Dr. Lasky's Blog

Folks,

In a recent posting we discussed that the higher melting temperatures of lead-free solder require reflow soldering temperatures to be higher, thus more electricity is used in lead-free assembly. However, as we calculated, this increased use of electricity is very small compared to all electricity used in the world.

An additional concern that some have voiced is the claim that RoHS, with its lead-free requirement, actually makes the environment worse because more tin and silver is used in lead-free solders.   They argue that the increased use of these metals, creates mining pollution and has driven the price of these metals sky high. Let’s examine these claims.

Prismark has estimated that approximately 90,000 tons of solder are used in electronics, with about 80,000 used in wave soldering and 10,000 tons for SMT soldering. It is important to remember that electronics solder is a subset of all solder. All solder (alloys for brazing pipes etc) uses about 190,000 tons of tin. Solder is the single largest user of tin. See Figure 1. 

Figure 1. Solder is the largest end use of tin. Tin is the base material for almost all solders. 

If tin-lead solder were still used predominantly, approximately 57,000 tons of tin (90,000 x 63% tin) would be used annually. With lead-free solder, about 88,000 tons (90,000 x 98% tin) of tin are used per year. This is an apparent increase of about 30,000 MT of tin used each year. However, an interesting thing to consider is that lead-free solder is about 14% lighter than tin-lead solder. Knowing that, and knowing that solder used in wave soldering (remember wave soldering accounts for almost 90% of all solder used in electronics assembly) is consumed by volume not weight (i.e. assuming approximately the same fillet size), about half of this increase is canceled out. 

This is all a bit confusing however, so it may be best to just to look at tin use. According to the United States Geological Survey (USGS), about 300,000 tons of tin are mined each year. Figure 2 is a graph of world tin production at mines per year (this graph does not show recycled tin.)  The amount of refined tin used each year in the US is depicted in Figure 3. Figure 3 includes about 15,000 tons a year of recycled tin. Recycling solder is very cost effective. Scott Mazur just pointed out (Printed Circuit Design and Fab and Circuits Assembly, p 36, August 2011), that recycling solder dross is 10 times as cost effective as recycling aluminum cans.

Looking at these graphs, it is hard to say that the amount of tin used has gone up since RoHS. It would appear that tin use is likely more affected by the economy and that it is really difficult to see an effect from RoHS’s July 2006 enactment.

Figure 2. World Tin Production at Mines. 

Most wave soldering solders have low or no silver. So, about 3% of the 10,000 tons of SMT solder, or 300 MTs of silver, are used in electronics. This is about 1.5% of the 22,000 MTs of silver produced each year. Silver use in electronics does not make anyone’s list of top silver usage.

Figure 3. US consumption of tin has decreased since RoHS was enacted.

So electronics solder use since RoHS has not caused tin use to increase, nor is it a significant factor in silver use. Therefore it is highly unlikely that electronics' use of tin or silver has been a prime driver in their stunning price increases in 2011.

 Folks,

In gathering information on the status of lead-free soldering, some helpful friends pointed out two great sources of information: NASA and The Navy. NASA sponsored an impressive lead-free reliability investigation: "Lead-Free Solder Testing for High Reliability Project 1." This project is finished and the reports are online. There is a follow-on project: NASA DOD Lead-Free Electronics Project 2 which is currently underway. The Navy sponsored a project with ACI and the summary is here. I am currently studying these documents to help develop the consensus.  Some preliminary info follows:

Regarding -20^°C to +80^°C thermal cycling, NASA concluded:

“Under the conditions of this test, Sn3.9Ag0.6Cu (SAC) and Sn3.4Ag1.0Cu3.3Bi (SACB) were always more reliable than eutectic SnPb regardless of component type (CLCC, TSOP, BGA or TQFP).

It has been shown that conditions that highly stress the solder joints by maximizing the CTE difference between the PWB and the component will favor SnPb over SAC6. Conversely, conditions that minimize the stress put on the solder joints (e.g., compliant components such as BGA’s and/or a thermal cycle with a small delta T) will favor SAC over SnPb. The current test falls into the latter category and we can say with some confidence that the lead-free alloys tested will outperform eutectic SnPb under field conditions that are even less stressful than the -20 to +80^°C thermal cycle test conditions.”

For -55^°C to +125^°C thermal cycling, the conclusions were more cautious, likely because the data were mixed:

“The feasibility of using Pbfree solder alloys in place of SnPb solder alloys for new product designs was demonstrated under thermal cycle test conditions. Additional investigation and characterization of Pbfree solder alloys will be required as a segment of a Pbfree solder alloy implementation plan. The application/introduction of Pb-free soldering processes for legacy product designs is not recommended without extensive materials characterization and product design review.”

These results seem to be consistent with what others report, lead-free assembly produces good thermal cycle results for commercial-type thermal cycling, but the results are mixed for harsh environment thermal cycling.

Folks,

You may remember that more than a year ago there was much speculation that tin whiskers may be behind the Toyota unintended acceleration problem. At the time I spoke out because there was no data to support the speculation. Now there is data, Mike Pecht and his CALCE Team at the University of Maryland, found numerous tin whiskers in the Toyota brake assemblies of concern.

Although the tin whiskers were not implicated in any failure, their presence is cause for alarm, and action should be taken to address this issue. Tin whiskers should not be found in mission critical devices. Pecht’s team has an algorithm that calculates the risk from tin whiskers that are discovered. The risk is 140 per 1 million not high, but with a million or so Toyotas on the road, clearly this is cause for alarm.

As you may know, I live in Woodstock, Vermont.  Many friends have asked how we are doing after hurricane Irene. Personally, my wife and I escaped with no damage to our house and only a bit of inconvenience (no water for 5 days). The town of Woodstock suffered considerable damage, but was, on the whole, fortunate. Some of the neighboring towns had all roads in and out washed away. Route 4 between Woodstock and Rutland has numerous sections destroyed. The flooding was declared by the governor to be the worst disaster in Vermont history. The photo is from the Valley News. It shows a wooden pedestrian bridge built to carry supplies into Bethel, VT by foot. There is no passable road, even for ATVs.

Best,

Dr. Ron

Folks,

Recently I posted a note about a flurry of Technet posts in which I was misquoted regarding the status of lead-free electronics assembly.  Harvey Miller then weighed in.  I responded. And this in turn raised more comments.

All of this caused me to wonder, is it possible to achieve a consensus on the state of Pb-free assembly?  I think it might be and am going to try.  The main thing that I think is important in this quest is that any points for the consensus, or lack thereof, be supported by data and analysis, not emotion.

If you have a point to add, that is backed by data and analysis, please share it with me.  One of the things I hope to accomplish is to develop a list of references, that can be referred to to support the consensus.

Stay tuned for more info on this effort.

Cheers,

Dr. Ron

Folks,

An obvious disadvantage of lead-free electronics soldering assembly is that the oven must be hotter and therefore will use more electricity (versus 63Sn37Pb soldering). But is the extra amount of electricity significant? Bill O’’Leary claims that a typical SMT oven uses $7K of electricity a year at $0.072/Kilowatt hour (Kwh) or about 100,000 Kwh. That number strikes me as about right, as a household uses about 5-20,000 Kwh per year.

In the late 1990s there were 35,000 SMT lines in the world, at a 3% growth rate that would be about 50,000 lines now. So worldwide SMT reflow oven use would be about 5E9 KWhr (50,000 ovens x 100,000 Kwh/per year) world wide.  

With most heat loss be due to convection, the increase in energy use will be approximately proportional to the difference between the oven temperature and the room temperature (25C). An oven processing tin-lead solder would run at about 210C versus lead-free’s 250C. So the added energy for a lead-free oven would be about (250-25)/(210-25) or about 22% more. So if all assembly lines in the world are SMT the added energy use would be about 0.22x 5E9 Kwh = 1E9 Kwh. The cost of this extra electricity would be about $100 million (US) at $0.10/ Kwh. The electronics industry generates about $1.5 trillion in sales. So this added cost would be about 0.0067% of sales. Since world electrical use is about 150,000 E9 Kwhr per year, this increase is about 1/150,000 of all of the electrical use or 0.00067%.

So although more electricity is used, the increase is not significant to the value of the electronics sold or the total world use of electricity.

Thinking about higher temperatures reminds me that my Indium Corporation colleague Dr. Ning-Cheng Lee is presenting a paper this week on a high melting temperature lead-free solder based on a BiAgX alloy system. Higher melting temperature solders are often needed in what is referred to as a solder hierarchy. Solder hierarchies have solders that melt at decreasing temperatures in multiple soldering steps, starting with the highest melting solder.

Cheers,

Dr. Ron

Folks,

It was five years ago today that RoHS was launched, amid concerns that the world of electronics would collapse due to the many challenges of lead-free (Pb-free) soldering. Well, we have five years of field data with no “the sky is falling” lead-free reliability events. But, has it been just five years?

No. As I mentioned in a recent post, Motorola implemented lead-free soldering around 2001 to take advantage of lead-free solder’s poorer spreading.  Hmmmm,  so it has been ten years! Not too bad!

Well it is actually better than that. SnAg3.5 solder has been used for decades in both:

1.     Step soldering:  with a eutectic temperature of 221C, SnAg3.5 can be used as the step previous to soldering with Sn63 or similar Pb-Free solder. The principle is to solder first with the SnAg3.5 and then with a lower melting temperature solder. The second soldering step is performed at a lower temperature, therefore not disturbing the SnAg3.5 solder joint or bond. 

2.      Mid-Temp Pb-Free alloy:  when a solder that melts somewhat above the melting point of a “standard” solder alloy is needed, and it must be Pb-free, SnAg3.5 is often the choice.  The automotive industry has used SnAg3.5 in these applications for decades.

While I still agree that lead-free solders need some time and experience, especially in harsh environments, to establish acceptable reliability for mission critical applications, the experience with SnAg3.5 is adding to lead-free solder’s reliability portfolio.

This information came to light with the recent announcement by a major solder materials supplier that they would no longer supply SnAg3.5. But take heart, Indium Corporation still supplies SnAg3.5.  

Folks,

Here is an interesting turn of events related to the reliability of lead-free (Pb-free) soldering reliability. 

I was reminded recently by something Carl Sagan said, or, actually, did not say: Billions and Billions Although this term is strongly associated with him, he never said it. Sagan believed that this term was connected to him because Johnny Carson mimicked him and used the term.

 
Although not even close to being in Sagan's league, I find that I am now equally unfairly associated with the term,  "lead-free solder is a grand success." This came about in an interview by Rob Speigel, which he summarized in a blog post.

In reading Speigel's post, you will see that,  "lead-free solder is a grand success," is Rob’s term, not mine. Well, Rob's post resulted in a string of postings on IPC’s Technet .

One person opined:

Irresponsible statements like "lead-free solder is a grand success" should NOT be ignored. Those who make such statements in the face of all of the contrary evidence should be noted, and treated as motivated only by greed. Lead-free soldering certainly has been known for many "thousand$" of successes.

I have learned that it is not even worth the bother to refute such statements with those who make them. It may be a "grand success" for PhDs who contract to solder paste companies, but it certainly has not been a "grand success" to literally thousands of companies dealing with the reliability elephant sitting in the room getting larger by the day, and the associated fallout as a result.

Ouch!

Another shared:

I disagree with the stated and implied affect of RoHS, on PWBs expressed in this article. Lead free assembly reduces reliability by 50%. There can be no doubt about that. There are too many studies that confirm lead free assembly significantly degrades reliability. There are so many studies that demonstrate a reduction in reliability that Rod's contention is almost laughable. We are now faced with increased failures of copper interconnections and dielectric material due to high assembly temperatures. There is an increase in crazing that can support CAF, significant copper dissolution, and cratering in assembly, Switching to lead free in most HDI applications is a significant challenge. Lead free assembly has a profound affect by degrading PWB's organic component (epoxy) due the temperature required and copper interconnection and also the exaggeration of the z-axis expansion of the dielectric.

I have asked for copies of the many reliability studies referred to. No response yet.

Finally someone hit the heart of the matter:

I'm curious if "grand success" were Dr.Lasky's words or Rob Spiegel's editorializing. Lasky does mention the lack of long term results, and Speigel, in the comments,  enumerates a number of reliability problems. ISTM that neither truly believes  those words.

Correct!, Thanks. 

Here was my response that I posted on Technet:

Folks,

Pete is correct. I never said lead-free implementation was a grand success. These were Rob's words in his blog post. 

I have said repeatedly that adequate lead-free reliability has been demonstrated for consumer products like mobile phones, PCs, portable electronics with service lives less than 5 years. This level of reliability has been demonstrated in numerous studies and more importantly with field data. Vahid Goudarzi, of Motorola, stated that field reliability of lead-free assembled mobile phones has been equal or better than leaded assembly units. His data go back to 2001 (not 2006. Motorola started early for reasons discussed below).

 The reason Motorola shipped early with lead-free products is due to the fact that lead-free solder does not spread as well. Because of this poorer spreading, Motorola was able to decrease lead spacings without getting shorts, thus increasing the amount of electrical function in a smaller space. Since increased function in a smaller space is the defining attribute of portable electronics, the importance of this lead-free advantage cannot be overstated. Admittedly, lead-free's poorer wetting is a challenge in other regards, especially hole fill in wave soldering, but the Motorola Droid X2 could not be assembled with leaded solder, there would be too many shorts. Since the packaging density of the iPhone and similar devices is on a par with the Droid X2, I suspect this statement is true for most mobile products.

I have also repeatedly stated that lead-free reliability for long term service, mission critical devices has not been demonstrated. As a result, these types of devices should not consider lead-free solder at this date.

I regularly discuss these topics in my blog (http://blogs.indium.com/blog/an-interview-with-the-professor). The most recent post shows a striking photo of leaded solders spreading -which is too "good" for portable electronics.

Cheers,

Dr, Ron

Folks,

I have often pointed out that SAC solder's poor wetting is both a curse and Godsend.  It is a curse when trying to fill a through-hole in wave soldering, and a Godsend when assembling close lead spacings as shown in the image (below). Indium Corporation colleague and friend, Mike Fenner (image below), pointed out that, when I say that, "SAC solder doesn't wet well", I should be saying, "it doesn't spread well". His explanation follows:

SAC is different from SN63, and I think it is helpful to explain the difference by making a subtle differentiation between wetting and spreading.

The way that solders spread and wet to a surface is a balance of competing forces. We have surface tension acting to make the molten solder shrink into a ball, and wetting forces trying to make it spread across the surface. Wetting is also the action of the solder dissolving into the surface to form an intermetallic. This intermetallic is essence of the solder joint. The balance changes with different alloys, surfaces, and processes.

Most people are very familiar with the way that tin lead solders behave - and that governs their expectations. The different balance in SAC means the solder tends to spread less for the same wetting and, therefore, can give the impression of a lower quality joint. This lack of spread is usually expressed as 'poor wetting'.

I would explain this by saying the “active ingredient” in both solder families is tin. SAC alloys have a ~50% higher concentration of tin than the Sn63 solder alloy. This gives them a higher surface tension which increases the balling (coalescing) force. At the same time, the less dilute tin, in SAC solders, dissolves into a surface faster. So the final SAC joint can have a well formed intermetallic, but not high spread. These relationships will vary with surface finish and, of course, flux chemistry and process conditions come into play, but that’s for another day. Meanwhile I hope this simplified explanation helps.

Thanks Mike!

Cheers,

Dr Ron

The solder image is courtesy of Vahid Goudarzi of Motorola.

Folks,

I met Peter Borgesen back in the mid 1980s when he was a research scientist at Cornell working with Professor Che-Yu Li. Later we worked together at Universal Instruments. Currently Peter is a Professor a Binghamton University.  All during this time, Peter has been working on materials science-related topics in electronics packaging and assembly, most notably reliability. In addition to his many technical skills, he is a gifted linguist, speaking multiple European languages. Etched in my mind is Peter talking to several European graduate students in their native European languages in the space of 5 minutes, switching from one to the other effortlessly.

Few people know more about lead-free solder reliability than Peter. So I thought I would get his perspective on my recent post on lead-free field reliability data. His comments follow.

Hi Ron,

I agree that the sky is not falling. Also, we should be talking much more (only?) about life in service. I realize that we don't know enough about this (and our predictions based on test results are much more off than people want to recognize). The vast majority of practitioners focusing on 'engineering tests' are doing worse than wasting time and effort if comparisons of test results do not translate to relative performances in service. There is a lot of ‘sticking heads in the sand’ here.

I am not concerned about the long term life of cell phones, and not very worried about in which respect they do better or worse in service than with SnPb. Intermetallic bonds have generally gotten weaker and more prone to sporadic defects, and cratering is greatly enhanced for the devices Vahid Goudarzi mentions when discussing Motorola field data. I agree those are limited concerns, no sky falling indeed.

What still scares me (in the case of critical applications) or concerns me (in the case of expensive applications) is the naivete with which many seem to think we can learn much about sporadic disasters or long-term reliability of those from consumer electronics experiences.

I am not often interested in comparisons to actual life of SnPb either (any more). We face ever more applications (designs and service conditions) for which we don't have sufficiently accurate critical experience with SnPb either. The first challenge becomes not to be surprised by effects of long-term aging, combinations of loading, minor differences in pad finish, joint configuration, latent damage, process, .... and their interactions for the specific solder alloy used (!).

While I can't yet extrapolate test results to life in long-term service (I think we are close, but I really need an extra $1M to prove my hypothesis and turn it into a quantitative model) I can show how current models may easily be off by 2-3 orders of magnitude or more (worse, how they may screw up comparisons of alternatives). It obviously depends on the application whether this really matters (I side with companies who have cut drastically back on testing for many applications).

Keep up the good work.

Peter

I will keep in touch with Peter in the future for updates on his perspective.

Cheers,

Dr. Ron

Folks,

I have occasionally written on calculating solder alloy density, as there is surprisingly more interest than I thought there would be in this topic. Recently, it occurred to me that it might be beneficial to compare the calculated densities to actual densities of a few alloys to see how accurate the correct formula is (for the derivation of the correct formula see below). The formula assumes “perfect mixing” (i.e. no interactions between the alloy elements). The alloys we investigated were tin-bismuth-silver, tin-silver, tin and tin-bismuth.

To measure the density, I obtained a few alloys from Indium Corporation. My student, Evan Zeitchik, determined that a good technique to measure density is to machine the alloy into a rectangular parallelepiped (see photo), weigh it, and calculate its volume from its dimensions.  The results agree with the correct formula to about 1 to 2 %. Some people would ask why there is any difference. The reason is that all alloys form different phases, and some form intermetallics. These phases and intermetallics would typically have different densities than that calculated for the alloy. I will have more detail on this work in a future post. 

Here is a derivation of the correct density formula:

Many people incorrectly assume that if you have an alloy of x % tin and y % silver, that the density of this alloy would be 0.x*Density tin +0.y*Density silver. This intuitive linear formula is incorrect however, as density has two units (mass and volume).  An easy way to understand the derivation of the correct formula (proposed by Indium Corporation engineer Bob Jarrett) is to consider a 96% tin, 4 % silver example.

Lets assume I have 1 g of this alloy, 0.96 g is tin and 0.04 g is silver.

The volume of the tin is 0.96 g/7.31g/cc = 0.131327cc

The volume of the silver is 0.04g/10.5g/cc = 0.00381cc

So 1 g of the alloy has a volume of 0.131327 + 0.00381 cc = 0.135137 cc

Hence it's density is 1g/0.135137cc = 7.39989g/cc

Hence, the general formula is:

1/Da = x/D1 + y/D2 + z/D3

Da = density of final alloy

D1 = density of metal 1, x = mass fraction of metal 1

same for metals 2, 3

The formula continues for more than 3 metals.

I have developed an Excel spreadsheet that calculates density automatically. If anyone wants a copy, send me an email at rlasky@indium.com

Based on a recent post I published regarding the use of bismuth (Bi) in solder alloys, John writes:

"If Bismuth comes from the production of Pb, and if the use of Pb is being reduced, won’t the availability of Bi be reduced…and the price would increase?"

"Just thinking…"

Dr. Ron responds:

Lead has been banned from many of its original uses, paints, solders, water pipes, gasoline, etc. However, its increased use in batteries has actually caused lead consumption to rise. The USGS estimates that 88% of lead produced is used for lead-acid batteries.

Many of us, in electronics assembly, have been focused on the 2006 RoHS lead ban. This may have caused us to believe that lead use in electronics was significant. About 9 million metric tons (MT) of lead are consumed each year, only about 20K metric tons were used for solders prior to July 2006, this amount is only about 0.22% of the total. Electronic lead use being so small, is likely why the lead industry had little visibility in fighting RoHS. Their important customers were making batteries.

Lead is quite effectively re-cycled, as about 60% of the 9 million MTs/yr are from recycling and 40% from mining.

Over 100 million lead-acid auto batteries are sold each year in the US alone. In addition, the use of lead-acid batteries in fork-lifts, electronic vehicles, and golf carts has increased demand for lead. So, the bottom line is that lead use is expected to grow at about 2% per year.

Considering that we calculated that bismuth use in solders would be at most 5% of total bismuth production, it is unlikely that this use, or lead production reduction, would affect bismuth supplies.

Folks,

 A few people asked some questions after my last post on bismuth solders. Here they are:

1.      The low melting point of these solders is encouraging. What are realistic field use conditions?

Bismuth solders tend to be brittle, so drop shock environments such as mobile phones would not be recommended. However, thermal cycle performance from 0 to 100C is good, so stationary office equipment, televisions, desktop computers, etc may be good candidates.

2.     I am working with your colleagues on an automotive application and I am curious whether you have any idea how this alloy will perform between -40 and 0°C? We have not been reviewing bismuth-containing alloys due to their lower sheer strength, but may need to look at them in the future.

We can find no information on thermal cycle performance at these low temperatures.

3.     I hear that bismuth is rarer than silver, if we start using bismuth in solders couldn’t that make it very expensive.

An old number from Prismark puts the world solder use at about 50,000 metric tons (MT) per year.  Assume bismuth solders took a 5% market share (I think this would be the highest) that is 2,500 MT of bismuth solder (Bi57Sn42Ag1) or 1,425 MT of bismuth.

Although bismuth's occurrence in the earth's crust is 0.009 ppm (silver is 0.075 and gold 0.004 ppm), about 22,000 MT are produced each year.  In comparison, about 2,000 MT of gold, 20,000 MT of silver, 400 MT of indium and 5 MT of rhodium are produced each year.  In comparison to more common metals, total lead production is 8,000,000 MT/year and tin a little less than 700,000 MT.

 Realistically, it would seem to me to be unlikely that use of bismuth in solder, at 1,425MT/year out of 22,000 MTs,  would affect the price much, especially if the adaptation rate is more like 1-3%, instead of 5%. 

For those interested in how bismuth is produced, this Wikipedia quote may be of interest:

"According to the United States Geological Survey, world 2009 mine production of bismuth was 7,300 tonnes, with the major contributions from China (4,500 tonnes), Mexico (1,200 tonnes) and Peru (960 tonnes).^[11] World 2008 bismuth refinery production was 15,000 tonnes, of which China produced 78%, Mexico 8% and Belgium 5%.^[9]

The difference between world bismuth mine production and refinery production reflects bismuth's status as a byproduct metal. Bismuth travels in crude lead bullion (which can contain up to 10% bismuth) through several stages of refining, until it is removed by the Kroll-Betterton process or the Betts process. The Kroll-Betterton process uses a pyrometallurgical separation from molten lead of calcium-magnesium-bismuth drosses containing associated metals (silver, gold, zinc, some lead, copper, tellurium, and arsenic), which are removed by various fluxes and treatments to give high-purity bismuth metal (over 99% Bi). The Betts process takes cast anodes of lead bullion and electrolyzes them in a lead fluorosilicate-hydrofluorosilicic acid electrolyte to yield a pure lead cathode and an anode slime containing bismuth. Bismuth will behave similarly with another of its major metals, copper. Thus world bismuth production from refineries is a more complete and reliable statistic."

So I don't think bismuth supply and price would be affected by its use in solders.

Folks,

When the industry was preparing to transition to lead-free solders almost ten years ago (can it have been that long), tin-bismuth solders were serious candidates. Their low melting point, of about 138C, made these solders interesting candidates to replace tin-lead solder. However, if contaminated with lead, tin-bismuth solders can produce a eutectic phase that melts at 96C. In such situations the resulting solder joint exhibits poor performance in thermal cycle testing. Since early in the transition to lead-free solders it was expected that there would be numerous components and PWBs with lead-based surface finishes, this property made tin-bismuth solders unacceptable.

Another aspect of tin-bismuth solders is that they expand on cooling. This phenomenon can result in fillet lift in through-hole solder joints.

However, as we are now well into 2011, almost no components or PWBs have lead-containing finishes and many portable electronic devices have no through-hole components, so it may be time to reconsider tin-bismuth for some applications.

Some years ago, Hewlett Packard (HP) had performed work to show that adding 1% silver to tin-bismuth solder enabled this alloy to outperform eutectic tin-lead solder in 0 to 100C thermal cycle testing. Even at these low reflow temperatures, HP demonstrated solder joint strength with SAC BGA solder balls that was 65% that of tin-lead solder. Expanding on this work, Indium Corporation's Ed Briggs and Brook Sandy performed stencil printing and reflow experiments consistent with the requirements of current miniaturized components using this 57Bi-42Sn-1Ag solder. All of their results were promising. Ed presented a paper at SMTA Toronto,summarized the Hewlett Packard work, and reviewed the results of this new work.

So for applications consistent with 0-100C thermal cycling, 57Bi-42Sn-1Ag solder may be something to consider if the high temperature of SAC solder paste is an issue to components or PWBs in a product

Cheers,

Dr. Ron 

PS: Read my follow-on posting about bismuth.

Folks,

Many people have been infatuated by the price of gold in recent months, but the price of silver has also skyrocketed. In 2000 silver was about $3.00 per troy oz. In the eight years that followed, its price grew to $15/oz. Today it is trading at over $41/oz! This price is almost an all time high, except for the time when the Hunt brothers tried to corner the silver market in 1980. The aberration of their efforts jolted the silver price to just short of $50/oz, but it settled down to $11 or so after the Hunts came under margin call and other pressures.

Unfortunately, the dramatic price increase today, does not appear to be an aberration. Although we may hope that it will soon drop to more historic levels, we may not have reason to expect that it will.

Although not as dramatic, tin and copper have experienced significant prices increases as well. The price of tin has doubled in the last year to $15/pound and copper has increased from about $3/lb to $4.50.  These metals are obviously key ingredients in critical electronic materials such as solder pastes, solder bar, and solder preforms.

In addition, oil, which is used for most organic electronic materials such as PWB resins, flip chip underfill, and epoxy fluxes, has increased to $110/bbl - approaching its all time high of $145/bbl.

All of these price increases have a significant impact on the electronic materials supply chain. Although we are used to price decreases in the cost of our mobile phones and PCs, at this point in time, the price of the materials that go into these devices will be increasing.

As one materials supply chain executive commented at APEX, “It’s not like we can be clever and somehow work around the price increase of silver and these other materials, we have to pass it on to our customer, or go out of business.”

Folks,

It was Wednesday evening and I had just finished a brief pitch on applications of SPC  to a group of twenty. I was followed by Jim Hall,  he spoke of process mapping using SIPOC.  So did these folks have solder paste under their fingernails, or wave solder flux stains on their shirts, or, perhaps, a solder preform or two stuck in their pant leg cuff? None of these souls would have had any of this type of trace evidence of electronic assembly on their person. You see, they were all medical doctors and students at Harvard’s  famed medical school.   (I hope it is OK that the proud dad shares that my daughter Jessica is a colleague of these folks.)

Jim and I were presenting to the doctors, because they are interested in process optimization in the healthcare industry. The event was hosted by Dr. Andy Ellner.  He is a professor and doctor at the medical school and is a focal point for these process improvement efforts. I was introduced to him in the summer of 2009 by Dartmouth’s  new President Jim Kim. 

In November of 2009, Jim Hall, our colleague Larry Parah, and I facilitated Andy’s team in dramatically improving the prescription refill process in Brigham and Women’s Hospital Clinic.  Jim and I plan on working with Andy in similar efforts over the next year or two.

The most striking thing that Jim and I left with on Wednesday evening was how profoundly interested these doctors and students were in healthcare process optimization. The Q&A session lasted nearly an hour.

Ah, yes, would that our many colleagues in electronic assembly were as interested in optimizing their processes!

Folks,

HIS iSuppli performed a teardown analysis of an iPad . They estimate that the cost of the bill of materials (BOM) is  $336.60 for a GSM 32 GB version. 

Observing the BOM is interesting. Many of us in the assembly business are disappointed that this innovative product is assembled in the Far East. However, assembly cost savings may not be the only reason for Far East assembly. If one looks at the bill of material, almost all of it comes from the Far East, the memory, the display, the applications processor, etc.

iSuppli estimates the assembly cost as $10 dollars. This amount strikes me as too low. A good rule of thumb is that assembly cost is at least $0.01 per I/O assembled. So if the all of the PCBs have more than 1000 I/O the assembled cost is likely higher. It is hard to estimate the I/O, but there are 100s of passives alone and typically the BGAs will have high I/O counts.  I just noticed that I don't see passives on the BOM.  Typically they are about 90% by part count of the electrical components. 
They are inexpensive, but with their high numbers, their price can add up.

Solder paste is typically about 0.05% of assembled cost or about $0.20. When one considers the importance of good solder joints, it always pays to use the best solder paste available.

Cheers,

Dr. Ron

The image is from iSuppi's article.

Folks,

Many people responded to my recent post, In Search of Tin Whisker Fails in Lead-Free Soldering.  A few pointed out that the recent NASA report on the Toyota Unintended Acceleration Issue  discussed numerous tin whiskers that were found, one implicated in a failure.  The tin whiskers were emanating from tin plating.

We don't know, however, if tin whisker mitigation techniques were used. In a mission critical application, such as this, it would appear unwise to use RoHS-compliant electronics, especially since they are not required for automobiles.  In other words, autos are exempt from RoHS.  Let me be very clear, from a tin whisker perspective, I am uncomfortable with RoHS-compliant tin plating in mission critical applications.  Much more work needs to be done before such tin plating should be used in mission critical applications.

In addition, in response to my post, a number of people pointed out the difficulty of proving a tin whisker fail and the reluctance of any manufacturer to admit that their products had them.

But my quest remains unfulfilled; the question remains:

"... who knows of any verified tin whisker fails when tin whisker mitigation techniques where used? Tin whisker mitigation techniques typically use 2% bismuth or antimony in the tin, assure that the tin has a matte finish and use a nickel strike plating between the copper and the tin to minimize copper diffusion into the tin."

Restated, here is my point.  Since RoHS, quite a few people take a position something like this:

With RoHS-compliant assembly, even the world of non-mission critical electronics is at considerable risk of numerous catastrophic failures, due to tin whiskers, that will cost $100s billions.

I still maintain, that with mitigation techniques, such as recommended by iNEMI, tin whisker control, for non-critical electronics, can be manageable.

As I pack up to leave my office today at Thayer Engineering School at Dartmouth, I am across the aisle from the chaps that provide our computers and IT support.  They buy millions of dollars of electronics a year.  In chatting with them they state two things:

1. They have noted no difference in electronics reliability since RoHS implementation

2. On the very rare occasion that they get an electronics failure, it is almost always a hard drive.

Bottom line: Except for hard drives, modern electronics are very reliable for their use life.

I expect my quest will uncover some tin whisker fails, even with mitigation, but the fails will most likely be isolated and not a significant threat to the industry at large.

Cheers,

Dr. Ron

The image is from Dr. Henning Leidecker of NASA, one of the world's leading tin whisker experts.

Folks,

In a recent post, I shared my perspective on the pluses, minuses and neutral aspects of lead-free solder assembly. In the minus category, I listed tin whiskers. A few people commented that tin whiskers were the biggest concern in lead-free assembly. I have trouble understanding this perspective. I’m not saying these folks are wrong, just that I don’t understand their viewpoint.

First, let me say that I appreciate the concern for tin whiskers in mission critical electronics such as military, aerospace and medical. I am also sympathetic to the fact that, even though these types of electronics are exempt from RoHS, they may have to use RoHS compliant products because non-RoHS compliant products may not be available.

When I discuss the topic of tin whiskers, people will point me to NASA’s tin whisker failures website . However, when one goes to the site, there are only about twenty tin whisker fails referenced, many due to bright tin plate. Bright tin plate should never be used in mission critical electronics as it is virtually assured of producing tin whiskers. In addition, many of the articles referenced do not talk about tin whisker fails. Few if any fails are discussed relevant to RoHS (i.e. almost all fails discussed are prior to July 2006.)

I do not want to minimize the significance of tin whisker fails, some of them cost 100s of millions of dollars (e.g. satellite failures). In addition, there have been a few papers that have discussed the formation of tin whiskers even if mitigation techniques are used. Tin whiskers clearly can cause problems, but do not appear to be common, especially if mitigation techniques are used.

So here is my question, who knows of any verified tin whisker fails when tin whisker mitigation techniques where used? Tin whisker mitigation techniques typically use 2% bismuth or antimony in the tin, assure that the tin has a matte finish and use a nickel strike plating between the copper and the tin to minimize copper diffusion into the tin.

Surely if tin whiskers are a major concern, there should be many fails in the over $3 trillion worth of RoHS compliant electronics manufactured since July 2006.

Folks,

I thought I would take a stab at listing the minuses, pluses, and “it’s a wash” aspects of assembling with lead-free (LF) solder. Here are my first thoughts. Please tell me what I missed or disagree.

Cheers,

Dr. Ron

Minuses

1.    Pb-Free requires higher reflow temperatures
The Tm for LF solders, in the 217-229C range, has created numerous challenges:

a.      PWB warpage and damage

b.      Component damage

c.      New defect modes such as graping and head-in-pillow defects (although concurrent reduction in solder paste deposit sizes for 0201 and 01005 passives and 0.3 mm CSPs also exacerbate these defects)

d.      Defects related to increased oxidation

e.      Increases in voiding

f.       Increases in tombstoning

2.      The higher cost of LF solder, mostly for wave soldering

a.      It’s not just the silver, tin is much more expensive than lead

3.      Poorer wetting of LF solders, creating the most significant challenges in wave soldering

4.      More rapid copper pad dissolution on PWBs in wave soldering

5.      LF solder attack of wave solder machine components

6.      LF reliability in harsh thermal cycle testing appears poorer than tin-lead solders

7.      Tin Whiskers

It’s a Wash

1.      Short-term reliability in consumer product-type environments

2.      Protection of the environment if discarded products are improperly disposed of

a.      Lead in electronics has never been shown to cause a problem in land fills

3.      Since July 2006, about $3 trillion of products have been manufactured with LF solder, with no “the sky is falling”-type of problems

Pluses

1.      LF solder's poor wetting enables finer lead spacings (see photo Courtesy of Motorola)

a.      It may be argued that some modern electronic products (e.g. smartphones) could not be made with tin-lead solder

2.      It is safer to recycle LF solders, especially if performed in a non-controlled environment