-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueBox Build
One trend is to add box build and final assembly to your product offering. In this issue, we explore the opportunities and risks of adding system assembly to your service portfolio.
IPC APEX EXPO 2024 Pre-show
This month’s issue devotes its pages to a comprehensive preview of the IPC APEX EXPO 2024 event. Whether your role is technical or business, if you're new-to-the-industry or seasoned veteran, you'll find value throughout this program.
Boost Your Sales
Every part of your business can be evaluated as a process, including your sales funnel. Optimizing your selling process requires a coordinated effort between marketing and sales. In this issue, industry experts in marketing and sales offer their best advice on how to boost your sales efforts.
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - smt007 Magazine
Inside Look: SMART Group's 26th Birthday Conference, Day II
October 13, 2010 | Pete Starkey, I-Connect007Estimated reading time: 11 minutes
Editor's Note: To read Part I of this article from European Technical Editor Pete Starkey, click here. It was a bright and sunny morning for the second day of the SMART Group's twenty-sixth Birthday Conference. The early golfers at the Oxfordshire Club in Thame, UK, looked considerably less bedraggled than the day before as they trundled off to the tees. Meanwhile, a keen and well-rested audience of electronics manufacturing specialists returned to the conference room for their further education and amusement. Whereas on Day One, delegates had enjoyed the experience of Rockwell Collins' Doug Pauls and Dave Hillman making solo presentations, the Day Two programme commenced with the "Doug and Dave Roadshow" performing their unique double-act. Introduced by SMART Technical Committee Chairman Sue Knight, they began on a jocular note with a short video indicating the fate that might befall a cell-phone ringing whilst they were presenting--scary!Turning to the serious business of cleaning and cleanliness specifications, Pauls and Hillman engaged, educated and entertained delegates with a review and analysis of IPC-A-600E, J-STD-001E, IPC-5704 and IPC-9202/3.IPC-A-610E gave inspection guidelines on visual cleanliness and, in essence, its messages were: Understand your materials and how they should be processed, understand your process and measure it and to be aware that certain materials and process interactions could result in problems. A-610-E had very good illustrations of the visual differences between tin-lead and lead-free joints and the training challenge was to recalibrate operators' eyes and experience. Pauls' and Hillman's best advice was not to go looking too closely for trouble. Magnification of 4x was the maximum recommended for routine inspection and 10x for referee inspection. "If you look close enough, you'll always find something that looks wrong!"--quoting examples of certain customers using high-magnification microscopes to detect signs of "contamination." Every production facility should have a standard, established for example by ionic contamination and SIR testing, which defined how much of each contaminant could be tolerated.J-STD-001E set standards for systematic cleanliness and cleanliness verification. Section 8 on cleaning process requirements was primarily concerned with post-solder cleanliness. There was a vagueness about the statement that "an item required to be cleaned shall be cleaned by a documented process to allow removal..." in that it did not define by whom--the standard or the customer? Theoretically, a no-clean assembly did not require to be cleaned, therefore J-STD-001E was not applicable. Because there had been some concern that the standard focused on the cleanliness of the whole assembly, such that localised residues might be averaged out over a large surface area, IPC's Ionic Residues Task Group was investigating localised extraction and evaluation methodsHow clean is clean? The new IPC-5704Cleanliness Requirements for Unpopulated Printed Boards specification defined recommended requirements for the cleanliness of unpopulated single, double-sided and multi-layer printed boards, and provided specifications for maximum limits of ionic contamination, using ion chromatography testing. Ionic chromatography, using the test procedure defined in IPC-TM-650 2.3.28.2, was considered a much more precise test than the old resistivity-of-solvent-extract ROSE test, since it was ion-specific. IPC-5704 gave recommendations on cleanliness testing for both process control and product acceptance and OEMs that had used this guidance had reported a reduction of failures attributed to bare board cleanliness.IPC-9202 was the new material and process characterisation and qualification test protocol for cleanliness, intended as an international protocol accepted for both IPC-J-STD-001 and IEC-61189. IPC-9202 allowed the use of test kits for process investigations, with actual material sets being used for process qualification. Significantly, the process qualification vehicle was an assembly, IPC-B-52, considered more meaningful than the B-25 bare board coupon. IPC-B-52 had break-off coupons for surface insulation resistance and ion chromatography testing, and optional coupons for tests such as solder mask adhesion. IPC-9203 was an overall guide to what goes into an intelligent process investigation and qualification effort--Doug Pauls in a can!Asked what they considered to be the biggest challenge facing the industry today, Pauls and Hillman responded instantaneously: Ignorance! Conscious that so much long-standing knowledge and experience had been lost to the industry as the old-timers retired or passed away, they were great advocates of education. "If you think the cost of education is high, check-out the cost of ignorance."The audience replete with state-of-the-art knowledge on cleaning and cleanliness specifications, it was perhaps appropriate that the question should be asked: "Why do we clean when we have a no-clean process?" Specialist in cleaning solutions for the SMT industry, Dr. Ralph Hoeckle from Zestron Europe set out to put no-clean into realistic perspective: No-clean was a process and, like every technical process, it had limits. He defined "no-clean" as a manufacturing process needing a solder paste which left only inert or encapsulated residues, which would be guaranteed by a solid soldering profile with a reasonable process window. "No-clean" did not mean that the PCBs used were electrically clean, or that no-clean solder pastes would supply a no-clean result under all circumstances or that contaminants other than cased by soldering were not harmful.Dr. Hoeckle discussed the typical ingredients of a flux formulation: rosin or resin, activators, thixotropes and solvents. All but the solvent would probably remain after soldering, but the activator residue should be encapsulated by an inert resin or rosin layer. If this layer was cracked because of a non-ideal soldering profile or climatic stress, then the activator residue could escape and lead to corrosion issues. A characteristic of lead-free pastes was that they tended to incorporate more-aggressive activators A simple qualitative test, in which a chemical reagent turned blue on contact with organic acids, could be used to detect the presence of activator residues.Eight percent of field failures of electronic assemblies had been attributed to contamination. J-STD-001 set limits for ionic materials less than 1.5 micrograms per cm², surface insulation resistance greater than 8 x 108 Ohm, resin residues less than 40 micrograms per cm² and other contaminants nil. So why clean? Cleaning removed activators and, therefore, avoided the corrosion risk; cleaning removed resins and, therefore, avoided delamination of conformal coatings; and cleaning removed fingerprints and dust and, therefore, avoided risks through unknown contamination. Dr. Hoeckle reviewed various cleaning processes then made some cost-comparisons between cleaning and conformal coating. If no-clean assemblies were cleaned before conformal coating, the coating thickness could be reduced, and it was possible to arrive at a cost-neutral situation where cleaning costs were balanced by savings in conformal coating costs. He concluded that no-clean generally worked reliably for Class 1 assemblies, but that most corrosion risks could be avoided by cleaning, and that high-reliability assemblies needed case-by-case assessment.From cleaning issues to tin whiskers: Barrie Dunn, manufacturing technology expert from the Product Assurance and Safety Department of the European Space Agency gave a presentation entitled "Mitigation of Whisker Growths Associated with High Reliability Electronics and Space Hardware." The first "whiskers" were seen on space hardware in 1974, and whiskers had been responsible for shorting-out relays and causing equipment failure in ESA spacecraft in 1998. Dunn had been working for many years to characterise whiskers, of tin and other materials, to determine the mechanisms of whisker initiation and to establish means of whisker mitigation.A series of long-term experiments had been carried out using tin-plated C-ring test pieces, similar to those used for stress-corrosion testing. The C-ring allowed simultaneous observation of the plating under compressive, tensile and neutral stress conditions. Tin-plated brass, tin-plated brass with a copper barrier, tin-plated steel and tin-plated steel with a copper barrier were all studied, over a period of 15½ years. All showed tin whisker growth to some extent, although in certain instances it took over three years before spontaneous whisker growth began. There was some mystery surrounding the origin, within the structure of the coating, of the actual tin atoms feeding the propagation of the single-crystal of the whisker, since they did not appear to come from the area immediately under the base of the whisker. Control samples with fused tin plating had shown no evidence of whiskering after 15½ years.Tin plated finishes had been forbidden in ESA spacecraft for many years. There were currently six ECSS-ESA standards prohibiting pure tin: it was required to have greater than 3% alloying elements. And the preferred finish for PCBs was fused tin lead--several of ESA's board suppliers had hung on to their old hot-oil reflow lines to support the requirement.First presentation of the afternoon session came from Jon Anderson of Humiseal and explored the phenomenon of creep corrosion. His title was "Silver versus Sulphur - Can Conformal Coatings Help?" and he began by reviewing the history. Creep corrosion was not a new effect; it had been recognised by Bell Laboratories as early as 1955. But it had become topical as a consequence of the growing use of SAC solders and immersion silver PCB finishes, and was caused by sulphidation--the corrosion of elemental metal in the presence of sulphur compounds to form conductive pathways between conductors.A study of the effects of various types of conformal coating, including water based and UV curable acrylics, urethanes and silicones, on test coupons processed by different suppliers of immersion silver finishes had indicated that, under mixed-flow-gas testing conditions, all uncoated immersion silver finishes showed significant creep corrosion within 10 days, even though their surface insulation resistance values were reasonably good. Of the coated samples, only silicone and UV-cured coatings gave consistently higher SIR readings than uncoated samples. Water-based acrylic appeared least effective in preventing creep corrosion. Anderson emphasised the point, however, that creep corrosion is just one of many considerations to be taken into account when selecting a conformal coating.Modern electronic assemblies have to operate in ever more demanding environments: Marion Quarrington, New Products Introduction Manager with Cooper-MTL, manufacturers of monitoring, control and product-protection instrumentation for safety-critical processes, explained that the trend was to place the electronics in a working, rather than a control-room, environment, largely for reasons of cost. She went on to describe some of her practical experiences in designing and building reliable and cost-effective assemblies to work in such diverse situations as deck cargo on ships, subject to humidity, salt, low-frequency vibration and shock, industrial chemical plant with corrosive gases, cold, low humidity environments like gas and oil fields in Canada and Russia, where ambient temperatures could be -50°C, dry desert heat, humid tropical heat, explosive environments, etcetera.Each situation presented its own challenges in terms of materials selection: printed circuit boards, components, solders and fluxes and assembly process chemicals. Circuit protection strategies and electronic and enclosure design had to be carefully considered, as did strategies for meaningfully testing products to prove the effectiveness of the design. Complex environments were difficult to simulate and considerable reliance had to be placed on beta-site testing. Design and production engineering had to work concurrently to find the best product protection specification without impacting price, weight or functionality."Head-in-Pillow or Head-on-Pillow: That is the Question" from Andrew Clarke of AIM looked at the causes of solder joint defects in BGA assembly resulting from an incomplete coalescence of solder ball and paste. During reflow, even though both the ball and the solder paste had melted, the paste would wet the pad but not fully wet the ball, leading to a weak solder joint with enough of a connection to pass functional testing, but liable to subsequent failure with very little mechanical or thermal stress. The prevalence of head-in-pillow defects had increased with the conversion of BGA components to lead-free. To improve drop-resistance, many BGA manufacturers had switched from SAC 305 alloy to SAC 105, with dopants such as magnesium to modify grain structure. These alloys had been observed to exhibit variable wettability.Head-in-pillow defects could be co-planarity related, solderability related or a combination of both, and the sequence of events by which a defect was formed was believed to be that the ball, which had initially been in contact with the paste, was lifted clear by thermal distortion of the package as the assembly approached reflow temperature. Although upon further heating, the package might subsequently flatten-out and bring the ball back in contact with the paste, by this time it would have formed a surface oxide layer which the residual flux activity was insufficient to dissolve. Because of the unpredictability of component warpage, investigations had been directed at interactions of process variables. The two most significant factors had been found to be solder paste flux chemistry and the wettability of the BGA alloy. Surface structure of the ball alloy was believed to be the most significant factor in determining its wettability. Balls with a matte surface, resulting from the presence of tin-silver and tin-copper intermetallics, were more difficult to wet. New solder paste chemistries had been developed which offered greater thermal stability over a wide range and these had been shown to help eliminate head in pillow defects by providing sufficient flux activity to reduce surface oxides.The final presentation of the conference came from Calum Drummond of Retronix. His presentation was "Component Refinishing/Retinning: Good Practice and Reliable?" Against a background of concern over cost structures, scrap reduction and stock utilisation, there was a growing need in the industry to use components that had either an uncertain past in terms of traceability and storage, or were excess inventory or old, slow-moving stock. Even if the electronic functionality of the components was perfectly acceptable, problems were likely to arise because of poor solderability.Drummond showed many examples of de-wetting, caused by contamination, oxidation, plating issues and poor de-golding and explained how solderability could be improved by stripping and re-tinning. The method could alternatively be used to convert tin-lead components to lead-free. At Retronix, every process step was carried out in line with recognised standards: baking to J-STD-033, stripping and re-tinning to GEIA-STD-006 or STD-0015 draft, cleaning to J-STD-0075 and packing to J-STD-033. Solderability was tested to J-STD-002. BGA re-balling was another of their specialities and was performed to similar standards. If stripping and re-tinning was done properly, the solderability of components would be restored without impacting their reliability.The proceedings were concluded, as the day before, with an energetic "Question Time" session, this time with Sue Knight as moderator. Once more it was clear from the questions asked that the conference programme had stimulated plenty of interest. Everyone stayed to the end and when Sue Knight called time, no-one was in a rush to leave.An excellent event all round!