metals Article
The Establishment of Surface Roughness as Failure Criterion of Al–Li Alloy Stretch-Forming Process Jing-Wen Feng 1,2,3 , Li-Hua Zhan 1,2,3, * and Ying-Ge Yang 1,2,3 Received: 17 December 2015; Accepted: 28 December 2015; Published: 7 January 2016 Academic Editor: Nong Gao 1 2 3
*
School of Mechanical and Electrical Engineering, Central South University, Changsha 410083, Hunan, China;
[email protected] (J.-W.F.);
[email protected] (Y.-G.Y.) State Key Laboratory of High Performance Complex Manufacturing, Central South University, Changsha 410083, Hunan, China 2011 Collaborative Innovation Center, Central South University, Changsha 410083, Hunan, China Correspondence:
[email protected]; Tel.: +86-731-8883-0254
Abstract: Taking Al–Li–S4–T8 Al–Li alloy as the study object, based on the stretching and deforming characteristics of sheet metals, this paper proposes a new approach of critical orange peel state characterizations on the basis of the precise measurement of stretch-forming surface roughness and establishes the critical criterion for the occurrence of orange peel surface defects in the stretch-forming process of Al–Li alloy sheet metals. Stretching experiments of different strain paths are conducted on the specimens with different notches so as to establish the Al–Li–S4–T8 Al–Li alloy, forming limit diagram and forming limit curve equation, with the surface roughness of characteristic critical orange peel structure as the stretch-forming failure criterion. Keywords: Al–Li–S4 Al–Li alloy; stretch-forming; orange peel; forming limit; surface roughness
1. Introduction Aircraft skin serves as the shape part of an aircraft and constructs the aerodynamic configuration of the aircraft, featuring big size and direct contact with air. Therefore, it requires an accurate shape, smooth streamline, and no surface defects, etc. [1]. As a relatively common aircraft skin-forming approach in the field of aeronautics and astronautics manufacturing, the technology of stretch-forming is widely used in the manufacturing of large-scale aircraft skin as part of aircraft aerodynamic configuration [2–4]. In the stretch-forming process, the clamps of the stretch-forming machine clamp both ends of the sheet metal and move along a certain track, or the die goes up to make the sheet metal contact the stretch-forming die, creating uneven plane stretching strain to make the metal sheet conform to the stretch-forming die so as to obtain the required part shape [5–7]. Compared with other approaches to aircraft skin forming, the stretch-forming technology might cause surface defects such as orange peel. Orange peel not only affects the appearance of the aircraft skin, but also damages its surface integrity. Especially in the case of mirror skin, orange peel easily appears due to mirror skin’s internal structure and polished surface, seriously affecting its service life, which is usually the main reason for the scrapping of parts [8]. The defect of orange peel is a kind of rough orange peel-like morphology found on the surface of shaped products. In general, coarse and unevenly structured grains on the alloy surface are considered the reason of the orange peel defect appearing on the alloy surface during the stretch-forming process, while the grain size of the alloy has a certain relation to the extent of the deformation. At a certain temperature, when there are relatively small deformations, recrystallization usually does not appear in the alloy and the grains maintain their original state; however, when deformation reaches a certain degree, recrystallized grains will become very coarse. In the manufacturing process of aluminum alloy Metals 2016, 6, 13; doi:10.3390/met6010013
www.mdpi.com/journal/metals
Metals 2016, 6, 13
2 of 10
skin sheet metal, there are usually multiple hot rolling and cold rolling processes as well as several heat treatments including annealing, solution and aging. Because of the imperfection in the control over cold deformation and the choice of heat treatment technology in the manufacturing process of aluminum alloy sheet metal, recrystallized grain structure tends to be coarse and show different sizes, resulting in the piling up of dislocation on large grains and the rapid increases of stress in the follow-up stretch-forming process. Thus, the areas of large grains reach and exceed the elastic limit in advance; the non-synchronous deformation between large grains and small ones gives rise to minor cracks on the surface of the material, manifesting as the orange peel structure at the macro-level [9]. The forming limit of sheet metal is a criterion used to describe whether the sheet metal fails to form or not. To identify the concept of forming limit, one should first determine the failure criterion of sheet metal’s forming process. Al–Li alloy as a new type of aluminum alloy has shown a wide application prospect in the fields of aeronautics and astronautics due to its good qualities of low density, high specific strength, and high specific stiffness, and it has become a hot subject in the research field of aluminum alloy materials, being regarded as one of the important candidate materials of modern aeronautic and astronautic structures [10,11]. However, due to the high cost, poor cold plasticity at ambient temperature, evident anisotropy and easy cracking in cold working compared with other regular aluminum alloys, Al–Li alloy at present can only be processed into relatively simple parts and faces great difficulty in the processing and manufacturing of more complex structural parts. Therefore, some of its own attributes also limit Al–Li alloy’s application in structural components [12–15]. Relevant researchers have conducted corresponding research on the formability of Al–Li alloy, which, however, mainly focus on the study of the Al–Li alloy’s structure property and the aspect of hot forming. Literature [16] explored the 2397-T87 Al–Li alloy with a thickness of 130 mm for the microstructure, stretching property and fracture toughness of layers with different thicknesses and at different orientations. By the uniaxial tension test under different hot deformation conditions, the forming limit test with cracking as a failure criterion, and the stretch-forming tests of 5A90 Al–Li alloy sheet metal, literature [17] built the Al–Li alloy forming limit model and determined the technological parameters for Al–Li alloy to acquire good deformation performance. Nevertheless, for aviation aircraft skin materials, cracking is not often taken as the criterion for judging whether the skin fails to form or not, especially for Al–Li alloy, the reason of which usually lies in that the forming failure is caused by the appearance of orange peel structure in the stretch-forming process. Currently, there are relatively few studies on the forming limit caused by the defect of orange peel in the stretch-forming process, and a forming failure criterion targeting the orange peel has not yet taken shape. As the occurrence of the orange peel phenomenon is a slowly changing and accumulated process, and all of the previous research on such a phenomenon was conducted by eye measurement, the results have certain randomness. In this paper, experimental research on the problem of orange peel in the stretch-forming process of Al–Li alloy is conducted, and a novel approach of critical orange peel characterization is proposed on the basis of the precise measurement of stretch-forming surface roughness. Furthermore, the judgment criterion of the orange peel defect is analyzed and established, the stretching tests on specimens of different strain paths are conducted combining the technology of optical deformation measurement, and, ultimately, the stretch-forming limit diagram and its forming limit curve equation are obtained with the surface roughness of orange peel structure with critical characteristics such as the stretch-forming failure criterion. 2. Experiments 2.1. Instrument and Methods In order to conduct synchronization tests on stress-strain and orange peel surface defect in the stretch-forming process of Al–Li alloy, a stretch-forming test and testing system are established to obtain the critical strain state of orange peel of the product, in which the optical deformation
Metals 2016, 6, 13
3 of 10
Metals 2016, 6, 13 Metals 2016, 6, 13
3 of 10 3 of 10
measurement instrument and the universal testing machine operate in collaboration, as shown in Figure 1. measurement instrument and the universal testing machine operate in collaboration, as shown in measurement instrument and the universal testing machine operate in collaboration, as shown in The Figure 1. optical deformation measurement instrument used in the experiment adopts the Figure 1. deformation measurement system based on computer visionused technology by the German The optical deformation measurement instrument in thedeveloped experiment adopts the The optical deformation measurementdeformation instrument used in the experiment adopts thesynchronized deformation company GOM for three-dimensional analysis. By controlling the deformation measurement system based on computer vision technology developed by the German measurement system based on computer vision technology developed the German company GOM operation of the optical deformation measurement instrument andBy theby universal testing machine, the company GOM for three-dimensional deformation analysis. controlling the synchronized for three-dimensional deformation analysis. By controlling the synchronized operation of the optical complete of monitoring the total stretch-forming deformation of thetesting specimens can the be operation the opticalofdeformation measurement instrument andprocess the universal machine, deformation measurement instrument and thesystem universal testing machine, the complete monitoring achieved and the computer image processing can be further utilized to obtain the true strain complete monitoring of the total stretch-forming deformation process of the specimens can be of the total stretch-forming deformation of the specimens can bethe achieved and the computer change rules different positions theprocess specimen surface process. achieved andat the computer image on processing system can bethroughout further utilizedstretch-forming to obtain the true strain image processing system can be further utilized to obtain the true strain change rules at different change rules at different positions on the specimen surface throughout the stretch-forming process. positions on the specimen surface throughout the stretch-forming process.
Figure 1. The stretching and testing system. Figure Figure 1. 1. The The stretching stretching and and testing testing system. system.
Compared with the stretch-forming mechanism, this experimental mechanism is to some extent simplified, in particular the top die removed. However, in the actual stretch-forming process of Compared with stretch-forming mechanism, this mechanism is extent Compared with the thewith stretch-forming mechanism, this experimental experimental mechanism is to to some some extent the sheet metal, with the stretch-forming die going up, the sheet metal gradually bends to attach to simplified, simplified, in in particular particular with with the the top top die removed. removed. However, However, in in the the actual stretch-forming stretch-forming process process of of the sheet die as shown inthe Figure 2. Due to the existence of the friction, the material flow oftothe the metal, with die going up,up, the sheet metal gradually bends attach to the the sheet metal, with thestretch-forming stretch-forming die going sheet metal gradually bends to attaching attach to segment limited relatively small deformation, whereas theoffree die as intends Figure 2.beDue theto existence of friction, the material flow of the attaching segment the dieshown asAA′ shown in to Figure 2.toDue to athe existence of friction, the material flow the segments attaching 1 without contact with the die AB and A′B′ tend to have relatively greater deformation without the AA tendsAA′ to betends limited relatively deformation, the free segments contact segment to to bea limited tosmall a relatively small whereas deformation, whereas thewithout free segments restriction ofAB friction. the peelgreater structure usually appears the free segments with the die and A1Therefore, B1 tend relatively deformation withoutfirst the on restriction of friction. without contact with the die to ABhave andorange A′B′ tend to have relatively greater deformation without the between the chucks and the die corners, and then slowly spreads toward the forming surface that Therefore, orange peel structure first on usually the free appears segmentsfirst between chucks and restrictionthe of friction. Therefore, theusually orangeappears peel structure on thethe free segments attaches to die. it can seen that thethen deformation of the free onthe thedie. sheet metal the die corners, andThus, then spreads toward theslowly forming surface thatsegments attaches to Thus, it between thethe chucks and slowly the diebe corners, and spreads toward the forming surface that is the principal factor affecting the appearance of orange peel structure. As the deformation of the attaches to the Thus, it can be seenfree that the deformation of the freeissegments on the sheet metal can be seen thatdie. the deformation of the segments on the sheet metal the principal factor affecting free segments the sheet metal is appearance similar to its the conventional stretching test of is the principalon factor affecting the ofstretch-forming, orange peel structure. As theon deformation of the the appearance of orange peel structure. As the deformation of the free segments the sheet metal is sheetsegments metal can the conventional research the influence the the orange peelbephenomenon on free onbe theused sheetinmetal is similaron tostretching its stretch-forming, conventional stretching testthe of similar to its stretch-forming, the test of of sheet metal can used in the research stretch-forming materials. sheet can of bethe used in the research on theon influence of the orange peelmaterials. phenomenon on the on the metal influence of the orange peel phenomenon the stretch-forming of the stretch-forming of the materials.
Figure 2. The schematic diagram of stretch-forming die. Figure 2. The schematic diagram of stretch-forming die.
Figure 2. The schematic diagram of stretch-forming die.
Metals 2016, 6, 13
4 of 10
Metals 2016, 6, 13
4 of 10
2.2. Experimental Design
2.2. Experimental Design
2.2.1. Experiment Design of Critical State Criterion of Orange Peel Defect 2.2.1. Experiment Design of Critical State Criterion of Orange Peel Defect
The experimental material was the Al–Li–S4–T8 Al–Li alloy which was used for aircraft skin The experimental material was the Al–Li–S4–T8 Al–Li alloy which was used for aircraft skin with awith thickness of 2 mm. See Figure 3 for the structure and size of the stretched specimens. The a thickness of 2 mm. See Figure 3 for the structure and size of the stretched specimens. The experimental specimens were polished to to make brightand andwithout without obvious scratches. experimental specimens were polished makethe thesurfaces surfaces bright obvious scratches. Meanwhile, in order to establish the same statefor forallall specimens, roughness Meanwhile, in order to establish the sameinitial initial surface surface state specimens, the the roughness of polished specimens waswas measured onon the profilometer ensure similar polishing of polished specimens measured theoptical opticalsurface surface profilometer to to ensure similar polishing effects for all the specimens. effects for all the specimens.
Figure 3. The shapeand andsize size of of stretched stretched specimen. Figure 3. The shape specimen.
Stretching tests of different strains were conducted on polished specimens to investigate the
Stretching tests of different strains were conducted on polished specimens to investigate the evolution situation of the orange peel phenomenon on material surfaces under the condition of evolution situation of the orange peel phenomenon material surfaces under condition different deformations. See stretching strains in Table on 1. Additionally, tests were also the conducted at of different stretching strains in Table the 1. Additionally, tests deforming were alsospeeds conducted fourdeformations. different strain See speeds on T8 alloy to investigate impacts of different on at four different strain on T8on alloy to investigate thethe impacts of different deforming speeds on the orange peel speeds phenomenon specimen surfaces in experiment. By contrasting the surface morphology and true strainon of the specimen at different speeds, when tiny By orange peel phenomena the orange peel phenomenon specimen surfaces in the experiment. contrasting the surface appear on the surface at different strain speeds, the true strain of the Al–Li–S4–T8 Al–Li alloy metal morphology and true strain of the specimen at different speeds, when tiny orange peel phenomena sheet can be measured with the results shown in Table 2. It is observed from the table that when themetal appear on the surface at different strain speeds, the true strain of the Al–Li–S4–T8 Al–Li alloy strain speed is 0.0005/s, the strain at the sampling point is the greatest when tiny orange peel sheet can be measured with the results shown in Table 2. It is observed from the table that when the structure on the specimen surface appears. Thus, in actual stretch-forming treatment of aircraft skin, strain speed is 0.0005/s, the strain at the sampling point is the greatest when tiny orange peel structure the said speed can be employed to alleviate the appearance of the orange peel phenomenon. on the specimen surface appears. Thus, in actual stretch-forming treatment of aircraft skin, the said speed can be employed to alleviate the appearance the of orange peel phenomenon. Table 1. Stretching of strains samples. Experiment Batch No. 1 2 3 4 Table 1. Stretching strains of samples. Stretching Strain 0% 3% 6% 9%
5 12%
Experiment Batch No. 1 2 3 4 5 Table 2. Critical strain and roughness of samples. Stretching Strain 0% 3% 6% 9% 12%
Specimen No. Strain Speed/s−1 Critical Strain/% Roughness Ra/nm 1 0.0001 strain and roughness 1.78 of samples. 841 Table 2. Critical 2 0.0005 1.87 827 3 0.001 1.14 973 ´1 Specimen No. Critical Strain/% Roughness Ra /nm Strain Speed/s 4 0.0015 1.19 1002
1 0.0001 1.78 841 2 0.0005 1.87 827 After the stretching tests, both sides 3 0.001 of the specimens 1.14were cleaned again, 973 and the side used for 0.0015 1.19optical profilometer 1002 for surface analysis, surface morphology4 observation was placed under the
obtaining the morphology, nephogram and roughness of the specimen surfaces. Combined with the criterion of the orange peel defect obtained by the experiment, the areas in which critical orange peel After the stretching tests, both sides of the specimens were cleaned again, and the side used structure occurred were found on the specimens, and the strain state of the areas can be also found at
for surface morphology observation was placed under the optical profilometer for surface analysis, obtaining the morphology, nephogram and roughness of the specimen surfaces. Combined with the criterion of the orange peel defect obtained by the experiment, the areas in which critical orange peel structure occurred were found on the specimens, and the strain state of the areas can be also found at
Metals 2016, 6, 13 Metals 2016, 6, 13 Metals 2016, 6, 13
5 of 10 5 of 10 5 of 10
same positionsononthe theother otherside sideof ofthe the specimens, specimens, which state of of thethe thethe same positions whichactually actuallyisisthe thecritical criticalstrain strain state the same positions on the other side of the specimens, which actually is the critical strain state of the appearing orange peel defect. appearing orange peel defect. appearing orange peel defect. 2.2.2. ExperimentDesign DesignofofStretch StretchForming Forming Limit Limit Diagram Diagram 2.2.2. Experiment 2.2.2. Experiment Design of Stretch Forming Limit Diagram During thestretch-forming stretch-formingprocess process of of aircraft aircraft skin, mainly fellfell During the skin, the the strain strainstate stateofofsheet sheetmetals metals mainly During the stretch-forming process of aircraft skin, the strain state of sheet metals mainly fell betweenuniaxial uniaxialstretching stretchingand andplane plane strain strain with with approximately approximately linear between linearstrain strainpaths paths[18]. [18].Thus, Thus, between uniaxial stretching and plane strain with approximately linear strain paths [18]. Thus, stretch-forming specimens that could implement different strain paths were designed with the stretch-forming specimens that could implement different strain paths were designed with the typical stretch-forming that size couldgiven implement different strain paths designed stretched with the typical specimenspecimens structure and Figure 4, among which thewere straight-edge specimen structure and size given in Figure 4,in among which the straight-edge stretched specimen was typical specimen structure and size given in Figure 4, among which the straight-edge stretched specimen was close to the uniaxial stretching state of strain, while the R = 10 notched specimen was close to the uniaxial stretching state ofstretching strain, while R = 10 while notched was close to thewas plane specimen close to the uniaxial statethe of strain, thespecimen R = 10 notched specimen close to thewas plane state of strain [19]. state of strain [19]. close to the plane state of strain [19].
Figure4.4.Size Size of of samples samples with Figure with different differentstrain strainpaths. paths. Figure 4. Size of samples with different strain paths.
In order to conduct observation and analysis of the surface morphology and strain state of the In In order totoconduct observation and and analysisofof surface morphology and strain state of orderboth conduct thethe surface morphology and strain state and of the specimens, sides ofobservation the specimens analysis were polished until without an obvious scratch, a thespecimens, specimens,both bothsides sidesofofthe thespecimens specimens were polished until without an obvious scratch, and a were polishedtountil without anpolishing obvious effects scratch,forand profilometer was used to measure the surface roughness achieve similar eacha profilometer was used the roughness to achieve achievesimilar similarpolishing polishing effects each profilometer was usedtotomeasure measure thesurface surface roughness effects forfor each specimen. The specimens with ethanol were cleaned aftertopolishing. As shown in Figure 5, the black specimen. The specimens with ethanol were cleaned after polishing. As shown in Figure 5, the black specimen. The specimens with ethanol were cleaned after polishing. As shown in Figure 5, the black and white speckle patterns were sprayed on half of one side of the specimens, so as to analyze the and white speckle patterns were on half of one one sideof of thespecimens, specimens, soasasto analyze and white speckle patterns weresprayed sprayed onstretching side the analyze the true strain on specimen surfaces during the process; another half was so used toto observe thethe true strainononof specimen surfaces during process; another halfhalf was used to observe the true strain specimen surfaces duringthe thestretching stretching process; another used to observe evolvement the orange peel phenomenon on specimen surfaces during the was stretching process. evolvement ofofthe peel phenomenon on specimen surfaces during the stretching process. theMeanwhile, evolvementanother theorange orange peel phenomenon on specimen surfaces during the stretching process. side, only polished, was used to measure surface morphology and roughness Meanwhile, anotherside, side,only onlypolished, polished, was was used used to roughness Meanwhile, another to measure measure surface surfacemorphology morphologyand and roughness with the profilometer. with the profilometer. with the profilometer.
Figure 5. Sample appearances along different strain paths after surface treatment. Figure strainpaths pathsafter aftersurface surfacetreatment. treatment. Figure5.5.Sample Sampleappearances appearances along along different strain
Metals 2016, 6, 13 Metals 2016, 6, 13 Metals 2016, 6, 13
6 of 10 6 of 10 10 6 of
3. Results and Analysis 3. Results and Analysis 3. 3.1. Results and Analysis The Equilibrium Diagram of Tensile Specimen before Deformation 3.1. The Equilibrium Diagram of Tensile Specimen before Deformation grain sizeDiagram grade of specimen canbefore be measured in accordance with the Metal Methods 3.1. TheThe Equilibrium of the Tensile Specimen Deformation The grain size grade of the specimen can be measured in accordance with the Metal Methods for Estimating the Average Grain Size, as shown in Table 3 below. From the measured grain size grain size of the specimen can be measured accordance the Metal Methods forThe Estimating thegrade Average Grain Size, as shown in Table in 3 below. Fromwith the measured grain sizefor grades, as shown in Figures 6 and 7, it is observed that the average grain size of the specimen is grades, asthe shown in Figures 6 andas7,shown it is observed the average grain size of the specimen is Estimating Average Grain Size, in Tablethat 3 below. From the measured grain size grades, 60–80 μm, suggesting that the specimen has already fallen into the category of open grain structure. μm, suggesting that7the specimen hasthat already fallen into the category of specimen open grainisstructure. as 60–80 shown in Figures 6 and it is observed the average grain size of the 60–80 µm, Meanwhile, judging from the equilibrium diagram, the grain structure of the specimen is extremely Meanwhile, judging from the equilibrium diagram, the grain structure of the specimen is extremely suggesting thatathe specimen hasinalready fallen the intolarge the category ofsmall opengrains. grain structure. Meanwhile, uneven with large difference size between grains and It is also discovered uneven withthe a large difference in size between thestructure large grains andspecimen small grains. It is also discovered judging from equilibrium diagram, the grain of the is extremely uneven from the measurement that the size of the large grains is over 100 μm, but that of the small oneswith is from the measurement that the size the large grains issmall over grains. 100 μm,Itbut the small ones isthe a large difference inμm. sizeThus, between theoflarge grains andcan is that also of discovered fromin merely around 10 the equilibrium diagram explain the appearance of orange peel merely around 10 μm. Thus, the equilibrium diagram can explain the appearance of orange peel in measurement the process. that the size of the large grains is over 100 µm, but that of the small ones is merely the process. around 10 µm. Thus, the equilibrium diagram can explain the appearance of orange peel in the process.
Figure 6. The exact exact location on the sample. Figure onthe thesample. sample. Figure6.6.The The exact location location on （2） （2）
（1） （1）
（5） （5）
（4） （4）
（3） （3）
（6） （6）
（7） （7）
Figure 7. The equilibrium diagram of tensile specimen before deformation. (1)(1) is correspondence with Figure 7. The equilibrium diagram of tensile specimen before deformation. is correspondence 7.6;The equilibrium diagramwith of tensile specimen before deformation. with (1) is3correspondence 1 inFigure Figure (2) is correspondence 2 in Figure 6; (3) is correspondence in Figure 6; (4) with 1 in Figure 6; (2) is correspondence with 2 in Figure 6; (3) is correspondence with 3 in Figure 6; is with 1 in Figure 6; (2) is Figure correspondence with 2 in Figurewith 6; (3)5 is with 3 in Figure with 6; correspondence with 4 in (5)Figure is correspondence incorrespondence Figurewith 6; (6)5isincorrespondence (4) is correspondence with 46;in 6; (5) is correspondence Figure 6; (6) is (4) is correspondence with 4 in Figure 6; (5) is correspondence with 5 in Figure 6; (6) is 6 incorrespondence Figure 6; (7) iswith correspondence in Figure 6. 6 in Figure 6;with (7) is7correspondence with 7 in Figure 6. correspondence with 6 in Figure 6; (7) is correspondence with 7 in Figure 6. Table tensile specimen. specimen. Table3.3.Grain Grainsize size analysis analysis of of tensile Table 3. Grain size analysis of tensile specimen. Heat Treatment Condition
Heat Heat Treatment Treatment Condition Condition T8 T8
T8
Type of Analysis Type Analysis Type of of Analysis Grain Size Grade Grain Size Grade m Average Diameter/µ Grain Size Grade Average Diameter/µ m
Average Diameter/µm
Sampling Point Sampling Point 1 2 3 Sampling 4 Point5 1 1 2 2 3 3 4 4 5 5.02 5.13 4.51 4.38 4.985 5.02 62 5.02 5.13 59 5.13 4.51 76 4.51 4.38 77 4.38 4.98 704.98 62 59 76 77 70
62
59
76
77
70
6 66 4.62 4.62 75 4.62 75
75
7 77 4.56 4.56 75 4.56 75
75
Average Average Average 4.7 4.7 71 4.7 71
71
3.2. The Establishment of the Criterion for Critical State of Orange Peel Defect 3.2. The Establishment of the Criterion for Critical State of Orange Peel Defect 3.2. The The Establishment the Criterion for Critical of Orange Defectand surface observation and specimensofwere cleaned after stretchState processing withPeel ethanol, The specimens were cleaned after stretch processing with ethanol, and surface observation and analysis on the optical profilometer were conducted. The photographed morphology can be seen in The specimens wereprofilometer cleaned after stretch processing with ethanol, and surface observation and analysis on the optical were conducted. The photographed morphology can be seen in Figure 8, which shows that when the strain is 3%, the specimen surface starts to grow rough and analysis the optical were conducted. photographed morphology can beand seen Figure on 8, which showsprofilometer that when the strain is 3%, theThe specimen surface starts to grow rough form a kind of evenly frosted surface morphology, but without evident minor cracks or significant in form Figure 8, which shows that when the strain is 3%, the specimen surface starts to grow rough and a kind of evenly frosted surface morphology, but without evident minor cracks or significant form a kind of evenly frosted surface morphology, but without evident minor cracks or significant orange peel phenomenon; when the strain reaches 6%, light black strip areas appear on the surface,
Metals 2016, Metals 2016, 6, 6, 1313
77 ofof 1010
orange peel phenomenon; when the strain reaches 6%, light black strip areas appear on the surface, which areare shallow grooves, andand an embryonic form form of orange peel structure can be roughly observed. which shallow grooves, an embryonic of orange peel structure can be roughly observed. When the strain reaches areappearing clear cracks on the specimen When the strain reaches 9%, there are9%, clearthere cracks on appearing the specimen surface, whichsurface, is rather which is rather severe despite the fact that they are relatively decentralized and independent; at the a severe despite the fact that they are relatively decentralized and independent; at the macro-level, macro-level, a very significant orange peel is phenomenon manifested. this stage, peel the orange peelon very significant orange peel phenomenon manifested.isAt this stage,At the orange structure the material surface affects the appearance of the stretch-forming parts, but thestructure materialon surface not only affectsnot theonly appearance of the stretch-forming parts, but also exerts great also exerts great influence on the performance and service life of the specimen, whereas the minor influence on the performance and service life of the specimen, whereas the minor cracks on the surface cracks therise surface can easily give riseand to stress concentration andgreater gradually evolve greater can easilyongive to stress concentration gradually evolve into cracks afterinto stress. When cracks after stress. When the strain amounts to 12%, the surface morphology of the specimen further the strain amounts to 12%, the surface morphology of the specimen further deteriorates with more deteriorates with more and deeperformed cracks, gully-like having already formed gully-like shapes. and deeper cracks, having already shapes. （a） （a）
(b) (b)
(c) (c)
50μm 50μm
50μm 50μm
50μm 50μm
(d) (d)
(e) (e)
50μm 50μm
50μm 50μm
Figure 8. 8.Surface Figure Surfacemorphology morphologyofofstretched stretchedspecimens specimenswith with different different strain 0%; (b) (b)3%; 3%;(c) strain variables: variables: (a) 0%; 6%; (d) 9%; (e) 12%. (c) 6%; (d) 9%; (e) 12%.
canbebeseen seenfrom fromthe theanalysis analysisresults results that that when when the It It can the strain strain reaches reachesaround around6%, 6%,the thespecimen specimen surface starts to form the orange peel defect in a real sense, but as the gap between the strain surface starts to form the orange peel defect in a real sense, but as the gap between the strain variables variables is relatively large, it is impossible to determine that the specimen is in the critical orange is relatively large, it is impossible to determine that the specimen is in the critical orange peel state peelthe state when the previous basis of the previous research, strain of were 5% when strain is the 6%.strain Thus,ison6%. theThus, basison of the research, strain variables ofvariables 5% and 7% and 7% a new their test to observe their surface morphology with thein added to awere newadded test to to observe surface morphology after stretching,after withstretching, the results given results given in Figure 9. When the strain was 5%, the specimen surface had not yet formed clear Figure 9. When the strain was 5%, the specimen surface had not yet formed clear grooves, but through grooves, but through comparing the stretched specimens of 6% and 7%, it could be found that, comparing the stretched specimens of 6% and 7%, it could be found that, though with roughly the though with roughly the same surface morphology, the scanning image of the stretched specimen same surface morphology, the scanning image of the stretched specimen with 7% strain clearly showed with 7% strain clearly showed darker grooves with relatively deep cracks starting to develop. darker grooves with relatively deep cracks starting to develop. （a） （a）
（b） （b）
50μm 50μm
（c） （c）
50μm 50μm
50μm 50μm
Figure Surface morphology of stretched specimens different variables: Figure 9. 9. Surface morphology of stretched specimens with with different strain strain variables: (a) 5%;(a) (b)5%; 6%;(b)(c)6%; 7%. (c) 7%.
On this basis, the roughness of specimen surfaces with different strain variables is measured, theOn results which as shownofinspecimen Figure 10,surfaces where Rwith aa is the arithmetic mean of the is absolute value this of basis, theare roughness different strain variables measured, the results of which are as shown in Figure 10, where Ra is the arithmetic mean of the absolute value of
Metals 2016, 6, 13 Metals 2016, 6, 13
8 of 10 8 of 10
of the distance between the dot theprofile profileand andthe thebaseline; baseline;RR q is the root mean square error of the the distance between the dot onon the q is the root mean square error of the profile, i.e., i.e., the theroot rootmean meansquare square value profile’s offset distance within the sampling profile, value of of thethe profile’s offset distance within the sampling range;range; Rz is R z is the micro-irregularity on the material surface, i.e., the sum of the average of the five greatest the micro-irregularity on the material surface, i.e., the sum of the average of the five greatest profile profileand peaks and theofaverage the five smallest profile withinlength. the sampling length. peaks the average the five of smallest profile valleys withinvalleys the sampling Combined with Combined the surface morphology by the profilometer, canatbeen at the the surface with morphology measured by themeasured profilometer, it can been seen it that, the seen stagethat, of 0%~5% stage of strain,was the that main was that the specimen surface to wrinkle, strain, the0%~5% main change thechange specimen surface began to wrinkle, withbegan dispersed convexeswith and dispersedappearing convexes as and concaves appearing assurface well asroughness; a rapid change surface roughness; the concaves well as a rapid change of at the of stage of 5%~6% strain,atthere stage of 5%~6% strain, there was little change in the surface roughness, but dispersed convexes and was little change in the surface roughness, but dispersed convexes and concaves started to gather to concaves started gather to form lumps while grooves appeared and started into cracks, form lumps whiletogrooves appeared and started to develop into cracks, whichto is develop also the major stage which is also the major stage of on qualitative change appearing theofmaterial at the stage of of qualitative change appearing the material surface; at the on stage 6%~12%surface; strain, cracks rapidly 6%~12% strain, cracks rapidly developed, gradually grew deeper, and joined with each other to form developed, gradually grew deeper, and joined with each other to form the gully-like morphology, the gully-like morphology, resulting thespecimens. rapid increase in Rz of the specimens. resulting in the rapid increase in Rz ofinthe 10 9
Ra
8
Rq
7
Rz
roughness/μm
6 5 4 3 2 1 0 -1 0
3
5
6
7
9
12
strain/%
Figure 10. Changing trend of stretched sample surface roughness.
Based on on the the above above experimental experimental analyses, be assumed assumed that that when when aa morphology morphology similar similar Based analyses, it it can can be to that that of of the the stretched stretched specimen specimen with with 6% 6% strain strain appears appears on on the the specimen specimen surface, surface, the the orange orange peel peel to phenomenon comes to a critical state. Thus, a roughness of R a = 700 ± 50 nm and Rz = 5.5 ± 0.5 μm can phenomenon comes to a critical state. Thus, a roughness of Ra = 700 ˘ 50 nm and Rz = 5.5 ˘ 0.5 µm can be taken taken as reference standard standard for for judging judging whether whether the the orange orange peel peel structure structure on on aa material material surface surface be as aa reference reaches aa critical critical state. state. reaches 3.3. The Establishment of Stretch-Forming Limit Diagram
the basis basisofofobtaining obtainingthe thecorresponding corresponding strain state surface roughness of orange the orange On the strain state andand surface roughness of the peel peel structure on the standard specimens, the research on the stretch-forming limit under different structure on the standard specimens, the research on the stretch-forming limit under different strain strain is paths is further conducted, whichdata strain data of different strain paths is measured paths further conducted, throughthrough which strain of different strain paths is measured with the with theroughness surface roughness of the characteristic critical peelasstructure as the forming failure surface of the characteristic critical orange peelorange structure the forming failure criterion, as criterion, shown shown in as Table 4. in Table 4. Table 4. Strain state state of of critical critical orange orange peel peel structure Table 4. Strain structure of of specimens specimens under under different different strain strain paths. paths.
Acquisition Acquisition Acquisition Acquisition Point 1 1 Point Point Point2 2 ε /% ε /% ε /% /% 1 2 1 ε1/% ε2/% ε1/% εε22/% Straight ´1.94 6.35 6.35 −1.98 ´1.98 Straight edge edge6.29 6.29−1.94 R = 10 3.17 ´0.21 3.17 ´0.18 R = 10 3.17 −0.21 3.17 −0.18 R = 15 2.95 ´0.19 3.04 ´0.13 R = 15 2.95 −0.19 3.04 −0.13 R = 30 3.31 ´0.55 3.76 ´0.58 R = 30R = 40 3.31 4.49−0.55 ´1.20 3.76 4.82 −0.58 ´1.59 R = 40 4.49 −1.20 4.82 −1.59 Sample Sample
Acquisition Acquisition AcquisitionAcquisition Acquisition Acquisition Point3 3 Point 4 5 Point Point 4 Point Point 5 εε11/% ε /% ε /% ε /% ε /% ε /% 2 1 2 1 2 /% ε2/% ε1/% ε2/% ε1/% ε2/% 6.31 6.31 2.65 2.65 3.27 3.27 3.64 3.64 4.66
´2.08 −2.08 ´0.15 −0.15 ´0.12 −0.12 ´0.77 −0.77 ´1.27
4.66
−1.27
6.256.25´1.84−1.84 6.33 2.762.76´0.13−0.13 3.17 3.31 ´0.23 3.09 3.31 −0.23 3.72 ´0.57 3.30 3.72 3.92 ´1.04−0.57 4.20
3.92
−1.04
´1.88 −1.88 6.33 ´0.15 −0.15 3.17 ´0.08 3.09 −0.08 ´0.55 3.30 ´1.23 −0.55 4.20 −1.23
Metals 2016, 6, 13
9 of 10
Metals 2016, 6, 13
9 of 10
Quadratic-multinomial fitting is conducted on measured data so as to establish the forming limit Quadratic-multinomial fitting is conducted on measured data so as to establish the forming diagram of the orange peel phenomenon on the aluminum alloy surface with the results shown in limit diagram of the orange peel phenomenon on the aluminum alloy surface with the results shown Figure 11 and the quadratic-multinomial fitting results given in Table 5. It can be seen from the figure in Figure 11 and the quadratic-multinomial fitting results given in Table 5. It can be seen from the that the orange peel structure on the aluminum alloy surface is correlated with the strain state of figure that the orange peel structure on the aluminum alloy surface is correlated with the strain state the materials; meanwhile, the materials at the plane state of strain more easily develop orange peel of the materials; meanwhile, the materials at the plane state of strain more easily develop orange structure than those at the uniaxial stretching strain state. peel structure than those at the uniaxial stretching strain state. Table 5. Binomial fitting results. Table 5. Binomial fitting results. Expression Expression yy==A + Cx A ++ Bx Bx + Cx22
AA
BB
CC
2.99 2.99
−0.389 ´0.389
0.652 0.652
7.0 6.5 6.0 5.5 5.0
e1/%
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -2.5
Straight sided specimens R=40 Notched specimens R=30 Notched specimens R=15 Notched specimens R=10 Notched specimens fitting curve -2.0
-1.5
-1.0
-0.5
0.0
e2/%
Figure Figure 11. 11. Stretch-forming Stretch-forming limit limit diagram diagram of of Al–Li Al–Li alloy alloy at at the the condition condition of of Al–Li–S4–T8. Al–Li–S4–T8.
4. Conclusions 4. Conclusions (1) (1) A stretch-forming experiment and testing system with with the the optical optical deformation deformation measurement measurement instrument and the universal testing machine operating in collaboration are constructed, collaboration are constructed, the the rule of stretched specimens with different strain variables is analyzed, surface morphology morphologychange change rule of stretched specimens with different strain variables is and the corresponding relation between critical orange peel defect surface roughness of analyzed, and the corresponding relation between critical orangeand peelthe defect and the surface specimens is It is discovered when critical orange peel defectorange appears on defect Al–Li roughness ofobtained. specimens obtained. Itthat is discovered that when critical peel alloy sheet the condition of Al–Li–S4–T8, the surface roughnessthe is Rsurface 50 nm and appears onmetal Al–Liatalloy sheet metal at the condition of Al–Li–S4–T8, roughness is a = 700 ˘ Rza ==700 5.5 ˘ 0.5nm µm. ± 50 and Rz = 5.5 ± 0.5 μm. (2) By processing different notched specimens, the stretch-forming limit tests with with different different strain strain (2) forming limit limit diagram diagram and and forming forming limit limit curve curve equation equation paths are conducted to obtain the forming ε11 == 2.99 roughness of of 2.99 −´ 0.389ε 0.389ε22 ++ 0.652ε 0.652ε222 2for for Al–Li–S4–T8 Al–Li–S4–T8 Al–Li Al–Li alloy, alloy, with the surface roughness characteristic critical orange peel structure as the forming failure failure criterion. criterion. Acknowledgments: This research was financially supported by the National Basic Research Program of China Acknowledgments: This research was financially supported by the National Basic Research Program of China (2014CB046602) and the National Natural Science Foundation of China (51235010). (2014CB046602) and the National Natural Science Foundation of China (51235010). Author Contributions: Contributions:Jing-Wen Jing-Wen and Li-Hua Zhan conceived and theJing-Wen experiments; Author FengFeng and Li-Hua Zhan conceived and designed thedesigned experiments; Feng and Ying-GeFeng Yang and performed the experiments; Jing-Wen analyzed the data; Jing-Wen wrote the paper. Jing-Wen Ying-Ge Yang performed theFeng experiments; Jing-Wen Feng Feng analyzed data; Jing-Wenof Feng wroteThe the authors paper. declare no conflict of interest. Conflicts Interest: Conflicts of Interest: The authors declare no conflict of interest.
References 1.
Wang, K.; Wan, M.; Hua, C.; Shao, X.F. Determination and application of coarse-grain critical pre-strain curve to aluminum alloy stretch forming. J. Beijing Univ. Aeronaut. Astronaut. 2013, 39, 508–511.
Metals 2016, 6, 13
10 of 10
References 1. 2. 3.
4.
5.
6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.
Wang, K.; Wan, M.; Hua, C.; Shao, X.F. Determination and application of coarse-grain critical pre-strain curve to aluminum alloy stretch forming. J. Beijing Univ. Aeronaut. Astronaut. 2013, 39, 508–511. Han, Z.R.; Dai, L.J.; Zhang, L.Y. Current status of large aircraft skin and panel manufacturing technologies. Aeronaut. Manuf. Technol. 2009, 25, 64–66. Araghi, B.T.; Manco, G.L.; Bambach, M.; Hirt, G. Investigation into a new hybrid forming process: Incremental sheet forming combined with stretch forming. CIRP Ann. Manuf. Technol. 2009, 58, 225–228. [CrossRef] Kurukuri, S.; Miroux, A.; Wisselink, H.; Boogaard, T.V.D. Simulation of stretch forming with intermediate heat treatments of aircraft skins: A physically based modeling approach. Int. J. Mater. Form. 2011, 4, 129–140. [CrossRef] General Editorial Board of “The Manual of Aeronautical Manufacturing Engineering”. In The Aviation Manufacturing Engineering Handbook—Aircraft Sheet Metal Process; Aviation Industry Press: Beijing, China, 1992. Hu, S.G.; Chen, H.Z. Manufacturing Technology of Aircraft Sheet Metal Parts; Beijing University of Aeronautics and Astronautics Press: Beijing, China, 2004. Chang, R.F. Sheet Metal Parts Manufacturing Technology; National Defence Industry Press: Beijing, China, 1992. Wan, M.; Zhou, X.B.; Li, X.X.; Wu, H. Process parameters in stretch forming of mirror skins. Acta Aeronaut. Astronaut. Sin. 1999, 20, 326–330. You, Z.H.; Huang, Y.S.; Wu, R.H. A study of orange-like roughness on front side on Airplane. Aviat. Maint. Eng. 2001, 6, 46–47. Zhang, R.X.; Zeng, Y.S. Development, technological characteristics and application status abroad of aluminum-lithium alloys (In Chinese). Aeronaut. Manuf. Technol. 2007. [CrossRef] Yin, D.F.; Zheng, Z.Q. History and current status of aluminum-lithium alloys research and development. Mater. Rev. 2003, 17, 18–20. Huo, H.Q.; Hao, W.X.; Geng, G.H.; Da, D.A. Development of the new aero-craft material—Aluminum–lithium alloy. Vac. Low Temp. 2005, 11, 63–69. Lyttle, M.T.; Wert, J.A. The plastic anisotropy of an Al–Li–Cu–Zr alloy extrusion in unidirectional deformation. Metall. Mater. Trans. A 1996, 27, 3503–3512. [CrossRef] Li, H.; Tang, Y.; Zeng, Z.; Zheng, Z.; Zheng, F. Effect of ageing time on strength and microstructures of an Al–Cu–Li–Zn–Mg–Mn–Zr alloy. Mater. Sci. Eng. A 2008, 498, 314–320. [CrossRef] Huang, J.C.; Ardell, A.J. Addition rules and the contribution of δ1 precipitates to strengthening of aged Al–Li–Cu alloys. Acta Metall. 1988, 36, 2995–3006. [CrossRef] Fan, C.P.; Zheng, Z.Q.; Jia, M.; Zhong, J.F.; Cheng, B.; Li, H.P.; Wu, Q.P. Microstructure, tensile property and fracture toughness of 2397 Al–Li alloy. Rare Met. Eng. 2015, 44, 91–96. Ma, G.S. Hot Forming Technology of Complex Aluminum Lithium Alloy Parts; Chemical Industry Press: Beijing, China, 2011. Jin, H.X. Basic Experimental Research and Simulation on Aluminum Alloy Mirror Skin Tensile Forming; Beijing University of Aeronautics and Astronautics: Beijing, China, 2009. Wan, M.; Han, J.Q.; Jin, H.X.; Wu, H. Determination of strain criterion of portevin–Le chatelier effect for aluminum alloy sheets. Trans. Nonferrous Met. Soc. China 2006, 16, 1499–1503. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
