Starship specs - weight, volumes, etc

[msinclair]

Boosters are going to get lighter, by 10s of tons.  See these tweets:

https://forum.nasaspaceflight.com/index.php?topic=47352.msg2612747#msg2612747

Thanks for that.

So the engines are lighter, but all that shielding and fire suppression must have been…hefty.
3630kg (Version 1) – 1720kg (Version 3) = 1910kg/engine reduction. Comes to 63t for 33 engines. Seems like a lot, but it doesn’t make that much difference to payload.
Any idea if “vehicle-side commodities” include some portion of residuals? Or is that the restart stuff for the center 13?

I was reading about how the point of hot staging was to prevent ullage collapse in the booster, over on the SpaceX ‘Star series’ simulations thread – and I noticed your nice performance spreadsheet.
https://forum.nasaspaceflight.com/index.php?topic=47179.msg2501323#msg2501323
Care to speculate on Version 3 dry masses at all?

[FutureSpaceTourist]

https://twitter.com/elonmusk/status/1832554031845097590

Max payload to standard Earth reference orbit is actually ~180 tons for Starship when it is fully reusable and ~300 tons if expendable. Latter number is the apples-to-apples number comparing Starship to Saturn V.

>100 tons to the Starlink orbit is the operational spec minimum.

Starship v3 specs I assume.

[wannamoonbase]

https://twitter.com/elonmusk/status/1832554031845097590

Max payload to standard Earth reference orbit is actually ~180 tons for Starship when it is fully reusable and ~300 tons if expendable. Latter number is the apples-to-apples number comparing Starship to Saturn V.

>100 tons to the Starlink orbit is the operational spec minimum.

Starship v3 specs I assume.

What are people's opinion on what 'the Starlink Orbit' means? 

Is Starship going to deploy Starlinks into the near operational orbit instead of long raising periods?

[InterestedEngineer]

https://twitter.com/elonmusk/status/1832554031845097590

Max payload to standard Earth reference orbit is actually ~180 tons for Starship when it is fully reusable and ~300 tons if expendable. Latter number is the apples-to-apples number comparing Starship to Saturn V.

>100 tons to the Starlink orbit is the operational spec minimum.

Starship v3 specs I assume.

What are people's opinion on what 'the Starlink Orbit' means? 

Is Starship going to deploy Starlinks into the near operational orbit instead of long raising periods?

Might mean the dogleg they have to go out of Boca Chica to put Starlinks in the correct inclination w/o going over inhabited land.

a 1.5x penalty is about what that penalty is.

Now, if they start launching from Florida, OTOH, they don't have this overhead.

There's also the overhead of the starlink deploy system.  I doubt it's 50-80t.

I suspect the 180t is for the fuel version of Starship, which has no payload deployment overhead.

So, some mix of the above.

[InterestedEngineer]

Update based on latest NSF update video on Florida permitting, (relevant snapshot below)

https://youtube.com/watch?v=CWYryy9vZJo

Estimate SECO mass:  150_{starship+header_tanks} + 200_cargo = 350t
estimated MECO mass:  (2650 + 350) + (200 + 500)  = 3700
estimated launch mass:  (2650 + 350) + (200 + 4100) = 7300

At average Isp for Booster of 340, and 369 for Starship, that gives deltaVs for the launch phase of:
Booster:  340*9.81 * ln(7300/3700) = 2.27km/sec
Starship: 369 * 9.81 * ln(3000/350) = 7.78km/sec

Total deltaV = 10km/sec, which is in excess of the nominal 9.2km/sec for LEO.  The extra may be for going to different orbits, bad assumptions above, etc.

So 200t to LEO is what that model of starship seems perfectly capable of.  Also note the emphasis of far more deltaV on the upper stage than on the booster.  The booster's job is basically to get a Starship full of fuel to altitude and with a trajectory such that cosine/gravity losses are small.

TWR at liftoff will be 10,500/7300 = 1.44.

TWR Starship immediately after startup is 0.95.

[xvel]

I made calculations for current "real" starship v3 (B18 S39 onwards)

Terminal velocity of Starship on flight 6 was 85m/s
Terminal velocity of Superheavy was 350m/s
Velocity at staging was 1438m/s at 32.5deg angle so horizontal v was 1212m/s

Superheavy prop 3650t
Starship prop 1550t
Starship payload 200t

Assuming:

Superheavy dry mass 300t
Starship dry mass 150t
Sea level Raptor ISP 350s
Vacuum raptor ISP 370s

Prop for landing Starship 6t for 116m/s deltaV (85m/s terminal velocity + margin)
Prop for landing Superheavy 40t for 430m/s (350m/s terminal velocity + margin)
Prop for boostback burn 210t for 1650m/s (zeroing that 1212m/s of horizontal v + 438m/s in launch site direction)

Prop mass for first stage burn 3400t
Fullstack mass 5850t
Fullstack mass at staging 2450t

First stage deltaV 2987m/s

Starship mass at staging 1900t
Starship burn prop 1544t
Starship dry mass 356t

Starship deltaV 6077m/s

Full deltaV 9064m/s which is just about what you need for LEO

All burns except starship burn calculated with 350s ISP

and looks like that version is marginally capable of 200t to LEO, version you calculated is probably supposed to take more

[InterestedEngineer]

I made calculations for current "real" starship v3 (B18 S39 onwards)

Terminal velocity of Starship on flight 6 was 85m/s
Terminal velocity of Superheavy was 350m/s
Velocity at staging was 1438m/s at 32.5deg angle so horizontal v was 1212m/s

Superheavy prop 3650t
Starship prop 1550t
Starship payload 200t

Assuming:

Superheavy dry mass 300t
Starship dry mass 150t
Sea level Raptor ISP 350s
Vacuum raptor ISP 370s

Prop for landing Starship 6t for 116m/s deltaV (85m/s terminal velocity + margin)
Prop for landing Superheavy 40t for 430m/s (350m/s terminal velocity + margin)
Prop for boostback burn 210t for 1650m/s (zeroing that 1212m/s of horizontal v + 438m/s in launch site direction)

Prop mass for first stage burn 3400t
Fullstack mass 5850t
Fullstack mass at staging 2450t

First stage deltaV 2987m/s

Starship mass at staging 1900t
Starship burn prop 1544t
Starship dry mass 356t

Starship deltaV 6077m/s

Full deltaV 9064m/s which is just about what you need for LEO

All burns except starship burn calculated with 350s ISP

and looks like that version is marginally capable of 200t to LEO, version you calculated is probably supposed to take more

Your dry mass for Booster is too high (it's closer to 200t), but your fuel mass for boostback and landing (210+40) is too low,  but the difference washes out mostly.

Was it really only 1.2km/sec horizontal at MECO?

My guess is this version is closer to 150t than 200t.

[Apollo22]

Thanks so much for these numbers. Damn interesting.

[Robotbeat]

I really like having this thread as a way to keep track of these figures, as official figures are often vague, outdated, or sort of optimistically looking at future optimizations.

[rsdavis9]

It looks like spacex quotes v3 raptor with the sl thrust 280t and the vacuum isp 350s. Sort of makes sense. You want to know the max thrust so it actually goes up from the launch pad and you want to quote the best isp(at meco of higher) to highlight the fuel efficiency.
I get an isp of 332 at sealevel btw.

https://x.com/SpaceX/status/1819772716339339664

      Raptor 3 (sea level variant)
         Thrust: 280tf
         Specific impulse: 350s
         Engine mass: 1525kg
         Engine + vehicle-side commodities and hardware mass : 1720kg

[InterestedEngineer]

It looks like spacex quotes v3 raptor with the sl thrust 280t and the vacuum isp 350s. Sort of makes sense. You want to know the max thrust so it actually goes up from the launch pad and you want to quote the best isp(at meco of higher) to highlight the fuel efficiency.
I get an isp of 332 at sealevel btw.

https://x.com/SpaceX/status/1819772716339339664

      Raptor 3 (sea level variant)
         Thrust: 280tf
   Specific impulse: 350s
         Engine mass: 1525kg
         Engine + vehicle-side commodities and hardware mass : 1720kg

From the screenshot of their Florida application it's looking like 300t of thrust from Raptor 3:

103MN/9.81/35 engines = 300t.

I'd label it all as aspirational.

I get an isp of 332 at sealevel btw.

   Specific impulse: 350s

I use 340 as the average between the two, as the booster stage spends about half its time at about sea level and about half time in near vacuum.  You can see the crossover shortly after max-Q, when the exhaust starts to widen out

[RobLynn]

Curious; with estimated mass and performance numbers what would be potential payload mass benefit of droneship booster landings vs RTLS?

(I suspect that the burgeoning AI robotics revolution is going to change the cost calculus on things like RTLS with highly streamlined and reduced cost offshore and dock transfer/transport and even crane ops.)

[Robotbeat]

Curious; with estimated mass and performance numbers what would be potential payload mass benefit of droneship booster landings vs RTLS?

(I suspect that the burgeoning AI robotics revolution is going to change the cost calculus on things like RTLS with highly streamlined and reduced cost offshore and dock transfer/transport and even crane ops.)

I think it depends on if they stretch the booster or otherwise adjust the ratio of booster to ship mass. V3 starship and beyond make the ship almost as big as the booster, which is an optimization in part to reduce the booster staging velocity (and to maximize ship performance from Mars and the fully refueled ship delta-v).

For efficiency, you’d want to reverse that. Actually make the booster land even further downrange than it does for Falcon 9. But that’s a kind of longer term optimization.

[xvel]

Curious; with estimated mass and performance numbers what would be potential payload mass benefit of droneship booster landings vs RTLS?

(I suspect that the burgeoning AI robotics revolution is going to change the cost calculus on things like RTLS with highly streamlined and reduced cost offshore and dock transfer/transport and even crane ops.)

For V3 at least +30t more payload, maybe +50t, it's not that much because starship is optimized for RTLS (relatively big second stage) unlike falcon.
Keep in mind that all those calculations are VERY approximate, there is too much we don't know, the only thing they show for shure is that starship with reasonable payload capability is physically possible with technology spacex already has.

[InterestedEngineer]

Curious; with estimated mass and performance numbers what would be potential payload mass benefit of droneship booster landings vs RTLS?

(I suspect that the burgeoning AI robotics revolution is going to change the cost calculus on things like RTLS with highly streamlined and reduced cost offshore and dock transfer/transport and even crane ops.)

For V3 at least +30t more payload, maybe +50t, it's not that much because starship is optimized for RTLS (relatively big second stage) unlike falcon.
Keep in mind that all those calculations are VERY approximate, there is too much we don't know, the only thing they show for shure is that starship with reasonable payload capability is physically possible with technology spacex already has.

The burdened cost of droneship catch&return is about $2M, so you'd be paying 67$/kg as opposed to < $25/kg for the rest of the cargo.  Why would yay 2-3x as much?

Note fully burdened includes amortizing the cost of building a huge drone ship and all the return infrastructure, so the "But I have this one inseperable cargo that's 230t" won't apply.  Probably better throwing away an old Starship for that type of launch, it'll be cheaper.

[Robotbeat]

There are ways to improve fully burdened droneship recovery cost dramatically.

[xcelr8]

Anyone remember this message by Elon from a few months ago? https://x.com/elonmusk/status/1903481526794203189

He mentioned 100t to Starlink orbit (i.e. LEO) for V3 with reusability. The "new" V3 he presented at the recent Starship update meeting is likely what he was referring to here. The 200t payload capability is, I believe, referring to the future Starship version that would have significantly stretched propellant tanks. This is what had been presented at last year's update by Elon as V3.

This is no longer the case. Starship V3 now is not much larger than the currently flying V2.

The Starship specs given in the EIS for SLC-37 likely represent an envelope of the design, as was speculated on in the NSF article on the EIS. For instance, the number of Raptors on the booster was given as 35, while they were still shown to be 33 in Elon's update for both the next gen (V3) and future gen. The number of Raptors on the Ship is not increasing to 9 until sometime after V3.

In the recent update, Elon also clearly stated that Starship V3 consists of Booster V3, Ship V3, and Raptor 3. He then stated that this version would be flying for the first time at the end of 2025.

[Robotbeat]

He said ~100t to the Starlink orbit, which isn’t the standard reference orbit of 26deg and 185km, but instead like 53deg and 400km. Also, I think that’s with the deployer mass.

That’s a pretty big difference. Using Silverbird Astronautics’ launch vehicle calculator and their “Evolved” (v2, I think) starship, that’s the difference between 149t to orbit and 8t.

[InterestedEngineer]

He said ~100t to the Starlink orbit, which isn’t the standard reference orbit of 26deg and 185km, but instead like 53deg and 400km. Also, I think that’s with the deployer mass.

That’s a pretty big difference. Using Silverbird Astronautics’ launch vehicle calculator and their “Evolved” (v2, I think) starship, that’s the difference between 149t to orbit and 8t.

the probably put in a huge dog leg.  Now do Florida, where a majority of those launches will come from.  It's about the same for 27 and 45 inclinations

[Robotbeat]

Maybe, but Starship suffers from much of the same problems as Shuttle and for the same reason. Its performance rapidly decreases beyond the lowest LEO orbit because the dry mass of the upper stage is very high, because it’s a reentry vehicle (heavy heatshield, aerosurfaces/“wings”, etc) with an integrated payload bay.

So the difference in performance between different LEO orbits is much larger than you might think.

Stretching the booster and doing down range booster landing would significantly mitigate this problem, tho.