Quote from: Warren Platts on 06/10/2010 10:11 amIt's a train wreck waiting to happen because it shoehorns you into to an exploration architecture that going to be exploring for ISRU propellant that it cannot use.Why use hypergolics? It's low Isp, it's heavy, it's toxic, it cannot be made from readily available lunar mineral resources. Boiloff is a nonissue. Boiloff will be used for station keeping in orbit, and for fuel cells during the lunar night.Sorry to be the bearer of bad news, but this idea will never fly.Depots filled with propellants manufactured off Earth will definitely fly. One day. In the 22nd century.
It's a train wreck waiting to happen because it shoehorns you into to an exploration architecture that going to be exploring for ISRU propellant that it cannot use.Why use hypergolics? It's low Isp, it's heavy, it's toxic, it cannot be made from readily available lunar mineral resources. Boiloff is a nonissue. Boiloff will be used for station keeping in orbit, and for fuel cells during the lunar night.Sorry to be the bearer of bad news, but this idea will never fly.
On the other hand, the FY2012 budget could fund construction of a hypergolic depot that would be filled with propellant manufactured on Earth.
Quote from: Warren Platts on 06/10/2010 10:11 amQuote from: Robotbeat on 05/29/2010 06:13 amA 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLOThe turnaround time is too slow. A 1MW SEP cannot supply propellant requirements for even a first generation lunar base. For example, if we take the one published lunar architecture out there that's been designed by actual engineers (Zegler et al., AIAA 2009-6567), by year 3, a 1st generation lunar base could be expected to go through close to 300 tons of propellant per year. A 1MW SEP can only deliver on the order of a few 10's of tons of propellant--every 7 months--if the VASIMRTM studies put out by Ad Astra are accurate....If you allow more time for turnaround (like a year or more), then you can have much, much more payload. Not good for time-sensitive payloads, but great for propellant. That's the thing about SEP: having a much longer turnaround gives you much more performance (since you are able to put proportionally more energy into the same amount of propellant), whereas a longer turnaround past a certain point for chemical is not helpful, since the energy you can put into the propellant is fixed.
Quote from: Robotbeat on 05/29/2010 06:13 amA 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLOThe turnaround time is too slow. A 1MW SEP cannot supply propellant requirements for even a first generation lunar base. For example, if we take the one published lunar architecture out there that's been designed by actual engineers (Zegler et al., AIAA 2009-6567), by year 3, a 1st generation lunar base could be expected to go through close to 300 tons of propellant per year. A 1MW SEP can only deliver on the order of a few 10's of tons of propellant--every 7 months--if the VASIMRTM studies put out by Ad Astra are accurate....
A 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLO
Quote from: Warren Platts on 06/10/2010 10:11 amQuote from: Robotbeat on 05/29/2010 06:13 amA 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLOThe turnaround time is too slow. A 1MW SEP cannot supply propellant requirements for even a first generation lunar base. For example, if we take the one published lunar architecture out there that's been designed by actual engineers (Zegler et al., AIAA 2009-6567), by year 3, a 1st generation lunar base could be expected to go through close to 300 tons of propellant per year. A 1MW SEP can only deliver on the order of a few 10's of tons of propellant--every 7 months--if the VASIMRTM studies put out by Ad Astra are accurate.Using the figures in Ad Astra's new paperhttp://www.adastrarocket.com/VASIMR_for_flexible_space_exploration.pdfA Survey of Missions using VASIMR® for Flexible Space ExplorationApril 2010JSC-65825A 500 kW VASIMR can deliver 14 mT of payload about every 6 months (or approx 28 mT per year) to low lunar orbit.300 mT / 28 = 10.7 tugsSo the Moon base can be supplied by launching 21 EELV year.
Quote from: Robotbeat on 05/29/2010 06:13 amA 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLOThe turnaround time is too slow. A 1MW SEP cannot supply propellant requirements for even a first generation lunar base. For example, if we take the one published lunar architecture out there that's been designed by actual engineers (Zegler et al., AIAA 2009-6567), by year 3, a 1st generation lunar base could be expected to go through close to 300 tons of propellant per year. A 1MW SEP can only deliver on the order of a few 10's of tons of propellant--every 7 months--if the VASIMRTM studies put out by Ad Astra are accurate.
Quote from: Robotbeat on 06/10/2010 05:30 pmQuote from: Warren Platts on 06/10/2010 10:11 amQuote from: Robotbeat on 05/29/2010 06:13 amA 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLOThe turnaround time is too slow. A 1MW SEP cannot supply propellant requirements for even a first generation lunar base. For example, if we take the one published lunar architecture out there that's been designed by actual engineers (Zegler et al., AIAA 2009-6567), by year 3, a 1st generation lunar base could be expected to go through close to 300 tons of propellant per year. A 1MW SEP can only deliver on the order of a few 10's of tons of propellant--every 7 months--if the VASIMRTM studies put out by Ad Astra are accurate....If you allow more time for turnaround (like a year or more), then you can have much, much more payload. Not good for time-sensitive payloads, but great for propellant. That's the thing about SEP: having a much longer turnaround gives you much more performance (since you are able to put proportionally more energy into the same amount of propellant), whereas a longer turnaround past a certain point for chemical is not helpful, since the energy you can put into the propellant is fixed.Great. And so how many modules of fuel tanks will have to be assembled in LEO to hold all this propellant? And how big/how many will the L2 depot have to be? This is the proplem: you guys will take a well-thought out plan like that proposed by the ULA engineers that have thought of everything. But then you'll obsess over one little detail that you think can be improved. So by solving one nonproblem, boiloff, you cause a whole slough of problems that cascade back through the architecture, and where it stops nobody knows.
Quote from: sdsds on 06/10/2010 05:41 pmOn the other hand, the FY2012 budget could fund construction of a hypergolic depot that would be filled with propellant manufactured on Earth.It could also fund a cryogenic depot.
If you have a three year time and only 20 tons of electric thruster propellant and no more than 22 tons of tug dry mass:...payloadm=(20.3+22-1.0779*(20.3+2))/(1.0779-1) = 234 tons of payloadyou could get hundreds of tons to EML2, but with the tug still able to just barely fit on a Delta IV heavy).
Quote from: Robotbeat on 06/10/2010 08:31 pmIf you have a three year time and only 20 tons of electric thruster propellant and no more than 22 tons of tug dry mass:...payloadm=(20.3+22-1.0779*(20.3+2))/(1.0779-1) = 234 tons of payloadyou could get hundreds of tons to EML2, but with the tug still able to just barely fit on a Delta IV heavy).Three years? To go to the freakin Moon?!? You've got to be kidding! Now that is one Inflexible Path. Hope your plans 3 years down the road are right on--that you won't need more or less than 243 tons. Also, hopefully your VASIMR won't break after 3 years of constant action--otherwise, those astronauts stranded on the Moon as a result are going to be SOL. And you'd probably want a heavy lift vehicle with gigantic farings to hold the tanks you'll need.
Quote from: Warren Platts on 06/10/2010 08:59 pmQuote from: sdsds on 06/10/2010 05:41 pmOn the other hand, the FY2012 budget could fund construction of a hypergolic depot that would be filled with propellant manufactured on Earth.It could also fund a cryogenic depot. Really? Construction of something like that sounds like something NASA could have in its 2012 budget? When would it have gone through PDR? In 2012 NASA could (and should) fund an umbrella shade for a Centaur flight, rather than forcing ULA to fund that themselves. And NASA could (and should) fund some design work for a small prototype cryogenic depot that might be constructed as soon as 2014 and flown in 2015. Based on the results of operating that prototype, NASA might design in 2016/2017, construct in 2018/2019, and fly in 2020 a full scale cryogenic depot. That gets you an empty depot in LEO ready to receive propellants by 2021. Maybe.
By then, a hypergolic depot could have been designed, manufactured, flown to LEO, filled, and flown to EML1 or EML2 using an existing cryogenic upper stage for the departure burn.
You still haven't explained what's the point of using hypergolics in the first place.
Quote from: A_M_Swallow on 06/10/2010 06:16 pmQuote from: Warren Platts on 06/10/2010 10:11 amQuote from: Robotbeat on 05/29/2010 06:13 amA 1MW SEP tug could move hundreds of tons of propellant from LEO to EML1/2/LLOThe turnaround time is too slow. A 1MW SEP cannot supply propellant requirements for even a first generation lunar base. For example, if we take the one published lunar architecture out there that's been designed by actual engineers (Zegler et al., AIAA 2009-6567), by year 3, a 1st generation lunar base could be expected to go through close to 300 tons of propellant per year. A 1MW SEP can only deliver on the order of a few 10's of tons of propellant--every 7 months--if the VASIMRTM studies put out by Ad Astra are accurate.Using the figures in Ad Astra's new paperhttp://www.adastrarocket.com/VASIMR_for_flexible_space_exploration.pdfA Survey of Missions using VASIMR® for Flexible Space ExplorationApril 2010JSC-65825A 500 kW VASIMR can deliver 14 mT of payload about every 6 months (or approx 28 mT per year) to low lunar orbit.300 mT / 28 = 10.7 tugsSo the Moon base can be supplied by launching 21 EELV year.Thank you, Mr. Swallow, for putting numbers to the absurdity of using SEP space tugs as tankers. The ACES-71 tankers can deliver 29 tons to the L2 depot, so the needs could be provided by 10-11 EELV's.
Quote from: Warren Platts on 06/10/2010 09:49 pmYou still haven't explained what's the point of using hypergolics in the first place.I thought that was clear.
The propellant would be used for propulsion on missions out-bound from the depot. Robotbeat provides a good explanation of the benefits of hypergolics. Primarily: hypergolic propulsion systems are known to work with high reliability even after long in-space storage.
No mission has soft-landed on the Moon using other than hypergolic propellant. No mission has ascended from the Moon using other than hypergolic propellant. No hydrolox system has provided propulsion more than twelve hours after it left the ground. Some people only want to attempt what has never been done before. Other people want to succeed.
Current EELV can only get a maximum of 30 mT to LEO, you are going to need fuel from LEO to low lunar orbit.
Meanwhile, the choice between SEP versus chemical propulsion is rather beside the point for the topic of this thread, which is whether the lunar landing architecture should rely on weak hypergolic engines or the beefy RL-10. Under no reasonable trade scenario would it be worthwhile to build a lunar architecture using hypergolics rather than LH2/LO2.
1. Hydrogen boil-off exists. With a 1% per day boil-off rate attempting liftoff on day 101 is likely to be difficult.2. Boil-off effects the Earth launch tankers, LEO propellant depot, cis-lunar tanker, EML depot, manned transfer vehicle, lunar lander and ascent stage waiting for 6 months.
1. According to my spreadsheet, if you start off at day 1 with 100%, with 1% loss per day, by day 101, you still have 36.6% left. Also, where did you get the 1% loss per day figure?
In sum, boiloff is manageable, and is indeed useful.
It is certainly no showstopper and is not a good enough reason to forgo LH2/LO2 architecture in favor hypergolics.
1. Warren, I suggest you lose the abrasive tone, it's not helping. 2. We're on the same side here. We have all seen the ULA architecture and we all like it. We all understand the importance of propellant transfer and the lure of LOX/LH2. 3.The point of using hypergolics is to get the essence of the ULA proposal (which is in line with lots of earlier proposals by the way), but to do it 3.1 faster, 3.2 more cheaply and 3.3 with less risk as I explained to you in another thread.
"The guiding philosophy that we have followed in developing the proposed sustainable explorationarchitecture is to use the least number of distinct elements. This meant not only all-up vehicles but the leastnumber of main engines, avionics systems, fluids systems, ECLSS systems, etc. By keeping the manyelements as common as possible development is foreshortened and costs suppressed. Recognizing that eachvehicle has unique functions it had to perform in addition to functions it shared with all other elementsflexibility and modularity had to be built in." (Zegler et al. "A Commercially Based Lunar Architecture, AIAA 2009-6567)
Despite the best engineering design and analysis activities it is amply clear that even highly vetted designs such as the Space Shuttle can fail catastrophically. Probabilistic analyses are spectacularly flawed in that they make sweeping assumptions about failure modes and the means to prevent them. Nature relentlessly renders these complex analyses moot when we find another hidden failure mode via flight experience. Ground testing can assure a baseline level of confidence but only extensive flight experience can truly generate a safe vehicle with high confidence in its overall reliability.
4. Sometimes less is more. 5. If we start using existing systems and components (EELVs, AJ-10, RL-10, Centaur, hypergolic propellant transfer) now, then more of us here will live to see manned exploration, cheap lift and commercial development of space.6. The whole idea about hypergolics is where to start, not where you would want to end up eventually. 7. The history of NASA exploration has been one of overreach and cancellation (Apollo, STS, SEI, SLI, Constellation). 8. Going back to the incrementalism of the early rocket pioneers, current New Space pioneers and ironically the incrementalism of Craig Steidle is what we need to succeed.
1. Sorry, but when a guy like sdsds insinuates that a guy like Frank Zegler doesn't want the American HSF program to succeed, I find that rather grating.
What bugs me is this constant willingness to question authority. When it comes to politics, I am a total anarchist. But when it comes to science and engineering, I tend to defer to the experts.
If I disagree with the experts, then I back up my counterargument with citations to other experts, along with detailed calculations of my own. I see neither citations to other work here nor calculations to back up what is being asserted here in this thread and elsewhere with respect to purported shortcomings in the ULA proposal.
3. I'm not sure you've quite grokked the essence of the ULA proposal.
So from the very get go, by proposing to mix in hypergolics, you are not getting at the essence of the ULA plan, but in fact you are running away from it.
Speed is good, but not when it comes at the expense of sustainability, affordability, and ROI for the American taxpayer. Moreover, you guys are offering nothing but assertions to back up your idea that your alternative idea is:
Why send up a hypergolic depot anyway? There is no spacecraft that exists for the hypergolic propellant to power. So to be faster, you've got to design a hypergolic spacecraft to use it. There is no reason to think a hypergolic spacecraft can be developed any faster than a spacecraft powered by RL-10's.
3.2. More cheaply: this is a red herring for sure. By unnecessary duplication of effort and increased complication is a sure-fire recipe for out of control cost spirals. In other words, you'll have to fund two development programs, instead of one, and you will have to fund to two separate support contracts. Now, I may be wrong in my reasoning, but if so, you need to prove that with actual numbers.
The RL-10 motor is among the most flight tested motors in the world. So as for risk, there is no way a hypergolic spacecraft is going to deliver less risk than an RL-10. Moreover, a single RL-10 is powerful enough to land a DTAL crewed lander. Since the design calls for 4 such engines in a lander, that results in quadrupally redundant engine-out capability. I don't think a hypergolic lander can deliver such engine-out capability. I may be wrong, but I want to see numbers before I change my mind.
4. No kidding. Unfortunately, you're proposing more (hypergolics and hydrolox) and calling it less. That's why I call your reasoning backwards. Not to be abrasive, but because literally, the arrows of implication in your arguments point in the opposite of the correct direction.
5. I see this argument all the time: If only NASA will do __________________, then at least I will be able to die happy; therefore, NASA should do _____________________. This is not a valid argument because it ignores the fundamental premise that the American taxpayer does not care whether you or I die unhappy. Furthermore, it leads to less than optimal decisions because it tends to sacrifice long-term afforadability and sustainability--and is thus unethical because it sacrifices the long-term benefits of future generations for short-term selfish gratification. (And such jejune versions of happiness that depend on space-as-edutainment are bound to lead to more ennui than happiness anyway, and are better off being reliquished.)
6. More backwards logic again. I don't mean to be abrasive, I'm merely pointing out the truth that the arrow of implication is drawn in the wrong direction. The ends should dictate the means. You put the means first, and it's bound to screw up where you're going to end up.
7. Overreach is exactly what a sensible and sustainable mission to establish a permanent presense on the Moon seeks to avoid.
8. Incrementalism is the very heart of the ULA philosophy. That's just a simple fact of life. It's what they do. It's their bread and butter. What is not incrementalism is the addition of an unnecessary excrescence to a streamlined and sensible proposal.
Quote from: Warren Platts on 06/11/2010 02:30 am1. According to my spreadsheet, if you start off at day 1 with 100%, with 1% loss per day, by day 101, you still have 36.6% left. Also, where did you get the 1% loss per day figure?1% is of the original mass, it's not an exponential decay. It is determined by solar heat flux, insulation, heat of vaporisation, density and the geometry of your depot.
Quote from: Warren PlattsIn sum, boiloff is manageable, and is indeed useful.You have not demonstrated that since you started from a false assumption.
The ULA architecture makes intelligent use of the boil-off in LEO and intelligent use of the superior thermal environment at L2. They believe a month in LEO with acceptable losses is feasible in the near term, but they have yet to demonstrate it, so there is still uncertainty associated with it. And even so, that still puts schedule pressure on the LEO to L1/L2 transport infrastructure.
Another interesting thing to look at is how things work for mild cryogens. ULA are the world's experts on cryogenic fluid handling in microgravity, so it makes sense for them to focus on liquid hydrogen. It is part of their competitive advantage and LH2 has a very high heat of vaporisation. This leads them to mixed fluid depots which use the boil-off of LH2 to cool the LOX and very low orbits for the depot which makes them slightly cheaper to reach for launch vehicles and allows the boil-off to be put to good use by compensating for the higher drag. But oxygen is much denser and thus easier to insulate and it catches less solar heat flux. Active cooling is also much more feasible for mild cryogens than for liquid hydrogen.
Quote from: Waren It is certainly no showstopper and is not a good enough reason to forgo LH2/LO2 architecture in favor hypergolics.No one claimed otherwise. But the benefits of LOX/LH2 are not worth delaying everything until it is operational. That's the crucial point you seem to be missing. We can always upgrade to LOX/LH2 later and we can even have development of that proceed in parallel, just not in series.
Look, you do not need to recite the virtues of the ULA architecture to us, we're all familiar with it. It's also not the last word on such architectures and the idea has been around for a long time and some of us have been studying it for a long time. Are you familiar with OASIS and its use of propellant transfer? It is even more advanced than the ULA architecture. You may want to look into it if you haven't done so already.ULA's plan is not the only way and not necessarily the best way. It's important to get this right, so we mustn't succumb to tunnel vision. Are you even aware of all the alternatives? Since we have similar goals it might make sense to try to collaborate instead of assuming you have all the answers and need to lecture us.