Hacker News Reader

Comment by bpodgursky | Hacker News Reader

The beauty of the Starship design is that if it works, it makes landing and taking off anywhere we want to go incredibly easy. You don't need a special lander module, complex docking manouvers, to replace disposable heat-shields, or anything. You land the Starship, and launch it again -- as is, no moving parts.

Earth's atmosphere and gravity are much heavier and stronger than either the moon, Mars, or even Mercury. If you have the regulatory controls to re-enter and land on earth safely, it's trivial to do it on any of those destinations.

It's still an incredibly hard challenge, and not might not work. But if SpaceX successfully reaches orbit and lands on earth with the Starship, landing on the moon and coming back is going to be an absolute cakewalk.

View on HN

It's the revenge of 1950's pop sci fi; gleaming (stainless steel), tail landing rockets thundering around the solar system. A fully fueled Starship can get anywhere in the solar system fast. Refueling in orbit is the key.

the cool part is, eventhough it needs refueling in orbit it is still cheaper by an order of magnitude than all the other options... some guy on youtube calculated an upper estimate of the cost to about 2000$ per KG - if i remember correctly.

Imagine dozens of consecutive unmanned launches boosted into orbit by the same 2-5 Boosters each with only a payload of fuel meant for the large tank of Starship.

I think the issue is less about building such a rocket system, and more about actually testing and certifying it to NASA's modern standards for carrying humans.

See the constantly stretching timeline for developing the Crew version of the Dragon capsule, despite the fact that SpaceX already had a working and tested rocket and pressurize capsule before they started on the crew version.

I'm optimistic that SpaceX can make some rapid progress here, but I wouldn't bet on the timelines holding for any of these human spaceflight projects.

Mars is hard. Enough atmosphere to burn you, but not to easily slow you down. Landing on the Earth and the Moon are both easier (as I understand it).

It depends - any atmosphere is a huge boon as long as your spacecraft can do aerobraking in any form. You can use it to change your orbit, capture into orbit from an interplanetary trajectory or even change the plane of your orbit as it was envisioned for some mission profiles of the Space shuttle or Dyna-Soar.

There are nice tables at Project RHO illustration how much delta-v you can save by aerobreaking when traveling between Solar System bodies:

http://www.projectrho.com/public_html/rocket/appmissiontable...

You loose all of that for an airless body like the Moon. In this case it's not that big of an issue as moon is close & has a low gravity. If Mars was airless, it could be much more of an issue, as you are coming much more quickly from an interplanetary trajectory & there would be more gravity to handle during landing than on the Moon.

If you have enough delta-v to land on a hypothetical Mars without an atmosphere you've got enough to slow down for a gentle atmospheric entry that wouldn't burn you up.

In terms of technical design creating a lightweight craft that can aerobrake to land on Mars is harder than the heavyweight craft that could do an airless powered landing. But once you factor in the booster requirements suddenly the atmosphere is making your job easier overall.

It's kind of a blessing in disguise that we live in the heaviest terrestrial planet in the solar system. It's like learning how to climb by starting with the most difficult mountain. You don't need as much energy to escape the gravity of the other planets and moons, and therefore can take less outlay of fuel with you for the return trip.

>> You don't need a special lander module, complex docking manouvers, to replace disposable heat-shields, or anything. You land the Starship, and launch it again -- as is, no moving parts.

Can anyone comment on how Falcon 9 knows where it is? Is GPS part of the game? Because they dont have that on the moon or Mars.

Don't know about Falcon but I know with Dragon one mechanism is viewing the star field.

On the Moon there's no atmosphere or erosion, so pattern recognition vs a known Lunar surface should work. Same should work to some extent on Mars.

IIRC the recent Chinese landers used LIDAR and feature matching during the terminal phase.

As for spacecraft navigation in general you are using star/sun/earth trackers (depending of where your spacecraft operates) to detect your orientation. This is important to correctly point solar panels to the sun, radiators, cameras and sensitive instruments away from the sun and directional cameras to Earth.

As for detailed absolute position in space, AFAIK this is usually done by ultra sensitive radio signal ranging back on Earth.

But other means could be used, eq. taking picture of your target or other bodies you can resolve and then tracking their magnitude/size and position, possibly over time, to calculate where exactly you are in the Solar System so that you are seeing what you are seeing.

I don't know the full story, but I know that for detecting the ground during landing they use radar (which was why they made sure the two falcon heavy cores didn't land at the same time).

Creating a gps (like) network on mars with a bunch of satellites might actually be an option before landing. SpaceX appears to be already developing that capability in the starlink satellites for the DOD [0].

[0] https://www.c4isrnet.com/battlefield-tech/space/2020/04/06/t...

The thing with a GPS network is that you need fixed ground stations receiving the signal to know the satellites' precise orbital parameters, and thereby know exactly where they are. So you'd need to land at least one Starship to get a "gps" network working.

I mean... wouldn't any method of determining the orbital parameters work? Stellar tracking, determining position based on known ground reference points via cameras, etc.

It's not like these are dumb birds, they can sit there for awhile collecting data to figure out their orbit, and then communicate that to the lander before it attempts to land.

Yes, you can determine it via other methods, but the accuracy won't be very high.

Reply on news.ycombinator.com