Crew Dragon Mission Life Extension: Free-Flying, Uncrewed Duration

[TheRadicalModerate]

...
That all gets back to temperature.  If some consumable's rate of consumption is temperature-dependent (e.g. water), then a low temperature can cause condensation and a variety of yucky stuff.

...

[2] Except maybe venting for some undefined purpose? Why would that be needed if existing thermal controls work?

This is why I keep wringing my hands about thermal.  I could see the thermal system working great as a free-flyer for 7-10 days, using a little bit of venting when insolation thermal load got a little too extreme, but with enough consumable reserve that the system has plenty of margin.  When the D2 is docked at the ISS, the ISS could provide the extra margin, either by circulating heat into the ISS pressurized volume, where its thermal system would clean it up, or simply by somebody adding water to the D2 as needed.

But if the thermal system isn't perfectly closed during long-duration free-flight, that margin would (ahem) evaporate.

NB:  I don't know anything about the D2 thermal system, and I could just be a nervous nelly.  (It wouldn't be the first time.)  But that's why I started the thread:  What kinds of things could go wrong, and how hard would they be to patch?

[cohberg]

I think you may want to read the ECLSS paper linked to earlier in the thread.

using a little bit of venting when insolation thermal load got a little too extreme

Dragon doesn't vent atmosphere into space for thermal control when the trunk is attached.
Heat is removed via a closed loop radiator system. Only electricity is consumed to cool the craft.
Removing the human metabolic loads will decrease the heating of the craft, lowering the duty cycle for the TCS.

human transferring the occasional bag of water back and forth.  Neither of those would be available during free flight.

simply by somebody adding water to the D2 as needed.

Dragon doesn't have a water tank or consume water.

Humidity is produced by humans and is only removed, not created, by Dragon.

[edzieba]

Not free flying, but a simple solution. Outfit the widget (depot tanker Starship) with a deck module that have a few radial docking ports above the propellant tanks. So the docked Crew Dragon capsule draws power and consumables from the depot waiting for the return of the OTV. Advantage of little modifications to the Dragon capsule.

There are several ways to sustain the Crew Dragon in LEO.
   1) dock it to ISS
   2) Fly another unmodified Starship HLS that will sit in LEO and act as a minimal CLD to sustain the Crew Dragon. No new engineering needed.
   3) Add a second IDSS port to all the Starship HLSs. I think this qualifies as a minimal change, and it adds operational flexibility. One of these HLSs acts as a slightly less minimal CLD and sustains the Crew Dragon.
   4) Go ahead and build a real Starship CLD with multple IDSS ports.

The advantage of a CLD with two or more ports is that the Dragon and the Starship OTV can both dock to it for the crew transfers. The Dragon stays docked until the crew returns to LEO for the homeward transfer. No need for three separate Dragon dockings.

5) Land it, refurbish it, then launch it again.
This seems like a lot of extra R&D and HSF certification and complicated CONOPS purely to avoid reusing the proven reusable spacecraft.

[DanClemmensen]

Not free flying, but a simple solution. Outfit the widget (depot tanker Starship) with a deck module that have a few radial docking ports above the propellant tanks. So the docked Crew Dragon capsule draws power and consumables from the depot waiting for the return of the OTV. Advantage of little modifications to the Dragon capsule.

There are several ways to sustain the Crew Dragon in LEO.
   1) dock it to ISS
   2) Fly another unmodified Starship HLS that will sit in LEO and act as a minimal CLD to sustain the Crew Dragon. No new engineering needed.
   3) Add a second IDSS port to all the Starship HLSs. I think this qualifies as a minimal change, and it adds operational flexibility. One of these HLSs acts as a slightly less minimal CLD and sustains the Crew Dragon.
   4) Go ahead and build a real Starship CLD with multple IDSS ports.

The advantage of a CLD with two or more ports is that the Dragon and the Starship OTV can both dock to it for the crew transfers. The Dragon stays docked until the crew returns to LEO for the homeward transfer. No need for three separate Dragon dockings.

5) Land it, refurbish it, then launch it again.
This seems like a lot of extra R&D and HSF certification and complicated CONOPS purely to avoid reusing the proven reusable spacecraft.

Each Crew Dragon flight appears to cost about $200 million, so that's the cost of the second flight. You need to find an approach that costs less than $200 million extra per Moon mission if you want to avoid that second flight. In the specific context of this thread, your life-extending trunk would need to cost less than $200 million, because the trunk is expended. This includes the amortized cost of development and crew-qualification.  I have no idea what the cost would be. For discussions about other approaches, please go to the other thread.

[Barley]

Two things come to mind that might limit endurance.

Station Keeping

Either the quantity of propellent or the number of cycles for the thrusters.

Mold and Mildew

And other biologicals.  These are dealt with by air movement and filtering, cleaning, or by making the cabin inhospitable to life.  Perhaps it's as simple as opening a large can of silica gel before leaving for the moon.

[TheRadicalModerate]

I think you may want to read the ECLSS paper linked to earlier in the thread.

using a little bit of venting when insolation thermal load got a little too extreme

Dragon doesn't vent atmosphere into space for thermal control when the trunk is attached.
Heat is removed via a closed loop radiator system. Only electricity is consumed to cool the craft.
Removing the human metabolic loads will decrease the heating of the craft, lowering the duty cycle for the TCS.

human transferring the occasional bag of water back and forth.  Neither of those would be available during free flight.

simply by somebody adding water to the D2 as needed.

Dragon doesn't have a water tank or consume water.

Humidity is produced by humans and is only removed, not created, by Dragon.

I went through the paper at your suggestion.  Some thoughts:

1) I was surprised to see that D2 only supports 20 person-days of consumables.  That's 5 days of crewed use, not the 7-10 that I'd heard (or thought I'd heard) elsewhere.  That should still be fine for D2/LSS kludgery, though.

2) The paper never mentions what the thermal control system coolant is.  Does anybody know?  I assume it's a single-phase coolant?  If not, there could be some issues with free-flight thermal load that could trigger some kind of gas venting, which would need to be replaced from a consumable reservoir.

3) Care and feeding of the nitrox COPVs (2 sets of 3 COPVs each).  What you said about D2 not venting isn't completely correct.  The ECLSS will vent atmosphere if pressure limits are exceeded.  It's unclear to me whether that's an exceptional event in free flight or not.  Presumably, if venting does occur, then atmosphere is back-filled from the nitrox system as things cool down.  Note that adiabatically cooled nitrox is used to keep the cabin cool (well, not intolerably hot) during reentry, so maintaining an adequate nitrox reserve is essential.

I assume that venting never occurs while the D2 is docked to the ISS, because the ISS's ECLSS is more-or-less in charge then, and the ISS consumable budget would include any needed atmospheric backfill.

4) It must be the case that the nitrox COPVs themselves don't need to vent for pressure-relief while docked to the ISS, because the ISS can't replenish them, and everything works fine for long-term docking.  I think that means that, as long as free-flying temperature limits don't have weird spikes in them, the nitrox system itself wouldn't vent during extended free-flying.  But that kinda relies on the answer to the questions in point #2 above.

5) Are there non-ECLSS systems that might have consumables?  Helium COPVs would be an example of one such system.

6) A lot of the stuff above is going to depend on lighting conditions.  I would assume that a docked D2 gets at least partial shading from the ISS structure, which would make peak thermal loads (direct sunlight and near-equatorial Earth albedo on the day side) last for less time than it would experience in free flight.  I'd guess that the nitrox supply for 20 crew-days of free flight would be fine for accommodating >1-2σ lighting conditions, which might generate any or all of the following:

a) Cabin pressure relief venting from a pressure spike due to direct heating.
b) Nitrox adiabatic cooling release to cool the cabin down from direct heating.
c) Nitrox cooing, which increases the pressure, which causes a pressure relief venting.
d) Nitrox tank pressure high, which might release directly overboard.

None of these would be a big deal if they happened a handful of times per crew free-flight.  But they might add up to a serious nitrox depletion if that handful of times was multiplied by a factor of 10-20 over the course of uncrewed free flight.

I'm of course guessing on whether such thermal spikes actually exist, and what their duration would be.  But it does seem that they'll be worse in free flight than they would be if the ISS were partially shading the vehicle in day-side equatorial conditions.

7) Leaks.  The nitrox COPVs are sized to deal with a fairly big MMOD-induced leak, with the criterion being that it can backfill enough for an emergency de-orbit and EDL.  But itty-bitty MMOD damage, which would otherwise be considered nominal for a 4-day free flight, could add up to to a big draw-down over 30-60 days.

There's also of course some nominal amount of leakage through valves and fittings that pierce the pressure vessel.

Two things come to mind that might limit endurance.

Station Keeping

Either the quantity of propellent or the number of cycles for the thrusters.

This could also be another thing where popping out on day-side near the equator could be something that required mitigating the thermal spike with something like a barbecue roll.  (I'm on real thin ice here.)

Do we have any data on how often they swap out the Dracos on the D2 currently?  It's an excellent point that extended free flight is a worst-case stressor on those systems.

Mold and Mildew

And other biologicals.  These are dealt with by air movement and filtering, cleaning, or by making the cabin inhospitable to life.  Perhaps it's as simple as opening a large can of silica gel before leaving for the moon.

What's the alien script for "Do Not Eat" that needs to go on the can of silica gel?

It's possible that they could dry the cabin air substantially with a big nitrox pulse to drop the temperature down quite a bit, with a pressure vent at the same time.  Dry air doesn't necessarily eliminate the risk of some kind of overgrowth, but it certainly reduces it.

Such a purge would obviously eat into the nitrox reserves, but if there aren't any nitrox releases in nominal free flight (i.e., my nervous nelly list above isn't a problem), it shouldn't be too bad.  I guess the cold might put a pretty hefty thermal stress on the cabin systems, though.

[TheRadicalModerate]

Why is there a need for station keeping? The craft is free flying and in a stable parking orbit.

See my lighting conditions hand-wringing above.

quantity of propellent

F9 + Dragon likely will not be launching to 51.6°. The additional relative performance of F9 S1 / S2 can be used to more directly inject Dragon into the rendezvous orbit.
Both craft can actively phase their orbits.
OTV LSS can actively take part in prox op docking, reducing Dragon fuel consumption
Dragon already does port relocations during typical ISS missions, simulating the fuel consumption for the "extra" docking with the returning OTV LSS.

I'd think that you'd want to keep the waiting D2 within some range of orbital tolerance, and failure to do so could result in an abort of all or part of the lunar mission.  So there's at least some stationkeeping that needs to occur.

Dragon performs "propellant wasting" burns at the end of typical ISS missions, indicating a surplus of fuel.

Yes, but how much surplus?  Enough to extend the mission by a factor of 10-ish?

[TheRadicalModerate]

I think MMOD risk needs to go near the top of the list of issues.  It's cumulative, so a longer loiter jacks it up.

Not completely on-topic for here, but worth bearing in mind:  Something bad happening to the loitering D2, from any cause, would have to go into contingency planning.

My guess for how to handle this is some kind of "warm standby" for an F9 and D2 on the ground.  This could be as simple as coordinating CRS or private missions so you can get a mostly-ready D2 out of processing and onto a waiting F9 within a day or two, to be launched after the OTV-LSS winds up in whatever LEO it's gonna wind up in.

Also sorta O/T:  This is yet another reason why the "just use two different D2s for departure and arrival" strategy isn't great.  Now you'd need two different D2s,  at advanced stages in the processing pipeline, one for the return, and one in case something bad happened to the return D2.  (If the departure one has a problem, that's a straight-up abort back to EDL--assuming it's not a fatal problem.)

[TheRadicalModerate]

What you said about D2 not venting isn't completely correct.  The ECLSS will vent atmosphere if pressure limits are exceeded.  It's unclear to me whether that's an exceptional event in free flight or not. 

Yes, but I qualified my statement with two conditions

Dragon doesn't vent atmosphere into space for thermal control when the trunk is attached.

1. "for thermal control" - venting exclusively for the purpose of thermal control
2. "when the trunk is attached" - state dependent on if trunk is still attached

Let's take a look at what the SpaceX ECLSS paper has to say about overpressure venting and see if my statement is completely correct.

The only times when active cabin pressure control is expected to be needed is for the nitrox cooling purge during entry and possibly during suit leak checks

So two conditions for Dragon to vent:

1. nitrox cooling purge during entry - the trunk is no longer attached
2. possibly during suit leak checks - not for thermal control

Per your terminology, venting is exceptional. It will occur very infrequently (and at prescribed times) only when there are humans onboard.
There is no regular venting for temperature / humidity control. That is the purpose of (and handled by) the radiator.
The spacecraft does not vent atmosphere to control temperature when the trunk is attached.

Everything you have above is correct--as long as the TCS can handle all of the >2σ thermal loads.  At the very least, there have to be contingency plans (and software) to handle off-nominal heating events on orbit.  Losing one of the TCS loops would qualify.  If that's the case, might they also consider an abnormally cloudy day over the equator, just as the orbital node crosses between the Sun and the Earth, an off-nominal event?

(Yeah, I'm lawyering the bejeezus out of this.  But I want to be sure you can prove me wrong.)

Another random thought:  You have longer peak heating events in a 28º-35º inclination orbit (an arm-wave for something that's suitable for TLI and TEI return), than in the 51.6º ISS inclination.  How much extra heat does that dump into the system on a maximum-albedo day?

[TheRadicalModerate]

Apropos MMOD mitigation for a D2 free-flyer:  How much delta-v would you need to station-keep a D2 in VLEO (<375km circular altitude) for a couple of months?  Debris distribution by altitude spikes up very quickly at about 375km.  Below that, drag cleans stuff out very quickly.

I have a feeling that SpaceX is going to want most, if not all, depot operations as low as possible, because a bad accident with a depot is a catastrophically large debris generator.  Tankage for an OTV-LSS is more than generous, so refueling very low works fine, and the depot doesn't have to move.  Therefore, keeping the D2 more-or-less co-orbital with the big guys makes a certain amount of sense, and certainly dramatically reduces MMOD risk.

The problem, of course, is that it costs more propellant for station-keeping, which is the opposite of what we want if we're to keep the modifications to D2 at a minimum, and hopefully to none at all.

[Eer]

Off the wall variant that <might> violate the "free flying" OP requirement:

1) Launch a StarShip equipped with reclosable chomper or other wide open fairing around the cargo area.
2) In the back (base) of the cargo area, store consumables and a docking port for the D2.
3) When D2 loiter is needed, dock it to the cargo-hold-based docking bay for power and consumable refresh, and close the chomper for additional protection against foreign object collisions
4) when D2 loiter is no longer needed, undock, transfer crew, and return.  Optionally, transfer crew, undock, and return - the sequence depends on how much cargo space is crew-friendly, and whether there is a second docking port (externally accessible - not also in the cargo area).
5) return StarShip for refurbishment and before transfer to alternate orbit as needed.

I think of this as a "garage" or "covered StarShip Park" alternative. Leverage the size, resources, long loiter capabilities StarShip gives you.

[TheRadicalModerate]

Off the wall variant that <might> violate the "free flying" OP requirement:

1) Launch a StarShip equipped with reclosable chomper or other wide open fairing around the cargo area.
2) In the back (base) of the cargo area, store consumables and a docking port for the D2.
3) When D2 loiter is needed, dock it to the cargo-hold-based docking bay for power and consumable refresh, and close the chomper for additional protection against foreign object collisions
4) when D2 loiter is no longer needed, undock, transfer crew, and return.  Optionally, transfer crew, undock, and return - the sequence depends on how much cargo space is crew-friendly, and whether there is a second docking port (externally accessible - not also in the cargo area).
5) return StarShip for refurbishment and before transfer to alternate orbit as needed.

I think of this as a "garage" or "covered StarShip Park" alternative. Leverage the size, resources, long loiter capabilities StarShip gives you.

Not free-flying is OK, as long as it doesn't require a bunch of development work.

Assuming that chomper is still a thing, this isn't terrible.  I like using the Starship fairing as an MMOD first line of defense.  But there are some problems:

1) First, you have to design a D2-docking and -supporting payload package for Starship.  That's not huge amounts of work, but it's probably more than tweaking the D2 itself to free-fly for long enough.  Note that the assumption for D2 docked operation is that it has ECLSS from the host, as well as thermal support.  If you're going to close up the chomper, you also probably need to provide enough heat to the exterior of the D2 that you don't exceed minimum temperatures.  That's probably a fair amount of power.

2) Second, if you bring the Starship down after each mission, its launch cost gets factored into the overall mission cost.  If you've drunk the $5M/launch Kool-Aid, that's not a problem.  But if, as I believe, the actual cost will be closer to $50M/launch, it's not nothing.  It is better, however, than launching two separate D2s, which would cost an extra $200M-$300M.

3) You could leave the Starship in orbit permanently, but then it has stationkeeping prop to manage, and it'll be pelted with MMOD itself.

4) It's one additional mission-critical RPOD.  There's a contingency if it fails (launch a second, warm-standby D2), but NASA hates RPODs, for reasons I don't completely understand.  The SLS/Orion folks are gonna beat the D2/LSS kludge over the head with the two required LEO RPODs already.  Not giving them more ammunition would be desirable.

[DanClemmensen]

Off the wall variant that <might> violate the "free flying" OP requirement:

1) Launch a StarShip equipped with reclosable chomper or other wide open fairing around the cargo area.
2) In the back (base) of the cargo area, store consumables and a docking port for the D2.
3) When D2 loiter is needed, dock it to the cargo-hold-based docking bay for power and consumable refresh, and close the chomper for additional protection against foreign object collisions
4) when D2 loiter is no longer needed, undock, transfer crew, and return.  Optionally, transfer crew, undock, and return - the sequence depends on how much cargo space is crew-friendly, and whether there is a second docking port (externally accessible - not also in the cargo area).
5) return StarShip for refurbishment and before transfer to alternate orbit as needed.

I think of this as a "garage" or "covered StarShip Park" alternative. Leverage the size, resources, long loiter capabilities StarShip gives you.

I don't think this is minimal. You are solving problems that don't really need to be solved for an early Artemis mission. The lives of today's  ISS crews already depend on Crew Dragons that are more-or-less unprotected in LEO for six months. For the kludge SLS/Orion replacement missions, you "protect" D2 by having another D2 available on the ground. The astronauts can stay in the OTV until the replacement D2 can be launched.

Your "garage" appears to be an (at most) minimally-modified version of a "standard" cargo Starship. But the cargo Starship is a new development in the context of the OP. The OP kludge depends only on the Starship variants that are already needed for Artemis III, namely Depot, Tanker, and HLS. It does not assume that an EDL cargo starship exists.

Yes, we know that SpaceX wants an EDL Starlink dispenser Starship as soon as possible, but that's outside of Artemis and is also likely to be simpler than a chomper.

[TheRadicalModerate]

I think MMOD risk needs to go near the top of the list of issues.  It's cumulative, so a longer loiter jacks it up.

Update from the other thread:  In response to something that Twark_Main found, I did a back-of-napkin on a 4-month probability of collision for a D2.  At 25º inclination it looks like the probability of collision per m² per year would have to be so high that D2 is OK at pretty much any altitude.  See here.

Not sure how I feel about this result;  seems a bit optimistic.  Not sure what the model used for the definition of "collision", and even very small objects could cause an air leak that would render the D2 free-flyer unusable.  But the number has a lot of margin in it.

[OTV Booster]

There is one consumable in the ECLSS not yet discussed. The activated carbon filters.


I expect that there's a lot of plastic in the D2 interior. Plastic outgasses just the stuff that the carbon filter traps. Should be an easy fix. More filters.

[OTV Booster]

Does anybody have a clue how the trunk radiator and the D2 are connected? I seriously doubt that it's through the heat shield. That leaves a connection somewhere on the side of the craft.


The connection can't be too close to the bottom or there will be thermal problems during EDL. Even if the materials can take it there would be unnecessary thermal transmission to the interior that would call for increased nitrox flow for cooling. What I picture is QD connectors up high and the connection lines staying with the trunk when it disconnects.


From the conversation to date it doesn't look like there are any insurmountable problems in extending D2 loiter time (except maybe MMOD) but the trunk is there 'just in case'.


A hunch: The loiter limits are self imposed because they are what was planned for and qualified.

[OTV Booster]

A late thought. Station keeping propellant shouldn't be a problem. Even at VLEO the orbit decay shouldn't be all that much and there should be plenty of propellant for a small boost or two. That's strictly a gut estimate.


What would be a problem would be attitude control - especially at VLEO. I'd classify modifying the current tanks as non-trivial. Adding in a tapoff from extra tanks in the trunk? Again not trivial, but trivial enough?

[MDMoery]

Does anybody have a clue how the trunk radiator and the D2 are connected? I seriously doubt that it's through the heat shield. That leaves a connection somewhere on the side of the craft.

Here is a good shot of trunk jettison on the in-flight abort test (2:54 video, 2:24 MET).  You can see the umbilical "claw" at the bottom of the trunk as it falls away from the spacecraft.


You can see the umbilical attached to the spacecraft below and to the right of the hatch.  https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=41016.0;attach=1492554;image

Here you can see all the connection ports without the trunk attached.  https://www.aerospacemanufacturinganddesign.com/remote/aHR0cHM6Ly9jZG4uZ2llLm5ldC9maWxldXBsb2Fkcy9pbWFnZS9zcGFjZXgtdjItdW52ZWlsMDUyOTE0LXNwYWNleF82MjB4LmpwZw.lstuMblkq6I.jpg?w=948&h=533&mode=pad&anchor=middlecenter&scale=both&bgcolor=F0F1F2

[Asteroza]

Does anybody have a clue how the trunk radiator and the D2 are connected? I seriously doubt that it's through the heat shield. That leaves a connection somewhere on the side of the craft.


The connection can't be too close to the bottom or there will be thermal problems during EDL. Even if the materials can take it there would be unnecessary thermal transmission to the interior that would call for increased nitrox flow for cooling. What I picture is QD connectors up high and the connection lines staying with the trunk when it disconnects.


From the conversation to date it doesn't look like there are any insurmountable problems in extending D2 loiter time (except maybe MMOD) but the trunk is there 'just in case'.


A hunch: The loiter limits are self imposed because they are what was planned for and qualified.

There's a side clamp doohickey on the edge of the heatshield that provides connectivity.

A late thought. Station keeping propellant shouldn't be a problem. Even at VLEO the orbit decay shouldn't be all that much and there should be plenty of propellant for a small boost or two. That's strictly a gut estimate.


What would be a problem would be attitude control - especially at VLEO. I'd classify modifying the current tanks as non-trivial. Adding in a tapoff from extra tanks in the trunk? Again not trivial, but trivial enough?

If the power budget can cover it, maybe small reaction wheels in the trunk?

[Zed_Noir]

<snip>

A late thought. Station keeping propellant shouldn't be a problem. Even at VLEO the orbit decay shouldn't be all that much and there should be plenty of propellant for a small boost or two. That's strictly a gut estimate.


What would be a problem would be attitude control - especially at VLEO. I'd classify modifying the current tanks as non-trivial. Adding in a tapoff from extra tanks in the trunk? Again not trivial, but trivial enough?

If the power budget can cover it, maybe small reaction wheels in the trunk?

Installed a few Starlink Argon thrusters in the trunk as RCS for fine attitude control. As long as you don't need quick attitude adjustments.

Bonus of some orbit change capability.