[About Sandia]
[Unique Solutions]
[Working With Us]
[Contacting Us]
[News Center]
[Search]
[Home]
[navigation panel]

[Sandia Lab News]

Vol. 54, No. 1        January 11, 2002
[Sandia National Laboratories]

Albuquerque, New Mexico 87185-0165    ||   Livermore, California 94550-0969
Tonopah, Nevada; Nevada Test Site; Amarillo, Texas

Back to the Lab News home page

Water resource simulator calculates water supply Modernized system to manage weapon use-control codes Sandia develops its first custom processor World's smallest microchain fabricated


Labs water resource simulator rapidly calculates tomorrow's supply given today's choicess

Back to topBack to Lab News home page.

By John German

By 2015, according to experts, half the world's population -- 3 billion people -- will lack access to fresh water. (See "Why water is a national security issue," on page 4. )

A team of Sandians is developing software models they think might help not only regions and nations with seemingly hopeless water shortages, but also water-wary areas such as the Southwestern US where sound resource management might still avert a crisis.

The simulations, called Dynamic Water Budget Models, allow decision makers to see how water policy options selected today will affect a society's water resources decades into the future.

The developers include Dick Thomas (6115), Steve Conrad (6515), Vince Tidwell (6115), Erik Webb (6115), and Cara McCarthy (University of Arizona).

Exploring policy options

The models are built on the commercial Powersim software tool, which Sandia has used to study everything from summer blackouts in California to global nuclear material inventories. The intuitive user interface allows easy changes to inputs and immediate extrapolation and visualization of results.

Making policy changes is as simple as fiddling with a few knobs, says Dick.

But underneath it all is a complex model not only of water uses but of the subtle interrelationships among ground and surface water sources, recharge rates, groundwater pumping, irrigation, climate, evapotranspiration, and demographics. Future models will include other factors, such as environmental impacts, water quality, economic productivity, and an area's social and cultural foundations.

"There are modeling tools that provide greater fidelity in modeling individual components of the water system," says Erik. "We've abstracted and combined those kinds of models to ask what the water resource picture might be 20 years down the road. This is the only model we've found that allows for big-picture, long-term planning."

Thinking globally, acting locally

Development of the tool, first envisioned by Dennis Engi (16000) in 1986, has been funded primarily through the Laboratory Directed Research and Development (LDRD) program.

Dick and Steve built the first model in the mid 1990s to examine water supply and demand trends for China's 10 major water basins, concluding that water will become a limiting factor in the country's ability to feed itself during the next two decades as China's major agricultural areas run increasingly large water deficits.

The simulations were part of a Harvard University study that helped alter the way some experts now think about China's future.

"China's goal is to be a self-sufficient nation with regard to grain production," says Steve. "Our conclusion was that China is not going to be self- sufficient unless something changes drastically."

The team then used the Middle Rio Grande Basin, the basin that supplies water to the Albuquerque area, as a testbed for developing the tool further.

"It's a way of helping our community with sustainability issues while also creating a tool that could help the nation and the world," says Dick.

Working with the US Geological Survey and the state engineer's office they built a model that shows, not surprisingly, that current water use practices in New Mexico are not sustainable.

The team continues to work with city planners in Albuquerque and Rio Rancho to apply the tool to Albuquerque-area policy-making efforts, in cooperation with the UNM Law School's Utton International Transboundary Resource Center.

Workable vs. unworkable options

A similar model of New Mexico's Estancia Basin, in cooperation with the Estancia Basin Water Planning Committee, is helping show farmers and developers the possible results of various development schemes and agricultural practices for the area.

"Different users have different ideas about what optimal use of the water resource is," says Dick. "We helped get them talking sooner about realistic approaches rather than dwelling on unworkable, unsustainable options."

The team expects to deliver to officials a web-based version of the Estancia Basin model in four to six months as part of a project sponsored internally by Corporate Outreach Dept. 12650.

They also are exploring, in cooperation with Sandia's Cooperative Monitoring Center, the possibility of modeling water issues for basins shared by countries of the Former Soviet Union, for nine countries that border the Nile River, and for the US and Mexico in the El Paso/Cuidad Juarez area.

The team has demonstrated the model to local school children as well.

"It's a great educational tool," says Steve. "Anybody can play their own 'what if' game. It allows different people with different stakes in the outcome to rapidly test the long-term effects of many policy options. It's very democratizing." -- John German

Back to topBack to Lab News home page.

Modernized system to manage codes for nation's nuclear weapons complete

Back to topBack to Lab News home page.

By Ken Frazier

An ambitious Sandia-led program to fully update the code management system that supports control over use of the nation's nuclear weapons has completed a major milestone.

The Code Management Systems (CMS) project completed a multiyear, full-scale engineering effort at Sandia with its first full system delivery of all hardware and software to two Department of Defense customers. On Nov. 30 the system became operational for the first time for weapons in Europe.

Code management systems and ancillary equipment are in place at headquarters command sites and at various bases in the field. They are used in conjunction with Sandia-designed permissive action links, or PALs, inside US nuclear weapons to recode, unlock, lock, and manage the weapons, while ensuring the secrecy and authenticity of command-and-control messages.

The systems allow those having custody of weapons with PAL systems to plan, store, change, interrogate, track, use, or otherwise manage all necessary code-related information. This is a critical part of ensuring that weapons can be used when authorized and cannot be used when not properly authorized.

The Code Management System coupled with the B61 ALT 339 retrofit enables the recoding of nuclear weapons in a fully encrypted manner. This new class of code management equipment designed by Sandia greatly simplifies use and logistics for personnel. It replaces a variety of different vintages of code-management equipment that had been produced and put into place at different times and for different weapon systems and users over the decades.

The new hardware and software has been redesigned from scratch in a systems approach intended to provide a common architecture, modular products, and reusable processes to facilitate future upgrades. Now that it is operational with USEUCOM (US European Command) and USAFE (US Air Forces in Europe), the CMS becomes the common foundation for all future upgrades of PAL system hardware and software.

A huge team effort

"To design and develop a system with an overall architecture to replace everything in the field was quite a challenge," says Doug Clark, CMS project lead engineer in Use Control Systems Dept. 2121. "We wanted to develop a system that was modular in nature, so that it could be maintained and upgraded in pieces as needed in the future. It was a huge team effort."

Fourteen custom products (nine software and five hardware products) were delivered, accepted by DOE/NNSA, and put into operation in Europe in November. All were designed at Sandia, and in addition all the software was implemented and produced at Sandia. The

custom hardware was manufactured at NNSA's Kansas City Plant.

The software contains about 160,000 lines of uncommented computer source code (260,000 including comments). About 570 documents and drawings were prepared in support of the requirements, development, production, and qualification of all CMS products.

The project started at Sandia in 1995 at a low level of effort, but became focused on its current development strategy in 1997. The use-control community realized that code management and PAL system equipment for recoding and managing nuclear weapons developed over 30 to 40 years was becoming difficult to maintain and depended upon a frustrating variety of different vintages of equipment.

"This culminates eight years of work in both code management and the B61 program that had to come together at the end of November," says Doug Mangum, Manager of Dept. 2121. "It all came together and worked as expected," he says.

The 14 products that make up the new CMS include a cryptographic processor (which was completed and delivered in Europe first, in 1997, to address some Y2K concerns), its software, host processor software, field processor software, file transfer software, a communication module, that module's software, a power module, an interface adapter, a field tester, a field training simulator, field training material, headquarters training material, and system operation definitions.

A kit the size of a small suitcase

The field hardware all fits in a kit the size of a small suitcase. Software and hardware products to support fifty of the kits have been delivered from Sandia and the Kansas City Plant.

The cryptographic processor, which contains three cryptographic chips and can support any current nuclear command and control cryptographic system, looks something like a large, sturdy, all-metal laptop computer and is, says Doug Mangum, "secure and trusted." He says the CMS system will support the use-control community's "end-to-end encryption requirement for crypto-capable weapons," in which PAL data are never exposed in the process of doing PAL recodes.

The Sandia CMS team will next focus on replacing the remaining code management equipment for all US military and NNSA users by the end of 2003 or early 2004.

At that point all the main objectives of the project will be complete. The nation will have new, modernized use-control code management capabilities and equipment providing greater flexibility and speed. The capabilities will incorporate advanced principles of nuclear surety (integrated anti-tamper features, enhanced use of encryption and no-knowledge systems, and improved equipment safety). And maintenance and logistic burdens will be eased, with personnel training and operation simplified. -- Ken Frazier

Back to topBack to Lab News home page.

Labs develops its first custom processor, the Sandia Secure Processor

Back to topBack to Lab News home page.

By Ken Frazier

The same department that developed the hardware and software for the Code Management System (see main story) has developed Sandia's first custom microprocessor architecture. The Sandia Secure Processor, or SSP, will have its first application in weapon use control, as a PAL-system component. But the architecture is generic, in the expectation that it will have multiple applications.

"It is important to note that this system is not designed for blazing speed as most commercial processors are," says Greg Wickstrom of Surety Electronics and Software Dept. 2125, which has project responsibility. "It is targeted for small, embedded, safety-critical systems. Any safety-critical system may find it useful." Satellites and medical devices are among the possibilities.

Although some of the newer use-control systems use modern processors, they are still constrained to older languages that make developing safe and secure software difficult. The Sandia Secure Processor is based on a subset of the newer Java language, which has inherent advantages for safety and security.

"Basically, we wanted to select a language that protects programmers from themselves," says Greg -- one that helps identify and correct errors. The Java language fits the bill. "We're leveraging its safety and security capabilities to the maximum extent," he says.

"Java usually requires a software virtual machine to execute, and we are turning that virtual machine into a physical one," says Department 2125 Manager Tom Perea. "This lets Java software run much more efficiently in very small systems."

The processor is not only secure but fully verifiable. "Since Sandia owns this design, we can analyze it to any level of detail," says Tom. In fact, a goal of this program is to enable the application of mathematically provable methods for verifying flawless, secure operation.

"We are now integrating all the parts of the processor into one coherent piece and testing the whole system in simulations," says Tom. The tools that synthesize the design into a producible part are currently generating a mere 40,000 logic elements capable of running at 50 MHz.

He notes that while major microelectronics chip companies might require large teams of developers to design a new processor, the SSP has been developed with the equivalent of three full-time employees over the last three years. This has been enabled through the use of modern development tools and close teaming between the three departments contributing to its development, 2125, 1735, and 2121.

Plans are to fabricate the SSP in a radiation-hardened technology at Sandia's Microelectronics Development Laboratory. . -- Ken Frazier

Back to topBack to Lab News home page.

World's smallest microchain drive fabricated at Sandia

Back to topBack to Lab News home page.

By Neal Singer

Except that each link and pivot could rest comfortably atop a human hair, a microchain that closely resembles a bicycle chain has been fabricated at Sandia.

(The distance between chain link centers is 50 microns. The diameter of a human hair is approximately 70 microns.)

Because a single microchain can rotate many drive shafts, the device could make it unnecessary to place many tiny microelectromechanical systems (MEMS) motors in close proximity. Usually, a separate driver powers each MEMS device.

"All those drives take up a lot of real estate on chips," says Sandia technician Ed Vernon (2614), who has received a patent for the silicon microchain.

The microchain also makes it possible to drive a MEMS device from a motor situated at a distance, again saving considerable space on the MEMS-bearing chip.

The microchain, says Ed, could be used to power microcamera shutters, as larger chains currently do in the macroworld. It could also be used in mechanical timing and decoding.

The 50-link silicon microchain is designed to transmit power somewhat like the drive belt in a 19th-century sewing factory. There, a central engine shaft powered by steam turned drive belts to power distant work stations -- for example, sewing machines -- before the dawn of the age of electricity.

Chain systems, unlike stroke systems, do not require back-and-forth movements but instead allow for both continuous and intermittent drive translation.

Ed fabricated a microchain rather than a microbelt because though silicon belts are tough and flexible, they are spring-like and produce too much torque on gears not aligned in a straight line. Each chain link, on the other hand, is capable of plus-or-minus 52 degrees rotation with respect to the preceding link, without creating pressure on the support structure. The wide angle means designers can be relatively unconstrained in positioning multiple devices.

The longest span unsupported by gears or bracing is 500 microns. A microchain tensioner is needed to accommodate longer spans.

The multilevel surface-micromachined silicon device was constructed with the aid of Sandia's patented Summit IV and Summit V technology, which enables construction of complicated MEMS devices.

-- Neal Singer

Back to topBack to Lab News home page.

Back to topBack to Lab News home page.

Last modified: january 11, 2002


Back to the Lab News home page


Browse current and past Lab News articles

View Sandia news releases and fact sheets


Back to top of page

Questions and Comments || Acknowledgment and Disclaimer