Starship Users Guide

[ZachF]

Most of the dV savings for lower latitudes comes from inclination changes. The savings on 28.6d vs 26d is about 57m/s.

This normally wouldn't be huge, but on a GTO trajectory the Starship itself will be like 85%+ of the mass.

I thought the only dv savings were the differences in rotational boost when you're launching due east.  Doing the latitudes a bit more carefully, I get 10 m/s.  How do you get 57?

Because to get to geostationary, you need to do an inclination change to 0 degrees.

This is why Ariane is such a good launcher to GTO.... going to a 200 x 35,786 from 6 degrees only takes a further 1500m/s to get to GEO, while 200 x 35,786 from 27 degrees takes 1800m/s. The boost from rotation is small compared to the gain from needing a smaller eventual inclination change.

https://gtocalc.github.io/

[warp99]

Starship will get a small boost for GTO payloads when launching from Boca Chica vs the Cape (28.6d vs 26.0d), about 65-70m/s.

Not a huge boost, but enough to theoretically add ~3 tonnes to the GTO-1800 payload vs the cape.

Does that include the losses from the dogleg it has to do to get out of the Gulf of Mexico without overflying anyone?

What dogleg?

I don't understand that trajectory.  It seems like it is curving to the north?  Or maybe that is just a trick of perspective. 

What orbital parameters did you use for this visualization?

I used google earth with a P2P ruler line between those two points and adjusted the atlantic point so that the line overflew basically nothing.  The curve is just a projection artifact.

I changed the angle to make it look straighter :-)

EDIT:  Added picture.  Now it is straight for everyone.

Is there not an issue with the fact that this trajectory is at 96 degrees from true North so at 6 degrees inclination to the equator when it needs to be at least 26 degrees when launching from Boca Chica at 26 degrees North?

[ZachF]

Starship will get a small boost for GTO payloads when launching from Boca Chica vs the Cape (28.6d vs 26.0d), about 65-70m/s.

Not a huge boost, but enough to theoretically add ~3 tonnes to the GTO-1800 payload vs the cape.

Does that include the losses from the dogleg it has to do to get out of the Gulf of Mexico without overflying anyone?

What dogleg?

I don't understand that trajectory.  It seems like it is curving to the north?  Or maybe that is just a trick of perspective. 

What orbital parameters did you use for this visualization?

I used google earth with a P2P ruler line between those two points and adjusted the atlantic point so that the line overflew basically nothing.  The curve is just a projection artifact.

I changed the angle to make it look straighter :-)

EDIT:  Added picture.  Now it is straight for everyone.

Is there not an issue with the fact that this trajectory is at 96 degrees from true North so at 6 degrees inclination to the equator when it needs to be at least 26 degrees when launching from Boca Chica at 26 degrees North.

A direct east trajectory from BC is going to look something like this:

[cuddihy]

So you’re going to cross US-1 somewhere in the Keys, then. Where would the dogleg take place to avoid this?

[freddo411]

So you’re going to cross US-1 somewhere in the Keys, then. Where would the dogleg take place to avoid this?

Is this necessarily a problem?   You are fly over shipping lanes as well that might have a higher person-density than the roadway connecting the keys.

[TheRadicalModerate]

A direct east trajectory from BC is going to look something like this:

Yup, that's what I got, too.  Note that it goes over the keys, and it's not incredibly far from Homestead.  So this trajectory may have range safety issues.  However, if you make the trajectory go just south of Key West, the azimuth is about 93º, and the extra delta-v to GEO is 5 m/s.  I suspect that that's tolerable.

BTW:  If you go by the FH trajectories that warpgg posted, you reach orbit at about the same longitude as the Keys, so it's possible that the impact point for late-stage failures is moving so fast that you can fly over the Keys and hit the 1/10,000 risk to the public with no problem.

The other issue, though, is matching 28.6º for refueling launches from both pads.  That requires an azimuth of 103º, which requires overflying western Cuba before burnout, or azimuth 77º, which overflies Tampa before burning out almost exactly over Canaveral.  However if you launch due east (90º) and then dogleg south about 670 km out, you should be able to skirt the northern coast of Cuba and hit 28.6º.  (Note that warpgg had also looked at a dogleg just south of Key West, but I think you can do better than that.)

Note:  I'm doing all of this graphically, drawing lines in Google Earth, based on knowing the needed azimuth from BC for a given inclination.  So I have no idea how much a dogleg 670 km out costs.  (FH speed that far downrange is about 4000 m/s.)  I'm guessing it'll be close to 200 m/s.  It might be more optimal to instead launch both the KSC and BC stuff to 30.8º, which would allow the BC launches to go south of Cuba with no doglegs.

Another obvious question is whether you actually need refueling launches from both pads.  Building up prop in your aggregation tanker (i.e., the tanker that actually docks with the payload Starship and does one refueling operation) will incur some amount of boil-off, so aggregating as quickly as possible counts for something.  There's also the military to consider, but I suspect that they're going to hate the idea of refueling until the risk gets quantified to within an inch of its life.

[warp99]

Is there not an issue with the fact that this trajectory is at 96 degrees from true North so at 6 degrees inclination to the equator when it needs to be at least 26 degrees when launching from Boca Chica at 26 degrees North.

A direct east trajectory from BC is going to look something like this:

OK - so it looks like the dog leg is the other way around so launch south east into the Gulf at something like 110 degrees until you are roughly level with a line 30 km off the north coast of Cuba then turn to a heading close to 90 degrees to follow a similar trajectory to the one you have shown but further south.  The change in heading is larger at 20 degrees but it occurs well before MECO and the velocity is much lower at perhaps 2000 m/s so the overall loss of delta V will be lower than doing the turn further out close to the Keys. 

This misses all the population centers by a wide margin and can track over the center of Inagua Island which is a wildlife sanctuary so almost completely unpopulated. 

To match the inclination of a launch from Canaveral do not fly quite so far south by launching at say 107.5 degrees and then do a dogleg to 92.5 degrees.

[niwax]

Note:  I'm doing all of this graphically, drawing lines in Google Earth, based on knowing the needed azimuth from BC for a given inclination.  So I have no idea how much a dogleg 670 km out costs.  (FH speed that far downrange is about 4000 m/s.)  I'm guessing it'll be close to 200 m/s.  It might be more optimal to instead launch both the KSC and BC stuff to 30.8º, which would allow the BC launches to go south of Cuba with no doglegs.

That distance poses the question what actually counts as a dangerous overflight. Boca Chica - Florida is ~1500km. During normal launches, the NOTAM goes nowhere near the 600km droneship. For the polar dogleg, they're avoiding highly populated Cuba at ~700km but then have no problem overflying South America at ~2000km.

[warp99]

Note:  I'm doing all of this graphically, drawing lines in Google Earth, based on knowing the needed azimuth from BC for a given inclination.  So I have no idea how much a dogleg 670 km out costs.  (FH speed that far downrange is about 4000 m/s.)  I'm guessing it'll be close to 200 m/s.  It might be more optimal to instead launch both the KSC and BC stuff to 30.8º, which would allow the BC launches to go south of Cuba with no doglegs.

That distance poses the question what actually counts as a dangerous overflight. Boca Chica - Florida is ~1500km. During normal launches, the NOTAM goes nowhere near the 600km droneship. For the polar dogleg, they're avoiding highly populated Cuba at ~700km but then have no problem overflying South America at ~2000km.

That makes sense as they are reaching orbit around 1500 km out.  Yes there will be a brief moment when the instantaneous engine out trajectory paints across South America but operating the self destruct would ensure that very few sizable pieces of S2 made it through re-entry.  Probably just the engine minus the extended bell. 

There is a similar situation during a normal geosynchronous launch when the engine out trajectory paints across Africa very briefly.

[ulm_atms]

Ok, try this.

Here is all the three western hemisphere GTO launch trajectories based on what I have seen (red lines).

Boca has two.  One (green) would be normal if flyover didn't matter in Florida and you were going for lowest inclination.  Purple is the path with minimal flyover of land.

For green, the inclination is slightly less then the Cape.  The purple, it slightly more then the Cape due to the angle.

If I did something wrong please explain...but these orbits are perfect circles and do go back over the starting spot if drawn around the globe completely.

NOTE:  If this needs to be in a different thread mods...please move..this is just the thread I saw this come up in.

[philw1776]

A direct east trajectory from BC is going to look something like this:

Yup, that's what I got, too.  Note that it goes over the keys, and it's not incredibly far from Homestead.  So this trajectory may have range safety issues.  However, if you make the trajectory go just south of Key West, the azimuth is about 93º, and the extra delta-v to GEO is 5 m/s.  I suspect that that's tolerable.

BTW:  If you go by the FH trajectories that warpgg posted, you reach orbit at about the same longitude as the Keys, so it's possible that the impact point for late-stage failures is moving so fast that you can fly over the Keys and hit the 1/10,000 risk to the public with no problem.

The other issue, though, is matching 28.6º for refueling launches from both pads.  That requires an azimuth of 103º, which requires overflying western Cuba before burnout, or azimuth 77º, which overflies Tampa before burning out almost exactly over Canaveral.  However if you launch due east (90º) and then dogleg south about 670 km out, you should be able to skirt the northern coast of Cuba and hit 28.6º.  (Note that warpgg had also looked at a dogleg just south of Key West, but I think you can do better than that.)

Note:  I'm doing all of this graphically, drawing lines in Google Earth, based on knowing the needed azimuth from BC for a given inclination.  So I have no idea how much a dogleg 670 km out costs.  (FH speed that far downrange is about 4000 m/s.)  I'm guessing it'll be close to 200 m/s.  It might be more optimal to instead launch both the KSC and BC stuff to 30.8º, which would allow the BC launches to go south of Cuba with no doglegs.

Another obvious question is whether you actually need refueling launches from both pads.  Building up prop in your aggregation tanker (i.e., the tanker that actually docks with the payload Starship and does one refueling operation) will incur some amount of boil-off, so aggregating as quickly as possible counts for something.  There's also the military to consider, but I suspect that they're going to hate the idea of refueling until the risk gets quantified to within an inch of its life.

In winter I'm just south of Tampa on FL West coast. Issuing written permission for SpaceX to overfly my property.

[TheRadicalModerate]

OK - so it looks like the dog leg is the other way around so launch south east into the Gulf at something like 110 degrees until you are roughly level with a line 30 km off the north coast of Cuba then turn to a heading close to 90 degrees to follow a similar trajectory to the one you have shown but further south.  The change in heading is larger at 20 degrees but it occurs well before MECO and the velocity is much lower at perhaps 2000 m/s so the overall loss of delta V will be lower than doing the turn further out close to the Keys. 

There's no dog leg. But if you launch due east (90º), you go over the Florida Keys (see his map).  If you launch at 93º, you miss the populated parts of the Keys, and likely are in orbit before overflying populated parts of the Bahamas.

[jthygesen]

Just a small note about the Users Guide: The author is registered in the meta data as Jessica Jensen, whom I believe is director of Dragon Mission management.

Could be a real sign of SpaceX moving personnel from Dragon to StarShip. Or not.

[[email protected]]

Just a small note about the Users Guide: The author is registered in the meta data as Jessica Jensen, whom I believe is director of Dragon Mission management.

Could be a real sign of SpaceX moving personnel from Dragon to StarShip. Or not.

FYI she's indeed assigned to the Starship program when she inaugurated the B1035 core on Space Center Houston. Her position is the director of Starship mission hardware & operations

https://m.facebook.com/groups/2387776317?view=permalink&id=10158427094176318

And I'm sure the workforce transition from Falcon & Dragon to Starship is already happening. For instance, Lars Blackmore who's the main person behind the amazing Falcon landings, his future projects is leading the Starship entry & landing (which is cool, because he already have an experience on Mars landing in JPL)

http://www.eng.cam.ac.uk/news/alumni-stories-meet-principal-rocket-landing-engineer-spacex

[Twark_Main]

OK - so it looks like the dog leg is the other way around so launch south east into the Gulf at something like 110 degrees until you are roughly level with a line 30 km off the north coast of Cuba then turn to a heading close to 90 degrees to follow a similar trajectory to the one you have shown but further south.  The change in heading is larger at 20 degrees but it occurs well before MECO and the velocity is much lower at perhaps 2000 m/s so the overall loss of delta V will be lower than doing the turn further out close to the Keys. 

There's no dog leg. But if you launch due east (90º), you go over the Florida Keys (see his map).  If you launch at 93º, you miss the populated parts of the Keys, and likely are in orbit before overflying populated parts of the Bahamas.

The criteria isn't about "overflying," it's when the instantaneous impact point passes over populated areas.

[TheRadicalModerate]

There's no dog leg. But if you launch due east (90º), you go over the Florida Keys (see his map).  If you launch at 93º, you miss the populated parts of the Keys, and likely are in orbit before overflying populated parts of the Bahamas.

The criteria isn't about "overflying," it's when the instantaneous impact point passes over populated areas.

Fair enough, but the impact point is moving downrange so quickly near burnout that the probability of impact anywhere in the Bahamas is extremely low.  Beyond that, the only areas of the Bahamas along the track are the mangrove swamps of South Andros, the southern tip of Exuma and a mile-wide strip of Long Island, half of which is a golf course.

[john smith 19]

Imo it would be better to change the sat design itself. Instead of using the best of the best components, go down 2 steps in capability (for example solar array efficiency) and make the sats cost only a fraction of today‘s expensive sats. Of course they would weigh then perhaps the double, but with cheap transportation, it could reduce the cost of space „stuff“ significantly. No longer the need to go for ultra lightweight with expensive components and manufacturing. That‘s the real benefit imo of (if successful) Starship and perhaps New Glenn.

An observation first made in the 70's at several shuttle conferences on space servicing.

Why do you need triple redundant systems when if it fails just recover it and bring it in for servicing?

[john smith 19]

By publishing this guide now, SpaceX is creating some level of commitment to a baseline specification for engineers and designers to engineer for. Deviating from these figures in a detrimental manner would be harmful for their business.

Correct.

[john smith 19]

I'm late to this party as I simply didn't expect them to have written a user manual this early.

So what' we seem to be looking at is a replaceable nose, rather like the SR71, which had several.

Fairchild did this in the late 40s/early 50s and the Sikorsky Skycrane helicopter also had a cargo module (never sure how many of those were actually built).

Obvious benefit is the "payload module" (it's a lot more than a fairing now) is a lot lighter than the fully load SS to crane up to the top. Say 10t for the module. That's 160t to move about 140m into the sky.

Two variants, cargo and passenger carrier with the implication that the cargo carrier becomes a tanker with additional tanks in the cargo area.  I'm guessing a trade off between what's left in the main tanks to refuel the payload carrying SS and what additional can be carried in the payload bay tanks (perhaps pressurant gas to speed up the expulsion?).

Vertical integration has to be in because they'll never get the biggest (most expensive) NSS payloads without it.

This all makes perfect sense to me.

As for the passenger carrier worries about windows SS comes in a lot like shuttle, very high AoA. I suspect it will fly that way to the moon as well to protect the window. Of course the downside is the belly TPS will get hit by MMOD, but I'm guessing the plan a short period in LEO before setting off and (I'm guessing here) MMOD levels drop a lot once reasonably far away from earth (like above GEO for example).

I also suspect there will be fast acting cover system to seal a window if it's blown in. Operating with the speed of an airbag during a car crash (might even be a similar design, but non deflating, to give time to evacuate the area and plan repairs.

The replaceable nose also makes P2P cargo or passenger management (in principle) much simpler. Load everyone on at ground level then lift the whole package to the nose.  A 20m long PM is nothing in terms of height to putting several hundred passengers on at full height one at a time. You can leverage most of what's been learned about the process in the last 90 years of commercial airline operations.


And if the wind picks up those aero surfaces will make great sails. Being inside the module will get quite interesting.  Like being in a cable car at a very windy location for example. I've never had that experience, but I'll bet someone here has.

And then of course we get to the join between the PM and the rest of SS. That will have to be made and broken for every flight.

BTW that plan with the trunions mounting to the walls of the PM? Failure to standardize those was a major cause of why shuttle payload integration was a PITA,

Perhaps SX should contact Hempsell Astronautics. They did quite a lot of work on this subject.