Starship heat shield

[Vultur]

Why use shingles (many separate parts) rather than have the cooling channels in the ship's hull or a one piece heat shield?

I think the biggest issue is thermal expansion when the ship's skin changes temperature through a huge range.  You either need a material with almost zero coefficient of thermal expansion, or a sufficient way to accommodate the changes in size/shape as things heat and cool, or active cooling, in which case cooling channels may make sense someday. 

Yes, I meant in a scenario where transpiration cooling (cooling channels) are used. In this case, the skin shouldn't be changing temperature through a huge range.

Another issue is vibration.  A single shield or one with large segments may react badly to launch vibrations if made of a brittle material. 

Sure, but I meant transpiration cooling using tubes integrated into the hull itself. The hull already has to survive launch vibrations. The tubes can be steel too, a transpiration cooling system hopefully can avoid brittle materials entirely.

I agree that a one piece ceramic heat shield wouldn't be a good idea. I was suggesting that a possible additional benefit to going to transpiration cooling wiuld be a much lower part count/greater simplicity.

[InterestedEngineer]

So if the regen gasses ARE being used in the test stand to pressurize the tanks (which would  match "test how you fly")- how do they clean water and CO2 out of the LOX tanks between testing?

[Slarty1080]

So if the regen gasses ARE being used in the test stand to pressurize the tanks (which would  match "test how you fly")- how do they clean water and CO2 out of the LOX tanks between testing?

And how annoying to have to dump any excess LOX after every test fire due to ice contamination. But probably for a different thread...

[SpaceLizard]

So if the regen gasses ARE being used in the test stand to pressurize the tanks (which would  match "test how you fly")- how do they clean water and CO2 out of the LOX tanks between testing?

And how annoying to have to dump any excess LOX after every test fire due to ice contamination. But probably for a different thread...

I think you two may have wandered into the wrong thread by accident,
Raptor Engine testing etc. is that-a-way -> https://forum.nasaspaceflight.com/index.php?topic=53555.0

[OTV Booster]

Major configuration changes would probably be reserved for a Starship successor vehicle if Starship in its current form factor can't be made as rapidly/cheaply reusable as desired.

Eh, I think in earlier days the name was more fluid (BFR / MCT / ITS), but with the current name well established I think it would just be considered "Starship version XYZ."


The nice thing about dramatically increasing the radius of curvature is that you can reenter Mars (and Earth) at a higher speed using the same heat shield tech.

Possibly, but this seems like a much larger change than just between versions.

If they continue to have heat shield problems, I'd expect them to go back to transpiration cooling rather than try a wildly different diameter & form factor vehicle which would need different ground equipment (chopsticks etc).

A different shaped (far larger diameter) vehicle would IMO be part of a fairly dramatic change.

Yes, I always thought that the tiles are just an interim solution to get something together that somehow works with limited new tech, even if it's not really conductive to quick and cheap reuse. You absolutely need to get the beasts to launch (and get them back) first anyway.

Transpiration cooling with methane or even water certainly is doable, it's just a heavy engineering effort you don't want to push while you're still solving other things and don't even have a final design for the ship. Once you get the things finalized and flying at cadence there's plenty room to experiment and do practical R&D while earning money with them.

I mean, thin perforated 3D-stamped stainless steel shingles spot-welded to the tanks bleeding methane or some quilted silica mats with water injected into them during reentry for transpiration cooling, keeping the plasma away and blocking IR radiation onto the hull are surely a kind of rocket science and would require lots of work but it would help a lot with not getting your craft especially hot at all in the first place. And this is what you actually want for limiting wear and tear and for longevity with a minimum of maintenance.

Anything that actually is able to survive extremely high temperatures while insulating the structure from them is a false start for that, you're really painting yourself into a material science corner here if you also want it to be durable with little or no refurbishment.

Anyway, right now it's the tiles and I have little doubt that they can make them kinda work for the interim.

It's not just a question of 'what works'. It's also a question of 'what is the mass penalty for any particular solution?'


Transpirational cooling was a hot topic (didn't mean to do that, but I like it) a few years back. Among the competing opinions was that tiles weigh less than the cooling fluid. Hmmm. I'm thinking. Always a bad sign.


For the ship at least, take the hit on CO2 & H2O tank contamination and find a way to recycle it for reentry protection. Not a stand alone solution but a piece of the puzzle.


Anyway, back to reality. Maybe the requirement is stupid and the question need to be reformulated as 'how much loss of mass to LEO is worth how much gain in turnaround time?'  As an arm wavy example, is only 75t to LEO worth reducing a 30 day turnaround to a two day turnaround?


Edit to add: according to M. Jagger and K. Richards "You can't always get what you want..."

[OTV Booster]

Major configuration changes would probably be reserved for a Starship successor vehicle if Starship in its current form factor can't be made as rapidly/cheaply reusable as desired.

Eh, I think in earlier days the name was more fluid (BFR / MCT / ITS), but with the current name well established I think it would just be considered "Starship version XYZ."


The nice thing about dramatically increasing the radius of curvature is that you can reenter Mars (and Earth) at a higher speed using the same heat shield tech.

Possibly, but this seems like a much larger change than just between versions.

If they continue to have heat shield problems, I'd expect them to go back to transpiration cooling rather than try a wildly different diameter & form factor vehicle which would need different ground equipment (chopsticks etc).

A different shaped (far larger diameter) vehicle would IMO be part of a fairly dramatic change.

Yes, I always thought that the tiles are just an interim solution to get something together that somehow works with limited new tech, even if it's not really conductive to quick and cheap reuse. You absolutely need to get the beasts to launch (and get them back) first anyway.

Transpiration cooling with methane or even water certainly is doable, it's just a heavy engineering effort you don't want to push while you're still solving other things and don't even have a final design for the ship. Once you get the things finalized and flying at cadence there's plenty room to experiment and do practical R&D while earning money with them.

I mean, thin perforated 3D-stamped stainless steel shingles spot-welded to the tanks bleeding methane or some quilted silica mats with water injected into them during reentry for transpiration cooling, keeping the plasma away and blocking IR radiation onto the hull are surely a kind of rocket science and would require lots of work but it would help a lot with not getting your craft especially hot at all in the first place. And this is what you actually want for limiting wear and tear and for longevity with a minimum of maintenance.

Anything that actually is able to survive extremely high temperatures while insulating the structure from them is a false start for that, you're really painting yourself into a material science corner here if you also want it to be durable with little or no refurbishment.

Anyway, right now it's the tiles and I have little doubt that they can make them kinda work for the interim.

Yeah, I agree. I don't think short term Starship really needs rapid reuse.

The early (say before 2030) goals are Starlink v3 (maybe with data center hardware added on?), Artemis HLS (Artemis III and maybe IV), and hopefully Mars cargo in the 28/29 launch window (mayyybeee a TMI/cruise test in the end of 2026 window?)

Starlink is apparently profitable even with F9 (no upper stage reuse) so it's worth doing regardless of the state of the heat shield development.

As long as they can build enough tankers to fully fuel an HLS without reusing any of them during the fueling campaign, rapid reuse isn't needed for early Artemis missions (III and IV).

An early cargo Mars ship presumably needs fewer tankers than a Moon one since it can use the atmosphere to slow down, and doesn't need a shorter time higher delta v trajectory to limit crew zero g time.

So they have time to work on a more optimized heat shield. I wouldn't be surprised if transpiration cooling ends up being the solution again.

About this part though:

I mean, thin perforated 3D-stamped stainless steel shingles spot-welded to the tanks bleeding methane or some quilted silica mats with water injected into them during reentry for transpiration cooling, keeping the plasma away and blocking IR radiation onto the hull are surely a kind of rocket science and would require lots of work but it would help a lot with not getting your craft especially hot at all in the first place. And this is what you actually want for limiting wear and tear and for longevity with a minimum of maintenance.

Why use shingles (many separate parts) rather than have the cooling channels in the ship's hull or a one piece heat shield?

Getting an optimized flow pattern in skin channels sounds like it would call for a 3D printed fractal pattern something like leaf veins. If on the outside surface it would be susceptible to damage. If on the inside surface it gets hard to weld in stringers and hoops. Maybe in conjunction with tiles in the problem areas, 3D fractal or not.


A hybrid approach makes sense. We're already getting a feel for the areas that need help. Treating them as individual but related problems would play directly into SpaceX development strengths. Try different approaches in different areas. Rough them in for now then build the better ideas into version IV. Remove, fine tune or combine as necessary and it's version V.


It might be a couple of years before we see a ship that can be reused after EDL let alone one that can be routinely reused. De Nada. As you say, there is no show stopper if reuse takes awhile.[/quote]

[Vultur]

Transpirational cooling was a hot topic (didn't mean to do that, but I like it) a few years back. Among the competing opinions was that tiles weigh less than the cooling fluid. Hmmm. I'm thinking. Always a bad sign.

Yeah I think that's why they went to tiles.

But transpiration might be safer + faster turnaround even if more mass.

[Action]

Transpirational cooling was a hot topic (didn't mean to do that, but I like it) a few years back. Among the competing opinions was that tiles weigh less than the cooling fluid. Hmmm. I'm thinking. Always a bad sign.

Yeah I think that's why they went to tiles.

But transpiration might be safer + faster turnaround even if more mass.

I believe Musk has subsequently said that the tile solution came in much heavier than expected, so the mass tradeoff was no longer as clear as they had originally thought.

Starship's long, slow, low-temperature high-total-heat-load reentry is better suited to solutions that don't expend a coolant, at least at first glance.  If tiles can be made to work, they seem like they'd have advantages.  The trouble is, we're a bunch of experiments into this tile adventure and it's still not obvious this can be made to work in a way that supports rapid, reliable reuse.

I've said before in this thread that, if it was me, I'd have three separate teams working on three diverse solutions: tiles; the simplest most-likely-to-work transpiration scheme I could invent; and an almost-guaranteed-to-work, but expensive, ablative.  Waste anything but time.  This is your biggest risk.

[spacenut]

Or redesign orbiter to work like Stoke Space upper stage.  Large size that Superheavy can launch.  Like 12m diameter giant capsule. 

I thought by now they would have the Starship figured out.  The current design might not be what works best. 

If using ablative material, would it have to be re-applied after every orbit? 

Also, would trade off's in costs make using an expendable upper stage more cost effective.  I do know they will have to land on Mars, but it's atmosphere is less dense and gravity lower.  So what they have will probably work there. 

[Vultur]

Also, would trade off's in costs make using an expendable upper stage more cost effective.  I do know they will have to land on Mars, but it's atmosphere is less dense and gravity lower.  So what they have will probably work there.

Well I think some early Ships, like HLS for the first two Artemis missions, will probably be essentially expendable. Early Mars ships would be "expendable" in the sense of never returning to Earth (though probably "reused" in terms of being converted to Mars structures) but would still need TPS and landing legs.

I doubt a general expendable upper stage would work for their desired extremely low costs though. Larger size and cheaper materials (stainless steel instead of Al-Li) would probably give a significantly lower cost/kg than F9, but not enough for their goals.

[Action]

If using ablative material, would it have to be re-applied after every orbit? 

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  You build a factory line making these things and maybe you refurbish them, or maybe you just use a new one every time.  You get full reuse of the ship except for this heat shield which you need to replace. 

Another advantage of this scheme is that you can have a lighter LEO version and a heavier high-speed version suitable for reentries from deep space.

Besides the fact that this would be more expensive, you also have the downside that each ablative is being used for the first time - it can't really be tested the way a reusable heat shield can.  That drives you to doing all sorts of expensive non-destructive testing.

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

[Sohl]

...

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  ...

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

I like this "Big Ablative Panels" (BAP) approach.  If I were Musk, I would have pushed BAP as gen 1 heat shield, reusable tiles as gen 2, and transpiration or other active cooling as gen 3.  Maybe BAP has a show-stopper flaw for the Starship application, but it seems the lowest risk and most pragmatic way to get something usable and buy time for a F&RR approach.

[xvel]

...

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  ...

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

I like this "Big Ablative Panels" (BAP) approach.  If I were Musk, I would have pushed BAP as gen 1 heat shield, reusable tiles as gen 2, and transpiration or other active cooling as gen 3.  Maybe BAP has a show-stopper flaw for the Starship application, but it seems the lowest risk and most pragmatic way to get something usable and buy time for a F&RR approach.

Easily replacable ablative heat shield was a idea I had in mind, it may be fine if reusable heatshield will not work, but SpaceX is in such position that they don't really have to play it safe and waste time, they are and will dominate launch industry for many years even without starship, so they will try to do the "impossible" straight away even if there is high risk.

[OTV Booster]

If using ablative material, would it have to be re-applied after every orbit? 

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  You build a factory line making these things and maybe you refurbish them, or maybe you just use a new one every time.  You get full reuse of the ship except for this heat shield which you need to replace. 

Another advantage of this scheme is that you can have a lighter LEO version and a heavier high-speed version suitable for reentries from deep space.

Besides the fact that this would be more expensive, you also have the downside that each ablative is being used for the first time - it can't really be tested the way a reusable heat shield can.  That drives you to doing all sorts of expensive non-destructive testing.

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

First time around everything's a kludge - unless it's a cost plus contract.

[uhuznaa]

...

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  ...

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

I like this "Big Ablative Panels" (BAP) approach.  If I were Musk, I would have pushed BAP as gen 1 heat shield, reusable tiles as gen 2, and transpiration or other active cooling as gen 3.  Maybe BAP has a show-stopper flaw for the Starship application, but it seems the lowest risk and most pragmatic way to get something usable and buy time for a F&RR approach.

Ablative heat shields are a solid matrix with some filler that evaporates and the generated gas keeps the plasma away to avoid conductive heat transfer while cooling itself at the same time AND blocks IR radiation to the hull.

This works great, but is not reusable.

You can make it reusable by making the matrix fixed while making the evaporating component refillable from tanks. Transpiration cooling basically is the same as an ablative heat shield, just that the ablative part comes from tanks (methane, hydrogen or water) you can fill up again after a mission.

[Action]

...

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  ...

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

I like this "Big Ablative Panels" (BAP) approach.  If I were Musk, I would have pushed BAP as gen 1 heat shield, reusable tiles as gen 2, and transpiration or other active cooling as gen 3.  Maybe BAP has a show-stopper flaw for the Starship application, but it seems the lowest risk and most pragmatic way to get something usable and buy time for a F&RR approach.

Ablative heat shields are a solid matrix with some filler that evaporates and the generated gas keeps the plasma away to avoid conductive heat transfer while cooling itself at the same time AND blocks IR radiation to the hull.

This works great, but is not reusable.

You can make it reusable by making the matrix fixed while making the evaporating component refillable from tanks. Transpiration cooling basically is the same as an ablative heat shield, just that the ablative part comes from tanks (methane, hydrogen or water) you can fill up again after a mission.

The quote from my post was trimmed, but to be clear, I am only proposing ablatives as a stopgap.

If (and I still think this is an "if") it looks like tiles aren't going to work for reliable, rapid reuse, transpiration cooling of some sort is a clear alternative.  But it will take time to develop.

Ablatives decouple that problem from the rest of the work of getting Starship into service.  You want Starship in service ASAP for the lunar program, for Starlink, and more generally to thoroughly debug the rest of its systems.  If you can do that with a bandaid-solution ablative, that's not a crazy idea. 

[InterestedEngineer]

...

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  ...

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

I like this "Big Ablative Panels" (BAP) approach.  If I were Musk, I would have pushed BAP as gen 1 heat shield, reusable tiles as gen 2, and transpiration or other active cooling as gen 3.  Maybe BAP has a show-stopper flaw for the Starship application, but it seems the lowest risk and most pragmatic way to get something usable and buy time for a F&RR approach.

Ablative heat shields are a solid matrix with some filler that evaporates and the generated gas keeps the plasma away to avoid conductive heat transfer while cooling itself at the same time AND blocks IR radiation to the hull.

This works great, but is not reusable.

You can make it reusable by making the matrix fixed while making the evaporating component refillable from tanks. Transpiration cooling basically is the same as an ablative heat shield, just that the ablative part comes from tanks (methane, hydrogen or water) you can fill up again after a mission.

The quote from my post was trimmed, but to be clear, I am only proposing ablatives as a stopgap.

If (and I still think this is an "if") it looks like tiles aren't going to work for reliable, rapid reuse, transpiration cooling of some sort is a clear alternative.  But it will take time to develop.

Ablatives decouple that problem from the rest of the work of getting Starship into service.  You want Starship in service ASAP for the lunar program, for Starlink, and more generally to thoroughly debug the rest of its systems.  If you can do that with a bandaid-solution ablative, that's not a crazy idea.

Turns out SpaceX sorta agrees with you, since they have an ablative layer underneath the tiles.

One with  characteristic color signature so they can optically inspect the tiles to determine if any ablation occurred.

[Nescio Erucis]

I know it's pie in the sky, but it would be amazing if rocket engineers could someday develop the re-entry equivalent of regenerative braking in the electric car industry.

The rocket has so much kinetic energy, purchased at such a high cost, and the current logic is to try to waste it all by turning it into heat. It's like wearing away your (ablative) disc brakes instead of using that energy for something.

I don't really have any ideas (although the transpiration discussion got me thinking about using the 'veins' to superheat some 'propellant' that could then be used to propulsively decrease velocity), but I think we may get there some day. There's just too much energy sitting there to not use it for something.

[philw1776]

...

While it's obviously less than optimal from a cost and safety perspective, I'm pretty sure a bolt-on ablative would work.

You design the Starship with fittings such that big ablative panels can be attached, and every flight you take the used ones off and replace them with fresh ones.  ...

But anyway, it would work.  It would be quick.  And it would get you through the lunar services contracts.  You can debug and optimize a better solution at your leisure - it takes heat shields out of the critical path.  At the cost of being a kludge.

I like this "Big Ablative Panels" (BAP) approach.  If I were Musk, I would have pushed BAP as gen 1 heat shield, reusable tiles as gen 2, and transpiration or other active cooling as gen 3.  Maybe BAP has a show-stopper flaw for the Starship application, but it seems the lowest risk and most pragmatic way to get something usable and buy time for a F&RR approach.

Easily replacable ablative heat shield was a idea I had in mind, it may be fine if reusable heatshield will not work, but SpaceX is in such position that they don't really have to play it safe and waste time, they are and will dominate launch industry for many years even without starship, so they will try to do the "impossible" straight away even if there is high risk.

I have long favored a "snap on" initial/temporary heat shield approach that is not reusable but enables the 2nd stage to be fairly rapidly re-flown as the used components are detached and new ones bolted on.  Starship cost is in the engines, plumbing and airframe and these get re-used as heat shield optimal designs get flight proven.

Dfferent areas of the operational, but developmental heat shield, portion of the Starship may each use different purpose built techniques. Over many flights real world flight data will validate the best heat shield designs for appropriate areas of the spacecraft. Low mass, low refurb time and cost goals can be approached over time.
Fly operationally as you develop was a very successful SpaceX Falcon 9 approach.

[JaimeZX]

There's an awful lot of hand-wringing over the current tile solution, given they haven't actually retrieved SS yet to closely inspect and debug a once-flown heat shield yet.  Sure they have lots of sensors and telemetry, but once they get their hands on a used SS I think they'll be able to make much faster progress.