You are not logged in.
There are active means and there are passive means, to keep sunlight from heating the tanks and boiling off the propellant inside too fast. The active means require some sort of power producer driving some sort of refrigeration equipment. I'm no expert on how that is done, but there is always a significant weight penalty, especially if the power producer is a heat engine (thermodynamics requires this).
Passive means involve interrupting the path of heat flow, which is always from hot to cold (again, thermodynamics). Shade the tank so the sunlight cannot reach it. Radiant energy is the thing to interrupt, in the vacuum of space with no other physical structures, conduction and convection are not an issue, but the radiant energy from a star surface at many thousands of degrees is. What you cannot forget is that your shade surface will get warm, and radiate in the infrared. You have to interrupt that too.
Many of the things you might do have both beneficial and adverse effects. You have to achieve a net balance in your favor. You interrupt the radiant heat from your warmed shade surface with insulation, on its shady side. The surface of that insulation still radiates, but it is cooler and radiates a whole lot less than the bare shade would. That is because radiation varies as the fourth power of absolute temperature. That is one huge effect you can manage!
You can make the sunlit side of your shade surface very reflective, so that most of the sunlight is reflected back to space. But not all of it, nothing is perfectly reflective. That will cool off the temperature of the sunlit side of the shade surface. But it will cool off a lot less effectively when it is shaded, because high reflectivity is associated with low emissivity, and that strongly impedes re-radiation cooling to space. Once warmed, it tends to stay warm longer even when in darkness.
You can mount a shade panel on some sort of extended arms, yes, but you still have to insulate its backside to interrupt the infrared from the warmed shade surface. There is a weight penalty associated with those mounting arms, as well as the shade panel and insulation itself. If instead you wrap the tank in a layer of low thermal conductivity insulation, and over-wrap that with a highly reflective metal foil, you get the same insulated shading effect at pretty much the same (or possibly less) shade panel and insulation weight, but without the weight penalty of the extended arms.
You do pay for this with a thermal conduction path through the insulation to the tank. However, the temperature difference through the insulation multiplies a very low conductivity and gets divided by a significant thickness, for a low heat flow value. Thickness divided by thermal conductivity (in mixed US customary units) literally is the R-value of the insulation. And if the shade is the highly reflective foil, that shade surface temperature is not very high. The colder your propellant, the higher the R-value of your insulation that is required. Simple as that!
If your fuel and oxidizer temperatures are greatly different, do not use a common bulkhead between the two tanks, even if insulated in some way. We already understand that to be a dominant heat leak path in the Centaur stage design, which limits the "stage life" to only hours to a day or so. Just accept the weight penalty of two bulkheads to completely interrupt that heat flow path.
The SpaceX idea of the header tank is the way to further interrupt heat flow from the heated shade material to the tank wall into the propellant, when you do it by wrapping the tank. You must correctly size the header, and it must be located nested inside the main tank, for this to work as a thermal control. Short-term, you burn all the propellant out of the outer main tank at the start of your mission. What you need for the rest of the longer-term mission is all in the header. Vent the outer tank to vacuum. That cuts off the conductive/convective heat flow through the propellant vapor, from the outer tank wall to the inner header tank wall. In effect, the outer tank wall and vacuum around the inner tank wall form a Dewar. The only possible heat flow path is infrared radiation from the outer tank wall, but that temperature will not be very high, so there is very little of that kind of heat to flow. We already know this Dewar technique works very, very well.
These suggestions, carried out properly, might get you a "stage life" measured in weeks or even a month or so, even with hydrogen! For years of "stage life", you might have to add the refrigeration equipment.
"Stage life" is the interval until there has been enough boiloff evaporation inside the propellant tank to raise its pressure dangerously, whereupon you either vent the propellant, or the tank bursts. This is a serious issue, and the source of a significant fraction of the space debris around the Earth.
Sorry, guys, there's no simple (one-issue) answer here. Thermal control is ALWAYS a very complicated tradeoff among many issues. You just have to do the necessary engineering, and you have to get it right. The answer will NEVER come from just plugging numbers into one single equation.
GW
I agree with the phased approach described in the video, but I disagree with the number of phases and what is done in each of them.
The unmanned phase started with Mariner 4 back in the 1965 flyby netting close-up photos, and still continues today 6 decades later! It is still incomplete, and has yet to put the proper prospecting rovers with real drill rigs onto the planet. The resources sought are most likely 10+ m under the surface. Scratching the surface will NOT evaluate them! Remote sensing is not the answer, because there are still disparities between it and the real ground truth that you need, to justify betting lives.
There are 3 manned phases, not two. The first is exploration, overlapping with the unmanned explorations. It would be best to visit multiple sites, prospect around, and use real ground truth to select the site (or sites) for the next phase, which is "experimental bases", NOT the start of any settlement!
It would be much more efficient to visit the multiple exploration sites in one mission from Earth, which inherently requires basing from low Mars orbit and using 1-stage reusable landers refilled on-orbit from supplies sent ahead previously. Redundant landers provides "the way out" (a rescue capability) that no other approach can supply. This from-orbit approach rules out the use of SpaceX's "Starship", which cannot carry large payloads into low Mars orbit, and would be stranded there if it attempted that mission.
The experimental base phase is where you experiment with your "live-off-the-land" technologies, while being supplied from Earth, until you get these technology things working "right". The base (or bases) are permanently manned, but with rotating crews who each go home after a bit of time. It is UNLIKELY IN THE EXTREME that such technologies, developed on Earth before this experimental base phase begins, will actually work "right" in the field! It is UNETHICAL IN THE EXTREME to send settlers there until all these technologies actually do work "right"! This phase is where you build the landing pads so that Starship landings can be reliably made, for direct cargo shipments to the surface. You need large, level, hard-surfaced pads to support a tall, narrow-stance ship, that would sink into soft sand and topple over. Rockets that topple over have always exploded violently here. Expecting different there is the height of ignorant idiocy!
The settlement phase should NOT begin until there is complete success with the experimental base phase! Only then can you reduce supply shipments from Earth, as the "live-off-the-land" technologies can shoulder the load and be expanded in capacity. Only in this settlement phase is it appropriate to send large numbers of people to go there permanently, instead of massive loads of supplies.
I really disagree with the video saying SpaceX's Starship is "designed for Mars". It is NOT! It is first and foremost a surface-to-LEO transport capability, of large payload masses and dimensions. That is its best use, especially in the exploration phase, and into the experimental base phase, until real landing pads can be built. It can only leave LEO if it is fully refilled on-orbit by tanker flights, and at a 1200 ton propellant load, it would take 6 tanker flights capable of delivering 200 tons each, to refill it. That would be 8 tankers if only 150 tons deliverable proves to be the case. More if they expand the propellant load to larger-than-1200-tons!
Numbers do not lie, but they can be misused and abused. Ethics doesn't lie to you either, but it is too often ignored.
GW
Merry Christmas and a Happy New Year to all. Hannukah, too. Whatever you celebrate!
GW
That "subway map" has reasonably good numbers, based on assumptions not explicitly given. Changing those assumptions changes the numbers, but not greatly. You're in the ballpark for getting started. You are much better off using more precise dV estimates, once you really start to size vehicles, though. And don't forget to look at vehicle thrust/weight at every burn, not just rocket equation dV, as that is how you make sure your burns are "impulsive" enough to use orbital dV's without gravity loss factoring. The thrust sets engine masses, which in turn affect your inert mass estimates. Those in turn critically affect your achievable mass ratios. Like everything else, even the rocket equation is subject to "GIGO" (Garbage In, Garbage Out).
GW
There is one other thing to think about, and that is the pressure at which your process or system does its thing, whatever that might be. Energies and power levels set the speed of the ejected stream. But the pressure at which the energy is transformed or transferred sets the density in the process, in concert with the temperatures involved. If the process density is low, the thrust is going to be low from a device of any given size, period! THAT is why the electric propulsion devices so far have given whispers of thrust (milli-N to at most N) from hardware weighing 1 to many tons, not including the KW-to-MW power supplies they need. The pressure inside the chamber where the ions are accelerated is almost indistinguishable from vacuum. Why would one ever expect a different outcome?
The solar thermal rocket must take that into account. You must do the heating of the hydrogen at significant pressure, since you are getting your exhaust velocity from a compressible flow expansion process. The NERVA XE reportedly did this at 560 psia "chamber pressure", although there would be a strong pressure drop through the reactor core. I'm only guessing the "chamber pressure" to be the aft pressure that feeds into the nozzle. Whatever, the higher that nozzle feed pressure, the better. That notion applies to ALL rockets that use nozzles: chemical, nuclear, or solar!
You only get a lot of force out of a reasonably-sized exit area if the exit plane density is high, so that there is a very appreciable massflow through that exit area (at the very high velocity determined by the energy transformations reflected in chamber temperature). That exit plane density is proportional to the chamber density, in a ratio fixed by the expansion ratio of the nozzle (as are the exit plane pressure and temperature and Mach number). Simple as that.
GW
If you can steer the spherical balloon collector relative to the core axis, the 90 degree thing is no longer a requirement. There are still angles that won't work, but there will be a majority that will work. But you get what you pay for: you must hold onto the balloon, and steer it to face where you need, and STILL get the rocket nozzle through the side of the balloon. That is NOT AT ALL easy! It will be heavy. Inherently!
GW
Bob:
Short answer about refueling flights: yes, but indirectly, from a depot that must be kept filled.
Starship carrying payload has only enough propellant left aboard after reaching LEO to effect a deorbit and a landing. Unrefueled, it can go NOWHERE else. The same would be true of any other orbital transport vehicle, whether partly or fully reusable.
Any sort of craft headed for low Mars orbit must depart from LEO, correct its course, and arrive in LMO. Without a tug, LEO departure requires at least 3.7 km/s (for Hohmann transfer), probably more (if you want to go faster), with course correction around 0.2 km/s and LMO arrival somewhere near 2 km/s. That's about 5.9-ish or more km/s just to get to LMO one-way.
A tug assist to leave LEO reduces that requirement on the interplanetary vehicle to something like 0.6 to perhaps 1 km/s (faster than Hohmann), plus the same correction and LMO arrival. That puts the 1-way dV nearer 2.8 to 3.2 km/s, a substantial reduction. That opens the door to a possible return from LMO to LEO, with things we already know how to build.
That's an important improvement! You do it with a tug vehicle that remains in the ellipse after releasing the interplanetary vehicle, at the ellipse perigee where speed relative to the Earth is highest. The tug can then return to LEO and get refueled there, which is easiest to do in low eastward circular. That's why an orbital assembly facility and refueling depot makes sense. Accumulate the propellants over time between interplanetary missions, then use it when you need it to mount one.
Starship is not primarily an interplanetary vehicle, despite what Musk claims. It is a large transport to LEO and back that is intended to be fully reusable. You can misuse it by refueling it to go elsewhere, but you WILL refuel it to go anywhere else! Nobody else has a vehicle yet that meets that reusable orbit transport definition, or to go anywhere else, but if they did, the VERY SAME refueling restrictions would apply. Everybody has to obey mother nature's rules.
The refueling requirements thing in LEO if you want to go anywhere else is EXACTLY why an assembly facility and propellant depot in LEO makes so much good sense! NASA failed to build one (the ISS is NOT such a facility!!!), because government bureaucracies micromanaged by congressional politicians are actually rather brain-dead, as we all know quite too well.
GW
The Newton is the unit of force defined in terms of the kg of mass, meter of distance, and second of time. One standard kg exposed to a gravitational acceleration of only 1 m/s^2 would weigh 1 Newton. Here on Earth, 1 standard kg of mass exposed to a gravitational acceleration of 1 standard Earth gee (some 9.80667 m/s^2) weighs 9.80667 Newtons. Check the definitions in SI if you don't believe me.
There's non-SI metric units in wide use, and the kg-force and the metric ton-force are a couple of them. The kg-force is the Earth weight of 1 std kg of mass exposed to one standard gee of gravitational acceleration. 1 kg-force = 9.80667 Newtons. Likewise, the metric ton-force is the Earth weight of one standard metric ton (1000 std kg) of mass exposed to 1 standard gee of gravitational acceleration. 1 metric ton-force = 9.80667 Kilo-Newtons. A lot of torque wrenches are calibrated to read in m*kg-f units, not the SI m*N units.
The existence of the kg-force (kg-f) unit is exactly why specific impulse is measured in seconds. It is kg-f of thrust divided by kg/s massflow (not weight flow !!!!). People "conveniently" divide-out kg-f with kg, although that is not right! Similarly, Isp = metric tons-force divided by metric tons of mass per second, dividing-out the metric ton-force with the metric ton, although that is not right, either. This was defined long before SI, by many decades.
The analog to this in US customary units is Isp = lb of thrust divided by lbm/sec of propellant flow, yielding sec if you divide-out lb with lbm, which isn't right. That is where the lbm is defined in terms of the lb force unit, feet, and seconds, improperly as the mass that weighs 1 lb at 1 standard Earth gee of gravity, which is really 32.174 ft/sec^2. The lb is the Earth weight of 1 lbm of mass.
Just goes to show you that the evils men do live long after them.
GW
A question or two: what is the tonnage that a tug needs to push from LEO into LLO (low lunar orbit), and still return to LEO unladen? Is it 900-some, or is it 600-some? Metric tons units of measure is what I am looking for.
Does all this need to go in one big chunk, or can it go in multiple smaller chunks? If it can go in multiple smaller chunks, what tonnages?
GW
NASA's mistake was not using bonded tiles of Avcoat in preference to hand-gunning the Avcoat into fiberglass hex glued to the capsule airframe! Bonded tiles work quite well. We've already seen that with Space Shuttle and X-37B. And now SpaceX's Starship. Doesn't matter whether the tiles are refractive or ablative; if your bonding scheme and your gap filler scheme work, you can pretty much do bonded tiles of any type desired.
NASA mistake 1: deleting the reinforcing hex from the Avcoat tiles they made. The fiberglass hex glass fibers tied the char to the virgin beneath more strongly, against any forces attempting to remove that char. Deleting it makes it much easier to lose chunks of char, simple as that! They were thinking inside the box of only hand-gun or not. They failed to consider that extrusion presses could force the Avcoat into hex cores without any hand-gunning! Pushing goo where you want it to go with an extrusion press, is a very old technology in the plastics industry. It is NOT new technology! It applies to ANY goo, even Avcoat. But you need to talk to some plastics guys to get yourself set up to do it. Use the kinds of presses they use.
NASA mistake 2: they wanted to use ultrasound to do QC inspections. But it was unreliable at the lower Avcoat density used in the first Orion flight test EFT-1. So they densified the virgin Avcoat for Artemis-1, and ultrasound was reliable for inspection. But the denser Avcoat was less permeable to gas percolation through the char, lending credence to the theory that gas pressure in the material blew off chunks of char on the second Orion test Artemis-1. Denser Avcoat did not fly on EFT-1, nor did it fly on Apollo! Both of those used Avcoat at lower density (I presume a higher microballoon content) hand-gunned into hex cores bonded to the capsule airframe. They did not think outside the box of only considering ultrasound! X-ray would likely have worked at any Avcoat density, although it is not as convenient as a hand-held ultrasound device. X-ray worked just fine inspecting solid propellant inside steel cases, in the tactical-size solid rocket industry, decades ago. It was a part of how we at the old McGregor plant routinely achieved 1-in-a-million failure rates with solid motors!
NASA mistake 3: for Artemis 1, they added a skip-type reentry, where there was a cooldown interval between two intense heating pulses. This increased range covered during entry by around a factor of 2, something they wanted to do. Such was NOT flown on EFT-1 Orion, or at any time during Apollo! Artemis-1 was the first such skip entry ever flown with capsules. Maybe they had no experience to guide them, but we in the solid rocket and ramjet industries had encountered the char embrittlement effects of cooldown decades ago! Embrittlement makes the char material more fragile, and therefore more susceptible to damage or removal by whatever forces get applied to it, during any re-heating. The thinking box here was only considering NASA experience, not other related industries that also use ablatives.
NASA has pretty much decided to fly the Artemis-2 mission with the pre-existing Artemis-1-type heat shield "as-is". But they have deleted the skip from the reentry trajectory, which is a step in the right direction. Their subscale ground tests say this will work, after duplicating the types of failures seen on the Artemis-1 entry. We will see what the real truth is, when it actually flies. I do think the risk of a fatal burn-through is rather low, but I am also sure that it is not as close to zero as it should be, and could be, if the heat shield were replaced. Yet, in experimental flight testing (and Artemis-2 is an experimental flight test, make NO mistake about that), risks like this are accepted every day. I just hope the odds pan out right, for the crew's sake.
GW
From AIAA’s “Daily Launch” for Monday 12-9-2024:
Here is the short item in the newsletter, which is a ink to a longer article posted on Ars Technica:
ARS TECHNICA
After critics decry Orion heat shield decision, NASA reviewer says agency is correct
Within hours of NASA announcing its decision to fly the Artemis II mission aboard an Orion spacecraft with an unmodified heat shield, critics assailed the space agency, saying it had made the wrong decision.
This is a short excerpt of 1 paragraph from the longer Ars Technica article:
"Based on the data, we have decided—NASA unanimously and our decision-makers—to move forward with the current Artemis II Orion capsule and heat shield, with a modified entry trajectory," Bill Nelson, NASA's administrator, said Thursday. The heat shield investigation and other issues with the Orion spacecraft will now delay the Artemis II launch until April 2026, a slip of seven months from the previous launch date in September 2025.
My take on it:
This is the gist of what has been going on: the investigation includes work by the regulars at NASA and an independent review team (IRT). NASA did a bunch of work, really good work, supporting flying “as-is”. The majority of the IRT agreed, although there were two who disagreed, one possibly a leader of some sort on that IRT.
I tend to think they are probably right to do this with the trajectory mod deleting the entry skip. But I also think they will see some cratering char loss anyway. How much, who knows? I think the odds of a fatal burn through are low, but not as close to zero as they should be.
The real problem is what changes will go into the heat shield for Artemis-3 landing mission? I’ve made my inputs to that, and I hope it helps them do this “right”.
-- GW
I could not make the 4 fold-out shipping containers around a single core work. I ended up with 2 shipping containers side-by-side with a shipping container-sized space in between them. I had room to put 3 rocket cores in that space, and did not get hoverable thrust, until I used 3 smaller engines on each core. The shipping containers are the landing pads. It's got the low cg height at touchdown, and would be capable for very rough-field landings on very soft sand dunes.
There has to be a foldable pair of heat shield doors between the two containers, that folds out of the way to let the engines fire, without flipping the vehicle (there isn't time for that, not at a ballistic coefficient near 1000 kg/sq.m, nor is there any time to use chutes). If the transfer stage does the departure and course corrections, then the cargo lander need only make the deceleration burn after the direct aerobraking entry, and finally to touchdown. I used storable propellants, and got a 115 ton item containing some 39 tons of deliverable cargo inside the two shipping containers.
The shipping containers open on one end, right at ground level. Easy to unload either manually, or with the analog to a forklift.
It's 1-way only. Not at all reusable. Although the transfer stage could be.
GW
Avcoat is an epoxy novolac resin filled with microballoons to reduce its density and "tailor" its ablation rate. I keep thinking silicone resin, but that was Gemini, and also the best ablative that I used in ramjet combustors. Sorry.
Avcoat is supposed to be gunned into the hex cells of a fiberglass hex core. The glass melting produces silica, and the epoxy pyrolizing produces carbon. The char this stuff produces atop the virgin is supposed to be composed of both silica and carbon, and it is inherently porous. That's the way it was during Apollo, and on the first flight test of the Orion capsule atop a booster other than SLS. Those performances were quite adequate for simple entries into Earth's atmosphere at its escape speed or thereabouts.
Plus, the glass fibers soften at a slightly higher temperature than the epoxy resin pyrolizes (real silica fiber even higher). What that means is that the glass fibers penetrate into the char a little ways, as the "pyrolysis front" moves into the material during heating. That effect helps to "tie" the weak porous char layer more strongly to the virgin beneath. Which in turn makes the char harder to rip off of the virgin beneath, by whatever forces get applied (for any reason).
If you leave out the fiberglass hex, the char has only carbon, and is less dense, and less strong. Anybody who was ever handled real charcoal that wasn’t made in a press, knows how crumbly this stuff really is. It is NOT strong at all! Plus, it usually embrittles upon cooldown, becoming even more vulnerable to applied forces. And without the hex, the char layer is no longer as strongly "tied" to the virgin beneath. THAT is exactly what they did when they went to Avcoat tiles without the fiberglass hex, on Orion for the Artemis 1 and 2 capsules!
Add to that the fact that Apollo and the first Orion test flight did not attempt skip-type entries, with two heating pulses separated by a cooldown coast. Artemis 1 did! It had a weaker char layer, tied more weakly to the virgin beneath, that further had embrittled and become even more fragile, during the cooldown coast between the two entry heating pulses. This exposure was outside all previous experience ever obtained with those heat shield materials and construction methods!
So, why should ANYBODY have been surprised if it didn't act as expected? And yet they were surprised, and then took 2 years investigating what actually happened. They have decided to fly the same misbehaving heat shield on Artemis 2, primarily because it is already built and installed, not for any other valid reason! They are going to delete the skip with its cooldown coast, which is a step in the right direction.
They think it was gas pressure buildup within the material as the thermal wave works its way inward and causes more virgin to pyrolize (during the skip cooldown), that actually pushed chunks of char off the heat shield, leaving those odd craters behind. They think that was the mechanism instead of my suggestion that fluid shear forces ripped the chunks out by a severe scrubbing action. But the actual cause makes no difference, the real "fix" is EXACTLY the same: put the hex back in the material, so that the glass fibers can help tie the char to the virgin more strongly! Simple as that!
We members of the public have only recently been allowed to hear more details about how they built this thing. I would have built molds for each tile shape, and just cast them almost right to finished dimensions. That's not what they did! They made big blocks of cast balloon-filled epoxy novolac resin, and literally machined every tile shape out of those blocks!
Doesn't matter, just put the hex back into the resin that you make the big blocks from! Then machine what you want from those blocks! The easiest way to get the resin into the hex cells without manual hand-gunning, and without (fatal) voids, is to use a plastics extrusion press, and just push it right through a big hex core, to make your big blocks! What could be simpler?
Nothing about my "fix" is different, except for exactly where in their more complicated process you do it. Pushing goo with an extrusion press to where you want it to go, is a very common processing procedure in the plastics industry. There is no new technology there!
Just do it!
GW
PS -- after posting this, I noticed the logout button is unresponsive. I've often seen that bbefore, in the older system. That fault is still there in this newer one.
Could be a simple bar truss. The Russians did a lot of those. Bars shipped up as bundles in payload bays, then snapped together in space by means of appropriate snap joints.
But you still need a way to refuel the Interim Upper Stage (IUS) on-orbit, before it departs to send the cluster to the moon. Why? Bigger cluster. It needs more dV than it has left after reaching LEO, in order to put an Orion/service module (and maybe a small lander) into low lunar polar orbit. The service module can get Orion back onto the trip home, from that low lunar orbit, but it cannot get both into and back out of low lunar orbit. It is too small.
GW
The links work fine.
But if you read the article, you see that I did not do this study for using a tug with Artemis-2 in any way. NASA is going to fly that one with the bad heat shield, that much is clear.
Actually, since the Orion capsule is not reusable, I don't see much usefulness for trying to slow it into orbit coming back from the moon or anywhere else. It'll have a "fixed" heat shield from Artemis-3 onward, and the odds favor success flying with the bad one on Artemis-2, although the probability of a fatal burn-through is not as close to zero as it ought to be.
I did the study to find out what might be possible in the near term. For the orbital data I used, getting onto an interplanetary trajectory at 11.5 km/s requires 3.7 km/s dV from LEO at 7.8 km/s, unassisted. The tug could get it to 10.9 km/s and subsequently be recover itself from an ellipse apogeeing near the moon's orbit. The interplanetary craft dV from that tug assist point is only 0.6 km/s! Tug-assisted departure makes a big difference!
I think Starship might actually have its flap burn-through problem solved, and its heat shield pretty much determined, and also have demonstrated propellant transfer from 1 vehicle to another, in about a year, which is very near-term. That means a tug modification could be flying not that long afterward. And the mass capacity it could fling onto hyperbolic departure is astonishing: just short of 500 metric tons or thereabouts. Arrivals, not so much: nearer only 175 tons.
The real long pole in the tent is not modifying Starship or some other upper stage to be used as a space tug, it is having a good facility in LEO to assemble interplanetary payloads and then dock tugs to them, using remote-operated mechanical arms, plus a ready means and depot from which to refuel the tugs and fuel the interplanetary craft.
That won't happen on a 2-year timescale! If Gateway gets built at the moon, the assembly and propellant depot facility in LEO will never get built. There is not enough $ to do both.
But with a tug assisting departures and maybe arrivals, why would we need Gateway? THAT is the really telling question no one is asking.
GW
The IUS is supposed to be enough to send Orion to Gateway in that halo orbit and still come home, which is supposedly enough for the lunar flyby without going to the halo orbit. But I cannot run the numbers for myself without data to populate weight statements.
Why not just keep a Starship in orbit and just use it as a space tug? You have to refill it fully, but so what? 120 tons inert, 1200 tons max propellant load, 3 sea level engines at about 230 m.ton-f thrust each in space, and 3 vacuum engines at about 250 m.ton-f thrust each. That's a max thrust capability of some 1440 m.ton-f with all 6 burning. You won't need anywhere near that much thrust.
The real trick is figuring out about what dead-head payload the Starship could push from LEO to an elliptic departure, and still get back unladen to LEO. And, what dead-head payload a Starship could go unladen to elliptic, and still retrieve back to LEO.
My crude numbers say 497 metric tons payload at departure, but only 175 metric tons payload at arrival. Both using full propellant loads as filled in LEO. I use 2 vac Raptors at full thrust while laden, but only 1 vac Raptor while unladen.
If you leave the heatshield and flaps on it, you can bring it home for repairs anytime anything wears out, like an engine.
GW
"Interim upper stage" (IUS) is the smaller second stage of the SLS as embodied in the Block-1 configuration that cannot reprise even Apollo-8 with Orion and its service module. This is supposed to be replaced with a substantially larger second stage in the Block-1B configuration, which will be able to do significantly more. Block-2 replaces the SRB's with liquid strap-on boosters, adding yet a bit more capability. I have my doubts as to whether there will ever be a Block-1B, much less a Block-2. Not with per-launch costs already climbing past $4B for a single SLS/Orion flight. I've heard the SLS price without Orion is nearer $2+B, but I rather doubt that to be true. So much has turned out to be false already.
GW
First: all 4 links work.
Second: Artemis-2 flight with Artemis-1-type heat shield -- Reading between the lines somewhat, from akl the publications I have seen, NASA managers boiled their decision down to 3 options: (1) do nothing, just fly as-is, (2) do nothing to the heat shield, but modify the entry portion of the trajectory to reduce the heating a little, or (3) fix-or-replace the heat shield. Choice 3 meant another year's delay and a lot of extra costs, so they went with either 1 or 2, I think probably 2. That prioritizes money and schedule above astronaut's lives, but they have a demonstrated history of doing that, apparently never learning the lesson that there is nothing as expensive as a dead crew, especially one dead from a bad management decision. Whatever "fix" there might be for the Orion heat shield will go on Artemis-3 (the manned landing) which hasn't yet been built. The Artemis-2 heat shield was built back in 2020, according to some dated photos I have seen.
Third: I seriously doubt NASA managers would be open to finding a means to slow the Artemis-2 Orion down all the way into LEO to ease the entry heating. Whether that was among the options they might have considered is unknown, but my hunch says unlikely. Such a thing might be done with a crew Dragon sent along as a wingman (docking to Orion on the return), but would require a lot more dV out of the Super Draco's than could possibly be loaded, even into the "super trunk" variant they are looking at for ISS deorbiting. There's nothing that could fly right now that could do the job of 3.1 km/s dV with a whole Orion added to the crew Dragon mass, to get from approach speed down to circular orbit speed.
GW
Well, they do have to make it all work, and then they must show it all to be reliable, before people ride it. We do that same process with new airplane designs, because it is the right thing to do.
Myself, I take claims like that with a bucket of salt. Musk dollars to do things seem to be off from reality by a similar large factor as Musk time to get something done.
GW
Considering how rife the internet and social media are with misinformation and disinformation, I look upon most of these sources with a jaundiced eye at the very least. Those sources are totally unpoliced for any truth.
What I have found in the trade journals (inflated as it is with marketing hype, but partly written by journalists with at least some standards), she probably has not lost much weight, but it does redistribute in zero-gee, to visually-noticeable effects. This has been seen in most astronauts up there for any significant number of days, since the 1960's.
That's not to say that she may be suffering from something that "they" are trying to hide. But there's little trustable evidence either way.
Crew Dragon was originally designed to fly with 7 seats. Some NASA "safety" type insisted on only 4 seats being installed, because of the "difficulty" the bottom 3 might have trying to get out quickly. 20-20 hindsight says that was a stupidly-bad decision: if crew Dragon still had 7 seats, Butch and Sunni could have ridden home with the last 4, even if in shirtsleeves (also a risk, as was found by a lost Russian Soyuz crew).
Starliner is a design travesty, yes. But there are others: (1) only 4 seats in crew Dragon, (2) spacesuits incompatible among capsules, (3) an ISS leaking air for 5 years because of metal fatigue cracks that cannot be effectively repaired.
Shall I go on?
GW
Those trajectories are exactly what I previously understood about SLS/Orion Block 1, and amazingly close to what I analyzed for modeling propellant return via Gateway to LEO. My 60,000 km apoapsis halo is gravitationally stable, but required a slightly-higher dV burn to enter it at periapsis.
Apparently that slight dV increase is too much for SLS/Orion Block 1, so they went to the gravitationally-unstable 70,000 km apoapsis halo, which requires frequent course corrections by Gateway just to stay in that orbit. Without them, it leaves the moon, going into a far orbit about the Earth.
GW
Artemis-2 will NOT be orbiting the moon. It cannot, that would be a 1-way suicide trip. Artemis-2 is a flyby of the moon, continuing on into cis-lunar space before returning home at about escape speed.
SLS Block 1 can put the Orion/service module only into trans-lunar injection. The interim upper stage is essentially out of propellant after doing that. The service module is undersized, compared to the mass of the Orion capsule, quite unlike Apollo. There is enough dV to enter low lunar orbit, but not to leave it.
This thing was designed only to barely reach that crazy halo orbit about the moon, it cannot reach an Apollo-like low lunar orbit.
GW
From what I have seen to read, NASA program managers were choosing from 3 options: (1) do nothing, (2) modify the entry trajectory to reduce the severity of the heating a little but keep the questionable heat shield, and (3) replace the heat shield with something better. Option 3 means unstacking the rocket and another year's delay at least. The other 2 options do not.
I believe they chose either option 1 or option 2, most likely 2. They're close-mouthed about this, because they do not want the general public to know they deliberately risked astronaut lives to save time and money, the direct cause of both fatal Shuttle losses.
Yes, I got my original letter-to-Bill Nelson into the hands of a real NASA heat protection engineer David E. Glass, along with some 3 decade old ramjet insulation test photos showing a very similar chunk loss pattern (from tests that I ran back then). I expect they might try a "better design" for the Artemis 3-and-subsequent heat shields, which haven't been built and installed yet, and which is where my information might be used. I have not heard back from him, after acknowledging receipt of the items.
Any sort of space tug rescue option would be something for the indeterminate future. There is nothing capable currently flying that we could use for this. I have not been able to figure the rendezvous dV for such a scenario, but I do know it would be enormous, and it would have to take place on a timeline only some minutes away from entry.
GW
As best I understand it, they used a Dragon to fire its thrusters to see what effects that might have as to pushing the ISS. I did not hear anything went wrong. I am presuming this was a crew Dragon, using its Super Draco thrusters.
If that presumption is correct, the "big Dragon" design that is supposed to deorbit ISS will be nothing but a crew Dragon flown unmanned, and controlled from the ground. This capsule would be flown with a possibly-bigger trunk, and that trunk containing a substantially-increased load of Super Draco propellants, connected to the thruster propellant supply in the capsule, where the thrusters are located.
Makes sense to me.
GW
Supersonic retropropulsion is not the brand new area of investigation that the Angry Astronaut says it is. It's only new if current people at NASA and SpaceX do not understand or appreciate history. Which is apparently the case!
My 1965-vintage edition of Sighard F. Hoerner's opus "Fluid Dynamic Drag" has a plot of drag coefficient versus jet thrust, for a retrorocket plume emanating from the heat shield center of a Mercury capsule shape, tested at Mach 2 in a wind tunnel in 8.2-inch diameter size. The reference Hoerner cites for this plot is a NASA Tech note titled "Thin Retro Jet", TN-D-751, dated to 1961, written by Charczenko. He also gives two other closely-related refences: Wasko "From a Sphere" TN-D-1535, and Peterson "Four Retrorockets" TN-D-1300. These citations did not give a date. The 3 relevant citations (numbers 22a, b, and c) are listed on page 20-9 of this edition of Hoerner's book. The plot is Figure 27 in Chapter 20, located on page 20-11 in Hoerner's book.
It is very clear that NASA was doing serious engineering testing of retropropulsion effects in supersonic flow, way back in the early 1960's.
The modern "experts" seem to be unaware of it. Some "experts" they are!
GW