GAO: Orion Program Plagued by Delays, Cost Overruns

NASA’s Orion with the European Service Module (Credit: ESA–D. Ducros)

by Douglas Messier
Managing Editor

Cost overruns and schedule delays continue to plague NASA’s Orion Multipurpose Crew Vehicle, according to a new assessment by the Government Accountability Office (GAO).

NASA expects the Orion program to exceed its $11.28 billion baseline budget, which covers expenditures through the Exploration Mission-2 mission, the report stated. The space agency expects to complete a new cost estimate by June.

A late arriving European service module (ESM) from ESA has been a major headache for the program.

“The module has proven more difficult to produce than expected and has been delayed numerous times. The service module is at risk of further delays that may affect NASA’s planned launch date for EM-1, which would begin to consume schedule reserve for EM-2,” the assessment stated.

The EM-1 mission, which will send an automated Orion spacecraft around the moon, will most likely occur in 2020. EM-2 will carry a crew into space in 2023.

“Program officials stated that recent ESM delays are due in part to late component deliveries from subcontractors, especially valves,” the document stated. “However, they also noted at least one of those valves is currently not meeting specifications, indicating that the design was not sufficiently mature prior to production. The valve, a heritage design from the Space Shuttle program that maintains pressure in the propulsion system, is not sealing properly due to the increased pressures necessary on the ESM.”

GAO also found that part of the cost increase can be attributed to the switch from a single-piece heat shield to one that uses blocks of materials.

“In addition, the program is addressing some parts material failures,” the report stated. “Specifically, the spacecraft’s avionics cards did not withstand vibration testing and required a materials change.”

The GAO’s assessment of the Orion program is below.

NASA: Assessments of Major Projects
Government Accountability Office
May 1, 2018
Full Report

Orion Multi-Purpose Crew Vehicle

The Orion Multi-Purpose Crew Vehicle (Orion) is being developed to transport and support astronauts beyond low-Earth orbit, including traveling to Mars or an asteroid. The Orion program is continuing to advance development of the human safety features, designs, and systems started under the Constellation program, which was canceled in 2010. Orion is planned to launch atop NASA’s Space Launch System (SLS). The current design of Orion consists of a crew module, service module, and launch abort system.

Project Summary

The Orion program continues to operate within its schedule baseline but NASA expects the program to exceed its cost baseline through Exploration Mission-2 (EM-2) due to new hardware and the program addressing development challenges. The extent of cost growth is unknown, but NASA plans to complete a new cost estimate by June 2018.  For example, according to NASA, the cost increases have been driven in part by moving from a single-piece, or monolithic, heat shield design to one that employs blocks in order to improve its structural strength.

The program’s service module — contributed by the European Space Agency — is currently driving the program schedule as well as the launch schedule for the first mission. The module has proven more difficult to produce than expected and has been delayed numerous times. The service module is at risk of further delays that may affect NASA’s planned launch date for EM-1, which would begin to consume schedule reserve for EM-2.

In addition, the program is addressing some parts material failures. Specifically, the spacecraft’s avionics cards did not withstand vibration testing and required a materials change.

Cost and Schedule Status

Credit: GAO

The Orion program continues to operate within its schedule baseline but NASA expects the program to exceed its cost baseline through EM-2, and is at risk for future schedule delays. The project’s life-cycle cost estimate is expected to increase beyond its $11.28 billion baseline — due in part to the EM-1 schedule delay—when a planned, new cost estimate for the program is complete in June 2018.

According to preliminary analysis, major drivers of the potential cost increase also include new hardware and addressing development challenges. For example, there was a cost impact when the program moved from a single-piece, or monolithic, heat shield design to one that employs blocks in order to improve its structural strength.

In December 2017, NASA announced December 2019 as the new internal launch readiness date for EM-1 and that the agency also allocated 6 months of schedule reserve to June 2020 for possible manufacturing and production schedule risks. While the Orion program did not have a committed launch date for EM-1, the recent delay has reduced the amount of time available to the program between EM-1 and its committed EM-2 launch date of April 2023.

Credit: GAO

In addition, the delay means that the program will continue to consume resources for EM-1 that would have otherwise been available for development on EM-2, thus increasing pressure on the EM-2 cost and schedule. If NASA delays EM-1 beyond December 2019 — which is likely given both Orion and the Space Launch System have no schedule margin to meet their deliveries for this date—the schedule reserve for Orion’s committed EM-2 launch date of April 2023 would continue to erode and put the program at risk for future schedule delays.

Developmental Partner

The late completion and delivery of the European Service Module (ESM) — a European Space Agency contribution via agreement with NASA — is driving the program’s schedule and may further delay EM-1. The European Space Agency has delayed delivery of the service module 14 months since the element’s critical design review in June 2016.

The ESM currently has no schedule reserve to support the December 2019 launch schedule, meaning that any additional delays will compress or delay integration activities prior to launch. Further, NASA is tracking a risk that the ESM could be delayed beyond the current estimated delivery date of June 2018. Such a delay would likely delay the EM-1 launch date beyond December 2019.

Program officials stated that recent ESM delays are due in part to late component deliveries from subcontractors, especially valves. However, they also noted at least one of those valves is currently not meeting specifications, indicating that the design was not sufficiently mature prior to production. The valve, a heritage design from the Space Shuttle program that maintains pressure in the propulsion system, is not sealing properly due to the increased pressures necessary on the ESM.

The program has previously stated that all sides believed that the development of the ESM would be easier than it has proven to be, being based on a prior European Space Agency spacecraft. However, the changes have been more substantial than expected and the production of the first flight unit has faced setbacks. The program stated that they expect the production of the flight unit for the second flight to be quicker, though it will require some additional elements to support crew.

Technology and Design

Avionics design has recently become an issue for the Orion program and is currently 2 months behind the ESM in terms of driving the program’s schedule. The avionics cards, along with many other onboard systems, have to withstand the vibration and radiation environment on Orion.

The program found that the circuits on the avionics cards were cracking under operating conditions due to a poor design. As a result, officials stated that the program determined the root cause of the failure and redesigned the cards’ base material to better withstand the environment. The program worked to reorganize the integration and test schedule to allow for the replacement of the cards.

Program Office Comments

In commenting on a draft of this assessment, Orion program officials stated that they believe the risk informed approach that they use to address and resolve issues has proven to be successful. Also, program officials stated the program remains on track to meet its April 2023 baseline for EM-2. Program officials also provided technical comments, which were incorporated as appropriate.

  • Michael Halpern

    Talking about a program that needs to be put out of its misery

  • Steve Ksiazek

    I wonder how much of this can be blamed on the decision to outsource the service module to the ESA. Didn’t Lockheed already have a SM design that was bought and paid for ? And how the hell are the avionics cards not holding up ?

  • newpapyrus

    Once NASA decided to support Commercial Crew Development, the Orion program should have been terminated in favor of using those funds to develop a reusable (LOX/LH2) lunar lander such as the ULA’s XEUS concept that could operate between EML1 and the lunar surface and would also be capable of transporting astronauts between propellant producing water depots located at LEO and EML1.

    Even if the Orion is developed, it will probably be obsolete within five years after it comes into operation.

    Propellant producing water depots and reusable LOX/LH2 vehicles are the future of transportation within cis-lunar space and probably the future of crewed transportation from the orbits of Venus to Jupiter.

    Marcel

  • Jeff2Space

    This mess is bizarre. ESA already had a flying, working design for the ATV. Adapting this for use as an Orion service module should have involved the minimum set of changes needed to get it to work. Yet the article says:

    “the changes have been more substantial than expected”

    I’d be interested to know if these “more substantial” changes were truly necessary or if the changes were dictated by NASA because they didn’t like the way ESA designed the ATV. The example of using a space shuttle heritage valve in an application which has higher pressure is concerning.

  • Michael Halpern

    Don’t fall into the ISP trap of hydrolox

  • passinglurker

    Probably a mix of both the ATV being a leo cargo taxi and needing an upgrade anyway, and nasa meddling/dictation I’d bet. as a result nothing is saved over designing an original service module from scratch…

  • ThomasLMatula

    Wow! Shocking news…

  • duheagle

    Put out of our misery.

  • duheagle

    There was a time when NASA and its contractor team were the 1972 Miami Dolphins of human spaceflight. They have now apparently devolved all the way down to a Pop Warner League scrub squad. It’s been so long since these organizations have actually played a game that they no longer have anyone on the roster who remembers how to block, kick and do the rest of the basics. I wouldn’t trust the current crew not to put their pads and jerseys on backwards.

  • passinglurker

    Well unless you know a process for efficiently synthesizing peroxide from water it is probably the best for isru in the long term.

  • passinglurker

    So… Vacuum optimized interorbital manned shuttles when? Cause let’s face it hauling a return capsule to the moon and back isn’t the most practical or necessary of ideas.

  • duheagle

    Very concerning.

    The Orion Service Module, described as “ATV-derived,” is actually a long way from being an ATV. The ATV didn’t need much delta-V given what its launcher and mission were. The Orion SM needs a lot more delta-V as it has a lot more to push and needs to be able to make multiple maneuvers in deep space, not just fiddly little adjustment burns in LEO.

    So the ATV “main engines” – which were essentially an Apollo SM maneuvering thruster quad with all the R-4D bells pointing in the same direction – was hopelessly overmatched. NASA specified a much more powerful AJ-10-series engine that was used in Shuttle OMS pods and which has heritage all the way back to Vanguard.

    This engine is also hypergolic, but – understandably – runs at higher pressures than a maneuvering thruster. I’m guessing not enough supporting-role NASA hardware was repurposed from the Shuttle and that some leftover ATV bits proved too weak to handle the loads the AJ-10 requires.

    This is all guesswork, mind you. But, given how many other bush league fumbles and intercepted passes the NASA/contractor Orion team has been responsible for, it seems quite plausible to me.

  • Michael Halpern

    Not necessarily, you can use a mass driver system with how low the lunar gravity is and being in vacuum, you can even take advantage of the conductivity of the regolith to use as a crude ion fuel, not the best for the engine but it can work

  • Larry J

    Why should Lockheed or any other of tge SLS/Orion contractors worry about change orders, cost overruns, or schedule slips? Cost plus, baby!

  • newpapyrus

    The utilization of extraterrestrial water is the key to opening up the solar system for humanity.

    Water can be used to produce rocket propellant (LOX/LH2). And the hydrogen and oxygen ullage gasses can easily be re-liquefied using crycoolers already developed by NASA.

    The hydrogen produced from water can be used for propellant in high ISP Nuclear Thermal Rockets.

    Water can be used for cosmic radiation shielding against heavy nuclei and can also significantly reduce proton, neutron, and alpha particle exposure.

    Water can be used to produce oxygen for air as it is on the ISS.

    Water can be used for drinking and food preparation.

    Water can be used for growing food.

    Water can be used for aquaculture.

    Water can be used for washing and cleaning.

    Water ice is available on the Moon, Mars, Mercury, Callisto, the asteroid belt, and the Greek and Trojan asteroids in the orbital arc of Jupiter.

    So a water based extraterrestrial economy clearly makes the most economic sense!

    Marcel

  • Someone in the USSR said Soyuz would be obsolete in a few years too. Never underestimate how long-lived some of these systems can be.

  • passinglurker

    ionic regolith and mass divers don’t really help on the landing leg of a sustainable cislunar infrastructure. Also you’d need something to circularize on the lifting leg, etc…

  • Michael Halpern

    if you have the carbon which you need for growing food make LCH4/LO2 not hydrolox

  • Michael Halpern

    for landing use ethanol

  • passinglurker

    That’s the point outside a planet with an atmosphere carbon is much harder to find than water.

  • passinglurker

    Where are you getting the carbon?

  • Michael Halpern

    not really some asteroids have a fairly high carbon content iirc

  • passinglurker

    It’s been speculated but with the cancellation of arm we aren’t really in the asteroid prospecting business anymore

  • William Douglass

    What heat shield? The EM does not have one!

  • Search

    More like NASA decided to use some ESA contribution to add an international component to make it harder to cancel and they figured the bus was “easy”. Well spacecraft aren’t Legos. Now its a millstone. $11B and one demonstration flight.

  • Search

    Giant programs that support many voters and warchest filling companies are a feature not a bug. If you think NASA is bad avert your eyes from DoD

  • Search

    Gee why hasn’t anyone else thought of that. Rolls eyes…

  • Search

    Simple answer – all of it. Involving ESA has the allure of International goodwill and on paper is free money – but reality is its always a setback. Better to have asked ESA to build some craft all by themselves and barter the whole service or something. ATV was working after all and able to resupply ISS with propellant – now we only have an aging ’57 Chevy – er um Progress to rely on to keep ISS from being put in the drink. What could possibly go wrong.

  • ThomasLMatula

    Blasphemy! How dare you talk about saving in a program designed to create jobs? Not just American jobs but European as well? Saving money means putting folk out of work and the basic purpose of the SLS/Orion/LOP-G triad is putting folk to work.

  • Tom Billings

    A good bit more than speculation goes into asteroid estimates for carbon. The reflectance spectra of about 75% of asteroids measured have very similar spectra to carbonaceous chondrite meteorites. CC meteorites contain from 5-18% water, 3-15% native metal bits, 10-35% kerogen-like substances, some nitrogen compounds, all mixed in with a balance of crumbly clay-like material containing various silicates and other minerals.

    Osiris-Rex is headed to a rendezvous with Bennu this August, which shares these reflectance spectra. We will begin acquiring data soon thereafter on many samples, some of which will be returned by 2023. “We” are definitely in the mining business in Space as a civil society. The only question is the speed of advance, as long as “we” isn’t confined to NASA.

  • mlmontagne

    “The EM-1 mission, which will send an automated Orion spacecraft around the moon, will most likely occur in 2020. EM-2 will carry a crew into space in 2023.”
    It is reasonably possible that, by 2023, SpaceX will be ready to land large space ships on the surface of the Moon. I’m sorry, although SLS and Orion both made sense ten years ago, they have been overtaken by technological developments outside of NASA. This is not NASA’s fault, indeed, NASA has been encouraging those developments, but it is time for the politicians to admit that their political reasons for continuing these projects are just resulting in a waste of money.

  • Michael Halpern

    It isn’t the organizations, it’s the contract types, how appropriate they are for the situation, urgency or lack thereof, and the fact Congress designed SLS. With DoD cost plus while still used more than it should be makes sense and works a little bit better overall as they have urgency and changing needs, as well as some practical concerns regarding budget, and many things have been cancelled because the end product was going to be too expensive for feasible deployment, such as the original ammunition for the main gun of the new stealth destroyers, would have costed over $1m per shell, so they said no. The problem with SLS there is no urgency and no reason for contractors to stay on schedule or budget.

  • Jeff2Space

    ESA has a funny rule that what when a certain level of money is put into ESA by a member country, that same level of money must be spent in that same member country. That’s pretty much the ultimate in pork spending.

  • ThomasLMatula

    Don’t worry, the BFR has lots of cargo capcity. They won’t notice an Orion capsule stowed in a empty corner for NASA. And of course it won’t need to have a service module or even need to work. Just be accessible to the astronauts sit in it for a few hours so they will be able to claim if flew with a NASA crew around the Moon when they give it to the Smithsonian. 🙂

  • passinglurker

    It can’t all be on brf a healthy space economy needs to have more than one way to get things done affordably.

  • ThomasLMatula

    The DC-3 did a good job in the 1930’s for the airlines, but there is also the New Armstrong. And I suspect others once SpaceX shows the way.

  • windbourne

    WHAT? No. Say it is not so.
    Part of a program that was started in 2004 and has cost America 32B, or more than 2B / year, and it might have delays and costs overrun? Naaaahhh.
    It is long past time to GIVE this program to these companies and allow them to use it to compete against other private space crafts.

  • windbourne

    GOP in America has similar rule. For any money put into the feds by blue states, all of it is to go to red states or at least red districts to shore up their votes.

    Worse case of pork spending going.

  • windbourne

    no. We are simply doing it wrong.
    If trump’s ppl have a clue, and backbone, they will give this to these companies and then put up COTs and Service contracts with decent amounts.
    For the amount of money that we spend on SLS EACH YEAR, we could have paid for 3 SHLV and have them done within 5 years and them be able to send 4-10 trips to the moon EACH YEAR.

  • windbourne

    DC-3 is far more likely to be BFR.
    Hopefully, both BO and SX will view their rockets as 707, not dc-3.

  • windbourne

    yes, but LH2/LOX is simpler and will work just as fine.
    As for the mass driver on the moon, great way to launch, but landing will suck without any engines.

  • windbourne

    which methane/LOX is great for mars, but a disaster for the moon. It is far more efficient to have a lunar lander/launcher that uses LH2/LOX from the moon.

  • Arthur Hamilton

    SLS will be obsolete by 2025. There will be no way to justify throwing away a a whole billion dollars per launch rocket. Orion should be operating by now.

  • Michael Halpern

    Unless you have a rock thrower on the lander, or just bring enough fuel for landing, you are already going to be limited with lunar activity because of lack of carbon nitrogen sulfur and phosphor, and we aren’t even sure we can effectively extract water from lunar ice, declaring it as a necessity for unlocking exploration and other activity throughout the solar system is a pretty big stretch for something we don’t even know if we can effectively obtain.

  • Michael Halpern

    Those ifs are pretty giant

  • Andrew Tubbiolo

    I would not count the Space X egg until it’s hatched. That said, I believe it will hatch given their past record. Both SLS and BFR will be late, and I’m ready to believe SX will come in ahead of NASA no matter what the sequence of events are. SLS should go, not because I think BFR will be a bright shining spaceship in two years. I think it will be five +. SLS should go because there are no ultra heavy payloads to use it. Falcon as flown now can handle any reasonable near term lunar efforts as will Vulcan/ACES or New Glenn. If NASA started designing manned lunar/interplanetary modules/landers now, it would take a decade to have them ready, longer with SLS eating their budget. You’re right, if this administration and this NASA administrator really intends on returning people to the Moon, they’ll have to address the political system that’s keeping SLS going.

  • mlmontagne

    “not because I think BFR will be a bright shining spaceship in two years. I think it will be five +.”
    I’ll be entirely satisfied with that. In the meantime we’ll have Dragon 2, Starliner, and Dreamcatcher.

  • Andrew Tubbiolo

    A luxuriously rich second rate space program. 🙂 We should have such problems.

  • Michael Halpern

    plus all of CHNOPS has been found in some meteorite samples