After Thursday's flight, Starship is the most revolutionary rocket ever built | Hacker News Reader

After Thursday's flight, Starship is the most revolutionary rocket ever built

arstechnica.com

47 pointsvanburen52 comments

> What Thursday's revelatory reentry footage promises is a world in which launch is cheap and abundant. No longer will we need to worry so much about mass or volume, which have been tyrannical overlords to mission planners since the inception of spaceflight nearly seven decades ago.

Using the metrics in the article, Starship costs come out to around $300 per kilogram launched.

This is an order of magnitude drop over existing launch costs, which are already lower than ever courtesy of the Falcon rockets.

(Wikipedia estimates Falcon costs are $3,000 per kg. Other providers still significantly higher.)

Any time you change such a fundamental pricing unit in a market, there is going to a huge amount of change in market opportunities.

People dismissing this have their heads in the sand.

Yes, and even $300 per kg on orbit seems rather conservative.

> SpaceX can likely build and launch a fully expendable version of Starship for about $100 million. Most of that money is in the booster, with its 33 engines. So once Super Heavy becomes reusable, you can probably cut manufacturing costs down to about $30 million per launch.

So, $30m per 100t is $300 kg. If Starship is reusable, the flight could cost closer to $10m. If the payload is 150t, the cost of kg is lower still. And all that is in relatively short term.

I used to be a Space X skeptic, pooh-poohing them for re-using RP-1 based rocket engine technology from the 1960s. Clearly, I’m a moron. Starting with known technology and scaling from there worked. Amazing.

They'd already done revolutions with this launch: https://www.youtube.com/watch?v=XXiB-Kgqvdk

The latest one was more successful than the previous tests partly due to not revolving.

What impact will this have on the design of future orbital vehicles? I can't imagine everyone needing to launch hundreds of tons into orbit, much in the way that people don't buy a lorry if they just want to pick up their kid from school.

When you send an envelope by Fedex, it usually flies in and out of Memphis that night. Somehow Fedex manages even though each plane has vastly more capacity than one envelope

Same will apply with Starship

SpaceX will be the biggest Starship customer for the foreseeable future (Starlink launches), so there’s already a critical mass of demand to support many flights

FedEx, huh. So what you're saying that SpaceX can now place my satellite into the wrong orbit using a gravity-boost flyby past Ganymede.

But FedEx wouldn't send a massive jet out if your goal is to mail that envelope from Brownsville to Boca Chica (of all places).

Yes, and?

Are you saying the whole planet by air looks like 40 km by road, so there's no point having other means of transport? Or something else?

a massive jet doesn’t need to pass through boca chica from brownsville, everything needs to pass through the atmosphere to get to space - it’s all on the way

When full reuse really gets going, it will be possible to launch the mass of a battleship into space for the cost of a single SLS launch.

You don't see this as having any effect on vehicle design? It's going to make the tradeoff between vehicle construction cost and vehicle launch cost very different. Spacecraft will become much more like ordinary tech than exotic aerospace devices, more like ships than aircraft.

You seem to be under the impression that mass is the main restriction - it isn't. Like at all.

Starship as a concept is nothing new - the Soviets developed the Energia in the late 1980s and the second version was supposed to be 100% reusable as well. After the downfall of the Soviet Union, Roskosmos couldn't find any customers for it, though, despite it being able to lift up to 100t to LEO and being super cheap.

The thing is that the larger (and heavier) the payload, the more expensive it usually is. No one is launching dead mass to orbit just for shits and giggles. The bigger the vehicle, the more (expensive!) tech is packed into it. To give you a hint of why Starship wouldn't change the landscape as much as many seem to think: compare the number of launches between Falcon 9 and Falcon Heavy: the launch cadence of F9 is bi-weekly and rising, whereas FH had 8 launches in 5 years or less than 2 per year.

The main purpose of Starship is to allow for Starlink to continue to exist and similar mega constellations to be launched. Single missions of 100t or more to LEO (Starship cannot go beyond that without refuelling) are usually multi-billion dollar flagship missions that simply don't launch very often.

You are assuming the cost/mass of a spacecraft is somehow a constant of nature. It's not, not at all. Reduce the cost of launch and there's enormous benefit of reducing the cost of the spacecraft, since that cost now dominates. Use cheaper (if heavier) materials, use higher margins to reduce design and testing costs. Use COTS technology and spend mass (shielding, pressure vessel) to make it work in space. Perhaps most importantly: launch large numbers of spacecraft with the same design to amortize design cost over more vehicles. Reduce design cost by hardware-rich development where problems are found by testing in space, not exhaustive preliminary analysis.

The problem is that the assumption holds true more than ever: https://www.marketsandmarkets.com/Market-Reports/small-satel...

Small sats are cheaper and a rapidly growing market. Technical question: what do you suppose a - say - 50t commercial payload would even be? Currently there's two options that I am aware of: space stations (where again launch cost is NOT the primary cost factor) and space telescopes.

Unless you can provide actual data and facts, I am not convinced at all this is going to change. Especially, since COTS hardware already IS used in the commercial spacecraft business. You can simply buy a standard bus and put your own hardware on it. Heck, you can even buy satellite components online: https://www.cubesatshop.com

You post a link about satellite construction at current launch costs. That has nothing to do with what I was talking about, which is what happens when launch costs fall to as little as $10/kg.

I want you to imagine what happens if two countries in conflict have cheap launch. Space assets are important, and it would become worthwhile to protect them. That means mass, the more the better. The market here would be almost unlimited. And that's just one possibility. Consider space manufacturing (like ZBLAN fibers), high power (multimegawatt) comsats, high resolution imaging from high orbit (mirrors tens of meters across). These are just off the cuff ideas; we could keep listing others. None of them have to be terribly convincing to rebut an argument there's obviously nothing to be done.

I feel like this is disingenuous. If launch costs truly go down to $300/kg - or even within an order of magnitude of that - you’ll see affluent hobbyists sending up microsatellites just because they can.

Hell, I’d be interested in doing it if I could put together a 2kg package for under $20k or so. Why not?

And $300/kg is the cost when they are still discarding the upper stage. The target for full reuse is more like $10/kg.

It would “only” cost $126 million to put the ISS in orbit at that price. At $300/kg, I could see a space hotel for the ultra rich (given the market for trips to the titanic). How much would a rich person pay to go stay at a space hotel for a week? Is $1 million out of the question? How many people paying that would you need to make it economically worth while? After 126 of them you’d pay for the cost to lift it to orbit, another 126 of them could pay for a lot of creating it and staffing, so we need 500 UHNW individuals to commit to taking a trip to space to have a space hotel be viable? I’m skipping some numbers here, but I’m sure there’s a pitch deck being circled around for that already.

Edit: oh someone’s already way ahead of me. They’re charging $5million for a the day trip.

https://www.businessinsider.com/voyager-station-first-ever-s....

> The main purpose of Starship is to allow for Starlink to continue to exist

Nope, the main purpose of Starship is to implement Musk's life goal: "to die on Mars, just not while landing".

Will that happen? Too early to tell.

But that's the goal.

If a price per kg into significantly drops, wouldn't this incentive to design heavier payloads at a lower cost? This could boost the affordability of science done in space or even make in space manufacturing a bit more closer.

It will also become much cheaper to put astronauts in space and support them there, so it will make sense to make vehicles maintainable and upgradeable in space, even beyond low earth orbit. For example, I expect we'll see space telescopes in high earth orbit that are more like terrestrial telescopes in that they could be maintained and upgraded rather than discarded after use.

How many reusable Starship launches have to be made before it exceeds the cost of one non-reusable rocket? Because last time I checked they’re going to have to make over a dozen additional launches to enable this one Artemis journey, assuming they ever get to that point.

IIRC, so far the total dev cost including three full test launches and 28 test vehicles has cost about the same as the price per launch of the SLS…

So it's still cheaper even without being reusable.

Starship's goal is to bring the cost per kg to space down to about $10, or about $4.50 per pound. If they succeed, and it's increasingly looking like they will, this will revolutionize space. The main reason we can't do anything fun in space right now is because costs are simply too high. The Space Shuttle cost about $54,500/kg to send something to orbit. [1] The Falcon 9 brought that way down to $2,720/kg. And the Falcon Heavy then brought it down even lower yet to around the $1400/kg mark. These are all near unbelievably reductions in costs, but the costs started out so astronomical that even at the point of $1400/kg you still just can't do anything too terribly interesting in space. It's just too much money. But at $10/kg? That's suddenly completely swinging open the gates to the mass commercialization and exploitation of space.

The ISS was a massive and global project largely because of the unbelievable costs involved in getting it up into space. But the ISS is actually relatively tiny, and only weighs about 450,000kg. We now may be able to send the combined mass of the ISS up into space for as little as $4.5 million (contrasted against $25 billion on the space shuttle)! This really feels surreal, like somehow I cannot even believe what I'm saying myself. But, to the best of my knowledge, it is all 100% accurate. And if so, we are on the verge of a revolution in human civilization that would be difficult to compare anything in our species' history. Cool time to be alive!

[1] - https://ntrs.nasa.gov/citations/20200001093

$10/kg is cheaper than a lot of long distance flights on an airliner. I weigh 90kg. $900 to go to orbit is the same cost as flying.

$10 is cost target. Price will likely be significantly higher.

> much in the way that people don't buy a lorry if they just want to pick up their kid from school

No, they buy multi-tons SUVs for that... and come up with all sorts of arguments to justify it.

I'm sure we'll find many reasons to put a lot of new things in orbit if it's cheap enough.

This website, of all means, talks surprisingly little about startups enabled with significantly reducing the price to get to orbit.

"Masses" sent to orbit roughly split into 3 groups: heavy simple materials - mostly metal structures and fuel, complex devices - mostly electronics, and humans. Significantly and closely human-related things are practically humans (can't have humans in space without those roughly in the same flight).

For the first group, the cost is mostly in delivering them to space, SpaceX mostly covers that. Second and third groups look quite promising. We already see several projects of orbital space stations; maybe more are in order? Also all kinds of in-space manufacturing - we already see profitable biological and high-tech electronics and communications products, but what about energy, satellite and deep space probes assembly and manufacturing, repairs of on-orbit things?

> I'm sure we'll find many reasons to put a lot of new things in orbit if it's cheap enough.

I'm sure we will - and hope we'll do it soon.

The ultimate limit on # of Starship launches may be from water pollution in the upper atmosphere.

Each Starship launches produces about 600 tons of water (either in the exhaust, or from oxidation of hydrogen in the exhaust). Let's ballpark this and say half is ejected in or above the stratosphere (any ejected above the atmosphere quickly returns to the atmosphere.)

The Feb. 2022 eruption of Hunga Tonga–Hunga Haʻapai injected an estimated 146 million tons of water into the stratosphere. So, under the ballpark assumption this would be equivalent to half a million Starship launches. Beyond that (if even that is acceptable, as the water stays up there for several years) materials in space might have to be obtained from space sources, not by launch from Earth.

EDIT: why was this downvoted?

4000 ish tons of propellant, half a million launches per year, would be 2 gigatons total propellant per year.

Until the Sabatier process plants are running at scale, that will also release a non-trivial fraction of current global CO2 emissions.

I got something wrong in the first post. I misread the propellant mass and was only looking at the second stage. So, multiply the water production by a factor of 4. Not half a million launches, but 1/4 of that. So the water constraint on launches is even worse than I was saying (this is likely what led to the downvote?)

Re: downvotes — I never ask when I get them, and I didn't vote either way.

(I think even asking that is frowned on?)

Well, I've never been dinged for it, but I'll stop if that's the case.

It would have been more useful to get a critique there; I could even have corrected the post before the edit window ended.

When is the next Starship launch?

Ship 29 is going through static fire testing but they're still building the booster. Two months maybe?

Musk says 6 more launches this year. That's considered optimistic but - and I'm not saying the progress so far hasn't been amazeballs - if you consider how far they need to go for Artemis and the theoretical Artemis time frame then 6 more this year might be what's what's required.

If you really wanted 6 this year it'd be more like 6 weeks for the next launch?

I was too optimistic in estimates for this flight :) and now I feel rather conservative. Third flight really pushed the progress forward - and also uncovered several areas for improvement.

Super Heavy return flight roll control and engine relight issues. What are those gases emitting from Starship after their engines are off on orbit? Orientation issues for Starship. Some thermal protection plates seem to be lost. The payload door open/close - was it successful enough? The cryogenic liquid transfer - Shotwell mentioned they're studying the telemetry to figure out how well it went, do we need improvements there as well? There are probably other issues as well.

I suspect FAA may have more that 17 - the number for the last launch - specific points to address for the next flight. Would be happy to be shown wrong.

recommend to watch Thunderf00t's live coverage https://www.youtube.com/watch?v=cfkadv8NHlw

Reply on news.ycombinator.com