Some questions about feasibility of lunar resource exploitation

[Styx]

I am sincerely curious about the numbers behind warrens idea of exploiting lunar resources. Our personal beliefs about whether something can or cant be done does not matter if the numbers say other said. If it can be proven the cost of placing a mining and launch facility on the moon is cheaper than mining and launching the same resource from Earth it could be a tremendous boon to spaceflight. But to do so I need to know the following.

1. How big a facility are we talking ? As in how many tons of payload must be placed into the lunar surface to make a facility to gather, extract and launch fuel/propellant from the moon ?

2. How many tons of fuel/propellant can this facility shoot back into space each year ?

3. We only have fuzzy indirect evidence of some water on the moon and not even a single picture of something recognizable as H2O. Where is this great reservoir of water that can be cracked for fueling a planets spaceflight needs ? Mind you the water needs to be concentrated. If you have a billion cubic liters spread across the entire moon you might as well believe you can make a fortunate mining gold out of the sands on a beach.

4. There needs to be manned presence on site. What kind of rocket will send all those engineers and workers to and from the moon ? How much does it cost compared to a Saturn 5 that can put two persons on the moon ?

5. What about supply food and supplies ? How much does moving a 1 kilogram of apples to the moon cost ? How many tons of food do a 100 people eat in a month ?

6. The ISS masses about 400 tons and can hold six people for extended periods of time. How many thousands of tons will a facility with enough electrical power, life support, radiation shielding for a 100 people going to be ?

7. How much does it cost to transport 1000 tons to the moon ?

8. Adding all these costs up how much lunar hydrogen and oxygen cost per kilo compared to something launched at 1000 dollars a year ? How many years before the facility pays for itself ?

9. Lunar fuel would only be useful to spacecraft headed out to interplanetary space. How many interplanetary spacecraft will still be burning hydrogen and oxygen instead of nuclear or solar powered engines ? In a future where there is enough space travel to justify mining the moon why would anyone make year long trips in a primitive spacecraft when they have far superior propulsion technology available ?

[Hop_David]

I am sincerely curious about the numbers behind warrens idea of exploiting lunar resources. Our personal beliefs about whether something can or cant be done does not matter if the numbers say other said. If it can be proven the cost of placing a mining and launch facility on the moon is cheaper than mining and launching the same resource from Earth it could be a tremendous boon to spaceflight. But to do so I need to know the following.

1. How big a facility are we talking ? As in how many tons of payload must be placed into the lunar surface to make a facility to gather, extract and launch fuel/propellant from the moon ?

2. How many tons of fuel/propellant can this facility shoot back into space each year ?

3. We only have fuzzy indirect evidence of some water on the moon and not even a single picture of something recognizable as H2O. Where is this great reservoir of water that can be cracked for fueling a planets spaceflight needs ? Mind you the water needs to be concentrated. If you have a billion cubic liters spread across the entire moon you might as well believe you can make a fortunate mining gold out of the sands on a beach.

Somewhat more than fuzzy indirect evidence.

This is one of the more persuasive links and quotes Warren has posted:

According to Dr. Paul Spudis, the Principle Investigator of the mini-SAR imaging radar experiment:

Our radar team has just finished the first round of analysis of data returned by the Mini-SAR for the north pole and results will soon be published in the technical journal, Geophysical Research Letters.

Mini-SAR is a lightweight, low power imaging radar. It uses the polarization properties of reflected radio waves to characterize the lunar surface composition and physical state. Mini-SAR transmits pulses of left-circularly polarized radar. Typically, reflection from planetary surfaces reverses the transmitted polarization, so that Mini-SAR radar echoes from the Moon are right circularly polarized. The ratio of received power in the same sense transmitted (left circular) to the opposite sense (right circular) is called the circular polarization ratio (CPR). Most of the Moon has low CPR (about 0.3), meaning that a reversal of polarization is the norm, but some specific areas have high CPR (greater than 1.0). These include very rough, rocky surfaces (such as a young, fresh crater) and ice, which is transparent to radio energy. In this latter case, the radar penetrates the ice and is scattered and reflected multiple times by inclusions and flaws in the ice, resulting in the reflection of many same sense polarization echoes, leading to higher CPR values than normal. High values of CPR are not uniquely diagnostic of either surface roughness or ice; we must take into account the geological setting of the high CPR signal to interpret its cause.

Many craters near the poles of the Moon have interiors that are in permanent shadow from the Sun. These areas are very cold and water ice is stable permanently there. Fresh craters show high degrees of surface roughness (high CPR) both inside and outside the crater rim, caused by sharp rocks and block fields that are distributed over the entire crater area. However, Mini-SAR found craters near the north pole that have high CPR values inside, but not outside their rims. This relation suggests that the high CPR is not caused by roughness, but by some material that is restricted within the interiors of these craters. It is not geologically reasonable to expect rough, fresh surfaces to be present inside a crater rim but absent outside of it. The craters that show this enhancement are all permanently cold and dark, where ice is stable. We thus interpret this high CPR to mean that water ice is present in these craters.

Over forty small (2-15 km diameter) craters near the north pole of the Moon are found to contain this elevated CPR material. The total mount of ice present at the pole depends on how thick it is; to see this elevated CPR effect, the ice must have a thickness on the order of tens of wavelengths of the radar used. Our radar wavelength is 12.6 cm, therefore we think that the ice must be at least two meters thick and relatively pure. At such a thickness, more than 600 million metric tones of water ice are present in this area.

This reinforces my opinion that are exploitable concentrations of water.

In addition to water, oxygen is quite abundant in the lunar crust. I have heard some estimates that it is 40% oxygen by mass. Oxygen can be extracted from lunar minerals by pyrolysis.

Getting back to your questions 1 and 2: I don't know the answer to either of those questions.

Occasionally folks propose storing propellent in orbital depots in the form of water. They mention ice is easier to store and the oxidizer and fuel can be separated via electrolysis. Such proposals are usually shot down quickly. Power source for splitting hydrogen and oxygen is massive and costly. Here is a NASA pdf

So long as the power source is of terrestial origin, it seems to me lunar water ISRU for propellent has the same problem.

Now if solar cells could be made from lunar resources, this could circumvent the need for continued lifting of earthly power sources.

How massive would pyrolysis plants to extract oxygen be? This I also don't know.

I have been hoping to find pdfs on minimum size plants needed for various lunar ISRU schemes. Splitting water, extracting oxygen from ilmenite, and making solar panels from lunar resources would be the 3 ISRU projects I'm most interested in.

4. There needs to be manned presence on site. What kind of rocket will send all those engineers and workers to and from the moon ? How much does it cost compared to a Saturn 5 that can put two persons on the moon ?

5. What about supply food and supplies ? How much does moving a 1 kilogram of apples to the moon cost ? How many tons of food do a 100 people eat in a month ?

6. The ISS masses about 400 tons and can hold six people for extended periods of time. How many thousands of tons will a facility with enough electrical power, life support, radiation shielding for a 100 people going to be ?

4, 5 and 6 are sure to be major expenses. How much can be done by telerobots? This is an open question. But I believe a large part of the expense in 4, 5 and 6 can be circumvented. I would hope a facility with 100 workers on the surface isn't a prerequisite.

7. How much does it cost to transport 1000 tons to the moon ?

There is a lot of debate on this. Large cost savings could be achieved if we attain a high flight rate. This would allow development and design costs to be amortized over many units. High flight rate would also give experience. There's no substitute for experience.

The Commercially Base Lunar Architecture paper that Warren frequently cites calls for relatively small vehicles using propellent depots.

At the moment, I can't find cost estimates on the paper. But I seem to recall Warren saying it would cost 7 billion a year.

Table 3 on page 20 says 40 tons on lunar surface the first year.
Table 4 on page 21 says 143 tons the 2nd year, cum 194 tons (their cum seems 11 tons off by my addition)
Table 5 on page 22 says 119 tons the 3rd year, cum tons 313 tons

8. Adding all these costs up how much lunar hydrogen and oxygen cost per kilo compared to something launched at 1000 dollars a year ? How many years before the facility pays for itself ?

My chief reason for wanting multiple propellent sources isn't cheaper fuel. It's to break up the exponent in the rocket equation. This enables smaller, simpler vehicles for trips to Mars or asteroids.

9. Lunar fuel would only be useful to spacecraft headed out to interplanetary space. How many interplanetary spacecraft will still be burning hydrogen and oxygen instead of nuclear or solar powered engines ?

Nuclear and solar rocket engines also use propellent. So Multiple propellent sources would enhance the capabilities of these engines also.

In my opinion Kirk Sorensen is one of the more credible writers on space travel. And he is constantly dissing VASIMIR. Not sure why. I would guess it's because low thrust doesn't enable good exploitation of Oberth effect. Also mass required for a big power source is a draw back. I'm hoping he will give some more detailed arguments.

But I don't always agree with Sorensen. I remain agnostic on VASIMIR.

In a future where there is enough space travel to justify mining the moon why would anyone make year long trips in a primitive spacecraft when they have far superior propulsion technology available ?

As I mentioned, the advanced transportation methods that use propellent could benefit from multiple propellent sources.

Given EML1 (or2)'s 2.4 km/sec advantage a two or three month trip to Mars with conventional chemical rockets is doable.

[Warren Platts]

1. In ten years there would be at least 20 modules with a useful mass on the order of 1,000 tons.

2. The goal would be to deliver on the order of 100 to 1,000 tons of propellant to the L2 depot annually.

3. One 2-kilometer diameter crater would have at least 6 million tons of bulldozable relatively pure ice.

4. EELV's would be used because they are cheaper than HLV.

5. It would cost on the order of $7 billion USD per year to keep the outpost supplied and upgraded.

6. There won't be 100 people since much of the work will be automated. Power levels will be on the order of a megawatt.

7. $70 billion USD.

8. Once the base is established with reusable SSTO's, the marginal cost will be much less propellant launched from Earth all the way to L2, and possibly lunar propellant to LEO would be cheaper as well. Since the base will be supporting NASA planetary science, it does not need to make a net profit in order to justify its existence any more than the ISS or a flags and footprints mission to Mars would use "profit" as a justification for those projects. That said, 1,000 tons of propellant to L2 @ $10 million ea = $10 billion USD compared to a $7 billion annual cost.

9. Lunar propellant could be used for any application within cis-lunar space with nontrivial propellant requirements. E.g., lofting satellites from LEO to GEO.

[NEOWatcher]

2. The goal would be to deliver on the order of 100 to 1,000 tons of propellant to the L2 depot annually.
5. It would cost on the order of $7 billion USD per year to keep the outpost supplied and upgraded.
7. $70 billion USD.

So, taking the conservative numbers you provided, the annual cost is the same as Earth based launches.

Being that there are still plenty of unknowns in there*, can we even say that the conservative numbers are achievable in the near term?

*(At least from the lack details of where your numbers are coming from)

8. Since the base will be supporting NASA planetary science, it does not need to make a net profit in order to justify its existence any more than the ISS or a flags and footprints mission to Mars would use "profit" as a justification for those projects.

I do agree with that. I can't see exploitation of resources to be self sufficient. I can only see it as a natural extension of other research.

[Hop_David]

2. The goal would be to deliver on the order of 100 to 1,000 tons of propellant to the L2 depot annually.

I am assuming that eventually this will all be lunar propellent. 7 tons per trip?

The round trip from Lunar surface to L2 and back will be about 4.8 km/sec. So propellent mass needed would be about double the dry mass of the ACES/Altair tanker.

Also 7 tons of propellent is being exported to EML2. This one way trip is about 2.4 km/sec. Propellent needed to move propellent would be about .73 * 7 tons.

So to get 7 tons of propellent to EML2 you need this mass of propellent: 2*dry mass ACES/Altair tanker + 5.1 tons.

Have they come up with estimates of the dry mass of the proposed tanker?

[Hop_David]

I do agree with that. I can't see exploitation of resources to be self sufficient. I can only see it as a natural extension of other research.

The poles are the coldest places in the solar system, colder than Pluto. I am guessing these are strange and wonderful places worthy of study.

They would also be great locations for infrared telescopes. An infrared scope at each pole would give almost 4 pi steradians of view.

[Warren Platts]

1. I am assuming that eventually this will all be lunar propellent. 7 tons per trip?

2. The round trip from Lunar surface to L2 and back will be about 4.8 km/sec. So propellent mass needed would be about double the dry mass of the ACES/Altair tanker.

Also 7 tons of propellent is being exported to EML2. This one way trip is about 2.4 km/sec. Propellent needed to move propellent would be about .73 * 7 tons.

So to get 7 tons of propellent to EML2 you need this mass of propellent: 2*dry mass ACES/Altair tanker + 5.1 tons.

Have they come up with estimates of the dry mass of the proposed tanker?

1. That's what the Frank Zegler paper, page 25 said. But I think that was based on the ordinary DTAL with the ACES-41. It would be better to design a dedicated unmanned tanker-lander based on the ACES-71 tanker.

2. the Frank Zegler paper, Table 1. and Table 2. (page 11-12) The ACES-71 holds 71 total tons of fuel. The dry mass for the basic version is 5.5 tons. Add 2 more tons for the descent equipment (lateral thrusters and landing gear). That makes a total dry mass of 7.5 tons.

Using your figures, the dry mass to go back and forth would require 15 tons of propellant. That leaves 55 tons. Thus: mass delivered to depot = 55 / 1.73 = ~30 tons per trip.

To supply the year 3 projected L2 depot propellant flux of 243 tons (the Frank Zegler paper, Table 5, page 22), that would only require 8 launches from the Moon's surface.

It would be interesting to see how a reusable basic ACES-71 tanker would perform making the run from L2 to LEO. Dry mass is only 5.5 tons, but the delta v is 3.5 km/s each way. Table 2 says the ACES-71 could deliver 29 tons to L2 from LEO, if you throw away the tanker at the end of the run. So how much propellant do you think it would take to dead head it back to the starting point?

ETA: Well, just interpolating by eye, it would take about 4 tons to send the ACES-71 tanker back empty to L2 after making a delivery to LEO, so it would take about 2 1/2 launches from the Moon in order to be able to deliver 20 tons to LEO. The year 3 projected LEO propellant flux is 485 tons, so it would take about 20 deliveries from the space tanker, which in turn would require 50 launches from the tanker lander. Thus for an aggressive plan, there might be 60 launches per year; about 5 per month; this would require a total production of over 4,000 tons of propellant. This would require digging a hole about half the size of a football field to a depth of 2 meters. The processing plant would have to process about one 20 ton truckload every other day.

This would eliminate the need for 22 launches of propellant and their associated tankers and RL-10 engines from Earth (728 tons to LEO @ 33 tons each). (Compared to my figure of 1800 tons to L2 per years.... I guess the difference is the wholly reusable architecture. No, the difference is I've got it backwards. Since the flow is from L2 to LEO, the LEO propellant requirement to support the lunar base would be vastly reduced since it would only be modules and people--rather than ever more propellant--that would need the propellant. So my analysis above is major overkill.)
_____________________________________
skylon.com

[Warren Platts]

1. So, taking the conservative numbers you provided, the annual cost is the same as Earth based launches.

2. Being that there are still plenty of unknowns in there*, can we even say that the conservative numbers are achievable in the near term?

*(At least from the lack details of where your numbers are coming from)


3. I do agree with that. I can't see exploitation of resources to be self sufficient. I can only see it as a natural extension of other research.

1. That's a good thing.

2. According to Dr. Spudis's latest article in the Smithsonean Air & Space Magazine:

[People tend to assume] an initial level of effort that implies more capability than is needed, at least initially. If the polar water is present in the quantities we estimate, a very small initial operation can harvest sufficient water to supply first the needs of the outpost, then to re-fuel the lunar lander, and finally make enough for export. A system that can collect and process a few metric tonnes per month is small and deliverable in one robotic landing. The machines needed can be teleoperated from Earth, something not possible for resource processing on asteroids or Mars.

People tend to have the impression that resource processing on the Moon requires a high level of industrial capability. Such is not the case; we start off small, use remote operations to build up an initial capability, and when people arrive, a resource processing infrastructure is already up and operating. It is then extended until needed capacity is reached.

Alternatively, we could just “study the technology” for ten years while doing nothing.

3. I think the lunar economy will eventually be running trade surpluses. However, one thing we can both agree on is that we will never find out unless we--or someone else--tries to do it.
________________________________________
"The Moon is on the critical path to human expansion into the Solar System."--Paul Spudis

[Antice]

as for robot's VS humans issues that is inevitable to come up again. humans on the moon are there to do science and engineering first and foremost. and repair the service bot's second. Automation can go a long way when enhanced by teleoperation. but it cannot go all the way. neither should it have to. all the telerobots needed for the robot + man scenario already exist on earth. so it is a case of adapting what we have rather than inventing anything new. the entire operation may not take more than 1 crew-member at any one time. the rest of the base is filled with scientists and geologists.
There may be times when the entire crew has to work on fixing emergencies with the ISRU equipment, but it will be offset by long periods where no crew-member has to do anything with it. the manned base is an insurance policy against major failures shutting the entire operation down for months at a time.

[Warren Platts]

There was a quote from the Frank Zegler paper that discussed why humans would be necessary at any Moon base where actual delivered functionality is desired and taken seriously:

Initial crewed missions have 120 day stay durations and overlap by approximately 10 days with the newly arrived crew. The intent was to have continuous human occupation of the lunar base once it was established. This allows maintenance to be done on the increasingly complex life support and scientific equipment on the lunar surface. It is never left unattended. This leverages one of the most powerful attributes of a human crew- the ability to rapidly address contingencies with minimal preplanning.

________________________________________
"The Moon is on the critical path to human expansion into the Solar System."--Paul Spudis

[Hop_David]

1. That's what the Frank Zegler paper, page 25 said. But I think that was based on the ordinary DTAL with the ACES-41. It would be better to design a dedicated unmanned tanker-lander based on the ACES-71 tanker.

2. the Frank Zegler paper, Table 1. and Table 2. (page 11-12) The ACES-71 holds 71 total tons of fuel. The dry mass for the basic version is 5.5 tons. Add 2 more tons for the descent equipment (lateral thrusters and landing gear). That makes a total dry mass of 7.5 tons.

Using your figures, the dry mass to go back and forth would require 15 tons of propellant. That leaves 55 tons. Thus: mass delivered to depot = 55 / 1.73 = ~30 tons per trip.

Looking over this, you math looks right. So for every 71 tons of propellent manufactured, 30 tons to EML1 (or 2)? That's not too bad. If the modified ACES 71 is a reusable lander/ascent vehicle that could be a game changer.

To supply the year 3 projected L2 depot propellant flux of 243 tons (the Frank Zegler paper, Table 5, page 22), that would only require 8 launches from the Moon's surface.

Well assuming 30 tons per launch is a deviation from the 7 tons per launch proposed earlier. But I believe this could be doable if the tanker is unmanned, as you suggest.

8 launches over 3 years. So a launch every 4.5 months? So the lunar ISRU plant would be making 71 tons of propellent every 4.5 months?

Has anyone done an estimate on the mass of infrastructure capable of making 71 tons lunar ISRU propellent each 4.5 months?

It would be interesting to see how a reusable basic ACES-71 tanker would perform making the run from L2 to LEO. Dry mass is only 5.5 tons, but the delta v is 3.5 km/s each way. Table 2 says the ACES-71 could deliver 29 tons to L2 from LEO, if you throw away the tanker at the end of the run. So how much propellant do you think it would take to dead head it back to the starting point?

3.5 km/sec from EML2 to LEO? I guess you're using Vanilla's (aka Kirk Sorensen) route? I haven't grokked or been able to replicate his results but I'll admit he's more knowledgeable than I.

At first glance it looks a round trip delta V budget of 7 km/sec. But in my opinion this is a vast overestimation since it relies on propellent to circularize the orbit in LEO. I believe most of the delta V needed for LEO circularization could be achieved by drag passes through the upper atmosphere.

It's true that the atmosphere would inflict some abuse on the craft during each perigee drag pass. But nothing remotely close to the aerobraking abuse the shuttle endures (shedding 8 km/sec over an hour).

I floated this notion at Nasa Spaceflight. As you know, there are experienced engineers in that forum that aren't shy about ripping a hare brained scheme to shreds. I was somewhat relieved when Hell's Own Puppy weighed in that MRO had used a similar scheme with only mild abuse suffered.

Using drag passes, a round trip from EML1 to LEO and back would take only 4.3 km/sec rather than 7.6 km/sec. (in that Nasa Spaceflight thread I made an arithmetic error: LEO circularization is 3.1, not 3.8. Dropping out of EML1 is .7 km/sec. So total delta V for one leg is 3.8 km/sec).

As I mentioned, I don't understand Vanilla's route to EML2. But I am guessing drag passes could shave about 3 km/sec off his round trip budget also. Using drag passes, I believe a Vanilla round trip would have a 4 km/sec delta V budget.

ETA: Well, just interpolating by eye, it would take about 4 tons to send the ACES-71 tanker back empty to L2 after making a delivery to LEO,

If I remember right, exhaust velocity of hydrogen and Lox is 4.4 km/sec.

Most delta Vs for LEO to HEO are around 3.1 km/sec (but maybe Sorensen's path is different, I don't know)

e^(3.1/4.4) - 1 is 1.023. So you'd need about the same propellent mass as the dry mass of the ACES tanker for the return trip. So I get 5.5 tons of propellent for the return trip to L2.

so it would take about 2 1/2 launches from the Moon in order to be able to deliver 20 tons to LEO. The year 3 projected LEO propellant flux is 485 tons, so it would take about 20 deliveries from the space tanker, which in turn would require 50 launches from the tanker lander. Thus for an aggressive plan, there might be 60 launches per year; about 5 per month; this would require a total production of over 4,000 tons of propellant. This would require digging a hole about half the size of a football field to a depth of 2 meters. The processing plant would have to process about one 20 ton truckload every other day.

This would eliminate the need for 22 launches of propellant and their associated tankers and RL-10 engines from Earth (728 tons to LEO @ 33 tons each). (Compared to my figure of 1800 tons to L2 per years.... I guess the difference is the wholly reusable architecture. No, the difference is I've got it backwards. Since the flow is from L2 to LEO, the LEO propellant requirement to support the lunar base would be vastly reduced since it would only be modules and people--rather than ever more propellant--that would need the propellant. So my analysis above is major overkill.)

First time I tried to read the above, my eyes glazed over. I think I could understand it with a little time and effort. But does it rest on the assumption of a 7 km/sec delta V budget for round trips between LEO and EML2? If so, I believe it's based on a wrong assumption. As I mentioned, with drag passes, round trip would be more like 4 km/sec.

[Warren Platts]

Looking over this, you math looks right. So for every 71 tons of propellent manufactured, 30 tons to EML1 (or 2)? That's not too bad. If the modified ACES 71 is a reusable lander/ascent vehicle that could be a game changer.

It's in the right ballpark, I think; Paul Spudis and Dennis Wingo did a study on the old NASA Altair and they thought a resuable SSTO version might be able to deliver 25 tons. This is using LH2/LO2. Supposedly, an AL/LH2/LO2 tripropellant can deliver an even higher Isp with more density.

Well assuming 30 tons per launch is a deviation from the 7 tons per launch proposed earlier. But I believe this could be doable if the tanker is unmanned, as you suggest.

I think it's worth a try.

8 launches over 3 years. So a launch every 4.5 months? So the lunar ISRU plant would be making 71 tons of propellent every 4.5 months?

No, I was thinking 8 launches per year. Also you would want to have at least 3 or 4 tankers on hand in case one breaks down or crashes.

Has anyone done an estimate on the mass of infrastructure capable of making 71 tons lunar ISRU propellent each 4.5 months?

Dr. Spudis claims that a single lander load can deliver an initial ISRU propellant plant that's totally teleoperated from Earth ought to be capable of "several" tons of propellant per month. So let's say it's 10 tons per month. In that case it would take several months to do one load. On the other hand, the Zegler (or Kutter) paper does make an intriguing claim about using the DTAL ascender in a reusable format for use as a sort of lunar "helicopter" for doing suborbital sortie missions away from the base. It would be interesting to wargame various scenarios for the minimum propellant needs for such a craft. Is there an easy way to calculate the delta v for suborbital "hops"?

3.5 km/sec from EML2 to LEO? I guess you're using Vanilla's (aka Kirk Sorensen) route? I haven't grokked or been able to replicate his results but I'll admit he's more knowledgeable than I.

I was just going by what the wiki "delta v budget" article says (which is in turn based on an older paper by Wendell Mendell. ) We're just doing round figures anyway. I figure 6 km/s LEO to lunar surface; 2.5 to L2; then 3.5 to LEO.

At first glance it looks a round trip delta V budget of 7 km/sec. But in my opinion this is a vast overestimation since it relies on propellent to circularize the orbit in LEO. I believe most of the delta V needed for LEO circularization could be achieved by drag passes through the upper atmosphere.

It's true that the atmosphere would inflict some abuse on the craft during each perigee drag pass. But nothing remotely close to the aerobraking abuse the shuttle endures (shedding 8 km/sec over an hour).

I floated this notion at Nasa Spaceflight. As you know, there are experienced engineers in that forum that aren't shy about ripping a hare brained scheme to shreds. I was somewhat relieved when Hell's Own Puppy weighed in that MRO had used a similar scheme with only mild abuse suffered.

Using drag passes, a round trip from EML1 to LEO and back would take only 4.3 km/sec rather than 7.6 km/sec. (in that Nasa Spaceflight thread I made an arithmetic error: LEO circularization is 3.1, not 3.8. Dropping out of EML1 is .7 km/sec. So total delta V for one leg is 3.8 km/sec).

As I mentioned, I don't understand Vanilla's route to EML2. But I am guessing drag passes could shave about 3 km/sec off his round trip budget also. Using drag passes, I believe a Vanilla round trip would have a 4 km/sec delta V budget.

This is interesting. I kind of forgot about possible aerobraking; the Zegler philosophy is not to worry about delta v overmuch; depots make possible aggressive use of rockets. However, the deltav from L2 to LEO in that case is only about 1/3 km/s. So you'd probably get a great mass fraction. Only problem is we're dealing with cryogenic fuels. Especially LH2 is going to have a hard time. Also, under the atmosphere grazing technique, that would require multiple passes, spiraling orbits, and a fair amount of time wouldn't it? Time factor wise, using propellant to slow down might be able to do several trips in the time it would take for one atmospheric grazing slow down.



If I remember right, exhaust velocity of hydrogen and Lox is 4.4 km/sec.

Most delta Vs for LEO to HEO are around 3.1 km/sec (but maybe Sorensen's path is different, I don't know)

e^(3.1/4.4) - 1 is 1.023. So you'd need about the same propellent mass as the dry mass of the ACES tanker for the return trip. So I get 5.5 tons of propellent for the return trip to L2.

First time I tried to read the above, my eyes glazed over. I think I could understand it with a little time and effort. But does it rest on the assumption of a 7 km/sec delta V budget for round trips between LEO and EML2? If so, I believe it's based on a wrong assumption. As I mentioned, with drag passes, round trip would be more like 4 km/sec.

OK, so if they can get 29 tons to L2 from LEO using a throwaway tanker, we should be able to send a full tanker from L2 to the LEO depot and have 29 tons left when it gets there. It drops off 23 tons, and then 6 tons should get it back with a little margin left over. That seems kind of extravagant, but I guess that's the price you pay. What if you put the first depot at C3 instead of L2 or L1?? That would break up the journey almost in half. Alternatively, maybe have a series of three depots.

[Hop_David]

I was just going by what the wiki "delta v budget" article says (which is in turn based on an older paper by Wendell Mendell. ) We're just doing round figures anyway. I figure 6 km/s LEO to lunar surface; 2.5 to L2; then 3.5 to LEO.

I believe sdsd, one of the participants at NasaSpaceflight Forum, works on a lot of Wikipedia orbital mechanics pages. I wouldn't be surprised if he got that figure from Vanilla (aka Sorensen).

If we're rounding figures, straightforward Hohmann's would give closer to 4 km/sec from EML2 to LEO. But I believe Vanilla has found a less expensive route to EML2 via lunar gravity assist. Counterintuitively, it's cheaper to get to EML2 than it is EML1 (at least according to Sorensen).

This is interesting. I kind of forgot about possible aerobraking; the Zegler philosophy is not to worry about delta v overmuch; depots make possible aggressive use of rockets. However, the deltav from L2 to LEO in that case is only about 1/3 km/s. So you'd probably get a great mass fraction. Only problem is we're dealing with cryogenic fuels. Especially LH2 is going to have a hard time. Also, under the atmosphere grazing technique, that would require multiple passes, spiraling orbits, and a fair amount of time wouldn't it? Time factor wise, using propellant to slow down might be able to do several trips in the time it would take for one atmospheric grazing slow down.

In the Nasa Spaceflight thread, gbalkie noted that perigee speed of transfer orbit to geosynch isn't that much less than transfer orbit to Luna. So with a just slightly more aggressive first drag pass, you would lower apogee to geosynch altitude (or below) and thus have less than 1 day waits between drag passes.

Shaving 3 km/sec off your delta V budget can make a huge difference if you're trying to design a reusable tanker for round trips between LEO and EML1 (or 2).

And Lunar oxygen doesn't mind too much if you take a couple of weeks to get to LEO. Hydrogen's another story. Although there's methane and other carbon volatiles at the poles, so fuels less vulnerable to boil off might be an option.

In the May issue of Discover there were some interesting articles on the moon. Alberto Saal had some arguments that the proposed collision forming the moon didn't boil off all the lunar volatiles those billions of years ago. I recall rolling my eyes in disbelief when you mentioned the possibility of lunar vents emitting volatiles. But after reading Saal's arguments, I no longer see that notion as so outlandish.

That issue had an article on Bill Stone and his Shackleton Energy Company (starting on page 61). Stone hopes to set up an ISRU lunar propellent business. I wouldn't give that horse even odds, but who knows? I was happy to see he hopes to employ aerobraking "...propellent could then be cheaply shuttle to stations in Earth orbit on inflatable craft that save on fuel by relying on aerobraking — briefly dipping into Earth's upper atmosphere — to slow down when entering orbit."

[Warren Platts]

I was just reading some stuff by Jon Goff and that Transterrestrial Musings guy. They point out an after the fact obvious point that a space tug combined with a fuel depot is functionally equivalent to running a string of two fuel depots. Above I was thinking it would be nice to break up the journey from L2 to LEO for a tanker delivering lunar propellant. So how about this: the lander tanker makes the run to the L2 depot: dV = 2.5 km/s. Then an ACES-41 space tug gives an ACES-71 space tanker a 1 km/s boost. Since the L2 to LEO dV is about 3.5 km/s, then the tug booster transforms the ACES-71 tanker into the top stage of a two stage vehicle. Thus from it's perspective, then ACES tanker only has to do 2.5 km/s, and would be able to deliver 30 tons with enough left over to make it back to L2.

So I'm leaning against using aerobraking. It's just too much trouble for what it's worth in savings. Either you have to haul around an expensive and heavy TPS , or it's going to take forever using the grazing technique. Meanwhile, we are talking about cryogenics here after all. Using frictional heat to slow cryogenic fuel is counterintuitive at best.

Plus the whole philosophy behind aerobraking is starting to bug me. Do you see what I mean? Aerobreaking is a product of the philosophy of scarcity, that we must always scrimp and save and do whatever is cheapest, the bare minimum of what is possible. We should approach space with the philosophy of abundance: space contains material resources beyond our wildest dreams. If it seems extravagant to skip aerobraking--I say: So what? The LEO to L2 run can't use aerobraking, so we shouldn't expect the architecture to have to support aerobraking for the reverse run. It just complicates things, and the savings are marginal at best.

[Hop_David]

Plus the whole philosophy behind aerobraking is starting to bug me. Do you see what I mean? Aerobreaking is a product of the philosophy of scarcity, that we must always scrimp and save and do whatever is cheapest, the bare minimum of what is possible. We should approach space with the philosophy of abundance: space contains material resources beyond our wildest dreams. If it seems extravagant to skip aerobraking--I say: So what?

So you're still holding out for an extravagant space budget? You're using the same Veruca Salt strategy employed by Bob Zubrin and his followers. And you're getting the same results as Veruca and Zubrin.

The LEO to L2 run can't use aerobraking, so we shouldn't expect the architecture to have to support aerobraking for the reverse run. It just complicates things, and the savings are marginal at best.

3 km/sec isn't marginal. If you want single stage reusable tanker, 4 km/sec may be doable, 7 km/sec is more difficult. Nearly doubling the exponent in the rocket equation also torpedoes the mass of your payload.

Either you have to haul around an expensive and heavy TPS , or it's going to take forever using the grazing technique.

Two or three weeks is forever? Too much time to waste for a measly 3 km/sec?

It takes an extra 4 or 5 extra days to reach EML2 from LEO (vs LEO to EML1). For what? A .3 km/sec savings? (3.5 for EML2 vs 3.8 for EML1). Using your reasoning, the ULA's choice of EML2 was a major blunder.

[Warren Platts]

So you're still holding out for an extravagant space budget? You're using the same Veruca Salt strategy employed by Bob Zubrin and his followers. And you're getting the same results as Veruca and Zubrin.

I can always tell I'm winning a debate when the opposition is reduced to gratuitous insults....

The LEO to L2 run can't use aerobraking, so we shouldn't expect the architecture to have to support aerobraking for the reverse run. It just complicates things, and the savings are marginal at best.

3 km/sec isn't marginal. If you want single stage reusable tanker, 4 km/sec may be doable, 7 km/sec is more difficult. Nearly doubling the exponent in the rocket equation also torpedoes the mass of your payload.

I don't know where you're getting these figures, but it doesn't matter. You have to understand that the rocket equation has more than one factor. Time and time again, people get engrossed with one factor of that equation, and then run with it, blinders on, and ignore everything else. With you and Isaac Kuo, your favorite factor is delta v. With Antice, it's Isp. With Robert Clark, it's fuel density. But the complicated fact is, that the outcome of any given trade study is going to depend on a lot of factors--and not all of these factors are even part of the rocket equation.

In the present example, we're wondering whether using aerobraking a cryogenic cargo is necessarily worth it. On the one hand, if it is worth it to not use aerobraking in one direction, then it would stand to reason that it must also be worth it to not use aerobraking in the other direction. Therefore, nothing is getting "torpedoed" by forgoing aerobraking.

The question then becomes whether the marginal increase in mass delivered is worth all the costs associated with aerobraking. If you use a full up TPS, then you must pay for it in dry mass, and construction and development costs. If you use the grazing technique, then you must pay for it: (1) in terms of turn around time: the delay may mean you have to have more tankers than would otherwise be necessary (an extreme example of this problem is the 7-month turnaround time of a VASIMR^TM space tug, where it requires a fleet of 8-10 VASIMR^TM space tugs to do the work of one ACES-71 space tanker); (2) there is the problem of exacerbated boil-off from (a) the friction associated with the atmosphere itself; (b) the delay induced by grazing in a thermally unfriendly environment. I don't know how much of a problem this is. But then, neither do you.

Then there is the economic factor. It may turn out that making propellant from lunar ice is more difficult than first imagined; in that case, the savings in propellant gained by aerobraking are worth a lot. On the other hand, it may turn out that making propellant from lunar ice is so easy, that it can be produced at will far in excess of the demand for it; in that case, given that we're using fully reusable SSTO's to launch the stuff, it doesn't matter how much we burn to get some of it to LEO.

Bottom line: the answer to the question whether aerobraking lunar cryogenic propellant exported to LEO is not obvious. It would require a study at the least to examine these issues. Just looking at delta v alone doesn't cut it as a trade study. That's just handwaving.

It takes an extra 4 or 5 extra days to reach EML2 from LEO (vs LEO to EML1). For what? A .3 km/sec savings? (3.5 for EML2 vs 3.8 for EML1). Using your reasoning, the ULA's choice of EML2 was a major blunder.

There you go again, taking one single factor out of the rocket equation, and let that decide everything. The fact is, L2 was selected for the favorable thermal environment so as to reduce boiloff rates of liquid hydrogen--not for a measly 300 meters per second savings in delta v.
_____________________________________
"If you can't handle the MIG's, then get out of MIG alley."--Dan Ackroyd

[Hop_David]

On the one hand, if it is worth it to not use aerobraking in one direction, then it would stand to reason that it must also be worth it to not use aerobraking in the other direction.

Let me get this straight. Since we don't use aerobraking for earth surface to LEO, we shouldn't use aerobraking for LEO to earth surface?

No, this argument doesn't stand to reason.

[Hop_David]

... lunar cryogenic propellant ...

In case you missed it the first time, I'll repost what I wrote earlier:

And Lunar oxygen doesn't mind too much if you take a couple of weeks to get to LEO. Hydrogen's another story. Although there's methane and other carbon volatiles at the poles, so fuels less vulnerable to boil off might be an option.

Lunar oxygen is abundant and is an obvious propellent commodity. It is the lion's share of propellent mass. Exporting lunar oxygen alone would be worthwhile.

[Hop_David]

There you go again, taking one single factor out of the rocket equation, and let that decide everything. The fact is, L2 was selected for the favorable thermal environment so as to reduce boiloff rates of liquid hydrogen--not for a measly 300 meters per second savings in delta v.

EML1 is also a benign thermal environment.

According to Near Term Propellent Depots by Zegler, Kutter, Goff, Chandler Marchetta:

The severe thermal environment in LEO causes a
substantial amount of propellant boil-off over the course of a year. EML-1 or -2 depots are in a much more benign
thermal environment, with very low boil-off levels. This leads to the conclusion that the best way to use a depot
system like this is to forward propellants on from the LEO depot to the EML-1 or -2 depot as quickly as possible.

[Warren Platts]

Let me get this straight. Since we don't use aerobraking for earth surface to LEO, we shouldn't use aerobraking for LEO to earth surface?

No, this argument doesn't stand to reason.

OK, let me get this straight: even though we don't use aerobraking for moon surface to LLO, we should use aerobraking for LLO to Moon surface?

Yes, your argument does not stand to reason.

In case you missed it the first time, I'll repost what I wrote earlier:

And Lunar oxygen doesn't mind too much if you take a couple of weeks to get to LEO. Hydrogen's another story. Although there's methane and other carbon volatiles at the poles, so fuels less vulnerable to boil off might be an option.

]Lunar oxygen is abundant and is an obvious propellent quantity. It is the lion's share of propellent mass. Exporting lunar oxygen alone would be worthwhile.

You still don't understand my point because you view the universe through delta-v tinted glasses--there's an idea for your next cartoon--and like the citizens of the Emerald City, you refuse to take them off because you would have to deal with reality as it is, rather than as reduced to a single factor of a rocket equation. I didn't miss your earlier point--I rejected it. For a number of reasons:

1. It's still going to take a disposable rocket from Earth with a huge tank to loft LH2 to LEO
2. There is no evidence despite your bald assertion that methane or other carbon volatiles exist at the poles in quantities sufficient for burning as rocket fuel (aluminum is the only viable alternative lunar rocket fuel
3. The RL-10 uses LH2
4. While LO2 may not mind a couple of extra weeks delay in getting to a LEO fuel depot, it may very well mind cruising through the Earth's atmosphere at hypersonic velocities in an uninsulated tank (BTW I wanted to ask you how you thought the ACES-71 should enter the atmosphere: forwards, backwards, or perhaps sideways?).

This is what I mean when I say that aerobraking may not be worth the trouble. Even if LO2 boiloff issues associated with heating a thin, uninsulated tank are negligible (a finding that must be shown, and not merely assumed), you have added another layer of cost and complexity that must be balanced: either ship LH2 from Earth subject to a 10 km/s delta v penalty, or substitute lunar aluminum, which will probably entail a major redesign of the RL-10 at a minimum. Either option costs money. Aerobraking is not for free delta v.

I'm not married to the idea that aerobraking not be used. I'm all for it--if the trade warrants it. My main point is that the philosophy of scarcity can result in a blinkered view of the possibilities of space travel. Ironically, the philosophy of scarcity is what gave us CxP. The idea was that space travel is just so damned hard and expensive, only monster SRB-powered HLV's can get the job done. It was all or nothing: either we had to bite the bullet and double the budget, or we can't do anything but sit around and produce billion dollar viewgraphs. The UMF architecture rejects that we must overeat or starve. There is a healthy medium. We skip the 24 oz steak and double portion of twice baked potatoes; yet our diet contains plenty of margin, so that we can always run that extra mile or two without running out of breath and our blood sugar crashing to zero. Delta v is not the sole currency. And this is not my idea. The ULA AIAA white papers make this point explicity.
_______________________________
"If you can't handle the MIGs, then get out of MIG alley."--Dan Ackroyd

[Warren Platts]

EML1 is also a benign thermal environment.

According to Near Term Propellent Depots by Zegler, Kutter, Goff, Chandler Marchetta:

The severe thermal environment in LEO causes a
substantial amount of propellant boil-off over the course of a year. EML-1 or -2 depots are in a much more benign
thermal environment, with very low boil-off levels. This leads to the conclusion that the best way to use a depot
system like this is to forward propellants on from the LEO depot to the EML-1 or -2 depot as quickly as possible.

Good job. You're finally starting to dig up some AIAA papers on your own. But it undermines your point that delta v is the sole overriding consideration, doesn't it? Clearly, you've uncovered evidcence that ULA is not married to the L2 location; apparently, L1 is also acceptible, despite the delta v cost.

I'm not exactly sure why L1 was not even mentioned in the Affordable Architecture paper. Apparently, it was a nod to the Zubrinistas; you would get a better Oberth effect boost around Planet Earth from L2 as opposed to L1 for a Mars transfer orbit, if I'm not mistaken.
___________________________
"If you can't handle the MIGs, then get out of MIG alley."--Dan Ackroyd

[Warren Platts]

1. How big a facility are we talking ? As in how many tons of payload must be placed into the lunar surface to make a facility to gather, extract and launch fuel/propellant from the moon ?

2. How many tons of fuel/propellant can this facility shoot back into space each year

People keep asking about this: I stumbled across a discussion of an actual NASA architecture drawn up in the 1990's: the LUNOX plan: http://www.astronautix.com/craft/lunox.htm

The NASA/JSC LUNOX proposal of 1993 tried to reduce the cost of maintaining a First Lunar Outpost by producing liquid oxygen propellant for the return to Earth from lunar soil. This allowed smaller launch and space vehicles, and reduced costs by 33% to 50%. . . . LUNOX tried to use the same [ISRU] strategy to reduce the cost of pioneering small-scale manned lunar exploration as well. This technique would permit drastic downsizing of the manned space vehicles and cost reductions of up to 50%.

The unmanned Phoenix/LUNOX lander would carry a lunar oxygen production plant/storage facility and a nuclear power reactor to the moon's surface. The total mass of this "package" was 12,454 kilograms. The mass summary was as follows:

• Phoenix Unmanned Lander: 4,717 kg dry, 16,578 kg propellants
• Cargo: 12,454 kg
• Translunar Injection Stage: 6,130 kg dry, 43,930 kg propellants Total Mass In Low Earth Orbit 83,809 kg

A second unmanned lander would deliver six small robotic lunar vehicles, after the LUNOX plant and reactor had been deployed. Two "Loader" bulldozers would collect and sort ~500kg/hour of ilmenite-rich lunar soil, which would be fed into the LUNOX plant and processed into liquid oxygen propellant. The process was based on H2 reduction of lunar regolith in a fluidized-bed reactor, solid-state high-temperature electrolysis and Stirling-cycle O2 liquefaction and refrigeration. Two "Tankers" transported the liquid oxygen from the LUNOX factory to other users on the lunar surface ( the annual production capability was 24 000kg, assuming an extraction efficiency of 4% after benefaction). Finally, two "Haulers" were available for moving heavy equipment across short distances. All rovers use regenerative power sources which were recharged from the nuclear reactor surface power supply.
. . .

• Phoenix Manned Lander (4 X 31 150kn Thrust Engines; 4:1 Throttling): 5505 kg dry; 16944 kg propellants
• Cargo: 2000 kg
• Lh2 Fuel for Return Trip: 2492 kg
• Lunar Oxygen Required: 10,165 kg

The Phoenix was rather unusual since it featured a side-mounted Apollo/FLO type crew capsule for improved pilot visibility during landing, plus centerline propellant tanks and cargo modules for better payload unloading characteristics. The vehicle consisted of a single descent/ascent stage since there was little or no performance advantage to Apollo-type separate stages and lunar orbit rendezvous when using lunar-based propellants. The propulsion and avionics systems were common with the unmanned cargo lander.

Interesting that the lander is similar to the ULA ACES/DTAL lander in that they both use the horizontal configuration with LO2/LH2.

But anyways, it shows how some significant ISRU can be accomplished with just a couple of modules. 24 tons per year from the very first stages of a Moon base wouldn't be too shabby.

But wait! That 24 tons is 4% of total regolith processed. 24 / 0.4 = 600 tons of regolith processed. So imagine if the plant were digging up 2 m thick blocks of relatively pure ice instead of regular regolith: The amount of fuel produced by a single such plant would be measured in hundreds of tons.

I think I am forced to revise my estimates of how fast it would take to get to 100-1000 ton per year production level. It probably could be producing at that level as soon as year 3 if the ULA Moon First (UMF) architecture was implemented in a serious way.
____________________________
"If you can't handle the MIGs, then get out of MIG alley."--Dan Ackroyd

[Hop_David]

OK, let me get this straight: even though we don't use aerobraking for moon surface to LLO, we should use aerobraking for LLO to Moon surface?

Yes, your argument does not stand to reason.

You made a very general conditional statement.

Here it is again:

... if it is worth it to not use aerobraking in one direction, then it would stand to reason that it must also be worth it to not use aerobraking in the other direction.

Does earth surface to LEO meet the condition? Why yes, it is worth it to not use aerobraking in one direction. In fact, you can't use aerobraking in the ascending direction, aerobraking is a one way ticket. This applies to any gravity well with an atmosphere at the floor.

However aerobraking can be and is used for the trip down.


For a general conditional statement, no amount of examples can conclusively prove it. But one counter example suffices to refute it.

For example, I can refute "If it bears live young rather than laying eggs, it's a mammal." by pointing to species of fish and snakes that bear live young. You could point to dogs and say they bear live young. But that doesn't counter my refutation. And neither does your contrived moon example counter my refutation of your general argument regarding aerobraking.

[Warren Platts]

Oh brother!

Thanks for the lesson in formal logic, Hop. But that's why I added the logical qualifier "it stands to reason", which is logically functionally equivalent to "ceteris parabus". The ceteris parabus qualifier implies that there very well might be exceptions to the general rule.

In your contrived example, it is physically impossible for most Earth-based launch vehicle designs to do fully propulsive landings. In that case, other things are not equal. However, the wide availability of supercheap lunar fuel would make DC-X style Earth-based FRSSTOFPLLV's much more doable.

More importantly, you missed my main point, which was to refute the generalization you appear to subscribe to, that if aerobraking is possible, then fully propulsive slow-downs never make sense.
_______________________________
"If you can't handle the MIGs then get out of MIG alley."--Dan Ackroyd

[Hop_David]

Oh brother!

Thanks for the lesson in formal logic, Hop. But that's why I added the logical qualifier "it stands to reason", which is logically functionally equivalent to "ceteris parabus". The ceteris parabus qualifier implies that there very well might be exceptions to the general rule.

I don't interpret "It stands to reasons" to mean "Sometimes, but not always".

The Free Dictionary says "It stands to reason" means "It is obvious or logical"

If someone were to say "If animals bear live young, it stands to reason they must be mammals" I would interpret that as "If animals bear live young, it is obvious they must be mammals."

In both cases, no number of examples can prove, but one counter example suffices to refute.

And so it is with

...if it is worth it to not use aerobraking in one direction, then it would stand to reason that it must also be worth it to not use aerobraking in the other direction.

More importantly, you missed my main point, which was to refute the generalization you appear to subscribe to, that if aerobraking is possible, then fully propulsive slow-downs never make sense.

I subscribe to the notion that shaving 3 km/sec off your delta V budget will enable a larger mass delivered to LEO. I subscribe to the notion that it is less challenging to design a reusable tanker for a 4 km/sec delta V budget than it is a 7 km/sec delta V budget.

I don't recall arguing that fully propulsive slow downs never make sense. At any rate, your flawed argument refuted nothing.

[Hop_David]

There is no evidence despite your bald assertion that methane or other carbon volatiles exist at the poles in quantities sufficient for burning as rocket fuel

For the record, this is what I wrote:

Although there's methane and other carbon volatiles at the poles, so fuels less vulnerable to boil off might be an option.

In what universe does "might be an option" equal a bald assertion they exist in sufficient quantities?

[Hop_David]

It's still going to take a disposable rocket from Earth with a huge tank to loft LH2 to LEO

For the fore seeable future, all upper stages will be lofted from earth with disposable rockets. The moon won't have the manufacturing infrastructure to make upper stages for some time.

This includes the dry mass of the upper stage's hydrogen tanks and oxygen tanks.

If lunar oxygen is available in LEO, this upper stage can be lofted from earth with an empty oxidizer tank. Oxidizer is about 8/9 of the mass of Lox LH2 propellent.

So lunar oxygen in LEO could have a huge impact on what sort of upper stages can be lofted to LEO.

[Warren Platts]

I don't interpret "It stands to reasons" to mean "Sometimes, but not always".

The Free Dictionary says "It stands to reason" means "It is obvious or logical"

If someone were to say "If animals bear live young, it stands to reason they must be mammals" I would interpret that as "If animals bear live young, it is obvious they must be mammals."

Then you would be wrong: dictionary.com defines "it stands to reason as": "It's reasonable or to be expected. For example, It stands to reason that if we leave late we'll arrive late. [Early 1600s]"

If we leave late, does it necessarily follow that we'll arrive late? No. But other things being equal, it does.

I subscribe to the notion that shaving 3 km/sec off your delta V budget will enable a larger mass delivered to LEO. I subscribe to the notion that it is less challenging to design a reusable tanker for a 4 km/sec delta V budget than it is a 7 km/sec delta V budget.

This is yet another example of simplistic slinging around of concepts like delta v because you ignore mass. The return trip mass is a fraction of the initial mass, because you drop off your cargo, and go back empty. The propellant requirements for each leg of the trip are quite different. To just add up the delta v as if that's informative--that's not how it works.

ETA:

Oxidizer is about 8/9 of the mass of Lox LH2 propellent.

Most LH2/LO2 motors run rich with mixing ratio (by mass) of LO2 to LH2 of 5/6, rather than 8/9.

____________________________
"I have learned to use the word 'impossible' with the greatest caution."--Wernher von Braun

[Hop_David]

Then you would be wrong: dictionary.com defines "it stands to reason as": "It's reasonable or to be expected. For example, It stands to reason that if we leave late we'll arrive late. [Early 1600s]"

If we leave late, does it necessarily follow that we'll arrive late? No. But other things being equal, it does.

So if it doesn't pay to use aerobraking on an upbound leg, it's reasonable that it doesn't pay for the downward leg?

Sorry, still don't follow this argument.

This is yet another example of simplistic slinging around of concepts like delta v because you ignore mass. The return trip mass is a fraction of the initial mass, because you drop off your cargo, and go back empty. The propellant requirements for each leg of the trip are quite different. To just add up the delta v as if that's informative--that's not how it works.

Well, aerobraking shaves 3 km/sec off the downward leg. The leg when is the tanker is full of lunox. On the trip back to EML1 or 2, the tanker is empty. It is the upbound leg aerobraking can't help with.

Your observation is correct, but it's an observation that favors my argument.

ETA: Most LH2/LO2 motors run rich with mixing ratio (by mass) of LO2 to LH2 of 5/6, rather than 8/9.

Even so, an empty oxidizer tank in an upper stage would be a big mass savings.

[Hop_David]

Good job. You're finally starting to dig up some AIAA papers on your own. But it undermines your point that delta v is the sole overriding consideration, doesn't it?

That paper addresses the plausibility of propellent storage, not the depots role in an architecture.

The role EML2 depots plays is addressed in An Affordable Exploration Architecture by Zegler, Kutter & Bar

Objects that are located at L2 have already had a substantial amount of energy pumped into them. They
have effectively reached earth escape velocity while still being bound to the earth. This energy constitutes a
substantial portion of the total delta V required to venture to Mars, for example. Many publications have
advocated the use of the LaGrange points for departure to the planets but a dedicated depot just for the
occasional Jupiter or Mars mission is clearly ineffective. With the road to L2 supported and maintained by
Lunar Exploration these departure strategies become very attractive.

Departure from L2 is further aided by incorporating a powered Earth gravitational assist into the mission
design. Objects being sent from L2 to Jupiter for example require only a small nudge to place them in a
trajectory towards earth with a very low perigee. As the departing spacecraft nears perigee it conducts a
burn when it is at an already high velocity. This activity effectively maximizes the delta V that can be
achieved from a fixed propellant mass. Rather than assembling stupendous propellant masses in LEO, as is
often proposed for a crewed mission to Mars, we can use this L2 departure strategy to gradually pump
energy into the objects we wish to send to Mars by caching them at L2 and also get a large delta V leverage
at the final departure. Much larger spacecraft can be sent to high C3 destinations using this approach.

Page 24

Clearly, you've uncovered evidcence that ULA is not married to the L2 location; apparently, L1 is also acceptible, despite the delta v cost.

The delta V advantage conferred by EML1 and EML2 is their chief virtue. It is a major virtue.

I'm not exactly sure why L1 was not even mentioned in the Affordable Architecture paper. Apparently, it was a nod to the Zubrinistas; you would get a better Oberth effect boost around Planet Earth from L2 as opposed to L1 for a Mars transfer orbit, if I'm not mistaken.

Falling from EML1 to a 300 km perigee gives a perigee velocity of 10.81 km/sec, just .11 short of the 10.92 km/sec escape velocity at that altitude. Falling from EML2 gives a perigee velocity of 10.84, which is .08 short of of the 10.92 km/sec escape velocity.

For exploiting the Oberth effect for Mars, asteroid or other destinations outside earth's neighborhood, EML2 has a .03 km/sec advantage over EML1.