Slashdot Mirror
Costs Soar on NASA Communications Upgrade Program
schwit1 writes A new GAO report has found that NASA's effort to upgrade the ground-based portion of its satellite communications system, used by both military satellites and manned spacecraft, is more than 30 percent over budget, with its completion now delayed two years to 2019. Worse, the GAO found that this problem program was actually one of three that have had budget problems. And that doesn't include the disastrously overbudget James Webb Space Telescope. "In its latest assessment of NASA's biggest programs, the U.S. Government Accountability Office identified the Space Network Ground Segment Sustainment (SGSS) as one of three — not counting the notoriously overbudget James Webb Space Telescope — that account for most of the projected cumulative cost growth this year. The others are the Magnetospheric Multiscale Mission, which launched March 12, and the Ice, Cloud, and Land Elevation Satellite-2, or ICESat-2, mission, the congressional watchdog agency said."
The actual GAO report is here [url]http://www.gao.gov/assets/670/669205.pdf[/url]
What it *really* says is that there are four big projects that NASA is doing that have significant cost overruns. SGSS is somewhat of a problem: while the report says they're having trouble retaining staff, I suspect the real problem is that NASA, and in particular, Space Network (who operate TDRSS and the ground stations) have a set way of doing things with existing 1970s designs and architecture and 1980s equipment. Anytime you're coming in to do an upgrade, there's a lot of undocumented lore and process that needs to be accommodated, and jumping 30 years in equipment technology leads to all sorts of unusual things. It's all well and good to say "we're going to replace that rack of discrete electronics with this whizbang 4U PC with a FPGA card in it", but that rack of discrete boxes with boards and boards of TTL and analog circuits has a lot of idiosyncratic behavior that must be duplicated.
I have a real gripe with the report too, because it talks about reducing risk. They are pleased in the report that in general NASA is making all their missions have high TRL (tech readiness levels) >=6 at PDR. Sure, this makes the mission cost/programmatic risk smaller and you come in on budget, but it also means you don't infuse new technology. It's cheap to get to TRL 4 or 5 (benchtop demos in the lab), it's really expensive to get to TRL 6 (demonstrated in relevant environment), and NASA's big bucks go to missions, not to technology development. Big technology development happens when you get a mission to take a bet on the new tech and pray that you can get it to TRL 6 by PDR/CDR.
I also have a gripe that GAO evaluates maturity of design by "percentage of released drawings at CDR". Given the increased role of software in modern spaceflight, I don't think that's a relevant metric, at least by itself.