Printer Friendly

Effect of thermal exposure time on the relaxation of residual stress in high pressure die cast AM60.


Magnesium alloys are becoming more commonly used for large castings with sections of varying thicknesses. During subsequent processing at elevated temperatures, residual stresses may relax and become a potential mechanism for part distortion. This study was conducted to quantify the effects of thermal exposure on residual stresses and relaxation in a high pressure die cast magnesium (AM60) alloy. The goal was to characterize relaxation of residual stresses at temperatures that are commonly experienced by body components during a typical paint bake cycle. A residual stress test sample design and quench technique developed for relaxation were used and a relaxation study was conducted at two exposure temperatures (140[degrees]C and 200[degrees]C) over a range of exposure times (0.25 to 24 hours). The results indicate that a significant amount of residual stress relaxation occurred very rapidly during exposure at both exposure temperatures.

CITATION: Hill, H., Zindel, J., and Godlewski, L., "Effect of Thermal Exposure Time on the Relaxation of Residual Stress in High Pressure Die Cast AM60," SAE Int. J. Mater. Manf. 9(3):2016.


Magnesium die castings are commonly used for part consolidation and significant weight reduction in production vehicles. In particular, the Aluminum-Manganese series of Magnesium alloys is favorable because of its low density and relatively high ductility as compared to the Aluminum-Zinc series [1]. There is an abundance of published research focusing on the characterization of microstructure and the mechanical properties in Mg castings that are used for automotive applications [2,3]. However, no studies were found which characterize the behavior of relaxation of residual stresses in Mg-Al during thermal exposure.

While analysis of creep behavior is outside the scope of the present investigation, understanding its tendency in AM60 is warranted as it is the primary mechanism behind the relaxation of residual stresses at elevated temperatures. For all metals, there exists a temperature above which time-dependent deformation can occur at stress levels well below the material's yield strength. This behavior, known as creep, generally becomes important in metals at temperatures above 0.4T_m (T_m= absolute melting temperature of metal or alloy) [4]. Numerous experiments have been conducted in order to characterize the creep resistance of Mg-Al alloys and identify the principal creep mechanisms. One such study investigated creep behavior in a commercial AM60 alloy as a function of time and temperature. Significant strain rates were observed at 150[degrees]C exposure with applied tensile stresses in the range of 20-75 MPa [5]. Another study suggests that creep of die-cast Mg-Al alloys can be thermally activated between 70[degrees]C and 150[degrees]C [6].

Use of high pressure die cast (HPDC) AM60 in the Ford MKT rear liftgate inner has highlighted the need for residual stress relaxation data for this material. The presence of large thermal gradients across a body, such as those imposed during the quenching of large components after casting, can induce high levels of residual stress within the material. Therefore, if substantial residual stress exists in a casting, there is a potential at high temperatures that these residual stresses may relax, inducing significant part distortion. Part distortion that may occur during the paint bake cycle when the part reaches 200[degrees]C could lead to residual stress as a potential culprit. Since temperatures reached during paint bake exceed those that have been shown to initiate creep in Mg-Al alloys, it is feasible that residual stresses present before paint bake could cause dimensional changes in components [5, 6, 7]. The goal of this study is to quantify the relaxation of residual stresses in a magnesium high pressure die casting through the use of thermal exposure.


Paint Bake Cycle

A typical paint application cycle involves a series of processes, of which the E-Coat and Prime processes expose a component to the highest temperatures. The set point temperatures for the E-Coat and Prime are 200[degrees]C and 140[degrees]C, respectively. Actual part temperatures of the E-Coat and Prime processes were obtained by using thermocouples that were attached at various locations on a rear Mg inner panel to record the temperature profiles during the paint bake process. The thermocouple data was used to characterize the temperatures during each step of the process in terms of part exposure at the selected temperatures, as outlined in Table 1. Based on this information, it seemed reasonable to select 140[degrees] and 200[degrees] as the exposure temperatures.


The residual stress test castings were produced with a 400 ton high pressure die casting machine using a cold chamber process. Prior to ladle transfer, the furnace melt temperature was set to 730[degrees]C. Samples were cast using a high speed plunger velocity of 3.0 m/sec, and the casting cycle time was approximately 40 seconds. After removal from the die, castings were cooled using forced air. Once the parts reached room temperature, the runner, overflows and flash was removed from each casting.

Baseline Residual Stress Sample

The experimental setup used to quantify residual stress relaxation is adopted from a similar study in which a unique sample geometry (Figure 1) and rapid quench technique (Figure 2) were specifically designed for the investigation of residual stresses [8].

In order to study stress relaxation, you must first be able produce samples with a controlled initial stress (baseline samples) with minimal variation. This consists of two steps: heating the entire sample to a desired temperature in a furnace then isolating and quenching the lower portion of the sample in water, as shown in Figure 2. The quench technique induces a substantial residual stress in what is known as the critical bridge region (Figure 1) of the casting. This technique induces an extreme thermal gradient throughout the sample causing the top region to plastically deform and during the final stages of the quench the upper bridge experiences a final tensile stress which induces the local residual stress.

For a specific combination of sample temperature at the time of quench and quench water temperature, the same level of residual stress can be consistently reproduced for numerous samples. By repeating this process for a sufficiently large number of samples under the same conditions, the average level of induced residual stress can be determined and considered characteristic of that sample population. This approach is used to establish what is referred to as the baseline condition for the samples after quenching.

Baseline data from the experiments conducted as part of the original test casting and quench technique development showed that repeatable stress measurements could be achieved with water temperatures in the range of 10[degrees]C-25[degrees]C [8]. In this experiment, temperature of the quench water was continuously maintained within approximately 20[degrees]C-26[degrees]C.

Thermal Exposure

The relaxation of residual stress was characterized for two temperatures, 140[degrees]C and 200[degrees]C. For each temperature, samples were aged at various times in increments from 0.25 - 24 hours. A complete thermal exposure schedule is provided in Table 2. Five samples were thermally exposed per condition to characterize the material.

A small hole was drilled (1 mm dia.) in the thick end of one sample per thermal exposure, where a type K thermocouple was inserted and used to monitor the samples temperature during thermal exposure cycles. All samples were in an electrical resistance furnace. For thermal exposure times of one hour or less, the samples were treated in a salt bath inside the furnace in order for the samples to reach thermal exposure temperature within the desired timeframe. The draw salt used for heat treating is a eutectic nitrate/nitrite mixture with a melting point of 135[degrees]C. With a wide working temperature range and high thermal conductivity, the salt allowed for rapid heating of the samples at both the 140[degrees]C and 200[degrees]C temperatures.

Residual Stress Measurement

A uniaxial pre-wired strain gage of nominal 120[OMEGA] resistance (type CEA-13-250UW-120) was mounted on the underside of the critical bridge region of each sample. Each gage was wired to a remote terminal block, which was connected to a portable Model 5100B Scanner from Vishay Precision Group. Each of the samples was sectioned to quantify the amount of residual stress still present after thermal exposure. This destructive method involves making a cut in the sample and recording the strain resulting from any internal stresses [9]. Sectioning was performed on a band saw by making a single cut on the bridge region of each sample adjacent to the strain gauge. Using StrainSmart data acquisition software, the microstrain experienced during sectioning was recorded for each sample.


Alloy Composition

Chemical analysis was performed on the AM60 alloy using a SPECTROMAXx arc spark OES metal analyzer. Table 3 gives the average composition from three readings along with the ASTM specification for AM60B [10].

Alloy Microstructure

In order to characterize the AM60 material in terms of microstructure, cross-sectional samples taken from the critical region section of as-cast pieces were mounted and polished for metallurgical examination. Micrographs obtained during optical observation at varying magnifications are presented in Figures 3, 4 and 5.

The general microstructure of an HPDC Mg-Al alloy contains primary magnesium ([alpha]-Mg) phase surrounded by the eutectic (Mg_17 Al_12) phase. With increasing Al concentration, the primary phase advances away from a cellular structure to become more dendritic in nature. Externally solidified grains are often present (Figure 4) in these alloys due to premature solidification in the shot chamber [11].

Baseline Conditions

Each sample was heated in an electrical resistance furnace and subjected to an initial quench as previously outlined, in order to induce a residual stress in the bridge area of the casting. Samples were quenched from 300[degrees]C and 400[degrees]C, resulting in two baseline conditions. Thirty samples (15 for each quench temperature) were quenched and sectioned without thermal exposure in order to determine the natural variation of the baseline level of residual stress; the elastic strain measurements for these baseline samples are presented in Figure 7. Quenching from 400[degrees]C results in a higher thermal gradient in the sample and, hence, a higher level of residual stress in the critical bridge region as compared to quenching from 300[degrees]C. Using a textbook value for elastic modulus of AM60B (45GPa), the average residual stress in the 300[degrees]C and 400[degrees]C baseline quench conditions was determined to be 52MPa and 96MPa, respectively [1].

Thermal Exposure Conditions

The relaxation profiles in Figures 8 and 9 characterize residual stress relaxation in the bridge region of the AM60 samples during thermal exposure. Each data point represents the average value for microstrain measured from five samples. The average strain measured for the baseline conditions is plotted on the vertical axis (0 hours of exposure) in Figures 8 and 9. Subsequent data points represent the relaxation of residual stress present in the samples after thermal exposure for the noted times.

Figure 8 shows that the samples from the 300[degrees]C baseline-quench condition experienced an average reduction of 123 measured microstrain (11%) after 15 minutes at 140[degrees]C; whereas samples thermally exposed from the 400[degrees]C baseline-quench condition experienced an average reduction of 391 measured microstrain (18%) after 15 minutes at 140[degrees]C. A similar trend occurs for exposure temperature at 200[degrees]C (Figure 9), wherein the samples exposed with higher states of initial stress, corresponding to the higher quench temperature, experience relaxation faster than those with lower states of initial stress.

Comparing the results in Figures 8 and 9, it is clear that the rate of residual stress relaxation is accelerated by increasing the exposure temperature from 140[degrees]C to 200[degrees]C. Over half of the residual stress initially present in the samples is relaxed after exposure for 15 minutes at 200[degrees]C. This is true for both the high- and low-level baseline conditions. The level of residual stress measured after exposure at each time and temperature is represented in Table 4 as a fraction of the residual stress initially present before thermal exposure.

Analysis was conducted to assess the applicability of the following power law creep relation to the experimental data for residual stress relaxation:


Strain rate, [??], is a function of both stress, [sigma], and temperature, T. Values of the material parameters A, n, and Q can also vary with stress and temperature. Calculations were performed to determine the values of the material constants from strain data. Equations were fitted to the experimental data (Figure 10) and their first derivatives were used to approximate the strain rates for each baseline strain condition and thermal exposure temperature.

The ln([epsilon]) vs ln([sigma]) was plotted for each baseline stress condition and exposure temperature (Figure 11). For each baseline stress condition and exposure temperature, the value of the stress exponent, n, is equal to the slope of the corresponding line; these values are shown in Table 5. An average value of 5.3 was computed for n. These values agree with those reported in other studies of AM60 within the same temperature and stress regimes [5,12].

The value of Q was determined by plotting the [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] vs 1/T. The slope of this plot for each of the exposure temperatures is equal to - Q/R, and the y-intercept is equal to ln(A).

The calculated activation energies were 116 kJ/mol and 154 kJ/mol for samples quenched from 300[degrees]C and 400[degrees]C, respectively. The average activation energy, Q, is equal to 135 kJ/mol, which is in the range of values reported for the activation energy of dislocation creep in Mg-Al alloys [5,12,13].


This study focused on stress relaxation from two different levels of initial residual stress in an AM60 alloy during thermal exposure at two different temperatures over a range of exposure times from 0.25-24 hours. Strain measurements recorded during sectioning of each sample were used to generate thermal exposure curves, from which the following observations can be made about the effects of exposure time on the relaxation of residual stresses:

1. The initial stress level of baseline samples quenched at 300[degrees]C was 52Mpa and baseline quenched at 400[degrees]C was 96MPa.

2. The rate of residual stress relaxation in the samples was predominantly influenced by exposure temperature. Increasing the exposure temperature from 140[degrees]C and 200[degrees]C resulted in an increase in the relaxation as shown:

a. Thermal exposure at 140[degrees]C reduced the initial residual stress by approx. 10% after 15 minutes.

b. Thermal exposure at 200[degrees]C reduced the initial residual stress by approx. 50% after 15 minutes.

3. The power law creep relation was applied to the experimental residual stress relaxation data for the AM60 test castings. Average values for the material parameters Q and n were calculated from the strain values measured during sectioning. Activation energy Q was calculated as 135kJ/mol. Average values for the stress exponent n were determined to be 6.9 and 3.8 for thermal exposure at 140[degrees]C and 200[degrees]C, respectively.

4. Results from this study indicate that if substantial amounts of residual stress are present in areas of an AM60 casting, they have the potential to be relieved by 50% under the conditions of this paint bake cycle. Further studies should be performed at the component level in order to understand the effects of thermal exposure on residual stress. Quantification of the amount of residual stress produced during quenching of a part is an important step in trying to correlate these results to those for production components.


[1.] Avedesian, M. and Baker, H., "General Engineering Properties of Casting Alloys," ASM Specialty Handbook: Magnesium and Magnesium Alloys, Materials Park, OH: ASM International, 1999, pp. 226-227.

[2.] Mao, H., Brevick, J., Mobley, C., Chandrasekar, V. et al., "Microstructural Characteristics of Die Cast AZ91D and AM60 Magnesium Alloys," SAE Technical Paper 1999-01-0928, 1999, doi:10.4271/1999-01-0928.

[3.] Dierks, M., Kuhn, C., and Etling, J., "Enhanced Mechanical Properties of Die Cast AM Series Magnesium; Through Part Design, Die Design and Process Control," SAE Technical Paper 940410, 1994, doi:10.4271/940410.

[4.] Callister, W., Jr., and Rethwisch, D., "Failure," in Materials Science and Engineering: An Introduction, 8th ed. Hoboken, NJ: John Wiley & Sons, 2010, ch. 8, sec. 12-13, pp. 265-268.

[5.] Chen, Z., Huang, J., Decker, R., LeBeau, S., et al, "The Effect of Thermomechanical Processing on the Creep Behavior and Fracture Toughness of Thixomolded AM60 Alloy," in Magnesium Technology 2011, Sillekens W., Agnew S., Neelameggham N. and Mathaudhu S., Eds. John Wiley & Sons, Inc., 2011, pp. 85-89.

[6.] Blum, W., Zhang, P., Watzinger, B., Grossmann, B., et al., "Comparative study of creep of the die-cast Mg-alloys AZ91, AS21, AS41, AM60 and AE42," in Materials Science and Engineering: A, vols. 319-321, Lavernia E.J., Ed., Elsevier, 2001, pp.735-740.

[7.] Saddock, N., "Microstructure and Creep Behavior of Mg-Al Alloys Containing Alkaline and Rare Earth Additions," Ph.D. dissertation, Dept. Materials Science and Eng., Univ. of Michigan, Ann Arbor, MI, 2008.

[8.] Godlewski, L., Su, X., Allison, J., Gustafson, P. et al., "A Test Method for Quantifying Residual Stress Due to Heat Treatment in Metals," SAE Technical Paper 2006-01-0319, 2006, doi:10.4271/2006-01-0319.

[9.] Rossini, N., Dassisti, M., Benyounis, and K., Olabi, A., "Methods of Measuring Residual Stresses in Components," in Materials & Design, vol. 35, Blundell M., Jerrams S. and Edwards K.L., Eds., Elsevier, 2012, pp.572-588, doi:10.1016/j.matdes.2011.08.022.

[10.] ASTM Standard B94, 2013, "Standard Specification for Magnesium-Alloy Die Castings," ASTM International, West Conshohocken, PA, 2013,

[11.] Dahle, A., Lee, Y., Nave, M., and Schaffer, P., "Development of the As-Cast Microstructure in Magnesium-Aluminium Alloys," in Journal of Light Metals, vol.1, issue 1, Elsevier, pp.61-72.

[12.] Kondori, B., and Mahmudi, R., "Impression Creep Characteristics," in Metallurgical and Materials Transactions A, vol. 40, issue 8, pp.2007-2015, Aug. 2009, doi:10.1007/s11661-009-9867-4.

[13.] Somekawa, H., Hirai, K., Watanabe, H., and Takigawa, Y., "Dislocation creep behavior in Mg-Al-Zn alloys," in Materials Science and Engineering: A, vol. 407, issues 1-2, Elsevier, 2005, pp.53-61, doi:10.1016/j.msea.2005.06.059.


Haley Hill



Thanks to Darryl Van Gaal from Oakville Assembly Complex Paint Department for his collaboration, as well as Mark Nichols and Mei Li from Ford Research and Innovation Center for helping make this work possible.


Table A1. Microstrain Measurements for All Aged Samples

Aging     Sample  300[degrees]C Quench
Time (h)  Number  140[degrees]C  200[degrees]C
                  Aging          Aging

0.25           1           1036            527
               2           1050            431
               3           1065            602
               4           1032            521
               5            968            556
               6                           487
               7                           505
0.5            1            812            312
               2            711            411
               3            666            341
               4            382            447
               5            898            292
               5            924            477
               7           1056            437
               8            989            476
1              1            686            210
               2            440            386
               3            643            411
               4            613            373
               5            458            441
               6            770
               7            900
               8            925
               9            907
              10            868
5              1            661            204
               2            575            199
               3            599            171
               4            656            204
               5            390            165
10             1            564            209
               2            456            105
               3            478            195
               4            354            162
               5                           163
24             1            444             85
               2            363            115
               3            400            140
               4            342            154
               5            316             91

Aging     Sample  400[degrees]C Quench
Time (h)  Number  140[degrees]C  200[degrees]C
                  Aging          Aging

0.25           1           1774            965
               2           1724           1033
               3           1735           1051
               4           1766           1084
               5           1762           1113
0.5            1           1671            862
               2           1663            884
               3           1619            584
               4           1491            740
               5           1503            892
1              1           1307            478
               2           1301            560
               3           1512            644
               4           1569            662
               5           1602            686
5              1           1292            347
               2           1279            250
               3           1313            449
               4           1294            319
               5           1222
10             1           1052            219
               2           1121            218
               3           1094            142
               4            935            210
               5           1087            281
24             1            956            160
               2            933            174
               3            834            260
               4            934            199
               5            841            224

Table A2. Average Measured Strain and Calculated
Residual Stress for AM60B Sample Sets

Quench          Aging     Aging     Average
Temp            Temp      Time (h)  Strain ([epsilon][mu])
                Baseline                              1153

30([degrees]C)       140      0.25                    1030
                     140      0.5                      805
                     140      1                        721
                     140      5                        576
                     140     10                        463
                     140     24                        373
                     200      0.25                     518
                     200      0.5                      399
                     200      1                        364
                     200      5                        189
                     200     10                        167
                     200     24                        117
                Baseline                              2143
40([degrees]C)       140      0.25                    1752
                     140      0.5                     1589
                     140      1                       1458
                     140      5                       1280
                     140     10                       1058
                     140     24                        900
                     200      0.25                    1049
                     200      0.5                      792
                     200      1                        606
                     200      5                        341
                     200     10                        214
                     200     24                        203

Quench          Residual      Relaxation
Temp            Stress (MPa)  Rate (MPa/h)
                          52  -

30([degrees]C)            46          22
                          36          41
                          32           7.6
                          26           1.3
                          21           1.0
                          17           0.3
                          23         114
                          18          21
                          16           3.2
                           9           1.6
                           8           0.2
                           5           0.2
                          96  -
40([degrees]C)            79          70
                          72          29
                          66          12
                          58           2
                          48           2
                          41           0.5
                          47         197
                          36          46
                          27          17
                          15           2.4
                          10           1.1
                           9           0

Haley Hill, Jacob Zindel, and Larry Godlewski

Ford Motor Company

Table 1. Thermocouple Data for Inner Panel from E-Coat and Prime Cycles

Process  Max Temp       Time Above     Time Above
         Recorded       140[degrees]C  200[degrees]C

E-Coat   203[degrees]C         21 min  7.5 min
Prime    163[degrees]C         20 min

Table 2. Thermal Exposure Schedule with Times and Temperatures Used in
Relaxation Study

Quench Temp ([degrees]C)    300             400
Exposure Temp ([degrees]C)  140     200     140     200

Exposure Time                 0.25    0.25    0.25    0.25
(hours)                       0.5     0.5     0.5     0.5
                              1       1       1       1
                              5       5       5       5
                             10      10      10      10
                             24      24      24      24

Table 3. Chemical Composition of AM60B Alloy (weight percent)

Element       Measured  Specification

Al               6.09        5.5-6.5
Mn               0.29        0.24-0.6
Si               0.006       0.1
Zn               0.06        0.22
Fe               0.003       0.005 max
Cu               0.003       0.010 max
Ni              <0.002       0.002
Other (each)    <0.02        0.02 max

Table 4. Microstrain Measured During Sectioning of Thermally Exposed
Samples (values expressed as fractions of initial baseline stress level
for the corresponding quench temperature)

Fraction of Initial (Baseline) Microstrain Measured in Thermally
Exposed Samples
ExposureTime  300[degrees]C Quench
(h)           140[degrees]C  200[degrees]C
              Exposure       Exposure

 0.25                  0.89           0.45
 0.5                   0.70           0.35
 1                     0.63           0.32
 5                     0.50           0.16
10                     0.40           0.14
24                     0.32           0.10

Fraction of Initial (Baseline) Microstrain Measured in
Thermally Exposed Samples
ExposureTime  400[degrees]C Quench
(h)           140[degrees]C  200[degrees]C
              Exposure       Exposure

 0.25                  0.82           0.49
 0.5                   0.74           0.37
 1                     0.68           0.28
 5                     0.60           0.16
10                     0.49           0.10
24                     0.42           0.09

Table 5. Calculated n Values from Experimental Data for Thermal
Exposure at 140[degrees]C and 200[degrees]C

Temperature   Baseline Quench               Average
([degrees]C)  300[degrees]C  400[degrees]C

140                     5.8            7.9      6.9
200                     4.0            3.6      3.8
COPYRIGHT 2016 SAE International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hill, Haley; Zindel, Jacob; Godlewski, Larry
Publication:SAE International Journal of Materials and Manufacturing
Article Type:Report
Date:Aug 1, 2016
Previous Article:Characterization of residual stresses on steel coil springs via x-ray diffraction techniques.
Next Article:Residual stress distributions in rectangular bars due to high rolling loads.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters