Hacker News Reader

Comment by irjustin | Hacker News Reader

The videos were incredible. My favorite part was watching the booster flip in such clarity. Normally we don't get full view of it, let alone 4k.

View on HN

Scott Manley pointed out that it seemed to flip in the wrong direction, with one of the grid fins passing through the plume and inducing a roll. Will be interesting to hear more about that.

The videos are great!, but the rest of it is never going to work lol, just never. Even without a rethink about how to get heavy payloads to another planet this is still good entertainment.

why won't it ever work?

The hardest problem in the entire design had yet to be solved. Having a robust human rated tile system that can be rapidly turned around is a huge engineering challenge that kind of breaks the whole point of the design if it doesn't work. I wouldn't be surprised if they eventually give up and go back to a cheaper throw away second stage, or throw out the tile design completely and try for some evaporative cooling approach, again.

Bear in mind that a lot of what's happening to the tiles now is deliberate experiments to see how much weight they can shave off and how many failed tiles they can survive. Given that the vehicle is routinely surviving reentry at this point, it doesn't seem "hard" to make the tiles more robust by paying for it with added weight. The question is whether they'll have enough weight budget to pay for it? But at this point...probably? Not my area ofc.

"surviving reentry" and being reusable are two very different things, particularly if this is to become human rated.

Human rating is irrelevant to what they want to do, it is only a NASA rating - if NASA wants to ride they will have to come up with a rationale but private astronauts can fly on it.

From day one they said their vision was to replace airlines for international air travel.

This vision is impractical, bordering on ridiculous, but there it is in the S-1.

Sure and we’ll have self driving cars in 2015

I really hope to see the evaporative cooling make a comeback but it seems unnecessary when it’s returning to earth right now.

It would also be interesting to see them do shallower dips into the atmosphere then pull back out and repeat. Like a skipping stone. Lots of expansion contraction, but might work better without tiles.

They did that during the return of Ship on this test flight.

They’ve already shown they can replace all the tiles in a couple of days with removal and new install. No reason they couldn’t do even faster turn around with just re-install if that was needed.

And human rating is a NASA requirement they won’t have to worry about for a few years.

Well at least for one part of that, 90-something percent of the launches don't need to be human rated.

Even if it landed perfectly how is it going to be rapidly reusable with all those tiles breaking and needing repair? Then if that problem was magically engineered-away through some sort of materials science breakthrough, it still makes more sense to me to keep your big ships in a space staging area and your smaller ones as atmospheric gophers.

All what tiles breaking and needing repair? There was remarkably little visible damage this time around compared with previous flights.

There's no materials science breakthrough needed -- the shuttle used ceramic tiles successfully its entire service life. What's needed is engineering work, and that's what SpaceX has been doing.

You know a whole the size of a quarter can wreck the entire spacecraft and make it effectively throw away? Also, you'd want to use this many times. Making a system robust while not requiring months of refurbishment is really really hard.

The Space Shuttle had that problem because it was aluminum with a much lower melting point. It’s one of the reasons they’re using steel.

We’ve seen much larger holes than that in previous tests. Some of the control fins burned completely through.

For some of the tests, they removed a few tiles before launch, presumably to test that. Starship did fine.

coming back in one piece, and being good enough to use for 5 more missions are two very different things. For example, all existing reentry vehicles come back "fine" but they need to be completely remade to go up again.

I remember this being the same argument used against the Falcon 9 when it first stuck its landings.

"Oh, they'll need to do it 10x to be profitable!"

Now they do, as a matter of course.

Just because you defeat popular consensus of what is possible in one place doesn't mean you can continue to do it again and again. This stuff is hard.

Yes reusing falcon 9 was hard too.

Being a cynical naysay isn’t hard.

uncritically believing this design will succeed, despite having all the signs of Elon interfering and massively shifting requirements, isn't "hard" either

They already demonstrated that entire tiles can be removed without wrecking the spacecraft.

The quarter thing may have been true for the space shuttle, that doesn't make it true in general.

Deliberately testing its survivability with that failure mode over different parts of the vehicle has been one of the major foci throughout the entire test campaign, and it has proven remarkably resilient. That generalisation pretty much does not hold for starship.

I doubt that. It's made of stainless steel. If it gets home safely they can patch it.

Weren't the tiles one of the worst obstacles to quick turnaround times for the shuttle? It was something like 18 months before one could be launched again, and that's if they were in a hurry.

SpaceX has been specifically engineering both the tiles themselves (e.g. manufacturing) and the way that are used on the ship to be much more rapidly repairable than the Shuttle.

By the end they could turn a shuttle around in ten weeks.

Small ships are less efficient, especially leaving the gravity well. Thats the whole point

Could you tell me more? I suppose a heavy two-stage rocket is not optimized from the point of view of the rocket equation, but I know nothing about this field.

In short, the more stages the better to discard mass once it isnt necessary, and the larger to the better to improve the ratio of (ship+payload) to fuel.

Here is a decent summary.

https://gemini.google.com/share/121466b300c1

This is only true to an extent. Yes, a larger rocket means a better mass:payload ratio, but a larger rocket also means more mass in absolute terms, and more mass means more fuel, and more fuel means more mass, and more mass means more fuel, and more fuel means more mass, and so on. This is "the tyranny of the rocket equation", and it places an upper bound on the size of rockets that need to carry their own fuel for a given gravity well. And because the larger absolute mass of a larger rocket means more fuel, which means more cost, it relies on actually being able to find enough paying customers to fill out that payload capacity every single time. This is why, for example, despite the existence of jumbo jets (which have a better mass:payload ration than smaller planes), most passenger flights are not on jumbo jets, because there's just not enough demand on most routes.

No it doesn’t matter if the payload is full or not. If they succeed in full reuse, flying on a mostly empty Starship will be ten times cheaper than flying on an F9 and that means everything will switch eventually.

If a jumbo jet was ten times more efficient than a smaller plane, they would go everywhere. If a giant pickup truck got a 100MPG, why would you take a 30MPG economy sedan anywhere?

> This is why, for example, despite the existence of jumbo jets (which have a better mass:payload ration than smaller planes), most passenger flights are not on jumbo jets, because there's just not enough demand on most routes.

Airlines used to use a hub and spoke model where it would make sense to have larger planes between hubs and smaller ones to get to and from the hubs, but consumers strongly preferred direct routing, so it didn't work out. For orbital payloads, most payloads probably do not mind too much if it takes a month or more to boost/deboost themselves to their intended orbits.

You fundamentally misunderstand the implications of the rocket equation. It does not put and upper limit on rocket size for a given gravity. Smaller rockets would not be successful where larger fail.

Due to strucrually efficiency, larger ones suceede where smaller fail due to higher m0/mf ratio

They’ve already demonstrated they can replace all the tiles in a couple of days - even if they continue to have some fall off it won’t be an issue.

The design is wrong in a fundamental level. The rocket equation says it all. By adding so much extra weight for “reusability” (which they don’t release numbers on in terms of cost and means nothing when the rockets end up in the Indian Ocean) they nerf the performance.

Hence tens of launches to make the moon and zero payload to orbit in a dozen flights.

This forum is full of fanboys but their arguments don’t hold up to reality. Saturn 5 had men on the mock. This much time and this many rockets in. Starship has launched a dozen times for $15B - over a billion per launch - and it hasn’t made orbit.

Say what you want about SLS, it works.

Musk is a conman and Starship is a failure. Like the F-35 supposed libertarians and conservatives will happily throw money at it until it achieves its targets, after multiple redesigns, and then claim success as if the neigh-sayers weren’t right all along.

While we're downing this I'd like to add I think the "I've written X in Rust" posts are also cringe :D

Reply on news.ycombinator.com