20Inside-Micrometer-54-860-276.html
  4. Yu, H., Lu, Y., Xu, Y., & Sisson, R. (2019). The Effect of Surface Preparation on Retained Austenite Measurement. In 100 Years of E04 Development of Metallography Standards. ASTM International

 

The Center for Heat Treating Excellence is an alliance between industry and university researchers that addresses the short- and long-term needs of the heat-treating industry. Membership is unique because members have a voice in selecting quality research projects that help them solve today’s business challenges.