SpaceX 'Star series' simulations

[OneSpeed]

This topic is intended to be along the lines of the  SpaceX Falcon Mission Simulations https://forum.nasaspaceflight.com/index.php?topic=42389.0 thread, but for the SpaceX Star series: StarHopper, Starship and Super Heavy.

First cab off the rank is a simulation of the StarHopper hopping to about 5kms altitude, hovering for a while, and landing back on the same pad. This simulation was first posted a week ago in L2, but since then many of the assumptions like 9mØ propellant tanks have become public knowledge, so I thought I'd post a minor update here. A week is like dog years at SpaceX! If the three Raptor engines provide 600t of thrust, then a GLOW of around 500t would provide a feasible T/W ratio of 1.2 at liftoff. Wet and dry mass fractions are a total guess at this stage, mine is 100t dry.

Because I've never simulated a hover for such an extended period, I've not noticed the drift that you will see in the video before. The rocket climbs vertically, and there is no wind in SpaceSim, so I assume the drift is actually caused by the rotation of the earth. Anyway, for that reason I needed to translate the rocket back towards the landing pad by about 200m on descent, and this was more difficult than you might assume. Pitching the rocket worked initially, but as the sink rate increased, so did the lift, in the opposite direction to the pitch. It 'felt' like descending in an aircraft with no tailplane, and could lead to tumble. So, since the hopper has no grid fins attached, I think it will need an RCS, perhaps pressure fed methalox vernier thrusters, if it is to descend with any velocity.

[jpo234]

Can your simulation figure out a rough estimate how high the hopper can actually hop? Could it break the Karman line and officially reach space similar to New Shepard?

[OneSpeed]

Can your simulation figure out a rough estimate how high the hopper can actually hop? Could it break the Karman line and officially reach space similar to New Shepard?

That's a good question. The sim assumes an available ΔV of 5.2 km/s. That's easily enough to reach the Karman line as long as you are happy to go supersonic, but maintaining control while descending might be another matter. If I simply set the throttle to 100% and burn to depletion, then theoretically the ship gets to about 960kms.

[philw1776]

960Kms amazing! Typo.

I did a simple spreadsheet assuming dry mass, propellant tonnage 400 tonnes limited to decent T/W and get ~5.2Km/sec.  Maybe (I doubt this) it could fly over the Karman line and use some remaining propellant to reduce velocity striking the atmosphere to retain control.  Beyond my ken.

EDIT: this crude hopper is not designed to go supersonic so the point is moot.  MaxQ would be "interesting"

Appreciate your excellent far better sims.

[envy887]

Can your simulation figure out a rough estimate how high the hopper can actually hop? Could it break the Karman line and officially reach space similar to New Shepard?

That's a good question. The sim assumes an available ΔV of 5.2 km/s. That's easily enough to reach the Karman line as long as you are happy to go supersonic, but maintaining control while descending might be another matter. If I simply set the throttle to 100% and burn to depletion, then theoretically the ship gets to about 960kms.

How high can it go while staying subsonic, both up and down? How about at a realistic dynamic pressure of perhaps a few psf? Ignoring the control issues of flying backward...

[OneSpeed]

How high can it go while staying subsonic, both up and down? How about at a realistic dynamic pressure of perhaps a few psf? Ignoring the control issues of flying backward...

A replacement nosecone is being constructed for StarHopper, and it looks more substantial than the first. Descending fins first with any velocity sounds like a recipe for disaster to me, but since you asked for it, here's a StarHopper sim limited to about Mach 0.75. Since there is no sign yet of vernier thrusters or grid fins, it also assumes the engines must be kept running in order to provide control authority. It could go quite a bit higher if it went supersonic at high altitude, and if the engines could be shut down and restarted.

[envy887]

Interesting... It looks like you're running all three engines the whole time? I'm assuming they will want to test inflight restarts. Maybe on just the outer engines, but possible on all three.

How high can it go if it accelerates to Mach 0.8 as quick as possible, holds that velocity for a while, shuts down all the engines and coasts to apogee (should be aerodynamicly stable thanks to the fins), restarts one engine at low throttle shortly after apogee to control the vehicle in free fall and avoid breaking Mach 0.8 on decent?

[edzieba]

this crude hopper is not designed to go supersonic so the point is moot.

Unless plans have changed (not impossible), the hopper is intended to go supersonic.

[docmordrid]

this crude hopper is not designed to go supersonic so the point is moot.

Unless plans have changed (not impossible), the hopper is intended to go supersonic.

That tweet could easily be about Starship Mk-1 - due this summer.

[dubya]

That tweet could easily be about Starship Mk-1 - due this summer.

Indeed. The definitive NO at the beginning of the sentence makes that conclusion pretty inescapable.

[edzieba]

The 'no' is in reference to the (now cancelled) S2 conversion mini-BFS vs. the now-actually-constructed BFS test article (AKA "BFR Test Ship"). That seems pretty definitive that the hopper will go supersonic, and I can see no reason why it could not do so. Its whole purpose is to replicate the supersonic-to-landing regime, with the follow-on Starship test vehicle then expanding that envelope to include the (sub)orbital through hypersonic re-entry regimes (replacing the previously proposed mini-BFS).

[Lar]

The 'no' is in reference to the (now cancelled) S2 conversion mini-BFS vs. the now-actually-constructed BFS test article (AKA "BFR Test Ship"). That seems pretty definitive that the hopper will go supersonic, and I can see no reason why it could not do so. Its whole purpose is to replicate the supersonic-to-landing regime, with the follow-on Starship test vehicle then expanding that envelope to include the (sub)orbital through hypersonic re-entry regimes (replacing the previously proposed mini-BFS).

I think that's an incorrect interpretation since the question asked was specific about hopper practicing the belly flop. His No seemed definitive to me.

[edzieba]

The 'no' is in reference to the (now cancelled) S2 conversion mini-BFS vs. the now-actually-constructed BFS test article (AKA "BFR Test Ship"). That seems pretty definitive that the hopper will go supersonic, and I can see no reason why it could not do so. Its whole purpose is to replicate the supersonic-to-landing regime, with the follow-on Starship test vehicle then expanding that envelope to include the (sub)orbital through hypersonic re-entry regimes (replacing the previously proposed mini-BFS).

I think that's an incorrect interpretation since the question asked was specific about hopper practicing the belly flop. His No seemed definitive to me.

The root question was on the modified S2 mini-BFS, not the hopper. Twitter is awful for linking with context, so a screenshot will have to do:

[OneSpeed]

This is a speculative simulation of a single stage Starship P2P flight. With 9 SL Raptors, and a full propellant load, the initial T/W is a healthy 1.6. So, throttle back for MaxQ occurs early, at the 36 second mark. If the ship were to continue to a purely ballistic trajectory, re-entry g forces would be prohibitive (~20gs). Instead, I've used negative pitch to flatten the trajectory, reducing the re-entry flight path angle. This allows the ship to skip like a stone on a pond, extending the range out to 10,000kms. The peak g force on the first 'bounce' is just over 4. If the Starship had larger (dragon?) wings, and hence a greater lift coefficient, the peak could be reduced further, and the range extended beyond 10,000kms.

[livingjw]

This is a speculative simulation of a single stage Starship P2P flight. With 9 SL Raptors, and a full propellant load, the initial T/W is a healthy 1.6. So, throttle back for MaxQ occurs early, at the 36 second mark. If the ship were to continue to a purely ballistic trajectory, re-entry g forces would be prohibitive (~20gs). Instead, I've used negative pitch to flatten the trajectory, reducing the re-entry flight path angle. This allows the ship to skip like a stone on a pond, extending the range out to 10,000kms. The peak g force on the first 'bounce' is just over 4. If the Starship had larger (dragon?) wings, and hence a greater lift coefficient, the peak could be reduced further, and the range extended beyond 10,000kms.

Did you hold AoA constant at 40 degrees? What was your L/D?

John

[envy887]

Did you hold AoA constant at 40 degrees? What was your L/D?

John

Lift and drag are both shown in the simulation. Hypersonic L/D looks like about 1.1.

[speedevil]

This reminds me of http://www.astronautix.com/s/saengerantipodalbomber.html - the description is somewhat fuzzy.
I have failed to find a source for "Concerning Rocket Propulsion for Long-Range Bombers" - and was wondering if the similarities might be more or less obvious with it.
Clearly the aimed for hypersonic L/D is much, much less with Starship.

[Keldor]

This is a speculative simulation of a single stage Starship P2P flight. With 9 SL Raptors, and a full propellant load, the initial T/W is a healthy 1.6. So, throttle back for MaxQ occurs early, at the 36 second mark. If the ship were to continue to a purely ballistic trajectory, re-entry g forces would be prohibitive (~20gs). Instead, I've used negative pitch to flatten the trajectory, reducing the re-entry flight path angle. This allows the ship to skip like a stone on a pond, extending the range out to 10,000kms. The peak g force on the first 'bounce' is just over 4. If the Starship had larger (dragon?) wings, and hence a greater lift coefficient, the peak could be reduced further, and the range extended beyond 10,000kms.

What would a flight with 6-9 SL raptors and 3 vac raptors look like?  Or what about dual bell, for allowing deep throttling at landing while allowing high thrust semi-vacuum operation the rest of the time, in the same manner as SSME?  I know Elon Musk said duel bell didn't make sense, but I think that was in the context of Starship sitting on top of Superheavy, where the engines would always be firing in either a very high altitude as a second stage where TWR is less important or else when landing at sea level with a very low thrust.  Single Stage suborbital changes everything.

Is converting an engine like raptor to use a toroidal aerospike likely to be feasible without a drastic overhaul?  Aerospikes have limited utility for normal two stage flight, but again, single stage changes everything.

[OneSpeed]

Did you hold AoA constant at 40 degrees? What was your L/D?

John

Actually I held pitch constant at 42 degrees, much the same as a spinning stone would do, so the AoA varies a little bit. I modulated the pitch between runs to see what gave the best range. This sim was the best result, with L/D at around 1.2. The best trade was to maximise lift regardless of drag, which kept the ship out of the sensible atmosphere for the longest periods between 'bounces'. On average, this minimised the effect of the drag, and gave the best range.

What would a flight with 6-9 SL raptors and 3 vac raptors look like?

Perhaps counterintuitively, not as good. With 9 Raptors and a 3g limit on acceleration, I'm either throttling back or shutting down engines from the 90 second mark, when the ship is already at 34kms altitude and just over 1km/s. More than 9 engines would just add mass for no benefit. Vacuum engines would have less thrust at low altitude, which is when maximum thrust is most useable.

Is converting an engine like raptor to use a toroidal aerospike likely to be feasible without a drastic overhaul?  Aerospikes have limited utility for normal two stage flight, but again, single stage changes everything.

Even if it was an easy conversion, I'm not convinced there would be much advantage. Although not everyone agrees with me on this, once you have sufficient engines symmetrically arranged, then above a certain altitude, plume interaction generates recirculative flow against the dance floor. This gives the effect of a single 'virtual' vacuum nozzle, not unlike the toroidal aerospike. That's why I'm suggesting an engine arrangement as per below. There would still be plenty of room for cargo bins, but not that many would be needed with the 10t payload this sim suggests.

[Ultrafamicom]

Did you hold AoA constant at 40 degrees? What was your L/D?

John

Lift and drag are both shown in the simulation. Hypersonic L/D looks like about 1.1.

Isn't that a bit too high? Even shuttle has only a Hypersonic L/D of less than 1.5

[Keldor]

Perhaps counterintuitively, not as good. With 9 Raptors and a 3g limit on acceleration, I'm either throttling back or shutting down engines from the 90 second mark, when the ship is already at 34kms altitude and just over 1km/s. More than 9 engines would just add mass for no benefit. Vacuum engines would have less thrust at low altitude, which is when maximum thrust is most useable.

I was picturing them shutting off the sea level engines once they reached a certain altitude and using just vacuum engines from there on.  Launch may have to be done with the vacuum engines initially off, which is why I suggested more SL engines, though Space Shuttle style semi-vacuum engines are a possibility to reduce the need for extra SL engines at liftoff.

6 SL + 3 semi-vac fireable at sealevel, shutting off the sea level engines over the course of the launch?

[OneSpeed]

Lift and drag are both shown in the simulation. Hypersonic L/D looks like about 1.1.

Isn't that a bit too high? Even shuttle has only a Hypersonic L/D of <1.5

For shuttle L/D varies with velocity, from about 1.8 at Mach 1 down to 1.1 at Mach 30. At the post SES-10 press conference Elon Musk said Falcon 9 had a lift-over-drag (ratio) of roughly one if flown at the right angle of attack.
I've calibrated SpaceSim as best I can against the SpaceX Earth EDL physics model shown in the Dear Moon presentation, and from that, L/D is about 1.2 at Mach 19, but your mileage may vary.

[hkultala]

Perhaps counterintuitively, not as good. With 9 Raptors and a 3g limit on acceleration, I'm either throttling back or shutting down engines from the 90 second mark, when the ship is already at 34kms altitude and just over 1km/s. More than 9 engines would just add mass for no benefit. Vacuum engines would have less thrust at low altitude, which is when maximum thrust is most useable.

I was picturing them shutting off the sea level engines once they reached a certain altitude and using just vacuum engines from there on.  Launch may have to be done with the vacuum engines initially off, which is why I suggested more SL engines, though Space Shuttle style semi-vacuum engines are a possibility to reduce the need for extra SL engines at liftoff.

6 SL + 3 semi-vac fireable at sealevel, shutting off the sea level engines over the course of the launch?

With what kind of engine layout?

Only two engines should be used for the final phase of the ascent to have both reasonable g-forces and good isp. But with 3-way symmetry, this would mean asymmetric thrust, requiring the craft to fly at an angle with engines gimballed quite a lot.

And having big nozzles that can gimbal a lot... then the base of the rocket easily runs out of space.

[speedevil]

Only two engines should be used for the final phase of the ascent to have both reasonable g-forces and good isp. But with 3-way symmetry, this would mean asymmetric thrust, requiring the craft to fly at an angle with engines gimballed quite a lot.

And having big nozzles that can gimbal a lot... then the base of the rocket easily runs out of space.

It would seem plausible to just go right to one engine operation after three, not two.
Start out with the engines canted out at 4 degrees as their launch position (mounted slightly inboard of where you would otherwise put them), and be able to gimbal to ~8 in one direction, and modestly less in others.

Assuming for the moment you are talking of upgraded non-throttleable 200 ton vacuum engines, and we want to avoid >3g, to make the worst case:
This means we have 600 tons thrust with all three lit, and a minimum total mass of 200 tons.
If we were then to go to two engines, highly gimballed, this takes the new total mass to 133 tons.

This is 1500m/s of flight under this regimen.

If we initially cant the vacuum engines out at 4 degrees during boost, they are 99.7% as effective as nominal. If we assume 4450m/s until the point we turn off one engine, that is a loss of ~2m/s or so due to reduced effective ISP.

The terminal phase of flight takes ~100s now, not ~50s, but I'm struggling to find a scenario in which this incurs gravity losses.

This implies that you can live with a sharply limited gimbal angle - ~5 degrees, not the ~9+ you might want if starting out at 0 gimbal, with essentially no penalty.

[Keldor]

Only two engines should be used for the final phase of the ascent to have both reasonable g-forces and good isp. But with 3-way symmetry, this would mean asymmetric thrust, requiring the craft to fly at an angle with engines gimballed quite a lot.

And having big nozzles that can gimbal a lot... then the base of the rocket easily runs out of space.

It would seem plausible to just go right to one engine operation after three, not two.
Start out with the engines canted out at 4 degrees as their launch position (mounted slightly inboard of where you would otherwise put them), and be able to gimbal to ~8 in one direction, and modestly less in others.

Assuming for the moment you are talking of upgraded non-throttleable 200 ton vacuum engines, and we want to avoid >3g, to make the worst case:
This means we have 600 tons thrust with all three lit, and a minimum total mass of 200 tons.
If we were then to go to two engines, highly gimballed, this takes the new total mass to 133 tons.

This is 1500m/s of flight under this regimen.

If we initially cant the vacuum engines out at 4 degrees during boost, they are 99.7% as effective as nominal. If we assume 4450m/s until the point we turn off one engine, that is a loss of ~2m/s or so due to reduced effective ISP.

The terminal phase of flight takes ~100s now, not ~50s, but I'm struggling to find a scenario in which this incurs gravity losses.

This implies that you can live with a sharply limited gimbal angle - ~5 degrees, not the ~9+ you might want if starting out at 0 gimbal, with essentially no penalty.

SpaceX has mentioned wanting to have ~1000 passengers on Starship.  This puts the payload into the 100 ton range by itself.  Add in the rest of the rocket, as well as the fuel needed for the landing burn, and Starship will almost certainly be 200 tons or more at the end of the vac engine burns.

Moreover, 200 tons is Raptor's peak performance.  Minimum throttle will probably be somewhere in the range of 130 tons.  This means that peak accelleration can be kept to ~2G's before they have to turn off engines, which seems perfectly reasonable.  They shouldn't have to shut down any engines.

Even if they did, the center of mass will be pretty high up when the tanks are nearly empty.  There are 100 tons sitting on the nose, after all!  My math estimates the deflection angle needed for single engine to be somewhere around 5.  Assuming the vac raptor has the gimbal 10 meters (it's likely less!) up from the end of the bell, 5 degrees means a bit less than 1 meter of deflection.  Since it will only ever have to rotate outward significant distances to compensate, there really is plenty of room for this.

For engine configuration, I would put 6 sea level engines in two rings, like a 3 pointed star.  3 vacuum engines in the spaces between the points.

In retrospect, Starship is going to need high TWR at liftoff, much higher than a standard two stage, which stretches the first stage as far as it can while the extra weight of tanks doesn't outweigh the extra fuel they provide.  But Starship P2P is single stage, so the extra tankage weight is much more significant.  9 liftoff engines seems reasonable.  This puts it at 9 SL, 3 Vac.  Let's see if I can whip up a decent layout.

[Keldor]

Thinking about it some more, engine out on the vac engines could be tricky to deal with.  Do you relight one of the sea level engines to compensate?  What's your margin on loosing ISP that way?  Should still be better than pure SL engines, right?

Anyway, the full expansion ratio 200 3 meter bells seem too large to fit and still have deep gimbaling, but it's pretty doable with 2.5m.

With 2-way symmetry, you're able to fit 2 3 meter vac engines with room for limited gimbal while having room for a pair of landing engines for engine out landing redundancy.  You can easily fit things like 2 Vac 10 SL.

[mikelepage]

This is a speculative simulation of a single stage Starship P2P flight. With 9 SL Raptors, and a full propellant load, the initial T/W is a healthy 1.6. So, throttle back for MaxQ occurs early, at the 36 second mark. If the ship were to continue to a purely ballistic trajectory, re-entry g forces would be prohibitive (~20gs). Instead, I've used negative pitch to flatten the trajectory, reducing the re-entry flight path angle. This allows the ship to skip like a stone on a pond, extending the range out to 10,000kms. The peak g force on the first 'bounce' is just over 4. If the Starship had larger (dragon?) wings, and hence a greater lift coefficient, the peak could be reduced further, and the range extended beyond 10,000kms.

Hi OneSpeed, great simulation! Reattaching your plot for reference.

With regard to optimising the trajectory, I have been wondering if the "skipping stone" analogy is the best one.  Going so deep into the atmosphere on the very first "skip" would surely have quite a dramatic (downward) effect on range?  Intuitively I wonder if performing a small burn just as Starship is passing through 80km altitude before that first skip - to zero out the vertical motion - would result in a much less "bouncy" profile, with Starship instead skimming like an air hockey puck across the top of the atmosphere, with (potentially?) increased range. 

I would love to see a simulation similar to what you've done, but with this re-entry "skim" burn in action. Obviously the real atmosphere wouldn't be so well behaved, but you could control that with real-time attitude adjustments/RCS burns the same way the F9 boosters use their grid fins/Cold gas thrusters.  Also not sure what the size of that re-entry "skim" burn would be, but perhaps it could be minimised by using negative pitch even more aggressively to get a lower apogee?  Gemini 3 was only a 161km x 225km orbit, so going to 260+km seems a little high.

[speedevil]

In retrospect, Starship is going to need high TWR at liftoff, much higher than a standard two stage, which stretches the first stage as far as it can while the extra weight of tanks doesn't outweigh the extra fuel they provide.  But Starship P2P is single stage, so the extra tankage weight is much more significant.  9 liftoff engines seems reasonable.  This puts it at 9 SL, 3 Vac.  Let's see if I can whip up a decent layout.

If the outer engines throttle well enough, 4 outer engines lit and hoverslam sort-of-works, one 200 ton thrust raptor in the middle fixed may almost work.
~4G peak, but if the passengers will put up with this, ...

Further on the more aggressive route.
Aft cargo compartment was stated at 88m^3.
If we fill this with six stubby raptors, and come off the pad at ~2g, not ~1.2 we can ~half gravity losses, or around 300m/s saving up till 300m/s. Some of this is lost by having to then throttle back to avoid the sound barrier, but it could in principle help considerably.

[Slarty1080]

Rules of engagement
Standing orders by squadron – attack bombers only
Don’t fire unless condition xyz - < 5 hexes and directly behind or advantaged
Out of ammo > head for home
>50% damage > head for home
> 75% damage bale out
Enemy directly behind 5 or less hexes > turn!
Fired at last turn > turn!

Stage 1 move to engagement from random position
Bombers continue
Escorts turn to interceptors
Interceptors turn to bombers
Continue until first burst fired or fired upon. Then switch to dogfight stage

Dogfight stage
Bombers continue fire at approaching enemy turn for home if damage > 50%
Escorts and interceptors check front aspect turn in behind nearest enemy
If no enemy visable within 7 hexes turn rnd left or right

draw cards turn left or turn right or straight on (a few)



Speed
Guns
Bursts
Range
Engines
Fuel

Turn rate
Hexes moved

Angle behind, advantaged, disadvanted, infront

Range 1-16
Options fire long or short range
Chase

Tailing in near range    fire
Tailing in far range  depending on strategy accelerate or fire
Tailing out of range   accelerate
Advantaged in near range   fire
Advantaged in far range         depending on strategy  manouvre or fire
Advantaged out of range   manoeuvre
Disadvantaged in close range of enemy
Disadvantaged in far range of enemy
Disadvantaged to out of range enemy
Tailed by enemy in close range  avoid!
Tailed by enemy in far range 
Tailed by enemy out of range

I would assume that SpaceX have been through all of the options for engine configurations many many times and will continue to review them as circumstances change (raptor performance and sl or vac, Starship mass, mission requirements and acceptable margins).

It wouldn't surprise me if we didn't see another new engine configuration at the next announcement, seems to be one of the things they change a lot. And I would not be surprised to see multiple versions with different engine configurations flying before long.

[OneSpeed]

I wondered if you had any estimates on which (if any) of those skipping events would generate sonic booms audible on the ground? But it seems to me it goes subsonic whilst still 25km altitude, which should minimise any complaints on the ground.  (Shuttle was still Mach 1.5 at 18km).

Answering your post here because the answer might have more general applicability. From:

https://www.nasa.gov/centers/armstrong/news/FactSheets/FS-016-DFRC.html

here are some more data points for supersonic aircraft and the Space Shuttle (STS):

Aircraft	Mach	Altitude(ft)	Altitude(m)	Pressure(lb/ft²)	Pressure(Pa)
SR-71 Blackbird	3.2	80,000	24,000	0.9	43
Concorde SST	2	52,000	16,000	1.94	93
F-104 Starfighter	1.93	48,000	15,000	0.8	38
Space Shuttle	1.5	60,000	18,000	1.25	60
XB-70	1.5	37,000	11,000	2.5	120

From 'NASA/CR–2011-217077 Measured Sonic Boom Signatures Above and Below the XB-70 Airplane Flying at Mach 1.5 and 37,000 Feet' by Maglieri and Henderson, the first image attached below indicates a range of pressure signatures generated by the XB-70 Valkyrie, and measured by an F-104 Starfighter at various distances, as well as at a static ground station. This is a great example of how complex shocks formed close to supersonic aircraft 'age' as they move through the atmosphere, and by the time they reach the ground, may approach the classic N shape.

This is fine for supersonic flight, but Starship P2P re-entry is likely to be much higher and faster than for any of the above.

From 'AIAA-89-1105 Review of Sonic Boom Theory' by K. J. Plotkin, who references:
1. Seebass. AR., "Hypersonic Boom". Boeing Scientific Research Laboratories Technical Communication 030. June 1970.
and
2. Tiegerman. B., "Sonic Booms of Drag-Dominated Hypersonic Vehicles". Ph.D. Thesis, Cornell University, August 1975.

"Currently, there is one analytic model for sonic boom at hypersonic speeds. It is based on a concept by Seebass that all hypersonic bodies have effectively blunt noses (both physical blunting, such as re-entry vehicles, and aerodynamically because of the entropy layer on slender vehicles), and the resultant drag dominates the sonic boom."

For hypersonic flight (Mach 5 and above), where:

Pg = ambient pressure at ground
ag = ambient sound speed at ground
D = vehicle drag
H = atmospheric scale height
h = flight altitude

the results for the N-wave overpressure Δp, positive phase impulse I, and duration T are:

Δp = 0.59 * ( Pg^5/8 * D^3/8 * exp(-h/8*H) / H^1/4 * h^1/2 )
I = 0.16 * ( Pg^1/4 * D^3/4 * exp(h/4H) / h^1/2 * ag )
T = 1.08 * ( Pg^-3/8 * D^3/8 * exp(3h/8H) * H^1/4 / ag )

"Calculations from this theory agreed well with available data from Apollo 15 and 16
re-entry. Flight Mach numbers were from 4.6 to 15.6. and the Apollo vehicle clearly matches
the postulated model."

From my simulation, the StarshipP2P ricochets are at:

Time(s)	Velocity(m/s)	Mach	Altitude(km)	Altitude(ft)	Drag(kN)
1180	6151	19.0	53.3	175,000	2,870
1589	4810	14.5	51.1	168,000	2,318
1859	3445	10.3	47.8	157,000	1,849
2082	2018	6.3	42.35	139,000	1,359

So, plugging these numbers into the Seebass/Tiegerman model:

flight altitude h(m) =	53300	51100	47800	42350
vehicle drag D(N) =	2870500	2318000	1849000	1359000
N-wave overpressure Δp(Pa) =	42.80135556	41.67204729	41.55288667	42.6139528
positive phase impulse I(Ns) =	1417762.39	1156171.097	915659.7167	657837.1682
duration T(s) =	1.114768856	0.933718945	0.741603564	0.519523846

So, for every ricochet, the overpressure felt at sea level will be about 42 Pa, or 0.9 lb/ft², just under that for the SR-71 Blackbird figure above. The impulse will be greatest at the highest velocity, as will the duration. For people on the ground, the overpressure figure is the most important in terms of perceived volume, and would be quite a bit less than for Concorde.

[OneSpeed]

Here's an update of the L2 StarHopper 200m hop simulation. From recent public footage, it looks like the distance from the launch pad to the new landing pad is about 160m. The audio is credit bocachicagal.

[OneSpeed]

Here's a speculative simulation of Starship Mk1 performing a flight to nearly 20kms in altitude, and returning using a 'Skydiver' profile.

[Jdeshetler]

Here's a speculative simulation of Starship Mk1 performing a flight to nearly 20kms in altitude, and returning using a 'Skydiver' profile.

Amazing simulation! 

With this Skydiver profile, how long will the flight be from launching to landing?

[OneSpeed]

Here's a speculative simulation of Starship Mk1 performing a flight to nearly 20kms in altitude, and returning using a 'Skydiver' profile.

Amazing simulation! 

With this Skydiver profile, how long will the flight be from launching to landing?

Thanks Jay!

You can see the elapsed time in the second row of output. From that, it would take about 1:50 to reach 18.8 kms, and it would land at 5:20, or 3:30 later.

[livingjw]

Here's a speculative simulation of Starship Mk1 performing a flight to nearly 20kms in altitude, and returning using a 'Skydiver' profile.

Amazing simulation! 

With this Skydiver profile, how long will the flight be from launching to landing?

Thanks Jay!

You can see the elapsed time in the second row of output. From that, it would take about 1:50 to reach 18.8 kms, and it would land at 5:20, or 3:30 later.

What direction is your zero reference AoA? Zero AoA appears to be measured relative to forward out the nose, but when you land it appears to be aft out the tail? Also positive AoA would normally be measured with windward surface towards the wind. Are you using the opposite?

John

[OneSpeed]

What direction is your zero reference AoA? Zero AoA appears to be measured relative to forward out the nose, but when you land it appears to be aft out the tail? Also positive AoA would normally be measured with windward surface towards the wind. Are you using the opposite?

John

The zero reference is always the flight path vector, regardless of orientation.

[livingjw]

What direction is your zero reference AoA? Zero AoA appears to be measured relative to forward out the nose, but when you land it appears to be aft out the tail? Also positive AoA would normally be measured with windward surface towards the wind. Are you using the opposite?

John

The zero reference is always the flight path vector, regardless of orientation.

Then how do you know when you are flying backwards (tail end first)?

John

[OneSpeed]

What direction is your zero reference AoA? Zero AoA appears to be measured relative to forward out the nose, but when you land it appears to be aft out the tail? Also positive AoA would normally be measured with windward surface towards the wind. Are you using the opposite?

John

The zero reference is always the flight path vector, regardless of orientation.

Then how do you know when you are flying backwards (tail end first)?

John

I hope I'm understanding your question correctly. It's when the angle between the flight path vector and the centreline of the ship (which is the AoA) exceeds ±90°.

[livingjw]

What direction is your zero reference AoA? Zero AoA appears to be measured relative to forward out the nose, but when you land it appears to be aft out the tail? Also positive AoA would normally be measured with windward surface towards the wind. Are you using the opposite?

John

The zero reference is always the flight path vector, regardless of orientation.

Then how do you know when you are flying backwards (tail end first)?

John

I hope I'm understanding your question correctly. It's when the angle between the flight path vector and the centreline of the ship (which is the AoA) exceeds ±90°.

Maybe I am missing something, but when you are about to land, the AoA is very near zero. I expected it to be around 180 degrees.

[OneSpeed]

Maybe I am missing something, but when you are about to land, the AoA is very near zero. I expected it to be around 180 degrees.

You are absolutely correct, thanks for pointing it out. It looks like there is a bug in the AoA display  when the ship has rotated over 360°, as it did in this profile. I'll fix that ASAP.

Edit: fixed the display issue, attached an updated video.

[OneSpeed]

The previous simulation was intended to explore the maximum possible envelope for a 200t dry mass Starship Mk1 with three 170t thrust Raptor engines, and a ceiling of 20kms. At full throttle, and a liftoff T/W of about 1.2, that allowed for about 250t of propellant.

However, the relevant Special Temporary Authority (STA) states:

Station Location

City	State	Latitude	Longitude	Mobile	Radius of Operation
Boca Chica	Texas	North  25  59  50	West  97  9  25	Boca Chica Pad Suborbital Test Veh Max Alt 22.5km	5.00

If the 'Radius of Operation' for communications is 5.00kms, and it is also the 'Radius of Operations' for the flight, then the envelope for the previous simulation would not be allowed. Also, having seen StarHopper fly twice now with throttle never exceeding about 70%, the Starship Mk1 test flight may use less than 100% throttle, and so the propellant load would need to be less than 250t. So, for this simulation, I'm assuming minimal downrange distance, and a maximum of 80% throttle. That allows for a GLOW of about 360t, and therefore 160t of propellant.

Some points of interest:

The simulation now displays both Inertial and Orbital velocities and accelerations (No more step change at the atmosphere height).
Maximum ascent velocity: Mach 1.34
Maximum descent velocity: Mach 0.83
Maximum acceleration: 1.85 gs
Terminal velocity on descent: 102m/s

The physics simulation in the recent Starship presentation has a terminal velocity of 66m/s. The difference could be largely down to the dry mass of Mk1 vs later variants (200t vs 105-120t).

[livingjw]

- Your drag seem low to me. On descent I picked off D=3081kN and q=7.1kPa. From these I calculated D/q=434m^2. This implies that CD is less than 1, since cross sectional area is greater than 434m^2. CD at 90 degrees should track similar to the chart below taken from Hoerner's drag book.

- You can use cross flow theory for obtaining CD and CL at any angles between ~55 degrees and ~125 degrees AoA which I believe is a good assumed range for SS. Use equation 23 to calculate CL and CD where CDbasic is the value from the first chart which is the CD at 90 degrees.

- Has anyone estimated the planform area?  I would guess it is in excess of 500m^2.

John

[OneSpeed]

- Your drag seem low to me. On descent I picked off D=3081kN and q=7.1kPa. From these I calculated D/q=434m^2. This implies that CD is less than 1, since cross sectional area is greater than 434m^2. CD at 90 degrees should track similar to the chart below taken from Hoerner's drag book.

- You can use cross flow theory for obtaining CD and CL at any angles between ~55 degrees and ~125 degrees AoA which I believe is a good assumed range for SS. Use equation 23 to calculate CL and CD where CDbasic is the value from the first chart which is the CD at 90 degrees.

- Has anyone estimated the planform area?  I would guess it is in excess of 500m^2.

John

Yes, my subsonic drag coefficient is closer to a constant 0.8, supersonic looks pretty close though. I'm actually treating the cross sectional shape including the flaps as more like a capsule, rather than a pure cylinder. Perhaps somewhere between the two is a better estimate?

Edit: just re-ran the sim with 105t dry mass, and 20t of landing propellant, and got a terminal velocity of 69m/s. Cd probably is a tad low.

[OneSpeed]

Now that we know that Starship orbital test launches are to be expected from Boca Chica, what would the minimum configuration be for such a test?

In particular, what if there was no payload? After some experimentation, I found that by reducing both the SS and SH propellant loads to about 2/3, it was no longer necessary for the Starship to have 6 engines. Even three is enough if you are prepared to let a 19 engine booster launch the ship into a relatively lofted trajectory.

Also, which stage is more likely to fail? The Super Heavy booster flight envelope is quite close to the Falcon 9 booster RTLS profile, and so not such a stretch goal. Starship orbital re-entry however, although similar in some ways to Shuttle, is new for SpaceX. By minimising the Starship engine count, perhaps the cost of failure can be reduced a little?

Raptor performance appears to be in a rapid state of flux, so it is difficult to say what performance will be available for the first orbital test. Anyway, I had to pick something, and the configuration in this simulation is:

Engine counts: SH 7 x 200t with gimbal, 12 x 250t fixed, SS 3 x 200t with gimbal.
Dry masses: SH 230t, SS 120t
Prop loads: SH 2200t, SS 700t

[envy887]

Thanks for sharing that, and nice work! Can you also do one with booster ASDS landing?

I'd expect initial SuperHeavy landings on orbital attempts to be at sea, as indicated in the EA for launching Starship from KSC. ASDS landings have lower environmental impact, lower risk to on-shore infrastructure, which means less insurance cost and better chance of getting a launch license, and also higher performance, which means less liftoff mass and fewer engines, or higher margins.

[OneSpeed]

Thanks for sharing that, and nice work!

Thanks for the feedback!

Can you also do one with booster ASDS landing?

I'd expect initial SuperHeavy landings on orbital attempts to be at sea, as indicated in the EA for launching Starship from KSC. ASDS landings have lower environmental impact, lower risk to on-shore infrastructure, which means less insurance cost and better chance of getting a launch license, and also higher performance, which means less liftoff mass and fewer engines, or higher margins.

All good points, but would the offshore platform at KSC necessarily be an ASDS? A fixed or floating platform say 20km offshore would cover most of the environmental concerns, but would make little difference to performance to orbit. Nevertheless, let's say they had a spare drone ship. Current F9 landings are certainly accurate enough, and the extra mass of the Super Heavy would not be noticed by an ASDS which can displace 20,000t+ of seawater.

So, again, after some experimentation, and with some slightly tighter margins, here's what the sim suggested:

Engine counts: SH 7 x 200t with gimbal, 6 x 250t fixed, SS 3 x 200t with gimbal.
Dry masses: SH 230t, SS 120t
Prop loads: SH 1600t, SS 600t

So, a saving of 6 Raptor engines, and about 700t of propellant.

[sdsds]

Attached is a visualization of the possible downrange landing location.

Recent comments from Musk suggest the production version of SH might have a center cluster of 8 engines. So a plausible SH prototype might have only those 8, and none on the outer ring. Is it right to think that would change the propellant loading only a little? Would it lead to any significant changes in the flight profile, e.g. conditions at stage sep or downrange distance to the landing location?

[OneSpeed]

Recent comments from Musk suggest the production version of SH might have a center cluster of 8 engines. So a plausible SH prototype might have only those 8, and none on the outer ring. Is it right to think that would change the propellant loading only a little? Would it lead to any significant changes in the flight profile, e.g. conditions at stage sep or downrange distance to the landing location?

It might be helpful to consider the conditions at liftoff. A rocket's thrust needs to exceed the mass of the vehicle to fly, in Starship's case by a factor of 1.5 or so. So, eight 210t thrust engines, giving 1,680t of thrust might be suitable for a gross lift of weight (GLOW) of around 1,120t. Adding another 20 300t engines would increase the thrust to 7,680t, for a GLOW of perhaps 5,000t. If the dry mass plus payload of the rocket is say 350t, then the difference in propellant loading is from 1,120 - 350 = 770t to 5,000 - 350 = 4,650t, six times greater.

Mainly because of the resultant differences in percentage mass fraction, the differences in flight profile would also be very significant. Hope this helps

[sdsds]

Good point: it certainly helps if the rocket can get off the ground!

A lift-off T/W ratio of 1.5 seems quite conservative though. 1.15 isn't outside the realm of possibility.

[envy887]

Good point: it certainly helps if the rocket can get off the ground!

A lift-off T/W ratio of 1.5 seems quite conservative though. 1.15 isn't outside the realm of possibility.

The reason for the high acceleration is to minimize the distance traveled during the first stage burn, reducing the fuel needed for booster RTLS. RTLS is a big deal for refueling launch turnaround time.

[abaddon]

For a point of comparison, based on Wiki I calculated the following thrust-to-weight ratios here: https://forum.nasaspaceflight.com/index.php?topic=46338.msg2081947#msg2081947:

Delta IV Heavy: 1.32
F9 Block 5: 1.42
Falcon Heavy: 1.65

[OneSpeed]

Here is a simulation of the upcoming Starship SN8 flight to 20kms altitude and return, assuming a 120t dry mass, significantly less than the 200t of Starship Mk1. We know SN8 will have three SL Raptor engines, but the sim showed that these would provide much more thrust than necessary to complete the mission. However, by running the engines at roughly 2/3 of full thrust, this mitigates the risk of a single engine failure. If the SN8 dry mass is as low as estimated, then a propellant load of as little as 100t would be sufficient.

[livingjw]

Here is a simulation of the upcoming Starship SN8 flight to 20kms altitude and return, assuming a 120t dry mass, significantly less than the 200t of Starship Mk1. We know SN8 will have three SL Raptor engines, but the sim showed that these would provide much more thrust than necessary to complete the mission. However, by running the engines at roughly 2/3 of full thrust, this mitigates the risk of a single engine failure. If the SN8 dry mass is as low as estimated, then a propellant load of as little as 100t would be sufficient.

What was your subsonic 90 deg drag coefficient?

John

[Twark_Main]

Here is a simulation of the upcoming Starship SN8 flight to 20kms altitude and return, assuming a 120t dry mass, significantly less than the 200t of Starship Mk1. We know SN8 will have three SL Raptor engines, but the sim showed that these would provide much more thrust than necessary to complete the mission. However, by running the engines at roughly 2/3 of full thrust, this mitigates the risk of a single engine failure. If the SN8 dry mass is as low as estimated, then a propellant load of as little as 100t would be sufficient.

What was your subsonic 90 deg drag coefficient?

John

Working backwards from drag force and dynamic pressure, it looks like the drag area at 90.1 degree AoA and M = 0.62 is about 432 m². Just plug in your estimate of frontal projected area to work out the drag coefficient.

[OneSpeed]

What was your subsonic 90 deg drag coefficient?

At 500m altitude in the sky diver orientation it is 0.965 form drag plus 0.002 for skin drag. This gives a velocity of 81m/s before rotation commences, somewhat higher than the 66m/s reached in the 2019 SpaceX simulation. However, my ship dry mass estimate is 120t, and the long term goal for Starship is more like 105t. As well, I'm carrying about 34t of propellant as ullage, and most of this is for ballast. To avoid landing off vertical like SN5 and 6, I'm running the three SL Raptors at 50% throttle for landing (apparently lower than that there is chugging). Even with 34t of ballast, that is a 1.7g hoverslam, quite a bit higher than for the Falcon 9 booster.

So, in short, I realise my terminal velocity seems high, but if the upcoming test confirms it, I'll certainly update the model to match.

[Twark_Main]

What was your subsonic 90 deg drag coefficient?

At 500m altitude in the sky diver orientation it is 0.965 form drag plus 0.002 for skin drag. This gives a velocity of 81m/s before rotation commences, somewhat higher than the 66m/s reached in the 2019 SpaceX simulation. However, my ship dry mass estimate is 120t, and the long term goal for Starship is more like 105t. As well, I'm carrying about 34t of propellant as ullage, and most of this is for ballast. To avoid landing off vertical like SN5 and 6, I'm running the three SL Raptors at 50% throttle for landing (apparently lower than that there is chugging). Even with 34t of ballast, that is a 1.7g hoverslam, quite a bit higher than for the Falcon 9 booster.

So, in short, I realise my terminal velocity seems high, but if the upcoming test confirms it, I'll certainly update the model to match.

This suggests you're assuming a frontal cross section of 447 m², correct?

[Okie_Steve]

Attached is a visualization of the possible downrange landing location.

The Gulf of Mexico get deep fairly quickly along the Texas coast but there is a shelf that continues from the Yucatan peninsula and reaches further North than Cuba. Also the West Florida Shelf is about the size of the peninsula. Depending on your definition of shallow/deep there are lots of places in the gulf for a platform, but there in the middle it is 4000+ meters deep. Definite ASDS territory there.

[OneSpeed]

This suggests you're assuming a frontal cross section of 447 m², correct?

At that altitude and velocity, yes. The drag coefficients and cross sectional areas are constantly being re-calculated by SpaceSim, based on local velocity, density and viscosity, as well as the shape and orientation of the components. The program has been tweaked over time by comparing it to as many real rocket launches as possible, and so the calculated cross sectional area may be slightly different from the planform area at any point in time. Further tweaking may well be required for the Starship model. If you are interested, an earlier version of the C# program source code is available at: https://github.com/zlynn1990/SpaceSim

[vaporcobra]

Given the knowledge that Raptor is currently limited to ~90 seconds of continuous operation at 300 bar, it would be interesting to see what the absolute ceiling of performance is if you assume that neither Super Heavy or Starship can burn for longer than that uninterrupted. Or, say, if current longevity permitted several more minutes of operation but only at ~80% throttle or ~250 bar. Basically, what's possible within those known or estimable constraints

[steveleach]

Given the knowledge that Raptor is currently limited to ~90 seconds of continuous operation at 300 bar, it would be interesting to see what the absolute ceiling of performance is if you assume that neither Super Heavy or Starship can burn for longer than that uninterrupted. Or, say, if current longevity permitted several more minutes of operation but only at ~80% throttle or ~250 bar. Basically, what's possible within those known or estimable constraints

It would be more accurate to say that the Raptor they tested was limited to ~90 seconds of operation at 300 bar.

The design is evolving all the time though. That test will have told them what failed, and they will now be working on ways to make that component survive longer. Then they will retest, and see what component to work on next. Rinse and repeat.

Only when they can't economically improve things significantly will we be able to say what the actual limitations are.

[OneSpeed]

Here is a simulation of the upcoming Starship Number 8 flight to 12.5km altitude and return. I didn't post a simulation of the projected 15km flight, because it seemed likely that the ship would go supersonic unless an engine was shut down prematurely, or there was a lot of propellant as ballast.

A 12.5km apogee makes a subsonic flight more likely, reducing the risk in what is already a risky enough proposition.

[Oersted]

Initial T/W of only 1.16? - Wauw, I had no idea it would lift off so slowly. That is going to be extremely impressive to watch. Saturn V vibes. Literally!

[leetdan]

The exclusion area extends a fair amount offshore, and isn't centered on the launch pad.  Isn't it reasonable that some amount of fuel would be spent in translating away from the ground infrastructure?  Wouldn't that translate into less deep throttling and higher TWR on liftoff?

[daveglo]

Here is a simulation of the upcoming Starship Number 8 flight to 12.5km altitude and return. I didn't post a simulation of the projected 15km flight, because it seemed likely that the ship would go supersonic unless an engine was shut down prematurely, or there was a lot of propellant as ballast.

A 12.5km apogee makes a subsonic flight more likely, reducing the risk in what is already a risky enough proposition.

THIS is the best explanation I've seen regarding the change from 15km to 12.5.  No dreamed-up safety rules or FAA administrative garbage, just pure aerospace engineering.

Nice work, OneSpeed (as usual)!

[OneSpeed]

The exclusion area extends a fair amount offshore, and isn't centered on the launch pad.  Isn't it reasonable that some amount of fuel would be spent in translating away from the ground infrastructure?  Wouldn't that translate into less deep throttling and higher TWR on liftoff?

It's certainly possible that the flight won't be straight up and down, as per some of the Grasshopper flights. As you can see from the sim, there is still about 40t of propellant remaining after landing. This ballast is required to keep the profile subsonic. So, there should be plenty of prop available to translate offshore a bit if they want, and literally glide back.

However, the T/W also needs to be quite low for the three engine hoverslam. From the sim, it will be quite sporty at around 1.7g versus Falcon 9 at around 1.3g. On that basis, I'm expecting a launch quite like Saturn V, as Oersted suggests.

[eriblo]

The exclusion area extends a fair amount offshore, and isn't centered on the launch pad.  Isn't it reasonable that some amount of fuel would be spent in translating away from the ground infrastructure?  Wouldn't that translate into less deep throttling and higher TWR on liftoff?

It's certainly possible that the flight won't be straight up and down, as per some of the Grasshopper flights. As you can see from the sim, there is still about 40t of propellant remaining after landing. This ballast is required to keep the profile subsonic. So, there should be plenty of prop available to translate offshore a bit if they want, and literally glide back.

However, the T/W also needs to be quite low for the three engine hoverslam. From the sim, it will be quite sporty at around 1.7g versus Falcon 9 at around 1.3g. On that basis, I'm expecting a launch quite like Saturn V, as Oersted suggests.

It is worth noting that the FCC permits that still apply specify a 2 km radius of operation from the pad which is significantly more restrictive than the exclusion zone or the TFR.

I wonder if they intend to have any significant propellant ballast - 40 t extra would be ~20 t sloshing around in the main tanks during the various flips and flops. I modified flightclubs 15 km profile to 12.5 km with a 3 Raptor ascent and 2 Raptor landing and adjusted the propellant load to 90 t have 25-30 t during descent (i.e. full headers and empty mains), ~7.5 of which are used for landing:

85% throttle, max velocity 350 m/s

and

50% throttle, max velocity 275 m/s

The 50% throttle case max velocity is the same as your simulation despite having 30 t of propellant less and 1 km higher cut off so I guess flightclub has higher ascent drag - I wonder if it might be related to the fact that it has a 9 m radius Starship Interestingly the propellant load works out to the same for both throttle settings suggesting that the increase in gravity losses are about the same as the decrease in drag losses.

[OneSpeed]

I wonder if they intend to have any significant propellant ballast - 40 t extra would be ~20 t sloshing around in the main tanks during the various flips and flops.

Perhaps, or maybe SN8 dry mass is more than my 120t estimate, and the required ballast is correspondingly less. For example, SN8 has been fabricated mostly from 4mm thick coil, and SpaceX is hoping to move to 3mm in the future. If her dry mass is actually 150t, then the prop load could be 30t less.

Either way, the sim tells us that the gross mass required to not exceed the speed of sound for a 12.5km apogee and 3 raptors at 50% throttle is around 240t. YMMV.

The 50% throttle case max velocity is the same as your simulation despite having 30 t of propellant less and 1 km higher cut off so I guess flightclub has higher ascent drag - I wonder if it might be related to the fact that it has a 9 m radius Starship

Interesting, why do they have a 9m radius Starship? Are they assuming skydiver orientation on the way up?

The exclusion area extends a fair amount offshore, and isn't centered on the launch pad.  Isn't it reasonable that some amount of fuel would be spent in translating away from the ground infrastructure?  Wouldn't that translate into less deep throttling and higher TWR on liftoff?

Thinking about this some more, the sub-tropical jetstream over Boca Chica could be as much as 25m/s on launch day, which would carry SN8 downrange somewhat for free.

[su27k]

Will there be attempts to simulate SN8's actual flight profile? It would be interesting to see how much performance they're holding back due to the need to burn all the way to apogee. I assume it's possible to get rough speed and altitude estimate from some of the amateur videos.

[RedLineTrain]

It would also be very interesting to see an updated Starship P2P flight based on what we know so far based on SN8.

This is a speculative simulation of a single stage Starship P2P flight. With 9 SL Raptors, and a full propellant load, the initial T/W is a healthy 1.6. So, throttle back for MaxQ occurs early, at the 36 second mark. If the ship were to continue to a purely ballistic trajectory, re-entry g forces would be prohibitive (~20gs). Instead, I've used negative pitch to flatten the trajectory, reducing the re-entry flight path angle. This allows the ship to skip like a stone on a pond, extending the range out to 10,000kms. The peak g force on the first 'bounce' is just over 4. If the Starship had larger (dragon?) wings, and hence a greater lift coefficient, the peak could be reduced further, and the range extended beyond 10,000kms.

[OneSpeed]

Will there be attempts to simulate SN8's actual flight profile? It would be interesting to see how much performance they're holding back due to the need to burn all the way to apogee. I assume it's possible to get rough speed and altitude estimate from some of the amateur videos.

I've started on one, it's tricky without the telemetry. From the frost lines, does anyone have a good number for how much propellant was on board?

It would also be very interesting to see an updated Starship P2P flight based on what we know so far based on SN8.

Perhaps unintuitively, SN8's flight changes the P2P model very little. So far, it actually appears to validate it.

[vaporcobra]

Will there be attempts to simulate SN8's actual flight profile? It would be interesting to see how much performance they're holding back due to the need to burn all the way to apogee. I assume it's possible to get rough speed and altitude estimate from some of the amateur videos.

I've started on one, it's tricky without the telemetry. From the frost lines, does anyone have a good number for how much propellant was on board?

Just knowing that SN8 was more or less hovering at apogee, that Raptor's nominal thrust range is 90-200 tons, and a maximum mass flow rate of ~550 kg, it should be relatively easy to calculate a reasonable range of propellant loads. Just splitting the difference and arbitrarily assuming an average Raptor mass flow rate of 400 kg/s (~72.5% thrust) during the test, I get a propellant mass of ~250 tons at liftoff.

If all Raptors were capped at either minimum or maximum throttle, propellant consumption would have been somewhere between 160 and 350 tons. Unfortunately, assuming SN8's empty weight is around 70 tons, I think any number in that range could be reasonable despite the slow acceleration observed at liftoff.

[OneSpeed]

Here is a simulation of the recent SN8 flight to 12.5km and return.

The simulation assumes a dry mass of 120t, and 230t of propellant. If that was the case, there would have been roughly 37t of residual propellant at landing.



Edit: changed plot to .jpg

[DreamyPickle]

SN8-telemetry.png fails to open unless extension is manually changes to BMP.

Also, the G-force meter seems to indicate passengers only experience 1.5G? The maneuver looks scary but might not be all that bad.

[wes_wilson]

SN8-telemetry.png fails to open unless extension is manually changes to BMP.

Also, the G-force meter seems to indicate passengers only experience 1.5G? The maneuver looks scary but might not be all that bad.

I was trying to imagine what that ride would feel like; would be interesting to hear it described in people terms.  You're weightless in orbit; you'd feel some semblance of gravity when re-entering and decelerating(?); back to zero gravity when you're free falling at terminal velocity; then a brief 1.5g as you rotate and back to full g as you land?

[steveleach]

SN8-telemetry.png fails to open unless extension is manually changes to BMP.

Also, the G-force meter seems to indicate passengers only experience 1.5G? The maneuver looks scary but might not be all that bad.

I was trying to imagine what that ride would feel like; would be interesting to hear it described in people terms.  You're weightless in orbit; you'd feel some semblance of gravity when re-entering and decelerating(?); back to zero gravity when you're free falling at terminal velocity; then a brief 1.5g as you rotate and back to full g as you land?

I suspect it would feel similar in many ways to being a passenger in a fast lap of a motor racing circuit. Terrifying, nauseating, exhilarating.

[cdebuhr]

SN8-telemetry.png fails to open unless extension is manually changes to BMP.

Also, the G-force meter seems to indicate passengers only experience 1.5G? The maneuver looks scary but might not be all that bad.

I was trying to imagine what that ride would feel like; would be interesting to hear it described in people terms.  You're weightless in orbit; you'd feel some semblance of gravity when re-entering and decelerating(?); back to zero gravity when you're free falling at terminal velocity; then a brief 1.5g as you rotate and back to full g as you land?

My understanding (FWIW - possibly not much!) is that you'd feel a heck of a lot more than "some semblance" of gravity during atmospheric entry and deceleration.  More like 3g to 4g, give or take.  Also, when falling at terminal velocity you will feel slightly more than 1g ... 1g for earths gravity, and slightly more because your actually decelerating all the way as the atmosphere get more dense as you fall.

[eriblo]

Since I have not yet seen it done: I took a shot at estimating the propellant load from SN8s flight from the visible frost.

For all flight attempts the LOX tank frost band reached just below the barrel weld above the aft bulkhead attachment weld. From Rafaels schematics the volume of the aft bulkhead is ~135 m^3 and the distance from attachment weld to barrel weld distance is ~0.5 m.

Combined main tank LOX volume (neglecting LCH4 piping): 166 m^3
Main tank LOX mass (density 1200 kg/m^3): 199 t
Total main tank propellant load (O/F = 3.6) : 254 t
Additional propellant in headers and down-comers: Up to 33 t

Rafael could give more exact numbers for the volumes but I estimate that the fuel would not quite fill the common bulkhead which is consistent with the lack of a corresponding frost line (making this close to an upper bound for propellant load). The other uncertainty is the height of the LOX frost band and how well it corresponds to liquid level - it looks to me like the barrel weld is visible so the height should be < 0.5 m. Using 0.4 m cuts total propellant by 10 t.

It is not too far of from the simulations but my feeling is that would be hard to get them to agree. I believe that they hovered at apogee specifically to minimize residuals (to prevent sloshing) and there is not much room for using more propellant - drag and throttle losses are minimal and the dry mass is already at the upper bound of the given numbers and best estimates.

[OneSpeed]

Here is a simulation of an orbital Starship launch, updated to have a 28 engine Super Heavy. The payload is 150t to LEO, and I'm assuming Super Heavy has 8 x 210t thrust Raptors with gimbal and throttle, and 20 x 300t thrust Raptors without either.

[vaporcobra]

Here is a simulation of an orbital Starship launch, updated to have a 28 engine Super Heavy. The payload is 150t to LEO, and I'm assuming Super Heavy has 8 x 210t thrust Raptors with gimbal and throttle, and 20 x 300t thrust Raptors without either.

Excellent work, as always! Out of curiosity, are you able to model performance to inclinations that would require dogleg maneuvers? Namely 53, 70, and 97.6 degrees for Starlink. Unclear if 97.6 degrees is even within the realm of possibility but that would be interesting to see

[OneSpeed]

Excellent work, as always! Out of curiosity, are you able to model performance to inclinations that would require dogleg maneuvers? Namely 53, 70, and 97.6 degrees for Starlink. Unclear if 97.6 degrees is even within the realm of possibility but that would be interesting to see

Thanks! Yes, SpaceSim can model any launch azimuth, as well as doglegs on top of those. For Starship, the range of allowable launch azimuths will depend on the launch site. Phobos and Deimos could potentially make nearly any azimuth achievable, and so doglegs might not be necessary. With that in mind, do you have a short list of scenarios?

[Thrustpuzzle]

I'd love to see a simulation of the same cargo launch but with SH landing roughly 250km downrange (ie launch from Boca Chica, land on downrange platform).  The two interesting questions it would answer is 1) what the optimal downrange distance would be and 2) how much extra orbital payload mass it enables.

[steveleach]

Here is a simulation of an orbital Starship launch, updated to have a 28 engine Super Heavy. The payload is 150t to LEO, and I'm assuming Super Heavy has 8 x 210t thrust Raptors with gimbal and throttle, and 20 x 300t thrust Raptors without either.

Epic!

ISTR Musk commenting on a booster-catching animation once that SH will come in almost vertically, rather than at an oblique angle.

[soyuzu]

Here is a simulation of an orbital Starship launch, updated to have a 28 engine Super Heavy. The payload is 150t to LEO, and I'm assuming Super Heavy has 8 x 210t thrust Raptors with gimbal and throttle, and 20 x 300t thrust Raptors without either.

Nice job as always!

Have you considered shutting down Raptor SLs on Starship mid flight to gain more Isp? What is your assumption of stage empty mass and residual propellant? I’m also interested in simulation of 21t to GTO and Dearmoon launch without refueling.

[Aeneas]

Here is a simulation of an orbital Starship launch, updated to have a 28 engine Super Heavy. The payload is 150t to LEO, and I'm assuming Super Heavy has 8 x 210t thrust Raptors with gimbal and throttle, and 20 x 300t thrust Raptors without either.

Nice job as always!

Have you considered shutting down Raptor SLs on Starship mid flight to gain more Isp? What is your assumption of stage empty mass and residual propellant? I’m also interested in simulation of 21t to GTO and Dearmoon launch without refueling.

After burning 2/3 of the fuel, the switch likely increases payload by ~7 t. 21 t GTO is possible but dearmoon without refueling not at all.

[_MECO]

Here is a simulation of an orbital Starship launch, updated to have a 28 engine Super Heavy. The payload is 150t to LEO, and I'm assuming Super Heavy has 8 x 210t thrust Raptors with gimbal and throttle, and 20 x 300t thrust Raptors without either.

9 Gs of deceleration for Super Heavy. Is that going to wreck the engines or the engine bay? It takes a lot of hot wind to put the mass of a Super Heavy through 9 Gs.

[OneSpeed]

Out of curiosity, are you able to model performance to inclinations that would require dogleg maneuvers? Namely 53, 70, and 97.6 degrees for Starlink. Unclear if 97.6 degrees is even within the realm of possibility but that would be interesting to see

From KSC at 28.5° latitude, and using β = asin(cos(mi)/cos(ϕ)):

Orbital inclination (from East)	Launch azimuth (from North)	ΔV (km/s)
28.5°	90°	7.407
53°	43°	7.542
70°	22.9°	7.666
97.6°	-8.6°	7.888

So a 97.6° retrograde orbit requires 481m/s more than I modelled. If that's what you are after, I could model it.

I'd love to see a simulation of the same cargo launch but with SH landing roughly 250km downrange (ie launch from Boca Chica, land on downrange platform).  The two interesting questions it would answer is 1) what the optimal downrange distance would be and 2) how much extra orbital payload mass it enables.

I guess it's a trade, the further downrange you are prepared to go, the more payload you can loft. But if they are going to relaunch boosters in an hour, they need to RTLS.

ISTR Musk commenting on a booster-catching animation once that SH will come in almost vertically, rather than at an oblique angle.

Do you mean this one? If so, then yes, it's coming in like a javelin, not quite vertically, and no sky-diver or late flip like Starship.

https://twitter.com/elonmusk/status/1375510416994967552
Tweet Contents:  Super Heavy doing the flip maneuver
Heavy comes in more like a javelin. Similar to Falcon 9, but caught by the tower vs landing on legs.

Have you considered shutting down Raptor SLs on Starship mid flight to gain more Isp? What is your assumption of stage empty mass and residual propellant? I’m also interested in simulation of 21t to GTO and Dearmoon launch without refueling.

Certainly, but until you have reached orbital velocity, shutting down engines increases gravity losses, so it's another trade. Once in orbit, then Isp is king. For Starship I'm assuming 120t dry, and 30t of residuals, about 10t of which is ullage. For Super Heavy, 230t dry, also 10t of ullage, cutting it a bit finer.

9 Gs of deceleration for Super Heavy. Is that going to wreck the engines or the engine bay? It takes a lot of hot wind to put the mass of a Super Heavy through 9 Gs.

Sure, but this is a RTLS profile, and not nearly as challenging as landing downrange. As you can see from the video, the peak heat flux for Super Heavy is 68.6kW/m². By comparison, from my as yet unpublished Starlink L18 ASDS sim, I get a peak heat flux of 44.2kW/m², at a much higher altitude and velocity, and that figure doesn't include any additional heat flux from the three running Merlin plumes.

It will be interesting to see how much protection the stainless Super Heavy skirt provides for the Raptors on entry, but in any case it will be about an order of magnitude less than the heat flux Starship will have to endure from orbital re-entry.

Edit: put the numbers in a table.

[steveleach]

ISTR Musk commenting on a booster-catching animation once that SH will come in almost vertically, rather than at an oblique angle.

Do you mean this one? If so, then yes, it's coming in like a javelin, not quite vertically, and no sky-diver or late flip like Starship.

Tweet Contents:  Super Heavy doing the flip maneuver
Heavy comes in more like a javelin. Similar to Falcon 9, but caught by the tower vs landing on legs.

No, there was another one a while back where he was responding to someone's visualisation of the booster catching mechanism. If I recall correctly, the concept wasn't sufficiently open at the top, so Musk commented that the booster will be descending almost vertically as it reaches the tower, rather than approaching from the side.

It isn't in the main tweet index though, but I'll see if I can dig it out.

[sdsds]

Here is a simulation of an orbital Starship launch

Fantastic!

I'm confused about the numbers shown at the end. Altitude is 165.3, thrust is zero, yet perigee is 168.3. An orbiting object can coast up to its apogee, but how can it be lower than its perigee?

[OneSpeed]

This is a cross post from the Starship Engineering thread, where InterestedEngineer asked:

Are there any "good" estimates of the glide slope ratio for Starship?

Google returns a Reddit article that seems to guess at 1:1, but IANAAE (I am not an aeronautical engineer) so I have no way of telling how wild a guess that is.

and Pueo responded:

L/D should be equal to glide slope in steady state.  If the maximum L/D is 1.2 at 25° AOA then the Starship should only be pitched -14.8° when in a gliding configuration because the glide slope should be -39.8° in respect to the horizontal.  The trick is getting there because starting pitched -14.8° puts you at an AOA of 75.2° where you're certainly stalled.

Another consideration is that when Starship is not powered, it is only controllable by the body flaps in a small range of AoA. I've tried modelling this, and found I needed to limit AoA to no less than 60°, otherwise I lost control of the ship. Perhaps a Kerbal Space Program expert could do better than I did? Anyway the maximum downrange I could get from 12.5km with good control was about 8km, and Starship was completely stalled through the entire descent.

Regarding steady state, that only happened once, instantaneously at T+06:16, and I've annotated that frame below. The entire video of the sim is on YouTube:

[livingjw]

This is a cross post from the Starship Engineering thread, where InterestedEngineer asked:

Are there any "good" estimates of the glide slope ratio for Starship?

Google returns a Reddit article that seems to guess at 1:1, but IANAAE (I am not an aeronautical engineer) so I have no way of telling how wild a guess that is.

and Pueo responded:

L/D should be equal to glide slope in steady state.  If the maximum L/D is 1.2 at 25° AOA then the Starship should only be pitched -14.8° when in a gliding configuration because the glide slope should be -39.8° in respect to the horizontal.  The trick is getting there because starting pitched -14.8° puts you at an AOA of 75.2° where you're certainly stalled.

Another consideration is that when Starship is not powered, it is only controllable by the body flaps in a small range of AoA. I've tried modelling this, and found I needed to limit AoA to no less than 60°, otherwise I lost control of the ship. Perhaps a Kerbal Space Program expert could do better than I did? Anyway the maximum downrange I could get from 12.5km with good control was about 8km, and Starship was completely stalled through the entire descent.

Regarding steady state, that only happened once, instantaneously at T+06:16, and I've annotated that frame below. The entire video of the sim is on YouTube:

What was causing loss of control? Seems like there should be plenty of pitch authority.

John

[OneSpeed]

What was causing loss of control? Seems like there should be plenty of pitch authority.

John

It was tending to pitch down irretrievably. The body flaps have no authority once the AoA gets too small. They are simply behaving like fins with variable dihedral, and the aft flaps are quite a bit larger, so it becomes a dart. Of course a better pilot might help

[niwax]

This is a cross post from the Starship Engineering thread, where InterestedEngineer asked:

Are there any "good" estimates of the glide slope ratio for Starship?

Google returns a Reddit article that seems to guess at 1:1, but IANAAE (I am not an aeronautical engineer) so I have no way of telling how wild a guess that is.

and Pueo responded:

L/D should be equal to glide slope in steady state.  If the maximum L/D is 1.2 at 25° AOA then the Starship should only be pitched -14.8° when in a gliding configuration because the glide slope should be -39.8° in respect to the horizontal.  The trick is getting there because starting pitched -14.8° puts you at an AOA of 75.2° where you're certainly stalled.

Another consideration is that when Starship is not powered, it is only controllable by the body flaps in a small range of AoA. I've tried modelling this, and found I needed to limit AoA to no less than 60°, otherwise I lost control of the ship. Perhaps a Kerbal Space Program expert could do better than I did? Anyway the maximum downrange I could get from 12.5km with good control was about 8km, and Starship was completely stalled through the entire descent.

Regarding steady state, that only happened once, instantaneously at T+06:16, and I've annotated that frame below. The entire video of the sim is on YouTube:

I just noticed your last simulation of single stage P2P was using the old triple fin design. Do you have any ideas on how the new design might change this profile? A new simulation would be much appreciated.

[livingjw]

One speed,

- It appears that there is an error in your lift display when the AoA exceeds 90 degrees. Probably a sign problem.

John

[mikelepage]

This is a cross post from the Starship Engineering thread, where InterestedEngineer asked:

Are there any "good" estimates of the glide slope ratio for Starship?

Another consideration is that when Starship is not powered, it is only controllable by the body flaps in a small range of AoA. I've tried modelling this, and found I needed to limit AoA to no less than 60°, otherwise I lost control of the ship. Perhaps a Kerbal Space Program expert could do better than I did? Anyway the maximum downrange I could get from 12.5km with good control was about 8km, and Starship was completely stalled through the entire descent.

Regarding steady state, that only happened once, instantaneously at T+06:16, and I've annotated that frame below. The entire video of the sim is on YouTube:

Great work OneSpeed.

Not sure if this has been done elsewhere, but it occurs to me that if one can get 8km cross range in the final 12.5km of descent, one could probably get considerably more cross range you combined this with your atmosphere skipping approach upthread.

Some kludgy math:
Looks like you get a dV of about 1km/s in that first atmospheric bounce. Perhaps if one angled the starship to get sideways deflection during that bounce, so you halve the height of the bounce, but get maybe 500m/s sideways dV for 4 minutes or so before you went in for the next entry interface (which you do at the expense of downrange distance)? 

That works out to at least 120km or so of cross range when coming in from orbit - just on that first bounce.  Seems like there ought to be a way to get several hundred km of cross range on the way in, then use that final 8km of "glide" during the vertical descent to fine tune your landing location. Not quite STS's 2000km of cross range, but still plenty.

[OneSpeed]

SpaceX recently released some details of the first orbital test flight. In particular, they released an event timeline, which I have attempted to model.

Event	T+ time (seconds)
Liftoff	0
MECO	169
Stage Separation	171
SES	176
Booster Touchdown	495
SECO	521
Ship Splashdown	5420

They also stated that their objective is to collect as much data as possible during flight to quantify entry dynamics and better understand what the vehicle experiences in a flight regime that is extremely difficult to accurately predict or replicate computationally.

This model represents the minimum viable product that I think can deliver that objective. Landing the booster offshore allows a slight reduction in booster engine count from 19 to 18, which is a convenient subset of the current 28 engine Super Heavy design. If SpaceX do attempt to recover the Super Heavy by landing it on a drone ship or platform, then perhaps 10 legs of the same design already used by Starship could be fitted?

I have assumed three SL engines only for the Starship, and that works well for the published event timeline. If the Vacuum Raptor is ready for the orbital test, then a much shorter Super Heavy burn with fewer engines would suffice.

[oses]

Great work!

Is it reasonable to say that it seems that the timeline of events in the FCC briefing is seemingly incompatible with a full set of raptors on SS + SH?

I'd definitely be interested to see how much margin they gain by having a full set of raptors onboard, especially with Chris's report of that being the plan.

[OneSpeed]

Great work!

Is it reasonable to say that it seems that the timeline of events in the FCC briefing is seemingly incompatible with a full set of raptors on SS + SH?

I'd definitely be interested to see how much margin they gain by having a full set of raptors onboard, especially with Chris's report of that being the plan.

They may well install all 34 Raptors, I'm just saying that the minimum viable product has fewer than that. They go from about 10.6 km/s of ΔV with 18 + 3 to about 12.7 km/s with a fully populated stack. Maybe they just run all those engines at about 65% of full throttle for reliability, and have a partial propellant load? If so, the performance would be similar to the simulation above.

[gsa]

SpaceX recently released some details of the first orbital test flight. In particular, they released an event timeline, which I have attempted to model.

Thank you for your great work!
I'm sorry, I've got a very stupid question to ask. Acceleration on your graph is measured in dm/s². IIRC 1 dm = 0.1 m, so 1000 dm/s² = 100 m/s² ≈ 10 g. I'm not sure this is right. Am I missing something?

[OneSpeed]

SpaceX recently released some details of the first orbital test flight. In particular, they released an event timeline, which I have attempted to model.

Thank you for your great work!
I'm sorry, I've got a very stupid question to ask. Acceleration on your graph is measured in dm/s². IIRC 1 dm = 0.1 m, so 1000 dm/s² = 100 m/s² ≈ 10 g. I'm not sure this is right. Am I missing something?

You're correct, it should be cm/s².

[edzieba]

If the flight profile can be met with a significantly sandbagged configuration of Raptors, what size of mass-simulator could be added to Starship with a full complement of Raptors and still complete the mission?

[OneSpeed]

If the flight profile can be met with a significantly sandbagged configuration of Raptors, what size of mass-simulator could be added to Starship with a full complement of Raptors and still complete the mission?

Good question! Perhaps not surprisingly, about 150t.

Edit: You might want to deploy the mass simulator before re-entry though.

[sferrin]

If the flight profile can be met with a significantly sandbagged configuration of Raptors, what size of mass-simulator could be added to Starship with a full complement of Raptors and still complete the mission?

Good question! Perhaps not surprisingly, about 150t.

Edit: You might want to deploy the mass simulator before re-entry though.

And make it conical in shape and paint it black for the lulz. 

[soyuzu]

If the flight profile can be met with a significantly sandbagged configuration of Raptors, what size of mass-simulator could be added to Starship with a full complement of Raptors and still complete the mission?

Residual fuel to simulate a tanker mission.

[beelsebob]

If the flight profile can be met with a significantly sandbagged configuration of Raptors, what size of mass-simulator could be added to Starship with a full complement of Raptors and still complete the mission?

Residual fuel to simulate a tanker mission.

They're not going to deliberately make a bigger boom than they have to when they land it.

[vaporcobra]

This raises the question: what's the maximum velocity a three-engine Starship can reach under its own power on either an expendable or recoverable trajectory?

https://twitter.com/elonmusk/status/1405588281622859778

[OneSpeed]

This raises the question: what's the maximum velocity a three-engine Starship can reach under its own power on either an expendable or recoverable trajectory?

I've just run a quick and dirty expendable sim. It all depends on the maximum throttle that the Raptors can sustain. E.g. if they can maintain 80% throttle, hypersonic re-entry at Ma 7.5 appears possible. However, I'm not sure if maintaining 80% throttle is feasible yet.

[OneSpeed]

Here is an update of the Orbital Launch Test simulation by Zach, the original author of SpaceSim. It includes an updated simulation, audio from the Falcon Heavy test flight, and amazing animations from Alexander Svan and C-bass Productions.

[OneSpeed]

The recent Everyday Astronaut interview with Elon Musk covered several updates to Starship, including:

Raptor2 Boost and Centre engines:
1. Common thrust is now 230t at sea level, or 2,255kN.
2. The thrust increase was obtained by opening up the throat, with consequent reductions in Isp to 327/8s, and chamber pressure to 298 bar.
3. This implies an increase in maximum propellant flow to around 700kg/s per engine.

Super Heavy and Starship dry mass:
1. Super Heavy dry mass is much less that previous estimates, at around 160t, including 29 Raptors of 2t each. Ullage is an additional 20t.
2. Super Heavy propellant capacity is increased (again) to 3,600t, but fuselage length is reduced to 69m.
3. Starship dry mass is slightly reduced to somewhere between 100 and 120t.

As well, it seems likely that staging and the boostback flip will become a single operation. I've modelled it in the sim, and separation works smoothly once the correct rate of pitch is reached. The order of events might be:

1. The booster shuts down most of its engines, keeping perhaps three centre engines running.
2. Those engines pitch up the whole stack, while keeping all propellant settled.
3. Once sufficient pitch rate is acheived, all engines are shut down, and separation is commanded.
4. The centripital force is sufficient to separate the ship and booster, and both continue to pitch.
5. At least one central ship engine is fired ASAP, resettling its propellant.
6. At least one central booster engine is fired, as per F9 boostback, also resettling its propellant.
7. As the ships pitch angle is restored, the RaptorVacs are lit, and launch proceeds normally.
8. Once the booster reaches a horizontal pitch, the remaining central engines are lit for boostback.

As for an entry burn, I've modelled this sim without one, and the sim gets right on 150t payload to orbit. If an entry burn is required, then payload will be somewhat less.

[vaporcobra]

Small tidbit in the GAO HLS protest denial hints at SpaceX's estimated boiloff during a Starship HLS mission: "tens of thousands of kg," or what I'd peg between 30 and 80 metric tons (and up to 20-90t). Might be useful for modelers/simulators

https://twitter.com/13ericralph31/status/1425244027956441093

[Keldor]

Small tidbit in the GAO HLS protest denial hints at SpaceX's estimated boiloff during a Starship HLS mission: "tens of thousands of kg," or what I'd peg between 30 and 80 metric tons (and up to 20-90t). Might be useful for modelers/simulators

https://twitter.com/13ericralph31/status/1425244027956441093

Is there more information about this?  Since this is a protest, they may be assuming a very pessimistic launch schedule for the tankers that results in an abnormally high amount of boiloff.

[vaporcobra]

Small tidbit in the GAO HLS protest denial hints at SpaceX's estimated boiloff during a Starship HLS mission: "tens of thousands of kg," or what I'd peg between 30 and 80 metric tons (and up to 20-90t). Might be useful for modelers/simulators

Is there more information about this?  Since this is a protest, they may be assuming a very pessimistic launch schedule for the tankers that results in an abnormally high amount of boiloff.

No. I believe GAO is citing directly from SpaceX's HLS Option A proposal, so the best glimpse we'll ever get is probably what GAO quotes in protests and what NASA itself brought up in their selection statement. We have to suffice with the crumb lol. Personally, I'd guess that "over the HLS mission" implies that tens of tons of boiloff occur after the Starship Lander has drained the depot ship. But again, we can only speculate from the crumbs.

[OneSpeed]

Here is a CompariSim™ between two Starships with 6 and 9 engines each.

The Super Heavy boosters both have 33 Raptor2 engines. However, the 9 engine Starship has three extra Vacuum Raptor2s, an additional 300t of propellant, giving a liftoff T/W of 1.4, and does not throttle back for MaxQ.

The simulation suggests that these combined changes increase the payload to a 26° inclination (e.g. Boca Chica) from 150 to 200t.

The payload for Starlink satellites to 53° would be about 6% less, or 188t.

[Nydoc]

The Super Heavy boosters both have 33 Raptor2 engines.

How tall would these 33-engine Super Heavy boosters be?

[xvel]

~70m similar to two already builded boosters, 33 engine boosters are build right now, they are not "made up" by OneSpeed

[Nydoc]

Not suggesting they are made up, just asking how tall they would be. My understanding is that the current boosters are 29 engines, about 70m tall and have 3600t of propellant. OneSpeed had mentioned the 33-engine boosters having an additional 300t of propellant so I assume they would be taller than 70m?

[steveleach]

Not suggesting they are made up, just asking how tall they would be. My understanding is that the current boosters are 29 engines, about 70m tall and have 3600t of propellant. OneSpeed had mentioned the 33-engine boosters having an additional 300t of propellant so I assume they would be taller than 70m?

He said the 9 engine Starship as an extra 300t, not the booster.

[eriblo]

Not suggesting they are made up, just asking how tall they would be. My understanding is that the current boosters are 29 engines, about 70m tall and have 3600t of propellant. OneSpeed had mentioned the 33-engine boosters having an additional 300t of propellant so I assume they would be taller than 70m?

He said the 9 engine Starship as an extra 300t, not the booster.

As can be seen in the simulation where the heavier Starship is at 1820 t at staging and 354 t at SECO.

[niwax]

Great to see your simulations again. Do you think you could update the old single stage P2P with the new configuration? The last one was still with the old tri-fin carbon fiber version.

[OneSpeed]

https://twitter.com/elonmusk/status/1484012192915677184
10:57 PM · Jan 19, 2022
Tweet Contents:  Simulation of Super Heavy Chopsticks Catch
Maybe something like this

720p video attached below:

[OneSpeed]

Here is a comparison between the recent SpaceX Super Heavy landing simulation, and the Falcon 9 S1 landing telemetry from the recent Transporter-3 mission.

The first diagram is the acceleration plot from the simulation, which is labelled 'Body Acceleration'. In other words, it is the inertial acceleration you would feel on board the booster, including gravity.

Although the resolution of the plots is quite low, and there is no plot of velocity, both velocity and altitude can be inferred from acceleration. This is the second plot.

The Falcon 9 telemetry however, is rotating Earth referenced, and does not include gravity. So, in the third plot I have offset the Super Heavy acceleration by 9.8m/s, and inverted it to match the Falcon 9 negative acceleration.

The SH landing burn starts at a lower velocity than F9, around 260m/s compared to 320m/s. This might be explained by the lower fineness, and hence ballistic coefficient of Super Heavy, giving it a lower terminal velocity.

The SH sim also appears have a 29 engine configuration, like B4. It ignites all nine gimbaling engines, producing 3.2 - 3.8g of initial deceleration. This minimises gravity losses for about 5 seconds until six engines are shut down, and deceleration reduces to around 0.8g for another 12 seconds, before hovering for around 5 seconds while the chopsticks grab the booster. In fractional thrust seconds (a unit I have just invented) that is 9 * 5 / 29 + 3 * 17 / 29 = 3.31.

Meanwhile, Falcon 9 has a 35 second single engine burn, at an average of 1g, with attendant gravity losses. In fractional thrust seconds F9 is 1 * 35 / 9 = 3.88, a larger fraction, so assuming similar throttle settings, less efficient. In short, it appears that even including 5 seconds of hover, the Super Heavy landing burn could be more efficient than the current Falcon 9 profile, helping to justify the deletion of landing legs.

Lastly, Super Heavy might not even have an entry burn. The Falcon 9 entry burn is 24 seconds long, and 18 of those seconds are with three engines lit, for an additional 3 * 18 / 9 + 1 * 6 / 9 = 6.66 fractional thrust seconds.

[volker2020]

Here is a CompariSim™ between two Starships with 6 and 9 engines each.

The Super Heavy boosters both have 33 Raptor2 engines. However, the 9 engine Starship has three extra Vacuum Raptor2s, an additional 300t of propellant, giving a liftoff T/W of 1.4, and does not throttle back for MaxQ.

The simulation suggests that these combined changes increase the payload to a 26° inclination (e.g. Boca Chica) from 150 to 200t.

The payload for Starlink satellites to 53° would be about 6% less, or 188t.

Around 90s into the simulation, shortly before stage separation, Starship does experience 3g in this simulation. Not sure, it is designed for that kind of load.

[eriblo]

Here is a CompariSim™ between two Starships with 6 and 9 engines each.

The Super Heavy boosters both have 33 Raptor2 engines. However, the 9 engine Starship has three extra Vacuum Raptor2s, an additional 300t of propellant, giving a liftoff T/W of 1.4, and does not throttle back for MaxQ.

The simulation suggests that these combined changes increase the payload to a 26° inclination (e.g. Boca Chica) from 150 to 200t.

The payload for Starlink satellites to 53° would be about 6% less, or 188t.

[Video link]

Around 90s into the simulation, shortly before stage separation, Starship does experience 3g in this simulation. Not sure, it is designed for that kind of load.

Any reason why it would not be? That is on the lower end of what is typical for crewed launches (Shuttle limited to  3 g, Saturn V and Soyuz peak at 3.5-4 g and Crew dragon is just above 4 g). Lighter payloads often go higher and the Starship Users Guide specifies loads up to 6 g.

EDIT: Never mind the above, I just realized you were looking at loads during SH flight. That is indeed a good question! Many launchers with large first stages have their maximum acceleration during first stage flight (Saturn V, Soyuz, Atlas V...) but Starship is an extremely large second stage:

If the LCH4 tank on Starship is pressurized to 6 bar and the LOX tank to 2.5-3 bar they would be in tension even at 3 g but the total pressure at the aft dome would approach 9 bar. If this is too much the LOX tank has to take the extra compressive load. There might also be complications if the LOX tank has to start pressing fully for flight before MECO.

If both tanks on SH are pressurized to 6 bar when approaching MECO that alone would support ~3800 t but 3 g on a heavier Starship would still put it into significant compression (3 * 1820 t = 5460 t in the simulation). This might need extra reinforcements/stringers on the LCH4 tank (the LOX tank should already be built to support a higher load on takeoff) which, together with the extra needed in the unpressurized skirt/interstage area, has to be traded versus the losses of throttling down further (extra gravity and RTLS dv).

I think that they have found the extra mass to be worth it if they decide to go for the stretched Starship even if they do not significantly throttle down SH.

[livingjw]

Around 90s into the simulation, shortly before stage separation, Starship does experience 3g in this simulation. Not sure, it is designed for that kind of load.

Four g is reasonable. I did some calculations which showed that the booster could handle it with only stringers in the lox tank.
With the added mass of the second stage they will likely need stringers in the CH4 tank as well. I think we have seen these already.

John

[RealTimeShepherd]

Hi guys, I was wondering about the likely points for beginning and ending of the SuperHeavy entry burn as I'm trying to simulate these flights in KSP with Realism overhaul and as many additional realism mods as I can (Principia, Ferrams etc.), I'm controlling the vehicles with k-OS and I want to create the entire SuperHeavy flight from launch to catch as accurately as possible

I can capture telemetry from these flights and it has taken me such a long time to solve the tower catching problem, I'm desperate to show it off

This is a 'short' so best watched on a phone in portrait format, but it is a genuine single flight of a StarShip model around the world from tower to tower controlled by a script solving simulated physical problems:
https://www.youtube.com/shorts/N6PHWHbtd44

I've also got a (regular aspect ratio) video describing how I finally pulled off the tower catch:


Can you guys help me?

[OneSpeed]

Hi guys, I was wondering about the likely points for beginning and ending of the SuperHeavy entry burn ...

That's a nice video, but why are you sure there will be a SuperHeavy entry burn? A lot of propellant could be saved if the stainless steel fuselage can tolerate deletion of that burn.

[daavery]

Hi guys, I was wondering about the likely points for beginning and ending of the SuperHeavy entry burn ...

That's a nice video, but why are you sure there will be a SuperHeavy entry burn? A lot of propellant could be saved if the stainless steel fuselage can tolerate deletion of that burn.

according to Spacex , deorbit burn , but no entry burn, and then a landing burn

[Hauerg]

Hi guys, I was wondering about the likely points for beginning and ending of the SuperHeavy entry burn ...

That's a nice video, but why are you sure there will be a SuperHeavy entry burn? A lot of propellant could be saved if the stainless steel fuselage can tolerate deletion of that burn.

according to Spacex , deorbit burn , but no entry burn, and then a landing burn

NO deorbit for SuperHeavy since NOT orbital.
No entry burn either, steel and hardened/simplified Raptor 2 should be able to take it with minimal shielding.

[halflife9]

Hi guys, I was wondering about the likely points for beginning and ending of the SuperHeavy entry burn ...

That's a nice video, but why are you sure there will be a SuperHeavy entry burn? A lot of propellant could be saved if the stainless steel fuselage can tolerate deletion of that burn.

according to Spacex , deorbit burn , but no entry burn, and then a landing burn

NO deorbit for SuperHeavy since NOT orbital.
No entry burn either, steel and hardened/simplified Raptor 2 should be able to take it with minimal shielding.

I believe they meant boost back burn by "deorbit burn"

[RealTimeShepherd]

Hi guys, I was wondering about the likely points for beginning and ending of the SuperHeavy entry burn ...

That's a nice video, but why are you sure there will be a SuperHeavy entry burn? A lot of propellant could be saved if the stainless steel fuselage can tolerate deletion of that burn.

according to Spacex , deorbit burn , but no entry burn, and then a landing burn

OK, great to know! I must admit, I was basing it on the Falcon 9 booster re-entry, just assuming it would be the same deal. I'll try again with no entry burn and see what it looks like!
Do you have a link to the info??

[RealTimeShepherd]

Oh! No worries, I found it on twitter from Elon himself

Thanks guys 😁

[soyuzu]

From FAA EA, Page 19

Following a suborbital launch, Starship would have LOX and LCH4 (approximately 10 metric tons)
remaining in the tank

I think this also indicates residule fuels mass of SS in an orbital mission.

[OneSpeed]

Here is a simulation of the first Starship Orbital Flight Test (OFT-1), informed by the recently published Flight Test Timeline and Environmental Assessment re-evaluation.

Some points of interest:
1. The maximum throttle setting will be at 90% of full thrust, and the maximum dynamic pressure (MaxQ) will be at T+0:55s.
2. To limit dynamic pressure, the booster will need to throttle back to 70% from the T+0:50s until the T+1:11s mark.
3. To limit acceleration, the booster will probably gradually throttle back from about T+2:09s to main engine cutoff (MECO) at T+2:49s.
4. The OFT-1 orbital parameters are unusual because the perigee at insertion needs to be in the atmosphere. The simulation orbital parameters are 235km x 80km at an inclination of 26° (the latitude at the Boca Chica launch site).
5. Because there is no payload in this simulation, the residual propellant mass is about 150t. From the Environmental Assessment re-evaluation, Starship will vent all but 14t of propellant. This happens in the simulation at apogee.
6. The Starship re-entry is performed with 70° of pitch.
7. Peak heating is at T+1:21:30, and peak deceleration is at T+1:27:00.
8. In the simulation, Starship splashdown occurs at T+1:30:23, quite close to the Flight Test Timeline best case scenario.

[OneSpeed]

Here is a comparison of the telemetry predicted from the previous OFT-1 simulation, and slightly smoothed telemetry from the first Starship test flight.

Differences include:

1. Reduced acceleration at liftoff. The reduction is very close to the difference between 30 and 33 engines at 90% throttle.
2. A more gradual reduction in throttle before MaxQ.
3. No perceptible throttle up after MaxQ.
4. Almost complete loss of thrust from 145s, and a subsequent tumble before flight termination.

[daveglo]

Elon said the decision to launch with 30 engines was acceptable based on the "minimum required", but at 90% throttle, that obviously had a big impact on expected acceleration and velocity.  Curious what the ultimate booster performance for the mission would have looked like if you re-ran the sim at those launch conditions, making the assumption that the rest of the flight goes as planned from an engine perspective.  I'm guessing it would push out the staging later in the flight and further downrange, since fuel consumption rate would be lower?  Also guessing a small impact on remaining fuel needs for the RTLS maneuver.  But it's not like the Starship didn't have enough fuel to recover if staging is done at underspeed conditions.

So a future test flight, even with a 3-engine-out launch, should still be successful, from a orbital perspective, but the launch event timeline would change.

[DanClemmensen]

Elon said the decision to launch with 30 engines was acceptable based on the "minimum required", but at 90% throttle, that obviously had a big impact on expected acceleration and velocity.  Curious what the ultimate booster performance for the mission would have looked like if you re-ran the sim at those launch conditions, making the assumption that the rest of the flight goes as planned from an engine perspective.  I'm guessing it would push out the staging later in the flight and further downrange, since fuel consumption rate would be lower?  Also guessing a small impact on remaining fuel needs for the RTLS maneuver.  But it's not like the Starship didn't have enough fuel to recover if staging is done at underspeed conditions.

So a future test flight, even with a 3-engine-out launch, should still be successful, from a orbital perspective, but the launch event timeline would change.

Do we know that the engines remained at 90% after the loss of three engines? They could make up the deficit by throttling up to 100% on 30 engines immediately.

[RamsesBic]

Elon said the decision to launch with 30 engines was acceptable based on the "minimum required", but at 90% throttle, that obviously had a big impact on expected acceleration and velocity.  Curious what the ultimate booster performance for the mission would have looked like if you re-ran the sim at those launch conditions, making the assumption that the rest of the flight goes as planned from an engine perspective.  I'm guessing it would push out the staging later in the flight and further downrange, since fuel consumption rate would be lower?  Also guessing a small impact on remaining fuel needs for the RTLS maneuver.  But it's not like the Starship didn't have enough fuel to recover if staging is done at underspeed conditions.

So a future test flight, even with a 3-engine-out launch, should still be successful, from a orbital perspective, but the launch event timeline would change.

Do we know that the engines remained at 90% after the loss of three engines? They could make up the deficit by throttling up to 100% on 30 engines immediately.

No we don't know. We don't even know if an attempt to throttle up above 90% caused 4 or 5 more Raptors to malfunction. The Raptors were not reliable enough, that is the only obvious conclusion. We can only hope the Raptors on B9 are better.

[envy887]

No we don't know.

We can calculate the vehicle acceleration from the velocity telemetry display, and SpaceX has published the vehicle's mass and propellant consumption rates, so calculating the thrust is simple. Based on this data, the engines, at least between liftoff and max-Q, remained at 90%.

[RamsesBic]

No we don't know.

We can calculate the vehicle acceleration from the velocity telemetry display, and SpaceX has published the vehicle's mass and propellant consumption rates, so calculating the thrust is simple. Based on this data, the engines, at least between liftoff and max-Q, remained at 90%.

You could use that telemetry, but how accurate is it? How can you see the difference between those that increase their thrust and those who fail? The switching could be fast enough not to show on the telemetry on the screen. I would not trust it too much anyway. So, I will rephrase, we are not sure we know the level of thrust of every single one of the 33 Raptors at any given time. But that is just my opinion.

[Herb Schaltegger]

No we don't know.

We can calculate the vehicle acceleration from the velocity telemetry display, and SpaceX has published the vehicle's mass and propellant consumption rates, so calculating the thrust is simple. Based on this data, the engines, at least between liftoff and max-Q, remained at 90%.

You could use that telemetry, but how accurate is it? How can you see the difference between those that increase their thrust and those who fail? The switching could be fast enough not to show on the telemetry on the screen. I would not trust it too much anyway. So, I will rephrase, we are not sure we know the level of thrust of every single one of the 33 Raptors at any given time. But that is just my opinion.

I think a lot of this was addressed in the Twitter Spaces talk Elon gave last week (there’s an AI-generated transcript elsewhere on the forum). 

[daveglo]

Do we know that the engines remained at 90% after the loss of three engines? They could make up the deficit by throttling up to 100% on 30 engines immediately.

We indeed do not KNOW, but it's pretty obvious from the data on Onespeed's graph that the acceleration was significantly below what would have been expected for a full 33-engine burn at 90% throttle.  So if the 30 running engines had throttled up to 100%, we should see the acceleration curve closer to Onespeed's predicted curve.  Since it never gets close, I think it's a safe assumption.

The questions about using the telemetry, well, that's exactly what Onespeed does to get his actual data, and he does a darned good job of it.  I'd trust his actual data analysis completely.

It's just that we now know (from the Spaces talk for sure), that a 30-engine launch is considered acceptable, at least for these early tests.  So maybe a simulation of what that kind of flight profile (30 engines @90%) would look like is informative for us rocketgeeks.

[aries1b]

Here is a simulation of the first Starship Orbital Flight Test (OFT-1), informed by the recently published Flight Test Timeline and Environmental Assessment re-evaluation.

Some points of interest:
1. The maximum throttle setting will be at 90% of full thrust, and the maximum dynamic pressure (MaxQ) will be at T+0:55s.
2. To limit dynamic pressure, the booster will need to throttle back to 70% from the T+0:50s until the T+1:11s mark.
3. To limit acceleration, the booster will probably gradually throttle back from about T+2:09s to main engine cutoff (MECO) at T+2:49s.
4. The OFT-1 orbital parameters are unusual because the perigee at insertion needs to be in the atmosphere. The simulation orbital parameters are 235km x 80km at an inclination of 26° (the latitude at the Boca Chica launch site).
5. Because there is no payload in this simulation, the residual propellant mass is about 150t. From the Environmental Assessment re-evaluation, Starship will vent all but 14t of propellant. This happens in the simulation at apogee.
6. The Starship re-entry is performed with 70° of pitch.
7. Peak heating is at T+1:21:30, and peak deceleration is at T+1:27:00.
8. In the simulation, Starship splashdown occurs at T+1:30:23, quite close to the Flight Test Timeline best case scenario.

What software did you use to run this simulation?

[OneSpeed]

What software did you use to run this simulation?

The program is called SpaceSim, and you can download it from https://github.com/zlynn1990/SpaceSim

[native chicken]

What software did you use to run this simulation?

The program is called SpaceSim, and you can download it from https://github.com/zlynn1990/SpaceSim

Can you provide the latest version of patch configuration?
I only found the ones two years ago. https://github.com/JohnnyOneSpeed/SpaceSim/commit/d7e4c5a7c19b821f67cc38d4d774b9c15d9e8c79
thinks.

[OneSpeed]

Here is a 'CompariSim'™ of two Starship Raptor 3 cargo missions to LEO, hot staging on the left, centrifugal on the right.

Both missions utilise the extra performance of the Raptor 3, which in turn enables a higher propellant load, for a GLOW of 6000t.

In order to maximise the chances of hot staging improving payload, I've kept the startup sequence as short as possible, to avoid the ship effectively towing the booster for any length of time. Although the Raptor 3 changes have increased the payload to LEO to 170t, and the hot staging has saved the booster some 2.4t of propellant, the benefit of hot staging to the ship appears to be a wash. The ΔV lost towing the booster cancels out the ΔV gained from reduced gravity losses.

Perhaps Elon Musk was considering some other factor when he said that hot staging could improve payload to orbit by as much as 10%?

Edit: fixed frame.

[InterestedEngineer]

Here is a 'CompariSim'™ of two Starship Raptor 3 cargo missions to LEO, hot staging on the left, centrifugal on the right.

Both missions utilise the extra performance of the Raptor 3, which in turn enables a higher propellant load, for a GLOW of 6000t.

In order to maximise the chances of hot staging improving payload, I've kept the startup sequence as short as possible, to avoid the ship effectively towing the booster for any length of time. Although the Raptor 3 changes have increased the payload to LEO to 170t, and the hot staging has saved the booster some 2.4t of propellant, the benefit of hot staging to the ship appears to be a wash. The ΔV lost towing the booster cancels out the ΔV gained from reduced gravity losses.

Perhaps Elon Musk was considering some other factor when he said that hot staging could improve payload to orbit by as much as 10%?

Edit: fixed frame.

I suspect restarting the engines after the flip costs more startup gas and tankage than your model is accounting for.

With hot staging there's less engines starting after the flip (if any)

[OneSpeed]

I suspect restarting the engines after the flip costs more startup gas and tankage than your model is accounting for.

Why? The restarts are modelled to take 0.5s, at the nominated throttle. What do you think the number should be?
The centripetal staging engines are shut down for around 6 seconds, and that is why the hot staging booster has so much more loft after the flip, and less work to do on the return to the launch site.

With hot staging there's less engines starting after the flip (if any)

We know there are only three engines left running at 50% throttle with hot staging. The boostback burn uses 13 engines, so there are at least 10 to be started after the flip.

[Nevyn72]

I don't really understand the concept of the Ship 'towing' the Booster, what does this mean?

[OneSpeed]

I don't really understand the concept of the Ship 'towing' the Booster, what does this mean?

If the ship engines are running, and most of the booster engines have been shut down, then most of the thrust will be from the ship. The separation point will be in tension, so the ship will be 'towing' the booster.

[Nevyn72]

I don't really understand the concept of the Ship 'towing' the Booster, what does this mean?

If the ship engines are running, and most of the booster engines have been shut down, then most of the thrust will be from the ship. The separation point will be in tension, so the ship will be 'towing' the booster.

This is what I don't understand, why at any stage of the ships' engines running would the clamps still be engaged?

How long in your simulation do you have this occurring for?

[OneSpeed]

This is what I don't understand, why at any stage of the ships' engines running would the clamps still be engaged?

How long in your simulation do you have this occurring for?

That's what hot staging means, the engines are hot before they stage. In the sim, I only hold the clamps for one second, to minimize the losses for the ship.

[JaimeZX]

That's super interesting and I wonder how the sim calculates all of that.  For that one second of "towing," you have the 3x SL engines firing at 50%?  And then what about... cosine losses? How much of the SL Raptor thrust percentage remains after venting out the ports? 

[OneSpeed]

That's super interesting and I wonder how the sim calculates all of that.  For that one second of "towing," you have the 3x SL engines firing at 50%?  And then what about... cosine losses? How much of the SL Raptor thrust percentage remains after venting out the ports?

For that one second of 'towing', and after separation:
3 x Booster SL @ 50%
3 x Ship SL @ 100%
3 x Ship RVac @ 100%

There will be some losses through the vents, but after separation there will also be some reactive forces due to plume interaction with the top of the booster. I have taken my best guess at those forces, using the same physical model as separation pushers in conventional staging. Those forces dissipate quite quickly though.

[sebk]

Perhaps Elon Musk was considering some other factor when he said that hot staging could improve payload to orbit by as much as 10%?

Edit: fixed frame.

I think Tom Mueller (the father of Merlin) is onto something: he tweeted that the likely gain is avoidance of ullage collapse. To fight ullage collapse in a regular SSH staging it could some dozens of tons of extra ullage gas dumped into the tanks. This gas is then wasted mass.

[OneSpeed]

I think Tom Mueller (the father of Merlin) is onto something: he tweeted that the likely gain is avoidance of ullage collapse. To fight ullage collapse in a regular SSH staging it could some dozens of tons of extra ullage gas dumped into the tanks. This gas is then wasted mass.

Isn't ullage gas required anyway to maintain propellant tank pressures? Hence the interest in autogenous pressurisation?

[rsdavis9]

Perhaps Elon Musk was considering some other factor when he said that hot staging could improve payload to orbit by as much as 10%?

Edit: fixed frame.

I think Tom Mueller (the father of Merlin) is onto something: he tweeted that the likely gain is avoidance of ullage collapse. To fight ullage collapse in a regular SSH staging it could some dozens of tons of extra ullage gas dumped into the tanks. This gas is then wasted mass.

Yes in straight flight you can have autogenous gas at 500k on top of 90k liquid propellant. Any slosh and that stratification is destroyed.
So you never want either stage in zero g and when you turn don't turn to fast.

You can calculate the tons that 500k gas at 6 bar weighs and compare that to 150k at 6 bar or so if it has been "collapsed" by slosh.

They have the same problem with the landing flip. Different is that they have all(?) the liquid in the header tanks and main tanks potentially(?) filled with 500k 6 bar gas.

[sebk]

I think Tom Mueller (the father of Merlin) is onto something: he tweeted that the likely gain is avoidance of ullage collapse. To fight ullage collapse in a regular SSH staging it could some dozens of tons of extra ullage gas dumped into the tanks. This gas is then wasted mass.

Isn't ullage gas required anyway to maintain propellant tank pressures? Hence the interest in autogenous pressurisation?

The problem is ullage gas gets rapidly cooled when propellant surface becomes turbulent and blobs and droplets of propellant  start falling through it.

Autogenous gas is normally pretty hot (400-500K). If you cool it down to 130-170K you have just reduced the pressure to 1/3, so you must pretty much doubled its mass.

IOW ullage collapse at staging likely requires severely increased production of autogenous ullage gas and once this gas is produced it's not going to recondense back in the next few minutes. It's thus wasted mass.

[InterestedEngineer]

The tweet in question

https://twitter.com/lrocket/status/1673035595231408128

Question:  Can you add ullage physics to the simulation?

[InterestedEngineer]

Perhaps Elon Musk was considering some other factor when he said that hot staging could improve payload to orbit by as much as 10%?

Edit: fixed frame.

I think Tom Mueller (the father of Merlin) is onto something: he tweeted that the likely gain is avoidance of ullage collapse. To fight ullage collapse in a regular SSH staging it could some dozens of tons of extra ullage gas dumped into the tanks. This gas is then wasted mass.

I have my own spreadsheet simulation.

I can trivially confirm that adding 40t of ullage drops the final orbit payload by about 10%.  About a 2:1 tradeoff. (2t additional ullage => 1t drop in final payload).

I think we finally got the root cause figured out on this one.


https://docs.google.com/spreadsheets/d/1bW0qWPjSl85lYLOwO9m6j9l71uWHWRSR1gMFtuR2bb0/edit#gid=1464826828

[OneSpeed]

Question:  Can you add ullage physics to the simulation?

Answer: Sure, that's a great idea. If someone has a formula that can relate pitch rate, acceleration, propellant remaining, temperature etc. to the amount of ullage collapse, I can add it to the simulation.

Edit: this paper could help: Modeling of Ullage Collapse Within Rocket Propellant
Tanks at Reduced Gravity, Anderson, Chintalapati and Kirk.

[xvel]

Wasn't the separation of stages by spin having the side effect of keeping the fuel at the bottom of the booster?

[Dancing Dog]

Once they complete the boost-back burn, a degree of ullage collapse could be a *good* thing.

Since the landing propellant comes from the header tanks, they only need enough pressure in the main tanks to maintain structural stability, maybe 1-2 bar?. Reducing ullage through controlled collapse, if practical, means less gas that must be vented/more props recycled after landing.

[rsdavis9]

Once they complete the boost-back burn, a degree of ullage collapse could be a *good* thing.

Since the landing propellant comes from the header tanks, they only need enough pressure in the main tanks to maintain structural stability, maybe 1-2 bar?. Reducing ullage through controlled collapse, if practical, means less gas that must be vented/more props recycled after landing.

If ullage collapse in main tanks then the question becomes what happens to header tank pressurization?

[Dancing Dog]

Once they complete the boost-back burn, a degree of ullage collapse could be a *good* thing.

Since the landing propellant comes from the header tanks, they only need enough pressure in the main tanks to maintain structural stability, maybe 1-2 bar?. Reducing ullage through controlled collapse, if practical, means less gas that must be vented/more props recycled after landing.

If ullage collapse in main tanks then the question becomes what happens to header tank pressurization?

I believe the Ship header tanks are pressurized from COPVs; I would be surprised if the Booster headers aren't also pressurized independently.

OTOH dumping the main tanks would reduce the landing props required, so it's just another potential trade. If the "OMG methane release" crowd gets noisy, maybe they'll dump the LOX and let the methane ullage collapse.

[OneSpeed]

Wasn't the separation of stages by spin having the side effect of keeping the fuel at the bottom of the booster?

Yes, that was the intent, but I'm wondering if OFT-1 saw ullage collapse during the earlier than expected tumble, and SpaceX are now convinced the centrifugal staging technique will not prevent ullage collapse, even at the correct staging point. So, they've abandoned it.

[Slarty1080]

I think ullage collapse is a major issue whenever a rocket cuts its engines and the bigger and emptier the tanks the more significant the issue is. The problem is not a simple one. During the boost phase the contents of the propellant tanks are far from equilibrium, the cryogenic liquid at the bottom is being forced into the engine inlets under 6 bar of pressure from a very hot gas at many hundreds of degrees C that is constantly replenished by the engines as the ullage space expands. But when the engine cut and the vehicle goes into free fall there is nothing to stop the cryogenic liquid floating around as blobs and droplets in the hot ullage gas. Any form of maneuvering only makes matters worse by stirring things up.

The hot gases have very low heat capacities compared to liquids so the cryogenic liquid cools the gas and (as the tank volume is constant) the pressure drops. The critical question is how long does this take? But that question is not easy to answer as it depends on the surface area of liquid in contact with the gas and many other things. But this pressure loss or ullage collapse is a major issue, because after the engines cut the only way to re-pressurize the tanks is by using cold COPV gas. Cold gas is a lot denser than hot gas and tens of tonnes would be required to re-pressurize Superheavy's main tanks when they were mostly empty. Hence the header tanks.

[edzieba]

On 'towing': there should be no reason not to unlatch the clamps immediately on - or even just prior to - upper stage startup to eliminate any towing losses. As there is no engine bell to clear from a retained interstage, as soon as separation occurs the upper stage is entirely free of the booster and would require an immediate and very rapid pitchover or yawover to even come close to recontacting (at which point things have gone horribly wrong and still being latched to the booster would be similarly problematic). The protruding guide pins also aid the upper stage in separating without sliding for the first few hundred mm of travel even after the latches are disengaged, as those guide pins are static and do not retract.
tl;dr 'towing losses' should be zero.

I think ullage collapse is a major issue whenever a rocket cuts its engines and the bigger and emptier the tanks the more significant the issue is.

Two extra caveats:
- Ullage collapse is an issue when there is a significant temperature differential between the propellant and ullage gas. Sub-chilled propellants (liquid temperature lowered), autogenous pressurisation (ullage gas temperature raised) both exacerbate this, and Starship and Superheavy add both. For non-autogenous non-subchilled cryogens, there will be an equilibrium head pressure below which ullage gas cannot occur (and this pressure can be used actively, as with VaPak)
- But that only applies after a burn. At the start of a burn (from ground) the ullage gas is Helium (so will not condense from sub-chilled LOX or LCH4), for both Superheavy and Starship.

Or in other words: at staging, Super Heavy has large tanks filled with a very large volume of hot gas (the preoonderence of tank volume and a little superchilled liquid, so very prone to ullage gas when agitated. Starship's main tanks (and headers) will have a small volume of ullage gas and that gas will be Helium (plus some fraction of propellant boiloff), so have greatly reduced vulnerability to pressure loss from ullage collapse - at worse, a drop back down to the initial ground pressure.

[OneSpeed]

On 'towing': there should be no reason not to unlatch the clamps immediately on - or even just prior to - upper stage startup to eliminate any towing losses. As there is no engine bell to clear from a retained interstage, as soon as separation occurs the upper stage is entirely free of the booster and would require an immediate and very rapid pitchover or yawover to even come close to recontacting (at which point things have gone horribly wrong and still being latched to the booster would be similarly problematic). The protruding guide pins also aid the upper stage in separating without sliding for the first few hundred mm of travel even after the latches are disengaged, as those guide pins are static and do not retract.
tl;dr 'towing losses' should be zero.

Firstly, if you unlatch prior to ignition, then by definition, it is not hot staging.

Secondly, the ship will be about 3 times heavier than the booster at separation. So, in order to separate, it will require greater than 3 times the thrust of the 3 booster Raptors at 50% throttle (about 1500t vs 400t of thrust assuming Raptor 3). In other words, it will require all six ship engines to be running at close to full throttle just to separate.

What happens if one or more of the vacuum Raptors fails or is slow to start? The torque while the stages are still in contact, but with the pins unlatched is potentially disastrous. However, if the stages are still latched, then there is still the option to shut down the booster engines, and separate with less than six ship engines running. This would lead to ullage collapse in the booster, and perhaps loss of the booster, but it may still be possible to save the ship.

[sebk]

One thing: the pushback from Starship exhaust onto the booster can't be ignored. Until Starhip is more than ~12m away (or the booster rotates from underneath) the vast majority of the exhaust of the SL Raptors trio will impact the top of the booster (and nearly stagnate there). Expect 50-75% efficiency of the pushback. If Starship blasted full throttle at the moment of latch release, the booster would experience negative g.

For SL Raptor the exhaust expands sideways at about 1/4 of it axial velocity (because at the nozzle exit plane the gas is at local Mach 4; Mach speed at 2000-2100K methalox combusion exhaust (mostly H20, then nearly equal CO2 and CO amounts) is ~0.9km/s, at the same time axial exhaust velocity is ~3.6km/s). So the exhaust cone sides have 4:1 slope, so for them to reach the booster edge ~3m to the side, they need ~12m vertical clearance. Only beyond that the exhaust cone would become wider than 9m Starship diameter and pushback would start to go down (quickly; at 17m it would be below half the initial value, at 24m it would be down to ~1/4, beyond that condensation and trace atmosphere effects would be significant, too).

[eriblo]

I see I agree with sebk:

Just because you unlatch does not mean you separate, the ship will stay put untill the combination of engine thrust and gas pressure in the interstage is larger than the force from the booster.

Speaking of this - when Starship starts its engines they will as a good approximation exert the same force on the top of the booster (and the exhaust will mostly be vented sideways). So if they start more than the three SL Raptors at 50% thrust Super Heavy might momentarily be forced backwards.

As soon as some separation is achieved and the "thrust transfer efficiency" drops you can throttle up Starship while maintaining positivt acceleration on Super Heavy. This might have to be timed in order to avoid contact as the booster swings to the side out of the plume but they might delay that slightly on early flights.

If I were to model it I would throttle the stack to 3x50% SL on the booster, start the same on the ship while dropping the booster to some much lower level and then throttle both up (ship to all engines and booster back to 3x50%) as the separation grows.

Only question is what function to use for throttle vs separation and when to start turning the booster.

[JaimeZX]

No doubt SpaceX will model that heavily and then update their numbers based on the next flight.
(Assuming there are no issues prior to stage separation this time.)

[OneSpeed]

Here is a Google Earth flyover of the IFT-2 ground track, at 30x actual speed, viewed from mostly the same altitude as the flightpath.

[OneSpeed]

Here is a 'CompariSim' of two Starship Raptor 3 cargo missions to LEO, informed by Elon Musk's recent talk at the Baku IAC, where he said that thrust at liftoff will be 20M lbs or 9072t. Given that SuperHeavy will have 33 Raptor 3 engines, 9072t / 33 = 275t of thrust per engine at sea level. He also said that the initial thrust to weight will be between 1.3 and 1.4. Taking the mean value of 1.35, the total propellant load will be increased for a GLOW of around 6,720t.

Musk also said that the ratio of booster to ship propellant will decrease to 2:1 (4120t:2060t), which is modelled by the Starship on the right in the video. I've extended the Starship to 64m in length, and kept the SuperHeavy at 72m, for a total length of 136m.

Then Musk said something that is completely contradictory, which was that the booster burn time would approach 100s. This would imply a prop ratio closer to 1:1 (actually 2900t:3300t), which I've modelled on the left. I've extended that Starship to 76m in length, and shortened the SuperHeavy to 60m, for the same total length of 136m.

A 100s burn time will mean staging at only 1,000 m/s (Mach 3), and that the ship will have an initial T/W ratio of only 0.55. For this reason, the 1:1 ship will fail to maintain its ascent, and will re-enter the sensible atmosphere at about Mach 5. With suitable flight path and body flap angles, the ship will 'bounce' or 'skip', as was implied in the Starship point to point proposal. From there it would theoretically continue to accelerate and climb to around 90km before SECO, and insertion to a relatively elliptical 90km x 300km coast orbit. Circularisation on the first orbit would be a priority.

The more conventional 2:1 Starship would stage after 148s at 1,840 m/s, and the initial ship T/W would be about 0.9. The ascent profile would be far more conventional, with SECO at an altitude of around 150km.

The payloads achievable for both of these configurations were impressive, with the sim estimating 260t for the 2:1 variant, and 240t for the 1:1 variant. The advantage of the 1:1 variant is that a ship with a far larger propellant capacity (3300t) would be delivered to LEO, enabling missions requiring higher ΔV than the 6km/s predicted for earlier Starship designs. If the ship dry mass was to increase to say 140t, then even with a 240t payload, the ΔV could be around 8km/s, and with no payload 11.5km/s.

[niwax]

Very interesting, it's surprising to see such a low TWR ship lose almost no upmass but gain so much capability once in orbit.  Unintuitive, but it's exactly the thing they are looking for, a very deep space capable vehicle that can serve as an Earth-focused launcher as well. And it makes GTO missions easier.

I would assume this is also much more sensitive to ship TWR? What engine configuration are you using for the ship? They could still go up to 9 engines.

[OneSpeed]

Very interesting, it's surprising to see such a low TWR ship lose almost no upmass but gain so much capability once in orbit ...

Yes, the ability of the Starship to perform a hypersonic glide could be truly revolutionary.

... What engine configuration are you using for the ship? They could still go up to 9 engines.

Yep, nine engines. The video displays the Starship SESU event as 3 Centre and 6 RVac. There are a lot of events in quick succession though, it's easy to miss

[daveglo]

Fascinating study, Onespeed!  I can visualize SpaceX having done something similar, with a range of SS-to-SH ratios, to see how they optimize.  Of course, as they improve the Raptor thrust levels, the optimization changes.

But the Baku2 configuration is entirely unexpected!  Would they really want to put the vehicle through an extended hypersonic phase like that?  VERY ballsy, but then again, kind of their method!  Certainly lowers the re-entry impact on the booster.  Staging at 15 miles up/downrange will be visually exceptional.  And getting a ground-observer view of that hypersonic phase would be a speed-junkie dream! 

Thanks for doing this!

[mikelepage]

Fascinating study, Onespeed! I can visualize SpaceX having done something similar, with a range of SS-to-SH ratios, to see how they optimize. Of course, as they improve the Raptor thrust levels, the optimization changes.

But the Baku2 configuration is entirely unexpected!  Would they really want to put the vehicle through an extended hypersonic phase like that?  VERY ballsy, but then again, kind of their method!  Certainly lowers the re-entry impact on the booster.  Staging at 15 miles up/downrange will be visually exceptional.  And getting a ground-observer view of that hypersonic phase would be a speed-junkie dream! 

Thanks for doing this!

Indeed. Great work! I'd sure there's value in a slightly more optimised version?: I could imagine some ratio of SS to SH that kept Starship's altitude more or less flat, instead of bouncing back into the atmosphere. If I was reading it right, the Baku2 variant was 600km downrange before it cleared 50km in altitude, and "bottomed out" at 26km. I was reminded of a conversation I had with Onespeed about sonic booms we had. I had to go back and find the reply from 4 years ago:

I wondered if you had any estimates on which (if any) of those skipping events would generate sonic booms audible on the ground? But it seems to me it goes subsonic whilst still 25km altitude, which should minimise any complaints on the ground.  (Shuttle was still Mach 1.5 at 18km).

Answering your post here because the answer might have more general applicability. From:

https://www.nasa.gov/centers/armstrong/news/FactSheets/FS-016-DFRC.html

here are some more data points for supersonic aircraft and the Space Shuttle (STS):

Aircraft	Mach	Altitude(ft)	Altitude(m)	Pressure(lb/ft²)	Pressure(Pa)
SR-71 Blackbird	3.2	80,000	24,000	0.9	43
Concorde SST	2	52,000	16,000	1.94	93
F-104 Starfighter	1.93	48,000	15,000	0.8	38
Space Shuttle	1.5	60,000	18,000	1.25	60
XB-70	1.5	37,000	11,000	2.5	120

<snip>

So, for every ricochet, the overpressure felt at sea level will be about 42 Pa, or 0.9 lb/ft², just under that for the SR-71 Blackbird figure above. The impulse will be greatest at the highest velocity, as will the duration. For people on the ground, the overpressure figure is the most important in terms of perceived volume, and would be quite a bit less than for Concorde.

I think the sonic boom from the "bounce" at 26km and 200km downrange might create unneeded regulatory hurdles for them, but if they keep it > (say) 35-40km, that issue would be much smaller.

[InterestedEngineer]

Very interesting, it's surprising to see such a low TWR ship lose almost no upmass but gain so much capability once in orbit ...

Yes, the ability of the Starship to perform a hypersonic glide could be truly revolutionary.

... What engine configuration are you using for the ship? They could still go up to 9 engines.

Yep, nine engines. The video displays the Starship SESU event as 3 Centre and 6 RVac. There are a lot of events in quick succession though, it's easy to miss

what did you use for non-payload orbital mass after adding all that tankage?

my estimate is 113t of fuel per ring and each ring masses about 1.1t dry plus .1t of landing fuel or 1.2t .  So the 2060t version has 8 extra rings and non-payload orbital mass increases by ~10t.

What are your numbers?

[RedLineTrain]

Nifty.  Interesting that at these low thrust to weight ratios on the ship, they would lose some abort capabilities.  I also wonder how much safety margin they would lose on engines out, etc.

When you refer to the SESU event, to which event are you referring?

[niwax]

... What engine configuration are you using for the ship? They could still go up to 9 engines.

Yep, nine engines. The video displays the Starship SESU event as 3 Centre and 6 RVac. There are a lot of events in quick succession though, it's easy to miss

Ah, I went by the 3300t mass and TWR which only comes out to ~200t per engine, but I missed the dry mass and payload. With those, it's still only ~225t per engine on average, so I guess there is still a bit to be gotten there but not much.

[OneSpeed]

But the Baku2 configuration is entirely unexpected!  Would they really want to put the vehicle through an extended hypersonic phase like that?

I was as surprised as you were when I first ran the Baku2 sim. With a 100s booster burn, the ship had no other choice but to glide!

I think the sonic boom from the "bounce" at 26km and 200km downrange might create unneeded regulatory hurdles for them, but if they keep it > (say) 35-40km, that issue would be much smaller.

True, but at least it would be in the NGA Launch Hazard Area, where sonic booms should already be expected.

my estimate is 113t of fuel per ring and each ring masses about 1.1t dry plus .1t of landing fuel or 1.2t .  So the 2060t version has 8 extra rings and non-payload orbital mass increases by ~10t.

What are your numbers?

The same. For the 2:1 configuration, I had 160t for the booster, 130t for the ship. For the 1:1 configuration, I had 140t for both the ship and the booster. SWAGs.

Nifty.  Interesting that at these low thrust to weight ratios on the ship, they would lose some abort capabilities.  I also wonder how much safety margin they would lose on engines out, etc.

Yep, Musk did say "trend towards 100s or so", so the 1:1 sim is probably the edge case, and reality will be somewhere between 2:1 and 1:1.

When you refer to the SESU event, to which event are you referring?

Second Engine Start Up, meaning the ship ignition at hot staging.

Ah, I went by the 3300t mass and TWR which only comes out to ~200t per engine, but I missed the dry mass and payload. With those, it's still only ~225t per engine on average, so I guess there is still a bit to be gotten there but not much.

I'm assuming the RVacs have less thrust than the SL engines, but higher Isp. At one point I think there was mention of reducing the RVac throat diameter to enhance the Isp, at the expense of thrust. A 225t per engine average for the ship wouldn't be all that far off.

[eriblo]

The numbers currently on the SpaceX website are 230 t_f for SL and 258 t_f for RVac which lines up fairly well with the I_sp split.

[OneSpeed]

Here is a 'CompariSim'™ of the first two Starship Integrated Flight Tests (OFT-1 and IFT-2), and Crew-7.

Attached are plots from the IFT-2 webcast telemetry, and this simulation, which utilises the IFT-2 events timeline, including all engine start-ups and shutdowns. To produce the simulations, I had to vary the pitch and throttle settings until they matched the telemetry. The fact that the real and simulated plots are so similar, suggests that there were no significant drop-outs or errors in the IFT-2 webcast telemetry.

The simulation confirms that the pitch was ≈ 30° at hot staging, and AoA only 0.2°. From that we can conclude that free-fall would have been -sin(30) or -0.5g on the plots. We can see that the booster experienced ≈ -0.7g for a couple of seconds, so around -0.2g in the reference frame of the booster, enough to cause propellant slosh.

Increasing the booster throttle to around 80% during those couple of seconds may well be sufficient to ameliorate the slosh (it works in the sim ).

[ATPTourFan]

Thank you, OneSpeed, for these excellent sims. So informative!!

[daveglo]

Really appreciate the confirmation (via a darn good sim) that:

1) the booster probably did see negative g, and the associated propellant slosh, which likely contributed to the loss of engines during the boostback burn, and

2) that there is a reasonable expectation that this staging method can certainly be tuned via engine settings to improve the odds for success.

Nice work as always!