 Momentum building in telecom
for boundary scan
By Alan Sguigna
Vice President of Sales and Marketing
ASSET InterTech
Somewhere in the global telecommunications industry there’s a technician who is holding his breath as he swaps out a line card in a wireless base station in a remote and hard-to-access location. He’s hoping he won’t have to come back for a long, long time. His boss and his boss’s boss harbor the same hopes.
That tech has been there before. He’s brought faulty line cards back to the repair depot and when they’ve been tested they were classified as “no fault found” (NFF). That starts the cycle of re-certifying the board, re-stocking it in an inventory of refurbished cards and eventually deploying it again into the field. That’s when the tech and his bosses hold their collective breath once again.
Fortunately, boundary scan can often break the NFF cycle. And even more importantly, NFF is just one of several areas where JTAG test and in-system programming are making a big difference in telecom. For example, system-level boundary scan, designed into systems from the ground up, is being included as an integral part of many element management systems for telecom networks. For service providers, the return-on-investment is huge. Once in place, system-level boundary scan can be deployed in a number of applications, like remote system management, remote diagnostics, remote software/firmware upgrades and reconfigurations, and as a secondary maintenance bus.
That Hidden Nemesis
One of the challenges of high-availability electronic systems like those in the telecom network, as well as other applications in other industries, is the problem of intermittent failures. This type of failure is caused by latent faults like a void of five percent in a solder joint or a conductive whisker of metal extending partway across adjacent traces. Under most operating conditions a latent fault will not cause a problem, but when the environmental conditions change and become extreme, latent faults often act up, causing outright failures or simply degrading the system’s performance.
Further complicating the problem, removing a board with a latent fault from the system in the field and subsequently testing it in a lab where conditions are usually in the middle of the board’s operating range often does not uncover the latent fault. As it did in the field, the board performs the way it should in the lab under conditions that are not extreme. Latent faults like these are often the cause of the NFF problem. Some equipment manufacturers have found that up to 20 percent of their boards can have NFF problems.
The ramifications of NFF can be quite costly. Boards must be re-certified and re-stocked each time they are returned from the field. Then, manufacturers must decide how long they will take a chance on re-deploying an NFF board. Some have policies that scrap an NFF board when it has returned from the field for the third time, for example.
Boundary scan testing can help uncover latent faults in several ways. If system-level JTAG has been embedded in the system, remote testing and diagnostics can be applied in the field while the system is experiencing the environmental conditions that may have brought on the fault condition. Without system-level JTAG, the faulty board can be returned to a lab where HALT/HASS tests in an environmental chamber can simulate conditions in the field. While in the environmental chamber and subjected to extreme conditions, the board can be connected to ScanWorks and test vectors applied to isolate any latent faults. Unlike functional tests which reveal that a board or assembly is not functioning to specification, JTAG tests can diagnose the structural cause of a problem down to an individual pin on a device.
System-Level JTAG
With advance planning to design-in the needed infrastructure, boundary-scan operations can transcend the board or chip levels and encompass system-level elements including backplanes and multi-board assemblies. Once in place, JTAG can assume the role of a secondary maintenance bus, performing remote or local structural tests and diagnostics, remotely reconfiguring complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs) and serial programmable read-only memories (PROMs) while the system is running, and tracking firmware versions, equipment serial numbers and other board characteristics.
Several telecom equipment suppliers, including Lucent, Motorola, Ericsson, Tellabs and others have adopted facets of system-level JTAG operations. With system-level JTAG in place, much of the system’s maintenance and troubleshooting can be performed remotely. In many cases, problems can be solved without dispatching a technician to the site, but sometimes sending a technician is unavoidable. When this is the case, system-level JTAG can identify the cause of the fault or performance degradation before the technician is dispatched. Technician time is kept to a minimum by eliminating on-site diagnostics and trial-and-error troubleshooting, and providing the basis for a solution that the technician can implement immediately.
 Embedded system-level JTAG can also be integrated into a life-cycle management process. From the time a system is assembled through its installation in the field, boundary-scan operations can generate information on structural faults and failures. Suppliers such as SigmaQuest have products that can gather this information into a database where it can be analyzed. Recurring faults or failures might alert manufacturers to quality issues in assembly techniques and materials, or in the components.
Another significant cost-savings feature of system-level JTAG involves remote in-system programming of logic devices and program memories. If a firmware upgrade must be done manually by technicians traveling to each installed system, the cost can be astronomical. Instead, systems in the field with system-level JTAG can be upgraded remotely from central offices. The cost savings from entirely eliminating the involvement of support technicians would be significant.
To implement system-level JTAG effectively, the system must be developed with this in mind from the beginning of design. Fortunately, ASSET’s DFT Analyzer makes it easier for developers to consider boundary scan design-for-test issues while the system is being developed. With DFT Analyzer designers don’t have to be experts in boundary scan to design-in the basic infrastructure. For more information on DFT Analyzer, click here.
In addition, for many years ASSET has worked closely with a number of telecom companies on system-level JTAG implementations. This has resulted in a significant body of knowledge in this area. For some tips from one of our experienced applications engineers, click here.
There’s also a white paper available in the Press Room on the ASSET web site. Titled “System Test with Boundary Scan (JTAG)”, the white paper was written by our technical product manager, Dave Bonnett.
The rise of the Internet and the proliferation of wireless communications have changed the telecommunications industry radically in recent years. Consumers now have more telecom choices than ever before. And competition for subscribers among service providers has reached a feverish pitch. At the same time, boundary scan technology has earned a central role in telecom as equipment suppliers strive to produce systems with the features and functionality that will support the service providers’ long-term goals.
|
