feature article
Subscribe Now

Cooler, Lighter, Faster

Actel Debuts Military-grade Flash FPGAs

We’ve all heard that time is money.  Also, given today’s energy market, we are painfully aware that power is money.  By the transitive property, we can therefore infer that power is time.  And, if we’re designing electronics for military and aerospace applications, it will come as no surprise that power is weight.  Why?  Because if your device consumes too much power, you have to start adding things like heat sinks and fans and bigger power supplies – all of which add weight and cost.  Adding fans also creates a new potential point-of-failure*, so really – power is weight is reliability is time is money.

This week, Actel decided to give military and aerospace designers a break on the front end of that process by announcing military-grade versions of three of their popular and very low-power flash-based FPGAs.  The new lines are dubbed Military ProASIC3/EL, and they include extended-temperature range versions of Actel’s commercial-lineflash-based FPGAs.  The new family includes the A3PE600L with 600K system gates, the A3P1000 with 1M system gates, and the A3PE3000L with 3M system gates. These devices bring some completely new options to power-, weight-, and safety-critical systems like avionics.

FPGAs have been a godsend to low-volume, high-requirement military systems.  Before viable FPGAs came along, mil-aero projects bore the unbelievable burden of designing custom ASICs in order to get the required performance, power consumption, weight, and integration level – and amortizing the substantial NRE costs across a very small number of production units.  With high-capacity, high-performance FPGAs, many of those projects could drop the ASICs, replacing them with much more cost-effective FPGAs.  FPGAs (sometimes) brought the additional advantage of late-cycle and in-system reprogrammability, providing the possibility of much cheaper and easier design changes to accommodate system evolution and bug fixing.

There were still limitations, of course.  FPGAs were limited in capacity and performance, brought a considerable power-consumption penalty, and usually required a few additional support chips for configuration.  Also, most conventional FPGAs are not particularly good at radiation-immunity – a critical factor even in normal-altitude airborne electronics where the probability of neutron-induced single-event-upsets (SEUs) affecting the configuration of FPGAs is a major concern.  The primary driver for military qualification, however, is temperature range.  Temperatures in airborne systems can fluctuate wildly from super-cold high-altitude ambient temperatures to super-heated conditions related to air friction from supersonic flight.  Before Actel’s recent announcement, two basic types of FPGAs had been certified for the extended military temperature ranges – SRAM- and antifuse-based devices.  SRAM brings penalties of high power consumption, low resistance to neutron-induced SEUs, and multi-chip configuration.  Antifuse carries penalties of high cost and lack of reprogrammability.  Now, flash-based FPGAs come in at a sweet spot in the solution space – with very low power consumption, large temperature ranges, reasonably high performance (~250MHz operation), reprogrammability, immunity from neutron-induced configuration upsets, single-chip operation, and density up to 3 million system gates. 

For designers of applications such as high-speed, high-resolution sensors, the new devices bring the capability to increase performance while reducing power consumption and weight.  If you follow the Joint Strike Fighter program, you’ll know that the F-35 has suffered considerable delays while engineers try to put the plane on a diet, attempting to lose as much as a ton from the empty weight of the aircraft.  Avionics make a significant contribution to that bloat, and the ability to maintain system performance while trimming size, power consumption, and weight are paramount in the program’s success.  Many of these systems cannot rely on cooling fans, so reducing the power footprint of individual components is extremely important.

Actel says the ProASIC3EL family delivers static power as low as 0.55mW and dynamic power less than 60mW at 100MHz (for the A3PE600L operating at 25C).  The devices also include Actel’s “Flash*Freeze” mode which preserves register states and drops the device to near power-off status in less than one microsecond.  The largest device includes as much as 504Kbits embedded memory and can hit speeds over 250MHz at 125C. 

Package options are limited at first, but Actel says they’ll be expanding the packaging options based on customer demand.  Initially, the A3PE600L will be available in an FG484 package, the A3P1000 is offered in either FG144 or PQ208 options, and the A3PE3000L can be delivered in FG484 and FG896 packages.  Since military parts must be separately qualified in each package option, suppliers often delay package offerings until there is proven customer demand for a particular package.

The company says the Military ProASIC3 A3P1000 is already shipping and software support is already in place.  The Military ProASIC3EL A3PE600L and A3PE3000L are following close behind with software support for the former in August and the latter in September and with datasheets and documentation already out.

If we read the tea-leaves correctly, this announcement could signal the continuation of a strategy shift at Actel.  Historically, the company made most of their money with very expensive antifuse-based FPGAs for military and space applications.  The reliability, speed, and radiation tolerance of antifuse devices made them attractive (even at very high price points) to the designers of these system because the alternative was a much more expensive and risky ASIC project. 

In recent years, however, Actel has put an increasing focus on its flash-based families, and most recently on the extremely power-efficient aspect of those devices.  With this announcement, the flash technology seems poised to jump over onto the high-rel/military/space side of the catalog, and that might signal a gradual replacement or augmentation of the company’s long-standing antifuse lines.  For applications like space (this announcement does not cover space-bound devices, as the requirements for space flight are considerably more stringent), flash-based FPGAs might possibly offer huge advantages like power efficiency, radiation immunity, and reprogrammability that are not well combined in current device offerings.  We’ll keep our eyes open for future announcements.

Leave a Reply

featured blogs
Aug 18, 2018
Once upon a time, the Santa Clara Valley was called the Valley of Heart'€™s Delight; the main industry was growing prunes; and there were orchards filled with apricot and cherry trees all over the place. Then in 1955, a future Nobel Prize winner named William Shockley moved...
Aug 17, 2018
Samtec’s growing portfolio of high-performance Silicon-to-Silicon'„¢ Applications Solutions answer the design challenges of routing 56 Gbps signals through a system. However, finding the ideal solution in a single-click probably is an obstacle. Samtec last updated the...
Aug 17, 2018
If you read my post Who Put the Silicon in Silicon Valley? then you know my conclusion: Let's go with Shockley. He invented the transistor, came here, hired a bunch of young PhDs, and sent them out (by accident, not design) to create the companies, that created the compa...
Aug 16, 2018
All of the little details were squared up when the check-plots came out for "final" review. Those same preliminary files were shared with the fab and assembly units and, of course, the vendors have c...
Jul 30, 2018
As discussed in part 1 of this blog post, each instance of an Achronix Speedcore eFPGA in your ASIC or SoC design must be configured after the system powers up because Speedcore eFPGAs employ nonvolatile SRAM technology to store its configuration bits. The time required to pr...