Tag Archives: printed circuit board

Good Times, Bad Times

Posted on by

Major printed circuit board fabricators are beginning to report their quarterly earnings, and the figures give some reason for concern.

By all accounts, 2010 was a banner year for PCBs in all regions. Sales were up 34% year-over-year in Germany through October, 18% in North America through November, and at least 10% overall worldwide. TTM and M-Flex, among others, are reporting record sales. DDi beat the pack to the recovery and has continued to spike.

But while the specialists are doing great, not all is well.

While Multek, a top 5 PCB provider, grew in the low-double digits sequentially, profits are scarcer, with the company saying the unit won’t be breakeven until the end of the current quarter (and that assumes the no dip in orders). Sanmina experienced problems in one of its factories (rumored to be Kuching, Malaysia) that sucked the margin out of what should be one of its more profitable businesses.

And that’s my concern: Even during a period where demand peaked, the largest players are still not consistently profitable. Based on experience, when the market slows, that means more factory closures or continued losses, or both. Another likely response is dropping their drawers on pricing, a move that inevitably ripples through the broader market.

Twenty years (!) in the PCB market has taught me this: If you can’t make a profit in an up market, you can’t make one in a down market.

There’s only one way to resolve this conundrum. The capacity increases have to stop. Let the factories remain full for a few years. Push back on OEMs that constantly demand price reductions with little regard for rising commodity prices and currency fluctuations. Try working together as an industry on this.

Are Soy PCBs Good for Your Health?

Posted on by

File under surreal.

A pair of University of Delaware researchers have developed a way to fabricate a low-Dk substrate made from soy.

Yes, soy.

Mingjiang Zhan and Richard P. Wool, who are part of the chemical engineering department of the University of Delaware, started with biobased resin acrylated epoxidized soybean oil (AESO), which they crosslinked with divinylbenzene (DVB) or chemically modified by phthalic anhydride. The DVB-crosslinked resins had a 14° to 24°C increase in their glass-transition temperatures (Tg), which was dependent on the crosslink densities. Tg increased linearly as the crosslink density increased. Phthalated acrylated epoxidized soybean oil (PAESO) had an 18 to 30% improvement in the modulus. The dielectric constants and loss tangents of both DVB-crosslinked AESO and PAESO were lower than conventional dielectrics used for printed circuit boards (PCBs).

The results, the researchers say, suggest that the new biobased resins with lower carbon dioxide footprint are potential replacements for commercial petroleum-based dielectric materials for PCBs.

Their work will be published next month in the Journal of Applied Polymer Chemistry. (It was published online in July.)

By the way, this isn’t Wool’s only attempt to trick nature. As part of another project, he is trying to carbonized chicken feathers so they can store hydrogen for fuel-cell vehicles.

Let’s Get Small

Posted on by

With 005005 components (metric 0201)  scheduled for production within two years, it becomes less attractive to place and solder what are truly microminiature parts using traditional means.

Robert Barbucha, a researcher at the Institute of Fluid-Flow Machinery Center’s Plasma and Laser Engineering in Gda?sk, Poland, is working on a laser device that may solve the problem. His method involves using lasers to project pathways on printed circuit boards covered with a layer of photopolymer. The concept isn’t novel — many engineers have suggested printing or lasering parts on boards — but in this case the laser itself is the focus of Barbucha’s research. He reportedly has begun working with a manufacturer in Gda?sk to produce a prototype.

Stay tuned.

On Markets and Makers

Posted on by

Despite my pleasure in seeing the forecast for printed circuit board production this year, I have to admit it’s depressing to contemplate how far the US market has fallen, both in size and share.

Dial back to 2000, and North America was head-to-head with Japan for world supremacy. Both markets were roughly $10 billion in size, give or take, and each had their specialty. In Japan, it was chip substrates, while the US dominated in high-layer-count boards.

I recall, at a meeting with Jack Fisher and a few others early that summer, word was out that Hadco’s lead times were as much as six months, and the industry forecast was for double-digit growth for the next two years. Hearing that, Jack surmised that no investments in HDI would be forthcoming any time soon, reasoning that if order books were maxed for “conventional” boards, owners wouldn’t see the value in investing in next-generation technology.

How right he was.

We will never know which factor had the biggest impact on the fate of the North American board industry since. Certainly, extended lead times pushed OEMs to consider Taiwanese and Chinese sources that, up to then, were looked upon more as fallbacks than primary producers for markets outside of PCs and some handhelds. The tech bubble decimated many companies, and revealed tremendous operating and management flaws among several US and Canadian fabricators. Wall Street’s push for OEMs to have a “China” solution (read: lower wages) didn’t help. And, of course, the lack of investment in HDI paved the way for better-financed Asian plants to take the lead.

If there’s a bright spot, it’s that 1) US engineers continue to amaze in their ability to get decent results from 20-year-old Excellon drills and 2) wage pressures are hitting China considerably sooner than probably anyone imagined. Bare board process equipment continues to improve, making it easier to fab boards without years of experience. Thus, the bar is slowly being lowered for future generations to jump in.

Maybe even the US.