Global SMT & Packaging South East Asia - July/August 2011 (#2.4)

www.globalsmtindia.in

Volume 2 Number 4 July/August 2011

South East AsiaSouth East Asia

Covering India, Thailand, Malaysia, Singapore, The Philippines and Hong Kong

www.globalsmtseasia.com

CoST fACTorS of THe CleANINg ProCeSS A reVIew of HAlogeN/HAlIde-free TeST MeTHodS ANd ClASSIfICATIoNS for SolderINg MATerIAlS

IMProVed effICIeNCy uSINg rooT CAuSe fAIlure ANAlySIS

ravi PagarInterview Inside

Is Your Cleaning Going DownThe Wrong Path?

Kyzen specializes in precision cleaning chemistries for electronics, advanced packaging, metal finishing and aerospace applications. Our industry expertise and dedicated customer support provide integrated cleaning process solutions that meet

any cleaning challenge. We offer a risk-free product trial that allows your business to witness Kyzen's results at no charge. Contact us today to learn how Kyzen can streamline your cleaning process.

Let Point YouIn The Right Direction!

Nashville, TN Manchester, NH Maldegem, BE Penang Shenzhen Shanghai Guadalajara = = = = = =

Kyzen Sdn. Bhd.Plot 47, Hilir Sungai Keluang 2

Phase 4, Bayan Lepas Ind. Park11900, Penang, Malaysia

[email protected]

www.kyzen.com

0

5

25

75

95

100

kyzen_GlobalSMT_SEAsia_2011_compass

Monday, May 16, 2011 8:31:19 AM

Global SMT & Packaging South East Asia – May/June 2011 – 1www.globalsmtseasia.com

Contents

ContentsGlobal SMT & Packagingis distributed by controlled circulation to qualified personnel. For all others, subscriptions are available at a cost of $19.99 for the current volume (six issues).

No part of this publication may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without prior written consent of the publisher. No responsibility is accepted for the accuracy of information contained in the text, illustrations or advertisements. The opinions expressed in the articles are not necessarily those of the editors or publisher.

© Trafalgar Publications Ltd

Designed and Published byTrafalgar Publications Ltd,Bournemouth, UK

Volume 2, No. 4

July/August 2011

EDITORIAL2 All about the tablets Pradeep Chakraborty

TEChnOLOGy FOCuS6 Cost factors of the cleaning process—part 1 Thomas Kucharek, Application Technology, Zestron Europe

10 A review of halogen/halide-free test methods and classifications for soldering materials Jasbir Bath, Christopher Associates; Gordon Clark, Koki Solder; et al

18 Improved efficiency using root cause failure analysis Gerry Padnos, Juki Automation Systems

30 X-ray inspection of semiconductor devices that use copper wire interconnections David Bernard, Ph.D., and Evstatin Krastev, Ph.D., Nordson DAGE

34 Solder paste stencil manufacturing methods & their impact on precision and accuracy Ahne Oosterhof, Oosterhof Consulting

38 Bare board contamination Sheila Hamilton, Teknek

SPECIAL FEATuRES32 Interview—Ravi Pagar, element14

REGuLAR COLuMnS24 2H’11: Uncertainty prevails but modest growth remains likely Walt Custer and Jon Custer-Topai

4 Industry News54 New Products60 International Diary

OThER REGuLAR FEATuRES

32

Visit www.globalsmtseasia.com for the latest news and more, every day.

South East Asia

6

12

www.globalsmtseasia.com [email protected]

editorial officesEuropeGlobal SMT & PackagingTrafalgar Publications LtdUnit 18, 2 Lansdowne CrescentBournemouth, Dorset BH1 1SAUnited KingdomTel: +44 7766 951665

United StatesGlobal SMT & PackagingPO Box 7579Naples, FL 34102, USATel: +1 (239) 245-9264Fax: (239) 236-4682

ChinaGlobal SMT & PackagingElectronics SecondResearch InstituteNo.159, Hepin South RoadTaiyuan City, PO Box 115, Shanxi,Province 030024, ChinaTel: +86 (351) 652 3813Fax: +86 (351) 652 0409

Editor-in-ChiefTrevor GalbraithTel: +44 7924 581 523 (Europe)Tel: +1 (239) 245-9264 x101 (US)[email protected]

Managing EditorHeather LackeyTel: +1 (239) 245-9264 [email protected]

Technical EditorPradeep ChakrabortyTel: [email protected]

Circulation & SubscriptionsKelly Grimm Tel: +1 (239) 245-9264 [email protected]

AdvertisingSouth India—Amitava [email protected] India—Jaideep Sinha+91 [email protected] & Hong Kong—Philip Lim+65 [email protected]—Amy [email protected]—Shana HarrisTel:+1 (408) 946-8109Mobile: +1 (510) [email protected]—Andy KellardTel: +44 7432 [email protected]

All about the tablets

www.globalsmtseasia.com

It’s all about tablets, isn’t it, today? Wherever you go, you can’t miss them! I have been meeting senior officials

from leading technology firms, and invari-ably, the discussions seem to lean toward tablets. While not everyone may be plan-ning to release a tablet soon, the interest is very much there.

Why, even at this year’s pre-Computex 2011 press conference, Stephen Su, gen-eral director of IEK (Industrial Economics & Knowledge Center), Taiwan, said that tablets were moving from ‘experience’ to ‘industry.’ Tablets will likely provide more innovative user-interface experiences as well. More cost will be spent on the user interface on tablets, as far as BoM is con-cerned, he added.

Tablets are also likely to change the transactional model of the ICT industry and emerge as the new ‘service platform’ for the integration of ICT and the other indus-tries as well.

There is likely to be a crossover evo-lution among mobile devices. The use of messaging and applications seem to have overtaken voice in smartphone usage. Also, there seems to be much more use of heavy media content on the Apple iPad than the iPhone.

While smartphones will likely impact growth of mobile phones, PMPs and PNDs, it is the tablet that will likely affect the net-book and e-reader, and gradually impact the notebook.

There have been a range of tablets announced or to be announced—at least, as per Warren East, CEO, ARM. Presenting at an industry meet in India, he presented an example, featuring the likes of the ASUS Memo, HTC Flyer, HP TouchPad, LG G-Slate, RIM Playbook, Dell Streak 7, and so on.

By the way, tablet applications revenues were projected to top $15 billion in 2015. However, there still seems to be no one to

at least beat the Apple iPad, at least, up until now!

Now, East also highlighted one other thing—that ~5.1 billion people on earth do not have access to the Internet. Mobile devices are currently said to be connecting ubiquitous environments, cloud computing storage and services. It presents all of us the opportunity to reshape the value chain and create growth.

First, what are tablets? Before the Apple iPad came into being, there was no such thing heard of. I do recall seeing one at Geneva, during ITU Telecom World 2003, but that seems to have ‘died its death.’

Now, following the Apple iPad, there are a whole lot of tablets on offer, as though they are the next big thing!

Perhaps, they are! In the backdrop of all of these, when

you look at the Indian electronics industry, what do you find? There are many users of tablets, but there are very few of those from Indian companies.

I recall one that was showed to me exclusively, by a local company in Bangalore, India. It was more of an intro-ductory model, at that time.

Then came Olive Telecom, with its OlivePad. Last heard, it is selling for con-siderably less than the Apple iPad. It runs on the Android Froyo 2.2 operating system, and makes use of the Qualcomm MSM7227 chipset and the ARM 600MHz processor. It also comes with a six-month free subscrip-tion to Zenga TV.

However, there are, at the moment, no or very few other releases from Indian firms. So, are Indian technology firms missing out on the ‘tablet rush’? For the moment, it appears to be so.

Oh, just in case, I also wonder how folks are using these tablet PCs as mobile phones! It may seem odd to use a big tablet PC as a mobile phone. Well, that’s up to the user, isn’t it?

Editorial

2 – Global SMT & Packaging South East Asia – May/June 2011

Pradeep ChakrabortyTechnical Editor

Title

www.globalsmtseasia.com4 – Global SMT & Packaging South East Asia – March/April 2011

Industry newsIndustry news

ZESTRON wins “Best of Conference” awardUmut Tosun, application technology man-ager at ZESTRON, provider of high preci-sion cleaning products and services for the electronics manufacturing industry, was awarded “Best of Conference” for his pre-sentation of “Fluid Flow Mechanics: Key to Low Standoff Cleaning” at the SMTA’s SE Asia Technical Conference held in Penang, Malaysia, on May 19-20, 2011. This is the second industry recognition ZESTRON has received recently. At the end of May, ZESTRON North Asia’s process engineer, Jerry Ji, was awarded the “Best Paper of Technology Conference One (CE11)” for his presentation of the technical study titled “pH-Neutral vs. Alkaline Cleaning Agents” at the SMTA China East Technical Conference during NEPCON Shanghai. www.zestron.com

Thai consumer electronics market expected to increase to $11.4 bn by 2015Thailand’s consumer electronics devices market is projected at about US$8.3bn in 2011, according to a new report from Research and Markets (“Thailand Consumer Electronics Report Q3 2011”). This is expected to increase at a CAGR of about 7 percent to US$11.4bn by 2015, driven by the growing affordability and popularity of flat-screen TV sets, low-cost smartphones and other digital lifestyle products.

According to the report, in 2011, con-sumer electronics spending is expected to grow 13 percent, driven by an improv-ing economy, and salary increases for civil servants and others. Government support through consumers through initiatives such as the Pracha Wiwat programme should help to boost consumer dispos-able incomes and consumer confidence. In 2010, demand for consumer electronics devices recovered strongly after political turmoil had earlier affected sales of smart-phones and some other high-end products. www.researchandmarkets.com

BTU International selects TSM Solutions as new Malaysia repBTU International, Inc., a supplier of advanced thermal processing equipment and processes to the alternative energy and electronics manufacturing markets, has

announced the selection of TSM Solutions as its new representative in Malaysia. TSM will represent BTU’s award-winning line of Pyramax™ convection reflow ovens. BTU has had direct support offices in Malaysia for 14 years and has a number of key cus-tomers in the country. TSM Solutions will further strengthen the company’s presence. www.btu.com

World Micro opens Asian hub in Penang, MalaysiaElectronics distributor World Micro, Inc., has sealed an agreement with the Northern Corridor Implementation Authority (NCIA), the statutory body responsible for the development of the Northern Corridor Economic Region of Malaysia to formalize the opening of its Asian distribution hub in Penang, Malaysia. World Micro’s Asian presence is expected to improve Malaysia’s ability to rid the global supply chain of counterfeit electronic parts through vari-ous training programs, provide excess inventory disposition programs to local and regional businesses and help sup-port the economic growth within the region. Through collaboration with the NCIA, World Micro will conduct train-ing, development and counterfeit aware-ness programs for small and medium enterprises as well as large multina-tional corporations throughout Malaysia. www.worldmicro.com

TÜV Rheinland expands in IndiaBangalore-based TÜV Rheinland India is substantially expanding its capabilities in India to meet the growing market demand and provide services as a one-stop shop. In the area of electro magnetic compatibility (EMC) & wireless testing, the company will be adding a 10 meters anechoic cham-ber in addition to its existing electrical & electro magnetic compatibility (EMC) labs in Bangalore. The 10 meters anechoic chamber enables the company to test any electronics/wireless products for their radiations in the range 9 KHz to 40GHz according various national/international standards such as CISPR, FCC, ETSI etc. www.ind.tuv.com

PLCs, SCADA & DCS to take the Automation & Control market in Malaysia forwardResearch and Markets has added Frost

& Sullivan’s new report “Automation and Control Market in Malaysia” to their offer-ing. This study provides an understand-ing of the automation market in Malaysia, one of Southeast Asia’s developed regions. It discusses the techniques such as PLC, SCADA, HMI, DCS and MES in detail. It also covers the major end-users such as oil and gas, water, electronics and semi-conductor manufacturers. The study also provides information on the latest develop-ments in the automation industry and the market share of major players besides pro-viding information on the challenges faced by manufacturers of these automation equipments. www.researchandmarkets.com

Panasonic establishes R&D center in IndiaPanasonic Corporation has established Panasonic Research & Development Center India (PRDCI) in Gurgaon in the northern state of Haryana. As Panasonic’s first R&D base in India, PRDCI will contribute to the company’s business expansion in the grow-ing market with efficient R&D tailored to local needs. Panasonic has been making a company-wide effort to cultivate its busi-ness in India, having enhanced its product lineup and marketing structure as well as opened a B-to-B showroom there. The company further plans to build full-scale manufacturing facilities called Panasonic Techno Park in Jhajjar, Haryana, in the fiscal year ending March 2013. PRDCI will help realize an integrated operation of product development, manufacturing and sales with its effective, locally-oriented R&D initiatives. www.panasonic.net

Altera and RV-VLSI join hands to promote FPGA/SOPC training in India, set up joint lab centreSetting up of state-of-the-art Innovation Lab to promote research, faculty enrich-ment and skill upgrading of students. Altera Corporation, known for program-mable logic solutions, headquartered in San Jose, CA, USA, and RV-VLSI, a research and training center, a unit of the Rashtriya Shikshana Samithi Trust, Bangalore, have forged a partnership to set up an Innovation Lab, the first of its kind facility in India, involving a joint invest-ment worth US $ 1 million. The Innovation Lab will promote the use of advanced design practices among corporates, aca-demic research institutions, and students,

Global SMT & Packaging South East Asia –July/August 2011 – 5www.globalsmtseasia.com

Industry news

thereby fostering an ecosystem for product creation, enhancing job opportunities for IT professionals and familiarizing Altera products to Indian customers. The Lab also plans to work closely with universities in India to upgrade the skills of faculty in the areas of VLSI design methodologies and embedded system design. www.altera.com

Research consortium to address advanced packaging challengesThe Institute of Microelectronics (IME), a research institute of the Agency for Science, Technology and Research (A*STAR) has launched its 11th Electronic Packaging Research Consortium (EPRC11) to address various technology challenges in advanced packaging technology in semi-conductor in enabling smaller and smarter devices. Since IME initiated the first EPRC in 1996, this resource and cost-sharing platform has injected invaluable R&D capabilities into the operations of many local enterprises and multi-national com-panies in the electronic packaging industry and its value chain. EPRC11 consists of 23 company members spanning the whole supply chain of the industry from system, integrated device manufacturer, foundry, assembly & test, to equipment and material companies. The consortium will engage in four projects over its 18-month durationL multiple chip embedded wafer level pack-aging, through silicon via (TSV) interpos-ers, fine pitch flip chip with Cu pillar, and high performance materials for advanced packaging. www.ime.a-star.edu.sg

Hitachi opens investment discussions with Hi-Rel ElectronicsHitachi, Ltd. and Hi-Rel Electronics Pvt. Ltd. have agreed to open discussions regarding a business alliance, including Hitachi’s investment on Hi-Rel and cre-ation of a new manufacturing facility. The aim of this move is to bolster both compa-nies’ power electronics businesses in India. This will be achieved by fusing the strengths of each company ‘Hitach’s medium- and high-voltage control technologies, advanced manufacturing technologies, and wealth of experience and expertise in sys-tems integration in large-scale plants, and Hi-Rel’s industrial-use UPS technologies, low-voltage and small- and medium-sized control technologies, existing manufac-turing facilities, and nationwide sales and service network in India. The two compa-nies will hold discussions with a view to entering into a final agreement around the end of September 2011. www.hitachi.com, www.hirel.net

ON Semiconductor opens global distribution center in SingaporeON Semiconductor, supplier of high per-formance silicon solutions for energy efficient electronics, has opened its new Singapore hub for global distribution. The US$3.5 million state-of-the-art facility was built and implemented in partnership with DHL. The Singapore Global Distribution Center (GDC) is dedicated to shipping finished goods to its worldwide customer base, acts as a global customer sample center, stores die and wafers for shipment to assembly and test operations, and serves as a local hub for consolidation and deliv-ery of products to customers in South Asia. This facility is now the largest distribution facility serving the semiconductor indus-try located in Singapore. More than half of the 38 billion devices ON Semiconductor produces and ships annually are expected to pass through the company’s Singapore GDC. www.onsemi.com

Ellsworth Adhesives authorized Cytec distributorin ThailandEllsworth Adhesives has become an autho-rized distributor for the Cytec Conap product line for the electronics and filter markets in Thailand, Australia and New Zealand. Cytec Conap is a leading manu-facturer of polyurethane and epoxy potting compounds and polyurethane and acrylic conformal coatings. Cytec Conap prod-ucts are proven systems with outstand-ing performance in the most demanding electronic and electrical applications. The Cytec Conap product line offers a reliable, cost effective solution for the aerospace, automotive, consumer, industrial and marine industries. www.ellsworth.com

Aehr Test Systems sells ownership in Singapore’s ESA ElectronicsAehr Test Systems, provider of semi-conductor test and burn-in equipment, has divested its shareholdings in ESA Electronics Pte Ltd, a Singapore private limited company, to IPCO International Limited, a Singapore company, represent-ing 12.5 percent of the outstanding shares of ESA Electronics. As a result, Aehr Test has no remaining ownership in ESA Electronics. ESA Electronics is in the busi-ness of distributing and servicing semicon-ductor back-end equipment, designing and manufacturing burn-in boards for semi-conductor back-end stress test and man-aging electronics turnkey projects. ESA Electronics has been a sales representative of Aehr Test in Southeast Asia since 1992. ESA Electronics will continue as a sales

representative of Aehr Test after the sale of Aehr Test’s ESA Electronics shares to IPCO International. www.aehr.com

Sono-Tek Corporation expands customer service globallySono-Tek Corporation, manufacturer of precision ultrasonic coating equipment, has expanded its customer services with new laboratory facilities in China, Korea, Germany and Taiwan. The new testing facilities will enable customers to test small samples of their liquids in a con-trolled environment, simulating process conditions as much as possible in order to prove concept and ensure compatibil-ity of liquids. Emerging applications such as advanced energy coatings and specialty medical device coatings often find these services particularly valuable. www.sono-tek.com

Flextronics Global Services continues rapid expansion into new and emerging marketsFlextronics Global Services (FGS), a busi-ness unit of Flextronics, has made a second strategic market expansion since open-ing new facilities in Europe and India in October 2010. FGS has expanded its new markets in Mumbai, India and Istanbul, Turkey in response to growing customer demand for post-manufacturing supply chain services in these regions. FGS’ new facility in Mumbai will support repair service requirements for a leading global OEM of smart phones and will serve as a satellite operation for FGS’ core facility in Bangalore. FGS’ new facility in Turkey will serve as the inbound hub operation for a global brand and leading computing OEM. The facilities in Mumbai and Turkey are both fully operational. www.flextronics.com

Scapa Group establishes new Indian subsidiaryScapa, a producer of adhesive components and bonding solutions for consumers, dis-tributors and OEMs in the automotive, industrial, electronics and medical sector, has esablished a new, wholly-owned sub-sidiary, Scapa Tapes India, to support the fast developing Indian market, the GDP of which is expected to increase more than 8 percent in the future years. The company will combine the existing sales offices at Mumbai and New Delhi into the new Indian division. Other subsidiaries of Scapa are located in Malaysia, Korea and China. www.scapa.com

6 – Global SMT & Packaging South East Asia – July/August 2011 www.globalsmtseasia.com

Cost factors of the cleaning process—part 1

Companies are constantly striving to reduce costs incurred during electronics production in order to increase cost effectiveness. Generally rising costs or the need to preserve competitiveness relative to low-wage countries further reinforce this development. Being part of the value chain and production, cleaning processes for electronic assemblies are also subject of cost analyses. The following article, the first in a three-part series, presents a detailed summary of all cost factors relevant to the acquisition and operation of cleaning processes. A detailed analysis for optimizing individual cost factors in batch spray systems and inline processes will be conducted in two additional studies.

Thomas Kucharek, Application Technology, Zestron Europe


The cost of a cleaning process depends on different factors. These can vary significantly depending on

the application or the cleaning equipment. Therefore, total costs have to be calculated relative to the throughput or to the number of cleaned parts in order to determine a cleaning process’s true costs. So, the general rule is:

The result of this calculation shows the cost incurred for each cleaned assembly depending on production output. Naturally, every company’s objective is to keep these costs at a minimum.

However, to establish the cost per cleaned part and to reduce it by implement-ing appropriate process improvements, the factors influencing a cleaning process’s investment and operating costs must first be determined.

Therefore, over the past few months, extensive studies on the subject of “cost fac-tors of the cleaning process” have been con-ducted at Zestron’s worldwide Technical Centers using different machine types (Figure 1).

The study’s objectives were: 1. to identify all major cost drivers in

cleaning processes,2. to determine factors influencing

the cost drivers, and 3. to identify approaches for realizing

cost reductions.

Step one, identifying the major cost drivers, can be illustrated quite quickly. As already

mentioned above, a distinction is made between the investment and the depre-ciation of a new process and the operating costs of an existing cleaning application.

With a new investment, users must consider the following factors as they have a major impact on the cleaning process’s type as well as size and therefore on later investment costs:

• Expected production throughput• Necessary degree of automation• Equipment size and/or available

floor space• Type of cleaning medium

Subsequently, operating costs occur during electronics production due to:

• the consumption by equipment peripherals such as exhaust sys-tems, etc.,

• the operating staff and mainte-nance costs, and

• the consumption of cleaning media and energy

Figure 2 illustrates and compares all rel-evant cost factors of a cleaning process.

After identification of the major cost drivers, a more detailed analysis will deter-mine the other influencing factors and possible options for cost optimization.

1. Cost factor: cleaning equipmentWhen the cost factor “cleaning equipment” is discussed, the initial investment in the cleaning machine is often considered first and is naturally the greatest cost factor

(investment and/or depreciation + operating costs)No. of cleaned parts

= costs per cleaned part

Figure 1. Zestron’s Technical Centers in Germany, the US and China.



when installing a new cleaning process. The selection of the equipment should

be based on the expected throughput and the user should first decide for one of the three basic machine types: ultrasonic benchtop, batch machine (i.e. one-chamber spray-in-air equipment), or inline machine. Table 1 illustrates the different application areas and serves as a basic guide.

Apart from the expected through-put, there are further requirements that the user should evaluate before select-ing the equipment. The approved invest-ment budget should not be exceeded and the machine’s size must be suitable for the available footprint. Furthermore, the clean-ing machine’s level of automation must be considered for the purchase decision since it influences subsequent operational costs of the process. The user must turn to the operational costs of the cleaning process once the equipment has been purchased.

Looking at the operating process, costs are incurred at various points depending on machine types construction. On the other hand, these can also provide options for savings.

Using the smallest process, the bench-top, cleaning and rinsing takes place in separate process tanks. Savings are thus realized primarily through the optimal positioning of the parts to be cleaned and the drip-off mechanism.

For batch machines, such as one-cham-ber spray-in-air machines, existing pipes, spray bars and pumps are often used for cleaning and rinsing. The drag out of the cleaning medium due to dead volumes in the machine thus influences the process costs. However, drag out does not just occur due to the machine’s construction, but also due to the configuration of the parts to be cleaned. Since the assembly geometry cannot be changed, the main option for

optimization is the drip-off time. Apart from the exhaust system, the

most important cost-driving factor of inline machines is the drag out of cleaning medium into the rinse stages. A suitable configuration of the air knives thus repre-sents a prerequisite for potential savings.

2. Cost factor: equipment peripheralsIn reference to the cleaning equipment’s periphery, the study’s results reveal that costs during the process operation essen-tially arise through processing, drag out and exhaust. The equipment periphery is structured differently depending on the machine type. Table 2 shows the peripheral systems that were investigated.

In summary, this means the following: 1. The cost aspect “equipment

periphery” is virtually non-existent with ultrasonic benchtops, since typically neither the cleaner nor the water are treated

2. The biggest cost driver in the batch process investigation, such as one-chamber spray-in-air machines, proved to be the processing or production of deionized water for rinsing. Using either an ion exchanger or reverse osmosis are reasonable depending on the amount of deionized water consumed. Costs also arise for processing the cleaning agent in closed loop, e.g. through filter changes.

3. In inline processes for high-vol-ume production, peripheral equip-ment costs have the largest impact

Machine size

Production through put

Degree of process

automation

New investment

Solvent or water-based

cleaner

Equipment peripherals

Energy consumption

Process operating

costs

Staff costsProcess

maintenance

Cleaning medium

New Investment Process Operation

Figure 2. Comparison of the investigated costs.

Cleaning equipment ultrasonic benchtop Batch process, e.g. one-chamber spray-in-air equipment

Inline process

Throughput per shift approx. 50–100 approx. 500–1,500 > 3,500

Investment volume 1,000–5,000€ 20,000–80,000€ > 100,000€

foot print <1 m2 1 m2 to 3 m2 >10 m2

degree of process automa-tion

none (manual)

medium (manual loading/unloading of cleaning goods)

high (fully automated, completely integrated into the production

process

Table 1. List of common cleaning systems. (These data are merely a guide).

ultrasonic benchtop Batch process, e.g. one-chamber spray-in-air equipment

Inline process

not available • Deionized water treatment (e.g. activated carbon, mixed bed resin)

• Processing of the cleaning medium (filtration)

• Machine exhaust• Deionized water treatment (e.g. activated carbon,

mixed bed resin)• Processing of the cleaning medium (filtration)

Table 2. Cost-relevant periphery by cleaning equipment.



on the total process cost. This is mainly due to the vapor losses through the machine’s exhaust system. As the investigation shows, the exhaust gas temperature and the spray nozzles’ atomization of the cleaning medium are the decisive factors. Furthermore, the results show that cost savings are possible by optimizing the treat-ment cycle of the cleaning agent and the rinse water.

The energy consumption and maintenance costs of the above mentioned cleaning pro-cesses were also investigated in the study. In addition to the operating costs incurred by machine peripherals during cleaning, the results for energy consumption and equipment maintenance are listed below.

3. Cost factor: energy consumption and equipment maintenanceDepending on the cleaning process, the amount of operational costs varies consid-erably. Figure 3 illustrates this.

Ultrasonic benchtop processes suit-able for the lowest throughput exhibit the lowest total operating cost in this con-text since relatively little power is needed to heat the bath and additional costs for compressed air and for drying are not necessary. On the other hand, staff costs comprise the largest share of the operating costs since all process steps such as clean-ing, rinsing and drying must be conducted manually. In addition, due to the low bath volume, this type of process does not have an integrated filtration system so that fre-quent bath changes are required and can lead to further maintenance costs.

In batch cleaners, the cleaning agent, and, if necessary, the rinse water must be reheated after each cleaning step. Due to this and the drying of the entire cleaning chamber, considerably higher energy costs arise in batch systems compared to bench-tops. In addition, discharging the water after each rinse causes further operational costs. As this type of process is mainly automated, staff costs usually only occur for loading, unloading and maintaining the machine. Furthermore, the integration of a treatment process for the cleaning agent, such as filtration, maximizes the bath life, so that expensive bath changes are required considerably less often than with benchtop systems.

In the studies, the consumption of water, energy and compressed air by inline systems had the greatest effect on the over-

all process costs (as expected). In general, these are closed loop circulation systems and the processing cost for rinse water is relatively high. Furthermore, most inline systems must be operated continuously to ensure consistent process results and even during times when no parts have to be cleaned. Due to this, continuous energy expenditures arise and are generated by the cleaning bath’s circulation and heating as well as by the exhaust system after the rinse stage and before drying. Additionally, costs are incurred when blowing off the cleaned parts, as air knives are integrated into the line multiple times. Due to the inline sys-tem’s permanent operation, comparatively high maintenance and repair costs occur as well. However, operator costs are relatively low due to the fact that inline machines are completely integrated into the production line and therefore easier to operate and monitor by a single person.

In summary, it can be stated that operating costs increase relative to the throughput and the equipment size. At the same time, the study confirms that personnel costs decrease as the cleaning process’s degree of automation increases. Furthermore, immersion systems and spray processes only incur costs during cleaning, whereas operating costs for inline systems arise even during idle phases.

After all costs associated with the cleaning equipment have been carefully

considered, as a next step, a suitable clean-ing agent must be selected. The relevant cost factors in this area will be shown in the following part of the study.

4. Cost factor: cleaning mediumIn the industry, primarily three types of cleaning agents are used for the cleaning of electronic assemblies. These include alco-hols, conventional surfactants and modern

Figure 3. Relationship of operating costs for the different machines per part.

Cost advantages Cost disadvantages

Solvents • long bath life• short process times

• high evaporation losses• expensive exhaust treatment• explosion protected equipment

required

Traditional surfactants

• low evaporation losses• low application concentra-

tions

• short bath lives, thus expen-sive bath changes needed

Moder, water-based cleaners

(e.g. MPC® media)

• low application concentra-tions

• long bath life• low evaporation losses

• filtration for enhanced bath life required, thus cost for filter changes, etc.

Table 3. Cost consideration of the operating process.

Figure 4. Comparison of bath lives.



water-based cleaning media. When a new cleaning machine is purchased and the tank has to be filled with a cleaning agent for the first time, companies often focus on the cleaning medium’s price per liter. However, this is not the only fact that determines the ultimate costs of the clean-ing agent.

For example, the bath life is crucial, i.e. how long the cleaning agent can be used before a bath change is required. Thus, a cleaning agent with a short bath life may be more expensive in the long run, despite its favorable price per liter, than a cleaning medium with a higher purchase price pro-viding a longer bath life.

Based on their respective chemi-cal properties, the three types of cleaning agents mentioned above have individual cost advantages and disadvantages during the cleaning process, as Table 3 shows.

Most solvents exhibit long bath lives and short process times due to their excellent drying properties (see Figure 4). However, there are cost disadvantages due to high evaporation losses, the neces-sary explosion protection for the cleaning equipment, which typically has an impact on the initial investment or even an expen-sive filtration of the exhausted air.

Conventional surfactants do not have these problems. At the same time, they are sold at an appealing price per liter and are often used with low application concentra-tions. Nevertheless, their bath life is mostly quite short, which leads to frequent bath changes and increases the costs due to the higher cleaner consumption (Figure 4).

On the other hand, modern water-based cleaning media provide very long bath lives (Figure 4) and can also be oper-ated at low application concentrations. Costly evaporation losses, as with solvents, are neglected. Also a long bath life can be achieved via a filtration cycle in the clean-ing machine for additional costs.

At this point, to choose the right clean-ing agent, users should seek advice from a cleaning expert, who can determine the most cost effective solution. Similarly, it is crucial to choose a cleaning agent that is suitable for removing the specific contami-nation and thus ensures the best possible cleaning results.

SummaryTo determine the cost per cleaned part and to take appropriate measures to reduce costs, all cost drivers and their parameters must be identified. Therefore this article has provided a fundamental basis.

The results of the previously men-

tioned technical studies point out that the throughput during assembly clean-ing essentially represents the basis for the equipment selection. At the same time, however, operating costs for energy con-sumption and maintenance are also largely determined by the equipment choice while specific measures for realizing significant savings cannot be taken.

Once the cleaning process is imple-mented, the cleaner, the equipment periph-ery and the process parameters prove to be the only cost factors that can be influenced

and thus optimized. Therefore, these three factors will be specifically investigated in two subsequent studies focusing on batch and inline processes to even better under-stand them and to detail possible savings potentials. These types of cleaning pro-cesses involve the highest cost cases due to their technology’s complexity and thus have the most potential for savings.


A review of halogen/halide-free test methods and classifications for soldering materials

In the electronics industry, there is a significant push toward halogen-free products. This movement is due

to legislation from various countries, and public outcry from well publicized negative third world recycling practices, as well as non-government organizations (NGOs) testing and publishing informa-tion on electronic devices regarding their content of various potentially hazardous materials. Halogen-free products are also being mandated by certain OEMs as a means to lessen poten-tial chemical effects on the environment.

In electronics assemblies, halogens can be found in the plastics for cables and housings, board laminate materials, components, and soldering fluxes and pastes. In solder pastes and fluxes, the haloge-nated compounds are used as activators that remove oxides to promote solder wetting. Eliminating the halogenated compounds can have a signifi-cant negative effect on the board assembly process. Process assembly challenges are not the only issues electronics assemblers face as they become halogen-free. The use of proper test methodologies to determine that the soldering products are actually halogen-free is currently not well defined as there are a variety of test methods and standards in the industry.

Halide content has been measured either qualitatively or quantitatively with halide testing being specified for more than

fifty years with standards such as the United States Federal Specification QQ-S-571 standard1 followed by MIL-F-142562 and IPC-SF-8183 standards and currently in standards such as IPC J-STD-0044. The specifications have listed requirements for the halide content of flux-containing soldering materials.

The terms halogen and halide have caused confusion in the electronics industry

with definitions to try and clear up the confusion provided by standards such as JEITA ET-73045 and IPC-J-STD-0044. The term halogen refers to all halogen family elements and halogen compounds including those which are present in nature. The JEITA ET-7304 standard5 specifically targets the halogen families of chlorine (Cl), bromine (Br) and fluorine (F) used as

the activators for soldering materials. The term halide is defined as the halide ion or halide salt compound having an ionic char-acter (e.g. Cl-, Br-, F-).

Covalently bonded halogens do not disassociate in water, and therefore the chloride, bromide and fluoride are still attached (covalently bound) to other species (typically organic), and will not be detected by techniques such as ion chro-matography or titration. Ionically bonded halogens do disassociate in water into the negatively charged halide ion (Cl-, Br-, F-, etc.) and the positively charged species (H+, Na+ etc.). Test methods used to look

This paper was originally presented at IPC APEX Expo 2011.

Keywords: halogen-free, halide-free, Classifications, Standards

Over the last few years, there has been an increase in the evaluation and use of halogen-free soldering materials. In addition, there has been increased scrutiny into the level of halogens and refinement of the definition and test-ing of halogen-free soldering materi-als. The challenge has been that there has been no common standard across the industry in terms of halogen-free definitions and the corresponding test methods to determine these. This has created confusion in the industry as to what end users want and what sol-dering materials suppliers can actu-ally provide.

This paper reviews the status of halogen-free and halide-free in terms of definitions, test methods and the limitations and accuracy of test methods used to determine if a soldering material is halogen/halide-free or not. The different industry standards, both currently available and in the process of being drafted, are also discussed.

Jasbir Bath, Christopher Associates; Gordon Clark, Koki Solder; Tim Jensen, Indium Corporation; Renee Michalkiewicz, Trace Laboratories; Brian Toleno, Henkel Electronic Materials


Process assembly challenges are not the only

issues electronics assemblers face as they become halogen-free.


Title



for ionic species, such as ion chromatog-raphy, will only detect halides.

A better understanding of the test methods, what they are capable of detecting in terms of halides and halogens, and how they relate to the various halogen-free defi-nitions and standards is required. A variety of these test methods and standards will be discussed in the following sections as well as some test preparation techniques.

Halogen-free definition and standardsA variety of halogen-free definitions and standards have been developed in the elec-tronics industry based on PCB laminates, components and soldering materials. The developed standards for PCB Laminates include IEC 61249-2-216, JPCA-ES-017 and IPC-41018. All three standards indicate less than 900 ppm Cl (<0.09wt %), less than 900 ppm Br (<0.09wt %) and less than 1500 ppm total Cl and Br (<0.15wt % Cl + Br).

The developed standards for compo-nents include JEDEC JEP709 standard9. This standard indicates that a solid state device must meet the following requirements to be defined as low halogen in terms of less than 1000 ppm Br (from BFR[Brominated flame retardants] sources), less than 1000 ppm Cl (from CFR[Chlorinated flame retardants], PVC[Polyvinyl Chloride] and PVC co-polymers sources). For the PCB laminates used in components the Cl and Br limits would follow the guideline in IEC 61249-2 standard6.

The developed standards for soldering materials include JEITA ET-70345 and IPC J-STD-0044. The JEITA ET-7034 standard5 states halogen content less than 1000 ppm Cl, less than 1000 ppm Br and less than 1000 ppm F. An updated draft of the JEITA standard, JEITA ET-7034A10, also includes iodine (I) with a value of less than 1000 ppm. In contrast, the IPC J-STD-0044 document does not currently have a requirement for halogen content. J-STD-004 standard4 only specifies a halide content less than 500 ppm total halide. The amendment to J-STD-0044 currently being added does include optional testing for halogen content and is leaning towards the 900 ppm Cl, 900 ppm Br and 1500 ppm total halogen content suggested require-ments.

Test Methods used to deter-mine halide /halogen contentThe test methods used to analyze for halides and halogens are outlined in the following section.

Silver Chromate paper test for bromide and chloride (halide)The Silver Chromate paper test method based on IPC J-STD-0044 and IPC TM-650 2.3.3311 is a qualitative test in which a sample of flux is applied to Silver Chromate Paper and allowed to remain on it for one minute. If the paper changes color then it indicates the presence of chloride or bromide. This test only identifies the halogen in the ionic form (halide) and is prone to false positives from chemicals such as amines, cyanides, and isocyanates. It also provides no indica-tion as to the total halogen present.

Fluoride spot test for fluoride (halide)The fluoride spot test method based on IPC J-STD-0044 and IPC-TM-650 2.3.35.112 is a qualitative test and is designed to deter-mine the presence (if any) of fluoride(s) in the soldering flux by visual examination after placement of a drop of liquid test flux in a zirconium-alizarin purple lake. This method only detects the presence of the fluoride ion.

Although the following test methods,

titration and ion chromatography, which are discussed in the next section are used to measure ionic halide, they may be used following oxygen combustion to determine total halogen content.

Titration method for chloride, bromide and fluoride (halide)These are quantitative tests that assess the chloride and bromide (IPC-TM-650 2.3.3513) and fluoride (IPC-TM-650 2.3.35.214) present in a flux expressed as chloride equivalents. A flux or flux extract is titrated to its endpoint using the appro-priate IPC test methods. The test methods are an improvement over Silver Chromate paper test and fluoride spot test methods in that it provides a value for how much halide is present. However, this test method detects only halides and not total halogens unless an oxygen combustion method is used to prepare the sample prior to titra-tion. Additionally, there are a wide variety of organic chemicals that can falsely be identified as halides.

Halogen content variation based on reporting

Mass (g) Mass of halogen (g)

Halogen content (in ppm)

Solder Paste (100g flux and 900g solder metal)

1000 0.045 45

Flux (base material) 100 0.045 450

Flux Residue (50% of 100g) (as 50% of the flux volatilized during reflow)

50 0.045 900

Table 1. Halogen content variation based on different reporting values for the solder paste, flux and reflowed flux residue.

Anions by ion chromatography

result (mg/kg)

reporting limit (mg/kg)

weight(g)

Bromide 1210 72 0.000607

Chloride <162 162 <0.000081

Fluoride <72 72 <0.000036

Iodide <700 700 <0.00035

Table 2. Solder paste extracted by centrifuge, prepared with oxygen bomb and analyzed via ion chromatography17. Oxygen bomb combustion test method: EPA SW-846 5050/9056/ SW505018.

Anions by ion chromatography

result (mg/kg)

reporting limit (mg/kg)

weight(g)

Bromide 2110 55.7 0.00105

Chloride <125 125 <0.0000625

Fluoride <55.7 55.7 <0.0000278

Iodide <700 700 <0.00035

Table 3. Solder paste reflowed at 240°C, prepared with oxygen bomb and analyzed via ion chromatography17. Oxygen bomb combustion test method: EPA SW-846 5050/9056 / SW505018.



Ion chromatography for chloride, bromide, fluoride and iodide (halide)This is a quantitative test method (IPC J-STD-0044 and IPC-TM-650 2.3.2815) for chloride, bromide, fluoride and iodide that can identify the total quantity of halides present in a flux. Based on the retention time in the ion exchange column, a chro-matogram is developed and peaks are iden-tified as various ions based on previously developed standards. This test method allows a quantification of how much halide ions are present and which particular halide is present. The challenge with ion chromatography testing by itself is that

it only identifies the ionic halide species and the covalently bonded halogen are not detected again, unless the sample has been prepared using an oxygen combus-tion method prior to ion Chromatography testing. In addition, there are chemicals that have similar retention times to Cl- and Br- which can result in non-halides being misidentified as a halide.

There is a growing practice of running ion chromatography on reflowed flux residue in terms of sample preparation before ion chromatography testing. There are two reasons that people typically utilize this type of method. First, they are examining the flux residues remaining on the PCB for any species that may lead to an increased occurrence of corrosion or dendrite growth from halide ions that do not volatilize. Secondly, any covalently bound halogens contained in the flux may disassociate during the reflow process and

then the subsequent extraction and chro-matography testing will detect these disso-ciated halogens as well as the halides that do not volatilize. However, if all of the covalently bound halogens are not disasso-ciated, then the amount of halogens will be underreported.

The IPC J-STD-004 standard4 mentions in Appendix B-10 that the IPC-TM-650 2.3.28 test method15 is intended for the

detection of ionic halides only and is not be confused with total halogen content determination [ionic halide plus non-ionic (covalent) halogen]. Total halogen content should be tested by oxygen bomb combus-tion testing using a test method such as EN14582 standard16 followed by ion chro-matography testing which is mentioned in the next section.

Sample No.

1st 2nd 3rd 4th 5th Average %rd

Test 1 - 13.1 12.7 12.4 11.3 11.3 12.2 6.6

Test 2 - 273 273 270 268 274 272 1.1

Test 3 - 1170 1190 1170 1200 1170 1180 1.4

Test 1 Residue after reflow

16.6 11.7 13.9 15.0 16.6 14.8 13.9


44.8 33.9 32.6 34.2 32.7 35.6 14.5


78.1 73.2 78.2 79.0 80.4 77.8 3.5

Table 4. Cl-concentration measured on raw flux and reflowed flux residue using ion chromatography only (not oxygen bomb combustion then ion chromatography)5.

No. Combustion Method

Amount of sample

mg

Combustion temperature

°C

Combustion time S

Conditions Cl (mg/kg = mass

ppm)

Br (mg/kg = mass

ppm)

1 Quartz tube 5 1000 300 Combustion tube 1 333 301







8 Oxygen flask 10 no Setting Time allowed to stand: 20 mins Flask 1 347 295

9 Oxygen flask 20 Time allowed to stand: 20 mins Flask 2 351 275

10 Oxygen flask 40 Time allowed to stand: 20 mins Flask 3 334 305

11 Oxygen bomb 100 no Setting Time allowed to stand: 20 mins Bomb 1 320 283

12 Oxygen bomb 200 Time allowed to stand: 20 mins Bomb 2 342 305

13 Oxygen bomb 400 Time allowed to stand: 20 mins Bomb 3 334 306

Table 5. A comparison of three common combustion methods using various combustion temperatures and times on the measured Cl and Br values for soldering flux5.

Combustion method

Average or Standard deviation

Cl (mg/kg = ppm)

Br (mg/kg = ppm)

Quartz tube Average 344.4 317.7

Oxygen flask Average 335.0 286.0

Oxygen bomb Average 338.0 305.5

Quartz tube Standard deviation 11.1 18.8

Oxygen flask Standard deviation 15.5 11.8

Oxygen bomb Standard deviation 5.7 0.7

Table 6. Average of the test result analysis values showing better repeatability for the oxygen bomb versus the quartz tube or oxygen flask test methods5.



Oxygen bomb combustion testing followed by ion chromatography testing (halogen)The use of oxygen bomb combustion followed by ion chromatography testing is growing in popularity in the elec-tronics industry. The oxygen bomb test method involves subjecting a sample of flux or solder paste to an oxygen bomb combustion in which all of the organic materials are burnt off at very high temperature. This process breaks the covalent bonds for all halogens. The remaining ash consists of the ionic halides and other inorganic materials. The dissolved ash is then run through ion chromatography to determine the total halide content of a material even if it originally contained covalently bonded halogens. Since most halide restrictions are based on the finished circuit board assembly, there has been a discussion on whether the oxygen bomb combustion test followed by ion chromatography test should be run on the reflowed flux residue rather than the unre-flowed flux.

To determine the halogen content of the flux residue, one could begin by testing the flux or the flux portion of a solder paste through TGA (thermo-gravi-metric analysis) equipment using a simu-lated reflow profile. This will provide an approximate value for the amount of flux residue remaining after reflow. Then, after testing the raw flux through oxygen bomb combustion followed by ion chromatog-raphy testing, a simple conversion could be done using the safe assumption that no halogen present will volatilize. For example, if the oxygen bomb combustion followed by ion chromatography test results show 450 ppm of Chloride present and the TGA results shows that the flux volatil-izes 50% during reflow, it would be deter-mined that there will be 900 ppm Cl- in the flux residue. Table 1 shows a hypothetical example of the halogen content variation based on different reporting values for the solder paste, flux and reflowed flux residue.

In a study run by Jensen et al.17 using oxygen bomb combustion testing followed by ion chromatography testing, they found the bromide concentration of the raw flux was lower than the reflowed sample. Many companies running the halogen content test are currently using raw flux for testing, as this seems to be the easiest to imple-

ment. It is important that those interpreting the results understand that there will likely be a higher ppm level in the flux residue due to the decreased mass of the tested sample. Results of the study are reported in Tables 2 and 317.

The oxygen bomb combustion test proce-dure mentioned in EN 14582 standard15 indicates that methods such as ion chroma-tography can be used for the determina-tion of halides after oxygen bomb combus-tion testing. There are various other oxygen bomb combustion test methods which can be

used in addition to EN 1458216, including EPA SW-846 5050/905618 and JPCA ES-01-20037 standards. JEITA ET-7304 standard5 mentions that any of these three oxygen bomb combustion test methods can be used. Most laboratories typically use EN 14582 standard16 which appears to gaining in popularity.

results and discussionHalogen-free definitions and standardsFor the definition of halogen-free for PCB Laminates when the IEC 61249-2-21 standard6 was being developed, there were discussions about the ability of the test methods to repeatability detect low chlorine and bromine levels in PCB laminates using the semi-open flask test method, which was the method used to detect bromine and chlorine in PCB lami-nates.

In some cases, a lower level of chlorine and bromine (200-300 ppm) was being pushed for by certain groups. Because of the difficulty in repeatability detecting these low levels of chlorine and bromine, certain other groups were pushing for a 1500 ppm to 2000 ppm range.

As a compromise, the maximum level

of both chlorine and bromine agreed upon in IEC 61249-2-21 standard6 as was already indicated in previous sections, was less than 900 ppm chlorine and less than 900 ppm bromine for a halogen-free PCB laminate with a total value of chlorine and bromine not to exceed 1,500 ppm.

This halogen-free definition was also used for halogen-free PCB Laminates included in JPCA-ES-017 and IPC-41018 standards. As already mentioned, the test method used for detection of bromine and chlorine in PCB laminates was the semi-open flask method. It has been found that this method of detection is not as accurate as the oxygen bomb combustion test methods and there are discussions to consider the inclusion of the use of the oxygen bomb test method for laminates in IPC 4101 standard8.

For the definition of halogen-free soldering materials, the JEITA ET-7304 standard5 discussed whether the value of 900 ppm should be used for both chlorine and bromine as is used for copper clad PCB laminate materials or if the 1000 ppm limit mentioned in the European Union RoHS legislation for the two brominated flame retardants PBDE (Polybrominated Diphenyl Ethers) and PBB (Polybrominated Biphenyls) should be used. The JEITA ET-7034 standards group decided that a difference of 100 ppm was not significant either technically or envi-ronmentally, so the 1000 ppm limit should be adopted for bromide and chloride. The committee who wrote the JEITA ET-7304 standard5 are also looking to add Iodine (I) into their standard in addition to chlorine, bromine and fluorine10.

For IPC J-STD-004 standard4, when quantitative requirements were placed on halide content, a product was deemed halide-free if the halide content measured was less than 0.05 wt% or 500 ppm. The 500 ppm definition most likely came from the typical detection limit for halides at that time and the fact that raw materials containing trace halide naturally typically fell below this limit.

The reasons why the IEC 61249-2-215, JEITA ET-70345, and IPC J-STD-0044 standards were not in line with each other included different times of standard publi-cation, different materials involved, and different reasons as to why the determina-tions were being run (I.e. environmental safety concerns versus determination of flux activity level).

Halogen testing data In order to understand halogen-free

To determine the halogen

content of the flux residue, one could begin by

testing the flux or the flux portion of a solder paste

through TGA equipment.



material testing detection methods, published data relating to halogen-free testing was reviewed. The JEITA ET-7304 standard5 has data which included:

• Ion chromatography testing of raw flux (unsoldered) versus flux residue(reflowed)

• Preparation using three combus-tion methods (quartz tube, oxygen flask, and oxygen bomb) at various temperatures and times

Ion chromatography only, raw flux versus flux residue testing: The JEITA ET-7304 standard5 study compared raw flux with reflowed flux residue. Three samples were tested with five replicates

each. Based on the information provided, it was assumed that the samples were either simply diluted versus reflowed and diluted and the chloride concentration was determined via ion chromatography. The samples were not prepared using an oxygen bomb combustion method. The analysis showed that the chloride concentration of the reflowed samples was less than that of the raw flux. Data is presented in Table 4.

Various paste and flux suppliers have completed specific analyses to try and address concerns that they have with halogen determination. There is still some debate as to whether the worst case ppm halogen in the unreflowed sample should be reported or that of the flux after reflow should be used. The after reflow

value more closely describes the amount of halogen that would be present on a soldered assembly.

Comparison of different combus-tion methods: Another study that was performed which indicated in the JEITA ET-7304 standard5 was a comparison of three common combustion methods, quartz tube, open flask and oxygen bomb with varying combustion temperatures and times. The data presented was assumed to be performed in a single laboratory. The sample tested was a soldering flux. A statis-tical analysis of the data showed that 95% of the Cl results, regardless of combustion method chosen fell in the range of 340 ppm + 22 ppm. The Br results showed a slightly larger range at a 95% confidence interval (307 ppm + 39 ppm). The oxygen bomb combustion method had the tightest range of results, as shown in Tables 5 and 6, so was the most repeatable of the test methods evaluated from the study.

As the JEITA test standard results5 were from a single laboratory, one of the major points of concern for those who are required to report halogen content would be the potential lack of repeat-ability between laboratories reportedly using the same test methodology. Toleno et al.19 reported findings of their lab-to-lab comparative analysis. Based on the adhesive material tested, halogens were not intentionally added, but it was known that there are halogens naturally present in the material. The samples of the same lot batch of material were sent to three different labs for analysis to determine the amount of halogens present. As can be seen in Table 7, two laboratories using the same method obtained very different results, whereas two laboratories using two different methods obtained results within experimental error of one another. Therefore, not only is the test method important, but also the sample preparation and halide ion detection meth-odology used.

In another study by Seelig et al.20 data was presented from a global round robin study of six laboratories using the EN 1458216 oxygen bomb combustion test method. A paste was prepared with 13,000 ppm bromine (NC-A sample) and a control paste prepared with 0 ppm bromine (NC-B sample). The pastes were oxygen bomb combusted and analyzed via ion chroma-tography.

Comparative data from the six labo-ratories is shown in Table 8. Laboratory 3 data shows a variation in reported bromine value for the NC-A paste sample compared with the other five laboratory results. In

Method utilized chlorine (ppm) bromine (ppm) fluorine (ppm)

Lab 1 En14582 (Oxygen bomb)16

nD nD nD

Lab 2 En14582 (Oxygen bomb)16

748 nD 2010

Lab 3 IEC612249-2-21 (Combustion flask)6

606 nD 1460

Table 7. Halide testing of an adhesive material from three test laboratories showing differ-ences in test results19.

NC-A Sample (13,000 ppm bromine)

NC-B Sample (0 ppm bromine)

Lab 1 11700 0

Lab 2 10906 0

Lab 3 7627 73

Lab 4 12700 0

Lab 5 10000 0

Lab 6 10993 0

Mean 10654 12

Standard Deviation 1735 30

Table 8. Solder paste bromine test data from six different test laboratories using the EN 1458216 oxygen bomb combustion test method for two no-clean solder pastes20.

NC-A Sample (13,000 ppm bromine)

NC-B Sample (0 ppm bromine)

Lab 1 11700 0

Lab 2 10906 0

Lab 4 12700 0

Lab 5 10000 0

Lab 6 10993 0

Mean 10993 0

(Lab 3 removed) 11260 0

Standard Deviation (Lab 3 removed)

1006 0

Table 9. Solder paste bromine test data from six different test laboratories using the EN 1458216 oxygen bomb combustion test method for two no-clean solder paste with the outlier Laboratory 3 test data remove20.



Table 9, the results for Laboratory 3 were omitted showing a relatively close set of bromine data results for NC-A sample paste for the five laboratories.

These findings are very useful, but a study of results for solder pastes that were closer to the halide-free pass/fail limit of 900 ppm Br and 900 ppm Cl would be more beneficial in determining the probability of false failures being reported. It would also be beneficial to know the accuracy limit, reproducibility (inter-laboratory and laboratory-to-laboratory) and uncertainty limits surrounding the acceptance levels of 900 ppm Br and 900 ppm Cl and total Br and Cl of 1500 ppm.

Based on this data review, an industry-wide gage repeatability and reproduc-ibility study is needed prior to establishing preferred halide test methodology and halogen-free pass/fail test limits.

ConclusionsIn terms of trends for halogen-free defini-tions and standards, most of the standards for components, boards and materials use either 900 ppm or 1000 ppm Br or Cl as the definition for halogen-free. Many OEMs use the 900 ppm Br, 900 ppm Cl and 1500 ppm total Br + Cl criteria in specifying halogen-free products. This is close to the restriction requirements for substances such as PBDE (Polybrominated Diphenyl Ethers) and PBB (Polybrominated Biphenyls) and lead mentioned in the European Union RoHS legislation which indicates less than 1000 ppm.

The differences between halogen-free definitions have varied based on different dates of standard publication from around the world as well as different amounts of data available in the determination of halogen-free. As the halogen-free defini-tions varied, the test methods by which to measure these halogens have also varied. There has been a movement to use oxygen bomb combustion testing followed by ion chromatography analysis. Based on the data reviewed, there have been varia-tions seen in test results for halogens in soldering materials based on laboratory to laboratory test differences.

future workFuture work would include conducting round robin testing to address inter-labo-ratory test variation. Testing would take place using raw and reflowed flux samples. The samples would be prepared for analysis using the EN14582 oxygen bomb test method15 as this would appear to be the most repeatable. The proposed round robin

testing would include samples which were halogen-free as well as samples containing 900 ppm Chloride and 900 ppm bromide. At least one of the halogen containing compounds could be run multiple times at each laboratory over several days to determine test method reproducibility. The main focus of this study would be to determine the source of any inter-labora-tory variability and how to resolve these discrepancies.

As already indicated, work should be done to standardize the halogen-free defi-nitions between IEC, JEITA and IPC stan-dards. Also, the determination of halogen content using the oxygen bomb combus-tion test method followed by ion chroma-tography testing on different soldering, board and component materials would be of benefit.

AcknowledgementsThe authors would like to thank the various persons involved in developing the data and standards discussed in this paper.

references1. US Federal Specification QQ-S-571

Solder, Electronic (95 to 485 deg. C) 2. MIL-F-14256 standard, Military

Specification: Flux, Soldering, Liquid, Paste Flux, Solder Paste and Solder-Paste Flux (For Electronic/Electrical Use).

3. IPC-SF-818 standard, General Requirements for Electronic Soldering Fluxes, 1991.

4. IPC J-STD-004 standard, Requirements for Soldering Fluxes, 2008.

5. JEITA ET-7304 standard, Definition of Halogen-Free Soldering Materials, 2009.

6. IEC 61249-2-21 standard, ED. 1.0 B: 2003 Materials for printed boards and other interconnecting structures – Part 2-21: Reinforced base materials, clad and unclad – Non-halogenated epoxide woven E-glass reinforced laminated sheets of defined flammability (vertical burning test), copper clad, 2003.

7. JPCA-ES-01 standard Halogen-free copper clad laminate test method, 2003.

8. IPC 4101 standard, Specifications for Base Materials for Rigid and Multilayer Printed Boards, 2009.

9. JEDEC JEP709 standard: A guideline for defining low halogen solid state devices (removal of BFR (Brominated Flame Retardant)/CFR (Chlorinated Flame Retardant)/PVC (Polyvinyl Chloride)), 2010.

10. JEITA ET-7304A draft standard,

Definition of Halogen-free Soldering Materials, 2010.

11. IPC TM-650 2.3.33, Presence of Halides in Flux, Silver Chromate Method, 2004.

12. IPC TM-650 2.3.35.1 Fluorides by Spot Test, Fluxes – Qualitative, 2004.

13. IPC TM-650 2.3.35 Halide Content, Quantitative (Chloride and bromide), 2004.

14. IPC TM-650 2.3.35.2 Fluoride Concentration, Fluxes – Quantitative, 2004.

15. IPC TM-650 2.3.28 Ionic Analysis of Circuit Boards, ion chromatography Method, 2004.

16. EN 14582 standard Characterization of waste. Halogen and sulfur content. Oxygen combustion in closed systems and determination methods.

17. T. Jensen et.al., Internal company study on Oxygen bomb combustion testing of solder paste.

18. EPA SW-846 5050/9056 standard, Bomb/ion chromatography method.

19. B. Toleno et.al., Internal company study on Oxygen bomb combustion testing of adhesives.

20. K. Seelig et.al., The Call for Halogen-Free Electronic Assemblies, www.aimsolder.com


Title

Milara Incorporated 4 Marc Road Medway, MA 02053

Phone: 508.533.5322 Fax: 508.533.8686 www.milarasmt.com Email: [email protected]

It’s LEAN It’s MEAN and ready to

PRINT

Our latest creation, the TouchPrint TD2929-E “Essentials” model was created for two specific reasons, to give you lightning fast speed with accuracy you demand. Milara has streamlined our high-end printing systems to deliver the finest quality printing the industry has yet to experience. MEAN because this platform is structurally stable combined with overall rigidity to ensure accuracy at high production speeds. LEAN because we only put in what you need to produce a high quality product and to hit the price points you require. Although lean, Milara still offers the best of our technology found in our high end system at an affordable price. Let us prove what our Essentials model will do for you.


Improved efficiency using root cause failure analysis

IntroductionRoot cause failure analysis (RCFA) is the process of determining the source of a problem. In many industries, especially those involving complex machinery, it is common to use root cause failure analy-sis when problems occur. Why is RCFA so important? Efficiency and accuracy. When it comes to the complicated and expen-sive machines used in the PCB assembly process, downtime equals increased costs, lost profits, missed deliveries, lost oppor-tunities or all of these. Better efficiency is one of the few ways to compete with lower labor cost manufacturers. With so many machines’ specifications looking so similar on paper, features that allow end users to improve efficiency in the production pro-cess are gaining importance in the selection process. The use of RCFA tools can assist in improving utilization, efficiency and maxi-mizing profitability.

Several systems to improve the produc-tion efficiency of the placement machine have been introduced over the years. There are tools designed for the direct improve-ment of utilization and indirect improve-ment. Twenty years ago, few pick and place machines had quick change feeder trolleys.

Why were they invented? To improve effi-ciency so the operator could change from one product to another faster. Intelligent feeders help the operator setup feeder trolleys offline and reduce the need for “buddy system” reviews of the feeder setup. Splicing feeders and tray changers that can have parts replenished without stopping the machine also were created. All of these systems were designed to directly improve machine utilization and are now common in the industry. Other tools such as self-cal-ibrations are indirect efficiency improve-ment tools. They reduce the chances of an error or the time needed for maintenance, but do not directly impact utilization. One thing these systems all have in common is that they are designed to improve the effi-ciency of the known or expected downtime.

RCFA deals more with the unplanned downtime due to unknown or unexpected causes. A failure has occurred. The goal is to find the problem as quickly as possible and get the machine back into production. These tools can be separated into two cat-egories also: after the fact and real time. After-the-fact tools include self-diagnostics that can be run after a problem occurs to see if a system is performing properly.

This paper was originally presented at IPC APEX Expo 2011.

A PCB fails final test. Why? Was it the solder paste? The screen printer? The PCB assembly machine? The reflow oven or none of the above?

Unplanned downtime is a costly fact of life. In order to minimize the length of downtime it is necessary to have clear details on exactly what the source of the problem is, not just the symptoms. This information allows operators and main-tenance personnel to go directly to take corrective steps more quickly and mini-mize downtime.

There are many machines involved in the assembly of a complete PCB: screen printers, conveyors, pick and place sys-tems, reflow ovens, and Automated Optical Inspection (AOI). Some of this equipment has the ability to check its results before, during or immediately after it has completed its task.

Until now there have been few real-time tools for the pick and place systems. In many cases, high speed movement on these machines makes it extremely hard to “see” exactly what is happening. Components misplaced by the assembly system could be caused by many different factors. Without tools to provide a clear view of very high speed placement, it is difficult to determine the cause of mis-placement.

How much easier would this task be if operators and maintenance personnel were armed with detailed information on the nozzles, feeders, and actual images of the picking and placing of parts on the PCB?

This paper will discuss tools available for the placement machine to assist in root cause failure analysis (RCFA).

Gerry Padnos, Juki Automation Systems, Inc. Morrisville, North Carolina, USA


Figure 1. Component present in tape and uncovered by cover tape.



Several tools exist to monitor machine per-formance in real-time such as tombstone detection and nozzle inspection, but these do not provide any insight into the cause of a problem. They only detect that there is a problem. Few systems exist to monitor the machine’s performance in real-time and provide useful information as to the cause of a problem. Unplanned downtime could be dramatically reduced with better root cause failure analysis tools.

The key to improving efficiency when unplanned downtime occurs is root cause failure analysis; determining the exact source of the problem. Knowing exactly what is causing a problem will allow main-tenance personnel to focus their efforts on a clear target rather than just follow-ing a checklist of possible problems. Many people have had the unpleasant, inefficient experience of talking to phone tech sup-port personnel who run down a checklist of “possible causes” for different problems. But the reason the checklist is needed is that there is no data available about the root cause of the problem. Without know-ing the root cause of any problem, main-tenance personnel are essentially working with blindfolds on. New systems designed specifically to give better insight into prob-lems can significantly reduce the time tech-nicians need to spend diagnosing prob-lems, and improve equipment utilization.

defect prevention in assembly linesOne of the most common errors from the placement machine is missing or incor-rectly placed components. There are two main types of defect prevention systems found in typical assembly lines: systems built into a machine to check itself and stand alone equipment whose sole func-tion is to check for defects. An example of a system within a machine to check itself would be post-print inspection on a screen printer. This system checks for defects after the printer has completed its task of applying solder paste to the PWB. This “post print inspection” cannot say why there isn’t enough solder paste in a certain location, only that there isn’t. Wouldn’t it be much more useful to know why there isn’t enough solder paste on the board? Clogged apertures, dry paste, squeegee pressure and many more problems could be the cause. Similarly, typical placement machines have several features to prevent defects, but most cannot provide insight as to why the defect occurred in the first place. Vacuum sensors are commonly used to detect when a component falls off the nozzle or isn’t

released on the board. Lasers can check for tombstoning, dropped components, and even incorrect or damaged nozzles. But there are times within the produc-tion cycle when these sensors cannot pro-vide useful root cause information. If the vacuum sensor reports an error, why did it occur? All the technician knows is that the required vacuum level was not achieved. Was the component not aligned with the pick position? Is there a problem with the vacuum system? Is the nozzle damaged? Is the feeder feeding correctly? Was the com-ponent even present at the pick position? All of these possibilities have to be consid-ered, but each takes time to evaluate.

An example of a machine that is used solely to check for defects would be an AOI or SPI machine. These systems all check for problems that were caused by another machine in the line. They detect missing, incorrect or misplaced components, sat-

isfactory solder joints, etc. AOI machines provide very useful evaluation of the final PCB, but also have limitations. Like the systems described above, they cannot tell why a component is missing or incorrectly placed. If an image shows a component placed at the wrong angle, why did this happen? Was the feeder loaded incor-rectly? Was the production program writ-ten correctly? Was the component prop-erly oriented in the tape, tube or tray? The images from AOI are only of the results, not the process. In addition, typically there is only one AOI machine for the entire line so when a component is missing, the oper-ator doesn’t even know which placement machine should have placed the compo-nent. The key to improving the efficiency of troubleshooting a missing or misplaced part is having actual images to see what happened during the pick and place pro-cess.

Figure 2. Image of component in tape allows technician to verify component presentation angle.

Figure 3. Images before and after picking. Note that position on nozzle can be seen in the after image to confirm acceptable pick location



reasons for failuresFocusing specifically on the placement machine, there are many reasons why a component may not be placed correctly on the PCB:

• Damaged nozzle,• Feeder malfunction,• Component incorrectly packaged

(wrong orientation in feeder), • Incorrect pick position teaching,• Incorrect component data,• Improper board support, or• Components sticking to the

nozzle.

Due to the high speed of today’s placement systems, it is extremely difficult for an oper-ator to know which of the above is actu-ally causing the problems. Therefore they are required to follow a pre-defined list of troubleshooting steps to eliminate one pos-sible cause after another. Throughout the process, they are completely unaware of the root cause. In the end, if they haven’t found the problem, they just guess or use experi-ence. This process is clearly imprecise and inefficient. Throughout the troubleshoot-ing period, the machine is down and the line is stopped.

The time required to precisely locate the source of the problem could be dra-matically reduced through better RCFA tools. One example is the use of simple pic-tures taken during the pick and placement process. Images of the various stages of the pick and place process instantly show the technician what happened and where the problem came from. This is made possible using tiny embedded high speed micro-cameras along with sophisticated image analysis. Along with the visual analysis by the technician, the software can provide automatic alerts when an error occurs and perform some RCFA on its own.

By capturing a series of images during the pick and placement operations, the technician can quickly see the following. While some of the questions below are the results and not the cause, they still help reduce the time the technician needs to diagnose the root cause of the problem.

1. Was the nozzle damaged when it picked the part?

2. Was the component present in the tape?

3. Was the component in the correct/expected orientation prior to pick?

4. Was the cover tape properly peeled back?

5. Did the nozzle lift the part out of the tape successfully?

6. Does the position of the part on

Figure 4. Top: Image showing solder paste prior to component placement. Middle: Image showing solder paste prior to component placement. Bottom: Component placed on PCB correctly.


TitleZESTRON Senior Cleaning Engineers

Aerospace, Military, Automotive, Medical, Communications – if you are in this line of business and the reliability of your

assemblies matters most – cleaning is a requirement. Whether implementing a new cleaning process or optimizing

an existing one, our local engineering teams stand ready to support you. With more than 30 years of experience and

over 2000 processes installed worldwide, they will find the most reliable cleaning solution for your requirements.

… ABOUT YOUR CLEANING PROCESS RELIABILITY

WE CARE...

Defluxing ∙ Stencil Cleaning ∙ Maintenance Cleaning Global Reach ∙ Local Support

For more info

www.zestron.com/gsea/we-care

[email protected]

1103-global-smt-sea.indd 1 22.03.2011 15:52:35



the nozzle indicate the pick posi-tion is good?

7. Was the part present and properly oriented on the nozzle just prior to placement?

8. Was sufficient solder paste present on the pads?

9. Did the nozzle come down enough/too much when placing the part?

10. Did the component release prop-erly when the nozzle moved up after placement?

In the time it takes to view a few pictures, the technician can accurately determine all of these points. The adage of a picture being worth a thousand words, or maybe a thou-sand seconds, is appropriate for this situ-ation. Checking just a single item without the pictures would take as long as checking for all of these points with pictures.

Compare the process of isolating the problem for a missing component using improved RCFA tools versus manually in Table 1.

TechnologyWhile RCFA is not necessarily a new con-cept, recent technology developments have made it much more affordable and effective. The use of digital cameras has exploded in recent years. Cell phones drove the wide-spread miniaturization of cameras to the point where it has become practical to embed them into very small spaces, includ-ing the placement head of a pick and place machine. Cameras are now widely used in situations where it isn’t practical or possi-ble for a human to watch something 100% of the time. Reports show that London has 10,000 security cameras throughout the city. Why? To help law enforcement find information that is simply not avail-able using people on the street because it isn’t practical or economical. This is why cameras are well suited. The days when a placement machine moved so slowly that a human could observe the process are long past. The movement of the typical pick and place machine is faster than any other machine in the production line, making it difficult to monitor the activity of the machine. The machine also moves over a large physical space making it impossible for a single camera to accurately see what actions are happening. Small, high speed cameras are needed and they have to be able to move with the placement head as parts are picked in one location and placed in another. Watching by eye would be extremely difficult due to the speed and design of most placement machines.

A component may appear to bounce after placing, but did it happen because the part temporarily stuck to the nozzle? Or was there too much placement force and the component? The still images can show what is otherwise impossible to see.

The size, field of view, and resolution are all critical technologies for the camera.

It must be small enough to fit inside the placement head without adding significant weight which would impact the through-put or design of the placement head. The field of view must be wide enough to cap-ture the component in the feeder and the component in the up position after being picked. At the same time, the resolution

Item using improved rCfA tools Task with manual rCfA

Was the nozzle damaged when it picked the part?

1. Look at nozzle image for damage. Software identi-fies the nozzle used for the missing component automatically.

1. Look up which nozzle was used for the missing component

2. Remove the nozzle from the machine and exam-ine it

Was the component present in the tape?

1. Look at tape pocket 1. Almost impossible to determine during high-speed assembly.

2. In some cases slow the machine down to watch.

Was the component in the correct/expected orienta-tion prior to pick?

1. Look at orientation of component in tape in “before” pick image

1. Almost impossible to determine during high-speed assembly.

2. In some cases slow the machine down to watch.

Was the cover tape properly peeled back?

1. Look at “before” pick image

1. Stop production and use teach camera to check pick position. If cover tape is present, it is likely the cause.

Did the nozzle lift the part out of the tape success-fully?

1. Look at the before/after pick images. Software automatically detects when the part isn’t picked.

1. use vacuum sensing and/or laser presence check.

Does the position of the part on the nozzle indicate the pick position is good?

1. Look at “after” pick image

1. Almost impossible to see on high speed machine

Was the part present and properly oriented on the nozzle just prior to place-ment?

1. Look at “before place-ment” image


Was sufficient solder paste present on the pads?

1. Look at pads in “before placement” image

1. use SPI machine or func-tion on screen printer OR

2. Check after using AOI or visual inspection

Did the nozzle come down enough/too much when placing the part?

1. Look at “during place-ment” image


Did the component release properly when the nozzle moved up after placement?

1. Look at “after place-ment” image

1. use vacuum and/or laser release check sensor

Table 1. Isolating the problem for a missing component using improved RCFA tools versus manually.



must be high enough to allow the image processing software to accurately detect even subtle differences required for accu-rate analysis.

While the images are the key to locat-ing the root cause, the software is also criti-cal in warning the operator of an error and providing analysis that may not be possible by eye. Powerful software analyzes each image to detect when the component is not picked, is not placed, and even calculates the amount of board deflection during placement. Through communication with the placement machine, it can also show trends such as identifying the feeders, nozzles, or placement sites experiencing the most problems. Although it is common

for a placement machine to identify feeder experiencing higher than normal errors, nozzle and placement errors are not as easy for typical placement systems to identify. The analysis of the images is what makes these functions possible or more accurate. This information can also help the opera-tor determine the root cause of problems more quickly.

Finally, the software must be accu-rately synchronized with the movement of the placement machine. Images must be captured in very high speed at very specific points in time; the exact moment the nozzle goes down to pick the part, the moment the head goes down to place the part, etc. This requires synchronization and

very high speed communication between the camera software and placement hard-ware.

ConclusionNew tools are invented constantly to “make life easier”. In a production environment, this often translates to “more efficient”. A compound miter saw makes a carpenter more efficient in his work. E-mail makes communications more efficient. Root cause failure analysis is a tool that can be used in many situations and industries to improve efficiency of troubleshooting. Technicians can quickly find the source of problems so they can be fixed fast and fixed right the first time. Airplane black boxes and cockpit recorders allow investigators to determine what happened even when there is no one to tell them anything. Cars now have some similar functions. Now imagine if every airplane or car had a camera and show-ing the interior for crash analysis. It would probably be even more effective than a simple voice recording. Similarly useful tools are now available for the high speed SMT assembly line with the same ben-efits; clear and quick understanding of the root cause of the problem, which leads to improved efficiency in fixing the problem. While some sensors and self-diagnostic tools exist, they are generally do not focus on finding the root cause. Root cause fail-ure analysis tools such as embedded micro cameras to show the exact details of the pick and place process along with powerful analysis software can save a huge amount of time versus methods commonly used today and dramatically increase efficiency and equipment utilization.

Figure 7. Example of micro-cameras embedded into placement head.

Figure 8. Analysis software showing results of all placements on a PCB along with statistical data.


2h’11: uncertainty prevails but modest growth remains likely

As we enter the second half of 2011, the near-term outlook is not clear. Arguments can be made for either

a double-dip recession or a 2H’11 upturn. We believe the later, but uncertainty still exists.

Per Chart 1 we are currently at the “zero growth” portion of the business cycle. The global economy and electronic equipment are still expanding at modest single digit rates, but electronic component and mate-rial growth rates are near “break even.”

Converting Chart 1 to real data, Chart 2 shows three-month (3/12) growth rates for global electronic equipment, printed cir-cuit board and semiconductor shipments compared to the global purchasing manag-ers’ index (PMI) as a leading indicator. The PMI suggests that “things will get worse before they get better”—at least short term.

world electronic equipment productionPer Chart 3 Taiwan/China electronic equipment shipments increased 8.1% (3/12 basis) in May 2011. In April (latest data available), Europe was up 10.6%, Japan was down 20.2% and the USA rose 2.8% (Chart 3). Recent performance is not encouraging but globally (Chart 4) electronic equipment

shipments should increase in the second half of this year, both due to normal sea-sonality and Japan’s recovery efforts from its March earthquake. It would appear that regional end-market growth rates may “bottom” in the single digit range in 2011 (except for Japan).

electronic componentsWorld semiconductor shipments peaked in early 2011 (Chart 5) but should soon see a seasonal upturn. Gartner recently pre-dicted 5.1% chip growth for 2011 (Chart 6). Printed circuit boards are behaving simi-larly. Current regional PCB growth rates (Chart 7) are approaching zero (except for Japan) but normal seasonality suggests a 2H’11 rebound leading to 5% global PCB growth in 2011 vs. 2010 (Chart 8).

forecastsChart 9 summarizes recent forecasts for the electronic supply chain. Assuming that Japan continues to recover and/or its critical components are available to world assemblers (from Japan or another source) and also that the world avoids major tur-moil and/or an economic meltdown we still see 2011 as a low growth (but not disas-trous) year.

end marketsChina’s top 100 companies produced 26.82 million computers, 73.67 million colored TV sets, and 174.52 million cell phones in 2010.—Ministry of Industry and Information Technology

Computers & peripherals• Worldwide PC shipment forecast was

adjusted down to 9.3% y/y growth and 385 million units by Gartner.

• Global PC shipments declined 1.1% y/y to 84.3 million in 1Q’11.—Gartner

• Workstation shipments fell 4.8% q/q and grew 18.6% y/y to 860 thousand units in Q1’11.—Jon Peddie Research

• Worldwide notebook shipments declined 13.2% m/m to 15.4 million units in April.—DRAMeXchange

• Worldwide server shipments grew 8.5% to 2.3 million units; revenue increased 17% y/y to 12.6 billion in 1Q11.—Gartner

• Computer hard drive revenue for PCs and servers are projected to grow 4.1% to US$28.1 billion in 2011.—IHS iSup-pli

• Total disk storage systems capacity shipments grew 46.3% y/y to 4,956 pet-abytes in 1Q’11, sales revenue increased 13.2% y/y to $5.6 billion.—IDC

• Worldwide enterprise storage sys-

Walt Custer and Jon Custer-Topai

2H’11: uncertainty prevails but modest

growth remains likely

BUSINESS CYCLE Supply Chain Effect

%GrowthOEM CEMPCBs Raw Material

0

RecessionExpansion

InventoryIncreases

InventoryReductions

Time

20050118

We are here

World Global PMI, Electronic Equipment, PCB & Semiconductor Shipments

Converted @ Constant 2008 Exchange Rates

3 6 9123 6 9123 6 9123 6 9123 6 9123 6 9123 6 9123 6 9123 6 9123 6 9123 6 9123 600 01 02 03 04 05 06 07 08 09 10 11

CALENDAR YEAR

0.5

0.7

0.9

1.1

1.3

1.5

3/12 rate of change

PCB "0" Growth SIAEl Equip Global PMI

Source: Custer Consulting Group

20110619

Chart 1. Chart 2.



tems market is expected to grow from $30.8 billion in 2010 to $37.3 billion in 2015.—IDC

• Worldwide external controller-based disk storage market grew 14.1% y/y to $5 billion in 1Q11.—Gartner

• DSL, PON, and Ethernet FTTH equip-ment market declined 2% to $1.96 bil-lion in 1Q’11.—Infonetics Research

• Ethernet switching market revenue was down 12% sequentially and 9% y/y to $4.2 billion in 1Q’11.—Infonetics Research

• Worldwide Ethernet switch market declined 9.1% y/y in 1Q’11; router market grew 6.9% y/y.—IDC

Mobile communications• Worldwide smartphone shipments are

expected to grow from 472 million in 2011 to 982 million units in 2015.—IDC

• Worldwide mobile communication device sales grew 19% y/y to 428 mil-lion units in 1Q’11.—Gartner

• Mobile broadband network infrastruc-ture annual investment is expected to

grow from $22 billion in 2011 to $48 billion in 2014.—Strategy Analytics

Consumer electronics• Global total set top box shipments

(including DTT, IPTV, cable, and satel-lite connections) are set to grow from 177m in 2009 to over 207m in 2014.—Business Insights

• Global TV shipments were down 29% Q/Q and 1% y/y to 55.2M units 1Q11.—DisplaySearch

eMS, odM & related assembly activityChina’s ODMs and EMS providers raised factory prices by between 5-10% to bal-ance increasing material and operation costs, but the end result caused them to lose market share in 1Q’11.—IHS iSuppliAutomated Circuit Design • added two new Juki lines and 13,500 SF

to its existing facility.• purchased four PCB vertical car-

rier carts with 20 carriers from Bliss Industries.

Badger Technologies is adding 16,600 SF and renovating its existing facility in Farmington, NY.Becom Electronics is investing HUF 480 million (EUR 1.8 million) to construct a new production hall in Hungary.Flextronics • donated $250K for Japanese Disaster

Relief.• plans to lay off 254 workers in

Massachusetts by Aug. 31, 2011 as a result of Verizon Wireless contract ter-mination.

Foxconn/Hon Hai• expanded production by 50% at its

LCD manufacturing facility in Nitra, Slovakia.

• halted plans to manufacture tablet computers in Hengyang, China.

• was removed from Hong Kong’s bench-mark Hang Seng index.

• invested in WIMM Labs.• is building a US$2.5 billion LCD moni-

tor factory in Chongqing, China.• laid off 225 employees from its FIHTK,

CDMA handset design subsidiary in Korea.

Global Electronic Equipment Shipment Growth

1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 400 01 02 03 04 05 06 07 08 09 10 11

CALENDAR YEAR

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.63/12 rate of growth in local currency

Taiwan/ChinaEuropeJapanUSA0 Growth

20110619

Europe = Eurostat EU27 NACE C26 (computer, electronic & optical products)

World Semiconductor Shipments Monthly US$

SIA

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 183 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11

0

5

10

15

20

25

30US$ Billions (3-month average)

20110530

2009 recession much sharper but shorter than 2001

Record high

World Electronic Equipment Monthly Shipments

Converted @ Constant 2008 Exchange Rates

1 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 5 7 9111 3 500 01 02 03 04 05 06 07 08 09 10 11

CALENDAR YEAR

60708090

100110120130140150160170180190

$ Billions

5

55

105

155

Thousands

Source: Custer Consulting Group

20110619

2010/2009 growth +20.9%

1Q'11/1Q'10 growth +4.6%

World Semiconductor Sales

255228

299 315

2008 2009 2010 20110

50

100

150

200

250

300

350$ Billions

Gartner 6/11

Total

20110622

-10.5% +31.1% +5.1%

Chart 3. Chart 4.

Chart 5. Chart 6.



• plans to invest $12 billion in Brazil over next five years.

• had a large explosion and fire at its Chengdu, China, factory.

• is investing US$80 million in its Chinese subsidiary Zhengzhou of Henan Province.

• is splitting into small firms to be com-petitive.

• was Taiwan’s largest enterprise in terms of revenue in 2010.

Hanza received a 34 MEUR defense order from Saab.IEC Electronics and Benchmark Electronics joined the Aerospace Industries Association.IMI acquired EPIQ assets in Bulgaria, Mexico & Czech Republic.Incap • became a supplier of electronic mod-

ules for Aidon’s smart metering and smart grid devices.

• received contract to manufacture electronics for sauna heaters made by Tulikivi.

Jabil Circuit • hired 670 workers and expanded oper-

ations in Hungary to increase produc-tion of high-speed mobile data trans-mission networks.

• received a manufacturing contract for Electro Precipitation Integrated Water Systems from Latitude Solutions.

Kimball Electronics purchased a SMT line with SIPLACE SX placement machines and SIPLACE CPP MultiStar heads in Tampa, Florida. Kitron • received a $3.9 million (NOK 22.5 mil-

lion) contract from Lockheed Martin to produce Integrated Backplane Assembly for the F-35 low rate initial production program.

• received US$ 3.2 million contract

from Kongsberg Defense Corporation to deliver electronics for common remotely operated weapon station.

Melecs opened a EUR 3 million manufac-turing facility in Gyor, Hungary.Nortech Systems named Jill Hesselroth VP of global supply chain and John Shelander senior business development manager of aerospace systems operations.On-Track Technology added a Rehm Thermal Systems VXS422 convection reflow solution.Partnertech received a U.S. development and production contract from Tomra Systems.Protonic Holland BV added a new Mydata MY100DX14 pick-and-place machine.Sanmina-SCI• former president and COO, Hari Pillai,

will continue working for the company in an advisory capacity through May 5, 2013 for $335,000 per year.

• received Strategic Supplier of the Year award from Ciena.

Scanfil • EMS closed its Vantaa, Finland, opera-

tions.• changed its name to Sievi Capital plc.SMT Holding ceased operations at its Changchun factory and scaled down oper-ations in Suzhou and Dalingshan.TES Electronics Solutions, Langon, France, was broken up and sold off to three different companies by order of the Tribunal de Commerce de Rennes.The Paragon Electronics Group received a contract to manufacture traffic signals and controllers for Peek Traffic.TPV will invest US$90 million to establish a factory in Chengdu, China, with produc-tion to begin in 4Q’12.Videoton • added three complete Fuji NXT inser-

tion lines, welding machines and auto-

matic optical test equipment.• acquired direct majority ownership in

STS Technology Kft, Győr, Hungary.Wistron is setting up a plant to produce notebook and desktop PCs, LCD TVs and communications devices in Chongqing City, China.

PCB fabricationJapanese are investing more than $1.0 billion in overseas PCB expansion in FY2011.—Dr. Hayao NakaharaTaiwan’s PCB production value is expected to increase 13.5% y/y to US$6.15 billion in 2011 due to order transfers resulting from March 11 Japan earthquake.—IEKWorld PCB production value will reach $59 billion in 2012 based on WSTS Semiconductor projection of $338 bil-lion.—Dr. Hayao NakaharaAll Flex added 5,000 SF plus capital equip-ment to its second production facility in Bloomington, Minnesota.American Standard Circuits received polyimide qualification under MIL-PRF-55110, amendment 3 and MIL-PRF-31032.APCB is building a second plant next to its existing plant in Kunshan, China.Aspocomp • appointed Tero Päärni as head of sales

and marketing and Antti Kangas qual-ity manager at Aspocomp Oulu Oy.

• purchased 10% of the shares in Aspocomp Oulu from TTM Technologies’ subsidiary MTG. MTG paid EUR 14.5 million for 20% of the shares in Meadville Aspocomp Holdings.

AT&S • bought a plater from Atotech for its

Cu18 IP2 line.• is building a microvia (conventional

ANYLAYER boards and ALIVH

Regional PCB Shipment Growth

1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 4 7 10 1 400 01 02 03 04 05 06 07 08 09 10 11

CALENDAR YEAR

0.40.50.60.70.80.9

11.11.21.31.41.51.61.71.81.9

3/12 rate of growth in local currency

Taiwan/ChinaEuropeJapanN America0 Growth

20110619

Sources: IPC, JPCA, Taiwan/China composite; modified SIA chip shipments to approximate Europe

World PCB Shipments (with forecast)Converted @ Constant 2008 Exchange Rates

1357911135791113579111357911135791113579111357911135791113579111357911135791113579111357911199 00 01 02 03 04 05 06 07 08 09 10 11 12

CALENDAR YEAR

20

30

40

50

60$ Billion

Source: Custer Consulting Group - base year expanded by monthly growth of N. American, European, Japanese & Taiwan/China monthly PCB shipments

20110619

51.742.937.631.729.4

Assumptions:Europe = composite European SIA & local PCB assoc dataJapan & N. America from JPCA & IPC dataTaiwan/China based upon 44 rigid & flex company compositeRest of Asia growth = Taiwan/China 44 company compositeData scaled to match Henderson Ventures annual totals2008 based upon sum of monthly totals

38.6 31.231.6 54.3 51.1 41.0 49.0

-20%

+19%

51.5

+5%

Chart 7. Chart 8.



board) plant in Chongqing, China.Bare Board Group released “Printed Circuit Boards for Dummies.”Beta Layout acquired Sprig Circuits.Camtronics Vale Limited was acquired by PhotonStar LED Group.Cicor • Electronic Solutions Division

appointed Erich Trinkler executive VP and head of the Cicor Electronic Solutions division.

• established Suzhou Cicor Technology in Suzhou, China.

Compeq is moving its rigid-flexible PCB plant in Taoyuan, Taiwan, to its HDI board plant in Huizhou, China. Eurotech acquired Dynatem.Founder Technology is building a second plant next to its existing plant in Chongqing, China.Global Brands Manufacture is expanding its monthly PCB production capacity from 4.1 to 4.7 million SF/month.Gold Circuit Electronics is investing US$24 million to expand its monthly HDI production capacity in Changshu, China, from 50,000 to 200,000 SF by 4Q’11.Graphic Plc • installed a hybrid collimation accutray

system for solder mask imaging at its Crediton, England facility.

• received a special ‘suppler recognition award’ from BAE systems.

GUH is investing RM16 million to expand 2-sided and 6-layered PCB bare board pro-duction in Penang and Suzhou, China to 50,000 m2 in 2H’11.HannStar Board • is building PCB plants in Chongqing,

China, with 3,000,000 SF/month capacity.

• transferred over-booked orders to Global Brands Manufacture.

Ibiden is spending RM1billion to build second 40,000 m2/month smart phone motherboard plant in Penang, Malaysia.Kinsus Interconnect Technology received NT$40-50 million (US$1.39-1.73 million)/month in revenues from FC CSP substrate orders from Intel.Meiko Electronics will install an automo-tive circuit board production line in Hanoi, Vietnam, in fiscal 2011.MFlex added a 290,000 SF facility in Chengdu and 610,000 SF plant in Suzhou, China.Multek • named Bill Beckenbaugh CTO.• was certified to manufacture PCBs with

Shocking Tech’s Voltage Switchable Dielectric™ (VSD) material.

Nippon Mektron exhibited a stretchable flexible printed circuit board prototype.

Panasonic is investing 10 billion yen in Taoyuan County, Taiwan, to increase its ALIVH high-density multilayer board manufacturing capacity.Prestwick Circuits GPS was named PCB global strategic partner for TT electronics.Shanghai Unitech added more laser drills after receiving large-volume HDI board orders that pushed PCB production lines to capacity.Taiwan PCB Techvest • and T-Flex Techvest PCB are jointly

investing US$17.0 million to estab-lish a 1.2 million SF/month PCB JV in Suining, China.

• will ramp up total capacity of its plants in China by 20% to 6.6 million SF/month by the end of 2011.

TC Cosmotronic expanded inner layer imaging, inner layer develop/etch/strip, inner layer AOI and alternative oxide application and added CAM and film plot-ting in Fullerton, California.Transline Technology acquired Hi Electronics.TTM Technologies • acquired remaining 20% stake in

Meadville Aspocomp Holdings.• Guangzhou facility was awarded

Nadcap AS9100B accreditation for aerospace quality.

Unimicron is building a plant for its German subsidiary RUWEL International in Kunshan, China.Wus Printed Circuit will open a plant in Kunshan, China, in 2012.

Materials & process equipmentSMT equipment market is expected to be in excess of $6.5 billion by the year 2017.—Global Industry Analysts3M named George Buckley COO.Agilent Technologies introduced the first modular embedded controller for AXIe systems based on the AdvancedTCA bus.Air Products • acquired Poly-Flow Engineering.• opened an electronic materials facility

in Banwol, Korea.Alpha-Cookson Electronics received the IPC APEX Expo Best Poster Paper award.Arlon Materials for Electronics intro-duced GenClad 280 bond-ply, woven fiber-glass reinforced, ceramic-filled composite material.

ASM Assembly Systems (formerly Siemens Electronics Assembly Systems) introduced SIPLACE Setup Center 4.0 for setup verifi-cation and fill level management.Baotek increased its production capacity for fiberglass fabrics to 6.5 million meters in July.BTU International • received Best Presentation award

from SMTA China for “Lowering Your Reflow Cost”.

• Zeping Zheng received “Officer of the Year 2011” award from SMTA China.

Cadence Design Systems acquired Altos Design Automation.Christopher Associates • released multiple camera retrofit kit for

Marantz AOI systems.• won 2011 NPI award for its Akila XR-3

X-Ray inspection system.Csun China subsidiary Suzhou Top Creation Machines received orders for wet processing equipment from Advanced Semiconductor Engineering, Foxconn Electronics, Gold Circuit Electronics, Kinsus Interconnect Technology, Nanya PCB and Wus Printed Circuit.Dow Corning named Robert Hansen CEO. Stephanie Burns remains chairman.Enthone • appointed Robert Haskins Sr. VP and

regional managing director, Asia.• was named “Top 10 International

Electroplating Brand”• Amy Tsang, Enthone Asia techni-

cal director—Performance Coatings was named one of the “Top 10 Most Influential People.”

Essemtec introduced Traqu, a high-resolu-tion digital inspection device for 3-D mea-surement and analysis in processes such as solder paste inspection.Fulltech is increasing its electronic grade fiberglass yarn capacity to 81,000 tons by year end 2011 with the addition of an

Global Electronic Supply Chain Forecast 2011 vs 2010

3

6

5

5

Combined GDP

Electronic Equipment

Rigid & Flex PCBs

Semiconductors

0 2 4 6-2% Change

20110619

Henderson Ventures

Gartner

Henderson Ventures

Custer Consulting Group

Chart 9.



800,000 m2/month fiber glass fabric plant.Head Invest acquired JOT Automation.Hitachi Chemical is investing 2 billion yen to double dry film photoresist manufactur-ing capacity in China to 200 million m2 per year.Indium Corporation VP of technology Dr. Ning-Cheng Lee was recognized for Best Presentation of Technology Conference One at NEPCON China (Shanghai) 2011 by SMTA.Keithley Instruments moved hundreds of manufacturing jobs from Solon, Ohio, to China.Kurabo is investing 6 billion yen to build a flexible circuit substrate manufacturing plant in Mie Prefecture, Japan.LORD Corporation introduced a low cost underfill encapsulant developed specifi-cally for the semiconductor packaging and assembly industries.LPKF • introduced its ProtoMat S-Series

entry-level milling machine for rapid PCB prototyping.

• Laser & Electronics acquired 100% shareholding of external stakes in subsidiaries LPKF Motion & Control in Suhl and in LPKF Distribution in Portland.

Marantz Business Electronics separated from its parent company D&M Europe BV in a management buyout. Its PCBA divi-sion will continue under the name of MEK Europe BV.Maskless Lithography appointed former Foxconn executive Richard Chenoweth, VP of global sales and marketing, and former Sanmina-SCI executive, William Pappani, CFO.MIRTEC • Europe appointed LTHD as Its exclu-

sive distributor for Romania.• received a 150 machine order at SMT/

PCB NEPCON KOREA.Mitsubishi Electric is expanding its PCB laser drilling system manufacturing capac-ity by 50% in 2H’11.Mitsubishi Gas Chemical restored BT-based laminate production to pre-earthquake volume.—Dr. Hayao NakaharaMitsui Metal is building an ultra-thin copper foil manufacturing line in Malaysia with a monthly capacity of 600,000 m2.Nordson • ASYMTEK introduced its Select Coat

SC-280 film coater.• DAGE established its Chinese website

www.nordsondage.com/china.• MARCH introduced MaxVIA™ plasma

treatment system.• opened a demo center in Dongguan,

China.

Orbotech • completed public offering of 7,705,000

ordinary shares at $12.50 per share.• Orbotech Pacific opened an office in

Suzhou, China.Panasonic Electric developed a halogen-free laminate series “MEGATRON2” with a low tangent loss of 0.010 for high-speed multilayer circuit boards.Park Electrochemical promoted Christopher Mastrogiacomo to Ex. VP and COO and Mark Carlson to director of sales—Americas.QSX Instruments introduced Quickshot XRF analyzer line for RoHS testing requirements.Rehm Thermal Systems introduced its CondensoX soldering system.Rogers opened a new production facility in Suzhou, China.Royce & Associates purchased a $7.5 mil-lion stake in BTU International.Seika Machinery appointed Koichi Koba executive VP and promoted Isao Muraoka to president and CEO.SIPLACE established a branch office in Bundang, Korea.Taiwan Glass is expanding its combined capacity for fiberglass yarns in Taiwan and China to 115,000 tons in 2011-2012.Tanaka Precious Metals began sale of SEA silver bonding wire.Technic• received exclusive sales and distri-

bution rights in the US, Canada and Mexico for STS vibratory technology.

• named STS Switzerland its European representative for its plating equipment products.

Tongtai Machine & Tool secured orders for NT$460 million for PCB drilling machines from tablet PC manufacturers.Topoint is investing US$6 million to set up a new PCB drilling plant in Suzhou, China.Universal Instruments relocated its European headquarters to Bratislava, Slovakia.Veeco opened its China Training Center in Shanghai, China.VJ Electronix moved into 1500 m2 facil-ity in the Suzhou Industrial Park near Shanghai, China.

Semiconductors & other components Semiconductor market is expected to grow by 7.6% to $338.4 billion in 2012.—WSTSIDC cut semiconductor annual revenue growth forecast to between 4-5% from 6-8%.

European semiconductor distribution grew 33.5% y/y to 1.81 billion Euros in 1Q’11.—DMASSJapan’s April chip sales fell 26.4% m/m and 18.8% y/y to $2.94 billion due to the March 2011 earthquake.—WSTSWorldwide semiconductor capital equip-ment spending is projected to grow 10.2% to $44.8 billion in 2011.—GartnerSEMI• Worldwide semiconductor manufac-

turing equipment billings increased 61% to US$12 billion in 1Q’11.

• Semiconductor fab equipment spend-ing is expected to rise 31% y/y to $44 billion in 2011.

• Worldwide semiconductor manufac-turing equipment sales grew 148% y/y in 2010 to $39.54 billion; semiconduc-tor materials sales grew 25% y/y to $43.55 billion.

• Worldwide semiconductor manufac-turing equipment sales are expected to grow 16% y/y to $45.81 billion in 2011.

• Worldwide silicon wafer area ship-ments increased 3% y/y to 2.29 billion square inches in 1Q’11

Semiconductor average content per auto-mobiles is expected to increase 15% to USD 350 in 2011 to $425 per vehicle in 2014.—IC InsightsWorldwide PC microprocessor unit ship-ments were up 7.4% y/y in 1Q11; IDC upgraded 2011 forecast to 10.3% growth.Connector industry’s sales grew +28.4% in 2010, its largest y/y growth in history.—Bishop & AssociatesGlobal NAND flash market grew 9.9% q/q to US$5.36 billion in 1Q’11.—DRAMeXchangeWorldwide DRAM revenues fell 4% q/q to US$8.3 billion in 1Q’11.—DRAMeXchange

Walt Custer is an independent consultant who monitors and offers a daily news service and market reports on the PCB and assembly automation and semiconductor industries. He can be contacted at [email protected] or visit www.custerconsulting.com.

Jon Custer-Topai is vice president of Custer Consulting Group and responsible for the corporation’s market research and news analysis activities. Jon is a member of the IPC and active in the Technology Marketing Research Council. He can be contacted at [email protected].


Title


X-ray inspection of semiconductor devices that use copper wire interconnections

In addition to the standard electronic packages that have been available for many years, new packages continue

to become available which have increased complexity in terms of their functionality as well as their construction. Until recently, whatever the package, be it a simple SOIC or a more complicated package on package (PoP) configuration (perhaps with multiple stacked die), all of the internal interconnec-tions within the packages were made with gold wires. However, copper wire is now increasingly being used to replace gold as the interconnection of choice, primarily because of its reduced cost (Figures 1 and 2). Part of the quality control standards used during device manufacture is to check for the presence and alignment of the inter-connection wires, especially after the outer package has been molded. This is accom-plished, non-destructively, by x-ray inspec-

tion. X-ray inspection also will be used once the device has been assembled onto the printed circuit board for failure analysis or counterfeit prevention applications, so that, once again, the insides of the package can be investigated non-destructively prior to further, possibly destructive, investiga-tion.

X-ray inspection is an analytical tech-nique that provides images based on the relative absorption of the x-rays by the various material densities contained within the sample. Thicker, and more dense mate-rials (e.g. solder paste), absorb the x-rays far more completely compared to thin¬ner, and lower density materials (e.g. fiberglass with embedded copper tracks), thereby providing a range of gray scale densities in the image to enable analysis of distinct fea-tures/flaws.

The use of gold, with its high density,

Copper wire is increasingly being used to replace gold as the intercon-nection of choice. This article dis-cusses the use of x-ray inspection to check for the presence and alignment of interconnection wires.

David Bernard, Ph.D., and Evstatin Krastev, Ph.D., Nordson DAGE, Aylesbury, Buckinghamshire UK


Figure 1. Flat panel detector x-ray image of equivalent components: 35 μm diameter copper wire interconnections (left) and 30 μm diameter gold wire interconnections (right).



has been sufficient to provide well contrasted x-ray images for analysis, even though the interconnec¬tion wire diam-eters used have continued to shrink (now to 25 μm, or less). However, as copper has less than half the density of gold, copper wires will appear far less distinct in the x-ray image (see images). This reduction in clarity will be even harder to see once the device has been assembled onto a board, as it will have to be seen against the background densities of the package and other parts of the board (Figure 2).

To combat this issue, and the continuing challenges of newer, smaller and less dense components, x-ray inspection systems are now available with alternative imaging detectors and improved x-ray tubes. With x-ray detectors, the image inten-sifier (II), so long the detector of choice in electronics, may now need to be replaced with a flat panel CMOS area array detector (FP detector). With x-ray tubes, there is the need for a tube with more x-ray power at lower accelerating voltages (kV), so as to retain a good rela-tive absorption through these thin, lower density materials but, vitally, the tube must also retain the resolution (or mini-mum feature recognition) at this higher power because of the need to image reduced fea-ture sizes. There are a variety of FP detectors and different x-ray tube types available in the market. However, the perfor-mance of each different tube/

detector combination in an inspection system is not nec-essarily equal for all electron-ics applications.

Generally, a good qual-ity FP detector will provide a bet-ter contrasted image than the best IIs and will do so more quickly, as the FP has less noise than an II. This may al-low faster inspection times. However, the gain within a FP detector is much lower than for an II. This means FP based x-ray systems have to operate, typically, at much higher tube powers. If the tube resolution then suffers, as is the case with some x-ray tube types, then observing the smallest features may not benefit from FP imag-ing. In addition, the FP detec-tor elements are usually much larger than in IIs. Therefore, there will be less image de-tail at the highest magnifica-tions using a FP. It should also be noted that the CMOS elements in the FP detec-tor will be damaged by the radia-tion used for inspection unless there is intrinsic shielding within the detector. The speci-fication of the individual FP will advise if this is the case. The best advice is always to try the system before any pur-chase, using good and ‘good bad’ examples of the types of components and assemblies that are to be inspected regu-larly. Overall, x-ray inspection of copper wire interconnec-tions within electronic com-ponents is possible but it may require a newer system speci-fication to accomplish it.

Figure 2. Flat panel detector x-ray images of QFN components as-sembled onto a PCB. Device with copper wire interconnections (left); device with gold wire interconnections (right).

From One Engineer To Another®

•answers

•blogs

•tech papers

•one-on-one

•support

•live chat

©2011IndiumCorporation

ASIA•CHINA•EUROPE•USA

scan code with mobile device

Dr.Ning-ChengLee,[email protected]

www.indium.com

“ Voiding under low stand-off components, such as QFNs, can be effectively minimized through material and process optimization. Let me show you how.”

Find out: indium.us/D154

Title


Interview—

Leading supplier of electronic compo-nents, element14, has launched “the knode”—said to be an amalgamation of design engineering and knowledge.

In an interview, Ravi Pagar, di-rector-India, element14, outlines the knode, a solution where design engineers can get trusted informa-tion on processes, and which acts as a knowledge source for them. Pradeep Chakraborty recently had the oppor-tunity to speak with Mr. Pagar about the knode.

What is the initiative all about? How will the knode bring together all software tools, development tools and kits, operat-ing systems?

The name for the solution was derived from the meaning of “node”—any point on a circuit where two or more circuit elements meet. The element14 knode brings knowledge (KNO) and design engineering (DE) together in one place.

The element14 knode was developed completely in-house by the element14 development team. It is the result of the need to create a solution that positively impacts the overall design flow and simultaneously streamlines working with component manufacturers and suppliers. It is one place where design engineers can quickly and easily get trusted information for their design process. It acts as a virtual workspace for the design engineering community.

Experts, design tools, IP, unique value-added design solutions, development platforms, components, services, hardware, software and up-to-date technical information have all been integrated into one convenient place: the element14 knode—knowledge for design engineers.

What else does the knode do to fulfill engineers’ needs for full suite of design tools and integrated solutions?

The element14 knode is poised to have a dramatic impact on the overall engineering design cycle and workflow, cutting design

times from weeks and months to days and hours in some cases. In fact, most engineers go to vendor websites to research products for a design, which is a very time-consuming process. By providing one place to get trusted information, element14 has eliminated the need for engineers to search hundreds of suppliers’ websites to receive the information they need to complete a design project.

The project-based area of the element14 knode is password protected, and access is limited to the user/design engineer.

How will these innovative solutions directly relate to different stages and functional requirements of the product design flow?

The element14 knode offers value-added engineering solutions through a complete design ecosystem combining an online delivery environment that helps today’s

Interview

ravi Pagar, element14


Title

engineers streamline the time-consuming and complex design ecosystem, from hardware assembly and manufacture to sourcing electronic components, to accessing technical information and libraries for hardware and software design tools.

It brings together development platforms, tools, software, IP cores, services, components and up-to-date technical information from domain experts and the global element14 engineering community, giving design engineers the ability to research, design, develop and manufacture in a single, intelligent environment. What are the benefits that this new initia-tive can bring to the customers?

Designers can use the element14 knode to quickly research, evaluate and purchase solutions, software and services. The availability of reliable technical information and solutions can save the designer hundreds of hours typically spent in searching and validating information.

Further, the innovative search engine returns all relevant and related contents for the search performed—in one click. The search bar allows users to enter keywords by supplier, architecture or part number, while search results can also be refined using categories and tag filters such as core architecture and silicon manufacturers.

Some of the features available include the expert Learning Centre, development platforms and kits, operating systems and stacks, development tools, CAD tools, PCB services and test equipment.

The element14 knode Learning Centre provides a library of rich content to help a user research various technologies, platforms and associated core components. This includes “How To” videos, application notes, technical documentation and much more. Forums host discussions in areas ranging from product/technology and applications to design recommendations, and technical support and designers can post questions to subject matter experts from element14’s enhanced technical support team and from suppliers across a number of industries and design environments.

Design engineers can access operating systems, RTOS, stacks and middleware for application software execution and interoperability.

The element14 knode also provides access to a comprehensive suite of software development tools for embedded systems: These include integrated development environments (IDEs), compilers and debuggers and the latest software design solutions from leading technology providers.

Through the knode, PCB design tools are available that enable designers to develop PCBs within a highly productive, scalable and easy-to-use environment.

The element14 knode delivers high quality, low risk solutions for PCB fabrication and assembly in partnership with industry leaders. element14’s partners not only provide rapid turn-around manufacturing using advanced technology and the application of state-of-the art design for manufacture (DFM) and test software, but they also deliver fabrication services at globally-competitive, affordable prices.

The element14 knode offers leading hardware analysis solutions for embedded systems or prototypes including oscilloscopes, multi-meters, signal generators and more.

Finally, a wide variety of development platforms and kits are available.

What are the full range of services and tools that element14 will provide through this global industry first initiative?

The element14 Knode is a portal for design solutions. It improves engineering design productivity by providing a single resource for all design needs. It streamlines the process of working with component suppliers. It raises the quality of the design by providing fast access to accurate technical information and community expertise. It enhances the overall design experience by unifying community, content and commerce in a sequence that is logical to engineers.

Providing resources that support the complete design flow from concept to

final production, the element14 knode is the industry’s first ecosystem helping design engineers accelerate design and development, bringing products to market faster than ever before. By not having to spend days and sometimes weeks researching technical resources and solutions, more engineering effort can be focused on application design and IP creation.

How has the experience and success with element14 platform contributed to the creation of the element14 knode?

The element14 knode is the culmination of feedback we have received from both our suppliers and design engineers to create a solution that positively impacts the overall design flow and simultaneously streamlines working with component manufacturers and suppliers. It is one place where they can quickly and easily get trusted information for their design process.

We are very excited about the launch and the impact it is poised to have on today’s workflows.

Why do you think this new service will be a landmark addition to element14’s exist-ing propositions?

The element14 knode is another industry first for us, one that further extends our leadership and offers an exciting new approach to design solutions in the web 2.0 era. Innovation is vital for design engineers as they constantly seek smaller, faster, better, cheaper, more environmentally friendly design solutions to meet the needs of their end customers. As their partner in innovation, we are re-defining the delivery of the solutions they need.

The services offered in the element14 knode v1.0 release include an engineering-focused alternative to standard search engines, an online ‘Learning Centre,’ design solutions that eliminate the need to visit multiple vendor-specific sites, development platforms and kits, operating systems and stacks, development and CAD tools, PCB services, and test solutions. The knode provides a full life-cycle electronics design flow solution from concept to production, easing design and accelerating reducing risk and time-to-market.

Thank you, Mr. Pagar.

Pradeep Chakraborty

Interview

By providing one place to get trusted information, element14 has eliminated the need for engineers to search hundreds of suppliers’ websites to receive the information they need to complete a design project.


Solder paste stencil manufacturing methods & their impact on precision and accuracy

Stencil manufacturingFor each method used to manufacture the stencil the important parameters are (1) the quality of the equip-ment used to manufac-ture the stencil, (2) the control over the process to fabricate the sten-cil, (3) the quality and behavior of the metal during the manufac-ture, (4) the tempera-ture differences during the various processes and (5) the varying ten-sion on the materials in the different process steps.

laser cuttingThe machine used to cut the stencils con-sists of two systems, the laser and the moving mechanism. It is very important that the laser has a small and very stable beam. The size of the beam determines whether very small details can be cut faith-fully. If the beam is not stable in size and the main power concentration moves around, the kerf will not be exactly where it is sup-posed to be, circular apertures are not round and straight wall apertures will have wavy sides (Figure 1).

Most of these lasers produce a stream of high power pulses to cut through the metal. Early lasers were pulsing at low fre-quencies, resulting in a scalloped cut when the metal was moved too quickly. Present day lasers employ a much higher pulse

frequency, allowing higher cutting speed without resulting in a scalloped cut line.

To verify that the laser beam is stable and produces constant power, close exami-nation of the aperture size, shape and wall is required.

With a 40 to 100x microscope it is easy to see whether the walls of an aperture are properly formed.

Various designs of the movement system exist. Early systems had a station-

Keywords: Solder Paste Stencils, Lasers, Laser-Cutting, Electro-Forming, Scanner

Stencil positional accuracy is a func-tion of the manufacturing process (Machines, Methods, Materials and Men). The various parameters that influence positional accuracy will be discussed. These include different lasers systems, various metals and processes, temperature variations and the effect of mounting a stencil in a frame. The total effect will be shown with measurement results from a number of stencils made according to today’s practices using available laser systems and processes. Additional problems may arise when stencils are not used correctly in the print-ing process. There are several param-eters that influence the matching (or mismatching) of the location of the solder paste bricks coming from the stencil with the location of the pads on the circuit board.

In the past, stencils were chemi-cally etched, and before that silk screening was used, but those pro-cesses have been replaced by more modern ones. These days most sten-cils are either laser-cut in stainless steel using an infra-red (IR) beam or electro-formed from nickel. Stencils have also been manufactured using polyimide, cut with an ultra-violet beam (UV) or IR1.

Ahne Oosterhof, Oosterhof Consulting, Hillsboro, OR, USA, and Stephan Schmidt, LPKF Laser & Electronics, Tualatin, OR, USA


Figure 1. Stencil cut using a laser cutting system with stable (top) and non-stable (bottom) positioning system.



ary laser beam while the table holding the metal sheet or stencil frame moved in both X and Y-axis. In some of the later systems the beam moves in X-axis and the table moves in the Y-axis. The next step is to hold the metal stationary and move only the beam in the X and Y axes.

The reduction in mass that has to be moved makes it easier to increase the cut-ting speed without sacrificing the ability to faithfully reproduce the detailed shapes of the stencil apertures.

In each design it is very important that X and Y-axes move perpendicularly to each other and that both move in a perfectly straight line (Figure 1). And of course the movement system has to be perfectly cali-brated to assure control over the amount of movement to within a few micrometers.

Most laser systems advertise location precision of 5-10 µm over a given distance.

MetalThe metal used for laser cutting has typi-cally been stainless steel, type 302 or type 304, produced in a rolling mill. The result-ing sheets are very uniform in thickness, but the specified thickness can typically vary by about 12 µm (0.5 mil). In order to improve paste release a number of post-processes have been tried, for example electro-polishing or chem-polishing, but not always resulting in improvements.

Other metals are now being introduced like nickel sheets and very fine grain stain-less steels. Especially these last ones have shown to bring significant improvements in the printing process. [Ref #2]

electro formingStencils made using the electro-forming (EF) process consist of pure nickel.

The EF process starts with a film that represents the aperture pattern to be man-

ufactured. Making the film introduces additional process steps with their inher-ent possibility of errors as the film mate-rial is both temperature and moisture sensitive. The film image is transferred in a photo process to a mandrel on which a metal layer is grown in an electro-chemical process. To get a uniform thickness stencil requires that the chemical actions in the bath are exactly the same over the full area of the stencil. This may at times be difficult, especially when the aperture density vari-ous greatly. Also the growth of the metal immediately around an aperture can be faster resulting in a small ridge or “dam” around the aperture. This dam has been used as a seal between the stencil and the pad on the board. However, if this dam is not exactly aligned with the pad or it gets damaged, paste can leak through, which may result in solder balls.

The process also has to be well con-trolled so that it can be stopped at the proper moment when the sheet has grown to the desired thickness. After that it has to be “peeled” of the mandrel without causing any damage to the sheet and then mounted in a frame.

TensionWhen sheets are laser cut, they are typically clamped and tensioned in one direction or they are cut already mounted in a frame.

If the sheets are mounted in the frame after cutting, the tension in both X and Y directions will often differ from the tension during cutting.

The same, but more so, is true for sten-cils made with the electro-forming process.

A stainless steel 125 µm (5 mil) sten-cil manufactured without any stress on the metal and then placed in a frame exerting a stress of 35 Newton/cm (common mesh tension) sees a strain (percentage change in

length) of 0.0131%. For a stencil image (or panel image) where apertures are 0.5 m (20 inch) apart, this can cause an error of up to 65 µm (2.5 mil)

TemperatureMost stencil manufacturers produce sten-cils in air-conditioned rooms where the temperature is about 20˚ C (68˚ F). In non-air-conditioned, small rooms the tempera-ture can easily vary by 5˚ C (9˚ F) or more. Similar variations can exist at the location where the stencils are being used.

The coefficient of thermal expansion of steel is approximately 17 and of nickel 13 ppm/˚ C. This number indicates the expan-sion or contraction of the metal in µm per meter for each degree C. If we have a stencil image (or panel image) where apertures are 0.5 m (20 inch) apart and the temperature difference between fabricating the stencil and using the stencil is 5˚ C, the change in dimension in a steel stencil can be 42 µm (1.7 mil). For a nickel stencil it would be about 32 µm (1.3 mil).

While laser cutting, the hot beam can cause a local temperature rise in the metal, which can lead to discoloration (innocent) or even deformation through local expan-sion of the metal (troublesome). Proper control of the beam and cooling of the metal (airflow or liquid cooling) can mini-mize this problem.

Impacts on useFor newer components, such as CSPs and very small passive components, the space between pads on the board can be less than 200 µm (8 mils)

To prevent significant errors as described above, it is imperative to employ the best stencil manufacturing equipment and practices possible. That also means working in a controlled environment, both

Figure 2. Stencil cut as loose sheet (left) vs. cut in frame (right).



at the stencil manufacturer and user loca-tions. To prevent errors due to possible tension differences it is desirable to cut the stencil while mounted in the frame. In short, as a stencil user it is becoming nec-essary to know what equipment and what process is being used and what checks are being made by the stencil manufacturer.

Figure 2 shows an example of one sten-cil cut from a sheet and then mounted and another cut in the frame on the same laser. A definite change in the error trend can be observed.

Stencil verificationThe simplest way to determine the preci-sion of a stencil is to scan it and determine the location and size of each aperture. Systems are available allowing such a test to be made with high accuracy (+/-5 µm or 0.2 mil) within a few minutes. A computer program can determine the centroid and size of each of the scanned apertures and

compare those to the original design. The resulting data can be used for an easy go-no-go determination or used to perform a statistical analysis.

The scanned, new, laser cut stencil may have remaining loose particulate in some of the apertures. This interferes with the centroid and area calculation but can easily be recognized and therefore excluded from the results (Figure 3).

A large stencil (about 460 mm x 300 mm/18” x 12”) with about 21,000 apertures has been selected for these tests.

These stencils have been mea-sured using a well-calibrated scanner (ScanCheck) with a resolution of 6,000 pixels per inch (12,000 with interpolation). The resulting numbers are then compared to the cutting data and errors beyond a given specification are presented. All data that has been collected can be exported for further analysis, as is done in this report. For this analysis only the location errors along the long stencil axis have been used.

Comparative measurementsA number of stencils have been produced using different methods, machines and processes. These stencils were produced using the commonly available laser cutting and electro forming production methods. Four different laser system brands for a total of seven different types of machines were selected. Of these stencils, five have been produced both as sheets and in a frame and two are only cut in a frame for a total of thirteen laser cut and one electro-formed stencils.

The stencils were produced in sev-eral different commercial facilities and

the environmental conditions were not recorded, therefore a temperature effect cannot be established separately from the machine accuracy and tension effects.

A specification of +/- 10 µm was used and for each stencil the extent and the dis-tribution of the location errors was calcu-lated. This above mentioned specification limit is a commonly used value for allow-able tolerance by many large EMS compa-nies. The value of Cp indicates how often this distribution of the data fits between the specification limits. Figure 4). For these very large and complicated stencils only one showed a Cp value greater than 1 (Figure 5). In the individual graphs the short green bars represent the three sigma limits. At those points the error rate is 2,750 ppm. Of course more desirable would be using Six Sigma where the error rate would be only 0.002 ppm.

As the stencil can be shifted and aligned to the board in the printer, the Cpk value, which uses the worst half of the distribu-tion and the deviation of the mean from the center of the specification, has not been determined.

resultsThe resulting Cp values for the whole group of stencils are shown in Figure 5.

The yellow bars show the Cp values for the stencils that were cut as loose sheets and the blue bars show the range for the stencils cut in the frame. For the measured apertures we see an error range (brown bars) varying from 35 to 185 µm (1.4 to 7.3 mil).

The data shows a noticeable grouping based on the chosen manufacturing tech-

Figure 3. Apertures with some debris.

Figure 4. Sample bell curve surrounding measured data. Red vertical bars show spec limits of +/- 10 µm.



niques. In general stencils that are cut in a mounted frame show significantly higher aperture positional accuracy then stencils that were cut as loose sheets and subse-quently mounted into a frame.

In Figure 6, the distribution of the data for a stencil cut in the frame and as a sheet using the same laser is shown. The notice-able change in the spread of the data shows the result of the change in tension between while cutting the stencil versus the ten-sion after the stencil has been mounted in a frame.

Another factor for the change in posi-tional accuracy is the choice of laser cutting system. In general we can observe that most newer generation systems (less than 3 years old) provide a higher positional accuracy compared to the older systems (3-15 years old). However even among new laser sys-tems we can observe significant difference in aperture positioning accuracy between different laser systems. These differences are probably related to system architecture

and calibration methods used. Figure 7 shows the change in the spread of the data for two stencils cut in the frame on two different laser systems.

ConclusionWhen printing on a board with components that have large pads and large spaces between pads, a significant align-ment error between the stencil apertures and board pads may not cause serious issues. It is like a form of overprint-ing and many solders, in

their molten state, will wick back onto the pad.

However, many of today’s boards have tiny parts with very small and closely spaced pads were such errors might cause bridging. On top of that, today’s lead-free solder does not spread as well as lead con-taining solders.

Therefore the size of the errors encoun-tered in several of these stencil samples will lead to production errors at an unaccept-able level.

For a stencil with optimum aperture positioning accuracy we can conclude that it is critical to choose the best manufactur-ing method based on three main factors: 1. Laser cutting shows better results than photo based processes, 2. Stencils cut in a frame show very little distortion and 3. Stencils cut on modern lasers showed sig-nificantly better positioning accuracy.

Note that additional printing errors can come from among others low mesh

tension, inadequate squeegee pressure or insufficient board support.

AcknowledgementsThe authors would like to thank Florian Roick of LPKF Laser & Electronics (www.lpkfusa.com), Mike Scimeca of FCT Assembly (www.fctassembly.com), and Frank Kurisu of Solder Mask, Inc. (www.soldermask.com) for their kind contribu-tions to this article.

references1. Ahne Oosterhof, et al, “Stencil cutting”,

Industrial Laser Solutions, June 2007.2. Robert F. Dervaes, Fine Line Stencil,

Inc.; Jeff Poulos, Alternative Solutions, Inc.; and Scott Williams, Ed Fagan, Inc., “Conquering SMT stencil printing challenges with today’s miniature com-ponents”, Global SMT & Packaging, April 2009.

Figure 5. Analysis of measurements.

Figure 6. Cutting in frame vs. cutting as loose sheet. Figure 7. Identical stencils cut on two different modern laser systems.

Bare board contamination

38 – Global SMT & Packaging South East Asia – May/June 2011 www.globalsmtseasia.com

Market trendsThere are a number of factors that mean bare board cleaning is becoming a neces-sity including:

• Sourcing from low cost economies where cleanliness standards are less stringent

• The adoption of laser marking, which is a considerable source of contamination

• On-going miniaturisation• Rising material, labor and fuel

costs squeezing margins, leading to demand for higher yield

• Adoption of 3D solder-paste inspection

Board contaminationIn general the contamination found on boards can be categorized as follows:

EnvironmentalDust, clothing and hair fiber

PeoplePeople are a considerable source of con-tamination including skin, hair and cloth-ing fiber

PackagingBoards are typically packaged in shrink-wrap and often with paper separating sheets, both of which can contribute loose contamination to the board surface.

Manufacturing processThe PCB manufacturing process typically does not conclude in a clean room and will involve much handling during AOI, repair, punching, routing and packaging. These steps will leave residual contamination.

Laser markingThis increasingly used process generates significant amounts of carbonated debris that constitutes a risk to your print process and end of line yields.

Finally, when packaged boards are sep-arated, they generate considerable amount of electro-static charge that will attract local contamination, be that from the envi-

ronment or people.

Solder paste printingIt is generally held that 75% of end of line faults are related to the solder-paste print process. It therefore stands to reason that a key component in solder-paste print be properly prepared for this crucial process step.

3d solder paste inspection (SPI) In an effort to improve quality and meet customer quality expectations there has been widespread adoption of 3D solder-paste inspection technology. 3D SPI can identify 100% of solder defects and will helpfully classify them as excessive, insuf-ficient, offset, misshaped or bridging. There are many factors within the printing pro-cess that can influence these fault codes, including stencil design, printer set-up and paste. What about contamination?

what influence does contamination have on yields at SPI?Established and sophisticated electron-ics assembly businesses manage the print process very closely and will typically have yields in the 91-94% range. The need to improve these yields is relentless as mar-gins are undermined by competition and increasing input costs. How then do we drive yields further in a well-run, con-trolled assembly environment?

A significant part of the answer in this scenario is to understand the influence of contamination and seek to address it through cleaning of the bare board.

Symptoms of contamination include:• Stencil apertures becoming

blocked and causing misprints• Craters after reflow caused by

contamination having volatized• Tomb-stoning (Manhattan

Effect)• Poorsolderwetting• Poorjointintegrity• Drysolderjoints• Excessorinsufficientsolder

Technology and processing develop-ments within SMT are moving bare board cleaning from a nice-to-have feature to an absolute necessity if yields are to be protected and/or improved.

This article examines the causes and impact of contamination and what solutions are available to tackle the problem.

Sheila Hamilton, Teknek, Renfrewshire, Scotland, UK




Process ImprovementWith over 10 years of experience in this application, Teknek can say with absolute confidence that removing contamina-tion as a variable will add 4-6% to yield at SPI. Time and time, again this assertion is proven in the field with leading automo-tive, OEMs and EMS companies around the world.

The effect of cleaning the boards is usu-ally measured using SPI equipment after the boards have been printed. Trials have shown a dramatic reduction in board fail-ures with contact cleaning before the solder paste print process. In one study using a Koh Young S.P.I., Teknek found that:

• Overall failure rates fell from an average of 9.5% to 5%—an improvement of 47%.

• Excessive solder dropped from an average of 3.4 to 1.3%—a 62% improvement.

• Bridging falls from an average of 1% to 0.7%—an improvement of 30%.

• Insufficient solder falls from an average of 2.1% to 0.8%—an improvement of 61%

Boards that fail SPIWhen a board fails SPI there are two pos-sible outcomes.

ScrapSome assemblers are required to scrap boards that fail SPI. This may be because of the safety critical nature of the end product or simply that they understand the full cost of rework. If the failure occurs during the first pass then the costs are easily assessed e.g. the board, paste, time, labor and dis-posal. On the second pass the same equa-tion applies but will include components and more line time so the numbers are considerably higher.

ReworkEach time a board is reworked, the quality and reliability is compromised. Misprinted boards can be washed down and wiped clean, but this will leave residual traces of solder paste, which are in turn a threat to inspection and end of line yields. (These will not necessarily be detected at the final electrical inspection at the end of the line.)

Contact cleaning technologyNow that we have established that PCBs arriving from the manufacturer are not pristine, what strategies can be put in place to reduce wastage, improve quality and increase yields?

Many manufacturers have tried brush, vacuum and blower systems to remove debris from the boards, but the most effec-tive way to achieve clean PCBs is to use spe-cialist contact cleaning equipment to clean the boards. This equipment uses a series of specially formulated elastomer rollers that make intimate contact with the bare board, gently removing dry unattached particles from the surface. (These rollers will remove contaminant particles down to one micron in size.) A reverse wound, pre-sheeted, adhesive roll then runs in contact with the elastomer rolls and transfers the particles to the adhesive. Once saturated a layer of the adhesive is removed, exposing fresh adhe-sive and the cycle starts again. Adhesive life is related to the contamination levels but our experience indicates a typical user will get between 400 and 600 panels for each sheet of adhesive. The adhesive sheets can also be used to investigate the root cause of the contamination—a useful diagnostic tool in its own right!

The cleaning device should also incor-porate anti-static systems to eliminate any charge as the boards move through the production line. Static monitoring sensors are available to measure static levels within your pre-determined criteria.

Contact cleaning equipment will pro-vide the highest levels of board cleanli-ness through the physical contact with the board. This contact overcomes the chal-lenge presented to non-contact system by the “boundary layer.”

return on investmentOur experience indicates that in most cases the return of investment can be measured in months rather than years. This will be a function of the preceding yields and those after installation of a bare board cleaner. Also included in the equation should be the cost of rework or scrap for SPI failure and the associate material, labour and overhead

losses. This of course can be broadly cal-culated during an initial consultation. The very best way to get certainty is to arrange a trial installation that will enable you to directly compare yields at SPI and end of line with the cleaner against the norm without.

AdoptionThe benefits of bare board cleaning are accepted by most automotive electronics businesses, yet the broader EMS market has been slower to adopt. Often they will cite a lack of customer demand for this feature. Bare board cleaning is a significant opportunity for the assembly businesses in respect to margins and is a clear indica-tion to prospective customers that you are in control of your process and capable of meeting their high quality requirements.

As the surface mount industry faces greater miniaturisation and demand for increased yield, contact cleaning will become increasingly accepted as a stan-dard process in a line. It is important that the concept is fully understood and used as a tool in the quest for continuing yield control of a modern SMT line.

Sheila Hamilton is Technical Director of Teknek and is responsible for keeping the company at the forefront through product performance, capability and applications. Sheila joined Teknek in 1987 as technical director after working as a product designer (yachting equipment) and power station engineer. She has also run her own consul-tancy in the field of electronics component packaging. Sheila has a BSc in Mechanical Engineering from Glasgow University and a MBA from Strathclyde University. In addi-tion, she is a recipient of two Smart Awards in the field of Electromagnetic Interference.

A bare board is fed into a contact cleaning system at the start of an automated pro-duction line to remove surface contamina-tion prior to the stencil printing process.

With contact cleaning, two tacky rubber rollers lift contaminants off both sides of the PCB at once. The rubber rolls then pass off the contaminants to the adhesive rolls.


Title

Title


Christopher Associates Inc. releases multiple camera retrofit kit for Marantz AOI systems

Christopher Associates Inc. announced the availability of a retrofit kit for the Marantz automatic optical inspection systems to add eight additional cameras. The com-bination of advanced lighting and angled cameras offers an unparalleled ability to capture defects while reducing false calls. The Marantz Multi-Cam option takes AOI to the next level, and includes real-time image processing and integrated software upgrades to offer even older systems the ability to inspect the most advanced tech-nology. www.christopherweb.com

Count On Tools now offers custom label nozzles for Amistar DataPlace systems

Count On Tools Inc. now offers custom label nozzles for Amistar DataPlace sys-tems. Count On Tools’ custom engineered nozzle service covers all types of labels. Each nozzle is built based on the customer’s specific labeling requirements. The Amistar

DataPlace system is based on proven pick-and-place technology for fast, reliable label-ing of circuit board assemblies for inline or standalone applications. www.cotinc.com

Solder paste deposition made easier by Ovation’s Magna-Print™ DeFlexIn another technology development break-through, Ovation Products announces that it has designed an enhancement to its proven Magna-Print™ universal blade hold-ing system that makes paste deposition even more robust. Current market-avail-able paste deflectors are rigid and can only be adjusted using a tool, which takes time and lacks precision. Magna-Print DeFlex and Magna-Print blades are simply and easily changed into and out of the universal blade holder with ease – no special tools are required. www.ovation-products.com

Tower-XL stores more 13” reelsTower is a fully automatic, modular and secure storage system for SMD components. The Tower-XL can store a maximum of 336 reels with 13” diam-eters, where the conventional Tower can store only 196. The footprint of the Tower-XL is 12 cm wider, which offers the additional room for larger reel storage cassettes. In total, the new Tower can store 406 trays and reels. When storing a reel, the Tower controls reel diameter and barcode. The access time for each reel is less than 12 seconds. www.essemtec.com

New, versatile epoxy resists high temperatures & harsh conditionsMaster Bond is pleased to announce the release of EP42HT-2FG, a new food grade

epoxy system. EP42HT-2FG has been inde-pendently tested and certified by a leading national laboratory to meet the stringent requirements of FDA CFR 175.300. It also was toxicologically evaluated to meet the NSF/ANSI 51.4.1 (2009) standard for food equipment materials. www.masterbond.com

Juki launches KE-1070/KE-1080 modular machines

Juki Corporation announces that its KE-1070/KE-1080 modular placement systems have greatly exceeded sales fore-casts since their release in late 2010. The KE-1070/KE-1080 machines are intended to meet the need for reliable, cost-efficient placement solutions for the mid-range market. KE-1070/KE-1080 machines are built using the same topology as the popu-lar KE-2070/KE-2080 machines that have had tremendous success in the market-place. www.jukiamericas.com

Speedprint Launches the SP700avi screen printer in North AmericaSpeedprint, part of the Blakell E u r o p l a c e r Group, introduces the Speedprint SP700avi Screen Printer to the North American market. The SP700avi features unique Look Down/Look Down vision that facilitates optimized board-to-stencil alignment, yielding 20 µm performance at 2 Cpk. Additionally, the system is fully equipped, as standard, with many features such as auto stencil load/unload, fully program-mable vacuum assisted stencil cleaner and optical inspection of the paste bead. 2-D post-print bridge detection and vari-ous tooling solutions also are available as options. www.speedprint-tech.com

New productsnew products


new products

Ellsworth Adhesives offers Devcon flame retardant productsEllsworth Adhesives now offers Devcon flame retardant products including Devcon 5 Minute I-FR and Devcon 10 Minute Epoxy I-FR. Devcon flame retardant prod-ucts are ideal for applications requiring a self-extinguishing structural system. The products are recommended for potting inserts and edge sealing for the aerospace industry. www.ellsworth.com

SEHO debuts rapid cooling function for MaxiReflow

SEHO Systems GmbH has equipped the successful MaxiReflow soldering system with a new feature that will lead to remark-able time savings, thus contributing to reduced manufacturing costs. The new rapid chamber cooling feature can help reduce changeover time to a minimum. Upon changing the soldering program, the system automatically compares the nomi-nal temperatures of the new program with the actual temperatures. If the new pro-gram’s temperatures are lower than a previ-ously defined temperature delta, the system automatically starts the rapid chamber cooling function. www.seho.de Condensation soldering is now more flexible than ever with CondensoX!

Rehm Thermal Systems is presenting its CondensoX Soldering System, which allows users to adapt vapour phase solder-ing to manufacturing conditions. Able to be implemented into existing SMD lines, the CondensoX-series is an advanced condensation reflow soldering system designed to meet the challenges of lead-free solder alloys for small and high ther-mal mass PCBs. The CondensoX combines vacuum profiling and thermal profiling in one machine. www.rehm-group.com

JST Corporation’s new LED lighting subminiature connector

The new LEA Series wire-to-board, crimp style disconnectable connectors recently introduced by JST Corporation provide optimum performance and subminiature size for low current, low voltage conditions required by LED light bars. These highly reliable, side entry SMT connectors are polarized and incorporate positive locking features including an audible click to pre-vent accidental disconnection. Molded in a RoHS compliant 94V0 9T Nylon (SMT headers) resin, the LEA Series is available in a 2 circuit size. www.jst.com

SIPLACE Precedence Finder—reliability from the first board on

As part of the current SIPLACE setup con-cepts campaign, ASM Assembly Systems is introducing the SIPLACE Precedence Finder. Some analysis and optimization processes can from time to time have a neg-ative effect on the throughput of SMT lines. The SIPLACE Precedence Finder helps to considerably improve this. From the very first board on, the software analyses the size and position of all components, PCBs, and nozzles and optimizes the placement sequence to prevent components from col-liding with each other. www.siplace.com

Microscan introduces three new machine vision innovationsMicroscan introduces the AutoVISION™ family of machine vision products. The product line includes the Vision HAWK and Vision MINI smart cameras, as well as the new AutoVISION™ machine vision software interface. The Vision HAWK and Vision MINI smart cameras fea-ture compact, fully-integrated hous-

ings, and can be operated by either the intuitive AutoVISION interface or the more advanced Visionscape® platform. This format allows users to set up jobs in AutoVISION™ and later migrate to Visionscape® if the application grows in complexity. www.microscan.com

LPKF introduces new laser for depanelingLPKF has added another laser system to its repertoire. The MicroLine 1000 S pres-ents a compact, cost-effective method for U V - l a s e r depaneling of thin-rigid and rigid-flex assembled PCBs. With UV-laser cutting of assembled and/or unassembled PCBs, those with limited space will benefit. The UV laser beam can cut along delicate components or circuit paths without inter-ference. The tool-less method makes any contour possible. Changes to the cutting paths are made by programming the soft-ware included with the machine or directly in the CAD software. www.lpkfusa.com New spray fluxer for wave soldering systemsEtek Europe’s branded spray fluxer for wave soldering systems, the EcoSpray, is a quality and cost effective system with full back up from the Etek Service and Support Division. Ecospray is avail-able in three sizes; 12", 17" and 20". www.etek-europe.com

Title

International diaryelectronica India + productronica IndiaSeptember 13-16Bangalore, Indiaelectronica-india.com

NEPCON VietnamOctober 6-8Hanoi, Vietnamnepconvietnam.com

electronicAsiaOctober 13-16Hong Kongelectronicasia.com

EPTC Electronics Packaging TechnologyDecember 7-9Singaporewww.eptc-ieee.net

India TelecomNew Delhi, IndiaDecember 7-9www.indiatelecom.org

Hong Kong Electronics April 13-16, 2012Hong Konghkelectronicsfairse.com

Volume 10 Number 2, February 2010ISSN 1474 - 0893

The Global Assembly Journal for SMT and Advanced Packaging Professionals

Fred HumeInterview Inside

NEW PRODUCTS | INDUSTRY NEWS | INTERNATIONAL DIARY

BGA ASSEMBLY RELIABILITY...PWB QUALITY IS THE KEY

METALLIZATION OPTIONS FOR OPTIMUM CHIP-ON-BOARD ASSEMBLY

Global Technology Awards

10.2-EU-temp.indd 1 2/9/10 10:14 AM

Bryan GassInterview Inside

Volume 10 Number 1, January 2010ISSN 1474 - 0893

2010 Company MilestonesAlso Global SMT & Packaging’s 10th anniversaryTRACK, TRACE & CONTROL: HIGH PRODUCTION OUTPUT AT LOW COSTS

FAILURE DEFINITION—NOT AS EASY AS IT SOUNDS

SYNERGY, THE OCCAM PROCESS AND TWISTED WIRE INTERCONNECT


10.1-EU-temp.indd 1 1/6/10 5:46 PM

www.globalsmt.net

SMTSMT

Industry NewsNew Products

International Diary

Volume 10 Number 2, February 2010ISSN 1474 - 0893


Fred HumeInterview Inside

NEW PRODUCTS | INDUSTRY NEWS | INTERNATIONAL DIARY

BGA ASSEMBLY RELIABILITY...PWB QUALITY IS THE KEY

METALLIZATION OPTIONS FOR OPTIMUM CHIP-ON-BOARD ASSEMBLY

Global Technology Awards

10.2-EU-temp.indd 1 2/9/10 10:14 AM

Bryan GassInterview Inside

Volume 10 Number 1, January 2010ISSN 1474 - 0893

2010 Company MilestonesAlso Global SMT & Packaging’s 10th anniversaryTRACK, TRACE & CONTROL: HIGH PRODUCTION OUTPUT AT LOW COSTS

FAILURE DEFINITION—NOT AS EASY AS IT SOUNDS

SYNERGY, THE OCCAM PROCESS AND TWISTED WIRE INTERCONNECT


10.1-EU-temp.indd 1 1/6/10 5:46 PM

www.globalsmt.net

SMTSMT

Industry NewsNew Products

International Diary

www.globalsmt.net


Volume 1 Number 1 Spring 2010

NEW PRODUCTS

INDUSTRY NEWS

INTERNATIONAL DIARY

Global SMT & Packaging Southeast Asia Vol. 1 No. 1

Spring 2010

Southeast AsiaSoutheast Asia

Guidelines for establishinG a lead-free wave solderinG process for hiGh-reliability

investiGation and development of tin-lead and lead-free solder pastes to reduce head-in-pillow defects

process challenGes and solutions for embeddinG chip-on-board into mainstream smt assembly

Mike Konrad Interview Inside

www.globalsmt.net


Volume 1 Number 1 Spring 2010

NEW PRODUCTS

INDUSTRY NEWS

INTERNATIONAL DIARY

Global SMT &

Packaging Southeast Asia Vol. 1 No. 1Spring 2010

Southeast AsiaSoutheast Asia

Guidelines for establishinG a lead-free wave solderinG process for hiGh-reliability

investiGation and development of tin-lead and lead-free solder pastes to reduce head-in-pillow defects

process challenGes and solutions for embeddinG chip-on-board into mainstream smt assembly

Mike Konrad Interview Inside

Six ContinentS

one Magazinewww.globalsmt.net


Title

A GLOBAL Technology Award sends the message that your product or service is an innovation and quality leader in an industry crowded with competing products.

Sponsored by Global SMT & Packaging magazine, the GLOBAL Technology Awards are now in their sixth year of recognizing and celebrating innovation in the electronics manufacturing industry. Entries are being accepted now through July 31st.

Entries are invited from equipment, materials and EMS companies of all sizes. In addition to the award statue, winners receive publicity in a special awards issue

of each of Global SMT & Packaging magazine’s five editions (US, Europe, China, Korea & South East Asia) as well as on the Global SMT & Packaging and GLOBAL Technology Award websites and in the Global SMT & Packaging email newsletters. Winners also receive a small poster for use at trade shows and an image and logo for use in advertising, websites and other promotional materials.

Enter now: http://awards.globalsmt.net

The ONLY global awards program in the industry

The 2010 GLOBAL Technology Awards

Presenting...

New for 2010: “Contract Services”

has been broken into two categories:

> $100 million and < $100 million.

full_page.indd 1 4/22/10 12:06 PM

HalogenFreeAd_603_605_044_Eng_EM_China.pdf 1 4/21/11 1:48 PM

Documents

Global SMT & Packaging South East Asia - July/August 2011 (#2.4)