Jump to content

SpaceX Discussion Thread


Recommended Posts

1 hour ago, YNM said:

Well they definitely weren't making airlines back then...

The Boeing Model 40 first flew in 1925. While originally built as a mail plane, in 1927 they started selling the Model 40A, which had seating in an closed cabin for two passengers. (The pilot sat in an open cockpit. Many pilots of the day felt like this was necessary, because the feel of the wind was part of how they flew.)

Edited by mikegarrison
Link to post
Share on other sites

 

Related to Starship and passenger safety. I will happily find a source for crow, and try to at least prepare it decently before I devour it should SpaceX fly people on SS at airline levels of safety, but I fully expect to expire before that happens.

Not saying SS won't work.

Not saying they won't fly people on it.

Just saying airline level safety is an astoundingly high bar.

Link to post
Share on other sites
56 minutes ago, tater said:

Just saying airline level safety is an astoundingly high bar.

Quibbling over safety and lack of guarantees of success, in a matter such as this, is insofar as my opinion, heresy.

And before you say I'm too harsh, consider that your ancestors, in matters such as these, took far greater risk, for a far lower payout, with little but a feeling of grim determination.

Link to post
Share on other sites
15 minutes ago, Nothalogh said:

Quibbling over safety and lack of guarantees of success, in a matter such as this, is insofar as my opinion, heresy.

And before you say I'm too harsh, consider that your ancestors, in matters such as these, took far greater risk, for a far lower payout, with little but a feeling of grim determination.

While it is worthwhile to engage on endeavours of great risk to further the reach of humanity, and that risk is definitely worthwhile for the purpose of space exploration, Starship is being proposed by SpaceX as a transportation system to compete with airlines. If they want to become a relatively popular means of mass transportation, the risk will not be balanced out by the payout of reaching amongst the stars, it will only be balanced by the payout of getting from point A to point B. That's where this airliner-level reliability idea comes from.

But a rocket is inherently less safe than an airplane, in the same way an airplane is inherently less safe than a car. (This is, of course, ignoring the safety attributed to the pilot, which is what makes cars more dangerous than planes in reality.) If your engine goes out while trying to park a car, you can put on the brakes, get out, and walk out alright. If your engine goes out while trying to fly a plane, you can glide down, hope there's a suitable landing area, land, and possibly walk out alright. If your engine goes out while trying to land a rocket, you die.

You can survive a lot of things going wrong in a car that you can't survive going wrong in a plane, and you can survive a lot of things going wrong in a plane that you can't survive in a rocket. Airliner-level reliability for a rocket would be incredibly difficult to achieve because they need to overcome the fact that the situation is more precarious with their system, but SpaceX needs to make it happen in order to for Starship to perform the functions they want it to.

Edited by RyanRising
Link to post
Share on other sites
1 minute ago, Nothalogh said:

Quibbling over safety and lack of guarantees of success, in a matter such as this, is insofar as my opinion, heresy.

And before you say I'm too harsh, consider that your ancestors, in matters such as these, took far greater risk, for a far lower payout, with little but a feeling of grim determination.

?

Space travel is dangerous, and will be for a while. A subset of people are willing to take that risk. Current commercial crew requirement is a 1:270 LOC.

This will certainly improve, and I think reusable rockets, with higher cadence can do that.

A 100X increase in safety to LEO would be 1:27,000 flights is a LOC incident. That would be fantastic for a space program with 2 human flights per year, and even for a space program at 2 per week that's awesome.

It needs to be one every few million departures to be "airline level" safety, so it needs to be 10,000X safer than crew Dragon. That's a high %[email protected]#in bar—and SpaceX talks about Point to Point. Mass use of rockets for P2P, or even mass flights of people into space requires a sense of safety that is substantially higher than 1:270.

Link to post
Share on other sites

Then we'll build orbiting cities :)

Build a 'stock' Starship.

Furnish its interior as a living space comparable to apartment block.

Launch it.

Launch modular docking adapter.

Start stacking Starships together, adding docking adapters in quantity and configuration as needed.

Add infrastructure. Hospital Starships. University Starships. Factory Starships. Warehouse Starships. Agriculture Starships. Power plant Starships.

Ideally, organize everything in a configuration that can rotate, providing artificial gravity.

Have fun with you Space Lego City! :D

 

Seriously though - i think it is not bad way to build a space station, that is both permanent and temporary. If every module is a fully functional Starship, you can add and remove modules on demand. That should solve many problems with maintenance experienced with stations we've built so far.

Module is malfunctioning due to wear and tear? Land it. Launch and dock another module performing the same duties. Problem solved quickly and efficiently.

Link to post
Share on other sites
On 2/6/2021 at 11:34 AM, tater said:

I assume the tiles are not light, either, though.

I’m pretty sure the tiles are VERY light. Like styrofoam and fiberglass light. 

On 2/6/2021 at 1:35 PM, grungar3x7 said:

Remember, the main reason for hex tiles in the first place, is to prevent hypersonic plasma being channeled in a straight line...

I'm not sure of a solution to the curvature problem, personally, but trapezoids and half hexes don't seem to be the answer, considering the reason for initial shape choice.

All the more reason to embrace the gaps. Trapezoids can work if you allow staggered positions. 

12 hours ago, tater said:

Soot I think.

Correct. Deposited primarily during the entry burn. 

Link to post
Share on other sites

This may be how they solve the curvature problem.

2010890.jpg

21 minutes ago, cubinator said:

Per the Inspiration4 raffle rules, Crew Dragon's astronaut height limit seems to be 6'6". This is three inches better than the Shuttle.

Guess I'm good then.

Link to post
Share on other sites
31 minutes ago, sevenperforce said:

This may be how they solve the curvature problem.

2010890.jpg

Guess I'm good then.

Yes but only on the tube part, not the nose cone where specially sized parts are need to close the pattern. 

Link to post
Share on other sites
16 minutes ago, RuBisCO said:

Yes but only on the tube part, not the nose cone where specially sized parts are need to close the pattern. 

The "tube" part can be covered entirely with equally-sized hex tiles. The smaller tiles account for the gaps to close the pattern around the cone.

Link to post
Share on other sites
Just now, sevenperforce said:

The "tube" part can be covered entirely with equally-sized hex tiles. The smaller tiles account for the gaps to close the pattern around the cone.

Oh I did not see that there was smaller ones there, neet

Link to post
Share on other sites
26 minutes ago, sevenperforce said:

This may be how they solve the curvature problem.

2010890.jpg

Guess I'm good then.

Don't see how smaller tiles solve it. You will get gaps and as you move up even the smaller tiles will start getting in each other way. 
The benefit of starship shape over the shuttle is that starship is an much simpler geometric. An  cylinder going over to an pointed dome at the front. You also have the joins and the raceway for the flaps.  Downside with the hexagonal is that they are not very good for an pointed dome.  Yes you can make them narrower at the top but then the next layer will not fit and you need an new type of tiles on each layer up. With square tiles in layers upward you don't care much if diameter become shorter you just use fewer tiles until you reach the nose who probably need other times as it curves so much. 

Link to post
Share on other sites
1 hour ago, cubinator said:

Per the Inspiration4 raffle rules, Crew Dragon's astronaut height limit seems to be 6'6". This is three inches better than the Shuttle.

Well, I'm out. 

 

 

Tanks are bigger than space ships, it seems 

14 minutes ago, magnemoe said:

Don't see how smaller tiles solve it. You will get gaps and as you move up even the smaller tiles will start getting in each other way. 
The benefit of starship shape over the shuttle is that starship is an much simpler geometric. An  cylinder going over to an pointed dome at the front. You also have the joins and the raceway for the flaps.  Downside with the hexagonal is that they are not very good for an pointed dome.  Yes you can make them narrower at the top but then the next layer will not fit and you need an new type of tiles on each layer up. With square tiles in layers upward you don't care much if diameter become shorter you just use fewer tiles until you reach the nose who probably need other times as it curves so much. 

Yeah - the fish scale tiles I referenced earlier look like an elegant solution - one that preserves a non-linear aspect 

Link to post
Share on other sites
2 minutes ago, JoeSchmuckatelli said:

Yeah - the fish scale tiles I referenced earlier look like an elegant solution - one that preserves a non-linear aspect 

Here's how trapezoidal tiles would look. It works quite well, really.

Spoiler

trap1.png
trap2.png
trap3.png

In the lower, purely conical section, I showed how the tiles can be slightly staggered to avoid any hypersonic flow path between tiles. I didn't stagger them in the properly curved/ogive section because it was waaaaay too complex, but the same thing is also possible there. 

If the gaps in the purely conical section are too severe, you can accomplish the same effect by alternately flipping the trapezoids like this:

Spoiler

trap-alternating.png

 

Link to post
Share on other sites
48 minutes ago, sevenperforce said:

In the lower, purely conical section, I showed how the tiles can be slightly staggered to avoid any hypersonic flow path between tiles. I didn't stagger them in the properly curved/ogive section because it was waaaaay too complex, but the same thing is also possible there. 

If the gaps in the purely conical section are too severe, you can accomplish the same effect by alternately flipping the trapezoids like this:

  Hide contents

trap-alternating.png

 

Now this is genial, Simply one piece, as you say you can offset to avoid vertical lines, you get horizontal one but not sure if this is an major problem, if it is its solvable with a bit more complex design and two more tile types. 

Link to post
Share on other sites
39 minutes ago, magnemoe said:

Now this is genial, Simply one piece, as you say you can offset to avoid vertical lines, you get horizontal one but not sure if this is an major problem, if it is its solvable with a bit more complex design and two more tile types. 

There's a horizontal line but it can be fixed by slightly offsetting the upper rings in the same way as the lower ones. It was just too complex a shape to do in Sketchup easily.

Link to post
Share on other sites
2 hours ago, magnemoe said:

Don't see how smaller tiles solve it. You will get gaps and as you move up even the smaller tiles will start getting in each other way. 
The benefit of starship shape over the shuttle is that starship is an much simpler geometric. An  cylinder going over to an pointed dome at the front. You also have the joins and the raceway for the flaps.  Downside with the hexagonal is that they are not very good for an pointed dome.  Yes you can make them narrower at the top but then the next layer will not fit and you need an new type of tiles on each layer up.

If you have two single differently-sized tiles, you can combine them to create dozens of shapes that tesselate around an ogive cone:

hex-tess.png

Obviously I didn't actually nose-over the curvature here but you can see the principle. 

Link to post
Share on other sites
17 hours ago, tater said:

It needs to be one every few million departures to be "airline level" safety, so it needs to be 10,000X safer than crew Dragon. That's a high %[email protected]#in bar—and SpaceX talks about Point to Point. Mass use of rockets for P2P, or even mass flights of people into space requires a sense of safety that is substantially higher than 1:270.

Refreshing to see this as I just finished explaining to someone on Youtube how airline failure rates are unlikely to be achieved any time soon given the nature of rocket engines.

Here are some excerpts for anyone who wants a more grounded and evidence-based approach to Starship and the current state of reliability in spaceflight:

Spoiler

"...for the simple case of when 1 engine failure is allowed, you would require a Raptor failure rate of 0.535% to achieve the 100 consecutive successful flights (99.01% success rate)"
"Alright, so at a 2% Raptor failure rate, we get a 89.25% success rate if allowing for 1 engine out event for the Super Heavy (28 x Raptor SL) and a 99.88% success rate for the Starship (3 x Raptor vac), multiplying them together gets us 89.15% for the whole launch. At this rate, you would expect around 11 MISSION failures in every 100 launches. Also, keep in mind this does not include the landing burns for both vehicles, but based on a quick analysis I did, this does not incur much of a penalty to the overall success rate, possibly only causing 1-2 extra failures per 100 flights." ~ the 2% Raptor reliability rate came from the other commenter, but is similar to current Merlin reliability, so I didn't change it.

"How Many Engines Should a Rocket Have?" by Geoffrey Landis is a great resource for anyone else wanting to do their own sums relating to engine reliability.

Link to post
Share on other sites

With 1 engine out allowed on ascent and landing, at 2% chance of failure per burn the Starship upper stage would have a 99.4% chance of successful ascent (6 failures in 1000) on 6 engines and a 99.9% chance of a successful landing on 3 engines. That's 7 failures in 1000 due to engines.

With 4 engines out allowed on ascent, the Superheavy booster, the Superheavy booster has a 99.98% chance of successful ascent. That's 2 failures in 10,000.

Loss of crew due to engines would therefore be 72 in 10,000 missions.

The booster landing on 4 engines would land successfully (up to 1 engine out) 99.77% of the time. ~25 lost boosters total ascent and landing per 10,000 flights.

 

If the engines are more reliable than 2% chance of failure then things improve a lot. Merlin 1D has flown 110 (990 engines) missions with 2 engine failures on ascent. Failures during static fire/landing are unknown and therefore excluded. That's 0.2%.

If Raptor can achieve Merlin level reliability of 2 failures in 990, then Superheavy might be expected to fail 24 in a million landings and suffer no failed ascents. The Starship upper stage would suffer critical engine related failures on 72 of a million flights.

 

That's not a very long way off where it needs to be TBH. In 1960 Boeing alone suffered over 40 accidents per million flights. Sure, engines are not everything that can go wrong with a rocket, but they are the most critical.

Link to post
Share on other sites

Here's the tesselation pattern to tile an ogive for two hex tiles of the same ratio as the ones on SN10:

hex-tesselation-good.png

You use the smaller tiles to slowly add gaps and then you build the larger tiles back in. The pattern repeats infinitely and gives you a flawless ogive.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...