Dr. Ron: Professor Coleman, it has been quite a long time since we last chatted.
Professor Patty: I agree, too long! And please, Dr. Ron, call me Patty.
DR: As long as you call me Ron.
PP: That would be impossible! Everyone knows you as “Dr. Ron.”
DR: (Chuckles) OK, you win.
DR: Patty, in statistics you teach how to perform hypothesis tests on one sample and on two samples. But, what if we are faced with finding, say the best solder paste, when we have 3 or more samples ?
PP: It is strange that you should ask that–I just added it to my lectures in the stats class I teach at Ivy U. When faced with this challenge, you have to use analysis of variance (ANOVA). Let’s consider an example where you have three solder pastes and you are measuring transfer efficiency (TE). (See Figure 1.) Typically, the goal would be at TE of 100%.
Figure 1. Transfer Efficiiency (TE) for Three Solder Pastes
DR: So it is quite obvious, from Figure 1, that vendor A has a lower TE than vendors B and C, but is the difference statistically significant?
PP: This is where ANOVA helps. With Minitab, we can statistically compare the samples using ANOVA. The results are in Figure 2.
Figure 2. Minitab Output of TE Data Comparisons for Three Solder pastes
DR: It looks like the difference between the TE of vendor B and C is not significant, as seen in the lower graph in Figure 2. Am I reading it correctly?
PP: Yes! And the two upper graphs comparing vendors A and B and A and C show that these differences are significant.
DR: So, we might drop vendor A from our evaluation and focus on B and C for further testing?
You have undoubtedly heard in the news that some predict AI will destroy civilization. Even the “Godfather” of AI, Geoffrey Hinton has quit Google and now feels compelled to warn of AI’s risks.
I recently spoke on this topic at Bio 2023 in Boston. I don’t want to minimize the stunning advances in AI as displayed recently in ChatGPT. But, its threat to civilization is at worst nuanced.
However, those of us in the teaching world are concerned of the impact of ChatGPT in education. It can write quite good papers on any topic, although errors are common. However, its success in tackling exams is nothing short of stunning. ChatGPT can pass the bar exam for lawyers at the 90 percentile, ChatGPT4 can also pass a wide variety of standardized tests such as SATs, APs, GRE’s and even a test for Sommeliers! These skills of AIs like ChatGPT raises the concerns of academic cheating and the hindering the learning of our youth.
An article by the New York Post sums up these types of concerns: “Long known for his warnings on the potential dangers of AI.” Tesla CEO Elon Musk on Monday cautioned that even a “benign dependency” on these complex machines can threaten civilization.” (NY Post May 2, 2023)
Another concern RE AI is crime. Criminals are very clever and will use AI to scam people among other things. One way scamming is done is through deepfakes. A typical deepfake is a digital file of a person’s voice or video image that is fake. There is no end of ways that criminals can use deepfakes to extort money.
So, the latest AI technology is indeed unsettling. However, there is one area in which AIs are utter failures and likely will be for generations: The Physical Embodied Turing Test. This test would essentially challenge an AI robot to assemble something like IKEA furniture from a kit with written instructions. The state of AI development is so far behind in this regime that it makes this task science fictional. In 2019, my then 8 year-old grandson, Nate Su, assembled the Apollo Spacecraft’s 1969 parts, see Figure 1, in 4 hours. No AI can come close to doing this. For AIs to be a real physical threat to humanity, significant advances would be required in interacting in the physical world.
Figure 1. Eight year old Nate Su assembled the 1969 parts of the Apollo rocket in 4 hours. No AI can come close to this feat.
One of the reasons AIs performs so poorly in the physical world is that they have no body. So much of human’s interaction with the physical world is through our bodies and our senses of taste, touch, sight and sound. In addition, humans have years of context about the physical world that an AI Robot would have to learn. As an example, as I write this post I can look out into my living room and see the hard wood floor, covered by carpets, the furniture, our Mumford fireplace, the TV, the phone, and on and on. Each of these items have a long story connected with them. For an AI to navigate our physical world, it would need to “learn” about the myriad aspects of it.
So, don’t have any concern that an AI robot like M3gan (Figure 2.) is on the horizon. Even more humbling to the AI world is Steven Pinker’s adage that no AI can empty a dishwasher.
Figure 2. Robots like Megan will not be on the horizon for decades, if ever.
When comparing the volume of solder paste provided by a circular versus square aperture, consider that if the side of the square is D and the diameter of the circle is also D, the square has greater than 25% more area. (i.e., (1-0.785)/0.785 = 0.274). See Figure 1.
Figure 1. Square vs. circle areas.
However, the greater area of a square is not the only reason square apertures deposit more solder paste. The curving of the circular aperture enables more surface of the stencil to contact more of the solder particle’s area. See Figure 2. So, the solder particles will adhere to a cicular aperture more readily and not adhere to the pad, resulting in a smaller solder paste deposit.
Figure 2. The curving of a circular aperture results in more contact area with solder particles than a square aperture
These two effects can result in dramatically different soldering results, as seen in Figure 3. Using the square aperture provides so much more solder paste; when compared to what a circular aperture provides, it is stunning in the soldering result.
Figure 3. Circular aperture/pad (left) and square aperture/pad (right), using the same Type 3 powder size, area ratio, flux chemistry (no-clean), and reflow profile (RTP)
I am reposting an updated blog post on Cp and Cpk calculations with Excel, as I have improved the Excel spreadsheet. If you would like the new spreadsheet, send me an email at [email protected].
One of the best metrics to determine the quality of data is Cpk. So, I developed an Excel spreadsheet that calculates and compares Cps and Cpks.
———————————————————————————————————-
Folks,
It is accepted as fact by everyone that I know that 2/3 of all SMT defects can be traced back to the stencil printing process. A number of us have tried to find a reference for this posit, with no success. If any reader knows of one, please let me know. Assuming this adage is true, the right amount of solder paste, squarely printed on the pad, is a profoundly important metric.
In light of this perspective, some time ago, I wrote a post on calculating the confidence interval of the Cpk of the transfer efficiency in stencil printing. As a reminder, transfer efficiency is the ratio of the volume of the solder paste deposit divided by the volume of the stencil aperture. See Figure 1. Typically the goal would be 100% with upper and lower specs being 150% and 50% respectively.
Figure 1. The transfer efficiency in stencil printing is the volume of the solder paste deposit divided by the volume of the stencil aperture. Typically 100% is the goal.
I chose Cpk as the best metric to evaluate stencil printing transfer efficiency as it incorporates both the average and the standard deviation (i.e. the “spread”). Figure 2 shows the distribution for paste A, which has a good Cpk as its data are centered between the specifications and has a sharp distribution, whereas paste B’s distribution is not centered between the specs and the distribution is broad.
Figure 2. Paste A has the better transfer efficiency as its data are centered between the upper and lower specs, and it has a sharper distribution.
Recently, I decided to develop the math to produce an Excel® spreadsheet that would perform hypothesis tests of Cpks. As far as I know, this has never been done before.
A hypothesis test might look something like the following. The null hypothesis (Ho) would be that the Cpk of the transfer efficiency is 1.00. The alternative hypothesis, H1, could be that the Cpk is not equal to 1.00. H1 could also be that H1 was less than or greater than 1.00.
As an example, let’s say that you want the Cpk of the transfer efficiency to be 1.00. You analyze 1000 prints and get a Cpk of 0.98. Is all lost? Not necessarily. Since this was a statistical sampling, you should perform a hypothesis test. See Figure 3. In cell B16, the Cpk = 0.98 was entered; in cell B17, the sample size n = 1000 is entered; and in cell B18, the null hypothesis: Cpk = 1.00 is entered. Cell B21 shows that the null hypothesis cannot be rejected as false as the alternative hypothesis is false. So, we cannot say statistically that the Cpk is not equal to 1.00.
Figure 3. A Cpk = 0.98 is statistically the same as a Cpk of 1.00 as the null hypothesis, Ho, cannot be rejected.
How much different from 1.00 would the Cpk have to be in this 1000 sample example to say that it is statistically not equal to 1.00? Figure 4 shows us that the Cpk would have to be 0.95 (or 1.05) to be statistically different from 1.00.
Figure 4. If the Cpk is only 0.95, the Cpk is statistically different from a Cpk = 1.00.
The spreadsheet will also calculate Cps and Cpks from process data. See Figure 5. The user enters the upper and lower specification limits (USL, LSL) in the blue cells as shown. Typically the USL will be 150% and the LSL 50% for TEs. The average and standard deviation are also added in the blue cells as shown. The spreadsheet calculates the Cp, Cpk, number of defects, defects per million and the process sigma level as seen in the gray cells. By entering the defect level (see the blue cell), the Cpk and process sigma can also be calculated.
Figure 5. Cps and Cpks calculated from process data.
The spreadsheet can also calculate 95% confidence intervals on Cpks and compare two Cpks to determine if they are statistically different at greater than 95% confidence. See Figure 6. The Cpks and sample sizes are entered into the blue cells and the confidence intervals are shown in the gray cells. Note that the statistical comparison of the two cells is shown to the right of Figure 6.
Figure 6. Cpk Confidence Intervals and Cpk comparisons can be calculated with the spreadsheet.
This spreadsheet should be useful to those who are interested in monitoring transfer efficiency Cpks to reduce end-of-line soldering defects. It is not limited to calculating Cps and Cpks of TE, but can be used for any Cps and Cpks. I will send a copy of this spreadsheet to readers who are interested. If you would like one, send me an email request at [email protected].
The Pareto Chart is a simple way to plot failure data that gives priority to the failure modes with the highest number of fails. This technique was developed by Vilfredo Pareto in the late 1800s to early 1900s. Pareto was studying social and economic data in Italy. He was one of the first to observe the 80/20 rule. In that, about 80% of property in Italy was owned by 20% of the people. Today many people use this rule. I have heard salespeople say that 80% of their business is from 20% of their customers as one of many applications of this rule.
In categorizing fails in electronics assembly, about 80% of fails are in 20% of the failure modes. Let’s look at an example (Figure 1). In this figure, we have plotted the number of fails versus the failure mode. Note that shorts is the most common failure at about 300, whereas opens is 75, missing components is about 50, and solder balls about 35.
Figure 1. A Pareto Chart of Electronic Assembly Failure Modes
These data should be used to develop a continuous improvement plan. Obviously, shorts should be focused upon first. Typically, one would use process data such as statistical process control (SPC) data to solve the shorts problem, most likely looking at a process metric like the volume of the stencil printed deposit.
I developed a graph similar to Figure 1 when I visited a client. The manager was convinced that solder balls were a big problem. When I asked the quality engineer for the supporting data, he said there was none. So, I asked if they collected failure data; he said they did. I then asked what they did with the data; he said they filed it away having never looked at it!
I asked to see the last several weeks of data and I plotted the data similar to that in Figure 1. It ended up that solder balls was the fourth biggest defect, not the first. As a result of using a Pareto Chart, the company focused on fixing their defect with the greatest number first, etc.
Pareto Charting is a simple yet crucial process in continuous improvement.
Here are the answers to theSMT IQ Test of a short while ago.
What does the “A” in SAC305 stand for? ANSWER: SAC stands for tin (Sn), silver (Ag), and copper (Cu). The “305” indicates 3.0 percent by weight silver, 0.5% copper, and the balance (96.5%) tin.
The belt speed on a reflow oven is 2 cm/s. The PCB with spacing is 36 cm. What is the maximum time that the placement machines must finish placing the components on the PCB to keep up with the reflow oven? ANSWER: Time (s) = product length (cm)/belt speed (cm/s) = 36 cm/2 cm/s = 18 sec.
In mils, what is a typical stencil thickness? ANSWER: In range of 4 to 8 mils.
BTCs are one of the most common components today; a subset of BTCs is the QFN package.
What does BTC stand for? ANSWER: Bottom terminated component
What does QFN stand for? ANSWER: Quad Flat Pack No Leads.
What is the melting temperature of tin-lead eutectic solder? ANSWER: 183° C.
In mm, what is the finest lead spacing for a PQFP? ANSWER: Most common is 0.4 mm. A few have 0.3 mm, but these smaller spacings are hard to process.
Are solder pastes thixotropic or dilatant? ANSWER: Thixotropic; the viscosity of solder paste drops when it is sheared (i.e forced through a stencil). Dilatant materials stiffen when sheared.
In stencil printing, what is response to pause? ANSWER: When stencil printing is paused, the viscosity of the solder paste can increase; this situation would be considered a poor response to pause. Pastes that have stable viscosities during pausing are considered to have good response to pause.
For a circular stencil aperture for BGAs or CSPs, what is the minimum area ratio that is acceptable? ANSWER: Typically greater than 0.66, although some solder pastes can print well a little lower than this.
What are the approximate dimensions of a 0201 passive in mils? ANSWER: Approximately 20 by 10 mils.
Soldering enables modern electronics. Without solder, electronics would not exist. Copper melts at 1085°C, yet with solder, we can bond copper to copper at about 235°C or less with current lead-free solders. These lower temperatures are required, as electronic packages and PWBs are made of polymer materials that cannot survive temperatures much above 235°C.
Before the advent of RoHS, tin-lead solders melted at about 35°C less than lead-free solders. So today, soldering temperatures are at the highest in history. For some applications, it would be desirable to have solders that melted at closer to tin-lead temperatures. This desire has increased interest in low-melting point solders, such as tin-bismuth solders. Eutectic SnBi melts at 138°C, so reflow oven temperatures in the 170°C range can be used. These lower reflow temperatures are easier on some fragile components and PWBs and will reduce defects such as PWB popcorning and measling. However, the lower melting point of SnBi solders limits their application in many harsh environments, such as automobile and military applications. As a rule of thumb, a solder should not be used above 80 to 90% of its melting point on the Kelvin scale. For SnBi solder, this temperature range is 55.8 – 96.9°C. These temperatures are well below the use temperature of some harsh environments. In addition, SnBi solders can be brittle and thus perform poorly in drop shock testing.
So, the electronics world could use a solder that can reflow at a little over 200°C, but still have a high use temperature. This situation would appear to be an unsolvable conundrum. However, my colleagues at Indium, led by Dr. Ning-Cheng Lee, have solved it. They used an indium-containing solder powder, Powder A, that melts at <180°C and combined it with Powder B that melts at ~220°C. By reflowing at about 205°C, Powder A melts and Powder B is dissolved by the melted Powder A. To achieve this effect, the 205°C temperature must be held for approximately two minutes. The remelt temperature of the final solder joint is above 180°C. I discussed the phenomenon of a liquid metal dissolving another that melts at a higher temperature before. An extreme example of this effect is mercury dissolving gold at room temperature. So, don’t drop any gold or silver jewelry into a wave soldering pot and expect to fish it out an hour later!
Powder A would not be a candidate on its own as it displays some melting at 113°C and some at 140°C.
Using the criteria above, the use temperature of this new solder powder mix can be in the 89.4 – 134.7°C range, after reflow, as the remelt temperature is above 180°C. Tests performed by Dr. Lee and his team have shown the resulting solder joints also have good to excellent thermal cycling and drop shock performance.
Figures 1-3 show schematically how the melting of the two powders would melt at a peak reflow temperature of 205°C.
Figure 1. Powder A and Powder B at room temperature.
Figure 2. At 205°C, Powder A has melted and it is starting to dissolve to Powder B.
Figure 3. After about a minute at 205°C, Powder B starts to dissolve. Given enough time, it will completely dissolve in Powder A, resulting in a new alloy that has a remelt temperature over 180°C, as well as good to excellent thermal cycle and drop shock performance.
To me, this invention is one of the most significant in SMT in a generation. It could be argued that it is like finding the holy grail of soldering: melting at low-temperature with a service life at high-temperature.
Cheers,
Dr. Ron
PS. I developed an Excel spreadsheet to calculate the use temperatures. It converts degrees C into K. I used it to calculate the use temperatures above. If you would like a copy, send me a note at [email protected].
Readers may remember that I have had an interest in tin pest for some time. Tin pest can occur if nearly-pure tin is exposed to cold temperatures (<13.2oC) for long periods of time. At the end of this post, I provide a short summary of the tin pest phenomenon. See this striking time lapse video of tin pest forming; I assume the time period is over many months.
The reason for this post is that a medieval scholar, Beata Lipi?ska, from Poland is studying tin pest and its effects on medieval culture, most notably in church organ pipes. She has contacted me to see if I can help her find papers that discuss tin pest from a historical point of view. If readers have any references that could help Beata, please contact her directly at [email protected].
Figure 1. Tin pest forms in Sn .05 Cu alloy from Plumbridge. See the paper referenced below.
What is Tin Pest? Tin is a metal that is allotropic, meaning that it has different crystal structures under varying conditions of temperature and pressure. Tin has two allotropic forms. “Normal” or white beta tin has a stable, tetragonal crystal structure with a density of 7.31g/cm3. Upon cooling below 13.2oC, beta tin slowly turns into alpha tin. “Grey” or alpha tin has a cubic structure and a density of only 5.77g/cm3 . Alpha tin is also a semiconductor, not a metal. The expansion of tin from white to grey causes most tin objects to crumble.
The macro conversion of white to grey tin takes on the order of 18 months. The photo, which is likely the most famous modern photograph of tin pest, shows the phenomenon quite clearly.
This photo is titled “The Formation of Beta-Tin into Alpha-Tin in Sn-0.5Cu at T <10oC” and is referenced from a paper by Y. Karlya, C. Gagg, and W.J. Plumbridge, “Tin Pest in Lead-Free Solders,” in Soldering and Surface Mount Technology, 13/1 [2000] 39-40.
This phenomenon has been known for centuries and there are many interesting, probably apocryphal, stories about tin pest. Perhaps the most famous of stories is that of the tin buttons on Napoleon’s soldiers’ coats disintegrating from tin pest while on their retreat from Moscow. Another common anecdotal story during the middle ages was that Satan was to blame for the decline of the tin organ pipes in Northern European churches, as tin pest often looks like the tin has become “diseased”.
Initially, tin pest was called “tin disease” or “tin plague”. I believe that the name “tin pest” came from the German translation for the word “plague” (i.e., in German, plague is “pest”).
To most people with a little knowledge of materials, the conversion of beta to alpha tin at colder temperatures seems counter-intuitive. Usually materials shrink at colder temperatures; they do not expand. Although it appears that the mechanism is not completely understood, it is likely due to the grey alpha tin having a lower entropy than white beta tin. With the removal of heat at the lower temperatures, a lower entropy state would likely be more stable.
Since the conversion to grey tin requires expansion, the tin pest will usually nucleate at an edge, corner, or surface. The nucleation can take 10’s of months, but once it starts, the conversion can be rapid, causing structural failure within months.
Although tin pest can form at <13.2oC, most researchers believe that the kinetics are very sluggish at this temperature. There seems to be general agreement in the literature that the maximum rate of tin pest formation occurs at -30o to -40oC.
I have enjoyed writing the Patty
and the Professor blog for about 10 years now. I’ve written about numerous
real-life electronics assembly examples that I have encountered in my career,
all disguised, of course.
To continue keeping things real, and to keep my readers
involved, I am inviting you to submit an authentic story from your career.
That’s right! You’re being invited to submit an idea, story, or experience that
can be built into the Patty & The Professor series.
Your experience will help many other electronics assembly
practitioners resolve their issues and avoid problems.
So, get your thoughts together, then shoot me an email at [email protected]. Share the details of your experience or observation. I may
ask a few questions to help me comprehend the full story. Then, I will write up
the segment and let you read it before posting. You will be credited, of
course.
Bonus: You will also receive either a Dartmouth hat or coffee mug (similar to, not exactly like, those pictured below)!
Contact me if you are interested in submitting a story. I
look forward to hearing from you!
“Well what should we do Professor?”
John said weakly.
“Clearly,
not shut the line down over the lunch break,” The Professor responded
quickly.
“We
can’t!” said John, “The operators are all friends and they
count on having lunch together.”
“How
much are they paid per hour?” asked The Professor.
“Ten
dollars,” replied John.
“You
can pay them $15 per hour and still make more profit if they keep the line
running over the lunch break,” The Professor opined.
“Fifteen
dollars per hour for the lunch time or the entire 40 hour week?” John
asked nervously.
“For
the whole week,” was The Professor’s reply.
“I
find that hard to believe,” John shot back.
“Consider
this,” said The Professor. “Your line is up only 9.7% of an 8 hour
shift, that’s only 47 minutes. Today you lost 95 minutes over the lunch hour.
You may be able to increase your uptime to greater than 15% by keeping the line
running over lunch. I modeled your business with ProfitPro3.0 cost estimating software. Your
company will make millions more per year if you keep the lines running over
lunch. I have worked with other companies to make this change; it is really
easy with a 30 minute lunch period. If 5 people normally run the line, you
have just one stay back during lunch. That way each person only misses the
regular lunch break once a week.”
John
thought optimistically, “There is such a thing as a free lunch.”
“Now,
let’s talk about what we can do to double the uptime from the 15% we will get
by running the lines over lunch,” said the Professor.
Patty
listened to all of this in amazement. The Professor was helping ACME more than
she thought possible.
Next
steps? Yes, John will keep his job. But, what is The Professor’s plan to get
uptime to 30% or more? And, we still haven’t learned where Patty will go to
dinner. Stay tuned for the latest.
Cheers,
Dr.
Ron
Dr.
Ron note: As surprising as this may seem, this story is based on real
events. The uptime numbers and improvements are from real examples. Any company
that can achieve 35% or more uptime can compete with anyone in the world, even
in low labor rate countries. Sadly, few companies know their uptime or have an urgency
to improve it.
