Battle of the Collossi: SLS vs Falcon Heavy
by Dale L. Skran
|There are partisans for each vehicle, but it can be difficult for the space-interested public to separate the facts from the rhetoric.|
Current plans call for the first test flight of SLS Block I1 to be no earlier than November 2018. This mission, dubbed “Exploration Mission 1” (EM-1), calls for an uncrewed Orion capsule to be lofted around the Moon. The configuration used for this flight includes the SLS core stage, two five-segment solid rocket boosters (SRBs), and the Interim Cryogenic Propulsion Stage, or ICPS. The ICPS is a modified Delta IV upper stage. The Block 1 system supports a five-meter payload fairing, and can put 70 metric tons (MT) into low Earth orbit (LEO).
The upgraded version of the SLS planned for EM-2 in 2021 (SLS Block IB) is based on a new, higher-power “Common Upper Stage” capable of supporting an 8.4-meter payload fairing, allowing for at least 105 MT to be launched to LEO. However, 2021 is six years from now, and six years ago SpaceX was flying the Falcon 1. The point here is that it is not appropriate to compare the Falcon Heavy to SLS Block IB since there is good reason to expect that whatever SpaceX is flying in 2021 it will be a considerable evolution beyond the Falcon Heavy, just as the Falcon 9 v1.1 flying in 2015 is a considerable evolution beyond the Falcon 1 flying six years ago. Thus, we will limit the scope of the comparison to the rockets likely to be flying in the 2018–19 time frame.
With SpaceX projecting a first Falcon Heavy flight in 2015,2 even with some slippage it seems reasonable to suppose that Falcon Heavy will be at least as “operational” in 2018–19 as the SLS. Both craft should have made at least one flight, allowing mission planners to ground future efforts in reality rather than speculation.
The comparison will be based on the SpaceX web site as a source for Falcon Heavy information, and on NASA document SLS-MNL-201 V1, the “Space Launch System Program Mission Planners Guide,”3 as a source of SLS information. The following types of arguments will not be considered as they are difficult to evaluate and also not always of great interest to potential customers:
An artist’s rendition of SpaceX’s Falcon Heavy, one of two heavy-lift vehicles under development in the US today. (credit: SpaceX)
Superficially, the SLS and the Falcon Heavy have a similar appearance. Both feature side-mounted boosters around a central core stage, topped by a second stage and then its payload. Both have similarly-sized payload fairings in the 2018 timeframe. However, these similarities belie vastly different design philosophies.
The SLS design is driven by two imperatives: to achieve particular lift goals (initially 70 MT, later 105 MT and finally 130 MT) and to re-use Space Shuttle and Constellation technology. This results in the use of two extended five-segment SRBs that are much the same as the shuttle SRBs, and a central core stage based around the same liquid-oxygen/liquid-hydrogen engines that powered the Space Shuttle. The ICPS is the only non-shuttle technology in evidence, and it is a modestly modified version of the proven Delta IV upper stage, featuring one rocket engine that burns liquid oxygen and liquid hydrogen.
One advantage of this basic design approach is that solid rocket boosters provide more thrust for a given size and weight, making them good candidates to initially boost a large rocket to orbit. The high-energy (high specific impulse) upper stage assures the maximum possible terminal velocity and opens up the widest possible range of destinations. Finally, re-using the Space Shuttle Main Engines, using liquid oxygen and liquid hydrogen, provides greater lift than a center core based on less energetic propellants such as RP-1 (refined kerosene).
|The SLS and FH appear to be designed with very different goals in mind, which may seem like an apples-to-oranges comparison. And yet, come 2018, customers will face the choice between these two launch systems.|
The net result is that the SLS makes good use of existing Space Shuttle technology and the resulting design is optimized for maximum terminal velocity and a wider range of destinations with maximum payload. This description may sound like the SLS is cobbled together from old parts, which is true on some level, but considerable efforts have been made to optimize and modernize those “old parts” to create a more manufacturable vehicle.4
The Falcon Heavy evolved from a radically different design heritage than the SLS, and is based on the technology that powered the Falcon 1 and Falcon 9 rockets. Inevitably, this discussion involves some speculation, as only Elon Musk and top SpaceX engineers can say for sure what their intention may have been in creating the Falcon Heavy. The FH gives every appearance of having been designed to maximize re-usability and minimize cost of manufacturing and operations while reusing as much Falcon 9 technology as possible. Both the center core of the Falcon Heavy and the attached boosters are more or less Falcon 9 core stages. The second stage of the Falcon Heavy is an unmodified Falcon 9 second stage. All engines used are the Falcon 9 Merlin 1D and are burning liquid oxygen and RP-1, which have lower specific impulse than engines powered by liquid oxygen and liquid hydrogen.
Thus, the FH is not optimized for maximum terminal velocity or maximum payload to beyond Earth orbit (BEO) locations; instead, the goal is maximum payload to LEO for minimum cost, coupled with a significant geosynchronous Transfer orbit (GTO) payload for a minimum cost. The usage of liquid oxygen/RP-1 instead of liquid oxygen/hydrogen engines provides less thrust and results in a lower terminal velocity, but it also runs cooler, putting less wear and tear on the engines and promoting the goal of reusability. Finally, the all-liquid-engine FH allows for main stage and side booster engine restarts associated with boost-back to the launch pad and eventual reusability, something that is not possible with solid boosters.
The SLS and FH appear to be designed with very different goals in mind, which may seem like an apples-to-oranges comparison. And yet, come 2018, customers will face the choice between these two launch systems, as well as other systems such as the Ariane 5, the Delta IV Heavy, the Atlas V, and the Falcon 9. How will customers evaluate the SLS and the FH then?
The difference between the SLS and the FH becomes obvious (well, mostly obvious!) just by looking at the charts and information provided in the various references.5,6 , A summary chart follows:
|Launch System||LEO capacity||GTO capacity||Moon capacity||Mars capacity|
|SLS||70 MT||N/A||25 MT||19–20 MT|
|Falcon Heavy||53 MT||21.1 MT||N/A||13.2 MT|
All this data confirms what the potential customer should already know from understanding the architectures of each launch system. If you want to launch a large probe to the outer planets at high velocity, SLS has that capability. If you want to put a large amount of mass into LEO, the throw weights are similar.
The “fairing” is a structure that protects the payload as the rocket rises through Earth’s atmosphere and generally correlates to the size of the payload being launched. This is a relatively simple comparison to make. SLS Block I provides a fairing that is 5 meters in diameter and 19 meters long. The FH offers a fairing 5.2 meters in diameter and 13.1 meters long. Although for some applications the longer length of the SLS fairing might be important, this does not appear to be an area where one system has a major advantage. It is also possible to imagine that SpaceX could provide a longer fairing if required by a customer.
Clearly cost to the customer will be a very important consideration. Since no SLS throw weight numbers are available for GTO, we will focus on cost to LEO. Although SpaceX puts list prices on its web page, the information provided for the FH only concerns launches to GTO. We will assume that SpaceX will charge the same price for a FH mission to LEO.
An added level of confusion results from SpaceX saying that the FH price is $90 million for “Up to 6.4 MT to GTO.” Given that the full capacity of the FH shown is 21.2 MT to GTO, it is not completely obvious how to interpret the published pricing relative to the full capacity. In particular, it is unclear what might create additional costs as the FH GTO payload rises from 6.4 MT to 21.2 MT. One possibility is that SpaceX is making use of “value pricing” where prices have little relationship to underlying costs, but instead are based on value to the customer.
|In the 2018–19 timeframe, it seems reasonable to assume that you can buy at least three FH launches for the price of one SLS launch.|
Previously, the SpaceX web site7 gave a list price of $77.1 million for a FH lifting under 6.4 MT to GTO, and $135 million for more than 6.4 MT to GTO. This suggests that SpaceX intends to charge more for the full capacity of the FH, so we will use the $135 million dollar figure for lifting 53 MT to LEO on a FH. It is also interesting to note that there has been an upward drift in the pricing for the FH over the last year, from $77.1 million to the current $90 million. The reason for this upward drift is unclear, but as the FH has been under development during this period, it would not be surprising to find that as cost estimates became more firm, they also became larger. Since SpaceX is no longer publishing a full capacity price for the FH, we will adjust $135 million upward by the same ratio that the “under 6.4 MT” price has risen, yielding a price of $158 million.
The task of discovering a real price for the SLS is also difficult. The method most favorable to the SLS8 would be to use the estimate provided by Dan Dumbacher, the former NASA deputy administrator for Exploration System Development: “What we are trying to do is get SLS into that $500 to $700 million per-flight range and some of us are working to actually get it even lower than that.”9 In order to use the best possible yet reasonable number for SLS, let’s take the average of Dumbacher’s numbers to get an estimate of $600 million for the SLS Block 1.
So let’s put these numbers in a table and take a look at them:
|Launcher||Mass to LEO||Cost to LEO||Cost per MT to LEO|
|Falcon Heavy||53 MT||$158M||$2.97M/MT|
On a per metric ton basis, the difference is dramatic. Keeping in mind that a real customer will pay additional fees for integration and so on, it seems clear that if you want to launch something that weighs less than 53 MT to LEO, you will save a lot of money—more than 50 percent—by using the Falcon Heavy. Further out in space, if you want a single launch, SLS seems to gain the advantage.
In the 2018–19 timeframe, it seems reasonable to assume that you can buy at least three FH launches for the price of one SLS launch. In other words, for around $600 million you could get 70 MT to LEO with SLS, or for about the same sum you could get 159 MT to LEO with the FH, but in three 53-MT packages. Which scenario works better for you as a customer is not simple to evaluate.
Now we come to the real crux of the “big launcher debate.” On one side we find the advocates for heavy lift vehicles, who argue that “bigger is better” and claim that SLS will be safer to use since there are fewer launches and less in-space assembly. Clearly, as SLS evolves into heavier variants, the larger the payloads it can launch become and the stronger this line of thinking would appear to be. On the other side we find the advocates of in-space assembly and fuel depots, claiming that it is cheaper to go with many launches of a less expensive system.
There is also a reliability aspect to this discussion, with the big rocket advocates arguing that fewer launches lead to a more reliable end result, while the smaller rocket proponents argue that more launches lead to a more reliable rocket. This discussion is not amendable to easy resolution and clearly depends on a number of still unknown variables, such as the fundamental reliability of the SLS and the FH.
There are a lot of risk-related similarities between the SLS and the FH. Both use proven existing engines: the SSME and the Merlin 1D, respectively. Both use proven existing second stages without significant modification: the ICPS and the Falcon 9 second stage. Both are doing some development on the first stage, although clearly since both first stages are built around existing engines, the amount of new development is modest. The SLS reuses SSMEs with new tanks, but engines are the main source of complexity and failure, and the bugs were worked out of the SSMEs a long time ago. It would appear from a customer viewpoint that the technical risk and readiness risk in SLS Block I and the Falcon Heavy are similar, at least up to first launch.
It is a virtual certainty that by the end of 2018, and indeed by the end of 2019, SLS will have flown only once. However, the flight manifest provided by SpaceX shows a total of five Falcon Heavy flights,10 including a test flight. Thus, it is possible that, assuming the test flight occurs in 2015 or 2016, by 2018 the Falcon Heavy may have flown five or more times successfully. This would substantially alter any reliability calculation made by a potential customer in favor of the FH.
Will SpaceX achieve its stated goal of making the F9 and FH re-usable by 2018? This is clearly a speculative topic, but given that SpaceX has been making continued progress toward this goal, a prospective customer has to consider the scenario that, by 2018, full first stage reuse for both F9 and FH will have been demonstrated. This has the potential to alter the cost equation very much in favor of the FH. Since the SLS has not been designed for reusability, a situation could evolve where the cost to launch on FH is not one-third of the SLS price, but perhaps far less, such as one-sixth the SLS. In this situation you could buy five or six FH launches for each SLS launch. Those planning a return to the Moon, take note.
|It is possible that a combination of a few SLS Block I launches with many FH launches would produce an optimal result.|
What then should the 2021 SLS Block IB (105 MT to LEO) be compared to? Or for that matter, the SLS Block IIB (130 MT to LEO)? Keeping in mind that significant aspects of the SLS Block IIB (advanced SRBs, exploration upper stage/common upper stage) are in a relatively early stage of development, with several competing proposals under consideration,11 it would seem most appropriate to compare these rockets to the notional next-generation Falcon. SpaceX plans to begin testing a powerful methane-liquid oxygen engine code-named “Raptor” at NASA Stennis.12 The target specifications for the Raptor have varied over time, and most recently have moved to 2.2 million newtons (500,000 pounds-force) of thrust, substantially greater than the 670,000 newtons (150,000 pounds-force) of the Merlin 1D. In any case, the fully reusable Falcon “Next” is expected be built around the new Raptor engine, creating a true heavy lift vehicle that will be the core of SpaceX’s planned trips to Mars.13 This new Falcon may or may not be operational in 2021, but it is fair to say that it is not appropriate to compare the Falcon Heavy of 2015 to the SLS Block IB/Block IIB of 2021 to 2030.
With high probability, both SLS Block I and the Falcon Heavy will be flying in 2018–19. The FH seems a clear winner for customers wanting to launch less than 53 MT to LEO and 21.1 MT to GTO. There is a significant probability that SpaceX will make first stage reusability work in this time period and the FH will become an even more attractive option. A reusable FH first stage has the potential to fuel a boom in the utilization of LEO and GEO for projects that were previously too expensive to merit serious consideration.
The SLS Block I will be an option for missions beyond GTO, and especially for deep space. The best approach for major cislunar activities such as an Earth-Moon L2 waystation or a lunar base will require a complex analysis to evaluate the tradeoffs between launching a few large payloads and a larger number of smaller payloads. It is possible that a combination of a few SLS Block I launches with many FH launches would produce an optimal result.