Starship itself will not be a great deep space delivery vehicle, in the same way current disposable second stages are used. Too big and too... reusable.

However, it'll make a great delivery system for some serious high-energy escape velocity "third" stages, rather like the Shuttle "Inertial Upper Stage" that delivered Magellan and Galileo and Ulysses.

Well, Starship is specifically designed to make it to Mars if refueled in LEO. So it could deliver monstrously large Mars rovers. If they're going for the outer planets, they could do as you suggest after the first burn using a refueled Starship. Either way you're sacrificing a Starship second stage (old, or better yet reduced mass version) to the gods of delta V.

A re-fueled Starship will be able to deliver ~100 tons to the surface of Mars, to be compared with the heaviest existing rover that was slightly over 1 ton. Yes, it will take ~10 Starship flights to fully refuel one Mars-bound Starship, but the fully reusable design is supposed to ensure this can be done for relative cheap.

The vision I have for NASA missions is to avoid these tiny one off probes and use starship to lift multiple boosters to orbit, assemble a server rack and multiple imagers, a radio thermal power source, and boost it to orbit around each outer planet (except perhaps Jupiter, too much radiation).

It's just not the right business model to build "small" probes like this that trickle back imagery when we can be downloading the same imagery AND querying data that's sitting in situ to run analysis on a planet -local server of TB of real time signals.

Imagine you see something interesting in one image, and just ask the server to find local similar things all over the planet from the imagery in storage. Instant frequency and geospatial analysis product without begging for observation time or (worse) a new mission.

Oh and a kilometer scale telescope in orbit.

I think that there will always be a desire for small probes, but you're right that one-offs aren't economical.

The advantages of Starship is that with such a massive payload capacity it can carry many small probes and combined with a low cost per kg it will justify setting up assembly lines that churn out small probes by the thousands.

We'll pepper the solar system with these probes and they'll act as nodes in a large interplanetary network relaying data between each other and larger, more powerful nodes as you describe.

The nodes in a network are important and more powerful nodes are more desirable, but the network itself is a powerful thing that it greater than the sum of its parts.

This combined with economies of scale will always justify more smaller satellites.

With economics, I think we have to be careful when questioning the financial efficiency of organisations like NASA, Universities, research, and the like. Some of these things should be societal cost centers.

The question for me therefore, based on your comment, is whether Europa Clipper’s expense can be justified because it is funding the current generation of custom space vehicle and mission experts, keeping alive their traditions and skills for the next generation.

The alternative view is that it’s a waste of money to continuously support the careers of what are possibly the spaceship equivalents of wheelwrights, stable masters, and saddlers.

I believe that it absolutely can be justified for precisely the reason you state -- it keeps the experts trained so that they can as a minimum produce things like Europa Clipper. But to me that's just keeping their expertise and the potential that it offers on life support.

Having the technology and infrastructure to produce the things that they are experts at producing but in an unprecedented scale is what we need if we want them to reach their full potential.

It allows them to more rapidly iterate over design ideas by seeing what works and what doesn't. It also allows cheaper destructive testing to work out the bugs instead of waiting a decade to find out if they just wasted billions of dollars and centuries of man years with intrinsically flawed designs.

And ultimately it allows us to build modular self-assembling spacecraft on a scale that we've only dreamed of allowing them to practice their trades in new ways that will unlock new science and knowledge for the human race.

With cheap and commonly launched rockets like starship we'll also be able to launch many more cheap test platforms that won't need billion dollar budgets. Take the antenna on this thing, they get one chance to deply it correctly. But now imagine launching a test antenna project where for a few million you can make sure they work before you put them on a more expensive probe.

Well, sure, but how does "peppering the solar system" with 100 probes at a time work when each insertion is wildly different? Lifting 100 probes is great, but only if they are going on same / similar boosts to their targets.

And how does it get the initial boost along its trajectory? We could do solar electric and wait a year for it to pick up enough delta-v to start the journey ...

Fully disclosing my ignorance here: How does it work to launch 100 probes to 10 planets from earth with one launch?

I'm not sure about that specifically, I've only taken an intro to orbital mechanics class so esoteric orbital maneuvers are a bit beyond me but I understand that in some cases it is possible. I would imagine that one technique that we have yet to try is putting fuel depots in long distance places to facilitate the execution of these manuevers.

With that said what I had in mind when I wrote my comment wasn't using one rocket to launch a 100 probes to different solar bodies but instead to launch 100 probes to a single body in one go. I envision a time where we'll deploy massive constellations of remote sensing satellites, like mini-star link constellations for each solar body be it a moon or planet so that we can get real-time data across the whole body instead of having to stitch together pieces of data from one probe. Like Starlink these constellations will be able to relay data between the nodes should they lose line of sight with Earth and will be able to relay data to other constellations orbiting other nearby solar bodies to get the data to Earth if the whole constellation is blocked.[0]

I could see a near future where we start sending these Starlink like communications/remote sensing craft at the same time we send craft designed to pepper the surface of a body with multiple rovers and stationary probes to survey the surface of planets and again relay the data they collect between themselves and the constellations above.

It would be cool to see terrestrial bodies explored with some sort of RTG/solar powered stationary device that has an automated lab for sample analysis and seismic measurements paired with a dozen or so of these sandflea type robots[0] that can roll and hop over the surface to survey and collect samples to bring back to the stationary labs for analysis.

You can even look at sample return missions where some of these stationary devices have the means to send samples into orbit around the solar body that they're located on where the sample carrying craft rendevous with an orbiting craft designed to return to Earth perhaps refueling with fuel depots previously put in place along the way.

[0] https://www.youtube.com/watch?v=XTmYm3gMYOQ

[1] https://www.youtube.com/watch?v=6b4ZZQkcNEo

This is a much better plan. One issue with 100 probes around, say, Enceladus, is the downlink bandwidth. This is why I suspect on site infra is going to be required.

But honestly let's do both. Infra and something like Planet Labs for outer planets. We can.

There will absolutely be a place for super nodes in the kind of networks that are going to be built out.

Their performance will likely be somewhat limited compared to what you're looking for at this point due to radiation and the excessive mass of the shielding required to protect them from it.

That can be mitigated with in situ resource production. Like burying a server rack scale processing node under ground on mars and relaying data to it with the network I described above or capturing water from comets and using it fill compartments for radiation shielding in orbital data centers.

Related to your question, but not responsive to the “speed” aspect: Here is a nice article from Ars [1] on the impact of starship on large space telescopes. This is focused on Webb successors (which in NASA fall under astrophysics, not planetary, like Clipper) not “planetary probes” per se, although a linked and more technical article [2] touches briefly on that.

One coauthor of [2] is the chief scientist for astrophysics at JPL, so it’s not just idle speculation.

The ability to launch an 8m telescope definitely has space mission designers interested. However, probe velocity itself is not addressed.

[1] https://arstechnica.com/space/2023/10/astronomers-say-new-te...

[2] https://pubs.aip.org/physicstoday/article/76/2/40/2869438/Ac...

Yes and yes. Starship offers about twice the delta-v and triple the mass capacity as Falcon Heavy. Capitalising on the former for a deep-space mission, however, would require sacrificing some reusable.

Given its relatively low cost, it would be entirely possible to use a non-reentry starship as a bus for an interplanetary probe with extra fuel capacity. They'd just have to strip down their lunar starship and possibly add segments for larger tankage.

A little, but mostly insignificant. The real practical benefit is that if Starship succeeds to be as reusable and rapidly launchable as claimed, its easier to line up launches with gravity assist windows as needed, and thats where you get the huge velocity increases

Starship lets you trade money and mission complexity for time. A fully refueled Starship in LEO should have around 5km/s in an expendable configuration. A gravity assist from Earth or Mars typically provides 2-5km/s. Looking at the https://www.reddit.com/r/space/comments/1ktjfi/deltav_map_of... 5km/s from LEO is barely enough to reach Jupiter. Looking at the mission trajectory direct launch would shave off roughly 2 years from 5 year mission - https://europa.nasa.gov/mission/timeline/

You could also launch much-much heavier probe with a dedicated boost stage and / or electric propulsion...

The largest probe ever launched afaik is Voyager 1, and it was less than a ton. Even 'massive' things like the James Webb telescope only weighs 7 tons.

Starship is aiming for being able to launch 200+ tons. So stuff like this on a dedicated mission would have vastly more delta-v available but also be kind of silly. There's no reason we couldn't scale things up and just launch hundreds of probes at once all through the Solar System. Alternatively instead of launching probes we could look to start launching modular observation stations that could one day even house humans, or perhaps even grander ideas.

Launching one off probes, rovers, and satellites is really something I think we should be aiming to move beyond, let alone on the ridiculous time intervals we operate on at present.

We shouldn't still be making basic, yet revolutionary, discoveries on the Moon, more than half a century after we set foot on it.

Comms is the reason. DSN dish time is not a particularly abundant resource even today.

The far side of the moon is prime real estate for radiotelescopes. You can create a sizable expansion of the DSN by sharing dish time with the antennas you can build from local materials.

Well I fully assume that the DSN itself will move to space, courtesy of Starship's large tonnage to orbit.

I've been thinking for years we should built a space network of probe and satellite like objects - imagine if we landed a small relay on every asteroid passing anywhere near earth - could be highly useful for communication and possibly also give us a GPS like location assist for additional spacecraft

Starship can be used to release a probe with gigantic fuel reserve and a bigger than usual energy source (solar or nuclear). Then the probe can use a VASMIR or other electric propulsion to gradually accumulate a vast amount of delta-v

A fully refueled starship in leo is a big big ask, akin to getting a full shuttle stack into orbit. On-orbit refueling at such a scale has not been done yet and is difficult to even think about doing. It isnt like KSP. The more efficient route would be to scrap starship and design a high-energy upper stage to be launched by the BFR instead, a purpose-built probe thrower. Starship is way to heavy to be a practical final stage for probes.

On-orbit refuelling at such a scale does sound rather difficult, but they also fully intend to do it in the near future. They're certainly going to give it a good try.

Also, if you don't want to make an entirely different second stage to put on top of superheavy, you should be able to fit a pretty substantial third stage in Starship's payload bay.

Right now the payload bay doesn't open anywhere near enough for that, but hopefully in the future that'll be an option.

And if you really want to make something go fast, you could refuel in orbit, give it a push from Starship, then deploy the third stage for another kick.

> Right now the payload bay doesn't open anywhere near enough for that

Shouldn't be too hard to design a starship with a blunt nose and place it under an expendable fairing.

With a 1-ton probe the budget is 99 tons for a kick stage. That's a lot of delta-v. With some creative propulsion options, maybe even interstellar probes can be done.

Everything you said is right, but on-orbit refueling is of course theoretically possible, and isn’t THAT huge of a leap given what SpaceX has already accomplished. Let’s not forget the space lasers, or the chopstick catch, or Starlink-live-streamed 4K views from the plasma shadow during reentry, or autonomous Dragon docking, etc.

Sure, it’s hard, but that is perhaps the best engineering team in the history of humankind, and I think they’re up to the task.

Additionally, people much smarter than me ALSO think they are up to the task, and they have more and higher-resolution data about the problem space, too.

My money’s on they get it working. Maybe it takes five years or a dozen failures, but I’m confident that they will succeed eventually and there will be orbiting methane and lox fuel stations sooner rather than later.

Those things were never seen as impossible, just difficult. Nobody has even a theoretical concept of how to move tons of liquid oxygen between rockets in orbit. And there is some talk of banning it as too dangerous. Remember all those pad explosions related to fuel issues? We cannot risk that much space debris as spaceX goes through a learning process.

If I'm not mistaken, SpaceX is already contracted by NASA to do on orbit refuelling, and one of the recent starships did initial trials of fuel transmission (internal for now).

Update: - source article - $58M contract to do that - https://www.teslarati.com/spacex-starship-nasa-contract-orbi...

- chatgpt confirmation + links to more sources - https://chatgpt.com/share/e/674b5c1b-d318-8004-b79d-c0eefc59...

- "A fully refueled Starship in LEO should have around 5km/s in an expendable configuration."

That can't be right; a direct application of the rocket equation gives (3.72 km/s) * ln(1,300 mT / 100 mT) = 9.55 km/s.

It's definitely higher than 5km/s, but I would not go up to 9.5km/s. My napkin math was - booster provides 2km/s and the LEO is 7.2km/s, so if the ship arrives nearly empty LEO it must provide a minimum 5.2km/s. I did not account for gravity + friction losses which could be guesstimated to 2km/s. If we assume that ship handles all of them (which is not true booster does some) we come up to 5 + 2 = 7.2km/s.

The equation you have uses block 1 prop + dry mass numbers. Block 1 does not have significant payload capability and the last one just flow on OFT 6. As such let's use Block 2 numbers, first one should fly early next year. Dry mass target 100t, payload 100t, prop mass 1,500t. So (3.72 km/s) * ln(1700 / 200) = 7.96 km /s. This is with using min advertised payload of 100t and Elon's forward looking 100t number for dry mass. Way above 5km/s, but reaching 9.5km/s with 100t of payload will be challenging.

I've pulled stats from https://en.wikipedia.org/wiki/SpaceX_Starship .

You're mixing up unrelated concepts. In particular, the payload capacity to LEO of 100 tonnes has nothing to do with this calculation.

Europa Clipper isn't a 100 tonne payload; it's about 6.

Speed is not necessarily the correct lever it might affect. A probe with much larger antenna, with much larger or multiple radioactive power sources, with much more solar panels, with much more onboard fuel, etc could be much more useful with the same speed.

More mass means more fuel means more speed?

More fuel may mean longer duration missions rather than getting to the destination faster.

Or the same duration mission with a more massive payload that can both do more and report more.