Vast, a Startup for "human habitation, first in LEO, and then beyond"

[JohnFornaro]

See the quote in my footer...

There's a viewpoint that the universe we live in is suitable for "all sites, people, and situations".

I think you're misunderstanding the quote.

I got your meaning.  I was pointing out that there's more than one "viewpoint"

[JohnFornaro]

Absence of evidence is not evidence of absence.

But of course.  Even so, the design as presented and as analyzed here leaves too much to the imagination and is faulty in several important ways, leading to a preliminary suggestion of, "evidence of absence".  Certainly, they don't want to reveal proprietary information, but consider that the nations ofthe world aren't exactly building space stations at an appreciable rate.  IOW, give me the entire set of engineering drawings and wait around for a few hours/days/decades while I build it, launch it, and populate it.

[JohnFornaro]

From that youtube video: "[Ring stations] are huge and heavy and require tons of launches."  Sigh.



How many times do I have to repeat myself: Mass is your friend!  To reduce the video's argument to the absurd:  We're not going to get much science done by strapping ourselves to a rotating turntable.  One at a time.

My initial idea for constructing a ring station incrementally, included the first design of two dragon capsules connected to tethers around a non-rotating hub.  To this basic structure, a 1K yard radius, four quadrant [or six hexants?] tether scaffolding would be added.  This would be strong enougth to gradually and symmetrically add modules to fill out the station.

Now I'm of the opinion [remembering that opinions aren't facts] that the concept of a ring station is sound. Build the prototype on Earth, and launch and install the pieces in an orderly fashion.  Either our modeling software is up to the task or it is not.

Another way of approaching the problem is to assume the VAST model as being an initial spoke of the ring station.  An important fault in their design is that the hub rotates with the spokes, when it should be not-rotating in the spoke reference frame.

BTW, What is the point of the "decorative" rings at either end of the station?

[edzieba]

How many times do I have to repeat myself: Mass is your friend!  To reduce the video's argument to the absurd:  We're not going to get much science done by strapping ourselves to a rotating turntable.  One at a time.

Well, we have. We even have labs dedicated to doing just that. Not just sit-in-a-seat-and-rearrange-your-face centrifuges, but up to and including entire rotating rooms. Plenty of science has been done on rotating environments combining rotational acceleration with a 1g field, and just the ability to do the same but with the 1g field eliminated would be very useful. Only budget cuts prevented such a lab being installed on the ISS.

(Reductio ad Absurdum only works if your Absurdum is actually absurd)

[Twark_Main]

See the quote in my footer...

There's a viewpoint that the universe we live in is suitable for "all sites, people, and situations".

I think you're misunderstanding the quote.

I got your meaning.  I was pointing out that there's more than one "viewpoint"

Yes. There's a right viewpoint and a wrong viewpoint.

The wrong viewpoint is that there's a single "perfect solution" for every imaginable situation. The right viewpoint is that the solution should match the context, customer, and purpose.

Every engineer knows this.

[Twark_Main]

the design... leaves... much to the imagination

Agreed.

The only difference between us seems to be whether one chooses to "fill in the details" with stuff that actually works (ie giving Vast the benefit of the doubt), or not.

You have chosen "not."

That's okay, of course. It just means you'll fail to predict any creative solutions Vast might have in the wings.

Still, if you didn't even try to come up with any solutions (just bailing out at the first sign of trouble instead), then how credible can the resultant feasibility estimate really be? 

[Twark_Main]

Mass is your friend! ... We're not going to get much science done by strapping ourselves to a rotating turntable.

So "mass" isn't actually your friend.

Experimental data is your friend.

If you can get the same experimental data with less mass then that's a win, not a loss.

[Twark_Main]

BTW, What is the point of the "decorative" rings at either end of the station?

I've mentioned them a couple times:

I still can't tell what the circles and lines on the end mean. I think they might just be artistically-done ellipses, indicating the possible continued addition of modules on each end. They're also asymmetrical — non-androgynous interface? Really guessing here.

Yes, I think those "ellipses" (both geometric and typographic, get it??) are meant to show that the design concept is scalable in that way.

[mikelepage]

Mass is your friend! ... We're not going to get much science done by strapping ourselves to a rotating turntable.

So "mass" isn't actually your friend.

Experimental data is your friend.

If you can get the same experimental data with less mass then that's a win, not a loss.

My two cents, (with the caveat that my startup is looking at a different spin-gravity concept): What is actually needed, is to optimise for usable volume (at a given gravity level) per moment of inertia unit (kg.m²)...

Which for us means we're pursuing a lightweight/smaller-scale torus concept. The utility of having every gravity level available simultaneously - possible in a baton station - is overrated. What you actually want is the ability to supply *any* gravity level between 0-1xG, and change between levels easily. This means changeable spin rates, so a minimised moment of inertia, but also a mechanism that can simultaneous maintain some experimental payloads at zero gravity - necessary as a control for many space experiments.

[JohnFornaro]

We're not going to get much science done by strapping ourselves to a rotating turntable.  One at a time.

Well, we have.

One of the reasons space exploration takes so long and costs so much is because of endless study under the assumption that all scenarios can be modeled and then select the best model.

We know that if you drop your ice cream cone on the hot summer sidewalk, it's going to melt.  We don't need to test all sidewalk temperatures, height of cone drop, effects of sugar cones versus waffle cones, etc. 

We know that terrestrial life can thrive on one gee, and that 1rpm is not disruptive to the inner ears of the vast majority of humans.

It would be a false choice to suggest that there's only two choices of study.

(Reductio ad Absurdum only works if your Absurdum is actually absurd)

Well, the image of the guy strapped to the turntable is pretty absurd on its own

[Paul451]

Can't see any sign of such a vehicle on their site.

Unsurprising, since the station render itself was only posted a few days ago.

I meant it isn't mentioned on the site, anywhere in the description of the station.

For all we know, they intend to despin during Starship resupply/recrew, and perhaps not even dock at the centre but to the "bottom" of the last module (normally 1g) at the end of the baton; and consider each module to be independent enough to not require an emergency evac vehicle(s).

Or just aren't bothering about such details at this stage, since this isn't actually a design, just a sketch (literally and figuratively) of a concept.

No radiators either.

As mentioned, I suspect the radiators are on the backside of the PV. If you look closely, you'll note that the "PV" is actually made up of two slightly different size panels, one in front of the other.

I think you are reading way more into a hand-drawn sketch than is there.

It looks like the panels might not pivot.

I'm not trying to read anything into the "look" of the sketch, my speculation about them being fixed is because they can be, so why wouldn't you. The rotational axis can be pointed at the sun, so the arrays will permanently face the sun. (Pedantically, the whole station will then need to rotate its axis once per year, but they need a way to control precession, so that capability seems like a given.) That said, you might want to be able to "feather" them, to match supply/demand with less storage. Or perhaps even to tilt them back and forth in the third axis, to use them as a crude momentum-exchange dampener.

Either way, they don't need to rotate constantly to follow the sun, the way Coastal Ron is apparently picturing. Which was both our points.

[Paul451]

My two cents [...]: What is actually needed, is to optimise for usable volume (at a given gravity level) per moment of inertia unit (kg.m²)...
The utility of having every gravity level available simultaneously - possible in a baton station - is overrated. What you actually want is the ability to supply *any* gravity level between 0-1xG, and change between levels easily. This means changeable spin rates

The graph gets populated at the same rate (assuming similar overall station sizes), just the way it's filled-in differs, depending on whether you optimise for as many samples as possible for each g-level, but a single g-level per "run" (eg, torus), or studying a wide range of g-levels but with fewer data-points at each g-level per "run" (eg, baton).

Both architectures can end up with the same range and fidelity of data, but with the the non-optimised thing coming at the expense of time.

My gut feeling is that we are so short of data for comparative analysis between g-levels, more-levels-sooner-with-less-data-at-each-level (eg baton) has a bigger payoff than more-fidelity-at-each-level-but-one-level-at-a-time (eg, torus). Especially to find effects with a j-curve or s-curve, where you want to find a specific g-level where the effect gets noticeably non-linear. You can then focus on those interesting g-levels to get more data over time, to pin down finer details, but the low-hanging-fruit is finding those interesting cross-over-levels quickly. Which favours the baton.

[But we've had this kind of debate before. And definitely not criticising your efforts. It's a design choice, not an absolute. And you guys are putting your money where your mouth is, I'm not.]


It occurs to me that if a company's ultimate goal is to develop a multi-ring torus, there's a similar "same result but different paths" for starting with a single-ring torus vs a multi-level baton.

A torus could start with a node and inner ring, spun-up and commercially operational while they develop/build/launch the hardware for a new, wider ring, and then another, and another.

For a baton, the initial baton serves as two spokes from which you expand sideways at different levels to eventually become multiple-rings. (The ring modules can have the same mould-lines as the vertical baton modules, but mounted sideways.) Again, the initial baton can be operated commercially while developing/building/launching the next set of hardware.

That said, I don't think this baton-station would be upgradable. Other than expanding the length. Indeed, they might be adopting the baton design specifically to avoid the cost/complexity of developing cross-junction modules (like ISS's "node" modules.) And if they are adopting SpaceX incremental development, then the early modules would be considered expendable anyway.

OTOH, if the vertical modules don't have to be permanently joined together, you could still insert cross-junctions/nodes between existing simpler modules for sideways expansion at a later date.

[JayWee]

My gut feeling is that we are so short of data for comparative analysis between g-levels, more-levels-sooner-with-less-data-at-each-level (eg baton) has a bigger payoff than more-fidelity-at-each-level-but-one-level-at-a-time (eg, torus). Especially to find effects with a j-curve or s-curve, where you want to find a specific g-level where the effect gets noticeably non-linear. You can then focus on those interesting g-levels to get more data over time, to pin down finer details, but the low-hanging-fruit is finding those interesting cross-over-levels quickly. Which favours the baton.

The baton has one big disadvantage - you have to physically move the experiment. Are you thinking of some kind of a guardrail to move it? Very space inefficient.
Multiple racks at set distances you manually swap? Takes astro time and might mess-up the experiment, likely will need some recalibration. And creates discreteness. So in the end you might end up spinning/despinning the baton anyway.
Thus the torus looks easier (docking in the stationary zero-g middle), can be easily spun/despun with very precise continuous g-control.

[Paul451]

[experiments across multiple g-loads]

The baton has one big disadvantage - you have to physically move the experiment. Are you thinking of some kind of a guardrail to move it? Very space inefficient. Multiple racks at set distances you manually swap? Takes astro time and might mess-up the experiment, likely will need some recalibration. And creates discreteness.

What kind of experiment would need to dynamically vary the g-load on a single sample? The only one I can think of would be "how do humans cope with moving between g-levels?" And obviously, it's the humans moving around the station.

For everything else, you are running different experimental batches in each g-level for their entire run. One set in each level. That's the whole point.

You aren't starting an experiment on level 2, then carrying it to level 5 for awhile, then down to 7...

You want to compare, for example, mice raised at 1g to mice raised in 0.7g, to those in 0.4g, to those in 0.2g, to those in zero-g.

And this would also apply to a variable-spin torus-station. Although in its case, all samples in each experimental run are at one g-load from start to finish. When you change the spin-rate, it's to do another run at the new g-load. Again, you aren't starting at one g-load, and changing the spin-rate multiple times for the same batch.

The difference between baton and torus (assuming the same number of mice in each station, and the same length of each run) is that the baton has fewer mice at each g-load, but tests many different g-loads simultaneously. The torus only does one g-load at a time, but can house more mice for each run, reducing error-bars. The baton takes more time to increase the depth of data at each g-load, the torus takes more time to sample different g-loads.

For example, say each run is a batch of 100 mice for 24 months (roughly the avg lifespan of a mouse.) The baton might have 20 mice in each of 5 different gravity levels. The torus has 100 mice all at one gravity level. Over 10 years, you run 5 batches of mice, and both stations can plot the results of 100 mice raised at 5 different g-loads. But after the first 2 years, the baton has a crude plot of 20 mice at 5 levels, possibly giving a much quicker insight into where different health effects kick in.

[Coastal Ron]

The panels (and the station) rotate in a flat plane. The rotational axis points in a constant direction. Point it at the sun and the panels are always perpendicular to the sun. They don't need to rotate at all. And indeed, might be fixed in place.

Yeah, that should have been obvious, and I missed it - I was too much in ISS mode. Plus, the designs I work on are not in orbit around a planet or moon, so I don't normally think about those issues... 

...which adds momentum to the plane of rotation...

Take any point on the solar panel, then measure across the center of gravity for the cross section of the station, and you'll find the momentum change is likely negligible regardless how the panels are rotated.

We still fundamentally disagree on this point, and the sun orientation of the panels was not a factor. The station does not have some sort of "keel" that that keeps the cylinder from rotating, and even if there was adding a mass on the side of that cylinder while it is rotating at 4 RPM is going to want to change orientation.

My opinion, which you disagree with, so we can move on...

If the solar panels have more potential angular momentum than the docked vehicle, then the panels are the second axis of inertia. Since they are in the plane of rotation, the design would be stable.

If the station is perfectly balanced when there are no vehicles (i.e. significant mass) attached to the side of it, that balance will change when a vehicle of any significant mass connects to the station.

Some thoughts:

1. The station could de-spin in order to take on supplies and exchange crew. This could be done with electric thrusters spooled out from the ends to some distant length, and let leverage of low thrust on a long moment arm do its thing. However that could take some time, and while you are de-spinning all of your experiments go out of testing bounds. Not sure if de-spinning with larger thrusters makes sense, because where do you store the propellant?

2. If the station is rotating at 4 revolutions per minute in order to create 1G at the ends, how do you spin a Starship up to 4 RPM and have a predictable center of rotation? And as you transfer mass between Starship and the station that is going to change the balance of the Starship+station, and I have no idea how it doesn't start to wobble and move about in ways that are not safe for undocking.

3. What I'm assuming for my rotating stations is a standardized cargo module that will dock with the station, and dock with visiting vehicles. That way visiting vehicles don't have to be designed to rotate at 4 RPM. The cargo module could be docked at a maintenance station close to the main station when it is not needed, and that could be where excess propellant and supplies are kept. Plus trash. 

Regarding 1G at 4.25 RPM, I don't understand why they would want to test for that, because we know that humans are already adapted to 1G, so I'm not sure why that should be a high priority. The priority should be Mars (38% of that of Earth) and our Moon (16.6% of that on Earth), since those are the places with the most interest for humans to spend significant time.

[JohnFornaro]

If the station is rotating at 4 revolutions per minute in order to create 1G at the ends, how do you spin a Starship up to 4 RPM and have a predictable center of rotation?

This is the part of their proposal which doesn't work at all.

Spinning the rotation up and down is not a good idea either. 

They need a central, non-spinning hub with electromagnetic bearings and enough play to balance the Starship when it docks.  An adjustable dead mass on the other side of the baton would also be helpful.

But what is the approach of the Starship?  Nose first, perpendicular to the axis of revolution? Or parallel, somehow at the center of mass of Starship? 

[Coastal Ron]

If the station is rotating at 4 revolutions per minute in order to create 1G at the ends, how do you spin a Starship up to 4 RPM and have a predictable center of rotation?

This is the part of their proposal which doesn't work at all.

Spinning the rotation up and down is not a good idea either.

Yeah, it will be interesting to see what solution they go with.

They need a central, non-spinning hub with electromagnetic bearings and enough play to balance the Starship when it docks.  An adjustable dead mass on the other side of the baton would also be helpful.

But what is the approach of the Starship?  Nose first, perpendicular to the axis of revolution? Or parallel, somehow at the center of mass of Starship?

They do need some way to capture visiting vehicles that are not spinning. This is the same challenge for larger stations too, since spinning up a vehicle takes too much weight and balance effort, since even the smallest of imbalances will create increasingly large wobbles. So some way to reduce the vehicle complexity would be optimal.

Which is the same situation with assuming any large spaceship should be allowed to dock with such a (relatively) small rotating space station. Any imbalance between the 4 RPM station and the non-rotating Starship could be VERY dangerous, so why even risk it? Just use an intermediary cargo vehicle that places the least stress on the station, and doesn't require any special modifications to any visiting vehicle.

[Paul451]

...which adds momentum to the plane of rotation...

Take any point on the solar panel, then measure across the center of gravity for the cross section of the station, and you'll find the momentum change is likely negligible regardless how the panels are rotated.

We still fundamentally disagree on this point, and the sun orientation of the panels was not a factor. The station does not have some sort of "keel" that that keeps the cylinder from rotating,[...]
If the station is perfectly balanced [...]
And as you transfer mass [...] that is going to change the balance of the Starship+station

since even the smallest of imbalances will create increasingly large wobbles.

Perfect balance isn't required. Rotation around the short axis is inherently stable. "Inherently stable", means it tolerates imbalance and moving mass.

Look, go get a paperback novel, or similar shaped object that won't break if it hits the floor but is dense enough to overcome air-resistance.

No, seriously, actually go do this right now.

You just wont get it unless you see it in person.

Throw the book up in the air, flat like a frisbee, rotating around it's short axis. That's how the baton-station rotates, with it's solar arrays/etc in the rotational plane. Note that it's highly stable, in spite of being flappy paper, in spite of not being "perfectly balanced". This is the most stable orientation for any object to rotate. It's highly tolerant of disturbances.

Throw it up in the air again, this time rotating around it's long axis like a BBQ-spit. Note that it's still stable. In reality, it's not as stable. Over time, if it were able to spin in free-space without falling to the ground, it will "nutate" around the short axis until it's rotating in the flat plane like a frisbee. This is because it's made of flappy paper and can exchange momentum between axes; if you used a solid object, it wouldn't even do that. There are tricks to stop it happening that don't rely on complex gyroscopes/etc, but that's not relevant for this design anyway, just good to keep in mind when talking about, say, Starships rotating on their long-axes. Highly tolerant of mass imbalances, but need to control long-term nutation.

Lastly, throw it up in the air again, now rotating around the second axis. Harder to describe this (pancake flip?), but it's the only one left. It will immediately flip around as it rotates. This is the classic "intermediate axis" issue. If you could "perfectly balance" it, it wouldn't do this. But here in reality you can't, so in practice it's instantly 100% unstable every time.

This is the important part: Now go back to the frisbee throw. But this time stick some silly-putty to the cover (or masking-tape some weights on). One side only. Completely unbalanced. Keep increasing the mass between throws. Note how much mass you can add (relative to the mass of the book) before you turn the highly stable short-axis rotation into an unstable intermediate axis rotation.

If you tape on longer masses, sticking further out from the cover, you'll notice that it's much easier to get the intermediate-axis tumble. The increased radius means the same mass will have more rotational inertia, eventually overcoming the inertia of the in-plane mass. With exactly the same mass closer to the cover, and the book is still stable.

When you get enough mass on the cover to destabilise the book, try adding some mass to the edges of the book (like solar arrays), near the ends, and notice how much stability it adds. The inertia in the plane of rotation matters. Even when balanced on both sides, they don't "cancel out".

Keep varying the masses, distances, positions and axes until it just makes sense in your head, until you can predict what is going to happen from any configuration, until you understand how it relates to the solar arrays on the baton station, docked ships, cargo moving, etc etc.

Now you can move on.

Some thoughts:
1. The station could de-spin in order to take on supplies and exchange crew. This could be done with electric thrusters spooled out from the ends to some distant length, and let leverage of low thrust on a long moment arm do its thing. However that could take some time, and while you are de-spinning all of your experiments go out of testing bounds. Not sure if de-spinning with larger thrusters makes sense, because where do you store the propellant?

A 100m long station at 4.25RPM, gives 22m/s at the tips.

Rotational acceleration vs linear, for this shape, reduces the effective pseudo-mass to roughly 1/3rd equivalent for linear acceleration of the same object. Assuming 100m length, 6m width, and thrust applied at the tips. F = I*a/r^2 = ~1/3*m*a. Since a = delta_v/t, required delta-v for 22m/s rotation is approx. 22/3 = 7-and-a-bit m/s.

Call it 15m/s for a full despin + respin.

I doubt you have to use anything fancy beyond regular 200-230s Isp mono-prop hot-gas thrusters.

2. If the station is rotating at 4 revolutions per minute in order to create 1G at the ends, how do you spin a Starship up to 4 RPM

No-one's assuming Starship docks to a rotating station. We've all said it's not. Either they despin, or, as TM suggested, use a dedicated logistics-carrier type module as a shuttle between ship and station.

Regarding 1G at 4.25 RPM, I don't understand why they would want to test for that, because we know that humans are already adapted to 1G, so I'm not sure why that should be a high priority.

Control group.

Lets you separate the side-effects of spin (if any) from the effects of partial-g, and controls for effects of the station itself: ECLSS, air-quality/VOCs, vibration, noise, people/carers, that annoying high-pitched buzzing from the lights they won't fix I keep putting in maint.reqs but noone ever does anything this wouldn't happen at NASA I hate my job if I wasn't under contract I'd...

It's not highly necessary (IMO) for early work looking for big-obvious effects. But if there is an unknown quirk on the station equivalent to the CO2 issue on ISS, you'll end up needing to do a 1g run to figure it out.

[JohnFornaro]

They do need some way to capture visiting vehicles that are not spinning. This is the same challenge for larger stations too, since spinning up a vehicle takes too much weight and balance effort, since even the smallest of imbalances will create increasingly large wobbles. So some way to reduce the vehicle complexity would be optimal.

Which is the same situation with assuming any large spaceship should be allowed to dock with such a (relatively) small rotating space station. Any imbalance between the 4 RPM station and the non-rotating Starship could be VERY dangerous, so why even risk it? Just use an intermediary cargo vehicle that places the least stress on the station, and doesn't require any special modifications to any visiting vehicle.

Yahbut:  The intermediate cargo vehicle adds complexity of it's own, plus the necessity of moving everythign twice.

Non-rotating hub.

[JohnFornaro]

Perfect balance isn't required. Rotation around the short axis is inherently stable. "Inherently stable", means it tolerates imbalance and moving mass.

Look, go get a paperback novel ... [and complete this simple analogous exercise]

Bottom line, which Paul hasn't gotten to yet.  Non-rotating hub.