Starship heat shield

[DanClemmensen]

Sorry, that indeed should have been 5/year. That could seem generous in early development, but now with whole stage 1s and 2s built it would seem quite restrictive.

Could they have just one left that they are holding in reserve to try and make orbit before the year ends? Not quite a SS to Mars, but a good way to close out 2022. In which case they'd want every possible failure mode closed off (AFAP) before using it.

Which universe are you from where 5 - 0 = 1?  There hasn't been a single orbital launch of Starship, or even attempt at orbital launch, so how could there be only one left out of five?

The 5/yr is a count of SH launches, not orbital launches. If they launched an SH 100 meters in the air with no SS on top to test a chopstick catch, it would still count as an SH launch. I suspect this is the reason they will only do full-up orbital launches of SH. The distinction is academic unless they want to start hopping SH out to a floating platform, in which case they will probably try for a change to the rule.

[john smith 19]

The 5/yr is a count of SH launches, not orbital launches. If they launched an SH 100 meters in the air with no SS on top to test a chopstick catch, it would still count as an SH launch. I suspect this is the reason they will only do full-up orbital launches of SH. The distinction is academic unless they want to start hopping SH out to a floating platform, in which case they will probably try for a change to the rule.

That is what I thought.

So has anyone been keeping count which ones would could toward the limit?

[DanClemmensen]

The 5/yr is a count of SH launches, not orbital launches. If they launched an SH 100 meters in the air with no SS on top to test a chopstick catch, it would still count as an SH launch. I suspect this is the reason they will only do full-up orbital launches of SH. The distinction is academic unless they want to start hopping SH out to a floating platform, in which case they will probably try for a change to the rule.

That is what I thought.

So has anyone been keeping count which ones would could toward the limit?

There have been no SH launches, so the count is zero. It's not a launch unless it leaves the pad.

In addition to 5/yr SH, They are also allowed to do 12/yr SS launches that do no include an SH. There has not been an SS launch this year. A full stack launch is an SH launch, not an SS launch.

[john smith 19]

There have been no SH launches, so the count is zero. It's not a launch unless it leaves the pad.

In addition to 5/yr SH, They are also allowed to do 12/yr SS launches that do no include an SH. There has not been an SS launch this year. A full stack launch is an SH launch, not an SS launch.

Thanks for that. 

It clearly means that SX is not limited by having run out of takeoff opportunities at the site. They could (in principle) stack an SS and SH tomorrow and take a shot at orbit.

It looks as if the TPS is proving more difficult to perfect than expected.

[Robotbeat]

The TPS isn’t stopping them from doing an orbital attempt. They don’t need to recover Starship intact for the first several dozen launches.

[DanClemmensen]

There have been no SH launches, so the count is zero. It's not a launch unless it leaves the pad.

In addition to 5/yr SH, They are also allowed to do 12/yr SS launches that do no include an SH. There has not been an SS launch this year. A full stack launch is an SH launch, not an SS launch.

Thanks for that. 

It clearly means that SX is not limited by having run out of takeoff opportunities at the site. They could (in principle) stack an SS and SH tomorrow and take a shot at orbit.

It looks as if the TPS is proving more difficult to perfect than expected.

Sorry, not tomorrow. They need both the FCC temporary frequency licenses and the actual FAA launch license. Both of these are routine and they don't take long, but they do result in public notices that we all get to see, just like at KSC. More than a day but less than a week(?)

[Vultur]

I think the actual pacing item is prep of the booster itself ("robustness upgrades" for B7) + static fires, nothing to do with heat shield.

[john smith 19]

Sorry, not tomorrow. They need both the FCC temporary frequency licenses and the actual FAA launch license. Both of these are routine and they don't take long, but they do result in public notices that we all get to see, just like at KSC. More than a day but less than a week(?)

And thanks for this as well. That gives a future timeline. I'm guessing those timescales would lengthen once we start getting closer to the Christmas period.

I guess the obvious question remains "Is the tile material, or how they are mounted?"

[john smith 19]

The TPS isn’t stopping them from doing an orbital attempt. They don’t need to recover Starship intact for the first several dozen launches.

Stage 2 reusability is the key distinguising feature of SS over F9. Without it you just have a very much larger, very much more complicated F9 architecture.

With it everything changes. All future SS programmes become possible. Provided it can demonstrate adequte levels of reliability and reasonable replacement costs.

[john smith 19]

I think the actual pacing item is prep of the booster itself ("robustness upgrades" for B7) + static fires, nothing to do with heat shield.

This seems doubtful. 

Having gone through so many iterations of Raptor it should be possible to deliver enough of a standard version that can run for what, about 10 mins total? That would get them early data on TPS under actual operating conditions.

It's a cliche but once you get to LEO you are "Halfway to anywhere." Payload can always be increased but if you can't survive reentry

TPS is looking like the long pole in the tent.

[edzieba]

I think the actual pacing item is prep of the booster itself ("robustness upgrades" for B7) + static fires, nothing to do with heat shield.

This seems doubtful. 

Having gone through so many iterations of Raptor it should be possible to deliver enough of a standard version that can run for what, about 10 mins total? That would get them early data on TPS under actual operating conditions.

It's a cliche but once you get to LEO you are "Halfway to anywhere." Payload can always be increased but if you can't survive reentry

TPS is looking like the long pole in the tent.

If you can't survive re-entry then SpaceX "merely" have a F9-style partially reusable vehicle capable of super-heavy-lift at a lower cost/kg than F9. Starship recovery and re-use are desirable in the long term for further cost reduction and for improved schedule (not needing to build an additional 6 engines between launches), but a partially reusable Starship + Super Heavy stack is still a viable system for payloads of sufficient mass and value.

I doubt SpaceX would deliberately remove TPS or flaps for a test launch, as any post-mission recovery testing is essentially 'free', but if the choice was not flying or flying a vehicle with TPS known to be sub-par, they'd still fly and get what data they can get for the phases of EDL it survives rather than waiting for a perfected vehicle.

[rsdavis9]

I think the actual pacing item is prep of the booster itself ("robustness upgrades" for B7) + static fires, nothing to do with heat shield.

This seems doubtful. 

Having gone through so many iterations of Raptor it should be possible to deliver enough of a standard version that can run for what, about 10 mins total? That would get them early data on TPS under actual operating conditions.

It's a cliche but once you get to LEO you are "Halfway to anywhere." Payload can always be increased but if you can't survive reentry

TPS is looking like the long pole in the tent.

If you can't survive re-entry then SpaceX "merely" have a F9-style partially reusable vehicle capable of super-heavy-lift at a lower cost/kg than F9. Starship recovery and re-use are desirable in the long term for further cost reduction and for improved schedule (not needing to build an additional 6 engines between launches), but a partially reusable Starship + Super Heavy stack is still a viable system for payloads of sufficient mass and value.

I doubt SpaceX would deliberately remove TPS or flaps for a test launch, as any post-mission recovery testing is essentially 'free', but if the choice was not flying or flying a vehicle with TPS known to be sub-par, they'd still fly and get what data they can get for the phases of EDL it survives rather than waiting for a perfected vehicle.

two comments:
1. Maybe they are leaving tps (and did I just hear copv's off of booster?) off so they can ship to Florida?
2. If SS reentry never works they have all the above advances over f9 and a much faster launch cadence then f9 because of no droneships.

edit: added reentry

[john smith 19]

Over the years there have been many attempts at improving flying efficiency by delaying the point on the chord of a wing where airflow goes turbulent.

This has created specifications for how smooth a surface (any surface) has to be in order to retain laminar flow.

This document, from the British "Aeronautical Research Council Reports & Memoranda" gives some figures for what has to be done. Which also implies that there will be trouble if they cannot be met, or maintained. 

Obviously this work was done from the PoV of subsonic vehicles. Above M1 it's unlikely that the limits would be less stringent, and quite plausible they would be more so.

[Robotbeat]

Laminar flow over wings has nothing to do with Starship. That only matters if you’re trying to get good lift to drag ratio, which makes no difference to Starship as the Elonerons basically act as big drag brakes.

Why the frowny face? LOL no concern trolling here…

[OTV Booster]

I think the actual pacing item is prep of the booster itself ("robustness upgrades" for B7) + static fires, nothing to do with heat shield.

This seems doubtful. 

Having gone through so many iterations of Raptor it should be possible to deliver enough of a standard version that can run for what, about 10 mins total? That would get them early data on TPS under actual operating conditions.

It's a cliche but once you get to LEO you are "Halfway to anywhere." Payload can always be increased but if you can't survive reentry

TPS is looking like the long pole in the tent.

If we unwind a year or so, NSF consensus (and Musk statements?) had TPS as the long pole. No surprises here.

[OTV Booster]

I think the actual pacing item is prep of the booster itself ("robustness upgrades" for B7) + static fires, nothing to do with heat shield.

This seems doubtful. 

Having gone through so many iterations of Raptor it should be possible to deliver enough of a standard version that can run for what, about 10 mins total? That would get them early data on TPS under actual operating conditions.

It's a cliche but once you get to LEO you are "Halfway to anywhere." Payload can always be increased but if you can't survive reentry

TPS is looking like the long pole in the tent.

If you can't survive re-entry then SpaceX "merely" have a F9-style partially reusable vehicle capable of super-heavy-lift at a lower cost/kg than F9. Starship recovery and re-use are desirable in the long term for further cost reduction and for improved schedule (not needing to build an additional 6 engines between launches), but a partially reusable Starship + Super Heavy stack is still a viable system for payloads of sufficient mass and value.

I doubt SpaceX would deliberately remove TPS or flaps for a test launch, as any post-mission recovery testing is essentially 'free', but if the choice was not flying or flying a vehicle with TPS known to be sub-par, they'd still fly and get what data they can get for the phases of EDL it survives rather than waiting for a perfected vehicle.

It's unclear if you're addressing the general case or the TPS naked builds we're seeing. If the latter, there's discussion in the StarLink thread that suggests that there might be a push to get some sats up via SS. If this is correct, some quick n dirty launches sans TPS might be in the works. There are good arguments both ways.

[Robotbeat]

If the current Starships in the pipeline are not intended to be the operational design anyway, there may not be a lot lost by making them expendable.

[Action]

Laminar flow over wings has nothing to do with Starship.

Not sure if this is what john smith 19 was getting at, but heat transfer goes way up when the flow goes turbulent, which is why surface roughness is an issue for reentry.  Smooth is good.

This is a known problem with fragile-tile systems like Shuttle and was studied extensively.  However it is also suspected to be a problem with transpiration cooled systems because the points where coolant is injected into the boundary layer may introduce turbulence.

Either way, I think this is just a "you really need more heat shield than you would if you assumed laminar flow" and not a fatal problem with careful engineering.

[john smith 19]

Laminar flow over wings has nothing to do with Starship. That only matters if you’re trying to get good lift to drag ratio, which makes no difference to Starship as the Elonerons basically act as big drag brakes.

Why the frowny face? LOL no concern trolling here…

Given the organization that did the work deals with aircraft and helicopters they normally talk in terms of wings, but in truth it's an issue with every part of an airborne vehicle.

The difference in heat transfer between laminar flow and turbulent flow can be be in the range 4x-10x. Or have you forgotten this?

Of course you can just assume worst case heat transfer on all tiles and just make them 10x the thickness that a laminar flow assumption gives you, but given the amount of skin that has to be wrapped with them that's going to start having an effect on payload, given the 1:1 struture to payload trade for stage 2's. I'm not sure how much peoples ability to identify where the laminar-to-turbulent point is has improved since Shuttle was designed. For a symentrical body it should be easier, but the more bumps, humps and protruberances you add the tougher it gets to model. It's not just the gross changes, it's all the little corners. For example, one of the worst hot spots on the shuttle was the corner area between the OMS pods and the rest of the rear engine section.

Turbulent flow also makes it more likely you'll get one of those tiles fluttering. Those forces can be quite high and  if one tile is ripped off you've now got a hole for flow to stagnate in. How serious this is depends on wheather it's virtually impossible, and needs no mitigation, or it's viewed as possible and something is done about it.

I'll give one last data point. The Piaggio P.180 is an aluminum alloy turboprop business aircraft. Laminar flow was a design goal. It's mfg said the wing manages to maintain laminar flow over 20% of the fuselage and double the chord of other aircarft of its size. They also said its skin is accurate to 0.010" which is part of how it achieves that flow. This is technology was available (at that scale) in the mid 1980's.

I thought the report was quite relevant, given this threads title.

[john smith 19]

and not a fatal problem with careful engineering.

That would be a fair description of the Shuttles safety record regarding its heat shield.