Embedded vetronics (vehicle electronics) subsystems for rugged deployed ground vehicles have long benefited from the reduced risk, faster time to market, and long lifecycle support offered by COTS board vendors. Recent advances in open standards developments, namely the OpenVPX/VITA 65 specification that defines system-level interoperability profiles for VPX architecture boards and backplanes, promise to take the COTS model from the board to the subsystem level, enabling faster deployment of proven, standards-based fully integrated enclosures.
The increasing functional density of today’s rugged, embedded single board computers (SBCs) and I/O boards is making it possible to integrate small form factor, cost effective solutions that can incorporate a wide variety of a ground vehicle’s vetronics functionality that previously would have required multiple, larger, boxes. Known as “LRU Consolidation,” this ability to recapture space, weight and power (SWaP) is enabling the inclusion of increased capabilities within SWaP constrained current ground force vehicles. Many of today’s vetronics subsystems can now be made smaller than the proverbial breadbox.
Advancing subsystem level interoperability is the key to taking these modern vetronics systems to the next level of lowered cost and faster time to market. Improving subsystem interoperability through flexible backplane configurations has long been a goal of COTS board and integrated systems Packaged COTS (PCOTS) vendors.
Long before the recent emergence of VPX as the higher bandwidth, more rugged heir to the VME bus architecture, board vendor and system architects strove for the flexibility that greater interoperability delivers. For example, Curtiss-Wright Controls developed variant versions of its 3U CompactPCI 124 and 1201 single board computers to enable full use of the card’s potential 64-bits of PMC I/O on non-system controller cards. Because the system controller 3U form factor card in Slot 1 is required to devote many of its limited I/O pins for system management, there would be a shortage of pins available to support the full PMC mezzanine card bandwidth. To increase flexibility and interoperability, Curtiss-Wright Controls developed PICMG 2.3 compliant variants of the 1201p and 124p SBCs. Identical with the originals in functionality, these peripheral-slot-only versions of the cards were stripped of all of the slot 1 card system management signals. This provided the needed pins to support 64-bit I/O routing from every PMC card in a system. In addition, the company developed the PICMG 2.3- compliant 201 PMC carrier card making it possible that every slot in the system, other than slot 1, could support a PMC card, and that the mezzanine card could reside on a PowerPC or x86-based SBC, or a PMC carrier card, which made the “mixing and matching” of a wide variety of cards within a PCOTS subsystem much simpler, and faster.
Customers understood right away that this approach gave them significant flexibility because they could begin a system design with one complement of cards and later, if need be, easily modify that selection. For example, if it turns out that a system doesn’t need as much processing power, a reduction in cost and maintenance issues can be realized by replacing an SBC with a PMC carrier card. Conversely, if the system needs additional processor bandwidth, another SBC can easily be added.
Open standards-based, highly interoperable PCOTS systems make taking the COTS model up from the card to the subsystem level a reality. All of the COTS value propositions that have been enjoyed for decades on the board level can now be successfully migrated up to the subsystem level. Development costs for fully integrated systems can be shared across multiple programs. Risk is also shared, as multiple users will deploy the same PCOTS system in numerous different ways so that no one customer is responsible for finding any deficiencies, weaknesses, or bugs in the system. In addition, all longevity and sustaining engineering costs are also shared. The result is a proven, fully qualified and cost effective solution. Even better, customers benefit from supply chain simplification since a fully integrated PCOTS system has a single part number instead of a laundry list of part numbers for each individual component.
In the example above, Curtiss-Wright Controls defined a slot profile for its 3U CompactPCI-based PCOTS system that then made it possible to craft boards and architect the system backplane so that no slot was dependent on a particular card. In a similar way OpenVPX is now taking the standardizing of flexible pinouts to the next level, bringing it to the rugged deployed VPX architecture.
OpenVPX provides a major breakthrough in the development of subsystem interoperability for VPX systems. OpenVPX, developed by VITA members working together outside of VITA, was introduced into the VITA Standards Organization (VSO) in October 2009 as VITA 65 for final comment, ballot, and ratification as a standard. OpenVPX brings system level interoperability to the popular new VPX bus architecture and is a key element in bringing the COTS model from the card to the subsystem level. The recently completed OpenVPX work, now under the leadership of the VITA 65 Working Group, defines the VPX Systems Specification, a scalable architecture that manages and constrains module and backplane designs. The VPX Systems Specification includes the definition of pin-outs and sets interoperability points within VPX while maintaining full compliance with the existing VPX specification.
OpenVPX will help drive the COTS proposition from the board to the system level. On the board level, flexible pin-out configurations and the use of IPM modules has enabled changes without the need to do costly board re-spins. Now, using OpenVPX interoperability profiles, it will be possible to develop truly flexible VPX-based general purpose PCOTS vetronics systems, rather than point solutions for a single customer or application. If the customer chooses to make changes to the I/O, they will be able to modify the architecture functionality of a system without having to change the backplane and the front panel I/O. As OpenVPX interoperability becomes widely embraced by customers, much of the costly and demanding re-qualification currently required if any change is made to the system may become unnecessary. Today, the re-qualification process for a system can typically take 2 to 3 months, or longer. Re-qualification requires a plan, procedures, equipment, a test laboratory, analysis of the results and, finally, a test report. The cost for all of this can often be as much as a couple hundred thousand dollars.This can often be quite costly. For many PCOTS customers, once they recognize that the care taken in the enclosure design is efficient and constitutes a proven solution, it may become possible to save the considerable cost and time of qualifying at the subsystem “box” level, and instead qualify the box at the same time as the entire system is qualified.
Another significant advantage that the VPX architecture brings to vetronics, in addition to its higher bandwidth and greater ruggedization, is its support for two-level maintenance. With OpenVPX, it’s now possible to envision flexible vetronics subsystems and enclosures, designed to an open standard, that can accommodate cards from different vendors and different pin-outs, that can be deployed throughout a ground vehicle and populated with a variety of two-level maintenance, replaceable in the field VPX-based line replaceable modules (LRMs). This will make it possible to change the personality of the subsystem, depending on what the needs are, choosing from a wide number of open standard cards. In this way, it becomes possible to deploy a same part number PCOTS enclosure integrated with a standardized selection of cards with which the system designer can achieve a very wide range of functionality for an array of ground vehicle applications, from fire control computers, to ammunition handling systems.
VPX’s support for two-level maintenance is also a great benefit for military technology insertion applications. It makes it much easier and cost-effective to upgrade fielded systems with the most advanced computing technology without having to redesign or re-qualify the subsystem. This can greatly extend the lifecycle of legacy systems. The combination of subsystem interoperability and two-level maintenance can eliminate the time, cost and engineering challenges associated with proprietary board and subsystem technology. Bringing the COTS model to the subsystem level promises to enable customers to upgrade individual subsystem slots and avoid custom development.
This article was written by Jacob Sealander, Chief Architect of Embedded Systems, Curtiss-Wright Controls Embedded Computing (San Diego, CA). For more information, contact Mr. Sealander at jsealander@ curtisswright.com, or visit http://info.hotims.com/28049-403.