NDT Solution
NDT of Precipitation Hardened Steels Using
Fluorescent Magnetic Particle Testing
by Bob
Potter*
This month's "NDT Solution" discusses the challenges in performing fluorescent magnetic particle testing of precipitation hardened steels. The author has provided a good overview of the primary types, heat treatment processes and commonly found discontinuities in this class of steel. The author has done an excellent job of presenting the basics and made the article interesting and useful for the engineers and inspectors involved in NDT of precipitation hardened steels.
G.P Singh
Associate Technical Editor
INTRODUCTION
This article concentrates on test findings of an aerospace finishing processor. The findings are reported based on historical test results supported by third party metallography. The findings are provided to the reader to better the understanding of magnetic particle testing (MT). One of the greatest challenges of testing aerospace parts using fluorescent MT is the testing of precipitation hardened stainless steels. Due to the nature of precipitation hardened steel manufacturing and subsequent refining, heat treating and cold working practices, it is imperative that the inspector have a basic knowledge of precipitation hardened steel production to accurately test and classify the associated discontinuities. Many discontinuities, when properly classified, are acceptable and do not affect part design function. Common discontinuities associated with precipitation hardened steels that are detected with MT include ferrite stringers, metallic and nonmetallic inclusions and segregation.
Precipitation hardened steels are a popular design choice in the aerospace sector.
Precipitation hardened steels, unlike ferritic, austenitic and martensitic steels, achieve both high strength and corrosion resistance using age tempering techniques. These steels are classified as either martensitic or semiaustenitic. Both types play key roles in the final heat treat condition and in discontinuity formation. Austenite is a solid solution, usually of carbon or iron carbide and is stable only at high temperatures. The presence of certain alloying elements, such as nickel and manganese, stabilizes the nonmagnetic austenite. Martensite is a phase formed by transformation of austenite during the cooling of the metal. It is the hardest constituent of precipitation hardened steel and is produced by rapid cooling following the heat treat cycle. Martensite is magnetic, with moderate to high permeability. Conversely, semiaustenitic steels are semimagnetic, with low permeability. Most precipitation hardened steels are supplied in the annealed condition (condition A) to facilitate subsequent machining, forming and hardening operations. Each of the precipitation hardened steels has a range of heat treatments that produce an equally wide range of strength, hardness, microstructure and magnetic permeability.
Hardening by aging is primarily accomplished after rapid cooling or cold working (ASM International, 1976). As an example, 17-7PH is aged using a solution anneal temperature of 1450 K (2150 ºF) for 4 h and by being air cooled. This process is followed with a 1350 K (1975 ºF) heat for 4 h, air cooling, a 1120 K (1550 ºF) heat for 24 h, air cooling and finally a 1060 K (1440 ºF) heat for 16 h and air cooling to complete the precipitation hardening process. Solution heat treatment is the process of heating an alloy to a suitable temperature, holding it at that temperature long enough to allow one or more constituents to enter into solid solution and then cooling it rapidly enough to hold the constituents in solution.
Precipitation hardened steels are a popular design choice in the aerospace sector. This is due to their high strength, corrosion resistance and mechanical properties. The following provides primary features, heat treat options and magnetic permeability of the most common precipitation hardened steels used in aerospace (Department of Defense, 1993).
PRIMARY FEATURE OF PRECIPITATION HARDENED STEELS
Martensitic
15-5PH steel transforms to a low carbon martensite, supersaturated with copper in the annealed condition. It is commonly refined by using vacuum arc remelting to achieve the desired properties and composition (free of delta ferrite). The magnetic permeability of this material significantly varies with heat treat condition, ranging from a permeability of 95 in condition A (solution treated) to 151 in condition H900. There are seven different heat treatments for this steel. Common aerospace applications include valves, shafts, fasteners, fittings and gears. Of the precipitation hardened steels tested, not only was this steel the most common, it also possessed the lowest amount of rejectable discontinuities.17-4PH steel has a composition similar to that of 15-5PH. It is a special chromium-nickel-copper alloy that achieves its strength and hardness through a combination of a martensitic transformation and precipitation hardening. Much like 15-5PH steel, the high percentage of copper content permits it to be hardened at low temperatures. At 1310 K (1900 ºF), the material is austenitic but undergoes transformation to martensitic at approximately 405 K (270 ºF). It has several of the same properties of 15-5PH steel, only it may contain delta ferrite. In fact, of the precipitation hardened steels tested in this study, this steel possessed the highest percentage of rejects. This is directly attributable to the lack of proper classification of the principal discontinuity observed - ferrite stringers. This is of concern to the NDT inspector because ferrite stringers have similar indication characteristics to nonmetallic inclusions and segregation. The magnetic permeability of this material remains consistent among the heat treat conditions ranging from 71 in the H1150 condition to 136 in the H1075 condition. This steel is usually produced in air melt furnaces and does not receive a secondary remelt. There are eight different heat treatments for this steel. As with 15-5PH, this alloy is normally supplied in the annealed or solution treated condition (condition A). Common aerospace applications include hinges, structural parts and fasteners.
Semiaustenitic
13-8PH steel has high transverse toughness, good resistance to stress corrosion cracking and high strength by a single low heat treatment. It is produced by vacuum induction melting plus consumable electrode vacuum arc remelting. Common aerospace applications include landing gears, structural sections, valves and shafts. Based on a review of test history, this steel presented very few rejectable discontinuities and was not included in this study.17-7PH steel has high strength and moderate corrosion resistance. Aluminum is added to this material to enhance weld slag formation during arc welding. The magnetic permeability of this material significantly varies with heat condition ranging from a permeability of 1.4 in condition A to 208 in condition TH1050. This steel is usually produced in an air melt furnace. Common aerospace applications include springs and washers. As with 13-8PH steel, this steel was not tested in this study due to low test frequency.
INHERENT PROCESSING
The inherent processing of precipitation hardened steel originates in either standard air melt or vacuum induction melt furnaces. At this stage, key alloy elements such as chromium, nickel, molybdenum and manganese are added. Typical chemical compositions and resulting mechanical properties of the common aerospace precipitation hardened steels are listed in Table 1 (ASM International, 1976).
Table 1 Typical properties of precipitation hardened streets
Precipitation
Hardened SteelsCarbon
Chromium
Nickel
Molybdenum Other Elements Ultimate Tensile Strength Hardness 15-5 Martensitic 0.07% 15.0% 4.0% - 2.5 to 44.5% copper
0.30 cadmium
and tantalum1030 MPa
(150 000 lb/in.2)38 Rc 17-4 Martensitic 0.07% 17.0% 4.0% - 4.0% copper, 0.15
to 0.45 cadmium
and tantalum1380 MPa
(200 000 lb/in.2)44 Rc 13-8 Semiaustenitic 0.05% 13.0% 8.0% 2.3% 1380 MPa
(200 000 lb/in.2)45 Rc 17-7 Semiaustenitic 0.09% 17.0% 7.0% - 1.0% aluminum 8960 MPa
(1 300 000 lb/in.2)85 Rc
During steel production, the molten steel is either continuously cast or cast using traditional ingot casting methods. For continuously cast steels, the liquid steel is tapped into a water cooled copper mold and begins to solidify as it travels down a conveyor system for subsequent hot working (rolling or extrusion) to the desired shape. This method provides good refinement but may contribute to the discontinuities described herein. With the conventional cast system, the molten metal is poured into stationary molds to form ingots. The ingots are then placed into an intermediate hold area for subsequent rolling into the desired shape. Conventionally cast products usually receive a homogenization treatment prior to rolling to improve the microstructure and thus reduce or eliminate several of the discontinuities identified with MT.
Air heat treating requires the steel to contain a sufficient amount of carbon and other alloying elements to harden fully during cooling (Boyer, 1984). Vacuum heat treating is performed to prevent contamination from air as well as to remove gases already dissolved in the metal. The solidification may also be carried out in a vacuum or at low pressure. The choice of the heat treat process is one of economics.
As previously stated, precipitation hardened steels are produced in either standard air melt or vacuum induction furnaces. Some of the steels, such as 15-5 and 13-8, receive additional refinement by the vacuum arc remelt process. In fact, 13-8 typically receives two vacuum arc remelts. The vacuum arc remelt process improves the homogeneity, fatigue and fracture toughness and the cleanliness of the steel and significantly reduces discontinuity formation. Due to the vacuum arc remelt process, there are fewer discontinuities identified with 13-8 and 15-5 than with the other precipitation hardened steels. As with the initial melt, the decision to use the vacuum arc remelt process is one of economics.
TEST CHALLENGES
Figure 1 provides a test breakdown by steel type. All of the parts tested were aerospace related. It should be noted that 17-4, while only accounting for 23% of the precipitation hardened steels tested in this study, accounted for 84% of the rejects. Of the discontinuities identified with 17-4, less than 1% were considered to be a result of secondary processing.
Figure 1 - Precipitation hardened steels tested by material type
Precipitation hardened steels have unique permeability properties. During production, there is a transformation from austenitic to martensitic (AK Steel, 1999). Second, when cold working steel that contains austenite, a ferromagnetic martensitic phase forms. As the steel is cold worked it becomes harder. This is contrary to the property of ferromagnetic materials in which the permeability decreases with increased hardness. Hence, the permeability increases with precipitation hardened steels when the steel is cold worked and becomes harder. Table 2 provides the mechanical properties of 17-4 and the resulting permeabilities.
Table 2 Properties of 17-4 precipitation hardened steel
Condition
Tensile Strength
Hardness
Permeability at 8 kA/m
(100 Oe)Permeability at 16 kA/m
(200 Oe)Permeability
at SaturationA
1100 MPa
(160 000 lb/in.2)35 Rc 74 48 95 H900
1450 MPa
(210 000 lb/in.2)45 Rc 90 56 135 H925
1380 MPa
(200 000 lb/in.2)43 Rc - - - H1025
1280 MPa
(185 000 lb/in.2)38 Rc - - - H1075
1210 MPa
(175 000 lb/in.2)37 Rc 88 52 136 H1150
1100 MPa
(160 000 lb/in.2)35 Rc 59 136 71
From a test standpoint, precipitation hardened steels require a higher magnetic force to achieve a desired flux density (American Society for Nondestructive Testing, 1989; Betz, 1967). While traditional calculations provide a good starting point for establishing the required test amperage, adequate field strength is usually higher then calculated and should be validated with either a digital hall effect probe or notched shims (ASTM International, 2001). The difficulty in establishing a repeatable technique is directly related to the alloy and heat treat condition. For example, the magnetic force to test a part produced from 17-4 in condition A would be less than that for conditions H1075 and H1150. When viewing Table 2, note that condition H900 has been cold worked, received subsequent heat treatments and, thus, has a higher permeability than the solution annealed (condition A) condition.
DISCONTINUITY CHARACTERIZATION
The following discontinuities may result during inherent processing of precipitation hardened steels and be detected during MT. From the ingot stage, they may travel through subsequent heat treatment, cold working and processing only to be detected during final end product testing. The discontinuities identified with MT primarily occur on the short longitudinal face. Figure 2 provides the grain directions and principal faces discussed.
Figure 2 - Grain direction and orientation (SL=short longitudinal, ST=short transverse, LT=longitudinal transverse).
Segregation or "banding" is an inherent discontinuity of the hot mill producer (Van Aken, 2002). It is principally caused by alloying elements in the material that haven't properly gone into solution. In past years, it was economically feasible to homogenize steel ingots in soaking pits, but today it is no longer practical as most steels are continuously cast and soaking pits would add cost and time to the process. Due to significant differences in permeability, this discontinuity type is very pronounced, sharp and usually runs through the center third of the part (Figure 3). Bands may wrap around the edges to the short transverse face. Segregation indications (Figure 4a) may be visually validated by using marbles etch and visually testing the area with 10x magnification. The indications will appear at 10x as gray shaded lines. Segregation indications differ from other inherent indications, such as ferrite stringers and nonmetallic inclusions, in that segregation may appear on multiple faces of the part. In addition, segregation is affected by etchants and will appear as valleys when viewed at 10´. The part shown in Figure 3 is 17-4 round stock. The indication is alloy segregation. Note the orientation, sharpness and spacing of the indications. These indications were confined to an area 51 mm (2 in.) circumferentially and ran the entire length of the part. Figure 4b is a cross sectional view of the indications. Note the pronounced band and pattern compared to Figures 3 and 4a. The band was approximately 0.03 mm (1 x 10-3 in.) deep. When this discontinuity was identified during MT, generally there were parts in the production (test) lot that were free of the discontinuity and dispositioned as acceptable. Microhardness testing of this discontinuity did not reveal significant differences between the banded and nonbanded area.
Figure 3 - Banding detected with magnetic particle testing
Figure 4 - Banding viewed at: (a) 10x following marbles etch; (b) cross section at 50x
The preceding part was tested at different ampere settings and the indications were as brilliant at half of the technique validated current value as at full current value. The parts were also fluorescent penetrant tested following proper cleaning procedures using a type I - method A - level 3 penetrant and a nonaqueous (form d) developer. The indications were not detected with fluorescent penetrant testing, but as noted above, were detected with visual testing following the marbles etch.
A subset of the preceding is microsegregation. A relatively high temperature gradient must be maintained during the melting process of the ingot to achieve a directed dendritic primary structure. The growth direction of the dendritics is a function of the metal pool profile during solidification. As pool depth increases with the melt rate, the growth of the directed dendritics can come to a stop. The ingot core then solidifies nondirectionally. With the increased dendritic arm spacing, microsegregation originates. This discontinuity is similar to segregation/banding in that it is generally limited to the center third of the part and may wrap around corners. This discontinuity is multidirectional and usually is present in dense patches. When testing at half the technique validated current value, the indications may diminish in fluorescence brilliance. Studies have shown that microsegregation has minimal effect on mechanical properties and may be considered an acceptable discontinuity.
Ferrite stringers are inherent discontinuities that are a result of ferrite not going completely into solution. This discontinuity is particularly common in 17-4 plate and bar stock (round and rectangular) due to refinement practices. The solid ferrite becomes elongated when the bar or plate is further rolled into the desired shape. These indications are principally identified in the short longitudinal face (Figure 5a) and run in the long axis. During metallography, microhardness readings were taken in the discontinuity area and in areas of stringer free material. There were no appreciable differences. Ferrite stringers are rarely detected with MT in the short or long transverse faces of the part due to directional orientation. The indications may be accompanied with banding (microsegregation) as shown in Figure 5b. The part in Figures 5a and 5b was manufactured from 17-4 bar and was supplied in heat condition H1150. Note that the indications are dense and consistent across the entire length and width of the part. The indications were as brilliant at half of the technique validated current value as at full current value. The indications were not detected with fluorescent particle testing nor were they identified during a 10x magnified visual test following marbles etch. When this discontinuity was identified during MT, generally the entire production (test) lot was rejected.
Figure 5 - 17-4PH steel: (a) manufactured from bar stock; (b) ferrite stringers with banding at 200x on the short longitudinal face.
Figure 6 is a part manufactured from 17-4 plate supplied in heat condition A. The MT indications were dense, fuzzy and concentrated in the center of the part. They were present in lesser detail across the entire length and width of the part. The indications were not quite as brilliant at half of the technique validated current value as at full current value. The parts were tested again using fluorescent penetrant testing and 10x magnified visual testing following marbles etch. No indications were observed. This may be attributed to the heat treat condition. The part underwent metallographic analysis to confirm the discontinuity type and orientation. With the exception of the concentration of the discontinuities in the center of the part, the results were similar to that of the bar stock example.
Figure 6 - Part manufactured from 17-4 plate stock
Nonmetallic inclusions are another inherent discontinuity of precipitation hardened steel. Nonmetallic inclusions are a result of deoxidizers that are added to the molten steel that were not dissolved or removed during the original cast. They may also be a result of improper homogenization. This discontinuity is common in 17-4 plate and bar stock (round and rectangular) materials. The inclusions become elongated when the bar or plate is rolled into the desired shape. These indications are principally identified in the short longitudinal face (Figure 7). During the metallographic analysis, microhardness readings were taken in the discontinuity area and in areas of inclusion free material. The appreciable differences confirmed why this discontinuity results in a reduction in fracture fatigue and should be rejected. Nonmetallic inclusions are rarely detected with MT in the short or long transverse faces of the part due to directional orientation. The indications were not as brilliant at half of the technique validated current value as at full current value. The indications were detected with fluorescent penetrant testing and were also visible at 10´ magnification. This may be attributed to the inclusions being torn or pulled out of the base material during the machining operation. When this discontinuity was detected with a test lot, the entire lot was usually rejected. Nonmetallic inclusions are usually not enhanced using etchants for visual testing.
Figure 7 - Short longitudinal view nonmetallic inclusion (200x).
CONCLUSION
The challenges of performing fluorescent magnetic particle testing on precipitation hardened stainless steels demand an informed inspector. By having a basic knowledge of the steel's history and associated discontinuities, the inspector is able to make sound classification decisions. The preceding examples may serve as an aid in this endeavor. While this study shares one processor's experience, reference photographs and reproducible techniques enhance the test organization's ability to make sound testing classifications.
ACKNOWLEDGMENTS
Special recognition goes to Chris Taylor and Adam Zellner of the Metal Finishing Company for their dedication and hard work in magnetic particle testing. Also, special recognition to Nick Roark of Arrow Laboratory for sharing his vast experience and expertise in metallurgical testing and analysis.
REFERENCE
AK Steel, AK Steel Product Data Bulletins, Armco Research Section, Technical Department, Advanced Materials Division, Middletown, Ohio, 1999.ASM International, Metals Handbook: Heat Treating, Cleaning and Finishing, eighth edition, Vol. 2, Materials Park, Ohio, ASM International, 1976.
ASTM International, ASTM E 1444-01, Standard Practice for Magnetic Particle Examination, West Conshohocken, Pennsylvania, ASTM International, 2001.
American Society for Nondestructive Testing, Nondestructive Testing Handbook, second edition: Volume 6, Magnetic Particle Testing, P. McIntire, ed., Columbus, Ohio, ASNT, 1989.
Betz, Carl, Principles of Magnetic Particle Testing, Chicago, Magnaflux Corporation, 1967.
Boyer, Howard E., Practical Heat Treating, Materials Park, Ohio, ASM International, 1984.
Department of Defense, Aerospace Structural Metals Handbook: Ferrous Alloys, West Lafayette, Indiana, Purdue, 1993.
Van Aken, Dave, "Engineering Concepts: Segregation and Banding in Steel," Industrial Heating, 2002.
References
Hirsekorn, S., S. Pangraz, W. Bernauer, G. Weides and W. Arnold, "Material Characterization and Nondestructive Testing by Acoustic Imaging Techniques," Acta Acustica, Vol. 2, 1994, pp. 195-204.Krautkramer, J. and H. Krautkramer, Ultrasonic Testing of Materials, fourth edition, New York, Springer-Verlag, 1990.
Panametrics, Inc., Ultrasonic Transducers P395, 1995.
* Metal Finishing Company, 1329 McLean Blvd., Wichita, KS 67213; (316) 267-7289; fax (316) 267-5127; e-mail <bobp@metalfinishingco.com>.
Copyright © 2003 by the American Society for Nondestructive Testing, Inc. All rights reserved.
Copyright © 2012 by the American Society for Nondestructive Testing, Inc. ASNT is not responsible for the authenticity or accuracy of information herein. Published opinions and statements do not necessarily reflect the opinion of ASNT. Products or services that are advertised or mentioned do not carry the endorsement or recommendation of ASNT.IRRSP, NDT Handbook, The NDT Technician and www.asnt.org are trademarks of the American Society for Nondestructive Testing, Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of the American Society for Nondestructive Testing, Inc. ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.
