PROGRAMUL CEEX CONTRBUTII STIINTIFICE IN TRANSPORTURI DFM SOLUTIONS FOR TOMBSTONING IN VAPOUR PHASE SOLDERING TECHNOLOGY BASED ON THE RESEARCH PROJECT Contract CEEX-X2C09 I. Plotog*, G. Varzaru**, C. Turcu**, T. C. Cucu*, N. D. Codreanu* *Center for Technological Electronics and Interconnection Techniques (UPB-CETTI), POLITEHNICA University of Bucharest, Spl. Independentei 313, 060042-Bucharest, Romania, Phone: +40 21 3169633 E-mail: norocel.codreanu@cetti.ro; ** Intrarom S.A, 17 Fabrica de Glucoza St., 020331-Bucharest, Romania, Phone: +40 21 2040702 Abstract The physics and chemistry of lead-free reflow soldering process phenomena’s revealed the interdependency between the solder paste, the component, the printed circuit board characteristics and reflow parameters, especially temperature versus time oven profile, going from the ambient value, approximately 25ºC, to over melting points, in order to assure liquid status of the solder paste. The paper presents the applied research performed in order to increase the quality of Lead-Free (LF) reflow/double-reflow VPS, by ruling out especially tombstoning. Key words: VPS, vapour phase technology, lead-free alloy, tombstoning, SMT soldering process 1. INTRODUCTION The goal of this paper is to determine the general requirements for PCBs and stencils design when using the „Vapour Phase Soldering” (VPS) technology in SMT reflow/double reflow process and to optimize VPS process parameters. The first experiments had shown an unacceptably large number of a specific defect, tombstoning (fig. 1), the partial or complete lifting of passive SMT components during reflow, which is the phenomenon whereby small chip components, which are being soldered on SMT pads at opposite or polar ends, suddenly flip up vertically; sometimes the lifting is partial, and sometimes the chips are staying on one end like tombstones in a cemetery, on one of the pads. a. b. Fig. 1a, b. Tombstoning after VPS 2. PHENOMENON ANALYSIS The tombstoning phenomenon is generally determined by an imbalance of the forces which take action over the composition when the solder paste laid-down in the form of paste is in a liquid state. An analysis of the forces (Fig. 2) shows that the surface tension of the solder will produce a turning moment. The condition under which the turning moment will be large enough to lift a component is: F1 + F2 < F3 (1) Where, F1 = vertical component of chip weight; F2 = vertical component of surface tension force acting under the chip; F3 = vertical component of surface tension force acting on the top of chip. 159 UCP AMTRANS - NOIEMBRIE 2007 PROGRAMUL CEEX CONTRBUTII STIINTIFICE IN TRANSPORTURI Detailing, (Fig. 3) with: m = chip mass; γ = surface tension of molten solder paste; H = thickness of chip; T = soldering zone of chip: Mg[(H+L)/2]cos(α+θ) + σT cos(α/2)< σ H cos(α+φ) (2) Fig. 2. The turning moment condition Fig. 3. Unbalancing forces in liquid phase Normally, the surface tension forces F2 and F3 take action in both ends of the chip component with the same values in opposite directions. All the surface tension forces are in equilibrium. F1, the chip weight in this situation, presses the chip on pads (Fig. 4) Fig. 4. Forces equilibrium The tomstoning phenomenon appears when the surface tension forces from the ends of the chip component are unbalanced by different causes and once lifting starts, the forces that cause the tombstoning effect will increase. Occurrence of phenomenon increases for smaller components only the solder paste from one end of the chip component becomes liquid passing through molten phase. The risk of tombstoning appearance is very high, especially for smaller chips. The thermal mass of the solder joints and for beginning of the process, the thermal mass of solder paste at either end of the passive component have a direct effect on the cause of tombstoning. The key for good quality reflow soldering process will be a result of the following congruently factors: solder paste characteristics, pcb design, components tolerances, parameters and tolerances of SMT reflow process in all phases. 3. EXPERIMENTS During the experiment, the authors have used NCAB type test PCBs, with the following classification: 1. (Chem. Ni/Au) Electroless Nickel Immersion Gold / 2. (Leadfree HASL) Hot Air Surface Leveling) / 3. (OSP) Organic Solderability Preservative / 4. (Chem. Tin) Immersion Tin / (Chem. Ag) Immersion Silver. Because the tombstoning affects especially the small chip components, the package types of lead-free chip components used was 0402, 0603, 0805, 1206. For the experiment lead-free was used , no-clean solder paste OM338T, type 3, (25-45µm per IPC J-STD-005) from Coockson Electronics, selected before after special experiments. The references for the dates of case dimensions (Tab. 1 – IPC 782) and pads geometry (Tab. 2) are given in fig. 5a, b. Tab. 1. Case dimensions Fig. 5. a. Case dimensions, b. Pads geometry 160 UCP AMTRANS - NOIEMBRIE 2007 PROGRAMUL CEEX CONTRBUTII STIINTIFICE IN TRANSPORTURI According to geometrical references for cases and pads, the authors take into consideration the following hypothesis for stencils apertures design (Tab. 2): - apertures dimensions identical with corresponding pads dimensions; with axial replacement to assure Smin (Ya4, Xa4); increasing 20% axial and 10% radial (Ya1, Xa1); increasing 10% axial and 20% radial (Ya3, Xa3,). axial reducing (Ya1, Xa1), according with: Xa1 = Wmax, Ya1 = Tmax+2(Cp-Lmin) (3) In order to assure different volume of solder paste two stencils were made with different thickness 0.125mm and 0.150mm according with apertures from Tab. 4. The authors also take into consideration one classical solution for stencils, 0.150mm thickness with 10% reducing apertures dimensions compared to pad dimensions and one stencil 0.125mm without reduction. Tab. 2. Stencil apertures dimensions For reflow soldering VPS machine type SLC504 from IBL company was used. 4. RESULTS The solder joints were analyzed according to the IPC-A-610 D. The solder joints grown from solder paste deposits follow out stencils apertures type 1 - 4 shown in fig. 6c, d. The results about tombstoning after thermal profile optimization (Fig. 7) are presented in Tab. 3 and Tab. 4. a. b. c. d. Fig. 6a-d. Solder joints follow up apertures type The apertures with tombstoning were type 2 for 0.125 stencil (Fig. 6a-0603 case, OSP finished) and for 0.150 stencil, type 1 (Fig. 6b-0402 case, OSP finished, 6d-0402 case, NiAu finished), also type 2 for NiAu , 0603 case and 3, 4, for chemical tin. Fig. 7. VPS thermal profile For NiAu pads finished, cases type 0402, 0.150mm stencil, the number of tombstoning was bigger, but at the limits, with a small angle lifting. It is important that NiAu for 0.125 stencil, chemical silver and HASL for both stencils have not tombstoning. The solder joints resulting after stencils apertures type 2 and 3 present to much solder. In connection to the tombstoning results, the conclusion is that 2 and 3 apertures type especially on 0.150mm stencil, NiAu pads finished and case smaller than 0805, must be avoided. The parameters of VPS process are position on levels, period on position and heating power. The SLC504 machine offers 20 levels with 0 – 100% power 161 UCP AMTRANS - NOIEMBRIE 2007 PROGRAMUL CEEX CONTRBUTII STIINTIFICE IN TRANSPORTURI range. In order to obtain the optimum thermal profile, the authors used temperature sensors to measure temperatures distributions on pcb surface. The result emphasizes differences between temperatures measured on PCB surface and machine reference temperature. Tab. 3. Tombstoning for 0.125 Stencil Printing Tab. 4. Tombstoning for 0.150 Stencil Printing 5. CONCLUSIONS The quality analysis and tombstoning appearances situations demonstrated that the type 2 and 3 apertures assure solder paste on pads in excess. As upshot the wetting angle of solder joints are close to 90°, the limit of quality acceptance. Also, the solder paste in excess offers condition for spread solder balls on pcb surface. The theoretical analysis and practical results show the importance of solder paste deposit’s axial dimension compared to chip component axial dimension as a tombstoning factor. Based on the practical results of experiments, the generally accepted criteria for a chip, component solder adequacy, the joint should extend up the component termination for a minimum of one-third of the component height, or 0.4 mm, whichever is less, the tolerance of chip components cases, the variability of pad dimensions, the authors are proposing the following requirements for stencil design regarding chip components assembling in vapour phase technology:  radial apertures dimension, Xa : Xa = W+H (4)  axial apertures dimension, Ya : Cp – Ya = S, Cp + Ya = L=T, Ya = (L+T) / 2 (5)  stencil thickness: 0.125mm As a result, the team defined the necessary criteria in order to obtain optimum thermal profile by measuring temperatures directly on pcb surface: preheating with 1.4 °C/s until 185-190°C and go to range of melting point with 0.66 °C/s until maximum temperature 230 °C. 6. REFERENCES [1] T. C. Cucu, N.D.Codreanu, I. Plotog, "Reflow process using lead free materials-basics and comparison with tin-lead process", Proceedings of the 2005 Interntional Symposium for Design Technology and Electronics Packaging (SIITME 2005), Cluj-Napoca, Romania, September 22-25, 2005, pp. 250 - 255. [2] I. Plotog, S. Jianu, C. Turcu, T. C. Cucu, N.D.Codreanu, “Multi-criterial Approach for Implementing of Lead-free Technology”, 4th European Microelectronics and Packaging Symposium, May 21-24, Terme Catez, Slovenia, pp. 301-306. [3] I Plotog, S. Jianu, C. Turcu, A. Stan, T.C.Cucu, N.D.Codreanu, DFM Concept for Wave Soldering Technology, SIITME 2006, Iasi, September 2006 Romania [4] Klein Wassink, Soldering in Electronics (2nd Edition), Electrochemical Publications, Port Erin, Isle of Man, Great Britain, 1989 [5] S. Jianu, I. Plotog, C.Turcu, A. Stan, T.C.Cucu, N.D.Codreanu, Reliability Aspects for optimal Implementation of Lead-free Technology, SIITME 2006, Iasi, September 2006 Romania [6] ***, Internet web sites related to DFM topics. [7] ***, www.ipc.org 162 UCP AMTRANS - NOIEMBRIE 2007