The status of SLS for its first launch in 2018

Please consider donating to Behind the Black, by giving either a one-time contribution or a regular subscription, as outlined in the tip jar to the right. Your support will allow me to continue covering science and culture as I have for the past twenty years, independent and free from any outside influence.

Link here. The article gives an excellent and detailed overview of where construction of SLS presently sits, what the problems are that still remain, and what they have done to overcome them.

What struck me most however in reading the article was how long it takes them to do anything. For example, it appears they will assemble the rocket in the spring of 2018 for a December launch. In fact, the description of this assembly in this article partly explains to me why SLS will have the ungodly slow launch rate of at best once every two years.

In addition, the article describes how NASA has handled a number of engineering issues that have come up, and for each the approach seems to me to be more complicated than necessary. However, I am not an engineer and have never been part of this kind of work. Maybe everyone does it this way.


  • Edward

    Robert wrote: “What struck me most however in reading the article was how long it takes them to do anything. … In addition, the article describes how NASA has handled a number of engineering issues that have come up, and for each the approach seems to me to be more complicated than necessary.”

    NASA may be overcautious in their approach, but they have had the bad experience of having lost three crews during test and missions. Considering the political fallout after each loss, I am hardly surprised at such caution.

    Generally, however, building the first of any new design tends to take longer, as the first unit is used to evaluate the effectiveness of the design as well as the processes for operation, and additional time must be put into the development schedule to account for the unexpected.

    In one case, the article suggests that a problem may not be real, “So we reopened this issue. We’re actually not even sure this [wayward retaining cup] is a credible failure because of the way the test was set up.”

    This FOD problem may have merely been a problem with their process. Sometimes the darnedest things go wrong.

    The article mentions that the Interim Cryogenic Propulsion Stage (ICPS) umbilical loads were greater than originally expected, but the article explains that this is because they had planned on using the same hardware as on Delta then had to make a change in that plan. This is one of those times when the plan does not survive contact with reality, and why development time is slow. I’m not sure why this happened, but it reminds me of a time when I suddenly had to fit an off the shelf computer board into a box that had been designed for a smaller sized board — it was a management decision made well after the board sizes and box dimensions had been frozen.

    This may have been when I started saying, “Some days just go better than others. This is not one of those days.”

    The Orion Service Module Umbilical had an issue that seemed solved but was reopened a second time. Perhaps testing showed an unforeseen condition. The solution had clearly seemed sufficient, but turned out not to be. I really hate when that happens. We spend way too much time on something that should have been designed right the first time, and I always wonder why we didn’t have the imagination to see this potential problem in the first place.

    Testing generally consists of qualification testing on the first unit, in which the hardware is tested well beyond expected loads or environmental conditions. Acceptance testing would be done on the first several units and tests the hardware a bit beyond the expected loads or conditions and generally tests workmanship.

    When many units are built, testing may be minimized. For instance, Iridium’s first order was for around 80 satellites, in the mid 1990s, and performed the expensive thermal vacuum and shake testing on only the first five, or so.

    Why Orion and SLS will take a couple of years for each production unit to be built is a mystery to me. It seems as though they plan to build only one at a time, starting the next only after completing the current one.

    The Saturns did not take so long, with four launches per year, and the Space Shuttle (the most complicated machine ever built*) did not take that long to build. The Space Shuttle first flew less than a decade after the contract was awarded, and the prototype was used for air-drop tests half a decade after the contract award.

    Perhaps NASA is overcautious, or perhaps there is another reason, but SLS seems to be taking a long time.

    To be fair, Orion’s first flight, unmanned, came less than a decade after the contract was awarded, but it awaits the SLS to be available for whatever mission it may be assigned by the president and Congress. SLS may also fly within a decade of contract award.

    For comparison, Dragon flew to the ISS about one decade after SpaceX’s founding, and half a decade after the Commercial Orbital Transportation Services concept (later: Commercial Resupply Services) was announced.
    “The program was announced on January 18, 2006 and successfully flew all cargo demonstration flights by September 2013.”

    Although the capabilities are less, development was more rapid and at considerably less cost.

    Notice that the Space Shuttle is considered more complex than even the Large Hadron Collider.

  • Localfluff


    But Russia hasn’t lost any crew since 1971. And the SLS is just the old space shuttle without the orbiter. Its boosters, tank and engines have been under development since the 1970s. It should be quicker, cheaper and safer to develop SLS than any other “new” launcher ever. (It just happens to be more politically corrupt than any other).

  • Orion314

    People need to see what the SLS really is , nothing but a make-work govt jobs program , with no intention of a launch, ever.

  • Localfluff

    “People need to see what the SLS really is , nothing but a make-work govt jobs program , with no intention of a launch, ever.”

    But it will make more works if it is launched. I bet that SLS will be launched a few times.
    And I love the idea of sending two of them to Europa. But maybe the lack of enthusiasm in NASA and all planetary scientists about this super grand plan, is telling. Pipe dream. JPL people have smoked enough to recognize it when they smell it. Don’t even need a spectrometer to do that as thick as the fog is.

  • Edward

    Localfluff wrote: “the SLS is just the old space shuttle without the orbiter.”

    Essentially correct, but the reconfiguration requires verification of the new design, new ground support equipment, and some changed mount points from the side of the fuel tank to the top and bottom.

    I’m not quite sure how Russia not losing anyone in four decades has to do with it, as they have not made significant changes since then.

    However, yesterday I distracted myself from the point of my essay, which was supposed to be that engineers like to *understand* the equipment, processes, and the system. If they do not understand the system, its performance relative to the intended performance, and the why and how it all works, then unexpected tragedies can occur.

    SpaceX has tested its Crewed Dragon spacecraft, but more tests are necessary before the engineers can be confident that they understand the spacecraft and its associated systems well enough to safely fly crews into space.

    NASA and Thiokol engineers were studying the performance of Shuttle SRB O-rings and the damage that they were taking despite the intent of the design being to prevent such damage. They believed that they had this understanding, but were wrong. Similar investigations were being taken about foam insulation falling off the Shuttle external tank. As the TV miniseries “From the Earth To the Moon” said about the Apollo 1 fire, NASA lacked the imagination that such a fire could occur.

    Plenty of unexpected things can go wrong in many endeavors. The Lockheed Electra had a couple of airplanes disintegrate in midair when the engine vibrations coupled into the airframes at the airframes’ natural frequency, causing the wings to come off.

    A Virgin Galactic pilot released the spacecraft’s feathering system prematurely, causing a deadly accident. The human in the system was the weak point.

    The parachutes on a Soyuz failed to deploy, and a valve accidentally opened when the descent module separated from the orbital module on another Soyuz. Both accidents killed the respective crews.

    Not all problems are found during test, but several additional problems can be discovered during the search for the cause of a test failure. All these problems need to be corrected in order to prevent life threatening trouble during operations of the spacecraft. Thus, test schedules may seem unnecessarily stretched out, but rushing a fix due to lack of schedule can lead to life threatening (or mission ending) trouble.

    This is one reason why we see so many delays in development programs.

    Another reason is insufficient funding. There have been times when Congress chose to reduce annual funding on projects and extend the time-frame for completion, only to have the overall cost of the project be greater than it would have been with sufficient funding each year.

    Other technical problems can happen to a project. Virgin Galactic tried to scale up an engine only to discover that the larger version may experience life threatening vibrations, so they slipped years behind schedule. The Space Shuttle also had difficulties implementing the design, resulting in a three-year delay in its first flight.

    We keep hearing that a target launch date for a new spacecraft has a 70% chance of being achieved, but sometimes I wonder if they really are achieved 70% of the time.
    From the article Robert linked: “The November 2018 date is tied to a review of the SLS program completed in August 2014, which concluded there was a 70 percent chance SLS would be ready for its first flight by November 2018.”

    I fear that Orion314 is half correct, that SLS is a Congressional make-work program, but I also suspect that Congress will fund a few launches. I also fear that without an actual mission, SLS will go the way of the Saturns, but only after we have spent huge sums to develop it — funds that could have been so much more productive on other exploration projects.

    We once had the equivalent of SLS with the Saturns, but four decades ago Congress and the president allowed it to be abandoned. Now we are spending huge sums remaking a capability that we once had, all at the expense of developing new capabilities and exploring the solar system.

  • pzatchok

    Even NASA doesn’t think the SLS/Orion craft will fly more than once.

    They are using old refurbished shuttle engines for the first flight only. After that they plan on using brand new and newly designed engines that will bring their own delays into the mix.

    Why not take the time and effort to design, build and use the new engines now on the first flight?

    I’ll tell you why. they don’t plan on having a mission or funding for a second flight. By the time a second flight is scheduled and the ship is built the private space industry will have passed them up and would already be doing the job of SLS.

    Plus all the BS about SLS being the rocket to take us to Mars. Please, its only the rocket that will take us to the ship that will take us to mars. Its a taxi..

  • Localfluff

    “Safe is not an option”, is the title of a book about spaceflight. If you cannot take a 1½% chance of blowing up at launch or landing, then you should not be an astronaut, just stay home in bed and hope you’ll live for ever.

    There are no technical reasons for SLS to be so very much more expensive and take so much longer to develop than any other launcher in world history. It is just a reconfiguration of the space shuttle, a great simplification since it doesn’t have the orbiter. SLS should have negligible development costs and much less than half of the operational costs as the shuttle system. The difficult part, the reusable orbiter, is gone.

    SLS was practically designed and ready to fly already in the 1970s. It is so extremely expensive and very late because it is CROOKED! And since corruption is pretty random, it depends on who talks to whom when, logic and technical knowledge are worthless for predicting if and when it will fly and at what cost. Even the Soviet Union was less corrupt when it so successfully developed and flew its Energia/Buran, which was a much more intelligent design than the shuttle or SLS.

    There are actually 16 RS-25, shuttle main engines, in storage. That’s enough for four SLS. I suppose they (some of them) were literally taken from the now scuttled shuttle orbiters. Proven engines, not just by design but the actual items. I’m not a rocket scientist but I only hear great things about the RS-25 engine. Other than its cost maybe, but that has political reasons as I said above. So I bet four SLS will fly successfully. If the redesigned unreusable engine version will work is another issue.

  • Alex

    Localfluff: The RS-25 engine is not such great design as you might assume. Working at the limit (for example its turbo pumps bearings), a very vulnerable and maintenance intensive design. An extreme expensive. It was the wrong decision to design it for maximum Isp, which would have not been necessary if over whole Shuttle system design as such would have been better. You are right Soviets/Russian made often more intelligent basic design. Even SpaceX is using at present intensively a superior Soviet/Russian design for its Falcon 9 first stage aerodynamic control fins for example!

  • Doug

    This gives me pause….

    When Mr. Hill noted that there are no hold down bolts for SLS, this prompted some members of the NAC to ask what was actually holding the vehicle to the pad and what the implications of no hold down bolts were.

    Mr. Hill noted that the immense weight of the SLS rocket (between 5.75 and 6 million pounds) is more than enough to keep it grounded firmly to the launch pad during the fire up sequence of the four RS-25 engines (which – at liftoff – will only be producing about 2 million pounds of thrust).

    The fact that SLS’s engines will be started in a staggered sequence to avoid pogo oscillations and shut down in a staggered sequence in the event of a pad abort will prevent any misalignment of the SLS the NAC was worried about.

  • Edward

    Actually, NASA has planned EM-1 (unmanned) and EM-2 (manned). Congress has decreed that there will be an SLS used for a Europa Mission (unmanned), and NASA and the president still think that there will be a mission (manned) to an asteroid put into cislunar space by the Asteroid Redirect Mission (this hasn’t been cancelled yet?). That’s three almost-certain launches and one possible launch.

    Future engines are not newly designed but are redesigned Shuttle engines, redesigned so that they are not reusable, thus supposedly not as expensive.

    You asked: “Why not take the time and effort to design, build and use the new engines now on the first flight?”

    That could delay the first flight even further, resulting in even more expense while everyone sits around waiting for the new engines. Using existing engines saves a couple of billion dollars per year.

    Localfluff wrote: “If you cannot take a 1½% chance of blowing up at launch or landing, then you should not be an astronaut, just stay home in bed and hope you’ll live for ever.”

    Actually, all the testing and redesigning is what makes the rocket, spacecraft, and the ground support equipment (GSE) work well enough that you have that 98.5% chance of not blowing up on launch. The article’s mention of FOD (foreign object debris) is an example of how GSE can cause damage that results in rocket explosion.

    SLS was not practically designed in the 1970s, unless you mean that the Saturn-V was practically the SLS. Compare the Shuttle external tank (ET) with the SLS, and you will see differences, and differences are critical to be verified. When it comes to vibration, for example, everything behaves like a spring. The natural frequency of the ET is different than the natural frequency of the SLS. More care must be taken to design and verify that these natural frequencies are not in the range that harm human crews. This is the problem that most plagued the Ares rocket of the Constellation program and is plaguing Virgin Galactic’s SpaceShipTwo.

    Without verification, astronauts cannot be assured of a 98.5% likelihood of survival. For instance, the FAA requires a whole lot of certification for seemingly minor changes on airliners. Winglets are a good example. They were shown to be a good idea, but they took about a decade to start showing up as retrofits to existing aircraft, because of the certification requirement. It might be OK for astronauts to take a 1.5% chance of death (although, Congressional post-accident investigations suggest otherwise), but airliners are far, far more reliable than that – and for very good reason.

    You guys might think that the new rocket is just an ET with the SRBs, but the core rocket is not just an ET. Not only is the core larger, the mount points for the engines and payload are new, too. The ET is not structurally designed for the redirected forces, it was designed for the side mounted Shuttle. Not accounting for the changes *will* result in tragic disaster and questions that make the Challenger investigation seem friendly.

    Adding a fourth engine requires even more changes. The engine support structure for SLS is for four engines, not three, so the Shuttle’s three-engine support structure is not appropriate. Fuel piping now has to go to the engines at the base of the rocket, not to the engines on the side-mounted Shuttle. There might even have to be a fairing designed for the top, if the Europa mission is to survive launch, and NASA had better verify that it opens properly, otherwise the Europa mission won’t get out of low Earth orbit.

    Much of the GSE must be redesigned for the larger rocket, and the fueling rate may also be faster, too, meaning larger fueling hoses, ports, and valves.

    Is it all crooked? Possibly. Congress certainly seems to like to spend more money than they receive in revenue – but then again, Congresscritters keep getting reelected, and that is not NASA’s, SLS’s, or the contractors’ fault. If you don’t like the overspending, then do like me and don’t vote for the incumbents.

    Why is SLS taking so long while ULA’s Vulcan rocket should be ready in half the time? Well, that’s government for you. The basic SLS requirements are by Congressional decree, not based upon civil, military, or commercial needs.

    Until today, I really had no idea how little people knew about the fundamentals of design, manufacture, test, and operations. Apparently the experience of building a tree fort leaves people with the impression that things are just slapped together with enough nails to hold it in place.

    My understanding of the use of hold downs was to prevent swaying of the rocket as it went from positive force (weight) on the pad to lifting forces. Releasing hold downs after full thrust was achieved allowed the rocket to “leap” off the pad, reducing the time it spent next to the gantry or umbilical tower, thus reducing the chance that it sways into the structure due to correcting any tipping forces*. It also allows for safe shutdown of the engines should one of them not achieve full thrust, as the rocket has not left the pad, yet.

    (Imagine a six million pound rocket falling an inch or so back onto the pad. After the explosion and the fire is out, there wouldn’t be much left of the pad and equipment.)

    It was also my understanding that many rockets use thrust vectoring (swiveling the rocket engines) to correct for tipping or imbalances. If this is the case with the SLS, then the base of the rocket has an increased chance of the engine bells contacting the pad’s mount points before they can clear them. The alternative is to have thrusters near the top of the rocket to counter any tipping.

    * For those new to rocketry, a rocket, propelled from the bottom, has a similar problem as when you try to balance a pencil on your finger. Goddard, in the early days, got around this problem by mounting his engine near the top of the vehicle structure.

  • pzatchok


    You do realize that all those problems that lead to delays were solved faster 50 years ago with brain power and slide rules.

    This isn’t NASA’s first rodeo and they have supercomputers now to do the hard number crunching faster.

    Was the first space flight of the STS shuttle manned or unmanned?
    And why would they risk it then but not now?

    SLS was always touted as the system to carry man to Mars. Any Mars missions planned? And exactly how are the SLS and Orion going to be used to take men to Mars and back?

    And why is the Falcon 9 being turned into the Falcon heavy without 20 billion in expence and all the excuses?

  • Edward

    You wrote: “You do realize that all those problems that lead to delays were solved faster 50 years ago with brain power and slide rules.”

    I’m not quite sure what you are talking about, here. If those supercomputers and slide rules were able to compensate for human error, then we wouldn’t need verification and test of hardware and processes.

    “And why would they risk it then but not now?”

    Could it be that they were overconfident then. The Shuttle does not have the best of track records.

    “SLS was always touted as the system to carry man to Mars. Any Mars missions planned? And exactly how are the SLS and Orion going to be used to take men to Mars and back?”

    Robert has a response to that:

    So far, only JPL has any NASA Mars missions, and SpaceX has a Mars mission that is also unmanned, which NASA will have some participation in. As for manned Mars missions, that is an idea that keeps kicking around NASA, but there is no funding for any specific mission or method of getting there — to the surface or to orbit.

    “And why is the Falcon 9 being turned into the Falcon heavy without 20 billion in expence and all the excuses?”

    To quote what I said above: “Well, that’s government for you.” SpaceX has a great financial incentive to not waste money and resources; government has none. We will reelect our Congresscritters whether or not they overspend, as the past three decades have demonstrated.

    I am not explaining or justifying why SLS and Orion are so *expensive*, I am explaining the methods and reasons for the test and verification that a cautious industry undergoes. Even SpaceX tests its newly manufactured engines. It would be less expensive to not test them, but it would also be much riskier. After all, SpaceX got a big surprise when an upper stage disintegrated, last year, said to be due to a poorly constructed set of struts, built below strength specifications. Apparently, SpaceX saved some money testing items that their vendor(s) was supposed to test, and if so, that policy was penny wise but mission-critical foolish.

    Not only do I not justify the cost, but I disagree with the development of them until we are ready to use them for something. We currently have no plans to use them for something useful; all we have are some ideas that Congress is *not* talking about funding, even in the future.

    SLS and Orion *are* expensive, and if they do not get used for a whole lot of missions that, collectively, make their high expense worth it, then we have spent a great deal of money that could have been used on other, more productive, exploration projects. So far, there are no missions or mission-proposals that would make even the concept of the SLS worth considering, much less worth developing; Mars is just an idea, and for the future, at that. Orion could probably still be used as transport craft for missions to the Moon, if they were once again planned, as they were before Obama, but it is not likely to be useful for reentry into Mars’ atmosphere for a mission to Mars.

Leave a Reply

Your email address will not be published. Required fields are marked *