Careful, hot wireless
Savannah River National Laboratory goes over technicalities for wireless at a nuclear facility
Storing nuclear material gets hot at government sites where security looms big.
Environmental and physical issues can complicate processes with interference.
Savannah River National Lab experts detail wireless regulations, processes.
Sensor requirements, technical hurdles clarified in wireless approval plans.
By Davis Shull and Joseph Cordaro
Introducing wireless technology into a government site where nuclear material is processed and stored brings new meaning to the term harsh environment. At Savannah River National Laboratory, we are attempting to address not only the harsh radio frequency and harsh physical environment common to industrial facilities, but also the harsh regulatory environment the nature of our site business necessitates.
The Department of Energy is responsible for the stewardship of the U.S.'s nuclear weapons program, from maintaining a strong nuclear deterrent to environmental cleanup and decommissioning of legacy nuclear materials and facilities in the Nuclear Weapons Complex (NWC). The National Nuclear Security Agency, which is part of the Department of Energy (DOE), manages the U.S. nuclear stockpile. The agency's nuclear weapons complex consists of eight major facilities across the country.
Like other industrial environments, NWC facilities present environmental and physical impediments to using wireless technologies. Most physical structures in the plant processing areas are constructed of thick concrete, rebar, and stainless steel components. Quite a few contain large stainless-steel tanks, glove-box containment systems, piping, valves, pumps, motors, fans, and ventilation ductwork. All of these types of structural materials and process equipment tend to create highly reverberant RF environments rich with opportunities for multi- path interference.
Wireless devices would also be subjected to many sources of electromagnetic interference (EMI) in the NWC facilities. In addition to the interfering electromagnetic emissions of various motor and motor-drive systems, different types of welding processes also commonly see use in plant areas. Process controllers, computing equipment, and handheld communications radios are potential sources of EMI that could hinder successful wireless technology deployments. Besides the harsh RF environment, potential RF devices must co-exist with other legacy wired process instruments and equipment, as well as any other previously installed RF equipment.
Some process areas contain sensitive instrumentation, such as airborne radioactive contamination monitors, nuclear criticality monitors, and personnel contamination monitors. Other security, communications, and process controls equipment must also avoid harm from proposed RF technology equipment.
The physical environment of the NWC processing facilities could present challenges of extreme temperatures, humidity, varying levels of pressure and vacuum, and corrosive vapors from chemicals in use and in storage. Some applications would also subject the wireless device to vibration and dynamic shock. Perhaps the most challenging environmental factor influencing deployment of wireless sensor technologies in the NWC, however, is the ionizing radiation level present in processing facilities.
It is especially important to consider electronics and materials that will be subjected to large cumulative doses of gamma radiation that can easily reach 106 rad (the lethal acute dose for humans is approximately 600 rad). Implementation options for RF technology in these radiological environments consist of the use of rad-hardened devices able to withstand larger doses of radiation before failure, incorporating appropriate radiation shielding to protect the sensitive components without rendering the RF communications ineffective, locating the sensitive parts of the design as far away from the higher dose areas as possible, or most likely, some combination of all of the above.
Because of the nature of NWC site work and the corresponding security emphasis, wireless devices are subject to restrictions regarding their operation and location. Proposed wireless deployments must pass a lengthy approval process and possess the appropriate level of documentation prior to installation in a facility. Early communication and involvement with the reviewing and approving parties will help avoid delays and ensure a successful deployment. As the number of successful wireless deployments across the NWC increases, we hope new projects can build from previous successes by using similar approved plans and documentation as templates and by taking advantage of lessons learned from these previous deployments.
A wireless technology approval process involves some or all of the following depending on the details and components of the wireless system. Each NWC site's process may vary slightly depending on the interpretation of requirements, site implementing procedures, and the level of acceptable risk as determined by that site's designated approval authority.
Spectrum supportability authorization: The local frequency coordinator must perform an assessment to determine whether the electromagnetic spectrum is available or will be available over the expected life cycle of the RF system. For federal government use, the National Telecommunications and Information Administration (NTIA) regulates spectrum management. The NTIA regulations, DOE requirements, and contractor's procedures at specific NWC sites will direct the use of spectrum-dependent equipment.
Procurement authorization: An additional approval by the local frequency coordinator is required for all purchases of RF transmitting devices. Additional procurement regulations may give preference to specific suppliers or require additional justification for purchasing from other suppliers.
Risk assessment: Prepare this for all wireless technology deployments to evaluate the risks to computing assets. It will identify the risks, analyze the risks for impact, develop mitigation and handling strategies, and determine any remaining risk after mitigation. You may need to layer mitigation strategies to achieve an acceptable level of residual risk.
Security plan: You will need to address specific restrictions in location and operation of wireless technologies at each NWC site. The security requirements and necessary controls will vary depending on the type and level of the security area and the sensitivity of the data the system processes or transmits. Controls will most likely be a layered combination of physical, technical, personnel, and administrative controls. Security approvals can involve many different site contractor organizations, disciplines, and potentially multiple DOE organizations at each NWC site, such as National Nuclear Security Agency and Environmental Management. Early involvement with these parties is the key to a successful wireless technology deployment. Approvals may also elevate to DOE Head-quarters depending on the specific aspects of the deployment.
Test plan: You must perform testing before introducing sensitive process measurements or data to the system to verify you have met all design goals and to requirements of the security plan.
The majority of radioactive material processing in the facilities of the NWC occurs in containment glove boxes or radioactively hot work cells. Glove boxes can be single standalone enclosures with one or two glove port working locations, or can extend to multi-room enclosures that have numerous glove-port working locations, possibly extending vertically to multiple working heights or platforms. These glove boxes are generally constructed of stainless steel with single or multilayered glass windows and are commonly maintained at some level of negative pressure to ensure containment. Electrical cables and mechanical piping must enter and exit these glove boxes at built-in penetrations that maintain the glove-box confinement. Similarly, hot work cells may have specially designed cable conduit penetrations that maintain cell confinement and provide radiation shielding by incorporating various turns and bends in the conduit path.
The need for more glove-box penetrations or work-cell cable conductors is a common problem. Upgrades and changes to existing facilities in the NWC to meet emerging needs or future missions and creating new sensor signal capacities can be especially costly as containment systems are usually seismically qualified designs and provide key safety protection for workers. Simply adding a few more sensors inside a containment environment can result in the need for additional cables, conduits or cable trays, containment penetrations, cable connectors, termination points, and termination cabinets in already crowded facilities.
Material costs for these modifications are also significant since cables and connectors are specialty-type items designed for the nuclear processing environment. To make these changes, workers perform in radioactively contaminated areas at great expense and increased hazardous exposure. Impacts to installation costs also come from additional quality control inspections, work planning and control, and project approvals required by modifications to these types of systems and structures, especially when you require a breach of containment.
The use of short-range wireless sensor networks in these glove-box and hot-work-cell environments has the potential to save millions of dollars in design and installation costs by providing an alternative means for signal transmission out of the containment environment. This would also reduce material costs as you eliminate thousands of feet of cable, penetrations, connectors, and conduit. Simplifying these facility modifications and minimizing the physical modifications to the containment systems can reduce worker radiological and non-radiological industrial hazardous exposure. Using wireless devices will reduce the amount of contaminated waste generated during installation as well as at the end of system life during dismantling and disposal. You can realize further longer-term cost savings by reducing maintenance associated with cable repair and connection rework during the lifetime of the system.
Likewise, in the case of new facilities, a designer faces the difficult task of managing signal cable, where space is at a premium and requirements for additional cables seem to be a daily occurrence. The use of wireless sensor networks in new facility designs could also reduce design, installation, and material costs, while helping conserve valuable facility real-estate.
Besides the transmission of signals through containment system boundaries, other applications involve sensor installations on component test equipment, such as thermally controlled vibration tables, shock machines, and centrifuges where sealed rooms and test chambers, as well as moving parts, make wired sensors difficult to manage.
In addition to wireless sensor applications involving physical parameter measurements, wireless technologies have the potential to prove beneficial in tracking various assets in the NWC. Radio frequency identification (RFID) is beginning to see use for tracking nuclear material storage containers, excluding prohibited personally owned electronics from security areas, and tracking accountable digital storage media. Other asset tracking applications involve tracking unique parts, components, and specialized tooling on the shop floor during a component assembly process. RFID could help in safely processing nuclear material, helping ensure users observe strict criticality mass limits when introducing material into operating rooms and processing areas. You can also improve control and safe use of chemicals with RFID to ensure proper storage, handling, and correct identification prior to use.
There are, however, some familiar technical hurdles associated with RFID applications in the NWC. Besides the familiar issues related to the performance of RFID on metal objects or containers, liquid-filled containers, and on or near personnel, one challenge is the RFID tag's susceptibility to ionizing radiation. In many nuclear material storage applications, the expected cumulative lifetime dose could exceed 106 rad of gamma radiation. You need radiation-hardened devices to ensure low-maintenance operation over the expected storage lifetime of the material containers.
Some tags with promise in this area are passive devices using radio frequency transmission with surface acoustic wave identification encoding. These devices are also suited for operation at temperatures of several hundreds of degrees Celsius, possibly up to 1000°C, and can survive radiation doses up to 5 x 106 rad. Drawbacks to this type of tag include limited identification bits, read-only, and difficulties in resolving multiple simultaneous tag responses. Some single-use, disposable components used in the pharmaceutical, bioprocess/biomedical, food and beverage, and medical industries have driven the development of a read/write RFID tag that can be subjected to sterilization process doses of gamma radiation up to 4.5 x 106 rad. Manufacturers need further development work, however, to provide additional RFID options for the nuclear industry environment.
Wireless sensor requirements
NWC sites see challenges associated with harsh RF, physical, and regulatory environments, and each deployment will vary in complexity and adversity of conditions. So it is important to consider the wireless sensor network applications as well as their requirements and potential attributes.
Robustness: Multi-path rejection capabilities, self-healing mesh network, spread spectrum modulation techniques, such as direct sequence spread spectrum (DSSS), frequency hopping spread spectrum (FHSS), hybrid spread spectrum (HSS), and ultra-wideband (UWB).
Latency: Response times of less than 1 second.
Security: Encryption built-in to the physical layer, minimum required power level to avoid detection, directional antennas where appropriate, and spread spectrum modulation techniques.
Adherence to a national standard: Non-proprietary hardware and multiple vendors.
Cost: Low cost is not a major requirement, considering the cost of cable and installation.
Two developing spread spectrum modulation techniques that might have advantages for NWC applications are HSS and UWB. HSS combines advantages of DSSS and FHSS. UWB uses low-power, short-duration transmissions to achieve high data rates at short distances—500 Mbps at distances of 10 feet.
ABOUT THE AUTHORS
Davis Shull is a principal engineer at the Washington Savannah River Co. in Aiken, S.C. (firstname.lastname@example.org). Joseph Cordaro is advisory engineer at the Washington Savannah River Co. (email@example.com).
Savannah River case study
Savannah River National Laboratory is currently developing a wireless sensor network to use in a radioactive work cell. Cable penetrations into the work cell are at capacity, pushing the cost of implementing a wired solution far above that of a wireless network. The desired deployment will provide a network of three temperature sensors and one oxygen level sensor. We will transmit the measurement data through the work cell viewing window and display it on a local computer display at the operator's work location. Oak Ridge National Laboratory Extreme Measurement Communications Center will provide analysis and design of the sensor network.
Researchers measured background RF inside the work cell before starting radioactive operations in the facility to determine any potential sources of electromagnetic interference and to assess the frequency transmission characteristics of the cell window. They defined the technical requirements for the sensor network and began conceptual design work. Initial discussions with the reviewing and approving organizations are also underway to help identify issues that might influence the conceptual design decisions.
Design challenges for this sensor network include gamma radiation fields on the order of 106 rad cumulative lifetime dose and the highly RF reflective environment present inside the work cell.
The wireless pace is picking up, and users are now beginning to realize they do not have to experiment with the technology any more. "They can start planning a strategy to move toward wireless more aggressively," said Hesh Kagan, director or wireless programs at Invensys Process Systems in Foxboro, Mass. Users are looking for ways to solve control problems. They need wireless I/O connectivity to PLC or islands of automation someplace. "It's all control related initially," he said. "But as soon as we get into a dialog, it expands to other parts of the enterprise looking for wireless solutions, such as safety, security, and maintenance, in addition to operations and IT."
That is part of the story behind two staffs integrating into one at the Lost Pines Power Park in Bastrop, Tex. The lack of integrated communications became a roadblock for optimizing efficiency and enhancing the staff productivity. The Sim Gideon power plant needed to replace an obsolete GAI-tronics public address system, and its neighboring Lost Pines power plant needed to add a paging system. Safety concerns emerged when workers could not broadcast emergency events across the entire facility. Between the two dilemmas, officials at the two plants decided wireless was the way to go.
The plants implemented a Wi-Fi and Wi-Max cloud over the entire site, "which allowed them to integrate communications through VoIP. Interoperability was the key in minimizing risks with employing a leading-edge technology," said David Runkle, production manager at the Lost Pines Power Park. After installing the wireless networks the plants implemented condition monitoring technologies and video capabilities in wireless formats. LCRA has started to integrate the wireless strategy into the conceptual design of a remote, unmanned peaking site they're constructing 24 miles from the Lost Pines Power Park. The wireless infrastructure will expand through WiMax long shot technologies and fiber connections to extend the VoIP and the other technologies between the two sites. The proposed site in Winchester Texas (Winchester Power Park WPP) will be remotely operated from the Sim Gideon site.
Cost savings through reduced wiring labor costs was the convincing factor behind going wireless.
Once installed, the infrastructure allowed implementation of future applications at an incremental cost, leveraging the wireless cloud, allowing for a continued return on the initial investment. "We continue to benefit in cost savings now and for many years to come," Runkle said. "This also provides a means to finally implement applications in single or low priority equipment that would have been cost prohibitive in the wired world."
The best way to integrate wireless, Runkle said, is to create an entire infrastructure that is managed and vendor-agnostic, and that provides the needed security of wireless and allows for a controlled roadmap for wireless implementations going forward. "Most wireless implementations from my experience are implemented in an ad-hoc manner that produces a series of different islands of standalone systems that are difficult to maintain and secure," he said. Since they added the wireless capability, they can scope all new projects with wireless applications where applicable. "To date, the plants have implemented condition monitoring, video for security and equipment monitoring, and communications. "Just last week, we reached an agreement with our IT group on how to implement security to satisfy the NERC requirements in bringing information into our DCS systems for alarming, level indication, and equipment monitoring alerts," Runkle said. "This was a major achievement for our company in moving forward integrating wireless into our control systems as a strategy."
Kagan's seeing wireless emerging in all industries equally because applications are in incremental measurements. "Everyone is interested in the mobile operators, such as handhelds that free someone from a control room or allow maintenance guy to get to data," he said.
One of the main challenges of wireless is it is still considered a new technology, "and this is a conservative industry," Kagan said. "There are still few shining examples of success. People are waiting to see critical mass of proof. But you know it's coming."