Join 3,382 readers in helping fund MetaFilter (Hide)


belters expanse trajectory: working up the Epstein Drive
April 14, 2013 11:11 PM   Subscribe

How NASA brought the monstrous F-1 "moon rocket" engine back to life - "The story of young engineers who resurrected an engine nearly twice their age."
The question, though, is whether or not the practical side of the equation can balance the romantic. The Advanced Booster competition will run through 2015, at which point a winner will be chosen, solid or liquid. The F-1B could be the engine sending astronauts to Mars—or it could wind up as one more Wikipedia footnote.
also btw...
-NASA's Space Launch System
-NASA's Space Launch System Advanced Booster
-Comparison of orbital rocket engines
-Comparison of orbital launch systems

oh and How NASA's Nuclear Rockets Will Take Us Way Beyond Mars
&c. Almost 40,000 People Have Signed up to Go to Mars—One Way
posted by kliuless (34 comments total) 34 users marked this as a favorite

 
Every time I think I understand something, something like this illustrates just how much I don't even have a solid grasp of the scale of things, much less anything else.

This is amazing. Particularly the lead Ars Technica article. They can put out some excellent writing & imagery when they bring attention to bear.
posted by CrystalDave at 11:20 PM on April 14, 2013


This is just fantastically written. Even when the subject exceeds my grasp, the writer has clearly worked hard to compose for a general audience. Science writing for a general audience, I miss you!
posted by mwhybark at 11:45 PM on April 14, 2013 [1 favorite]


This is amazing.

yea! while john baez comes down on the side of XMLicious...
"I know it's an amateurish question many people may have asked before, but: wouldn't it be better to spend the money on even better telescopes and robotic probes? What's the point of heaving a person into space to die? We seem to be struggling the keep earth's climate in check. Shouldn't we first deal with that before thinking about terraforming Mars?"

I basically agree. To get scientific knowledge, unmanned probes and telescopes are a much better use of money. The only really good reason for sending people to Mars is to start spreading intelligent life across the universe. I think we should wait until we develop intelligent beings that can enjoy the climate of Mars before going there... it's not like sending people to Australia, it's worse than sending people to Antarctica.

So, I don't think our governments should pay to send people to Mars now. They should spend money on unmanned probes, telescopes, AI, biotech, nanotech, and fighting global warming. But if people want to fund a Mars mission privately using a reality TV show, I don't see the point in arguing with them.
i'm starting to side with NdGT :P
posted by kliuless at 11:56 PM on April 14, 2013


"The gas generator spoke with a deep rumbling, topped with a rocket's crackle-crackle-crackle—a sound I'd always thought was just the microphone clipping when listening to recordings of rocket launches."
posted by mwhybark at 12:00 AM on April 15, 2013 [1 favorite]


Rocket surgery.
posted by bardic at 12:38 AM on April 15, 2013


Awesome!
posted by Kevin Street at 12:57 AM on April 15, 2013


It's a small point, but I hope they're going to reassemble the originals when they're done with them.
posted by InsertNiftyNameHere at 1:17 AM on April 15, 2013


Nice Expanse series nod with the post's title. Love those books.
posted by snwod at 1:41 AM on April 15, 2013 [2 favorites]


wouldn't it be better to spend the money on even better telescopes and robotic probes?

In the short term I agree. However, big rockets are useful to deploy telescopes and probes. More throw weight means a much bigger payload of probes delivered to your destination. Bigger is sometimes better.

A Saturn V used to be able to deliver ~15 metric tons to the moon. Fifteen tons of remote controlled mining equipment digging for water in Shackleton crater would be very exciting indeed.
posted by justsomebodythatyouusedtoknow at 2:10 AM on April 15, 2013


From: How NASA's Nuclear Rockets Will Take Us Way Beyond Mars

That's 4GW of power, equal to the total production capacity of Chernobyl—enough to power 3 million homes—generated in less than a quarter of an hour.

I think Gizmodo should stick to reviewing phones.
posted by atrazine at 4:03 AM on April 15, 2013 [6 favorites]


They should have tried to get this one up and running.
posted by grajohnt at 5:06 AM on April 15, 2013


These articles made me so happy. Great to hear that engineers are learning from those original engines.
posted by Brandon Blatcher at 5:29 AM on April 15, 2013


One little thing I caught from a slide is that they expect to reduce the parts count somewhat, in the new edition. How much? Well, the old engines took more than 5,000 parts. They claim they'll be able to build an almost identically functioning machine with less than a hundred.

Apparently, a lot of the complexity was building, essentially, a fluidic computer in the engine, one that would only allow each step in the ignition sequence to fire if the preceding one had succeeded. They didn't have computers, so they had to do it manually; if something didn't work, it had to physically prevent the next step from happening.

I mean, think about that a minute. The sheer complexity of that machine must be far beyond the ability of most of us to grasp. Those guys and gals had no computing power at all. But they still had to build IF/THEN clauses into their rocket engine, and by god, they figured it out, and went to the goddamn Moon on that engineering.

When people call the Apollo program one of mankind's most significant achievements, they are not exaggerating. The Egyptians built the Pyramids with primitive technology; America of the 1960s went to the Moon on steel, kerosene, and brainpower.
posted by Malor at 5:56 AM on April 15, 2013 [16 favorites]


I read these articles earlier today. This is the stuff that will lead to the next generation of engineers and explorers. Inspirational.
posted by michswiss at 6:06 AM on April 15, 2013


The amount of hand welding it what stunned me, along with the cases where the welders made mistakes, realized it and then fixed the mistake. They didn't start over from scratch! And these engines flung men at the Moon. My appreciation for the technical aspects of the program just grew incredibly larger.
posted by Brandon Blatcher at 6:06 AM on April 15, 2013


America of the 1960s went to the Moon on steel, kerosene, and brainpower.

Many of us in other countries would love to see more of this particular version of the US in action. It was hugely inspiring.
posted by Wolof at 6:17 AM on April 15, 2013 [7 favorites]


Malor: one of the dirty little not-really-a-secrets about the Apollo program?

Those computer screens at the consoles in the mission control center weren't actually computer screens. The computers used in the Apollo mission control system were late 1950s vintage mainframes. VDUs hadn't been invented when they were built; instead, they sent their output to dials and pen plotters, along with direct data from various sensors, all in a data center at some remove from mission control. NASA installed TV cameras facing the monitor instruments, connected them via coax to mission control, and sent the output to the screens in front of the mission control flight engineers.

So what looked like computer terminals in a too-sophisticated-for-1968 control room were actually 1940s kluges, cameras pointed at dials and counters.

(Rocket science was new and bleeding-edge when Wernher von Braun was pioneering it. Today? The computer screens in SpaceX's mission control are each driven by a computer that probably has more raw processing power than every mainframe in North America back during the Apollo program ...)
posted by cstross at 6:33 AM on April 15, 2013 [7 favorites]


I mean, think about that a minute. The sheer complexity of that machine must be far beyond the ability of most of us to grasp. Those guys and gals had no computing power at all. But they still had to build IF/THEN clauses into their rocket engine, and by god, they figured it out, and went to the goddamn Moon on that engineering.

They iterated the design. "Let's try this." BOOM. "Okay, no, this" RummmBOOM. "Okay, better."

When they first started testing the F-1, combustion instability would destroy the engine within seconds. When they were doing the final tests -- multiple burns longer than the mission burn would be, they would put small explosives in to attempt to cause instability, and the plume would self-damp the instability in under a tenth of a second. Basically, they found out what made that boom happen, built around that, and kept doing that until the engine would boom, at least, not for the mission duration.

But the cost? HIGH. All those parts cost. All the welding time cost the square of that. The development cost was huge. The big question is "can we leverage enough out of a 1950s design?" to make it worth our while to study and possibly redesign the F-1 into the F-1B?

I'm not sure if this is the right path. Yes, the F-1 is mighty, but the RD-171 is mighty as well -- 1.84Mlbf vs. 1.55Mlbf -- costs a fraction of the cost of the F-1, and has a much higher specific impulse. The current RD-180, as flown on the Atlas V, has over half the thrust of the F1, twice the burn time, and an even higher specific impulse, thanks to the staged-combustion rather than the gas generator design.*

And, of course, if you decide to go with LH2 rather than RP-1 as your fuel, then the F-1 has very little to help you with.

I'm just afraid we will be, in essence, completely redesigning the engine for modern use, and if that's the case, the right answer is to clean-sheet it. Between the advances we've made in manufacturing (most importantly, precision casting), metallurgy, and controls, I'm wondering if there really are any good lessons to be learned from the original F-1 design. Basically, you're going to need to take the original design, change every part, change every material, and probably double the burn time.

We should *learn* from it, yes. I'm just wondering how much we will actually learn about designing engines in the 2010s. As a historical project, I find it fascinating.

But I'm still very torn about the idea of building an F-1B. Unless, of course, they basically ignore the F-1 and just use the name to get support. It's not the first time something like that has happened -- the F/A-18E and F models have very little in common with the F/A-18A through D, including being about 20% larger. But if they called it by a new number, they figured Congress wouldn't approve. But this was new, this was a "upgrade" of the F-18.

Sure, if you can say"Well, I replaced the CPU, the mainboard, the power supply, all the drives, and then upgraded to a larger case...."






* The difference between combustion cycles. All big liquid fuel engines need massive amounts of fuel and oxidizer, so they all use turbine driven pumps (turbo pumps) to deliver it. The F-1 pumps can empty an olympic swimming pool in 15 seconds.

The simplest design is the gas generator. You basically have a mini-rocket engine, hooked to the turbo pumps. You take some fuel and oxidizer and burn it in that, that thrust drives the turbo pumps, which feed the engine. The advantage is that this is the simplest way to do it for big engines, which makes it cheaper and more reliable. The disadvantage is that you lose performance, because some of your fuel and oxidizer is exhausted outside of the combustion chamber, not adding thrust.

Staged Combustion uses a rich fuel mixture, which is burned in the turbine driving the turbo pumps. The exhaust is then fed into the combustion chamber, where it finishes combustion and goes out the same path as the main exhaust. The advantage is the efficiency increase, which is seen as a higher specific impulse. The problem is that rather than just exhausting the turbo pumps drivers, you have to pipe that very hot mixture into the combustion chamber, and the back pressure makes life very much harder for the turbine driving the turbo pumps. This was the big cost adder to the SSME/RS-25. It had to handle the extra stress of staged combustion *and* reuse.

An unflown variant is full flow staged combustion, where *all* the fuel and oxidizer flows through the turbo pump's turbine, and a small amount is swapped and combusted to drive the system. As a bonus, the rest is also turned into a gas, which burns faster. The theoretical advantages is cooling -- all that uncombusted mass cools the turbine, and the faster combustion meaning higher thrust. The Russian RD-270 was close to flight ready when it was canceled.

Most of these use regenerative cooling of the nozzle -- a small amount of the fuel is pumped around the nozzle, which cools the nozzle and preheats the fuel. The expander cycle takes this further by routing some fuel into channels around the nozzle, which turned it into a gas, then using that to drive the turbo pumps before finally being pumped into the engine. The open cycle version of this, the expander bleed cycle, throws the fuel driving the turbo pumps away, which reduces the back pressure. This makes for a thrust-to-efficency tradeoff, you get more thrust, but less specific impulse. The LE-5 on Japan's H-I and H-II rockets uses this.

And, since I've almost completed the series, the last, useful only for small engines, is the pressure cycle, where a gas (nitrogen or helium) pressurizes the fuel/oxidizer tanks and that pressure forces them into the engine.
posted by eriko at 7:23 AM on April 15, 2013 [9 favorites]


"The gas generator spoke with a deep rumbling, topped with a rocket's crackle-crackle-crackle—a sound I'd always thought was just the microphone clipping when listening to recordings of rocket launches."

That kind of jumped out at me as well. I saw a space shuttle launch a fewe years ago and I was surprised at how much the sound matched what I had seen on TV; I, too had assumed the crackling was some sort of artifact. What is different live is that you also feel the sound throughout your body, even from miles away. The other thing about seeing a launch in person is that during the several second delay between seeing the light and hearing the sound, you can actually see the sound waves coming toward you across the Indian River (where the causeway is).
posted by TedW at 7:27 AM on April 15, 2013


Tangentially related, later this week, rocket engines from the 1970s, originally intended for the Soviet moon program might finally get a chance to put something into orbit. Mind you, the engines have been refurbished and are not all original NK-33 parts, but the bulk of the engines being used are literally several decades old.

PWR actually licensed the rights and acquired the relevant information to be able to domestically manufacture the RD-180 engine used in the Atlas V. There was a pilot program to validate that they could manufacture key components. However, given the set-up costs to be able to begin production (probably $500m-$1b) and higher labor costs, it would be far more expensive than continuing to buy them in 'large' batches from Russia.
posted by resplendentoops at 7:55 AM on April 15, 2013


Many of us in other countries would love to see more of this particular version of the US in action. It was hugely inspiring.

+1000 from this US resident.
posted by scose at 8:28 AM on April 15, 2013 [1 favorite]


1962 NASA documentary on the Saturn Propulsion System; includes a full-scale test fire at the end.
posted by TedW at 8:58 AM on April 15, 2013


The real thing in action at 500 frames per second.
posted by TedW at 9:00 AM on April 15, 2013


The real thing in action at 500 frames per second.

That music is horrible. Here's the same with a voice over describing what's happening.
posted by Brandon Blatcher at 9:06 AM on April 15, 2013 [1 favorite]


Yes, please follow Brandon's link instead of mine!
posted by TedW at 9:08 AM on April 15, 2013


Well, your link has some nice/interesting annotations, which Brandon's link does not. For those of us watching without sound, the music/voiceover doesn't matter.
posted by Steely-eyed Missile Man at 9:11 AM on April 15, 2013


America of the 1960s went to the Moon on steel, kerosene, and brainpower.

What I try to keep in mind, is that they did the engineering on slide rules with one or two digits after the decimal place. Then they had a draftsman draw it up, and THEN someone welded it together.

That. And never underestimate the power of cigarettes and benezrine.
posted by mikelieman at 9:37 AM on April 15, 2013 [1 favorite]


I think this link belongs here. (I was going to make an FPP on it, some day, but this post is a good excuse to get it linked right now.)

IGNITION! An informal history of liquid rocket propellants, by John D. Clark.

I just ran across it a few days ago, but it looks great. Some 200 page on rocket fuels from Von Braun to the 1970s. Here's an excerpt to get a flavor for his writing.

He starts off with two photos illustrating
"This is what a test firing should look like. Note the mach diamonds in the exhaust stream."
versus
"And this is what it may look like if something goes wrong.".
A scanned .pdf is here. (Not sure if that link has a right to exist, but the book is out-of-print, and used copies cost $375 on amazon.com, so ....)
posted by benito.strauss at 10:05 AM on April 15, 2013 [1 favorite]


For those who dig technical manuals - here are some your browsing pleasure:

Saturn V flight manual [18MB PDF]

Rocketdyne F-1 Engine Familiarization Training Manual [16MB PDF]

Good pics and a lot more to get lost in at the Rocketdyne Archives
posted by chambers at 11:08 AM on April 15, 2013 [1 favorite]


Benito.strauss, I have linked to that .pdf a few times over the years and it is still around, so apparently no one is too concerned about it ( I did download a copy for safekeeping, though). I hunted it down after following the link in this comment. That is where I learned the term hypergolic (if eriko has any insight there I would love to hear it), not to mention that the forward was written by Isaac Asimov and is worth a read on its own. Really gives a good feel for the early days of rocketry. The fact that early rocket fuels and uranium enrichment contributed greatly to the use of fluorine as an industrial chemical is also important to me personally and professionally, but I am starting to veer off topic.
posted by TedW at 2:52 PM on April 15, 2013


Heh, I also ended up there via Derek Lowe. And the excerpt I just happened to use as an example in my comment details the origin of the word "hypergolic' (great word).

I'm thinking of trying to get it printed off, as it looks like a great reading-while-on-public-transit book and I don't have a Kindle.
posted by benito.strauss at 3:21 PM on April 15, 2013


Ignition is a great book. There's a great bit where he discusses the early days of Red Fuming Nitric Acid as an oxidiser and how they found out that you can store it safely by adding HF to it. When you're *adding* hydrofluoric acid to something for safety reasons, you know you're in exciting territory.
posted by atrazine at 4:55 AM on April 16, 2013 [3 favorites]


"Red Fuming Nitric Acid" sounds like one of Captain Haddock's curses in Tintin.
posted by benito.strauss at 1:30 PM on April 16, 2013


They didn't have computers, so they had to do it manually; if something didn't work, it had to physically prevent the next step from happening.

Or they didn't trust the computers they had. Which was definitely wise then and would probably be wise now. You want the shortest logic possible from sensing to output. A computer is small but has a very long chain if you include all the electronics and software, wires, sensors, motors/actuators, etc. Plus you may be relying on inferences about the world vs actual physics that prevents bad things from happening.

tldr: KISS
posted by DU at 6:04 AM on April 17, 2013


« Older What would a poverty map of India look like?...  |  We have made the act of killin... Newer »


This thread has been archived and is closed to new comments