Back in March as Jared Isaacman was settling into his post as NASA administrator, he had spent enough hours looking at what was being spent on the Artemis program and developing his own ideas of what was needed for his vision (along with the administration's) of what a settlement on the moon needed and began efforts to "trim the fat." Thus began a presentation called "Ignition" or "the Ignition Event" and Isaacman let the collection of programs he intended to cancel be known. Sure, there was grumbling from contractors who saw their gravy train being cut off. Completely to be expected.
“For too long we tried to satisfy every stakeholder,” he said during the Ignition event in March. “Billions of dollars wasted. Years lost. Hardware that never launched. Fewer flagship science missions. And fewer astronauts in space, which means fewer kids dressing up as astronauts for Halloween. I don’t like it. The president doesn’t like it. The American people have waited long enough.”
We've talked about this to some degree, and a key part was canceling the Lunar Gateway - talked about in March but only officially told to shut down last week.
On Wednesday, NASA’s Office of the Inspector General prepared a memorandum on the elements of the Artemis Program that NASA was canceling as its focus shifted to the Moon’s surface. These were:
- Exploration Upper Stage, an upgrade for the Space Launch System rocket
- Universal Stage Adapter, which links the Orion spacecraft to the Exploration Upper Stage
- Mobile Launcher 2, a larger launch tower for the upgraded Space Launch System rocket
- Habitation and Logistics Outpost, a habitation module for the Lunar Gateway
The memorandum notes that each of these projects has experienced substantial cost increases and numerous delays over the last decade.
“Over the course of their life cycles, the combined contract values for these efforts ballooned from nearly $2.8 billion to $5.9 billion and NASA extended their contracted delivery dates by up to seven years,” states the report by the inspector general. “However, our projections indicate that if NASA allowed work to continue to completion, the systems would have cost more and taken longer than what was on contract.”
Eric Berger at Ars Technica has done a feature piece on this, and for illustration chooses the least expensive module of the four, the Universal Stage Adapter. NASA contracted with Dynetics in June 2017 to design, test, and build this piece of spaceflight hardware. Not particularly big or exotic, the original contract awarded to Dynetics totaled $131 million, to which NASA added another $9 million for a payload separation system, so $140 million.
At the time the program was canceled earlier this year, the contract value had grown to $353 million, with a delivery date delayed to September 2028. The inspector general’s report projected that the project would likely cost $497 million and not be ready until May 2030.
Left to its previous price trajectory, $497 million is probably optimistically low, but it's pretty easy "in your head math" to see $497 is closer to four times the $131 ($524 million) than to three times the initial contract (or $393 million).
To reduce it to its essence, NASA was probably going to pay half a billion dollars for a tall (33 feet) but relatively straightforward stage adaptor. This doesn’t have propulsion or anything like that on board. On top of the cost, it was probably going to take 13 years to complete.
Admittedly huge, it's a tapered pipe for a fancy, overpriced system. The only important thing it has to do is match the pieces it's installed between.
A test version of the Universal Stage Adapter is seen at Marshall Space Flight Center. Credit: NASA

So a 30' tall pipe that costs more than a Falcon9?
ReplyDeleteThat's legacy aerospace for you.
Now do this to every other part of the SLS/Artemis/Lunar Gateway/Lunar Settlement.
Space is hard.
ReplyDeleteEspecially when the companies are spending all that sweet money on Cost Plus contracts and produce nothing that works.
Delete$500 Million? SpaceX would have set up a factory, produced 20 test objects, tested, redesigned, retooled and produced 5 flight articles for that amount. For a 30 foot tall conical shell with no other features.
I was gonna say space is harder when you're stupid, but this isn't stupid, it's organized crime. Get everything on a cost plus contract, hire people who aren't qualified to do the work, and get paid with a bonus for all the rework you have to do. Yeah, I'm extrapolating wildly based on a bunch of other stories I've read about the SLS.
DeleteIn 1986 I joined Rocketdyne to work on the ISS Power System ( Work Package 4). The core of our design (20kHz AC distribution) was driven by ONE engineer at Lewis Research Center who was going to die on the hill of 20kHz power in space. Not 50Hz/60Hz commercial standard, not 400Hz aircraft and fire control, but 20Khz so we could save weight on transformer magnetics. No concern that there were (and are) no system power supplies designed for that frequency of mains power either.
DeleteThe other WPs considered it a significant technical risk, but we blithely carried on.
It wasn't until he retired that Rocketdyne was allowed to acknowledge we had no Litz wire cables or connectors that were suitable for space operation nor did we know how to build the cables in ways that they could be folded for launch and unfolded for operation. That is when sanity took over and we went to a 160V DC bus design.
Multiply that waste for every project and amplify for 40 years.
Thanks! A perfect example story.
DeleteWhen the only ones lobbying for cost plus contracts are congress critters, who must be siphoning off money for their favorite funds - like themselves, this is why projects get so eff'ed up. From a very, very narrow viewpoint, the weight of the flight hardware, he probably had a point. Step back a little for a slightly higher overview and it does away.
He was a loon... At that frequency you are using surface conduction (not bulk) so you need heavy braid of finer wires (Litz wire) and very high contact count connectors. Neither exist off the shelf so WP4 would have to braid them up from kapton-insulated wire and figure out how to build ~300 contact space rated mechanically operated connectors. The mass of the connectors alone would eliminate any potential weight savings, not to mention the excess cost for all the custom builds that would have to be vac and vibe tested.
DeleteRocketdyne spent millions of dollars to build a faraday-isolated (~140dB??) high power test facility at the DeSoto facility so we could run multi-KW devices without blacking out radio and TV signals over the San Fernando Valley.
Stupid on first look, even stupider the deeper you looked, and millions wasted because NASA management wouldn't tell ONE engineer to sit down and shut up or quit.
"...figure out how to build ~300 contact space rated mechanically operated connectors."
DeleteOne of those "I heard it but can't swear it's true" things is that since SpaceX now has more experience operating satellites in various earth orbits (mostly LEO) than the rest of the world combined over the last 60 years, they've concluded that they really don't need anything other than the typical connectors - like the ones rated "industrial", not "Class S" or space rated. Not even rated for Minnesota winters.
HPE's Spaceborne Computer project has sent COTS servers to ISS three times to test memory and storage reliability and high performance computing. The systems did just fine. Space-rated components made sense for Voyager and Pioneer but are just a legacy requirement today.
DeleteAnd, of course, even if SpaceX is using space-rated hardware, the amount they are using will force said space-rated hardware to become cheaper as the sheer volume of hardware being used will force manufacturers to industrialize and mass produce and with SpaceX buying most all of the hardware it will allow SpaceX to set the price.
DeleteOr, well, SpaceX will either buy the producing companies or, if given enough BS from said companies, will stand up their own pieces-parts manufacters.
Nobody is leaving near earth orbit. Never has, never will.
ReplyDelete