The Space Reviewin association with SpaceNews

F-1 engines
NASA collected detailed records about the F-1 engine at the end of the Apollo program should it later decide to revive it, but doing so may not be cost-effective. (credit: NASA)

Thunder in a bottle: the non-use of the mighty F-1 engine

NASA is currently considering whether to switch its planned heavy-lift booster from Rocketdyne’s Space Shuttle Main Engine (SSME) to the same company’s RS-68 engine. Although the Exploration Systems Architecture Study originally indicated that the SSME would power the rocket, the shuttle engines are expensive and were never designed to be produced in bulk, which is why the RS-68 is now in play.

However, one engine from the agency’s past was apparently never under consideration—the powerful F-1 built by Rocketdyne for the Saturn 5. The last time that the agency semi-seriously considered returning to the Moon, the F-1 was actually under serious consideration. So much, in fact, that Rocketdyne, then part of Rockwell International, actually studied the costs of returning the F-1 to service. The costs were high, which might explain why the F-1 is still on the shelf for the new lunar plan.


One of the myths of space history is that NASA lost or destroyed the blueprints for the Saturn 5 launch vehicle and could not build it again even if they wanted to. Critics of the agency use this myth to attack what they call petty or narrow-minded bureaucrats.

Except that it is not exactly a myth. The claim is partially true, but too complex for the soundbite version that “NASA lost the blueprints,” a charge that readily suited the agency’s critics.

Much of the technical design documentation for the Saturn 5 and the Apollo spacecraft has indeed been lost or destroyed. In fact, much of it was never saved, nor meant to be saved. Certainly there are hundreds of thousands of pages of technical documentation preserved on microfilm and in boxes at the National Archives facility near Atlanta. But these records (I’ve seen them) are incomplete, and they contain considerable gaps. There is nothing close to a complete set of “Saturn blueprints” in federal archives. I have also been to the archives at Kennedy Space Center, Marshall Space Flight Center, Houston, and Fort Worth. There is nothing approaching a “complete set of Saturn 5 blueprints” anywhere.

But then, there never was a complete set of the technical documentation.

There is nothing approaching a “complete set of Saturn 5 blueprints” anywhere. But then, there never was a complete set of the technical documentation.

One problem is the definition of a “blueprint.” Lots of information goes into the design of a complex vehicle like a rocket, and it is not all blueprints. There are technical drawings, but also lists of calculations and specifications for materials used to produce parts, as well as instructions for how to assemble the pieces. In addition, although complex machines require significant documentation, some of the information required for their production resides in the minds of the workers who build them and will fade as they retire and die. Imagine the construction of a house. There are the blueprints of what the house should look like, but also the knowledge of the carpenters, electricians, and plumbers that is important to the construction but not contained in the actual instructions. The plans for a house usually do not include instructions on how to pound a nail with a hammer, but if there is nobody alive who knows how to do this, then the blueprints themselves are useless until somebody re-learns that skill.

Perhaps more importantly, much of the technical documentation for the Saturn was developed by contractors and kept at their facilities, and not provided to NASA. When the contracts ended, they were under no obligation to keep materials unless the contract required it. This is the same with any government contract. The Air Force buys fighter planes and manuals for how to maintain and repair them, but does not also acquire the detailed documentation on how to build them. That stays with the contractor.

The Saturn had many contractors. The first and second stages were built by different contractors, and they did not share their documentation with each other. So when the program shut down, detailed technical documentation was spread around at various contractor facilities. NASA did not collect it all and save it.

Another issue is the tooling for the vehicles. Documents and blueprints are not the only things necessary for building a rocket. They also require tools, many of which are built exclusively for assembling the vehicle and many of which are large, such as jigs that can hold ten-meter-diameter fuel tanks. Those tools also have to be designed. When the contract is over, the tools take up space that can be devoted to other tasks. So the tools are either put into storage and later scrapped, or simply scrapped immediately. The tooling for the Saturn 5 was destroyed over three decades ago. If a complete set of technical documentation existed, the tooling to build it would have to be designed (more blueprints!) and then built.

Technology evolves, however. A Saturn 5 was not simply a piece of technology, or even many pieces of technologies, it was the product of many other technologies, many of which were evolving or becoming obsolete. For instance, welding evolved throughout the 1960s and continues to evolve today. Friction stir welding and laser welding did not exist in the 1960s yet are commonly used in aerospace manufacturing today. Similarly, aluminum-lithium alloys and carbon fiber composites are common materials today that were unavailable in the 1960s. Welding, milling, and bending machines are now computer-controlled and produce more accurate parts. Materials are cured and chemically treated in ways that had not been invented in the 1960s.

A Saturn 5 was not simply a piece of technology, or even many pieces of technologies, it was the product of many other technologies, many of which were evolving or becoming obsolete.

If NASA wanted to build a new Saturn 5 today, the agency would not want nor need the original blueprints. They would want to, and would have to, do things differently. They would want to develop computer-assisted drawings of the pieces, for starters. And they could build pieces lighter and stronger than in 1966. The plans, the blueprints that the agency “lost,” would not be all that useful in developing similar equipment using technology that has evolved and improved over four decades.

But NASA officials were also not stupid when they shut down the Saturn program. They realized that key parts of the vehicle were likely to be useful in the future and they made a strategic decision to preserve that technology. In particular, they preserved most important part of the Saturn 5, the engines.

The F-1A

The longstanding story that NASA lost or destroyed the Saturn 5 plans quickly falls to pieces when one learns about the F-1 Production Knowledge Retention Program. This was a project at Rocketdyne, the company that built the F-1 engine, to preserve as much technical documentation and knowledge about the engine as was possible. According to an inventory of records, this produced twenty volumes of material on topics such as the engine’s injector ring set, valves, engine assembly, and checkout and thermal insulation and electrical cables, among others.

But the project went beyond simply preserving documentation. Rocketdyne actually sought to preserve the knowledge inside the heads of the people who designed and manufactured the engines. They conducted tape-recorded interviews with them, asking about parts that were difficult to produce and manufacturing tricks that they had learned in the process of building multiple engines. In addition to all this material, NASA also had several F-1 engines in storage, plus the ones that have ended up in museums that could be disassembled and examined. Five engines were in storage at NASA’s Michoud Assembly Facility, with ten others mounted on stages on external display.

Rocketdyne delivered 98 production engines to NASA, of which 65 were launched. A total of 56 equivalent development engines were tested. The company conducted 2,771 production and R&D firing tests of single engines, 1,110 total full duration tests, and accumulated 239,124 seconds—over 66 hours—of engine firing experience. The five-engine cluster used on the Saturn 5 was fired at the Mississippi and Alabama test facilities 34 times, with 18 full duration tests for a total of 15,534 seconds of engine experience. Rocketdyne estimated in 1992 that the eight-year F-1 engine development program had cost $1.77 billion in FY91 dollars.

If NASA wanted an F-1, it could hire Rocketdyne—now part of United Technologies Corporation—to build it. Actually, what Rocketdyne would build is the F-1A. The F-1 engine was designed to produce 6.7 million newtons of thrust. Rocketdyne uprated the engine and increased the thrust to 8 million newtons at sea level, or about 8.9 million newtons in vacuum. The design changes that increased the engine power were successfully demonstrated on two engines.

NASA considered using the F-1 engine back in the late 1980s and early 1990s as part of its Space Exploration Initiative (SEI). Even when SEI was politically dead, the agency still continued what are called “trade studies,” meaning the trades between costs and capabilities of different choices. The F-1A was considered as part of these trade studies, and a senior study conducted in 1992 recommended that it be pressed back into service. But the cost of using the F-1A in a heavy-lift launch vehicle was high.

Each SSME would have less than a quarter the thrust of a single F-1A—for perhaps twice the cost. But one must add to that the nearly half billion dollar F-1A reactivation cost, which has to be amortized over the production run of the engines.

In 1992 Rocketdyne, then a part of Rockwell International, conducted a study of the company’s ability to place the F-1A back into production. The company surveyed its personnel with F-1 engine experience in three areas—engineering, quality, and manufacturing—and identified how many were active and how many were retired and available to work on the program if asked (the totals were 248 and 76, respectively). The company pointed out that both the Atlas and Delta engines had been out of production for many years and had been restarted. For instance, the Delta engines had been out of production from 1968 until 1989, when the company restarted the RS-27/27A production line.

But the F-1A would have been expensive. Rocketdyne estimated that activation of the production line would cost $315 million in 1991 dollars. A significant chunk of that money, $100 million, would be required to pay for four test engines and a spare. These costs apparently did not include reactivation of the special test stands that had been used for the F-1.

The per-unit cost for a production engine was difficult to estimate, however, because it depended upon the quantity ordered and the production rate. Rocketdyne estimated that the cost of each engine would be $15 million, assuming an order of 40 or more engines at a rate of 10–12 a year.

Taking these figures and adjusting them for inflation to 2005 dollars, the activation cost would be approximately $445 million and the per-engine cost for a large batch would be $21 million. Of course, these numbers should be taken with several grains of salt. Part of Rocketdyne’s cost estimate was based on its available workforce, and certainly there are far fewer people with F-1 engine experience today than there were fourteen years ago. The same might also be true for some facilities.

One problem in using the SSME in any future launch vehicle is that the production line for that rocket has to be reactivated too—no new SSME’s have been built in years because they are reused, not thrown away like most rocket engines. However, it is doubtful that this is a major cost for Rocketdyne, because the refurbishment of existing engines requires many of the same facilities and personnel as producing new rockets. But the SSME is still a very expensive engine. Rocketdyne has apparently estimated that it could build a “down-engineered” disposable version of the Space Shuttle Main Engine for $40 million a copy. Each SSME would have less than a quarter the thrust of a single F-1A—for perhaps twice the cost. But one must add to that the nearly half billion dollar F-1A reactivation cost, which has to be amortized over the production run of the engines. The cost of the RS-68 is not currently public information, but it is undoubtedly lower than the cost of the SSME, because it was designed from the start to be disposable.

What all of these back-of-the-envelope comparisons demonstrate is the limitations of back-of-the-envelope comparisons. Without better knowledge of Rocketdyne’s actual production costs, it is hard to draw firm conclusions about how an F-1A engine would compare to other options. But the comparison does highlight one important fact: of all three engines—the F-1A, the SSME, and the RS-68—the only one that is currently in production and requires no activation costs is the RS-68. More importantly, the RS-68 is used for a rocket, the Delta 4, which already has other customers. Increasing the RS-68 production line to power a new heavy-lift launch vehicle should result in lower per-unit costs for the engines than Rocketdyne already charges. This is undoubtedly one of the reasons that NASA is considering switching from the expensive SSME to the RS-68. But it’s too bad that the F-1A seems destined to remain shelved. Those who experienced the roar of the Saturn 5 have said that it is an unforgettable experience.