Why not just bring back the F-1 and J-2?

[dwight_looi]

In retrospect, it appears that the last 40 years has seen NASA, the USAF and private US companies subcontracting for them running around in circles trying to find more expensive and unsafe ways of accessing space. The Shuttle is the most expensive method of getting a pound of anything into orbit and it has the worst safety record of any man rated launch vehicle. The Atlas V and Delta IV offered nothing that a good old Saturn IB didn't. The ARES I again was another 15~20 tonner with nothing new to offer other than stability problems and resonance issues with the extended length SRB. Even the latest kid on the block -- the Falcon 9 -- is basically rehashing the medium lift option albeit using nine very small engines.

It seems that NASA and the US government is more interested in preserving aerospace jobs in existing and unnecessary programs than finding the best and most efficient means of flying orbital missions. I mean, let's turn the clock back to 1970. Let's say the EELVs didn't exist, the Shuttle didn't exist, SpaceX didn't exist and ARES wasn't even thought of...

Can you imagine what we could have done and how many more missions we could have flown if we simply rolled out a simple 2-stage vehicle using a single F-1 engine on the first stage and a single J-2 engine on the upper stage? That's a 20-ton to LEO Proton class vehicle. It'll have all manrated engines with extensive flight histories. We could have launched the Hubble with it. We could have launched Chandra with it. We certainly could have pieced together the ISS with it. There wouldn't have been a need for the Delta IV, the Atlas V or the Falcon 9 for the most parts. There wouldn't have been a need for the RS-68 engine or the license built Russian RD-180. Most of the shuttle payloads would have flown on a 20 ton vehicle. If you really want, you can probably strap three side by side for 50~60 tons of LEO capacity.

 

[bdewoody]

The safety record of the Space shuttle is about the same as the Russian soyuz since as I recall they have lost at least 2 crews that they admit to.

 

[jimoutofthebox]

I couldn't agree more. NASA has spent enough money re-inventing the wheel in the last 30 years to build a base on the moon. I think we could use a core booster with two J-2 engines and 4 strap ons based on the falcon 9 or a booster using the F-1 to get 55 tons to LEO. A centaur booster could then send a payload to the moon. We would need to launch two payloads (one for the command and service module and one for the lunar lander). They would link up in lunar orbit to transfer the astronauts to the moon and back. This plan would not need a lot of R&D to do.

 

[stevekk]

Well, if they complete the J-2X update, then why not do a F-1 update also. How much extra lift do we get going from the J-2 to J-2X ? We should be able to get a comparable increase with a F-1X.

 

[jimoutofthebox]

Well, if they complete the J-2X update, then why not do a F-1 update also. How much extra lift do we get going from the J-2 to J-2X ? We should be able to get a comparable increase with a F-1X.

The J-2X update is an example of the problem. NASA spent $1.6 billion to tweak an existing design and they still don't have an engine to show for it. We need to stop tweaking, developing, and reengineering and just build the rockets based on existing technologies. What NASA spends most of its money on is little more than intellectual masturbation.

 

[stevekk]

Well, if they complete the J-2X update, then why not do a F-1 update also. How much extra lift do we get going from the J-2 to J-2X ? We should be able to get a comparable increase with a F-1X.

The J-2X update is an example of the problem. NASA spent $1.6 billion to tweak an existing design and they still don't have an engine to show for it. We need to stop tweaking, developing, and reengineering and just build the rockets based on existing technologies. What NASA spends most of its money on is little more than intellectual masturbation.

I just looked up the J-2X engine. It's time to change the name. This new beast weighs 1000 lbs more than the original J-2. What were they thinking ??

 

[oldAtlas_Eguy]

The J-2X unlike the J-2 will be a variable thrust engine. The added weight is for the new turbo pump and changes to the thrust chamber to make it stable through the variable thrust range. The J-2X is not actually a J-2 engine its really a brand new engine that shares some engineering with the J-2. The F-1 is not the engine to use either. It’s also not a variable thrust engine. Most modern vehicle design goes for variable thrust engines to be able to manage loads at the end of burn by dialing back the thrust. This makes it possible to lower the static vehicle weight improving performance overall. Launch vehicle design is a matter of tradeoffs.

Actually if you looking to revive the F-1 use the F-1A which was designed and fully tested and ready for flight that has an improved thrust of 1.7+ million lbf over the F-1 with only 1.5 lbf.

Space X Merlin 2 sounds a lot like an F-1A, same thrust level and basic thrust chamber design, but has a variable thrust turbo pump. It’s this turbo pump that Space X has estimated will cost $1 Billion to develop and test.

 

[dwight_looi]

Really, does efficiency matter? I mean really. Let's say the J-2X is better and it allows 23 tons instead of 20 tons to LEO with the same launch mass. Does that matter? It's still the same class of vehicles. Burn a little more fuel, carry a little less weight into space. So what? If it's me, I'll design the mission around the lift capacity rather than the other way around. 1.6 Billion is 12~18 extra flights. Plus, a less complex engine with more flight history is probably the cheaper and safer engine.

The SSME is a more efficient engine. The R-180 too and so is the RS-68. But, think about it. If we take all the money spent on the Atlas V, the Delta IV and Shuttle vehicles. All the money spent on the RS-68, SSME, Shuttle SRB, and all the abortive programs like the RL-60 and RS-84. All the money spent on the ground infrastructure of the aforementioned. If we had taken that money and simply flew a single 20 ton F-1 + J-2 vehicle for everything in the past 40 years. How many missions to mars is that? How many manned orbital flights is that? How many space station modules is that? How many solar system probes is that? Well... it'll be abut 50~75 missions a year for 40 years; we would have flown about 2000~3000 missions with that money.

With that kind of flight history and the economies of scale, we would probably have eaten the Ariane's lunch in the commercial launch arena too -- especially if after 10 or 20 years NASA simply buys and license the F-1, J-2 and vehicle blueprints to any US company who wants to build the launch vehicle for $1 to create a commercial, firm fixed price bidding environment on launch services.

 

[oldAtlas_Eguy]

Really, does efficiency matter? I mean really. Let's say the J-2X is better and it allows 23 tons instead of 20 tons to LEO with the same launch mass. Does that matter? It's still the same class of vehicles. Burn a little more fuel, carry a little less weight into space. So what? If it's me, I'll design the mission around the lift capacity rather than the other way around. 1.6 Billion is 12~18 extra flights. Plus, a less complex engine with more flight history is probably the cheaper and safer engine.

The SSME is a more efficient engine. The R-180 too and so is the RS-68. But, think about it. If we take all the money spent on the Atlas V, the Delta IV and Shuttle vehicles. All the money spent on the RS-68, SSME, Shuttle SRB, and all the abortive programs like the RL-60 and RS-84. All the money spent on the ground infrastructure of the aforementioned. If we had taken that money and simply flew a single 20 ton F-1 + J-2 vehicle for everything in the past 40 years. How many missions to mars is that? How many manned orbital flights is that? How many space station modules is that? How many solar system probes is that? Well... it'll be abut 50~75 missions a year for 40 years; we would have flown about 2000~3000 missions with that money.

With that kind of flight history and the economies of scale, we would probably have eaten the Ariane's lunch in the commercial launch arena too -- especially if after 10 or 20 years NASA simply buys and license the F-1, J-2 and vehicle blueprints to any US company who wants to build the launch vehicle for $1 to create a commercial, firm fixed price bidding environment on launch services.

There are some interesting things going on when the government purchases development of an item from a contractor. The government only has limited intellectual rights to the item. The company has unlimited rights because it is the original designer. So for a commercial company to use an F-1 or J-2 engine all they have to do is contract with Rocketdyne to build one. With it having been greater than 20 years the patents have entered the public domain but the designs are still protected as part of Rocketdyne trade secrets.

As I stated earlier, rocket design is driven by what tradeoffs a designer is willing to accept. If higher acceleration loads is ok, then variable thrust engines is not necessary. If pushing the envelope in engine performance is not desired to get as much performance as possible then older engine designs would work. In fact this is the design philosophies the Russians have used. Instead of variable thrust engines, they used many smaller engines and shut a few of them down to reduce the acceleration to reduce the loads. They accepted less performance trading it for more reliability.

Applying these philosophies can definitely lower design costs and mitigate some operational costs as well. Space X followed some of these and lowered their design and operational costs.

 

[Valcan]

Dwight,

One of the things your forgetting thought is that rocket engines fail. With just one engine when it goes out you loses hundreds of millions if not billions of dollars if your lucky. If there are crew on board.................

One of the good things about falcon 9 and falcon 9H is that they can have a engine loss and still reach orbit. This also means its safer and probably cheaper in the long run.

The falcon series all uses the same engine. That means these engines can be mass produced which drives down cost per launch.

Also they have shown plans to develope the falconX and Falcon XX a heavy design with a larger more powerful rocket. These could also be used on the same basic vehicle as falcon 9. However they again would not have the ability to attain orbit after a engine outage sense like you wished there would only be 3.

 

[dwight_looi]

Is the Falcon a safer ride into orbit than an F1/J2, Delta IV or Atlas V? In some ways it is, like having engine out capability over 90% of the 1st stage burn. But more engines also means a higher frequency of singular failures affecting the mission. In any case, engine out isn't the only failure mode or the most common one. Having multiple engines won't save you from ruptured propellant lines, structural failures or staging anomalies. Honestly, I don't think SpaceX went with nine engines because it is safer or cheaper. They went with nine engines because they have the Merlin and they didn't (yet) have a million pound class motor. Clustering what they have is the most expedient and lowest risk way of getting a medium lift vehicle into service. Already, they are soliciting funding for a million pound class motor and have brandished slides showing a Falcon 9 sized vehicle with a single 1st stage motor.

Regardless... this wasn't just about the Falcon 9. This is about the fact that -- I believe -- we would have been better off with a single F1/J2 vehicle over the Shuttle, Atlas, Delta, Titan and Falcon. It's about not the chasing after of incremental little tweaks and sticking to a standardized vehicle built on proven technology, and developing new stuff only when you have to. It's about using money to fly missions instead of reinventing 5 different ways to bring 10~24 tons into orbit. Or, alternatively using the money to make the next big leap like going to Mars or getting to 1/10th the speed of light so we can send a probe to Alpha Centauri in 50 years or Gliese 581 in 200. Really... 50 years is something we can live with. If the mission left when Apollo was in business, we'll be almost there.

 

[oldAtlas_Eguy]

If Space X decided to just purchase F-1A’s which are already developed just requiring a recertification of production, they could bypass a great deal of development costs. I wonder why they have not gone in that direction. There must be some overriding reason.

I think I found the reason: cost. With other problems such as training up a new production team in the methods to produce all of the parts for the engine of $500+ million (2010 dollars), which also may increase the costs further. At ~$25 Million (2010 dollars) or more per engine of production costs the F-1/F-1A may be too expensive. That would be for 50 to 100 engines (the number used in 5 years) total costs of $30 to $35 million. The Merlin 1 looks like it costs a total of $2 million per engine ($ 1 mil to produce and $1 mil toward payoff of development). Putting the total cost for a first stage engine of F-1 size to be no more than $18 million ($8 mil to produce and $10 mil toward payoff of development) to maintain the launch price of $56 Million. With an F-1A it would be $75 to $88 million.

http://www.thespacereview.com/article/588/1

 

[dwight_looi]

SpaceX likes to develop their own engines. I mean, they could have licensed the NK33 or something like some of the fizzled upstarts. But they decided that they can and should do their own to bring a critical core competence in house. The thing about the Merlin is that it is small enough to manufacture out of standard CnC mills and the like without expensive tooling or tricky welding procedures. The Merlin also went with the pragmatic approach of keeping the engine simple rather than trying to squeeze out the last few seconds in specific impulse. For instance, the Merlin dumps the gas generator exhaust overboard instead of discharging it into the periphery of the nozzle bell like the F-1.

Having said that the general consensus seems to be that one big engine is cheaper than a lot of small ones. The RS-68 and RD-180 were created on that premise. One possibility is that SpaceX can license the F-1 and simplify it. For instance, they can chop the nozzle off at the level where the gas generator exhaust dumps into it. A smaller nozzle is less efficient, but also lighter. The elimination of the complex gas generator also makes the engine cheaper. You'll probably lose 5~10 seconds of impulse at higher altitudes, but that is unlikely to be a deal killer. The Merlin II (or perhaps they'll call it the Gryphon) may be just that, but it's too early to try to judge which direction they'll take.

 

[Valcan]

.

The merlin is cheap because it can be mass produced and simply refined over time this lowers cost and engineering problems in future launch systems. Also i believe they are supposed to be reusable they just dont have a reason to reuse them until they have more buyers.

 

[oldAtlas_Eguy]

The F-1 demonstrated long life capability, several thousand seconds of operation without rework. So It is actually a reusable design though that was not its original intent. Its a byproduct of a stable and reliable engine. The big difference between the F-1 and Merlin and Merlin 2 engines is the injectors. The Merlin's use a pintel injector and the F-1 does not. The pintel is easy and cheap to manufacture and is non-critical in the control of its dimensions whereas the the injectors on the F-1 are critical in deminsions making them difficult to manufacture and expensive. There are now easier ways to do regenerative cooling bell manufacture than what was used on the F-1. Its all these new design items that lower the cost of manufacture that makes the F-1 obsolete.

 

[dwight_looi]

I wouldn't call the F-1 obsolete. Something becomes obsolete when it is replaced by something else of equal or superior capability or economy (or both). Right now, the US does not have a million pound class hydrocarbon engine. Until we develop one there is nothing to obsolete the F-1 with.

The manufacturing of the F-1 may also be easier today than in the 60s. The shower head injector needs precisely placed holes of precise dimensions. Back then they didn't have CnC mills and it's both labor intensive and tricky on the QA side of things. Today, CnC mills are everywhere. Heck you can even buy a desktop unit to play with for $3000.

Having said that, pintles are not the exclusive feature of the Merlin. In fact, the Merlin is not even the biggest pintle engine. The NK33 (aka AJ26 when license built for the upcoming Taurus II) is a pintle injector design.

 

[vulture4]

The F-1 design has been supplanted by the RS-84. http://www.astronautix.com/engines/rs84.htm which was developed by Rocketdyne and then not produced. SpaceX has licensed the design and has indicated that it may be used to replace the nine Merlins in future versions of the Falcon. Whether a new version of the J-2 is needed depends on whether the upper stage uses hydrogen. Generally the multiple-pintle design is lighter but somewhat more complex to manufacture than the single-pintle design, but either can provide adequate performance. The RS-84 I believe uses multiple pintles and staged combustion rather than the gas-generator cycle of the F-1, but in other respects it is designed for manufacturability with channel-wall construction replacing the brazed tubes of the F-1 and J-2.

What's really impressive is that SpaceX didn't simply ask to buy the engines from PWR the way Boeing would; SpaceX believes they can do a better job building it in-house. When was the last time a single company made both airframe and engines?

 

[oldAtlas_Eguy]

Now I understand why SPace X did not go with the F-1, they already hadc a line on a much cheaper engine.

 

[dwight_looi]

Something like this?

 

[vulture4]

Hey, that looks great. So why are we spending at least $10B taxpayer dollars to design the already-obsolete Ares V which was mandated by Mike Griffin to use only ATK SRBs and expensive tube-wall liquid engines?

 

[stevekk]

Hey, that looks great. So why are we spending at least $10B taxpayer dollars to design the already-obsolete Ares V which was mandated by Mike Griffin to use only ATK SRBs and expensive tube-wall liquid engines?

Because the SRBs were a relatively proven design that was capable of the lift necessary to get the Orion capsule off the ground.

Falcon 1 and Falcon 9 can't. SpaceX might have a new rocket and motor design on paper, you can see my the dates of the SpaceX powerpoint, that they still won't have the necessary lift capability until later in this decade. They say 2015 for their "Eagle 1", but given that they were a couple of years late with the Falcon 9 and the "new/improved" Falcon 1e hasn't made it's first flight yet, I wouldn't count on them coming within 2 years of any of these development schedules.

You guys say you want to get out of LEO, but that's not happening with the Dragon capsule.

 

[dwight_looi]

Well, I am not sure the Ares V is still alive...

But... it is definitely the wrong idea. Here's why a rocket like the "Eagle 7" above is better.

• This is an all hydrocarbon rocket (no LH2 fuel to keep cold, no flaky insulation, no need to quickly launch after fueling)
• While it may have an inferior payload mass fraction, it carries just as much and the rocket is not actually larger it is just heavier (because kerosene is denser).
• There are only two stages vs three in the Ares V (when you count the SRBs)
• Both stages use the same Gryphon engine (just different numbers of it)
• This is a mono-diameter design which tends to be more mass efficient.

 

[oldAtlas_Eguy]

I did a cost estimation for the cumulative development to get to a Falcon XX which includes structural design, engine design, and flight tests. The cumulative total to get to the Falcon XX a 140MT LEO lifter was $3.4 billion. This even if it is off by a factor of 2 which is possible but I don’t think it will be more than that, it is still incredibly inexpensive when compared to the Ares V expenditures of $10 to $20 Billion.

Also here is some links that are fascinating if you are technically inclined about the F-1 and J-2 original engine developments. Also some data on SpaceX engine development.

http://history.nasa.gov/monograph45.pdf

http://images.spaceref.com/news/2010/SpaceX_Propulsion.pdf

 

[vulture3]

The SRBs are justifiable as long as the Shuttle is in service, since they are part of an operational design. But they have unmanageable failure modes, as can be seen in the explosion of a Delta-II a few years ago. More important, they have extremely high processing costs, requiring many hours of hazardous operations and their enormous weight requires vertical assembly and a vast and expensive launch infrastructure. Reuse of solid fueled rockets has never been economical. SRBs were chosen for the Shuttle because of their supposedly lower development cost. But their cost of operation was simply not understood, because no prototype system was flown before the design was chosen. To use large segmented SRBs in any future design would be a mistake.

The current HLV design, mandated by Congress, will, unfortunately, be a failure since it is much more expensive to develop and operate and intrinsically less reliable than the competing all-liquid HLV proposals from SpaceX and Boeing. Consequently any further funds spent on the "Ares V" or similar SRB-based HLVs is money wasted.

 

[oldAtlas_Eguy]

The SRBs are justifiable as long as the Shuttle is in service, since they are part of an operational design. But they have unmanageable failure modes, as can be seen in the explosion of a Delta-II a few years ago. More important, they have extremely high processing costs, requiring many hours of hazardous operations and their enormous weight requires vertical assembly and a vast and expensive launch infrastructure. Reuse of solid fueled rockets has never been economical. SRBs were chosen for the Shuttle because of their supposedly lower development cost. But their cost of operation was simply not understood, because no prototype system was flown before the design was chosen. To use large segmented SRBs in any future design would be a mistake.

The current HLV design, mandated by Congress, will, unfortunately, be a failure since it is much more expensive to develop and operate and intrinsically less reliable than the competing all-liquid HLV proposals from SpaceX and Boeing. Consequently any further funds spent on the "Ares V" or similar SRB-based HLVs is money wasted.

The Titan IV had stackable SRB’s just not anywhere as big as the Shuttle SRB’s. In Fact the reason NASA thought they knew enough to go with SRB’s is that the AF tested Shuttle sized SRB’s during the Apollo years. But those were all single use not the reusable ones the Shuttle design went with.