As I recall, Lindbergh got the money to pay for the Spirit from a bunch of St. Louis bankers and businessmen. I think he chipped in $2000-3000 of his money. All told they put together $25 large. I also recall they needed a little more but I forget how much so let’s go with $25,000.
How much is that $25,000 in 2010 dollars?
$315,661
Not much towards the $130 million needed but then circling the moon is a much harder deal. It’s *possible* one could have an airplane that can fly NY to Paris for $316 large, but not a sure thing. A Spirit replica with a modern engine and 1927 instruments would probably cost a lot more than $315G.
Okay, so (for the not-rocket-scientists) what he’s saying is that the Dragon capsule has (almost) enough onboard thrust to get one man to Lunar orbit and back? Like – “take out the extra seats” close?
$130 million? Only a sufficiently motivated lottery winner away.
I agree with the underlying premise that such a stunt would go a long ways to put the lie to space travel being “NASA-hard.”
Of course, an accident could have the opposite effect; but then again so does not trying. I don’t see a downside here, and there’s plenty of upside potential for the company whose technology could pull this off.
Do you actually need to “enter orbit” at the moon? A quick sling around the backside without entering a stable orbit would seem possible and just as “press-worthy” with a substantial fuel savings. Yes?
The article talks about how it was just the Lone Eagle and his trusty airplane:
“He didn’t have a co-pilot. He didn’t have an army of engineers monitoring his plane or a flight surgeon monitoring his heart rate, it was just him and his trusty single-engine monoplane, the Spirit of St. Louis.”
And the implication, to me at least, is that it can be done the same way today and therefore more cheaply than we might imagine.
However you aren’t going to be able to dispense with the ‘army of engineers’ to monitor the spacecraft:
One thing to remember is that while the Spirit of St. Louis was a pretty advanced airplane for it’s time, it was still simple enough for one person to thoroughly understand it and figure out what the cause of any in-flight problem might be, and what effective remedies might be available, select one and implement it. Navigation was simple enough for the pilot to handle and since there was no GPS, no infrastructure was needed.
This will not be true of the circum-lunar flight. The systems will be too complex for one crew to thoroughly understand. They will need ground based support and this costs money.
What ground-based support will be required for navigation is worth some thought. We have much more computer power in smaller lighter packages than the Apollo days but I don’t know if it’s enough to completely manage the task. Plus ground based radar was used to locate the spacecraft in the Apollo days.
Lastly, one would probably need the Deep Space Network or TDRSS or some gubbmint-controlled communications network to stay in contact with the vehicle at all times (to supply that ground based anomaly support).
So while I like the idea and the analysis in the article seems reasonable, I don’t think it tells the entire story.
Gregg raises some good points. I have not seen any information on the size of the mission support staff SpaceX uses for Dragon missions. If it’s like everything else I’ve seen from their company, it’ll be very lean by NASA standards. Part of the determination depends on the degree of autonomy built into the Dragon capsules. Back in the Apollo era, the on-board computers were primarily used for navigation. All of the vehicle systems were monitored from the ground (and in the vehicle) but all commands were executed by the crew. Today’s computers are orders of magnitude more capable than their Apollo counterparts especially at monitoring systems. Likewise, navigation systems today are both more capable and much smaller/lighter than those available for Apollo. The Apollo computer alone weighed 70 pounds. It’s likely the whole navigation system with the computer was well north of 200 pounds. By contrast, modern technology like AHRS, laser ring gyros, and good star sensors coupled with a good computer could likely do the same job and weigh less than 20 pounds.
For communications and precision navigation, they’d need to either buy expensive time on the Deep Space Network or come up with a private solution.
As for flying non-stop from New York to Paris, back in 2002 Lindbergh’s grandson make the flight in an essentially stock private plane with some additional ferry tanks fitted.
However you aren’t going to be able to dispense with the ‘army of engineers’ to monitor the spacecraft:
This will not be true of the circum-lunar flight. The systems will be too complex for one crew to thoroughly understand. They will need ground based support and this costs money.
A modern airliner is just as complex — in fact, more so. They don’t have a mission control. Some clueless types at NASA actually suggested developing a “mission control for airliners,” a few years back. The airlines simply smiled and ignored them. The aircraft is flown entirely from the cockpit, and the number of guys in the cockpit has gone from four or five to two.
You overglorify the “army of engineers.” In the television version of “Failure is Not an Option,” Gene Krantz talked about the first spacewalk and the feeling of helplessness since there was nothing mission control could do if the astronaut ran into a problem.
In typical Krantz fashion, he failed to note the irony that NASA was paying two dozen guys to sit in front of consoles and watch teevee even though their boss said there was nothing they could do to help.
I remember Max Hunter saying that he used to launch Thor missiles with a crew of eight, none of whom ranked higher than sergeant — and he designed it to be launched in the field with even smaller crews. When the Air Force turned it into Delta, they had over 20 people doing the same job, and when NASA started launching Deltas, they had over 100. Compare the ground crew for DC-X or SpaceShip One to the number of people NASA uses to launch a rocket.
What ground-based support will be required for navigation is worth some thought. We have much more computer power in smaller lighter packages than the Apollo days but I don’t know if it’s enough to completely manage the task.
Not even close. A short time ago, one gigaflops (1 billion floating-point operations per second) was considered the definition of a supercomputer. The holy grail of supercomputing was one terraflops — 1000x that. Today, video gamers have that much power in their graphic cards. Even cellphone have more computing power than supercomputers of not that long ago — vastly more power than the guys who designed Apollo ever dreamed of — in a package that weighs about five ounces with multiple sensors, wireless telemetry, etc. built in.
That’s what happens when you have an environment that allows competition, innovation, and incremental development. If we ran computing the way we ran the space program, Mike Griffin would be telling us we need to build another vacuum-tube machine because “it’s the only thing we know that works.”
For communications and precision navigation, they’d need to either buy expensive time on the Deep Space Network or come up with a private solution.
Commercial communication satellites have done lunar-swingby maneuvers to get to GEO, and I don’t recall their renting time on the DSN.
The United States Navy was successfully bouncing signals off the Moon for radio communications in the 1950′s.
The Moon is not deep space. Apollo did not use the Deep Space Network, it used the Manned Space Network. (An exception was during the Apollo 13 emergency, when they needed to save power and required the bigger antennas of the DSN.)
Well, if we’re talking about something with the complexity of a modern airliner and dealing with three-dimensional astrogation problems, I can see needing a small team on the ground to help with the maths and provide a second set of eyes on the gauges to make sure the pilot doesn’t miss anything. Lindberg himself was pretty occupied with just flying the Spirit, and I seem to recall the Mercury astronauts also had some workload issues. It would probably also be useful to have some kind of ground-based check on the spacecraft’s position as well.
With decent automation it’s probably going to be more like a “squad” of engineers than an “army” supporting the mission, though.
[And yes, you could get the same effect with extra crew members onboard the spacecraft and possibly dispense with ground control entirely. I'm assuming for the sake of argument that this pilot mission is limited to one crew member to reduce dry mass]
The origional idea for a one-man mission draws a parallel to Lindbergh but, as Frank Glover suggests, why limit yourself to one if refueling in orbit is possible? Or, dock with a small ‘earth-departure’ stage once in orbit.
The United States Navy was successfully bouncing signals off the Moon for radio communications in the 1950′s.
Amateurs routinely bounce signals off the moon today. I’d go as far as to say that there’s plenty of private entities who could do deep space communication with a Mars expedition (which is one to two orders of magnitude further away).
The $25,000 was the prize, if I recalled they only raised spent around $15,000 to win it.
The Wapedia gives the cost of the Spirit of St. Louis at only $10,000, which is about right since is was based on the Ryan M-2 mail plane that was already in service
Actually, according to Lindy himself the cost of the plane as delivered was $10,580+ the cost of extras. I quote his book:
“That will make $10,580, with a J-5 engine – special equipment extra at cost.”
In addition, Lindbergh wrote be bought 500 gallons of special fuel which developed a little more power and had it shipped form California to NY. I don’t know what av gas cost per gallon back then.
And it was definitely not a warmed over M-2. I quote again [ Donald Hall - the chief designer speaking]:
” Now we can’t use the standard Ryan fuselage, ” he says. “Also the wind span will have to be considerably increased so as to reduce the wing loading for take off and increase the aspect ratio for range. That means we’ll have to move the tail surfaces aft to maintain satisfactory stability and control. And that means the engine will have to be moved forward. When it comes right down to it, I’ve really got to design a completely new fuselage structure to meet your requirements. We will have to design a different type of landing gear while we’re about it,” he continues. “The M-2 type of gear would be too heavy when it’s designed to go with the longer wing span and to take the load you’re going to carry….”
“A modern airliner is just as complex — in fact, more so. They don’t have a mission control. Some clueless types at NASA actually suggested developing a “mission control for airliners,” a few years back. The airlines simply smiled and ignored them. The aircraft is flown entirely from the cockpit, and the number of guys in the cockpit has gone from four or five to two.
You overglorify the “army of engineers.” ”
As it turns out, today airliners actually radio all kinds of information about the status and condition of the aircraft without the pilots’ contribution at all. It’s done automatically. This allows maintenance to be ready at the aircraft’s destination.
And if you think the airline and other airplane pilots do not use ground based support when problems occur you are sadly mistaken. I, as a pilot, have gotten all sorts of recommendations from ATC when problems occurred. As PIC it was ultimately my decision as to what to do.
And I’m flying planes not much more complex than the Spirit. Some (like a J-3 Cub) less complex.
Not all the time of course – there are plenty of emergencies where there’s no time for long deep conversations.
But even when Sully ditched in the Hudson, they were communicating with ATC and ATC made recommendations which the PIC elected to not take.
And I’m not over-glorifying anything. I’m simply saying you will have to have ground based support to analyze and/or simulate problems, that might occur, and work out solutions. I said nothing about Krantz nor anything about the “glory” days of Apollo.
You’ve made the grave, grave error of overlaying on my remarks someone else’s deification of the Apollo days.
“Gene Krantz talked about the first spacewalk and the feeling of helplessness since there was nothing mission control could do if the astronaut ran into a problem. ”
And the farther out we go, the more that speed of light delays will mandate autonomy by crews (and/or, someday, advanced AI). Best get used to it now.
The planetary science people work with this by planning (and often executing) their moves very slowly and carefully, especially with surface rovers. Helpless is not being able to do anything about what you’re seeing in nearly real-time. Worse than helpless is knowing that whatever the outcome of it will be, many light-minutes away, has *already* happened…
How certain/uncertain is the estimate of the Draco thrusters performance?
Also, not sure how the life-support resources scale, but I wonder if two people could go, if they were very lightweight (but still healthy adults). I’m thinking of Dick Rutan half-joking that Jeana Yeager cut off her ponytail before their round-the-world flight to gain an extra four miles worth of fuel.
Yes, I really didn’t even bother talking about the difficulty of plotting a TLI maneuver or LOI, etc. because it’s not hard. In terms of computing power, I expect these days you could probably get an iPhone app that would do the job.
If you’re interested in seeing how easy it is to do the calculations that took an army of “computers” back in the days of Apollo, just go download one of the many free software packages that are available.. there’s even free versions of the software the big boys use now.
Al, yes it would be press worthy. Anyone doing anything in space is press worthy. The reason I chose the Apollo 8 profile over the Apollo 13 profile is that free-return-trajectory is just too damn easy. It’s often described as 3.2km/s of delta-v (because you have to course correct a few m/s along the way), which means your vehicle could be as much as 3669kg dry for a single Falcon 9 launch. I estimate the dry mass of a stock standard Dragon at 2296kg.. you could take 7 people on that joy ride and still have mass for luggage.
rickl, thanks, that was my goal.
On the other end of the spectrum, I’m trying to convince myself that a 511kg dry mass design for a single person lunar lander doesn’t exist. 9137m/s of delta-v to go from LEO to the lunar surface and return from the surface to Earth is a lot for a single stage. I expect you’d need to drop tanks along the way to have sufficient thrust-to-weight to land and take off again. Jon Goff had the right idea with his one-way one-person lander.
Well, if we’re talking about something with the complexity of a modern airliner and dealing with three-dimensional astrogation problems, I can see needing a small team on the ground to help with the maths
You don’t need a small team on the ground — just a small computer. No one does those type of calculations by hand these days.
and provide a second set of eyes on the gauges to make sure the pilot doesn’t miss anything. Lindberg himself was pretty occupied
Why do you obsess with Lindbergh and ignore Mercury, the X-15, and SpaceShip One? Or hundreds of military fighters that are flying with a single pilot every day?
Any pilot will tell you that the most critical aspects of flight are takeoff and landing, not “straight and level.” Once it’s left Earth orbit, the capsule is going to be coasting. Maybe one or two mid-couse maneuvers. What do you expect to be out there that will make it such a high workload environment?
The main difficulty in this is GN&C, and the problem has been overwhelmed by modern computational capability. The Apollo Guidance Computer was a 2.04 MHz machine with 34 K words of ROM, and 2 K words of RAM (a word being 16 bits). Kit computers of the late 70s had surpassed that, and a modern laptop PC dwarfs the capability of the IBM 360s used by ground control to actually process GN&C data.
The Dragon heat shield is, indeed, designed for lunar reentry. This mission would indeed be feasible on the same basis as Lindbergh’s flight. It should be attempted before the USG gains the authority to regulate human space flight….
As it turns out, today airliners actually radio all kinds of information about the status and condition of the aircraft without the pilots’ contribution at all. It’s done automatically. This allows maintenance to be ready at the aircraft’s destination.
Sending a maintenance log back to base has nothing in common with a mission-control room full of engineers telling the pilot how to fly the plane and a flight surgeon telling him when to scratch his right buttocks. Are you just being argumentative, or do you really not understand the difference?
And if you think the airline and other airplane pilots do not use ground based support when problems occur you are sadly mistaken. I, as a pilot, have gotten all sorts of recommendations from ATC when problems occurred.
Yes, ATC is likely to say “all sorts” of things. ATC is air *traffic* control, not air mission control. Most air-traffic controllers are not pilots, let alone aeronautical engineers. You don’t really think ATC has a mission control room with Gene Krantz and two dozen engineers, do you?
If you’re a pilot, you must know it’s possible to fly VFR from an uncontrolled field and not even contact ATC, let alone some imaginary mission-control room. I’ve flown in airplanes that didn’t even have radios. (And there is no ATC for space yet, by the way, although the FAA has put some thought into creating one.)
But even when Sully ditched in the Hudson, they were communicating with ATC and ATC made recommendations which the PIC elected to not take.
That’s nice, but Sully didn’t talk to mission control because airliners don’t have mission controls. We weren’t talking about pilots calling ATC, their squadron-mates, or their wives.
I think there’s no doubt a Dragon could do the Apollo 8 mission including sending three people rather than just one (many hands make for lighter work.) This could be a preamble to a regular tourist option.
Going to the surface should wait until we have a dedicated lander orbiting the moon rather than trying to bring a one person pogo stick along.
This could also be used to prove in a practical way the concept of on orbit refueling. Doing with an F9 what used to require a Saturn V.
Navigating to the moon is not a trivial issue, even if you have unlimited computing power with 0 weight and volume cost. Star sightings and laser gyros are all fine for attitude, but that doesn’t get you position and velocity. You need a transponder and ground network. You can use (earth or lunar) landmark sightings but this is not as accurate.
Trent makes a good point about flying to the moon for LOI vs. free-return. If you just want to whip around the moon, call up Space Adventures and they’ll set you up with the Russians, who are apparently willing to push a modified Soyuz on a free-return around the moon.
If Trent and others are pondering about the feasibility of a private Apollo 11 as opposed to a private Apollo 8, I have put some thought into a Centaur-based double-EOR architecture. I have been struggling to find enough information to flesh it out, although I have committed a spreadsheet with real calcs and a CAD model to the idea (not just doing abstract mass ratios, but actually doing configuration work and rough estimations of load and sizing structures). The Centaur would provide TLI, LOI, and possibly even lunar decent to just shy of terminal landing phase (i.e. crasher stage) for a 2-man slightly-larger-than-Gemini capsule w/ propellant and rockets for terminal landing, lunar ascent, and TEI (with similar configuration to ULA’s dual-axis horizontal lander).
I personally believe it would be an undue risk to attempt any lunar landing without at least 1 or 2 test flights leading up to the landing. This adds significantly to the cost, but it would be difficult to see any responsible way of putting human life on the lunar surface without it.
There are commercial options for space communications (which usually includes range and/or range rate data, important for calculating orbital trajectories), even to Lunar distances. Universal Space Network was checked out for use with The Lunar Reconnaissance Orbiter (LRO), though I’m not sure to what extent they’re used:
Of course, to complete the Lindbergh analogy, the only rations allowed on board would be tuna sandwiches. Don’t know the effect that would have on the weight budget…
So Tom, you’re saying the days of turkey salad sandwiches after thanksgiving are astronaut training?
Figuring the minimum resources needed is fun, but the right architecture is going to be robust. A general use zero-g, 7+ km/s delta-V spaceship (aka spaceship) can be a vital supply link providing fuel and supplies for the lander in lunar orbit and a transfer point of people and material from the lunar surface.
The fact that a Dragon has enough delta-V to go into moon orbit and back to earth means it was designed for robust earth orbit operations.
The fact that a Dragon has enough delta-V to go into moon orbit and back to earth means it was designed for robust earth orbit operations.
I’m surprised (due to my ignorance) that it is possible at all. What does it mean that Dragon can do it? Does it mean, as Ken says, that Dragon has lots margin for normal LEO operations, or does it really just show how close other capsules have been to being able to achieve the same thing — that is, go to lunar orbit (or even a free return trajectory) without a TLI stage? Can a version of Soyuz or Zond do it without any extra staging? Could a stripped-down Apollo or Gemini have done it with just one person and no extra staging? I’ve been reading this to help answer my own question: http://www.astronautix.com/craft/soyz7kl1.htm
I’m also reading this: http://www.astronautix.com/articles/bygemoon.htm to figure out why it wasn’t done with Gemini — at first glance, it looks like all the different Gemini plans used Centaurs as a TLI stage.
Edward, I think we’re either in violent agreement or arguing over a pretty small difference. I agree with you that 60s-style mission control isn’t needed, and I think the way NASA does it currently has a lot more to do with the tradition of a large flight control team than actual technical requirements. I’m not defending that.
What I am suggesting is that you may want two pairs of eyes on the systems for your lunar spacecraft, particularly in an emergency situation. I think the example of the X-15 is a good one- it did have a “ground control”, in the sense of a few guys watching the trajectory and letting the pilot know if he was getting too far off, rather than a NASA-style army. Modern airliners are not allowed to fly without two pilots, because the systems may need more than one pair of eyes and the workload may be two high for one person in an emergency. In both cases, the pilot-in-command is the one making the final calls, which is the way I think it should be. NASA’s concept of the flight crew as switch-flippers for the engineers on the ground never made a whole lot of sense to me.
The optimum solution, IMO, would be to take another person or two with you and do away with the ground control bit altogether. The only reason I suggested otherwise was because for a mission like this my impression was that every kg of dry mass was important. Ergo, one person only.
The most exciting thing to me is that it’s apparently doable, with hardware that’s either available now or is in development- real development with hardware being tested, not PowerPoint development. If I win the lottery…
Navigating to the moon is not a trivial issue, even if you have unlimited computing power with 0 weight and volume cost. Star sightings and laser gyros are all fine for attitude, but that doesn’t get you position and velocity. You need a transponder and ground network. You can use (earth or lunar) landmark sightings but this is not as accurate.
Laser ring gyros would be part of an inertial navigation system that is capable of determining position and velocity. Even homebuilt planes often have glass cockpit technology that is more advanced than what Apollo had. As for the difficulty of navigation to the moon, the US, Russia, Japan, China and India (am I forgetting anyone>) have all sent unmanned spacecraft to orbit the moon. It is proven technology that no longer requires a cast of thousands to achieve.
I agree the pilot model for a spaceship should be like an airplane. But don’t forget the Vortacs. When two or more are available they give the pilot a positional fix. When not available, pilots still have to fly and navigate the plane.
We don’t have much infrastructure for navigating in space but we don’t need much. Besides the stars, radio sources in known fixed locations can be used not just for position but velocity as well (the difference in fixed positions over time.) Solar position satellites may happen over time.
Spaceship pilots just need to be qualified to do their job and have a minimum of equipment needed to support that job. I would think that would include on-board navigational software with a database of reference points. Speaking to ground control would be considered an emergency backup to on-board equipment.
“The Dragon could do it” is an oversimplification. The Dragon is a good base to think about what is possible because it exists and has a few numbers available or easily inferred.. and it’s light.
As I said in the post, you’d need to pull that heavy common berthing mechanism off it, and cut out any other mass you can. That gets the mass down to something manageable – in fact I really don’t have enough information to tell how much is unnecessary. And yes, tuna sandwiches sounds like the kind of light weight rations to keep you alive.. and you can forget about the fancy toilet.. “lookin’ at stars, pissing in jars” as they say.
Next, the Dragon doesn’t have enough propellant in the standard configuration.. you need to extend the existing tanks or add new ones. There’s good techniques for estimating tank mass, so we can crunch the numbers.
The biggest problem with the Dragon is that its Draco thrusters are not design for “high impulse” maneuvering. The burn to get to the Moon would take 46 minutes with stock thrusters and that’s around 7.5 times too long for an impulse burn. So you either have to start in a highly elliptic orbit, which means you get less mass margin, or you have to spiral out, which takes longer and more delta-v. I prefer to just assume that 7.5 times as much mass flow can be put through the Dracos.. more customization.
Larry said Laser ring gyros would be part of an inertial navigation system that is capable of determining position and velocity. Even homebuilt planes often have glass cockpit technology that is more advanced than what Apollo had.
Inertial navigation is not accurate enough to get you to the moon and back unaided. The drift is too great, which is why they take star sightings to recalibrate for attitude. So you are still left with position and velocity errors. ICBMs (which use inertial navigation exclusively) already push the limits of what is possible from gyros and accelerometers, and even reconaissance aircraft use star trackers to re-calibrate their platforms.
Thanks Trent. I was quite taken with the idea that an “off the shelf” standard configuration Dragon could go to the moon just by reducing the crew and eliminating the docking mechanism, without as of yet undeveloped (or at least unpaid for) extra stages and exra launches. All the discussions of “lunar Dragons” on nasaspaceflight.com and other forums always postulate multiple launches, extra stages, etc. Oh well!
It was an exciting fantasy. And it led me to learn that the Dragon is surprisingly light.
Now I’m wondering about the details of the customization steps you suggest — it would be great if someone looked into the details in greater depth. I think Space Adventures would want to look into it. Their $200 million dollar proposal is cheaper per seat, but it relies on (I think) many more unbuilt & untested components.
US, Russia, Japan, China and India (am I forgetting anyone?)
Don’t forget Hughes’ Asiasat 3. I would think that here, of all places, the first commercial mission to the moon would get a mention.
See http://en.wikipedia.org/wiki/PAS-22
Inertial navigation is not accurate enough to get you to the moon and back unaided. The drift is too great, which is why they take star sightings to recalibrate for attitude. So you are still left with position and velocity errors. ICBMs (which use inertial navigation exclusively) already push the limits of what is possible from gyros and accelerometers, and even reconaissance aircraft use star trackers to re-calibrate their platforms.
Space navigation isn’t as hard as you think. For example, over 10 years ago the Deep Space 1 probe demonstrated auto navigation to include rendezvous with a comet. Multiple technologies are available and proven to make the navigation exercise of going to the moon almost simple.
This discussion got me to thinking about space navigation infrastructure. In particular I was curious whether GPS could be used. As usual, the “experts” are way out in front of me. I found slides from the Munich Satellite Navigation Summit in 2006. (http://www.pnt.gov/public/2006/2006-02-munich/NASA.pdf) Take a look at slides 11 and 12. It appears that GPS signals could be used to a point just beyond L1. Eventually, lunar NavSat’s could cover the remaining distance. However, in the meantime and more germane to this topic, existing GPS appears as if it could support the translunar correction maneuver both outbound and inbound.
So, in addition to much greater computing power, we also have GPS as an addition to the technologies making this feat easier now than in the late 60′s. I’m definitely not an expert, so additional thoughts or opposing facts would be welcome.
Running some numbers just for fun – not debating:
As I recall, Lindbergh got the money to pay for the Spirit from a bunch of St. Louis bankers and businessmen. I think he chipped in $2000-3000 of his money. All told they put together $25 large. I also recall they needed a little more but I forget how much so let’s go with $25,000.
How much is that $25,000 in 2010 dollars?
$315,661
Not much towards the $130 million needed but then circling the moon is a much harder deal. It’s *possible* one could have an airplane that can fly NY to Paris for $316 large, but not a sure thing. A Spirit replica with a modern engine and 1927 instruments would probably cost a lot more than $315G.
Okay, so (for the not-rocket-scientists) what he’s saying is that the Dragon capsule has (almost) enough onboard thrust to get one man to Lunar orbit and back? Like – “take out the extra seats” close?
He could toss out an RC-car while he swings by and claim the Google Lunar X-Prize while he’s at it.
Needs a beefier heat shield too for Lunar.
$130 million? Only a sufficiently motivated lottery winner away.
I agree with the underlying premise that such a stunt would go a long ways to put the lie to space travel being “NASA-hard.”
Of course, an accident could have the opposite effect; but then again so does not trying. I don’t see a downside here, and there’s plenty of upside potential for the company whose technology could pull this off.
Do you actually need to “enter orbit” at the moon? A quick sling around the backside without entering a stable orbit would seem possible and just as “press-worthy” with a substantial fuel savings. Yes?
The article talks about how it was just the Lone Eagle and his trusty airplane:
“He didn’t have a co-pilot. He didn’t have an army of engineers monitoring his plane or a flight surgeon monitoring his heart rate, it was just him and his trusty single-engine monoplane, the Spirit of St. Louis.”
And the implication, to me at least, is that it can be done the same way today and therefore more cheaply than we might imagine.
However you aren’t going to be able to dispense with the ‘army of engineers’ to monitor the spacecraft:
One thing to remember is that while the Spirit of St. Louis was a pretty advanced airplane for it’s time, it was still simple enough for one person to thoroughly understand it and figure out what the cause of any in-flight problem might be, and what effective remedies might be available, select one and implement it. Navigation was simple enough for the pilot to handle and since there was no GPS, no infrastructure was needed.
This will not be true of the circum-lunar flight. The systems will be too complex for one crew to thoroughly understand. They will need ground based support and this costs money.
What ground-based support will be required for navigation is worth some thought. We have much more computer power in smaller lighter packages than the Apollo days but I don’t know if it’s enough to completely manage the task. Plus ground based radar was used to locate the spacecraft in the Apollo days.
Lastly, one would probably need the Deep Space Network or TDRSS or some gubbmint-controlled communications network to stay in contact with the vehicle at all times (to supply that ground based anomaly support).
So while I like the idea and the analysis in the article seems reasonable, I don’t think it tells the entire story.
“Needs a beefier heat shield too for Lunar.”
I believe it was designed for that from the beginning (which means pretty good margin for LEO return).
But if it’s barely possible almost as-is, if this isn’t an argument for orbital refueling, what is?
Well, if you’re willing to wait for the Falcon 9 Heavy (with the tripled first stage) there’s plenty of delta-v, yes?
Of course, that adds another $40M or so (list) price…
Gregg raises some good points. I have not seen any information on the size of the mission support staff SpaceX uses for Dragon missions. If it’s like everything else I’ve seen from their company, it’ll be very lean by NASA standards. Part of the determination depends on the degree of autonomy built into the Dragon capsules. Back in the Apollo era, the on-board computers were primarily used for navigation. All of the vehicle systems were monitored from the ground (and in the vehicle) but all commands were executed by the crew. Today’s computers are orders of magnitude more capable than their Apollo counterparts especially at monitoring systems. Likewise, navigation systems today are both more capable and much smaller/lighter than those available for Apollo. The Apollo computer alone weighed 70 pounds. It’s likely the whole navigation system with the computer was well north of 200 pounds. By contrast, modern technology like AHRS, laser ring gyros, and good star sensors coupled with a good computer could likely do the same job and weigh less than 20 pounds.
For communications and precision navigation, they’d need to either buy expensive time on the Deep Space Network or come up with a private solution.
As for flying non-stop from New York to Paris, back in 2002 Lindbergh’s grandson make the flight in an essentially stock private plane with some additional ferry tanks fitted.
However you aren’t going to be able to dispense with the ‘army of engineers’ to monitor the spacecraft:
This will not be true of the circum-lunar flight. The systems will be too complex for one crew to thoroughly understand. They will need ground based support and this costs money.
A modern airliner is just as complex — in fact, more so. They don’t have a mission control. Some clueless types at NASA actually suggested developing a “mission control for airliners,” a few years back. The airlines simply smiled and ignored them. The aircraft is flown entirely from the cockpit, and the number of guys in the cockpit has gone from four or five to two.
You overglorify the “army of engineers.” In the television version of “Failure is Not an Option,” Gene Krantz talked about the first spacewalk and the feeling of helplessness since there was nothing mission control could do if the astronaut ran into a problem.
In typical Krantz fashion, he failed to note the irony that NASA was paying two dozen guys to sit in front of consoles and watch teevee even though their boss said there was nothing they could do to help.
I remember Max Hunter saying that he used to launch Thor missiles with a crew of eight, none of whom ranked higher than sergeant — and he designed it to be launched in the field with even smaller crews. When the Air Force turned it into Delta, they had over 20 people doing the same job, and when NASA started launching Deltas, they had over 100. Compare the ground crew for DC-X or SpaceShip One to the number of people NASA uses to launch a rocket.
What ground-based support will be required for navigation is worth some thought. We have much more computer power in smaller lighter packages than the Apollo days but I don’t know if it’s enough to completely manage the task.
Not even close. A short time ago, one gigaflops (1 billion floating-point operations per second) was considered the definition of a supercomputer. The holy grail of supercomputing was one terraflops — 1000x that. Today, video gamers have that much power in their graphic cards. Even cellphone have more computing power than supercomputers of not that long ago — vastly more power than the guys who designed Apollo ever dreamed of — in a package that weighs about five ounces with multiple sensors, wireless telemetry, etc. built in.
That’s what happens when you have an environment that allows competition, innovation, and incremental development. If we ran computing the way we ran the space program, Mike Griffin would be telling us we need to build another vacuum-tube machine because “it’s the only thing we know that works.”
For communications and precision navigation, they’d need to either buy expensive time on the Deep Space Network or come up with a private solution.
Commercial communication satellites have done lunar-swingby maneuvers to get to GEO, and I don’t recall their renting time on the DSN.
The United States Navy was successfully bouncing signals off the Moon for radio communications in the 1950′s.
The Moon is not deep space. Apollo did not use the Deep Space Network, it used the Manned Space Network. (An exception was during the Apollo 13 emergency, when they needed to save power and required the bigger antennas of the DSN.)
Well, if we’re talking about something with the complexity of a modern airliner and dealing with three-dimensional astrogation problems, I can see needing a small team on the ground to help with the maths and provide a second set of eyes on the gauges to make sure the pilot doesn’t miss anything. Lindberg himself was pretty occupied with just flying the Spirit, and I seem to recall the Mercury astronauts also had some workload issues. It would probably also be useful to have some kind of ground-based check on the spacecraft’s position as well.
With decent automation it’s probably going to be more like a “squad” of engineers than an “army” supporting the mission, though.
[And yes, you could get the same effect with extra crew members onboard the spacecraft and possibly dispense with ground control entirely. I'm assuming for the sake of argument that this pilot mission is limited to one crew member to reduce dry mass]
The origional idea for a one-man mission draws a parallel to Lindbergh but, as Frank Glover suggests, why limit yourself to one if refueling in orbit is possible? Or, dock with a small ‘earth-departure’ stage once in orbit.
The United States Navy was successfully bouncing signals off the Moon for radio communications in the 1950′s.
Amateurs routinely bounce signals off the moon today. I’d go as far as to say that there’s plenty of private entities who could do deep space communication with a Mars expedition (which is one to two orders of magnitude further away).
Greg,
The $25,000 was the prize, if I recalled they only raised spent around $15,000 to win it.
The Wapedia gives the cost of the Spirit of St. Louis at only $10,000, which is about right since is was based on the Ryan M-2 mail plane that was already in service
ttp://wapedia.mobi/en/Spirit_of_St._Louis
I don’t really have anything to add to this discussion, but I’d just like to say that it’s very exciting reading.
Thomas M:
Actually, according to Lindy himself the cost of the plane as delivered was $10,580+ the cost of extras. I quote his book:
“That will make $10,580, with a J-5 engine – special equipment extra at cost.”
In addition, Lindbergh wrote be bought 500 gallons of special fuel which developed a little more power and had it shipped form California to NY. I don’t know what av gas cost per gallon back then.
And it was definitely not a warmed over M-2. I quote again [ Donald Hall - the chief designer speaking]:
” Now we can’t use the standard Ryan fuselage, ” he says. “Also the wind span will have to be considerably increased so as to reduce the wing loading for take off and increase the aspect ratio for range. That means we’ll have to move the tail surfaces aft to maintain satisfactory stability and control. And that means the engine will have to be moved forward. When it comes right down to it, I’ve really got to design a completely new fuselage structure to meet your requirements. We will have to design a different type of landing gear while we’re about it,” he continues. “The M-2 type of gear would be too heavy when it’s designed to go with the longer wing span and to take the load you’re going to carry….”
Edward Wright Says:
“A modern airliner is just as complex — in fact, more so. They don’t have a mission control. Some clueless types at NASA actually suggested developing a “mission control for airliners,” a few years back. The airlines simply smiled and ignored them. The aircraft is flown entirely from the cockpit, and the number of guys in the cockpit has gone from four or five to two.
You overglorify the “army of engineers.” ”
As it turns out, today airliners actually radio all kinds of information about the status and condition of the aircraft without the pilots’ contribution at all. It’s done automatically. This allows maintenance to be ready at the aircraft’s destination.
And if you think the airline and other airplane pilots do not use ground based support when problems occur you are sadly mistaken. I, as a pilot, have gotten all sorts of recommendations from ATC when problems occurred. As PIC it was ultimately my decision as to what to do.
And I’m flying planes not much more complex than the Spirit. Some (like a J-3 Cub) less complex.
Not all the time of course – there are plenty of emergencies where there’s no time for long deep conversations.
But even when Sully ditched in the Hudson, they were communicating with ATC and ATC made recommendations which the PIC elected to not take.
And I’m not over-glorifying anything. I’m simply saying you will have to have ground based support to analyze and/or simulate problems, that might occur, and work out solutions. I said nothing about Krantz nor anything about the “glory” days of Apollo.
You’ve made the grave, grave error of overlaying on my remarks someone else’s deification of the Apollo days.
“Gene Krantz talked about the first spacewalk and the feeling of helplessness since there was nothing mission control could do if the astronaut ran into a problem. ”
And the farther out we go, the more that speed of light delays will mandate autonomy by crews (and/or, someday, advanced AI). Best get used to it now.
The planetary science people work with this by planning (and often executing) their moves very slowly and carefully, especially with surface rovers. Helpless is not being able to do anything about what you’re seeing in nearly real-time. Worse than helpless is knowing that whatever the outcome of it will be, many light-minutes away, has *already* happened…
How certain/uncertain is the estimate of the Draco thrusters performance?
Also, not sure how the life-support resources scale, but I wonder if two people could go, if they were very lightweight (but still healthy adults). I’m thinking of Dick Rutan half-joking that Jeana Yeager cut off her ponytail before their round-the-world flight to gain an extra four miles worth of fuel.
Yes, I really didn’t even bother talking about the difficulty of plotting a TLI maneuver or LOI, etc. because it’s not hard. In terms of computing power, I expect these days you could probably get an iPhone app that would do the job.
If you’re interested in seeing how easy it is to do the calculations that took an army of “computers” back in the days of Apollo, just go download one of the many free software packages that are available.. there’s even free versions of the software the big boys use now.
Al, yes it would be press worthy. Anyone doing anything in space is press worthy. The reason I chose the Apollo 8 profile over the Apollo 13 profile is that free-return-trajectory is just too damn easy. It’s often described as 3.2km/s of delta-v (because you have to course correct a few m/s along the way), which means your vehicle could be as much as 3669kg dry for a single Falcon 9 launch. I estimate the dry mass of a stock standard Dragon at 2296kg.. you could take 7 people on that joy ride and still have mass for luggage.
rickl, thanks, that was my goal.
On the other end of the spectrum, I’m trying to convince myself that a 511kg dry mass design for a single person lunar lander doesn’t exist. 9137m/s of delta-v to go from LEO to the lunar surface and return from the surface to Earth is a lot for a single stage. I expect you’d need to drop tanks along the way to have sufficient thrust-to-weight to land and take off again. Jon Goff had the right idea with his one-way one-person lander.
Well, if we’re talking about something with the complexity of a modern airliner and dealing with three-dimensional astrogation problems, I can see needing a small team on the ground to help with the maths
You don’t need a small team on the ground — just a small computer. No one does those type of calculations by hand these days.
and provide a second set of eyes on the gauges to make sure the pilot doesn’t miss anything. Lindberg himself was pretty occupied
Why do you obsess with Lindbergh and ignore Mercury, the X-15, and SpaceShip One? Or hundreds of military fighters that are flying with a single pilot every day?
Any pilot will tell you that the most critical aspects of flight are takeoff and landing, not “straight and level.” Once it’s left Earth orbit, the capsule is going to be coasting. Maybe one or two mid-couse maneuvers. What do you expect to be out there that will make it such a high workload environment?
The main difficulty in this is GN&C, and the problem has been overwhelmed by modern computational capability. The Apollo Guidance Computer was a 2.04 MHz machine with 34 K words of ROM, and 2 K words of RAM (a word being 16 bits). Kit computers of the late 70s had surpassed that, and a modern laptop PC dwarfs the capability of the IBM 360s used by ground control to actually process GN&C data.
The Dragon heat shield is, indeed, designed for lunar reentry. This mission would indeed be feasible on the same basis as Lindbergh’s flight. It should be attempted before the USG gains the authority to regulate human space flight….
As it turns out, today airliners actually radio all kinds of information about the status and condition of the aircraft without the pilots’ contribution at all. It’s done automatically. This allows maintenance to be ready at the aircraft’s destination.
Sending a maintenance log back to base has nothing in common with a mission-control room full of engineers telling the pilot how to fly the plane and a flight surgeon telling him when to scratch his right buttocks. Are you just being argumentative, or do you really not understand the difference?
And if you think the airline and other airplane pilots do not use ground based support when problems occur you are sadly mistaken. I, as a pilot, have gotten all sorts of recommendations from ATC when problems occurred.
Yes, ATC is likely to say “all sorts” of things.
ATC is air *traffic* control, not air mission control. Most air-traffic controllers are not pilots, let alone aeronautical engineers. You don’t really think ATC has a mission control room with Gene Krantz and two dozen engineers, do you?
If you’re a pilot, you must know it’s possible to fly VFR from an uncontrolled field and not even contact ATC, let alone some imaginary mission-control room. I’ve flown in airplanes that didn’t even have radios. (And there is no ATC for space yet, by the way, although the FAA has put some thought into creating one.)
But even when Sully ditched in the Hudson, they were communicating with ATC and ATC made recommendations which the PIC elected to not take.
That’s nice, but Sully didn’t talk to mission control because airliners don’t have mission controls. We weren’t talking about pilots calling ATC, their squadron-mates, or their wives.
I think there’s no doubt a Dragon could do the Apollo 8 mission including sending three people rather than just one (many hands make for lighter work.) This could be a preamble to a regular tourist option.
Going to the surface should wait until we have a dedicated lander orbiting the moon rather than trying to bring a one person pogo stick along.
This could also be used to prove in a practical way the concept of on orbit refueling. Doing with an F9 what used to require a Saturn V.
Good job Trent.
Navigating to the moon is not a trivial issue, even if you have unlimited computing power with 0 weight and volume cost. Star sightings and laser gyros are all fine for attitude, but that doesn’t get you position and velocity. You need a transponder and ground network. You can use (earth or lunar) landmark sightings but this is not as accurate.
Trent makes a good point about flying to the moon for LOI vs. free-return. If you just want to whip around the moon, call up Space Adventures and they’ll set you up with the Russians, who are apparently willing to push a modified Soyuz on a free-return around the moon.
If Trent and others are pondering about the feasibility of a private Apollo 11 as opposed to a private Apollo 8, I have put some thought into a Centaur-based double-EOR architecture. I have been struggling to find enough information to flesh it out, although I have committed a spreadsheet with real calcs and a CAD model to the idea (not just doing abstract mass ratios, but actually doing configuration work and rough estimations of load and sizing structures). The Centaur would provide TLI, LOI, and possibly even lunar decent to just shy of terminal landing phase (i.e. crasher stage) for a 2-man slightly-larger-than-Gemini capsule w/ propellant and rockets for terminal landing, lunar ascent, and TEI (with similar configuration to ULA’s dual-axis horizontal lander).
I personally believe it would be an undue risk to attempt any lunar landing without at least 1 or 2 test flights leading up to the landing. This adds significantly to the cost, but it would be difficult to see any responsible way of putting human life on the lunar surface without it.
There are commercial options for space communications (which usually includes range and/or range rate data, important for calculating orbital trajectories), even to Lunar distances. Universal Space Network was checked out for use with The Lunar Reconnaissance Orbiter (LRO), though I’m not sure to what extent they’re used:
http://uspacenetwork.com/
Of course, to complete the Lindbergh analogy, the only rations allowed on board would be tuna sandwiches. Don’t know the effect that would have on the weight budget…
So Tom, you’re saying the days of turkey salad sandwiches after thanksgiving are astronaut training?
Figuring the minimum resources needed is fun, but the right architecture is going to be robust. A general use zero-g, 7+ km/s delta-V spaceship (aka spaceship) can be a vital supply link providing fuel and supplies for the lander in lunar orbit and a transfer point of people and material from the lunar surface.
The fact that a Dragon has enough delta-V to go into moon orbit and back to earth means it was designed for robust earth orbit operations.
The fact that a Dragon has enough delta-V to go into moon orbit and back to earth means it was designed for robust earth orbit operations.
I’m surprised (due to my ignorance) that it is possible at all. What does it mean that Dragon can do it? Does it mean, as Ken says, that Dragon has lots margin for normal LEO operations, or does it really just show how close other capsules have been to being able to achieve the same thing — that is, go to lunar orbit (or even a free return trajectory) without a TLI stage? Can a version of Soyuz or Zond do it without any extra staging? Could a stripped-down Apollo or Gemini have done it with just one person and no extra staging? I’ve been reading this to help answer my own question: http://www.astronautix.com/craft/soyz7kl1.htm
I’m also reading this:
http://www.astronautix.com/articles/bygemoon.htm to figure out why it wasn’t done with Gemini — at first glance, it looks like all the different Gemini plans used Centaurs as a TLI stage.
Edward, I think we’re either in violent agreement or arguing over a pretty small difference. I agree with you that 60s-style mission control isn’t needed, and I think the way NASA does it currently has a lot more to do with the tradition of a large flight control team than actual technical requirements. I’m not defending that.
What I am suggesting is that you may want two pairs of eyes on the systems for your lunar spacecraft, particularly in an emergency situation. I think the example of the X-15 is a good one- it did have a “ground control”, in the sense of a few guys watching the trajectory and letting the pilot know if he was getting too far off, rather than a NASA-style army. Modern airliners are not allowed to fly without two pilots, because the systems may need more than one pair of eyes and the workload may be two high for one person in an emergency. In both cases, the pilot-in-command is the one making the final calls, which is the way I think it should be. NASA’s concept of the flight crew as switch-flippers for the engineers on the ground never made a whole lot of sense to me.
The optimum solution, IMO, would be to take another person or two with you and do away with the ground control bit altogether. The only reason I suggested otherwise was because for a mission like this my impression was that every kg of dry mass was important. Ergo, one person only.
The most exciting thing to me is that it’s apparently doable, with hardware that’s either available now or is in development- real development with hardware being tested, not PowerPoint development. If I win the lottery…
Navigating to the moon is not a trivial issue, even if you have unlimited computing power with 0 weight and volume cost. Star sightings and laser gyros are all fine for attitude, but that doesn’t get you position and velocity. You need a transponder and ground network. You can use (earth or lunar) landmark sightings but this is not as accurate.
Laser ring gyros would be part of an inertial navigation system that is capable of determining position and velocity. Even homebuilt planes often have glass cockpit technology that is more advanced than what Apollo had. As for the difficulty of navigation to the moon, the US, Russia, Japan, China and India (am I forgetting anyone>) have all sent unmanned spacecraft to orbit the moon. It is proven technology that no longer requires a cast of thousands to achieve.
Shit,
We are threading needles in the outer planets with unmanned craft.
Entering Lunar Orbit with a sober Hobo and an IPhone should be baby-shit today.
I agree the pilot model for a spaceship should be like an airplane. But don’t forget the Vortacs. When two or more are available they give the pilot a positional fix. When not available, pilots still have to fly and navigate the plane.
We don’t have much infrastructure for navigating in space but we don’t need much. Besides the stars, radio sources in known fixed locations can be used not just for position but velocity as well (the difference in fixed positions over time.) Solar position satellites may happen over time.
Spaceship pilots just need to be qualified to do their job and have a minimum of equipment needed to support that job. I would think that would include on-board navigational software with a database of reference points. Speaking to ground control would be considered an emergency backup to on-board equipment.
“The Dragon could do it” is an oversimplification. The Dragon is a good base to think about what is possible because it exists and has a few numbers available or easily inferred.. and it’s light.
As I said in the post, you’d need to pull that heavy common berthing mechanism off it, and cut out any other mass you can. That gets the mass down to something manageable – in fact I really don’t have enough information to tell how much is unnecessary. And yes, tuna sandwiches sounds like the kind of light weight rations to keep you alive.. and you can forget about the fancy toilet.. “lookin’ at stars, pissing in jars” as they say.
Next, the Dragon doesn’t have enough propellant in the standard configuration.. you need to extend the existing tanks or add new ones. There’s good techniques for estimating tank mass, so we can crunch the numbers.
The biggest problem with the Dragon is that its Draco thrusters are not design for “high impulse” maneuvering. The burn to get to the Moon would take 46 minutes with stock thrusters and that’s around 7.5 times too long for an impulse burn. So you either have to start in a highly elliptic orbit, which means you get less mass margin, or you have to spiral out, which takes longer and more delta-v. I prefer to just assume that 7.5 times as much mass flow can be put through the Dracos.. more customization.
US, Russia, Japan, China and India (am I forgetting anyone>)
ESA (SMART-1, 2006).
Larry said Laser ring gyros would be part of an inertial navigation system that is capable of determining position and velocity. Even homebuilt planes often have glass cockpit technology that is more advanced than what Apollo had.
Inertial navigation is not accurate enough to get you to the moon and back unaided. The drift is too great, which is why they take star sightings to recalibrate for attitude. So you are still left with position and velocity errors. ICBMs (which use inertial navigation exclusively) already push the limits of what is possible from gyros and accelerometers, and even reconaissance aircraft use star trackers to re-calibrate their platforms.
Thanks Trent. I was quite taken with the idea that an “off the shelf” standard configuration Dragon could go to the moon just by reducing the crew and eliminating the docking mechanism, without as of yet undeveloped (or at least unpaid for) extra stages and exra launches. All the discussions of “lunar Dragons” on nasaspaceflight.com and other forums always postulate multiple launches, extra stages, etc. Oh well!
It was an exciting fantasy. And it led me to learn that the Dragon is surprisingly light.
Now I’m wondering about the details of the customization steps you suggest — it would be great if someone looked into the details in greater depth. I think Space Adventures would want to look into it. Their $200 million dollar proposal is cheaper per seat, but it relies on (I think) many more unbuilt & untested components.
US, Russia, Japan, China and India (am I forgetting anyone?)
Don’t forget Hughes’ Asiasat 3. I would think that here, of all places, the first commercial mission to the moon would get a mention.
See http://en.wikipedia.org/wiki/PAS-22
Inertial navigation is not accurate enough to get you to the moon and back unaided. The drift is too great, which is why they take star sightings to recalibrate for attitude. So you are still left with position and velocity errors. ICBMs (which use inertial navigation exclusively) already push the limits of what is possible from gyros and accelerometers, and even reconaissance aircraft use star trackers to re-calibrate their platforms.
Space navigation isn’t as hard as you think. For example, over 10 years ago the Deep Space 1 probe demonstrated auto navigation to include rendezvous with a comet. Multiple technologies are available and proven to make the navigation exercise of going to the moon almost simple.
This discussion got me to thinking about space navigation infrastructure. In particular I was curious whether GPS could be used. As usual, the “experts” are way out in front of me. I found slides from the Munich Satellite Navigation Summit in 2006. (http://www.pnt.gov/public/2006/2006-02-munich/NASA.pdf) Take a look at slides 11 and 12. It appears that GPS signals could be used to a point just beyond L1. Eventually, lunar NavSat’s could cover the remaining distance. However, in the meantime and more germane to this topic, existing GPS appears as if it could support the translunar correction maneuver both outbound and inbound.
So, in addition to much greater computing power, we also have GPS as an addition to the technologies making this feat easier now than in the late 60′s. I’m definitely not an expert, so additional thoughts or opposing facts would be welcome.
Comments are closed.