Wiki Super Heavy/Starship - General Development Discussion

[JohnnyEnglish]

Interesting comments by Elon regarding Starlink 2 satellites - at 1.2 tonnes they are around 4x heavier than the current Starlink sats but will have an order of magnitude greater performance (somewhat vague but seemed to be in terms of bits throughput). So about 2.5x performance per kg.

Combine that with Starship's lower cost per kg to orbit (relative to Falcon 9) and the Starlink 2 service economics should be much improved.

If Starship can loft 100 tonnes to orbit that would be around 80 Starlink 2 sats, equivalent in performance to 800 Starlink 1 requiring around 16 Falcon 9 launches.

 

[Cosmacelf]

Interesting comments by Elon regarding Starlink 2 satellites - at 1.2 tonnes they are around 4x heavier than the current Starlink sats but will have an order of magnitude greater performance (somewhat vague but seemed to be in terms of bits throughput). So about 2.5x performance per kg.

Combine that with Starship's lower cost per kg to orbit (relative to Falcon 9) and the Starlink 2 service economics should be much improved.

If Starship can loft 100 tonnes to orbit that would be around 80 Starlink 2 sats, equivalent in performance to 800 Starlink 1 requiring around 16 Falcon 9 launches.

Yeah, I was kinda blown away by the 10x throughput figure. Knowing how Elon rolls, part of that 10x has got to be from utilizing satellite to satellite transport via lasers to route packets to less used base stations. Any one base station typically has a 100 Gbps Internet link. The ones in heavily congested areas are probably pretty loaded, while there are probably many that have lots of spare capacity, allowing Starlink to route packets to them.

It's one heck of a network!

 

[Grendal]

 

[ecarfan]

At about 11:40 in this video you can see a group of tiles falling off S24 during pressure testing (not a cryo test). They were attached to a side of the vehicle that is not visible on camera but about 20 of them can clearly be seen falling away.

 

[JRP3]

I guess from the expansion of the SS skin under pressure. That adhesive has to take a lot of abuse, huge temperature differentials and physical expansion and contraction of the SS.

 

[Cosmacelf]

At about 11:40 in this video you can see a group of tiles falling off S24 during pressure testing (not a cryo test). They were attached to a side of the vehicle that is not visible on camera but about 20 of them can clearly be seen falling away.

Yeah I’m surprised Tim Dodd didn’t ask this question of Elon. We’ve seen lots of tiles that seem to adhere properly.

 

[mongo]

I guess from the expansion of the SS skin under pressure. That adhesive has to take a lot of abuse, huge temperature differentials and physical expansion and contraction of the SS.

They later pulled a bent pipe out. May have been a shock load from mechanical failure .
Tiles are attached to three studs previously welded to the skin. Is the tile to stud connection adhesive vs mechanical? That would seem to be less thermally resistant and worse for automation/ replacement.

 

[JRP3]

I assumed adhesive because I thought a mechanical fastener would provide a heat pathway through the tile to the skin.

 

[JRP3]

Some pics appear to show visible fasteners on the tile face while others do not.

The lower image is older.

 

[mongo]

I assumed adhesive because I thought a mechanical fastener would provide a heat pathway through the tile to the skin.

Interesting point. The studs definitely penetrate the tile volume, so that results in a higher overall thermal conductivity at those points; however, total area of stud is quite small overall so average insulating value approaches a stud free setup.

Some pics appear to show visible fasteners on the tile face while others do not.

View attachment 809797
View attachment 809798

Lower pic is the first test tiles which look glued on. Upper pic is the current setup. Robot installs all the studs and I think they have another to do the tiles, but that part might still be manual.

 

[JRP3]

In this video they are just slapping them on, doesn't even seem to be any studs, maybe just a test piece?

https://twitter.com/x/status/1370853087410278407

 

[JRP3]

Another video showing push in place. Worker on the right seems to do something to the surface first. Can't tell if there are studs or not.

https://www.reddit.com/r/SpaceXLounge/comments/pli35a

 

[JRP3]

Interesting point. The studs definitely penetrate the tile volume, so that results in a higher overall thermal conductivity at those points; however, total area of stud is quite small overall so average insulating value approaches a stud free setup.

I was thinking of a stud with external bolt on the surface, but the newer tiles don't have that. The studs seem to be two types, a slightly conical one and then two split studs that might work like an auto trim fastener? Hard to believe those would have enough holding power on their own without an adhesive.

 

[mongo]

I was thinking of a stud with external bolt on the surface, but the newer tiles don't have that. The studs seem to be two types, a slightly conical one and then two split studs that might work like an auto trim fastener? Hard to believe those would have enough holding power on their own without an adhesive.

Yeah, studs do look ... unexpected. Maybe a split pin and 2nd person places the wedge? The slap into place attachment method seems like a locking/ one way mechanical setup, unless it's friction fit with a little glue (only on solid pin?) for permabond, but I would expect to see gloves it they were dealing with strong adhesives.

 

[JRP3]

I guess the highest force would be during re-entry in the belly flop position where pressure would be pushing the tiles onto the studs, at least near the middle. Closer to the edges forces would be trying to slide the tiles up the sides which the studs would resist.

 

[mongo]

I guess the highest force would be during re-entry in the belly flop position where pressure would be pushing the tiles onto the studs, at least near the middle. Closer to the edges forces would be trying to slide the tiles up the sides which the studs would resist.

Yeah, plus a tight connection may not be good if the stud expands more than the tile.

 

[ecarfan]

Part 2 of Tim’s latest interview with Elon. Frankly, disappointing. He does not seem to have many prepared questions. A lot of time is spent standing around saying “It’s just insane” while looking at the OLT. While watching I kept thinking of questions that I would like Elon to answer.

 

[EVCollies]

In this video they are just slapping them on, doesn't even seem to be any studs, maybe just a test piece?

https://twitter.com/x/status/1370853087410278407

That video is from 3/13/21. Probably about 100 iterations old...

 

[mongo]

That video is from 3/13/21. Probably about 100 iterations old...

That looks like the current setup, but the insulation blanket is hiding the studs.

 

[Cosmacelf]

Part 2 of Tim’s latest interview with Elon. Frankly, disappointing. He does not seem to have many prepared questions. A lot of time is spent standing around saying “It’s just insane” while looking at the OLT. While watching I kept thinking of questions that I would like Elon to answer.

I agree! At the very least, I want to know what Elon thinks of all these FCC delays. And whether it has held back Starship.