Institute of Circuit Technology
39th Annual Symposium.
Gaydon, UK, th June 2013
The Institute of Circuit Technology Annual Symposium has become a must-attend conference and networking event for the UK printed circuit industry.
Technical Director Bill Wilkie had chosen a splendid venue for the ICT’s 39th Annual Symposium – the Heritage Motor Centre in Gaydon, Warwickshire, in the Midlands of England, home to the world’s largest collection of British motor cars from the classic, vintage and veteran eras.
The UK PCB industry was remarkably well represented, with an impressive collection of fabricators and suppliers (including a few personalities from the industry’s classic, vintage and veteran eras…) gathered together to absorb up-to-date knowledge from eminent presenters, to renew acquaintances and to share information, gossip and opinion with their peers.
ICT Chairman Professor Martin Goosey welcomed the gathering and remarked upon the continuing growth of the Institute’s membership, now well into the 300’s, and the wide and increasing range of services to members: annual symposium, evening seminars, foundation course, technical journal and website as well as collaboration in research projects such as ASPIS, coordinated by ICT and now in its final stages.
Professor Goosey introduced a well-chosen programme of technical presentations, commencing with an up-to-the-minute review of PCB substrate materials for thermal management from Ventec Europe Technical Support Manager Ian Mayoh. He explained that, besides general trends to miniaturisation, higher density and embedded technologies in multilayer circuits, a major driver was the continuing growth in high-power LED lighting applications. Insulated metal substrates (IMS) enabled the reduction of system costs through the reduction in size or elimination of cooling fans and heat sinks. Compared with standard FR4 laminate, which had a thermal conductivity around 0.25 Watts per metre Kelvin (W/mK), IMS materials were currently available with thermal conductivities between 1 and 5 W/mK, and 7 W/mK materials were in development. Mayoh stressed the importance of understanding that the actual thermal impedance of a material, which was a function both of its thermal conductivity and its thickness, was the meaningful practical consideration in determining the suitability of an IMS for a particular application, and a cost-effective solution could often be found without having to use the highest thermal conductivity (and most expensive) material. He discussed the relative attributes of glass-reinforced and non-reinforced materials in terms of thermal and mechanical performance, formability and cost/reliability trade-offs, and advised against specifying a material purely on data-sheet information without carrying out proper trials.
Next to speak was Stuart Hayton, Sales and Marketing Director for Mutracx, with a thought-provoking discussion of the dilemma faced by companies choosing whether to pursue “sustaining” or “disruptive” technology routes, with some sobering examples of big names with successful histories who had concentrated on putting too much emphasis on customers’ current needs, had failed to adopt new technologies or business models that would meet customers’ unstated or future needs and had lost their leading position as a result. Quoting Clayton Christensen, he commented that in their efforts to provide better products than their competitors and earn higher prices and margins, suppliers often “overshot” their market and gave customers more than they needed or were ultimately willing to pay for. Hayton illustrated this with his own mobile phone, which he said had many more functions than he ever wanted and some which he had not even know existed. For established companies it was easier to rationalise the decision not to invest aggressively in disruptive technologies because these tended to have a lower initial profit margin for the supplier than contemporary technology and were often conceived and commercialised in different market areas. Indeed, in most cases failure could be attributed to classical “good management”.
A specific case in point was Kodak’s failure to realise the opportunity presented by digital photography, and to cling to film whilst their technology was effectively obsoleted by their digital competitors. And film had a particular relevance to PCB imaging – Hayton described how the technology had progressed from hand-laid artwork and process cameras, through vector and laser plotters to laser direct imaging (LDI). He did not consider LDI to be a truly disruptive technology, since it still involved the wasteful use of photoresist and its associated processing. And although LDI had been commercially available for over 15 years, it still only represented 5% of the world PCB imaging market, the remaining 95% relying on silver halide film. The real disruptive technology presently emerging was ink-jet primary imaging, which eliminated 11 of 15 process steps in the imaging of inner layers, and only placed resist where it was actually needed. The technology itself was not new – it had been developed to a high level in the graphic arts industry – but it brought a new value proposition to the printed circuit industry and offered the capability for inner layers be ready for etching within 5 minutes of the image data being output from the CAM station.
The ASPIS project, mentioned by Professor Martin Goosey in his introduction, was in its final stages and Professor Karl Ryder described how his research group at University of Leicester had evaluated immersion gold processes based on ionic-liquid chemistry as a potential means of overcoming the hyper-corrosion effects observed when aqueous gold chemistries were used in the ENIG process. A series of immersion gold formulations had been prepared using Ethaline 200, an ionic liquid composed of ethylene glycol and choline chloride in 2:1 molar ratio, and gold in the form of gold chloride, gold cyanide and potassium cyanoaurate , and these had been used to deposit gold on a standard electroless nickel surface. Quartz crystal microbalance techniques had been used to study plating rates, scanning electron microscopy and atomic force microscopy to study surface morphology, and wetting balance measurements to test solderability. The roughness of deposits from ionic liquids was less than the equivalent from aqueous solutions, and there was less evidence of corrosion of the electroless nickel. Soldering tests had indicated that coatings from ionic liquids wetted faster and more reliably than those from aqueous processes. Spun out of the ASPIS work, a current project supported by IeMRC was exploring applications of ionic liquids as soldering fluxes, where their ability to readily dissolve metal oxides without the use of acids potentially offered significant advantages.
Another IeMRC-funded project was concerned with the functionalisation of copper nanoparticles for applications in electronics manufacturing. Dr John Graves described work being carried out at the universities of Coventry and Loughborough together with industrial partners with the objective of substituting commercially-available nano-copper powder for tin-palladium catalysts in electroless plating processes. It was necessary to coat the nanoparticles to improve dispersion stability, inhibit oxidation and to promote adhesion to surfaces. The chemistry of self-assembled monolayers (SAM), of which there was in-depth experience at Loughborough, was being evaluated as a means of achieving an appropriate functional coating, and remarkably good initial results had been observed in the catalysation of glass and polymer surfaces. A particular challenge was the successful dispersion and de-agglomeration of the SAM-functionalised nano-powder, and Dr Graves demonstrated how Coventry University’s expertise in sonochemistry had helped overcome this problem using high-power ultrasonics,
From ultrasonics to megasonics, a related technique mainly used in the silicon industry, which operated in a higher frequency range (typically 0.8–2 MHz compared to 20-200 kHz for ultrasonics) and gave less-aggressive cavitation effects. Dennis Price from Merlin Circuit Technology reported his work with Heriot-Watt and Greenwich universities on the ASPECT project, studying the effects of megasonic agitation in the fabrication of high-aspect-ratio blind microvias. Their objective was to reduce the number of sequential bonding operations required to build complex multilayer microvia PCBs, by carrying out a single bonding operation then controlled-depth mechanical drilling to various levels to achieve interconnection to several layers in a single plating operation, rather than through a series of drill, plate, and build-up stages. The metallisation and electroplating of high-aspect-ratio blind holes presented particular challenges and conventional agitation could not guarantee consistent results. Ingenious methods had been developed for modelling, observing and measuring solution flow in and around via holes, and megasonic agitation had been demonstrated to facilitate the break-up and extraction of entrapped air bubbles. On a micro scale, acoustic streaming could give an enhancement effect in through vias although its influence was small compared with diffusion. Improved performance was achieved due to the high level of bulk concentration next to the mouth of the via, supported by acoustic streaming along the board. Price concluded that, within microvias, megasonic agitation gave improved ion transport, mainly due to the presence of resonant bubbles rather than by acoustic streaming.
The focus moved from high-frequency sound waves to design considerations for high frequency signal integrity in the presentation by Martyn Gaudion, CEO of Polar Instruments, who began by saying “Don’t always trust the data sheet” when considering impedance and loss. Regardless of whether impedance was modelled or measured, a multiplicity of factors needed to be taken into account when frequencies higher than 100MHz were involved, and results were likely to be inaccurate unless specialised 2D field-solver programmes were used for the calculations. A range of loss mechanisms came into play as data-rates moved into the multi-GHz range. Skin depth became a serious consideration and the surface roughness of copper conductors, particularly that resulting from bonding treatments, became a crucial factor. Solder mask could have a dramatic effect on impedance, and this could be further influenced by moisture absorption. It was preferred to route critical traces on inner, rather than outer layers. Even then, the natural dielectric inhomogeneity of woven-glass reinforced laminates was a limitation, although this could be mitigated to an extent by non-orthogonal routing. In general, losses could be reduced by the use of short traces, as wide as possible, and the smoothest copper consistent with adequate adhesion to resin. Gaudion discussed how to minimise crosstalk and explained mode conversion and field distribution in differential pairs. Finally, he demonstrated how to properly document modelled results into meaningful stack-up specifications and technical reports.
Back to practical aspects of PCB manufacture, as Chris Serre, Managing Director of Union Tool, described current technology and latest developments in drilling and routing tools, and discussed the technical and functional details of tool geometry, materials and coatings. Another advocate of being careful what to believe in data sheet statements, he listed the attributes and benefits of single-flute drills, which were capable of better registration accuracy, higher capacity for swarf removal and longer life than their conventional 2-flute counterparts and certain “parallel-flute” offerings. Furthermore, single-flute drill bits could be re-pointed on existing machinery. Proprietary lubricant coatings further assisted in swarf evacuation, and Serre illustrated their effect using high-speed video. Taking the example of a 0.1mm single-flute drill, an uncoated tool was capable of 6000 hits and one re-point, giving an effective total of 12,000 hits, whereas the coated equivalent could be used for 16,000 hits before re-point and re-pointed twice, giving an effective total of 48,000 hits. Flute design was not the only physical consideration: the drill shank was a critical component in its own right. Traditionally, tools had been made from solid tungsten carbide. New composite tools were becoming popular, with the shank made from stainless steel. This offered not only a significant material cost benefit over tungsten carbide, but was better at absorbing spindle vibrations and run-out. Diamond-based wear-resistant coatings on routing tools gave significant improvements in life and resistance to breakage, and with the increasing requirement to machine aluminium-backed IMS substrates, lubricant coatings substantially reduced the tendency to clogging of tools with aluminium.
When the symposium drew to a close, and Bill Wilkie had thanked presenters for their contributions, delegates for their attention, and Ventec Europe for their generous sponsorship, there seemed a certain reluctance for attendees to rush off home – many took their time to browse the spectacular selection of museum exhibits on their way out of the conference area. A very satisfying and informative day all round.
Pete Starkey. I-Connect007. June 2013
Author’s note: I am grateful to Alun Morgan for kindly allowing me to use his photographs.
Members can download the presentations here: Symposium Papers