columns | June 21, 2005

evertiq at the EIPC Summer Conference - Day 1

Paul Waldner maintains that conferences are about meeting people as much as they are about learning from people, so he got everyone on their feet to say hello to one another.
Paul talked about the success of the recent Technology Trip to China for the CPCA show, and now the EIPC are doing the same thing in Hong Kong in conjunction with the HKPCA at the beginning of December. In October, 27th and 28th, there is to be a joint IPC/EIPC TMRC in Berlin, and a Call for Papers has just been issued. He reminded everyone that EIPC has its bi-annual General Meeting coming up which will be held during Productronica, and a new board of directors will be elected, as well as a new Chairman. All members are most welcome. His recent trips to Korea and Japan showed that the technology there is outstanding, with FPC in a whole class of its own, when it comes to equipment and technology. The world is now flat, says Paul, when it comes to competitiveness; Europe no longer has the advantage; the use of call centres and other information tools makes it all possible for everyone to do anything, anywhere. That includes us all.

Walt Custer looked at the present business conditions, including the supply chain effect, and the shift to Asia; difficult times for us all. Sure, there is growth in electronic equipment, manufacturing has moved up from $1218 billion dollars in 2004 to $1312 billion in 2005; Europe makes 19% of world electronic equipment, but the situation in the States is not so good.

Happily inventory levels are now down to where they should be, with no huge stocks anywhere. The period 2005-2006 shows a slowing but not a disaster. In the telecom sector the top six companies have shown 12% growth over 2003 to 2004, Ericsson up 12% but margins slim, and the cell' phone market will be some 750 million units this year. In market share, Nokia enjoy 30% and continue to be aggressive, Motorola have 16.8%, and they are the two giants. He showed us how Skype works, immediate free telephone access to your laptop. Wireless is getting faster, now there are air cards; Asia has about 45% of the world's chip consumption, and the assembly market shifting worldwide with Asia growth in this sector of, again, 45%. He mentioned Elcoteq
in Finland as an EMS company that is going places.

Of the 3000 PCB companies in the world, 1000 are in China, 400 in Europe, and 500 in the USA. Japan currently running flat, but of course China is showing linear growth. However, less than 10% of companies in China are indigenous, the remainder are all there through foreign investment. Europe PCB production is led by companies such as AT&S, Aspocomp, and Eltek. Capital expenditure in the PCB industry will be negative for the next year, but interestingly Central Europe is emerging as a major EMS sector.

Mats Ehlin came over from Du Pont PCM in the USA, but as he was born in Stockholm he felt quite at home. He wanted to discuss dry film photoresist technology for profitable PCB production, and began by looking at the value of European PCB production. What are the value drivers? They include quick turnaround service, local military/defence, medical electronics, a need for high product mix, demand for local production, some design involvement, and applications with advanced intellectual property.

Competing with China is not a recipe for success, to do this you have to be extremely capable. These attributes include production flexibility combined with a broad capability, a short set-up time, real processing, 'zero' defects, a high resolution capability, use of thin materials, sound possess technology know-how, and an ability to make strong personal relationships and instil brand loyalty. He explained the QFD - quality function deployment - process where customers determine the relative importance of attributes, low cost and high yield being the outcome. Where should the focus of your offer be? Direct imaging resists are technologies that offer easy start-stop capability, and for the highest possible yields.

LDI is a reality; there are now100 systems operating worldwide at the leading companies. LDI offers production flexibility with yield improvement, and their Riston® LDI500 is a product they have brought in to meet the demands for maximum benefit. Mats described the semi-additive process, where15mm lines and spaces after plating are obtainable with an accuracy of ± 1mm. Customers claim a 5% reduction in yield loss with their products.

Stephen Street works for UL International in the UK, Environmental legislation world-wide is increasing, with WEEE, RoHS and ELV in Europe, the CA Prop 65b in the USA, the Asia Pacific Green programs, and corporate social responsibility. The power of green consumerism should not be underestimated. It all impacts upon information technology, telecommunications, consumer electronics, and domestic appliances. If your products are used in these applications then you need to know about the RoHS directive, which he went on to explain.

For the record the limits are:
Lead - MCV - 1000ppm
Mercury - MCV -1000ppm
Cadmium - MCV - 100ppm
Hexavalent Chromium - MCV - 1000ppm
PBB - MCV - 1000ppm
PBDE - MCV - 1000ppm

Homogenous materials are termed as materials which cannot be mechanically disjointed into different materials. As to who enforces it all, this should be announced shortly. It might be your competitors! He described test methods to detect restricted substances, and took a look at UL796, a test that, for a new laminate, takes a year and costs $40,000.

The new designation for FR4-like materials is called IPC4101A, and the proposed groups are being classified as (a) brominated epoxy laminates, (b) non-brominated epoxy laminates, and (c) epoxy blend laminates. The UL Restricted Substance Compliance Solution acts as type of buffer between the suppliers and the customer and the regulator and is where UL offers risk management, compliance and cost avoidance. This is a low cost single programme to meets the needs of producers and eliminates the need for suppliers to participate in many test programmes. So begin thinking about making your products greener; use your environmentally friendly attitude as a selling tool, and always keep an eye open for new technologies.

Thomas Sutter is the R&D director at Rohm & Haas Electronic Materials, and presented his thoughts on advances in inkjet materials. R&H has several digital imaging programmes running, and are active in the graphic arts field. They have been busy looking at solder mask inkjet applications as well as etch resist. With solder mask they need fine pitch solder pads need to be defined, and to explain how this is important he made a comparison between hard tooling solutions against inkjet application of soldermask.

R&H have taken a different approach, with a hot melt water-dispersible process, and he explained how it works. The circuit board is pre-cleaned, then coated uniformly with solder mask and pre-dried. The vision system then captures the fiducials on the board, and coveys this data to the inkjet head which will then scan the board and apply U/V blocking solder mask to the desired SM openings. The board is then exposed and developed, so that you have a finished board with clean SM openings or pads.

The advantages are that quite conventional coating can be jettable, and there is no need to re-qualify a 'special' resist. The system is compatible with dry film or liquid solder mask. The Inkjet UV blocking mask has variable drop volume and shape, and the hot melt technology gives you great versatility. He showed the lab scale Spectra Galaxy print head, which had defined 500mm pad width, with 200mm dams. There is no image distortion under vacuum pressure in a PC printer; the first prototype will be ready for testing later this month, with optimisation of the process during the third quarter this year. They will therefore be demonstrating this exciting new system at Productronica.

Harri Latonen came down from Aspocomp. What are the drivers and trends in OSP surface finishes? There are several: reliability; repeated mechanical wear, all devices run the risk of handling, impact etc. ergo solder joint failure. Then there is the environment, with RoHS; materials management, process management, and managing lead-free assembly, where solder joint reliability is critical. It's a complex situation with many different parameters, and surface finish options are many and varied. Most are good, some are limited, but none are perfect for all applications.

At Aspocomp they have been producing selective finishes for the last 5 years, and whilst lead-free soldering reflow profiles depend on the application, typically they are 20-40°C higher than in normal soldering. In considering material selection, the parameters of laminates have to be taken into account, as are the processing aspects. Are they compatible? Most often the answer is yes, but you do need to specify your requirements, and not rely just upon manufacturer's claims.

The thermal expansion of a laminate can impact upon component retention, and higher melting temperatures can cause higher Delta T. Reflow during the assembly processes can be as many as three times, all at higher temperatures, so much depends on application. At Aspocomp they use a copper-selective OSP process called Entek Cu 106A(X) HT. Field testing of soldered examples showed that up to 4 reflows could be performed without detriment. In conclusion he stated that both OSPs and NiAu can be used for reliable solder joints.

Katsumasa Suyama hails from the Espec Corporation in Japan, Well ahead of the European RoHS directive, they have been using SAC (Sn-Ag-Cu) solder alloy since 2001. However, aware of the difficulties placed upon assembly by the higher temperatures, they have been running tests with Tin-Zinc-Bismuth leadfree solder, on an OSP surface finish. In a detailed presentation, Mr. Suyama he concluded that in high temperature testing Tin-Zinc-Bismuth showed a lack of joint strength against SAC, with Sn-Zn-Bi resistance decreasing after 530 cycles, and failing after 650 cycles, whereas others passed 1000 cycles. SAC cracked after 1000 cycles, and SZB peeled at toe. It was Ni-Pd-Au that looked good for intermetallic compound, and is much better for solder joint strength.

Orbotech-Schuh GmbH sent along Harald Eppinger to tell us about how they test lead-free soldering with AOI. Reflection values can change with changes in process, brightness of solder joint etc. so they needed to create a 'model' which would give them a comparison between lead-containing and lead-free solder joints. Orbotech tested component presence, position of solder joints, and focused on BGAs on their test panel, being able to look under the BGA, In lead-containing and lead-free solder joint comparisons, solder joints looked more matt, were darker, and more grainy.

Lead-free soldering can be astonishingly good, with the differences in reflection due to Pb-Sn being only small. Recognising problems in advance. The differences between Pb-Sn and SAC are hardly visible, but tests have shown that with 3D AOI on reflow solder joints there are no problems, on wave soldering no problems, and with pick and place there are no problems, either. Their technology can live with the leadfree requirement.

Jurgen Friedrich of ERSA talked about lead-free flip-chip and chip scale packaging designs, which
require a new inspection technique. The tiny solder joints for CSPs always crack on the component side, not on board side. Why? Well, different materials have different thermal expansion characteristics, (CTE), with huge differences, up to 6 times that of the component that sits on the board. It is that CTE mismatch that leads to shear strain at corners of solder-joints.

So how does lead-free change anything? The process temperatures are higher; the lead-free solder joint solidifies at a higher temperature. Is there an alternative to a non-destructive test? Yes, happily there are now some new microscopes available on the market that allow inspection of bottom joint, and look up at the top joint, which is more important. Problems come with not the heating up, it's the cooling down, you get just as much stress during cooling as you do in heating,
so the risk of component side ball delamination will increase with a lead free process. To see is to survive.

As the saying goes, you want your customer to come back, not your boards.

The man from Nokia Technology Platforms, Finland came to tell us about how they manage data flow for PCB design. His name is Tero Kärkkäinen. They use Gerber RS274X or RS274D, Excellon NC, and DRC, (design rules check). For embedded resistors, they use thin film technology, such as Ohmega Ply, and thick film polymers. For embedded resistors they have to bear in mind space saving on the surface of the PCB, there are potentially better signal integrity costs savings (resistor density is the factor here) and improved application reliability, i.e. fewer solder joints. We are now seeing a new kind of detailed information for PCB manufacturing, which includes resistor value, resistor tolerance, material definition and power handling capabilities. Data flow automation is therefore needed to avoid human error, and whilst what we have now is an old format such as IPC-2581, happily this will lead to a new standard called IPC 2584, which although not ready yet, is needed before embedded resistor technology production can enter mass production. The way forward is for embedded capacitors, embedded active components, and
embedded optics, and all of these are setting up new challenges for design data infrastructure, so there is a need for better data transfer between OEMs and PCB manufacturers.

Martin Cotton is a designer, too. If he wasn't in circuits he should be in haute couture, such is his style. Sanmina-SCI has this concept about RFID. By replacing bar codes with RFID means that you are adding value to the chip, and thereby you remove price pressure from the system. If companies such as Flextronics are starting to look at RFID, then we should be looking at it too. If you bury an RFID into an item, it cannot be removed, which offers great security. This 'label' could be mounted on a PCB type laminate. There is no connection to the PCB; the PCB is the carrier. RFID can 'revolutionise the pipeline'.

By embedding the RFID system into the product, you can offer 'tracking' through the whole manufacturing process, through assembly sub-assembly, shipping, retail sales and consumption (purchase). The RFID chip can act as a sensor, and you can have it as an active, or passive component. RFID can be used during the product manufacturing process or through the product lifecycle. It can be used as a third party interrogator, using wireless. It changes the PCB supplier from a service provider to a product developer; it is both a production device, and a security. We are not talking labels; here we are talking PCB opportunity. RFID is part of a LAN matrix, like the Internet. Needless to say a lot of questions to a lot of interesting ideas.

Marcus Bärtele is the man from EADS who came to talk about embedded passives in BGA interposer structures for avionic applications. He described the differences between avionics equipment, which in its overall format demands include very high density for assembly, high data rates, minimum weight, operates in harsh environment, has to be highly reliable, and has to have a long life, more than 25 years. It will be maintained over long periods, it must include state of the art manufacturing technologies, there are a wide variety of them, with low volumes per lot sometimes. System considerations include thermal a management, manufacturing technologies, miniaturisation, embedded components in interposer PCBs.

It is with these BGA interposer structures that bow and twist is one of the problems that are encountered, Before balling it is stable between -50mm to +150mm, and the assembly process of double pack (IP and BGA) does not negatively influence the component. Mr. Bärtele explained the steps for the assembly process of a Double (Interposer) Pack, and it is by this means that they have created a balanced CTE between the BGA and the PCB. They have, of course, used lead-free solder between the two interposers

Frando van der Pas is the face of Enthone, and looked at the new reliable generation of PCB surface finishes for lead-free soldering applications, a subject about which he is obviously passionate. He stated that HASL really isn't a starter, and of the other processes the usage is currently as follows: Immersion Silver 63.3%, Immersion Silver 31.1%, ENIG only 3.5% and OSP 19.0%. He claims that Immersion silver and OSP will grow at HASL's expense, and HASL will decline, as it is a difficult alloy to control, and lacks the capability of providing a planer finish.

ENIG has many advantages, including a durable finish, with good shelflife. It is wire bondable, and easy to test, but it does have relatively high costs and is not reworkable, important in the lead-free solder field, but on the other hand it does demonstrate inconsistent solder joint strengths. Immersion tin is relatively low cost, good solderability, is compatible with lead and
lead-free solders. Tin whiskers are of course a major limitation, and horizontal process equipment is expensive.

Furthermore, the process is difficult to control. So, on to immersion silver - and a new generation. What are the advantages? It has relatively low cost, is a highly solderable finish, it does not affect solder joint reliability, and is already in use in some lead-free pastes. It is compatible with solder resists, and all 'no clean' assemblies, is wire bondable and has good colour. Limitations might include silver electromigration, but it is however readily oxidised. He explained how Cookson have improved bath life and stability, deposition speed, thickness deposit rate and have reduced sensitivity to drag-in contamination etc.

Performance has been improved for SMT reflow, wave soldering, and wetting balance, and it is highly compatible with SAC. The electromigration figures looked good, tarnish resistance was much improved, even after 6 pb-free reflows. Cases of galvanic attack were examined, where solder mask undercut and poor side mask adhesion, corrosive environments such as etching. They have developed a new chemistry that minimises solder mask undercut, with a milder pH, called Entek plus HT for Pb-Free.

They also have a new OSP(Organic Solderability Preservative) for lead-free, where the advantages include the highest joint strengths, with good solderability. But it is not wire bondable and there are some contact resistance properties. Less robust than thick metal coatings, this new OSP give a PCB a 12-month shelf life, and by careful consideration of parameters it will be perfect for SAC adhesion.

Next on was Hiroshi Otake is from Uyemura who knew about final surface finishing for lead-free material solder joint applications. The process ENIG and ENEPIG were described, the latter is ideally suited for SAC alloy solder materials in which higher solder joint reliability was recorded. ENEPIG showed continuously good test results whereas ENIG did not, when under extreme conditions of frequency of application and high temperatures during multiple reflow. ENEPIG gives a uniform, thin and even IMC (intermetallic compound) layer with SAC, giving better adhesion, bond integrity. ENEPIG is very good for solder reliability where SAC alloys are used in continuous high volume production.

Don Cullen is from MacDermid in Connecticut. He discussed the effect of lead-free surface finishes on solder joint microvoids. The use of a powerful X-ray will always show where you have voids, but how are they caused? You have different finishes, at different temperatures, for a start. Microvoids are usually seen along the foot of the solder ball where it sits on the conductor, and can be caused by insufficient temperature during solder reflows, by oxides on sliver underneath, by solder mask contamination, by tin contamination, machine oils, foams, etc. It came down to cleanliness, surface preparation, and surface finish. Flux gassing on Ag showed higher outgassing than with gold. The causes and creation of microvoids was explained at length and in detail, it has much to do with processing, the creation of microvoids can be minimised by running thinner film of solder, from 0.15 - 0.45mm. No one simple answer to a complex problem, as Don told us.

Dr. Peter Meeh of MacDermid GmbH spoke to the conference about the application of test methods and comparative results for immersion tin, immersion silver and ENIG. Another MacDermid man, but this time from Germany. They are using a MUST II solder wetting balance from Concoat in the UK and a test coupon for measurements with the solder globule method. He demonstrated how they run their test methodology, and described what they were looking for and what they had found. All three surfaces are compatible with leadfree solders. The solder globule method is a sensitive method for the examination of surface finishes with lead free solders, and he concluded that ENIG, immersion silver and immersion tin are compatible with lead-free solders; all of them have their own characteristics. However, all the samples were made under production conditions so this is not just another laboratory report.

A Review of the proceedings held on Day 2 of the conference will be published on evertiq within shortly.

John LingCarter-Lane Consulting,
The Reddings, Old Colwall,
Worcestershire, England WR13 6HH
Telephone & Fax: +44 (0) 1684 540901 Mobile: 07968 190649

The European Institute of Printed Circuits (EIPC), based in The Netherlands, is an international service provider to the European Interconnection and Packaging Industry. The EIPC is a network of PCB professionals providing platforms to exchange business and technology information for the success of the European electronics industry. Since 1968, the EIPC is servicing almost 130 member companies, including suppliers of machinery and materials to the PCB industry, PCB Manufacturers, contract electronics manufacturers and OEM's.


Please note the following: Critical comments are allowed and even encouraged. Discussions are welcome. Verbal abuse, insults and racist / homophobic remarks are not. Such comments will be removed.
Further details can be found here.
Load more news
September 15 2017 9:25 AM V8.7.1-1