MIL/Aero Backplanes - SFF vs. OpenVPX
- Monday, 04 March 2013
Today we are pleased to have a guest blog on military backplane technology from Justin Moll, vice president of U.S. market development for Pixus Technologies.
3U OpenVPX is the 800 lb gorilla in all types of heavy signal processing Mil/Aero applications for SIGINT, C4ISR applications, etc. that are deployed in mobile (air, land vehicle, sea) systems. Some benign environment ground-based systems in a communications station or command center will also use OpenVPX (often 6U), but increasingly they will use less ruggedized open architectures such as AdvancedTCA. Many other applications utilize CompactPCI, VME64x, or MicroTCA. As a supplier of these advanced backplane systems, we are always thinking about what technology might displace these highly successful technologies (and what new concepts are going to confuse an already fragmented market niche even more?). But, let’s focus on the smaller size technologies.
3U OpenVPX is such an ideal technology for compact applications for RADAR, SONAR, signal intelligence (SIGINT), weapons tracking, measurement, etc., that it will be hard to displace. Its open architecture, proven technology, VME compatibility, Eurocard rugged form factor and connectors, compact size, wealth of products and options, (phew! deep breath....), high IO, and high speed design are unmatched. However, there are some applications where there is very limited space and even 3U OpenVPX doesn't fit well. Enter Small Form Factor (SFF).
We reviewed participating in any of the 3 leading SFF initiatives in VITA (VMEbus International Trade Association): VITA 73, 74, and 75. The small size is attractive for mobile robotics (such as mobile IED robots), very small mobile surveillance systems, etc. The concepts are very interesting, but frankly it was frustrating that individual companies were taking their stakes in different specifications instead of working together on one common standard. There is already enough confusion about VME vs. VXS vs. VPX. Then VPX became OpenVPX. Huh?? Now with SFF, you have three VITA-based open specifications that are significantly different. Talk about a niche of a niche of a niche. No thanks!
At DesignCon a few weeks ago, I reviewed the connectors proposed in the VITA 74 specification. The connector offers some features similar in scope to OpenVPX (using the Multi-Gig RT series connectors) including the speed, pin count, differential or signal ended signaling options. Plus, the compact size and light-weight they allow for modules are great advantages. But when I inquired about test data, particularly to MIL-STD-810F for shock and vibration, I was referred to other (non MIL-spec) test reports that had different parameters.
In general the SFF architectures are a mixed bag. Some are proven in the field and some are not. None of the SFF initiatives have the wealth of products and options, compatibility, and are as proven and trusted as OpenVPX. If only one or a couple of companies are truly leading the SFF initiative, then where is the "openness" of the specification? Overall, we'll stick with OpenVPX. I can see SFF being used in smaller, mobile applications, but as for taking over VPX's realm of the process-heavy applications, I see for the near future SFF being selected mainly in cases where the size and space requirements absolutely demand it.