An interview with Ken LaBel at NASA's NEPP program: Rocket science applied to componentsStory
February 10, 2009
Editor's Note: Group editorial director Chris Ciufo caught up with Ken LaBel, NEPP co-manager, at last year's Components for Military and Space Electronics (CMSE) trade show and again at the recent FPGA Summit to get the 411 on NEPP's multifaceted mission: to analyze and test nascent (even commercial) technologies for radiation tolerance; develop qualification insurance guidance; and to make its testing results public (believe it or not).
MIL EMBEDDED: To start, can you provide an overview of the NEPP program?
LaBEL: We have two sides of the program: The NASA Electronic Parts and Packaging (NEPP) program and its sister portion called NASA EEE [Electrical, Electronic, Electromechanical] Parts Assurance Group (NEPAG).
NEPAG looks at problems with electronics that are already currently in design or in systems, for example cracked capacitor or DC-to-DC converter issues on a device that's already deployed. They also hold weekly telecons with the community to share the latest information on tasks, failures, DSCC audits, GIDEP [Government-Industry Data Exchange Program], and so on. Once a month, they have an inter-national version of this telecon that includes the European Space Agency, Japanese Space Agency, Canadian Space Agency, and others.
MIL EMBEDDED: So the intention of NEPAG is to talk about component-related issues in deployed or in-design systems?
LaBEL: That' a pretty accurate take on it. It's really part of our attempt to continue the infrastructure within the community. If you go back 10 years or so when we started having a big push for COTS and went away from MIL standards for parts and testing, the infrastructure program started falling away in favor of nothing but high technology and pushing the state-of-the-art envelope without really dealing with day-to-day issues where we bend metal.
MIL EMBEDDED: When you say, 'bend metal, what are you really referring to?
LaBEL: Build and design of electrical systems used in space. We are trying to ensure that the community stays together at least within the NASA side, and as best as we can within the entire aerospace community.
MIL EMBEDDED: So in essence, NEPAG is one big components engineering group?
LaBEL: A virtual one across the agency, yes, and we're trying to make it into the community so we can say, 'Hey, does anyone have a spare part,' for example. Or 'We need a spare to do this, that, or the other thing.'
MIL EMBEDDED: Are you aware that NSWC [Naval Surface Warfare Center] Corona through GIDEP actually has the components working group for obsolescence? How are you different from that, other than the fact that your platforms fly higher than any others?
LaBEL: Part of the NEPAG community includes NSWC in multiple forms. Another difference is that our sharing of what's available for parts is more to stock spares from flight programs within NASA as opposed to obsolescence.
MIL EMBEDDED: So you're focused on sourcing those components as opposed to cost avoidance and other sorts of things?
LaBEL: Right. For NASA, obsolescence has only been a big issue for the shuttle and space station programs, where we're supposed to build 'onesy-twosy' things.
MIL EMBEDDED: Getting those Am29050 CPUs for the shuttle cockpit is kind of tough when AMD obsoleted those 10 years ago. So tell us about NEPP.
LaBEL: So the other half of the program is the NEPP side. And that's focused on research tasks, looking at new and emerging technologies and devices with the intent of gaining knowledge on failure modes within the technology or device, learning how to qualify, and developing guidance for qualification insertion.
MIL EMBEDDED: So you're referring to things like upgrading, SRAM derating, DRAM guard-banding, and so on, but over a wider temperature range while looking at various failure modes?
LaBEL: Right where you can and can't use these technologies. A hot question right now is: Are current radiation test methods adequate? And the answer is blatantly no. We can't change them fast enough to keep up with technology.
But we're concerned with both the radiation and reliability sides, device temperature range, package outgassing - everything that goes through parts qualification.
MIL EMBEDDED: Does NEPP publish a database of components that you qualified?
LaBEL: The NEPP program doesn't qualify components. We're just determining if something can be qualified or if something needs to be done differently for qualification. The NEPP website [http://nepp.nasa.gov] is public and has links to all the test reports my group has done on radiation testing, for example. And there are quarantined areas of the website that have ITAR concerns and are password protected. However, we try to make as much data as available as we can.
MIL EMBEDDED: Do you make recommendations to various users based on research?
LaBEL: We'll say,This is great in this temperature range but not in this one, so when you go for your application, make sure you utilize it only in this temperature range. Or we might say,Radiation is the same for this total dose level but not this type of radiation effect.
MIL EMBEDDED: Do you have a newsletter that you publish or any frequent communique that goes out?
LaBEL: That's a tougher question than it should be. In the past, we have had electronic newsletters. We are considering restarting those, looking to see how we can approach those and still try to keep our management sanity when trying to keep track of a program.
MIL EMBEDDED: Which types of technologies does NEPP consider in its testing and reporting, and what's the downside?
LaBEL: Among the technologies we look at are strictly commercial technologies, so we do testing. But there is a sensitivity if we (speaking hypothetically) find something tha's bad or find a part that fails.
MIL EMBEDDED: A company can come in and say, How dare you say our part failed at such and such. You probably get threats of lawsuits, in which case you have to take it seriously.
LaBEL: Right, especially within the governments. If private concerns scream at the government, it has to be taken seriously regardless of the claim. So we have this issue of trying to get approval for release of information in those types of conditions.
MIL EMBEDDED: Is that why most can't get information out of GIDEP-unless you're on a need-to-know basis?
LaBEL: We have a softened version of GIDEP - the NASA alert system - which does less impact commentary but shares only with NASA and its immediate contractor family. We are working hard to get a better way of doing that, but we're not making lots of progress lately. Mike Sampson, NEPAG co-manager, is championing better information flow.
MIL EMBEDDED: Publically available information on NEPP is kind of revolutionary because every other branch of the service has its own database, but they don't make it public or share it with other services.
LaBEL: There are things that obviously we do not put into public consumption, things that will have legal ramifications. But any data that is strictly performance data or scientific information, we put into the public domain. That's really our goal. If we take a data point and nobody sees it, that's not a good thing.
We tend to push the state-of-the-art of technology more so than others. And that's because NASA's scientific instrumentation by definition is pushing the state-of-the-art. So we have to take a few more risks, learn a few more things, and try to push it as best we can.
Ken LaBel is co-manager, NASA Electronic Parts and Packaging program, and group leader, Radiation Effects and Analysis Group, NASA/GSFC. He has worked at NASA since 1983 after graduating from The Johns Hopkins University with a BES in EECS (minor in Mathematical Sciences). His career at NASA has included development of hardware/software for ground systems; advanced technology; flight hardware; systems engineering; and radiation hardness assurance/research for more than 50 NASA projects. Ken has published more than 100 papers as author/coauthor, has taught multiple short courses, and is a recognized expert in radiation effects systems engineering.