- Email: [email protected]
Investigation of Effects (Welding Sequence, Fixturing, Welding Points) on Distortions after Spot Welding for Determining Individual and Cumulative Tolerances
Available online at www.sciencedirect.com
ScienceDirect
Investigation of Effects (Welding49-29 Sequence, Fixturing, Welding Points) on IFAC-PapersOnLine (2016) 030–035 Distortions after Spot Welding forSequence, Determining Individual and Cumulative Investigation of Effects (Welding Fixturing, Welding Points) on Investigation of Effects (Welding Sequence, Fixturing, Welding Points) on Investigation of Effects (Welding Sequence, Fixturing, Welding Points) on Tolerances Distortions after Spot Welding for Determining Individual and Cumulative Distortions after Spot Welding for Determining Individual and Cumulative Distortions after Spot Welding for Determining Individual and Cumulative Tolerances CemTolerances Yurci*. Anil Akdogan** Tolerances
Numan M. Durakbasa*** Cem Cem Yurci*. Yurci*. Anil Anil Akdogan** Akdogan** Cem Yurci*. Anil Akdogan** Numan M. Durakbasa*** Numan M. Durakbasa*** M. Durakbasa*** *Yildiz Technical University,Numan Mechanical Engineering Department, Istanbul, Turkey (Tel: +905324176635; e-mail: yurci_cem@ yahoo.com). *Yildiz University, Mechanical Engineering Department, Istanbul, Turkey *Yildiz Technical Technical **Yildiz TechnicalUniversity, University,Mechanical MechanicalEngineering EngineeringDepartment, Department,Istanbul, Istanbul,Turkey Turkey *Yildiz Technical University, Mechanical Engineering Department, Istanbul, Turkey (Tel: +905324176635; e-mail: yurci_cem@ yahoo.com). (Tel: +905324176635; e-mail: yurci_cem@ yahoo.com). (e-mail: [email protected]) (Tel:University, +905324176635; e-mail: yurci_cem@ yahoo.com). **Yildiz Mechanical Engineering Department, Istanbul, Turkey **Yildiz Technical Technical University, Mechanical Engineering Department, Istanbul, Turkey *** Vienna University of Technology, Institute of Production Engineering and Laser Technology, Department for **Yildiz Technical University, Mechanical Engineering Department, Istanbul, Turkey (e-mail: [email protected]) (e-mail: [email protected]) Interchangeable Manufacturing and Industrial Metrology, Vienna, Austria (e-mail: [email protected]) (e-mail: Production [email protected]) and Laser Technology, Department for *** *** Vienna Vienna University University of of Technology, Technology, Institute Institute of of Production Engineering Engineering and Laser Technology, Department for *** Vienna University of Technology, InstituteMetrology, of Production Engineering(e-mail: and Laser Technology, Department for Interchangeable Manufacturing and Industrial Interchangeable Manufacturing and Industrial Metrology, Vienna, Vienna, Austria Austria (e-mail: [email protected]) [email protected]) Interchangeable Manufacturing andand Industrial Metrology, Vienna, (e-mail:and [email protected]) Abstract: The manufacturing GPS (Geometrical Product Austria Specifications Verification) have to be thought and analysed together. Welding and Metal Forming are two of manufacturing methods, which Abstract: The manufacturing and GPS (Geometrical Product Specifications and Verification) have to Abstract: manufacturing (Geometrical Specifications have to be be are used toThe create sheet metal and partsGPS assembly groups. Product The dimensions do not and stayVerification) steady and they change Abstract: The manufacturing and GPS (Geometrical Product Specifications and Verification) have to be thought and analysed together. Welding and Metal Forming are two of manufacturing methods, which thought and analysed together. Welding and Metal Forming are two of manufacturing methods, which during these two manufacturing methods. firstly, are aftertwo forming (for example bending) the thought andcreate analysed together.parts Welding andBecause Metal Forming of not manufacturing methods, which are sheet assembly groups. The dimensions do and they change are used used to to and create sheet metal metal assembly groups. spot The welding) dimensions not stay stay steady steady and they change springback secondly, afterparts welding (for example thedo distortions exist. In this study, the are used to create sheet metal parts assembly groups. The dimensions do not stay steady and they change during two manufacturing methods. Because firstly, after forming (for example bending) the during these these two manufacturing methods. and Because firstly, on after forming (forand example bending) the distortions after spot welding are researched their effects tolerance chain assembly tolerances during these two manufacturing methods. Becausespot firstly, after the forming (for example bending) the springback and secondly, after (for welding) exist. In study, springback andThese secondly, after welding welding (for example example welding) the distortions distortions exist.some In this thissheet study, the are analysed. distortions are determined with spot analysis software after designing metal springback and secondly, after welding (for example spot welding) the distortions exist. In this study, the distortions after spot are researched and effects on chain and tolerances distortions after spot welding welding are researched and their theirare effects on tolerance tolerance chain and assembly assembly tolerances part assembly examples. After that, these distortions achieved with real measurement (scanning) after distortions after spot distortions welding areare researched andwith theiranalysis effects on tolerance chain and assembly tolerances are analysed. These determined software after designing some sheet metal are analysed. distortions determined with analysis software after designing some sheet metal welding of theThese sheet parts. At theare final step, the analysis and measurement results are compared. are analysed. These distortions are determined with analysis software real after designing some sheet metal part part assembly assembly examples. examples. After After that, that, these these distortions distortions are are achieved achieved with with real measurement measurement (scanning) (scanning) after after Keywords: Statistical Tolerance Analysis Methods, Stackup, Distortions, Spotreserved. Welding, part assembly examples. After that, these distortions areTolerance achieved with real measurement (scanning) after welding of the sheet parts. At the final step, the analysis and measurement results are compared. © 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights welding of the sheet parts. At the final step, the analysis and measurement results are compared. welding of the sheetClamping parts. At the final step, the analysis and measurement results are compared. Welding Sequence, Keywords: Statistical Statistical Tolerance Tolerance Analysis Analysis Methods, Methods, Tolerance Tolerance Stackup, Stackup, Distortions, Distortions, Spot Spot Welding, Welding, Keywords: Keywords: Statistical Tolerance Analysis Methods, Tolerance Stackup, Distortions, Spot Welding, Welding Sequence, Clamping Welding Sequence, Clamping Welding Sequence, Clamping individual and cumulative tolerances is aimed. Some of the 1. INTRODUCTION factors, affecting on distortions also on the shape deviations, individual and cumulative tolerances is Some of individual and welding cumulative tolerances is aimed. aimed. of the the 1. INTRODUCTION are clamping, sequence and the numberSome of welding Sheet metal industry is a very important and economical 1. INTRODUCTION individual and cumulative tolerances is the aimed. Some of the factors, affecting on distortions also on shape deviations, factors, affecting on distortions also on the shape deviations, 1. INTRODUCTION So, to determine these effects andthe their dimensions, at sector. Spot welding takes a big place by the assembly of points. factors, affecting on distortions also on shape deviations, are clamping, sequence and the of welding Sheet metal industry is important economical are clamping, awelding welding sequence and group the number number ofdesigned. welding Sheet metal parts industry isas aainvery very important and and economical the beginning sheet metal assembly will be sheet metal such the automobile or white goods are clamping, welding sequence and the their number of welding Sheet metal industry takes is a very important and assembly economical So, these effects at sector. Spot welding big place by of points. So, to to determine determine these formed effects and and their dimensions, dimensions, at sector. welding takes aa that big today place the by the the This group have simple individual parts, which sector. Spot It is being observed spotassembly welding of is points. points. So, towill determine these effects and their dimensions, at sector. Spot welding takes a the bigautomobile place by the assembly of the beginning a sheet metal assembly group will be designed. sheet metal parts such as in or white goods the beginning a sheet metal assembly group will be designed. sheet metal parts such as in the automobile or white goods will be combined with spot welding. This spot welding made metal approximately fully by automobile robots especially ingoods the the beginning a sheet metal assembly group will be designed. sheet parts such as in that the or white This group will have simple formed individual parts, which sector. It is being observed today the spot welding is This group individual parts, which sector. It is industry. being observed today spot parameters welding is analysis willwill behave madesimple with formed the software ‘Weld Planner’. automobile So, the that design of the welding group will have simple formed individual parts, which sector. It is being observed that today the spot welding is This will be combined with spot welding. This welding made fully robots especially in will be combined with spot Analysis welding. Methods, This spot spotmentioned welding made approximately approximately fully by bywelding robots process especially into the the After that, Statistical Tolerance especially before automated has be will be combined with spot welding. This spot welding made approximately fully by robots especially in the will be made the ‘Weld Planner’. automobile industry. So, of welding parameters analysis will be be used made towith with the software software ‘Weld individual Planner’. automobile carefully industry. because So, the the design design welding above, will progress from these determined willof a role parameters during the analysis analysis will be made with the software ‘Weldmentioned Planner’. automobile industry. So, the they design ofplay welding parameters After that, Statistical Tolerance Analysis Methods, especially before automated welding process has to be After that, Statistical Tolerance Analysis Methods, mentioned especially before automated welding process has to be to cumulative tolerances and compared. At the first whole process inherently. In this context, the toprior After that, Statistical Tolerance Analysis Methods, mentioned especially before automated welding process has be tolerances above, will be used to progress from these individual determined carefully because they will play a role during the will be used to progress from these individual determined carefully because they willtolerances play a roleand during the above, step, the designed individual parts will have the same length determination of sheet metal parts’ critical above, will be used to progressandfrom these At individual determined carefully because they a role during the tolerances to tolerances whole process inherently. In this context, the prior tolerances to cumulative cumulative tolerances and compared. compared. At the the first first whole process inherently. In will this play context, thebecause prior in a dimension so that the computations can be implemented dimensions will provide an economical advantage tolerances to cumulative tolerances and compared. At the first whole processof inherently. In thistolerances context, and the critical prior step, the designed individual parts will have the same length determination sheet metal parts’ step, the designed individual parts will have the same length determination of sheet metal parts’ tolerances and critical for 2-D. some trigonometric calculations can belength used, the rate of waste decrease. Following that, theand Statistical the Also, designed individual parts will have theimplemented same determination of will sheet metal parts’ tolerances critical step, in a dimension so that the computations can be dimensions will provide an economical advantage because in a because dimension thatwill the be computations can be implemented dimensions Analysis will provide an economical advantage because the so parts designed according to that. The Tolerance Methods are very important in too in a2-D. dimension so thattrigonometric the computations can be implemented dimensions will will provide an economical advantage because Also, calculations can be the Following that, the Statistical for 2-D. analysis Also, some some trigonometric calculations can sequence be used, used, the rate rate of of waste waste will decrease. decrease. Following that, the Statistical for welding will be made for several welding Measurement Technique. These methods are ‘arithmetic’, for 2-D. Also, some trigonometric calculations can be used, the rate of waste will decrease. Following that, important the Statistical too because the parts will be designed according to that. Tolerance Analysis Methods are very in too because the parts will be form designed according to that. The The Tolerance and Analysis Methods according are veryto Renault importantnotes in combinations to progressing singular to cumulative. In ‘quadratic’ ‘probabilistic’ because the parts will be designed according to sequence that. The Tolerance Analysis Methods are veryareimportant in too welding analysis will be for welding Measurement Technique. These methods ‘arithmetic’, welding analysis willgroup be made made for several several welding sequence Measurement Technique. These methods areCase ‘arithmetic’, the final step, sheet specimens will be manufactured (2009). Arithmetic method is known as Worst Analysis welding analysis will be made for several welding sequence Measurement Technique. Theseaccording methods toareRenault ‘arithmetic’, combinations to form to cumulative. In ‘quadratic’ and ‘probabilistic’ to progressing progressing form singular singular cumulative. In ‘quadratic’ and method ‘probabilistic’ according to Renault notes notes and the deviations will be measured with to various methods and quadratic is known as Root-sum-square by combinations combinations tosheet progressing form singular to cumulative. In ‘quadratic’ and ‘probabilistic’ according to Renault notes the final step, group specimens will be manufactured (2009). Arithmetic method is known as Worst Case Analysis the final step, sheet group specimens will be manufactured (2009). Arithmetic method is known as Worst Case Analysis and final these step, results will group be compared withwill analysis results. This Fischer (2004). There is another method calledCase Monte Carlo the sheet specimens be manufactured (2009). Arithmetic method isknown known as Root-sum-square Worst Analysis and the deviations will be measured with various methods and quadratic method is as by and the be measured with and various methods and itquadratic method is for known as Root-sum-square by study willdeviations provide a will preliminary impression methodology but is especially used computer-based simulations the deviations will be measured with various methods and quadratic method is known as Root-sum-square by and these results will be compared with analysis results. This Fischer (2004). There is another method called Monte Carlo and these results will be compared with analysis results. This Fischer (2004). There is another method called Monte Carlo researching ofwill some of the spotwith welding process effects (according to Fischer) andanother is not included in thisMonte study. Carlo With for and these results be compared analysis results. This Fischer (2004). There used is method called study will provide a preliminary impression and methodology but it is especially for computer-based simulations study will provideChain. a preliminary impression andthis methodology but it methods, is especially used for computer-based simulations on the Tolerance During these all steps, study will these called ‘Tolerance Stackup’ or ‘Tolerance study will provide asome preliminary impression and methodology but it is especially used for computer-based simulations for researching of of the spot welding process effects (according to Fischer) and is not included in this study. With researching of create some of the spot welding process effects (according to Fischer) and is not included in this study. With for depend on and relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding researching ofChain. some During of the these spot welding process effects (according to Fischer) is not included in this With for the steps, will these calledand ‘Tolerance Stackup’ or study. ‘Tolerance on the Tolerance Tolerance Chain. During these all allGeometrical steps, this this study study will these methods, methods, ‘Tolerance ‘Tolerance on Standards (2012), which introduce product from individual called tolerances. An Stackup’ assembly or (cumulative) on the Tolerance Chain. During these all steps, this study will these methods, called ‘Tolerance Stackup’ or ‘Tolerance depend on and create relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding depend on and create relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding specifications (GPS) – Geometrical tolerancing – Tolerances tolerance can be for example between two parts, which are depend on and create relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding Standards (2012), which introduce Geometrical product from individual tolerances. An assembly (cumulative) Standards (2012), which Geometrical product from individual or tolerances. An that assembly (cumulative) form, orientation, locationintroduce and run-out, and with EN ISO not in a place can have a critical of Standards (2012), which introduce Geometrical product fromside-by-side, individual tolerances. An assembly (cumulative) specifications (GPS) – Geometrical tolerancing – Tolerances tolerance can be for example between two parts, which are specifications (GPS) – Geometrical tolerancing – Tolerances tolerance can be for example between two parts, which are 13920 Standards (1996) which introduce general tolerances importance (for instance a midpoint of a hole). In this study, specifications (GPS) –location Geometricalrun-out, tolerancing with – Tolerances tolerance can be for between two parts, which are of orientation, not side-by-side, or example in that can aa critical of form, form, orientation, location and and run-out, and and with EN EN ISO ISO not side-by-side, in aa place place thatwelding can have have critical welded constructions. analysing how theor after affect on the for of form, orientation, location andintroduce run-out, general and withtolerances EN ISO not side-by-side, ordistortions in aa midpoint place that can have a critical 13920 Standards (1996) which importance (for instance of a hole). In this study, importance (for instance a midpoint of a hole). In this study, 13920 Standards (1996) which introduce general tolerances Standards (1996) which introduce general tolerances importance (for instance a midpoint ofwelding a hole). In thison study, 13920 analysing for welded welded constructions. constructions. analysing how how the the distortions distortions after after welding affect affect on the the for analysing how the distortions after welding affect on the for welded constructions.
2405-8963 © 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2016.11.066
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
2. PREPARATION OF CALCULATION METHODS and ANALYSING In this study some mathematical models will be prepared. The aim is to proceed from the simplest to the more complex. Firstly a beginning example will be detailed. So, the first model (Fig. 1) consists from 3 straight parts (lines) as if they have been assembled. Here it is accepted that not springback or distortion during montage have occurred and the end points (conjunction points) are subjected to a displacement (deformation) (for example: thermal expansion).
Fig. 1. The first mathematical model For this example model the tolerance interval is given for arithmetic, quadratic and probabilistic calculation methods. While the left side of the matrix equations (1-4) shows the changing area, the right side illustrates the tolerance interval limits within it has to be stayed. l10, l20, l30 : Nominal values; l11, l21, l31 : Values after deformation IT values represent the tolerance of each part.
cos γ sin γ
l11 − l10 l 21 − l 20 ≤
cos β sin β
IT1 X + IT2 X + ( IT4 X − IT3 cos α ) (3) IT1Y + IT2Y + ( IT4Y − IT3 sin α )
cos γ sin γ 2
2
l11 − l10 l 21 − l 20 ≤ l31 − l30 2
2
2
2
IT1Y + IT2Y + ( IT4Y − IT3 sin α ) 2
(4)
Besides, some tolerancing problems, solutions and examples are introduced by some studies. In one study, Thimm et al. (2006) investigated the parallelism tolerance and angular error analysis. There are other studies, belonging to Dahlström et al. (2007) and Shaoyun et al. (2006), too which showed tolerance analysis by changing process parameters and compared with actual and real results.
2
2
2
IT1 X + IT2 X + ( IT4 X − IT3 cos α ) 2
(1)
IT1 cos 2 α + IT2 cos 2 β + IT3 cos 2 γ 2
( X12B − X12 A ) + ( X 23B − X 23 A ) + (l4 cos β2 − l3 cos β1 ) − (l4 − l3 cosα ) (Y12B − Y12 A ) + (Y23B − Y23 A ) + (l4 sin β2 + l3 sin β1 ) − (l3 sin α )
≤
Quadratic Method:
2
Arithmetic Method:
( X12B − X12 A ) + ( X 23B − X 23 A ) + (l4 cos β2 − l3 cos β1 ) − (l4 − l3 cosα ) (Y12B − Y12 A ) + (Y23B − Y23 A ) + (l4 sin β2 + l3 sin β1 ) − (l3 sin α )
l31 − l30
IT1 cos α + IT2 cos β + IT3 cos γ IT1 sin α + IT2 sin β + IT3 sin γ
2
Fig. 2. Mathematical model (beginning position and position after distortion)
Quadratic Method:
cos β sin β
cos α sin α
show the coordinates of the endpoints of singular parts for both positions. Here also the coordinate calculation that has been made for other parts could be made for the third part, too. However, trigonometric calculations and (formulas) have been used because of the appropriate geometry of this part. Proceeding from here, not only dimension but also angular tolerances can be considered, included and put in the calculations.
≤
Arithmetic Method:
cos α sin α
31
2
IT1 sin α + IT2 sin β + IT3 sin γ
(2)
The secondly dealed model shows the assembly of three parts, which have been bent with different bending angles and have different lengths, in 2 dimensions. Their length in one dimension is the same. The Fig. 2 shows the beginning position (with the index A) and the position after occurring of distortions for welded construction (with the index B). They
The simulation software programs for welding are being developed newly. Before that, some works including FEA based on electric and thermal formulas have been generated for Aluminium Resistance Spot Welding Process by Sun et al. (2000). There is another FEA study of Nied (1983) dividing the workpiece and electrode contact region into small quadrilaterals. Also, spot welding has been analysed, for other aims, too, for example to study nugget formation by modelling the process for the software ANSYS by Thakur et al. (2010). The resistance spot welding has been modelled
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
32
and analysed for giving a gap and not giving a gap by Moos et al. (2012). Also Moos et al. (2015) worked about the modelling of spot welding for interference condition of the sheet metal parts, too. 3. EXPERIMENT and ANALYSIS EXAMPLES and THEIR RESULTS 3. 1 Experimental Procedure and its Results The drawings of practice singular parts are as follows:
Fig. 5. The locations of spot weld points for Part 1and Part 2 For the assembly groups 6, 7 and 8 the last weld line in three parts downwards has not been applied (Fig. 6). For the assembly models above, it has been tried to consider the symmetry to provide staying in 2D deviations because at the beginning the analysis of 2 dimensional deformations has been aimed. But in future it has been observed that this assumption was not true while the distortions are affected by many factors such as welding sequence.
Part 1
Part 2
Part 3
Fig. 3. Technical drawings of designed singular parts These parts have been manufactured in Abkant with bending in the firm Öztiryakiler. The thickness of every part is 0.8 mm and its material is Stainless Steel 304. There width is 150 mm. The assembly after spot welding is shown in Fig. 4. The 3 parts are welded according to the sequence 1-2 and 2-3.
Part 1
Part 3
Fig. 6. The spot weld points examples for assembly groups 6,7 and 8
Fig. 4. The assembly model after spot welding By spot welding, to obtain bigger distortions and so to observe the distortions easier, the electrode with the biggest surface has been used. But this opinion at the beginning was false because it depends on other various effects. A fixture has not been employed not to create restrictive effect on distortions and so a deformation release has been provided. Besides, the fingers holding the parts can be modelled as clamping points or not. They can be modelled so as if they have a very low stiffness. In the simulations in the software Weld Planner they have been modelled as free clamps. For the assembly groups marked with number 3, 4 and 5, the butt contact surface (blue circles) of electrodes has been shown in the Fig. 5. These blue areas do not represent the size of nugget diameters. They show the contact area of the electrode.
The singular parts have been numbered and assembled with spot welding process according to these numbers. The singular parts before spot welding and assembly groups after spot welding have been 3D-scanned. For these purpose, Atos 3D Scanner of GOM has been used. To reach to the deviation after spot welding the singular parts and the groups have been aligned based on RPS and Best Fit method in GOM Inspect V8 Software. During this work, the scanned data in the STL format of the singular parts have been accepted as CAD data and the other STL data of assembly groups as Mesh format. Applied Spot Welding Parameters are as follows: Approaching time: 10, Squeeze time: 15, Start current: % 15, Current rise time (Upslope): 20, Weld time: 15, Weld current: % 35, Impulse Number: 0, Finish Current: % 0, Current fall time (Downslope): 15, Hold time: 5, Opening time: 25 The time unit is 0.02 s because the current frequency is 50 Hz. The current magnitude is not certain and it is determined with practice because it changes according to the operation. The real magnitude has to be measured with ammeter (for example %35 weld current can correspond approximately to 9 kA).
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
33
Table 1. An example for the data obtained with RPS Method, for Part 1
Fig. 7. The current-force diagram of the used spot welding machine The current-force diagram in Fig. 7 is very similar but different from the ones shown in the DIN EN ISO 17677-1 Standard (2010). Also the manufacturer of the used spot welding machine applies this diagram and the used machine works according to this diagram.
The deviation results can be compared with the values in EN ISO 13920 Standards. For example for these practice parts the tolerance value in Table 1 for Class C (± 3 and ± 4 mm) can be considered. For the parts in this study, it is accepted that the Class C provides the satisfactory geometric specifications according to functional specifications. As it is seen in Fig. 9, after bending the form of the singular parts does not match exactly with the CAD model. For example, the whole of vertical face of Part 1 (Fig. 3) does not possess the angle with the magnitude 90° to the horizontal face. Also its form is curved. Because of that the equations foreseen (1,4) can not be used for the practice division of this study. For these equations, the tip of the parts were thought that they would move together after welding distortions as if they were combined. But the movement of the tip of the parts can be compared to the opening of a rose. Also the tips are narrowing but not exactly touching (Fig. 10). If the parts after manufacturing had the exactly the same form like the CAD part, again they would not have the foreseen displacement movement in Fig. 2. This can be supported with the distortion results after the welding simulation in Fig. 14. In Fig. 11, in some regions 4 and in some regions 2 surfaces are illustrated if the result is magnified with a very big factor. But at the beginning of the simulation two middle surfaces were existing. Also this shows too, that there is an open area even if it is very small. Also the distortions change along the edge of the part (effect of welding sequence) which matches with the scanning results in spite of very big magnitude differences.
Fig. 8 Surface Comparison on CAD - Part 1 – Results for RPS of Assembly Group 4 After the applying of the aligning method as it is seen in the Fig. 8 the locations on a face of the parts are showing very different deviation results. Because of that we have to make an assumption and take the value on a certain point into consideration. Like in the Fig. 8 the results at the upper, middle and lower points can be taken into account. But one of these points has to be chosen when to make transition to tolerance calculations. For now, the deviatons at the middle point are thought to be considered. However, to take the biggest deviation into consideration can be logic, too. The deviation results are as follows:
Fig. 9. The STL format of the manufactured part in GOM Inspect
34
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
Sequence 1:
Sequence 2:
Blue Part: The scanned STL format of the singular Part 1 Grey Part: The assembled part group Red Region: The movement of the tip Fig. 10. The tip of Part 1 narrows to the other part.
Fig. 12. The illustration of welding sequences
Fig. 13. The formed nugget The process parameters were exaggerated. So, the peel test has been applied to control the nugget. It has been observed that a stabile nugget (Fig. 13) has generated and its diameter is 3 mm. Because of that two simulations with 3 mm and 5 mm spot diameter have been applied. Fig. 11. The illustration of sheet surfaces 3. 2 An Example for Spot Welding Analysis: In the spot welding analysis the software Weld Planner is used. This software belongs to the firm ESI Group. However the software Sysweld which considers the phase transformations could be used but Weld Planner has been preferred because its properties are enough for this study’s aims. Weld Planner works according to the Local Shrinkage Analysis. It is based on the shrinkage following the thermal expansion after reaching the melting temperature. Jackson et al. (2009) sustained that its biggest advantage is the very short calculation time when compared with Sysweld and its results are satisfactory especially if the welding sequence and clamping locations have to be investigated.
3. 2. 1 The results of analyses in Weld Planner showing two effects on distortions after spot welding 1. The number of Points (with the same welding sequence) The translational displacement at middle points (on Part 1) (according to Fig. 15): (Spot Diameter: 5 mm) Assembly group 3,4 and 5: 0.01932 mm (Fig. 14) Assembly group 6,7 and 8: 0.01489 mm (Spot Diameter: 3 mm) Assembly group 3,4 and 5: 0.008501mm Assembly group 6,7 and 8: 0.005267 mm
The middle surface models of manufactured parts have been created in SOLIDWORKS. After that, the appropriate mesh has been created in VISUAL-MESH. Especially for the locations around the spot weld points a precise layer mesh has been created. Before that, the spot welding points have been assigned in the interspace software VISUAL-WELD. Then, this example has been imported into the software WELD PLANNER and the distortions after the spot welding have been obtained. The welding sequences and the number of lines consisting welding points have been changed to see their effects.
Fig. 14. An Analysis Example in Weld Planner for the sequence 1 and number of weld points according to the assembly groups 3,4 and 5
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
35
The studies explained here are continuing. So, to see the effects of the clamping, welding sequence and welding points on distortions various welding simulations will be implemented. Besides, the parts manufactured with different width and material will be inspected and their results will be interpreted, too. Also this study will be broadened. REFERENCES
Fig. 15. The model for extracting the displacement at the middle point 2. The welding sequence (according to Fig. 12) (with the same number point) The translational displacement at middle points (on Part 1): Sequence 1: 0.01932 mm (Spot Diameter: 5 mm) Sequence 2: 0.01938 mm Sequence 1: 0.008501 mm (Spot Diameter: 3 mm) Sequence 2: 0.007823 mm 4. CONCLUSIONS This was an initial study to create a beginning methodology for sheet metal part manufacturers. Because of that the experiments detailed here were not conducted according to the rules of Design of Experiments. Also the scanning and aligning process can be bounded to a specific methodology, too (an example by Hammett et al. (2009)). At the beginning it was aimed to use the mathematical formulas above to make transition between the distortions (deviations) after spot welding and the cumulative tolerances. But these formulas can not be used for our model. Because the tip of the parts (the end conjuction points) do not move together. They are opening like a rose. The scanning results and the welding simulations confirmed that. If the results are inspected, also there is a very big difference between the scanning and simulation result. It can be that the scanning results have been affected by many factors and the measurement accuracies might be unsatisfactory. It has been seen, too that the RPS aligning gave better results than Best Fit aligning. Also there is a clear difference between the specimens, too. The distortion values in the simulations are very small because only free clamps have been applied here. The distortions are being affected in a big magnitude by clamps. According to the welding simulations the distortion at the middle point increased when the welding point line increased. The welding sequence did not affect so much on the distortion. But these observations will vary according to the form and construction of sheet metal parts. Besides, as it has been foreseen, the distortions increase when the spot diameters increase.
Dahlström S., Lindkvist, L., Söderberg, R. (2007). Practical Implications in Tolerance Analysis of Sheet Metal Assemblies: Experiences from an Automative Application. Models for Computer Aided Tolerancing in Design and Manufacturing, 311-320. DIN EN ISO 17677-1 Standard. (2010). Widerstandsschweissen – Begriffe – Teil 1: Punkt-, Buckel- und Rollennahtschweissen. EN ISO 1101 Standards. (2012). Geometrical Product Specifications (GPS) – Geometrical Tolerancing – Tolerances of Form, Orientation, Location and Run-out. EN ISO 13920 Standards. (1996). General Tolerances for Welded Constructions. Fischer, B. R. (2004). Mechanical Tolerance Stackup and Analysis, 61-107. Marcel Dekker, New York. Hammett P. D., Guzman L. M. G., Geddes S. W., Walsh P. T. (2009). Quantifying Alignment Effects in 3D Coordinate Measurement. University of Michigan Transportation Research Institute, Manufacturing Validation Solutions, LLC, 22. Jackson, K., Darlington, R. (2009). Advanced Engineering Methods for Assessing Welding Distortion in AeroEngine Assemblies. IOP Publishing. Moos, S., Vezzetti E. (2012). Complaint Assembly Tolerance Analysis: Guidelines to Formalize the Resistance Spot Welding Plasticity Effects. Int J Adv Manuf Technol, 61, 503-518. Moos, S., Vezzetti E. (2015). Resistance Spot Welding Process Simulation for Variational Analysis on Compliant Assemblies. Journal of Manufacturing Systems., Vol. 12, 44-71. Nied, H. A., (1983). The Finite Element Modeling of the Resistance Spot Welding Process. Welding Research Supplement, 123-132. Renault – Tolerance Chain Education Lecture Notes. (2009). Shaoyun, C., Zhongqin, L., Yizhu, Z., Yongbing, L. (2006). A Parametric Study of Sheet Metal Joints for Dimensional Integrity. Int J Adv Manuf Technol, 29, 446-452. Sun, X., Dong, P. (2000). Analysis of Aluminium Resistance Spot Welding Processes Using Coupled Finite Element Procedures. Welding Research Supplement, 215-221. Thakur, A. G., Rasane, A. R., Nandedkar V. M. (2010). Finite Element Analysis of Resistance Spot Welding to study Nugget Formation. International Journal of Applied Engineering Research, Dindigul, Vol. 1, no. 3, 483-490. Thimm, G., Rui, W., Yongsheng, M. (2006). Tolerance Transfer in Sheet Metal Forming. International Journal of Production Research, Vol. 2, no. 5, 3289-3309.
ScienceDirect
Investigation of Effects (Welding49-29 Sequence, Fixturing, Welding Points) on IFAC-PapersOnLine (2016) 030–035 Distortions after Spot Welding forSequence, Determining Individual and Cumulative Investigation of Effects (Welding Fixturing, Welding Points) on Investigation of Effects (Welding Sequence, Fixturing, Welding Points) on Investigation of Effects (Welding Sequence, Fixturing, Welding Points) on Tolerances Distortions after Spot Welding for Determining Individual and Cumulative Distortions after Spot Welding for Determining Individual and Cumulative Distortions after Spot Welding for Determining Individual and Cumulative Tolerances CemTolerances Yurci*. Anil Akdogan** Tolerances
Numan M. Durakbasa*** Cem Cem Yurci*. Yurci*. Anil Anil Akdogan** Akdogan** Cem Yurci*. Anil Akdogan** Numan M. Durakbasa*** Numan M. Durakbasa*** M. Durakbasa*** *Yildiz Technical University,Numan Mechanical Engineering Department, Istanbul, Turkey (Tel: +905324176635; e-mail: yurci_cem@ yahoo.com). *Yildiz University, Mechanical Engineering Department, Istanbul, Turkey *Yildiz Technical Technical **Yildiz TechnicalUniversity, University,Mechanical MechanicalEngineering EngineeringDepartment, Department,Istanbul, Istanbul,Turkey Turkey *Yildiz Technical University, Mechanical Engineering Department, Istanbul, Turkey (Tel: +905324176635; e-mail: yurci_cem@ yahoo.com). (Tel: +905324176635; e-mail: yurci_cem@ yahoo.com). (e-mail: [email protected]) (Tel:University, +905324176635; e-mail: yurci_cem@ yahoo.com). **Yildiz Mechanical Engineering Department, Istanbul, Turkey **Yildiz Technical Technical University, Mechanical Engineering Department, Istanbul, Turkey *** Vienna University of Technology, Institute of Production Engineering and Laser Technology, Department for **Yildiz Technical University, Mechanical Engineering Department, Istanbul, Turkey (e-mail: [email protected]) (e-mail: [email protected]) Interchangeable Manufacturing and Industrial Metrology, Vienna, Austria (e-mail: [email protected]) (e-mail: Production [email protected]) and Laser Technology, Department for *** *** Vienna Vienna University University of of Technology, Technology, Institute Institute of of Production Engineering Engineering and Laser Technology, Department for *** Vienna University of Technology, InstituteMetrology, of Production Engineering(e-mail: and Laser Technology, Department for Interchangeable Manufacturing and Industrial Interchangeable Manufacturing and Industrial Metrology, Vienna, Vienna, Austria Austria (e-mail: [email protected]) [email protected]) Interchangeable Manufacturing andand Industrial Metrology, Vienna, (e-mail:and [email protected]) Abstract: The manufacturing GPS (Geometrical Product Austria Specifications Verification) have to be thought and analysed together. Welding and Metal Forming are two of manufacturing methods, which Abstract: The manufacturing and GPS (Geometrical Product Specifications and Verification) have to Abstract: manufacturing (Geometrical Specifications have to be be are used toThe create sheet metal and partsGPS assembly groups. Product The dimensions do not and stayVerification) steady and they change Abstract: The manufacturing and GPS (Geometrical Product Specifications and Verification) have to be thought and analysed together. Welding and Metal Forming are two of manufacturing methods, which thought and analysed together. Welding and Metal Forming are two of manufacturing methods, which during these two manufacturing methods. firstly, are aftertwo forming (for example bending) the thought andcreate analysed together.parts Welding andBecause Metal Forming of not manufacturing methods, which are sheet assembly groups. The dimensions do and they change are used used to to and create sheet metal metal assembly groups. spot The welding) dimensions not stay stay steady steady and they change springback secondly, afterparts welding (for example thedo distortions exist. In this study, the are used to create sheet metal parts assembly groups. The dimensions do not stay steady and they change during two manufacturing methods. Because firstly, after forming (for example bending) the during these these two manufacturing methods. and Because firstly, on after forming (forand example bending) the distortions after spot welding are researched their effects tolerance chain assembly tolerances during these two manufacturing methods. Becausespot firstly, after the forming (for example bending) the springback and secondly, after (for welding) exist. In study, springback andThese secondly, after welding welding (for example example welding) the distortions distortions exist.some In this thissheet study, the are analysed. distortions are determined with spot analysis software after designing metal springback and secondly, after welding (for example spot welding) the distortions exist. In this study, the distortions after spot are researched and effects on chain and tolerances distortions after spot welding welding are researched and their theirare effects on tolerance tolerance chain and assembly assembly tolerances part assembly examples. After that, these distortions achieved with real measurement (scanning) after distortions after spot distortions welding areare researched andwith theiranalysis effects on tolerance chain and assembly tolerances are analysed. These determined software after designing some sheet metal are analysed. distortions determined with analysis software after designing some sheet metal welding of theThese sheet parts. At theare final step, the analysis and measurement results are compared. are analysed. These distortions are determined with analysis software real after designing some sheet metal part part assembly assembly examples. examples. After After that, that, these these distortions distortions are are achieved achieved with with real measurement measurement (scanning) (scanning) after after Keywords: Statistical Tolerance Analysis Methods, Stackup, Distortions, Spotreserved. Welding, part assembly examples. After that, these distortions areTolerance achieved with real measurement (scanning) after welding of the sheet parts. At the final step, the analysis and measurement results are compared. © 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights welding of the sheet parts. At the final step, the analysis and measurement results are compared. welding of the sheetClamping parts. At the final step, the analysis and measurement results are compared. Welding Sequence, Keywords: Statistical Statistical Tolerance Tolerance Analysis Analysis Methods, Methods, Tolerance Tolerance Stackup, Stackup, Distortions, Distortions, Spot Spot Welding, Welding, Keywords: Keywords: Statistical Tolerance Analysis Methods, Tolerance Stackup, Distortions, Spot Welding, Welding Sequence, Clamping Welding Sequence, Clamping Welding Sequence, Clamping individual and cumulative tolerances is aimed. Some of the 1. INTRODUCTION factors, affecting on distortions also on the shape deviations, individual and cumulative tolerances is Some of individual and welding cumulative tolerances is aimed. aimed. of the the 1. INTRODUCTION are clamping, sequence and the numberSome of welding Sheet metal industry is a very important and economical 1. INTRODUCTION individual and cumulative tolerances is the aimed. Some of the factors, affecting on distortions also on shape deviations, factors, affecting on distortions also on the shape deviations, 1. INTRODUCTION So, to determine these effects andthe their dimensions, at sector. Spot welding takes a big place by the assembly of points. factors, affecting on distortions also on shape deviations, are clamping, sequence and the of welding Sheet metal industry is important economical are clamping, awelding welding sequence and group the number number ofdesigned. welding Sheet metal parts industry isas aainvery very important and and economical the beginning sheet metal assembly will be sheet metal such the automobile or white goods are clamping, welding sequence and the their number of welding Sheet metal industry takes is a very important and assembly economical So, these effects at sector. Spot welding big place by of points. So, to to determine determine these formed effects and and their dimensions, dimensions, at sector. welding takes aa that big today place the by the the This group have simple individual parts, which sector. Spot It is being observed spotassembly welding of is points. points. So, towill determine these effects and their dimensions, at sector. Spot welding takes a the bigautomobile place by the assembly of the beginning a sheet metal assembly group will be designed. sheet metal parts such as in or white goods the beginning a sheet metal assembly group will be designed. sheet metal parts such as in the automobile or white goods will be combined with spot welding. This spot welding made metal approximately fully by automobile robots especially ingoods the the beginning a sheet metal assembly group will be designed. sheet parts such as in that the or white This group will have simple formed individual parts, which sector. It is being observed today the spot welding is This group individual parts, which sector. It is industry. being observed today spot parameters welding is analysis willwill behave madesimple with formed the software ‘Weld Planner’. automobile So, the that design of the welding group will have simple formed individual parts, which sector. It is being observed that today the spot welding is This will be combined with spot welding. This welding made fully robots especially in will be combined with spot Analysis welding. Methods, This spot spotmentioned welding made approximately approximately fully by bywelding robots process especially into the the After that, Statistical Tolerance especially before automated has be will be combined with spot welding. This spot welding made approximately fully by robots especially in the will be made the ‘Weld Planner’. automobile industry. So, of welding parameters analysis will be be used made towith with the software software ‘Weld individual Planner’. automobile carefully industry. because So, the the design design welding above, will progress from these determined willof a role parameters during the analysis analysis will be made with the software ‘Weldmentioned Planner’. automobile industry. So, the they design ofplay welding parameters After that, Statistical Tolerance Analysis Methods, especially before automated welding process has to be After that, Statistical Tolerance Analysis Methods, mentioned especially before automated welding process has to be to cumulative tolerances and compared. At the first whole process inherently. In this context, the toprior After that, Statistical Tolerance Analysis Methods, mentioned especially before automated welding process has be tolerances above, will be used to progress from these individual determined carefully because they will play a role during the will be used to progress from these individual determined carefully because they willtolerances play a roleand during the above, step, the designed individual parts will have the same length determination of sheet metal parts’ critical above, will be used to progressandfrom these At individual determined carefully because they a role during the tolerances to tolerances whole process inherently. In this context, the prior tolerances to cumulative cumulative tolerances and compared. compared. At the the first first whole process inherently. In will this play context, thebecause prior in a dimension so that the computations can be implemented dimensions will provide an economical advantage tolerances to cumulative tolerances and compared. At the first whole processof inherently. In thistolerances context, and the critical prior step, the designed individual parts will have the same length determination sheet metal parts’ step, the designed individual parts will have the same length determination of sheet metal parts’ tolerances and critical for 2-D. some trigonometric calculations can belength used, the rate of waste decrease. Following that, theand Statistical the Also, designed individual parts will have theimplemented same determination of will sheet metal parts’ tolerances critical step, in a dimension so that the computations can be dimensions will provide an economical advantage because in a because dimension thatwill the be computations can be implemented dimensions Analysis will provide an economical advantage because the so parts designed according to that. The Tolerance Methods are very important in too in a2-D. dimension so thattrigonometric the computations can be implemented dimensions will will provide an economical advantage because Also, calculations can be the Following that, the Statistical for 2-D. analysis Also, some some trigonometric calculations can sequence be used, used, the rate rate of of waste waste will decrease. decrease. Following that, the Statistical for welding will be made for several welding Measurement Technique. These methods are ‘arithmetic’, for 2-D. Also, some trigonometric calculations can be used, the rate of waste will decrease. Following that, important the Statistical too because the parts will be designed according to that. Tolerance Analysis Methods are very in too because the parts will be form designed according to that. The The Tolerance and Analysis Methods according are veryto Renault importantnotes in combinations to progressing singular to cumulative. In ‘quadratic’ ‘probabilistic’ because the parts will be designed according to sequence that. The Tolerance Analysis Methods are veryareimportant in too welding analysis will be for welding Measurement Technique. These methods ‘arithmetic’, welding analysis willgroup be made made for several several welding sequence Measurement Technique. These methods areCase ‘arithmetic’, the final step, sheet specimens will be manufactured (2009). Arithmetic method is known as Worst Analysis welding analysis will be made for several welding sequence Measurement Technique. Theseaccording methods toareRenault ‘arithmetic’, combinations to form to cumulative. In ‘quadratic’ and ‘probabilistic’ to progressing progressing form singular singular cumulative. In ‘quadratic’ and method ‘probabilistic’ according to Renault notes notes and the deviations will be measured with to various methods and quadratic is known as Root-sum-square by combinations combinations tosheet progressing form singular to cumulative. In ‘quadratic’ and ‘probabilistic’ according to Renault notes the final step, group specimens will be manufactured (2009). Arithmetic method is known as Worst Case Analysis the final step, sheet group specimens will be manufactured (2009). Arithmetic method is known as Worst Case Analysis and final these step, results will group be compared withwill analysis results. This Fischer (2004). There is another method calledCase Monte Carlo the sheet specimens be manufactured (2009). Arithmetic method isknown known as Root-sum-square Worst Analysis and the deviations will be measured with various methods and quadratic method is as by and the be measured with and various methods and itquadratic method is for known as Root-sum-square by study willdeviations provide a will preliminary impression methodology but is especially used computer-based simulations the deviations will be measured with various methods and quadratic method is known as Root-sum-square by and these results will be compared with analysis results. This Fischer (2004). There is another method called Monte Carlo and these results will be compared with analysis results. This Fischer (2004). There is another method called Monte Carlo researching ofwill some of the spotwith welding process effects (according to Fischer) andanother is not included in thisMonte study. Carlo With for and these results be compared analysis results. This Fischer (2004). There used is method called study will provide a preliminary impression and methodology but it is especially for computer-based simulations study will provideChain. a preliminary impression andthis methodology but it methods, is especially used for computer-based simulations on the Tolerance During these all steps, study will these called ‘Tolerance Stackup’ or ‘Tolerance study will provide asome preliminary impression and methodology but it is especially used for computer-based simulations for researching of of the spot welding process effects (according to Fischer) and is not included in this study. With researching of create some of the spot welding process effects (according to Fischer) and is not included in this study. With for depend on and relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding researching ofChain. some During of the these spot welding process effects (according to Fischer) is not included in this With for the steps, will these calledand ‘Tolerance Stackup’ or study. ‘Tolerance on the Tolerance Tolerance Chain. During these all allGeometrical steps, this this study study will these methods, methods, ‘Tolerance ‘Tolerance on Standards (2012), which introduce product from individual called tolerances. An Stackup’ assembly or (cumulative) on the Tolerance Chain. During these all steps, this study will these methods, called ‘Tolerance Stackup’ or ‘Tolerance depend on and create relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding depend on and create relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding specifications (GPS) – Geometrical tolerancing – Tolerances tolerance can be for example between two parts, which are depend on and create relationship with EN ISO 1101 Chain’, cumulative tolerances can be reached by proceeding Standards (2012), which introduce Geometrical product from individual tolerances. An assembly (cumulative) Standards (2012), which Geometrical product from individual or tolerances. An that assembly (cumulative) form, orientation, locationintroduce and run-out, and with EN ISO not in a place can have a critical of Standards (2012), which introduce Geometrical product fromside-by-side, individual tolerances. An assembly (cumulative) specifications (GPS) – Geometrical tolerancing – Tolerances tolerance can be for example between two parts, which are specifications (GPS) – Geometrical tolerancing – Tolerances tolerance can be for example between two parts, which are 13920 Standards (1996) which introduce general tolerances importance (for instance a midpoint of a hole). In this study, specifications (GPS) –location Geometricalrun-out, tolerancing with – Tolerances tolerance can be for between two parts, which are of orientation, not side-by-side, or example in that can aa critical of form, form, orientation, location and and run-out, and and with EN EN ISO ISO not side-by-side, in aa place place thatwelding can have have critical welded constructions. analysing how theor after affect on the for of form, orientation, location andintroduce run-out, general and withtolerances EN ISO not side-by-side, ordistortions in aa midpoint place that can have a critical 13920 Standards (1996) which importance (for instance of a hole). In this study, importance (for instance a midpoint of a hole). In this study, 13920 Standards (1996) which introduce general tolerances Standards (1996) which introduce general tolerances importance (for instance a midpoint ofwelding a hole). In thison study, 13920 analysing for welded welded constructions. constructions. analysing how how the the distortions distortions after after welding affect affect on the the for analysing how the distortions after welding affect on the for welded constructions.
2405-8963 © 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2016.11.066
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
2. PREPARATION OF CALCULATION METHODS and ANALYSING In this study some mathematical models will be prepared. The aim is to proceed from the simplest to the more complex. Firstly a beginning example will be detailed. So, the first model (Fig. 1) consists from 3 straight parts (lines) as if they have been assembled. Here it is accepted that not springback or distortion during montage have occurred and the end points (conjunction points) are subjected to a displacement (deformation) (for example: thermal expansion).
Fig. 1. The first mathematical model For this example model the tolerance interval is given for arithmetic, quadratic and probabilistic calculation methods. While the left side of the matrix equations (1-4) shows the changing area, the right side illustrates the tolerance interval limits within it has to be stayed. l10, l20, l30 : Nominal values; l11, l21, l31 : Values after deformation IT values represent the tolerance of each part.
cos γ sin γ
l11 − l10 l 21 − l 20 ≤
cos β sin β
IT1 X + IT2 X + ( IT4 X − IT3 cos α ) (3) IT1Y + IT2Y + ( IT4Y − IT3 sin α )
cos γ sin γ 2
2
l11 − l10 l 21 − l 20 ≤ l31 − l30 2
2
2
2
IT1Y + IT2Y + ( IT4Y − IT3 sin α ) 2
(4)
Besides, some tolerancing problems, solutions and examples are introduced by some studies. In one study, Thimm et al. (2006) investigated the parallelism tolerance and angular error analysis. There are other studies, belonging to Dahlström et al. (2007) and Shaoyun et al. (2006), too which showed tolerance analysis by changing process parameters and compared with actual and real results.
2
2
2
IT1 X + IT2 X + ( IT4 X − IT3 cos α ) 2
(1)
IT1 cos 2 α + IT2 cos 2 β + IT3 cos 2 γ 2
( X12B − X12 A ) + ( X 23B − X 23 A ) + (l4 cos β2 − l3 cos β1 ) − (l4 − l3 cosα ) (Y12B − Y12 A ) + (Y23B − Y23 A ) + (l4 sin β2 + l3 sin β1 ) − (l3 sin α )
≤
Quadratic Method:
2
Arithmetic Method:
( X12B − X12 A ) + ( X 23B − X 23 A ) + (l4 cos β2 − l3 cos β1 ) − (l4 − l3 cosα ) (Y12B − Y12 A ) + (Y23B − Y23 A ) + (l4 sin β2 + l3 sin β1 ) − (l3 sin α )
l31 − l30
IT1 cos α + IT2 cos β + IT3 cos γ IT1 sin α + IT2 sin β + IT3 sin γ
2
Fig. 2. Mathematical model (beginning position and position after distortion)
Quadratic Method:
cos β sin β
cos α sin α
show the coordinates of the endpoints of singular parts for both positions. Here also the coordinate calculation that has been made for other parts could be made for the third part, too. However, trigonometric calculations and (formulas) have been used because of the appropriate geometry of this part. Proceeding from here, not only dimension but also angular tolerances can be considered, included and put in the calculations.
≤
Arithmetic Method:
cos α sin α
31
2
IT1 sin α + IT2 sin β + IT3 sin γ
(2)
The secondly dealed model shows the assembly of three parts, which have been bent with different bending angles and have different lengths, in 2 dimensions. Their length in one dimension is the same. The Fig. 2 shows the beginning position (with the index A) and the position after occurring of distortions for welded construction (with the index B). They
The simulation software programs for welding are being developed newly. Before that, some works including FEA based on electric and thermal formulas have been generated for Aluminium Resistance Spot Welding Process by Sun et al. (2000). There is another FEA study of Nied (1983) dividing the workpiece and electrode contact region into small quadrilaterals. Also, spot welding has been analysed, for other aims, too, for example to study nugget formation by modelling the process for the software ANSYS by Thakur et al. (2010). The resistance spot welding has been modelled
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
32
and analysed for giving a gap and not giving a gap by Moos et al. (2012). Also Moos et al. (2015) worked about the modelling of spot welding for interference condition of the sheet metal parts, too. 3. EXPERIMENT and ANALYSIS EXAMPLES and THEIR RESULTS 3. 1 Experimental Procedure and its Results The drawings of practice singular parts are as follows:
Fig. 5. The locations of spot weld points for Part 1and Part 2 For the assembly groups 6, 7 and 8 the last weld line in three parts downwards has not been applied (Fig. 6). For the assembly models above, it has been tried to consider the symmetry to provide staying in 2D deviations because at the beginning the analysis of 2 dimensional deformations has been aimed. But in future it has been observed that this assumption was not true while the distortions are affected by many factors such as welding sequence.
Part 1
Part 2
Part 3
Fig. 3. Technical drawings of designed singular parts These parts have been manufactured in Abkant with bending in the firm Öztiryakiler. The thickness of every part is 0.8 mm and its material is Stainless Steel 304. There width is 150 mm. The assembly after spot welding is shown in Fig. 4. The 3 parts are welded according to the sequence 1-2 and 2-3.
Part 1
Part 3
Fig. 6. The spot weld points examples for assembly groups 6,7 and 8
Fig. 4. The assembly model after spot welding By spot welding, to obtain bigger distortions and so to observe the distortions easier, the electrode with the biggest surface has been used. But this opinion at the beginning was false because it depends on other various effects. A fixture has not been employed not to create restrictive effect on distortions and so a deformation release has been provided. Besides, the fingers holding the parts can be modelled as clamping points or not. They can be modelled so as if they have a very low stiffness. In the simulations in the software Weld Planner they have been modelled as free clamps. For the assembly groups marked with number 3, 4 and 5, the butt contact surface (blue circles) of electrodes has been shown in the Fig. 5. These blue areas do not represent the size of nugget diameters. They show the contact area of the electrode.
The singular parts have been numbered and assembled with spot welding process according to these numbers. The singular parts before spot welding and assembly groups after spot welding have been 3D-scanned. For these purpose, Atos 3D Scanner of GOM has been used. To reach to the deviation after spot welding the singular parts and the groups have been aligned based on RPS and Best Fit method in GOM Inspect V8 Software. During this work, the scanned data in the STL format of the singular parts have been accepted as CAD data and the other STL data of assembly groups as Mesh format. Applied Spot Welding Parameters are as follows: Approaching time: 10, Squeeze time: 15, Start current: % 15, Current rise time (Upslope): 20, Weld time: 15, Weld current: % 35, Impulse Number: 0, Finish Current: % 0, Current fall time (Downslope): 15, Hold time: 5, Opening time: 25 The time unit is 0.02 s because the current frequency is 50 Hz. The current magnitude is not certain and it is determined with practice because it changes according to the operation. The real magnitude has to be measured with ammeter (for example %35 weld current can correspond approximately to 9 kA).
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
33
Table 1. An example for the data obtained with RPS Method, for Part 1
Fig. 7. The current-force diagram of the used spot welding machine The current-force diagram in Fig. 7 is very similar but different from the ones shown in the DIN EN ISO 17677-1 Standard (2010). Also the manufacturer of the used spot welding machine applies this diagram and the used machine works according to this diagram.
The deviation results can be compared with the values in EN ISO 13920 Standards. For example for these practice parts the tolerance value in Table 1 for Class C (± 3 and ± 4 mm) can be considered. For the parts in this study, it is accepted that the Class C provides the satisfactory geometric specifications according to functional specifications. As it is seen in Fig. 9, after bending the form of the singular parts does not match exactly with the CAD model. For example, the whole of vertical face of Part 1 (Fig. 3) does not possess the angle with the magnitude 90° to the horizontal face. Also its form is curved. Because of that the equations foreseen (1,4) can not be used for the practice division of this study. For these equations, the tip of the parts were thought that they would move together after welding distortions as if they were combined. But the movement of the tip of the parts can be compared to the opening of a rose. Also the tips are narrowing but not exactly touching (Fig. 10). If the parts after manufacturing had the exactly the same form like the CAD part, again they would not have the foreseen displacement movement in Fig. 2. This can be supported with the distortion results after the welding simulation in Fig. 14. In Fig. 11, in some regions 4 and in some regions 2 surfaces are illustrated if the result is magnified with a very big factor. But at the beginning of the simulation two middle surfaces were existing. Also this shows too, that there is an open area even if it is very small. Also the distortions change along the edge of the part (effect of welding sequence) which matches with the scanning results in spite of very big magnitude differences.
Fig. 8 Surface Comparison on CAD - Part 1 – Results for RPS of Assembly Group 4 After the applying of the aligning method as it is seen in the Fig. 8 the locations on a face of the parts are showing very different deviation results. Because of that we have to make an assumption and take the value on a certain point into consideration. Like in the Fig. 8 the results at the upper, middle and lower points can be taken into account. But one of these points has to be chosen when to make transition to tolerance calculations. For now, the deviatons at the middle point are thought to be considered. However, to take the biggest deviation into consideration can be logic, too. The deviation results are as follows:
Fig. 9. The STL format of the manufactured part in GOM Inspect
34
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
Sequence 1:
Sequence 2:
Blue Part: The scanned STL format of the singular Part 1 Grey Part: The assembled part group Red Region: The movement of the tip Fig. 10. The tip of Part 1 narrows to the other part.
Fig. 12. The illustration of welding sequences
Fig. 13. The formed nugget The process parameters were exaggerated. So, the peel test has been applied to control the nugget. It has been observed that a stabile nugget (Fig. 13) has generated and its diameter is 3 mm. Because of that two simulations with 3 mm and 5 mm spot diameter have been applied. Fig. 11. The illustration of sheet surfaces 3. 2 An Example for Spot Welding Analysis: In the spot welding analysis the software Weld Planner is used. This software belongs to the firm ESI Group. However the software Sysweld which considers the phase transformations could be used but Weld Planner has been preferred because its properties are enough for this study’s aims. Weld Planner works according to the Local Shrinkage Analysis. It is based on the shrinkage following the thermal expansion after reaching the melting temperature. Jackson et al. (2009) sustained that its biggest advantage is the very short calculation time when compared with Sysweld and its results are satisfactory especially if the welding sequence and clamping locations have to be investigated.
3. 2. 1 The results of analyses in Weld Planner showing two effects on distortions after spot welding 1. The number of Points (with the same welding sequence) The translational displacement at middle points (on Part 1) (according to Fig. 15): (Spot Diameter: 5 mm) Assembly group 3,4 and 5: 0.01932 mm (Fig. 14) Assembly group 6,7 and 8: 0.01489 mm (Spot Diameter: 3 mm) Assembly group 3,4 and 5: 0.008501mm Assembly group 6,7 and 8: 0.005267 mm
The middle surface models of manufactured parts have been created in SOLIDWORKS. After that, the appropriate mesh has been created in VISUAL-MESH. Especially for the locations around the spot weld points a precise layer mesh has been created. Before that, the spot welding points have been assigned in the interspace software VISUAL-WELD. Then, this example has been imported into the software WELD PLANNER and the distortions after the spot welding have been obtained. The welding sequences and the number of lines consisting welding points have been changed to see their effects.
Fig. 14. An Analysis Example in Weld Planner for the sequence 1 and number of weld points according to the assembly groups 3,4 and 5
Cem Yurci et al. / IFAC-PapersOnLine 49-29 (2016) 030–035
35
The studies explained here are continuing. So, to see the effects of the clamping, welding sequence and welding points on distortions various welding simulations will be implemented. Besides, the parts manufactured with different width and material will be inspected and their results will be interpreted, too. Also this study will be broadened. REFERENCES
Fig. 15. The model for extracting the displacement at the middle point 2. The welding sequence (according to Fig. 12) (with the same number point) The translational displacement at middle points (on Part 1): Sequence 1: 0.01932 mm (Spot Diameter: 5 mm) Sequence 2: 0.01938 mm Sequence 1: 0.008501 mm (Spot Diameter: 3 mm) Sequence 2: 0.007823 mm 4. CONCLUSIONS This was an initial study to create a beginning methodology for sheet metal part manufacturers. Because of that the experiments detailed here were not conducted according to the rules of Design of Experiments. Also the scanning and aligning process can be bounded to a specific methodology, too (an example by Hammett et al. (2009)). At the beginning it was aimed to use the mathematical formulas above to make transition between the distortions (deviations) after spot welding and the cumulative tolerances. But these formulas can not be used for our model. Because the tip of the parts (the end conjuction points) do not move together. They are opening like a rose. The scanning results and the welding simulations confirmed that. If the results are inspected, also there is a very big difference between the scanning and simulation result. It can be that the scanning results have been affected by many factors and the measurement accuracies might be unsatisfactory. It has been seen, too that the RPS aligning gave better results than Best Fit aligning. Also there is a clear difference between the specimens, too. The distortion values in the simulations are very small because only free clamps have been applied here. The distortions are being affected in a big magnitude by clamps. According to the welding simulations the distortion at the middle point increased when the welding point line increased. The welding sequence did not affect so much on the distortion. But these observations will vary according to the form and construction of sheet metal parts. Besides, as it has been foreseen, the distortions increase when the spot diameters increase.
Dahlström S., Lindkvist, L., Söderberg, R. (2007). Practical Implications in Tolerance Analysis of Sheet Metal Assemblies: Experiences from an Automative Application. Models for Computer Aided Tolerancing in Design and Manufacturing, 311-320. DIN EN ISO 17677-1 Standard. (2010). Widerstandsschweissen – Begriffe – Teil 1: Punkt-, Buckel- und Rollennahtschweissen. EN ISO 1101 Standards. (2012). Geometrical Product Specifications (GPS) – Geometrical Tolerancing – Tolerances of Form, Orientation, Location and Run-out. EN ISO 13920 Standards. (1996). General Tolerances for Welded Constructions. Fischer, B. R. (2004). Mechanical Tolerance Stackup and Analysis, 61-107. Marcel Dekker, New York. Hammett P. D., Guzman L. M. G., Geddes S. W., Walsh P. T. (2009). Quantifying Alignment Effects in 3D Coordinate Measurement. University of Michigan Transportation Research Institute, Manufacturing Validation Solutions, LLC, 22. Jackson, K., Darlington, R. (2009). Advanced Engineering Methods for Assessing Welding Distortion in AeroEngine Assemblies. IOP Publishing. Moos, S., Vezzetti E. (2012). Complaint Assembly Tolerance Analysis: Guidelines to Formalize the Resistance Spot Welding Plasticity Effects. Int J Adv Manuf Technol, 61, 503-518. Moos, S., Vezzetti E. (2015). Resistance Spot Welding Process Simulation for Variational Analysis on Compliant Assemblies. Journal of Manufacturing Systems., Vol. 12, 44-71. Nied, H. A., (1983). The Finite Element Modeling of the Resistance Spot Welding Process. Welding Research Supplement, 123-132. Renault – Tolerance Chain Education Lecture Notes. (2009). Shaoyun, C., Zhongqin, L., Yizhu, Z., Yongbing, L. (2006). A Parametric Study of Sheet Metal Joints for Dimensional Integrity. Int J Adv Manuf Technol, 29, 446-452. Sun, X., Dong, P. (2000). Analysis of Aluminium Resistance Spot Welding Processes Using Coupled Finite Element Procedures. Welding Research Supplement, 215-221. Thakur, A. G., Rasane, A. R., Nandedkar V. M. (2010). Finite Element Analysis of Resistance Spot Welding to study Nugget Formation. International Journal of Applied Engineering Research, Dindigul, Vol. 1, no. 3, 483-490. Thimm, G., Rui, W., Yongsheng, M. (2006). Tolerance Transfer in Sheet Metal Forming. International Journal of Production Research, Vol. 2, no. 5, 3289-3309.