It sounds great on paper. After getting slammed for buying $900 oneof- a-kind hammers and other custom-made parts, the government’s ongoing push for commercial-off-the-shelf (COTS) components is successfully slashing the cost of new military equipment. At the same time, it’s greatly speeding up time to market. But the military’s buy-in of COTS also is creating some interesting problems. They range from software security issues to compatibility problems with encryption, export-regulation red tape, and the inability to find components containing lead.

“We used to worry when a plane went down that you didn’t lose topsecret documents,” says Robert Day, vice president of marketing at LynuxWorks. “Now you have to do the same for software used in the plane. You have to make sure no one can get in.” That goal is particularly challenging because the military is now using the same global information grid as everyone else—everyone from the soldiers on the front lines to the generals directing military action to hackers sitting in their living rooms. If that isn’t difficult enough, coalition forces need to share information on the same network, explains Day.

“At the moment, a lot of systems are not compatible,” he states. “They don’t talk to each other well. We’re working on next-generation systems where there will be different levels of security.” Even with all of the compatibility problems, Day notes that the advantages of using COTS components are significant—particularly for speeding up time-to-market and time-to-battle.

With system complexity increasing, the difficulty and expense of writing software code for each new project is enormous. “The amount of software code in a military system is huge,” he says. “You don’t want to have to rewrite millions of lines of code each time. Because of that, we’re getting to a turnaround time that is more toward what you find in the commercial world. It’s less about how much you can do yourself and more about how much you can re-use.”

Encryption is another challenge because it all depends on who holds the keys and how often they are changed. Gilles Garcia, director of marketing at AMCC, says that encryption is regulated by country or by region and varies by each of those. “European regulations are different from the United States,” Gilles explains. “They are not following the same standard. Europe is using different keys and encryption mechanisms.”

Complicating matters are the export controls on COTS parts. While it’s well understood which parts can be sold where, parts are often sold through brokers or distribution and the trail gets more convoluted. “This has required us to put in place processes and rigorously check which channel is using which parts,” Garcia says. He notes that all sorts of mechanisms have been developed for tracking sensitive parts— including radio-frequency identification (RFID)—so authorities know where parts are at all times.

Unfortunately, a potential overlap in military and non-military uses of parts also can turn this into a very gray area. A bank may have security needs that are as tight as a military operation, for example. It also may have operations in countries where sales of military parts are restricted. The silver lining is that parts made for commercial customers also can be sold to military customers in a variety of countries. This greatly expands the market for makers of some components.

“There are military markets where they will only buy organically,” states Jay Swenson, director of global marketing communications at GE Fanuc. “But as with the United States, they are also buying components all over the world and their suppliers are selling all over the world. Most of these are the same big-name players, but there will be new players coming in as well. There will be some emerging companies because of the market for in-country components too.”

The Search For Lead

One such opportunity is for leaded components. After getting the lead out of electronic components in compliance with the European Restriction of Hazardous Substances (RoHS), some companies and government agencies are now looking for new suppliers of leaded parts—or at least for ways to put the lead back in. At issue are the tiny tin whiskers that can grow out of pure tin. What makes those whiskers grow and how to stop such growth is the subject of intense scientific study and debate. So far, there are no clear answers. But there is a growing body of evidence for the magnitude of the problems that they can cause. One of the most celebrated examples occurred in May 1998, when tin whiskers caused a Galaxy 4 satellite to short circuit, rendering 40 million pagers useless and interrupting ATM transactions at banks. Tin whiskers also were blamed for the unexpected shutdown of a nuclear power plant in Waterford, Conn., in 2005. In addition, they have been implicated in pacemaker failures around the globe.

Nor are whiskers just confined to tin. NASA has been compiling papers on whiskers emanating from a variety of types of metals (http://nepp. nasa.gov/WHISKER/). Growing metal whiskers is bad enough when you can replace a component or even an entire board. But for an incredibly expensive spacecraft—paid for with taxpayer dollars, launched deep into outer space, and surrounded by a massive marketing campaign—it can devolve into a particularly sensitive subject. It’s also a very touchy subject when it comes to applications like missile-guidance systems and other potentially explosive devices. With the number of lead-free parts increasing, the threat of equipment failure is causing insomnia for many military planners and administrators.

Tin whiskers have been documented to grow from components plated with pure tin. They emanate out of “hillocks” that develop under compressive stress when tin is deposited through electroplating. Why exactly the lead acts to mitigate this is unknown. Tin whiskers weren’t even on the radar screen when the U.S. government began trimming costs with COTS parts. Enter RoHS regulations. While the military and aerospace are exempt, many manufacturers of off-the-shelf parts have trouble justifying the manufacture of leaded parts for the military when the bulk of their sales are non-military. Even within the military, not all sectors are concerned about occasional failures. For some of them, however, a short circuit can be catastrophic.

For critical applications, there are still many leaded parts available—at least for now. Lisa Duckett is director of marketing at VersaLogic, which makes embedded computers for the military. She says that part of her company’s strategy has been to stock up on leaded parts. “We have a five-year guarantee on products,” states Duckett. “All products we’ve launched have an end-of-life guarantee that we will continue to provide them during that time.” She also notes that VersaLogic’s end-of-life strategy was initiated with the rollout of RoHS standards on July 1, 2006. “Eventually, like it or not, government agencies are going to have to accept it. We can stock as much as we can, but there will be a finite end. Until then, we will try to make leaded parts available.”

In cases where there are no leaded parts available, there’s a growing business for adding the lead back in. Corfin Industries has seen a big uptick in its business of adding lead back into parts. It replaces pure tin with tin-lead through a molten alloy process. The company’s process actually dissolves the pure tin and—in the same step—applies a new coating with a metallic bond.

Don Tyler, Corfin’s general manager and vice president of marketing, says the pickup in business began even before RoHS took effect because many suppliers wanted to be compliant ahead of time. That brought in a variety of agencies whose focus ranged from medical to space and nuclear power. “There is a lot of experience out there with tin whiskers,” Tyler explains. “There are fewer people that have actually seen shorts, but a lot of them have seen furry boards and are worried about them. This can mess up optical devices. Whiskers can break off and fuse things together and they can cause devices to momentarily fail.”

Tyler says that the phenomenon has been well known since the 1950s, which is why lead was added in the first place. He points out that there have been numerous attempts to solve that with more accepted metals and alloys. But such approaches often create different whiskers, weaken the connection on printed-circuit boards, or both.

The Future

According to Bryan Brady, vice president of Avnet Electronics Marketing Americas’ defense aerospace business unit, there’s still plenty of inventory left for agencies and companies looking for leaded components—for the moment. He says that the real issue is that new technologies aren’t being introduced with leaded versions. “We’re still seeing some vendors introducing components with military specifications, and some say they will continue producing it as long as there is demand,” Brady explains. “But every time I go to a conference, I ask the audience what they’re going to do. With commercial avionics, there’s less pressure to use lead. But when you get to radar and missile applications, they definitely want lead. We’ve been surprised by the lingering demand for leaded parts.”

Ed Sperling is a regular contributing editor to Chip Design magazine. Ed has spent the past two decades immersed in technology. He is the recipient of numerous awards for journalistic excellence.