The Return of PCB Chat

After a hiatus, we have relaunched PCB Chat as a podcast.

Our first guest is Mike Konrad of Aqueous Technologies, who shares his experience with what happens when a contract manufacturer follows its customer’s instructions to the detriment of the product. The outcome: Product failures, blame, drama, and a really big lawsuit.

We apologize in advance for the imperfect audio; we are still getting up to speed on the editing tools. But we think you’ll enjoy this, the first in what will be a regular series of interviews and conversations.

Mistakes Were Made — Too Much Ground Isolation

I recently ran a batch of my Neo Pixel clock boards through the factory here. It’s an Arduino UNO-based design that I made for myself not long ago. It sports an Atmega328P, with bootloader, an FT231X USB chip, and a DS3231 real time clock (RTC) chip. Pretty standard stuff. It doesn’t even use small parts. All the passives are 0805 size. There’s nothing exotic here. So, where did I go wrong?

I also used my 3D printer to make a clock frame to hold this board and a 60-pixel ring of NeoPixels, from Adafruit. I found that with the micro USB connector on the top of the board, it’s a little awkward to plug in the USB cable, so I put pads for the connector on the back side of the board. Depending on exactly where and how the board will be used, the micro-USB, button switches, and clock backup battery can all go on either the front or back surface of the board.

Programming the bootloader worked as expected, so I assumed it was just a job well done. Except it wasn’t. When I plugged in the micro USB cable, the RX and TX LEDs flickered briefly, but the board wasn’t recognized by my PC.

Take a look at the back side of the PCB and see if you can find my mistake (spoilers after the photo).

I ran a 24 mil trace around the back side of the board to supply power to the NeoPixels. That’s not a problem, except that I closed the loop on that trace, and didn’t put a path for the ground to get across the trace.

Follow it around, and notice that the ground connections to the u-USB connector don’t go anywhere except to this part of the plane. Ugh.

Duane Benson
Cassini’s gone now.

Dont’ Forget PCB West Next Week!

Next week is the 26th annual PCB West, the preeminent trade show in the Silicon Valley for the electronics supply chain.

As those who have attended before know – and there are quite a few of you – PCB West focuses on the design and manufacture of PCBs, HDI, electronics assembly and printed circuit board test, and gives engineers, designers, fabricators, assemblers and managers an opportunity to improve skills, increase knowledge and network with peers, colleagues and experts. With an emphasis on training – half the presentations
are at least 2 hours in length – there is no place better to get real, practical, in-depth information.

Our three-day conference features:
• More than 70 presentations on the hottest topics, including noise control, flex circuits, and diagnosing assembly defects. This is our largest conference yet!
• More than 15 day-long tutorials or half-day seminars

• Sessions for all levels of experience and training, from novice designer and engineer to seasoned pro
• Speakers from Analog Devices, TTM Technologies, NXP Semiconductor and many more top companies
• The ever-popular Rick Hartley, Doug Brooks and Susy Webb
• An all new PCB/EMS Management track with special sessions aimed at helping executives make the capital investment and hiring decisions that shape their companies
• Three free day-long tracks on Sept. 13, with topics ranging from signal integrity and IoT PCBs to 3D printing technologies.

Also next week, a special 2.5-day IPC Designers Council Certification Program powered by EPTAC.

All conference attendees receive free admittance to the one-day exhibition Wednesday, Sept. 13, which includes a complimentary luncheon and evening reception, both on the show floor.

For more information or to register, click here.

Looking forward to seeing you at the show! And as always, please feel free to share your thoughts.

QFN Center Pad Revisited

The QFN (quad flat pack, no leads) package can no longer be considered exotic. It was when I first wrote about it a decade ago, but not anymore. In fact, with the wafer-scale BGA, it’s one of the more common packages for new chip designs.

Not all QFNs come with an exposed metal pad underneath, but most do, and that can cause problems with reflow solder. The pad itself isn’t the problem, but improper solder paste stencil layer design can be.

The default stencil layer in the CAD library footprint might have an opening the full size of the metal pad. If that’s the case, modify the footprint so that there will be 50% to 75% coverage with solder paste (Figure 1). If you don’t, it may result in yield problems. With a 100% open area, the likely result is too much solder in the middle. The part will ride up, or float, and may not connect with all of the pads on the sides of the part.

Figure 1

Figure 1. The optimal QFN footprint will have 50% to 75% solder paste coverage.


Figure 2 shows a stencil with too large an opening in the center, a segmented paste layer in the CAD footprint, and the resultant segmented stencil.

Figure 2

Figure 2. Stencils shown with too large an opening in the center (left), segmented paste layer (center), and the resultant segmented stencil (right).


You may note that I said to shoot for 50% to 75% coverage and ask: “Well, is it 50% or 75%? What gives?”

True, that is a bit of ambiguity. Anything in that range should be fine for prototype boards, however. If the assembly is headed for volume production, work with the manufacturer to tweak the design for best high-volume yield.

The good news on this front is that many QFN manufacturers and parts library creators have taken notice. It’s far more likely now than it was 10 years ago to find a datasheet correctly illustrating this, and footprints created correctly. But, always check your footprints to make sure.

Duane Benson

So Long, Sola

I have to say, I didn’t think Jure Sola would or could last this long. The cofounder of Sanmina, Sola was one of the poster boys for wanton M&A excess, snatching up more than a dozen companies or OEM plants during the late 1990s and early 2000s. The spree culminated in the purchase of SCI Systems in mid 2001, a $6 billion deal that saddled the company with so much debt, when the ensuring tech collapse occurred, it was forced to take 20 straight quarters of “one-time” charges.

Most execs couldn’t have survived such a bloodletting. Sola wasn’t most execs, however. He continued to place his bets on fabricating in the US — in a memorable line, he told an IPC Printed Circuit Expo audience that “plating was in his blood” — and Sanmina remains the second (or third) largest board supplier in North America. Moreover, he correctly swung to the military and aerospace markets, eschewing the PCs that SCI was so dominant in.

Today the company is half the size in revenue of its peak, but consistently profitable.

Come October Sola will ride off into the sunset with his legacy intact, perhaps not the most beloved man to run a major PCB company, but a success nonetheless. In this era, that’s no small thing.


Green Herring

For those newbies, Bob Herring was the perfect example of good timing, building up and selling not one but two board shops. The first, Industrial Circuits, was sold in 1989 for $60 million. The latter one, Herco Technology (which we profiled multiple times in PC Fab), went for $122 million in 2000, just a year before the tech crash. (The buyer of Herco, Teradyne, closed it less than two years later. The former Industrial Circuits lasted less than one year longer before Toppan shut the doors.)

In case you were wondering where Bob went, well, he started his own network cable news channel. It now is televised in some 30 million homes.

Guess there is life after PCBs!



Gerber: The Format that Just Won’t Die

I’m a big believer in standards but I’m not so sure why IPC is pushing a Gerber Coupon Generator when it has spent so many years developing IPC-2581, a much more comprehensive electronic data format.

Now in its B revision, IPC-2581 has been implemented in trial and production, and represents the most comprehensive set of industry requirements for printed circuit board fabrication, assembly, and test in a data-centric, open, license-free, industry driven standard format. Moreover, the consortium supporting its adoption boasts more than 90 members, including all the major PCB software vendors, plus a host of major OEMs, equipment suppliers, manufacturers, and service suppliers.

It’s time the emphasis be placed on moving the industry out of the buggy era. (Pun intended.)

Full disclosure: I’ve been a member of the IPC-2581 task group since its inception, and spent several years at IPC working on the predecessors to IPC-2581.

Where the Jobs Are

This news item from the Associated Press cuts to the heart of the matter when it comes to reshoring of manufacturing and why skeptics (including this humble writer) abound over whether Foxconn, among others, truly intend to set up large manufacturing plants in the US:

WASHINGTON (AP) — President Donald Trump brought two dozen manufacturing CEOs to the White House on Thursday and declared their collective commitment to restoring factory jobs lost to foreign competition.

Yet some of the CEOs suggested that there were still plenty of openings for U.S. factory jobs but too few qualified people to fill them. They urged the White House to support vocational training for the high-tech skills that today’s manufacturers increasingly require — a topic Trump has seldom addressed.

“The jobs are there, but the skills are not,” one executive said during meetings with White House officials that preceded a session with the president.

The truth is there are hundreds of thousands of manufacturing jobs available in the US today. The US Census Bureau puts the figure at just shy of one million. In talking with circuit board fabricators and assemblers over the years, the biggest impediment to hiring is not lack of work but rather lack of qualified workers.

My belief is that the demographics of electronics design and manufacturing resemble a bimodal distribution (two humps), whereby workers over 50 years old represent the largest group by age and workers aged 20 to 30 the second largest. Those aged 30 to 50 are the smallest group (the valley in the graph, see below). My thesis is that workers in that segment were coming online right about the time the North American electronics industry cratered — late 2001 to early 2004, leaving them either out of jobs or unable to crack the then much-smaller workforce that was left after the tech recession.

(The graph below illustrates the basic concept, although in reality the right hump would be higher than the left as there likely are more workers over 50 than under 30 in electronics design and manufacturing today. But you get the idea.)

With the older wave starting to retire, coupled with an upturn in the industry’s fortunes starting around 2008, a new wave of workers has entered the industry. And while we often speak of the lack of millennials in manufacturing, a tour of Silicon Valley area shops takes the air out of that conversation. There, workers don’t ask where the young people are; they just look around — they are everywhere. And manufacturers are catering to them, setting up coffee (and more) bars inside their plants, creating workspaces that resemble outdoor atria that offset the traditionally sterile assembly lines.

Moreover, there is some concern that widespread move of manufacturing back to the US will only accelerate the implementation of robots, leaving thousands of operators on the sidelines. In anticipation, robot makers are ramping capacity, in some cases by as much as two times. This is not without precedent. Those of us who were around when PCB fabrication and assembly migrated to China en masse in the late 1990s/early 2000s recall how common semiautomatic machines were then. It was a nod to the Chinese government, which was adamant about protecting employment.

What’s your experience? Is your company weighing a return to the US? If so, will it come with an increase in automation?

(Please, no political comments.)

The Top 10 of 2016 — PCD&F

As we did with CIRCUITS ASSEMBLY, each year we review the 10 most-viewed features.

Keep in mind that the counts are not adjusted by the date of publication. Therefore, an article published in January has an advantage over one published in December. The month of publication is listed in parentheses.

10. “Designing PCBs for DDR Busses,” by Nitin Bhagwath (June 2016)

9. “Ensuring Reliable Products with New Rigid-Flex Design Rules,” by Jim Frey (September 2016)

8. “Empirical Confirmation of Via Temperatures,” by Doug Brooks, Ph.D. (February 2016)

7. “6 Pillars of PCB DfMA Success,” by John McMillan (March 2016)

6. “Insertion Loss Modeling,” by Jeff Loyer (January 2016)

5. “Price vs. Function in Today’s EDA Software,” by Chelsey Drysdale (December 2016)

4. “IPC-1782: The New Dawn of Electronics Traceability,” by Michael Ford (December 2016)

3. “5 Common Mistakes in Board Design,” by Arbel Nissan (December 2016)

2. “DC Analysis of PDN: Essential for the Digital Designer,” by Jeff Loyer (March 2016)

1. “Via-in-Pad Design Considerations for Bottom Terminated Components on PCB Assemblies,” by Matt Kelly, Mark Jeanson and Mitch Ferrill (March 2016)

Thanks, as always, for reading!

Mentor’s Final Sale

In the end, Paul Singer did what Carl Icahn couldn’t: Got Mentor sold.

Singer, the hedge fund manager known for taking large positions in companies and pushing for tough changes, breakups or sales, started accumulating shares of the EDA CAD company earlier this year. In September, it was revealed that his company Elliott Management, had bought up 8.1% of Mentor’s stock. Elliott immediately started lobbying for changes.

For Mentor, it could have seemed like a recurring bad dream. The company had been through this before, starting six years ago, when Carl Icahn, himself a famed corporate raider, began acquiring shares and issuing accusations of waste throughout the organization.

Icahn’s relationship with Mentor was public and acrimonious. Soon others joined the fray. Everything went under the microscope, from spending on marketing to the personal wealth of the directors. CEO Wally Rhines came under attack for pocketing $65 million from Mentor while the company generated only $113 million in free cash between 2001 and 2011. Icahn even offered to buy the company outright for $1.9 billion, a figure Mentor’s board dismissed as too low.

The board, however, couldn’t outright avoid Icahn and the others, who at their peak owned more than 20% of the company. Instead, they executed a “poison pill” amendment to its bylaws, making a hostile takeover more expensive and risky.

Icahn managed to land three directors on Mentor’s board but never affected the breakup or sale he had hoped for. Mentor bought back half his shares in February for $146 million, and he sold the last of his holdings in May.

Icahn certainly made a pile of money off Mentor. It took Singer, however, to fundamentally change the trajectory of the company.

Upon Elliott’s announcement, Mentor charted a different course. Instead of waging another attempt to fend off the barbarian at the gate, this time it signed on with Bank of America as an advisor to a possible sale. The deal with Siemens came quickly thereafter.

Singer’s stance was Mentor was undervalued by 20%. The price Siemens is paying — $4.5 billion — suggests even he was low.

Siemens was never a stretch as a suitor. As far back as 2011, we suggested the German conglomerate was one of a few companies that made sense to possibly acquire Mentor.

For some involved, the deal completes a circle. Mentor will become part of Siemens PLM, whose president Tony Hemmelgarn is a former Integraph executive. In fact, he was director of sales and marketing when the company spun off its Electronics Division into a wholly owned subsidiary known as VeriBest. Mentor then acquired VeriBest for $19 million in 1999.

It does spell the end to Mentor after 35 years as a standalone company. Founded by a trio of Tektronix engineers — Tom Bruggere, Gerry Langeler and Dave Moffenbeier — in 1981, Mentor added the PCB division through a merger with CADI in 1983. (Just after, Mentor hired the legendary John Cooper, who with partner David Chyan eventually developed the first shape-based router.)

In all likelihood this also means an end to Wally Rhines’ 23-year tenure as head of Mentor. He will be remembered as a steady leader during a period of great upheaval and M&A in EDA. On his watch, Mentor’s revenues grew from $340 million to nearly $1.2 billion. That’s a pretty darn good run.

Less clear is how the rest of the industry will react. Siemens gives Mentor exceptionally deep pockets, a buffer against meddling shareholders, and an extensive market for technology both as a customer and to partner with. The focus on “concept to system” just got a big boost.

By comparison, on the PCB side, the door has been slammed shut on one of the exit strategies for Cadence and Altium. Dassault has been rumored to be kicking the tires on Altium; this could trigger a move. Will PTC, which shares a Boston area neighborhood with Cadence, be compelled to act as well in order not to get shut out of ECAD? As one longtime industry observer noted to me recently, “It’s about the form factor.” OEMs want to design product in its entirety, not in silos of electrical, electronics, mechanical and wire harness. Given that, it’s a safe bet the M&A in ECAD won’t stop with this deal.