Starship mass budget

[SpacegeekV]

SpaceX is well known for its openness, shearing technical details, putting on public display its open-air production facilities. However, the most important parameter of aerospace systems, the mass properties, are rarely discussed. Beyond few numbers dropped by Elon Musk, the detailed masses (weights) of the test articles are only estimations from third-party enthusiasts. They vary depending on how optimistic or not is the estimator.

The decision to switch on welded stainless steel and the pressure-stabilized tank was the tipping point toward fast prototyping and agile progress. As usual, Elon Musk does not refer to its predecessors – the first American ICBM and later space launcher Atlas and upper stage Centaur still in use by Launch alliance. These two rockets show the unprecedented weight (of tanks) to propellant ratio.

Like Atlas, the Superheavy is basically a flying tank. And have a few problems achieving the necessary structure/propellant mass ratio. In contrast, the Starship has large portions of its structure to use semi-monocoque design – payload cone, engine bay, and most important flaps. The welded steel is not the best material for such kinds of structures. Easy observable loss of stability of the thin shell is not a cosmetic issue but a sign of structural problems.

In one year only, SpaceX made remarkable improvements in welding technics and shaping sheet metal. The new nosecone has shown in the latest interview looking as seamlessly welded as the Centaur stage. The capacity for further improvements is possibly exhausted. In my view, the “strength of materials” is a significant obstacle to the ambitious goals of Elon Musk. Looking peculiar arrangement of hanging the rocket is an attempt to tweak the problem. Structural engineers always prefer tension vs compression in order to avoid buckling.

Comprehensive structural analysis is not possible but a simple comparative analysis demonstrates how far is Starship from the indicative weight budget. I am attaching a table comparing Shuttle and Starship. Other possible references might be the Atlas main tank, Centaur upper stage and also few aircrafts made of stainless steel – XB-70 Valkyrie and MIG 25/31.
Anyone who has reliable data on the weight budget of Starship, please write in the comments.

[Jim]

Way off base.  The Shuttle orbiter only carried only 14 tonnes of propellant.  It is not possible to compare the two, since the Shuttle used SRBs and engines that burned the whole way to orbit.

[c4fusion]

Actually, in terms of mass fractions it’s actually not that far off of normal upper stages. The Centaur is around 9%, the EPS on Ariane 5 is around 11%, Falcon 9 second stage is about 4-6%, and Starship using your numbers is 8.5%.

That alone is extremely incredible, since it has to carry the extra weight of the fairing, TPS, flaps, and be robust enough to be reused multiple times.

[Stan-1967]

Overall this is not going to be helpful comparison.  Musk is certainly aware of the use of stainless on historic Atlas & Centaur vehicles, as well as the very good weight fractions that ballon tanks can deliver.   The previous vehicles ( Atlas/Centaur) decision to use stainless is not necessarily why he moved to SS/SH stainless, so why should be feel compelled to credit those designs?  The use of stainless on SS/SH was decided by the superiority of stainless at high temperatures vs. composites.  Ballon tanks do not have this reqirement.  Stainless traded well at cryogenic temperatures against compisites,  but not as nearly as decisive as the benefits at high temperature.

Dry mass estimates may be guesses, but my observation for many of the reliable posters here in NSF & elsewhere, is that pretty good mass estimates can be achieved with what is public knowledge.  If engine thrust & ISP are known, and the makers of the vehicle are claiming capability to put specific payload mass into known orbits, the physics puts a pretty tight bracket around the uncertainty.  The uncertainty margin for enthusiastic assumptions gets shrunk to the estimates for aero losses, gravity drag, & prop residuals.

[SpacegeekV]

"The Centaur is around 9%, the EPS on Ariane 5 is around 11%, Falcon 9 second stage is about 4-6%, and Starship using your numbers is 8.5%." Exactly. The simple question is: how starship will achieve the same EPS as Centaur having in mind an additional weight of ~1000 m³ payload bay, engine bay, thermal protection, flaps with powerful mechanism and batteries, a propellant for landing, additional tanks and plumbing for landing propellant, landing gear etc.?

Starship is built by virtually the same SS 301/304 as Centaur. The structural design and detailing are very common. Even bullheads have the same shape as early Centaur. So from where comes the extreme structural efficiency. From twits?

http://www.b14643.de/Spacerockets_2/United_States_3/Centaur/Centaur%20A.jpg 

[Tommyboy]

Let's go back to first principles; what do we want to optimize for, mass fractions or cost per ton to orbit?
If cost is what we want to optimize for, why the obsession for the mass fraction?

[SpacegeekV]

Way off base.  The Shuttle orbiter only carried only 14 tonnes of propellant.  It is not possible to compare the two, since the Shuttle used SRBs and engines that burned the whole way to orbit.

The proposed parallel is addressed to engineering intuition.

[Jim]

Way off base.  The Shuttle orbiter only carried only 14 tonnes of propellant.  It is not possible to compare the two, since the Shuttle used SRBs and engines that burned the whole way to orbit.

The proposed parallel is addressed to engineering intuition.

It doesn't work as a parallel, because the orbiter is the payload and not a launch vehicle stage. It is not like Starship.  If you want a relevant parallel, look at Agena when it is used as spacecraft bus

[envy887]

Starship has large portions of its structure to use semi-monocoque design – payload cone, engine bay, and most important flaps. The welded steel is not the best material for such kinds of structures. Easy observable loss of stability of the thin shell is not a cosmetic issue but a sign of structural problems.

Are you talking about dents that pop out when it's pressurized? I would not necessarily conclude that those represent structural problems, since the loads while unpressurized are very different than when pressurized.

Musk claims that Starship is already around 100 tonnes with Ship 20. There will undoubtedly be incremental improvements in future ships, but those are not really necessary for the system to perform. 100 t is fine.

[SpacegeekV]

Starship has large portions of its structure to use semi-monocoque design – payload cone, engine bay, and most important flaps. The welded steel is not the best material for such kinds of structures. Easy observable loss of stability of the thin shell is not a cosmetic issue but a sign of structural problems.

Are you talking about dents that pop out when it's pressurized? I would not necessarily conclude that those represent structural problems, since the loads while unpressurized are very different than when pressurized.

 with Ship 20. There will undoubtedly be incremental improvements in future ships, but those are not really necessary for the system to perform. 100 t is fine.

Where Musk claims that Starship is already around 100 tonnes? I hope he do not such claims in front of his investors.

[su27k]

"The Centaur is around 9%, the EPS on Ariane 5 is around 11%, Falcon 9 second stage is about 4-6%, and Starship using your numbers is 8.5%." Exactly. The simple question is: how starship will achieve the same EPS as Centaur having in mind an additional weight of ~1000 m³ payload bay, engine bay, thermal protection, flaps with powerful mechanism and batteries, a propellant for landing, additional tanks and plumbing for landing propellant, landing gear etc.?

Starship is built by virtually the same SS 301/304 as Centaur. The structural design and detailing are very common. Even bullheads have the same shape as early Centaur. So from where comes the extreme structural efficiency. From twits?

There's a mass breakdown estimate in the Starship specs - weight, volumes, etc thread, total mass for tanks and engines is estimated to be just 56.1t, you can try to figure out the rest yourself.

[SpacegeekV]

"The Centaur is around 9%, the EPS on Ariane 5 is around 11%, Falcon 9 second stage is about 4-6%, and Starship using your numbers is 8.5%." Exactly. The simple question is: how starship will achieve the same EPS as Centaur having in mind an additional weight of ~1000 m³ payload bay, engine bay, thermal protection, flaps with powerful mechanism and batteries, a propellant for landing, additional tanks and plumbing for landing propellant, landing gear etc.?

Starship is built by virtually the same SS 301/304 as Centaur. The structural design and detailing are very common. Even bullheads have the same shape as early Centaur. So from where comes the extreme structural efficiency. From twits?

There's a mass breakdown estimate in the Starship specs - weight, volumes, etc thread, total mass for tanks and engines is estimated to be just 56.1t, you can try to figure out the rest yourself.

I know this thread. It was made with the wrong assumption that there is only a shell made of the same SS coil spotted on the construction site. In the same thread, you can find another estimation of 124.6 t. It looks more realistic but again does not include the landing propellant. The internal structure is complex and heavy [ref pictures]. The internal structure of “fins” was never revealed but should be massive. There is no reason they to be lighter than Shuttle wings.

 As with any budget, the weight budget is doomed by hidden costs. Plumbing for example. 


 

[TorenAltair]

To put in some real data for comparison by "our own" Dr. Pietrobon.. you know his signature: "Akin's Laws of Spacecraft Design #1:  Engineering is done with numbers.  Analysis without numbers is only an opinion."

https://www.nasa.gov/pdf/382034main_018%20-%2020090706.05.Analysis_of_Propellant_Tank_Masses.pdf

[rakaydos]

As with any budget, the weight budget is doomed by hidden costs. Plumbing for example.

What I get from your posts is that you dont believe Starship is possible.

And you know what, fair enough. Your belief is not required. As long as you restrict your "debunking" to forums and youtube, and dont go around harassing spaceX employees (like that video where Neil Armstrong Buzz Aldrin punches a moon landing denier for calling him a coward) then I hope you will at least accept their successes when they occur.

[kevinof]

Was actually Buzz Aldrin but I get your point.

As with any budget, the weight budget is doomed by hidden costs. Plumbing for example.

What I get from your posts is that you dont believe Starship is possible.

And you know what, fair enough. Your belief is not required. As long as you restrict your "debunking" to forums and youtube, and dont go around harassing spaceX employees (like that video where Neil Armstrong punches a moon landing denier for calling him a coward) then I hope you will at least accept their successes when they occur.

[tbellman]

"The Centaur is around 9%, the EPS on Ariane 5 is around 11%, Falcon 9 second stage is about 4-6%, and Starship using your numbers is 8.5%." Exactly. The simple question is: how starship will achieve the same EPS as Centaur having in mind an additional weight of ~1000 m³ payload bay, engine bay, thermal protection, flaps with powerful mechanism and batteries, a propellant for landing, additional tanks and plumbing for landing propellant, landing gear etc.?

Starship is built by virtually the same SS 301/304 as Centaur. The structural design and detailing are very common. Even bullheads have the same shape as early Centaur. So from where comes the extreme structural efficiency. From twits?

It is constructive instructive to compare not just empty mass to propellant mass, but also empty mass to propellant volume.

(NOTE: Below I'm going to make some assumptions about propellant ratios and temperatures; they are unlikely not perfectly correct, but I believe they are good enough to get in the right ballpark.)

According to http://www.spacelaunchreport.com/atlas5.html, Centaur on Atlas V carries 20.8 tonnes of propellant, and has an empty mass of 2.03 tonnes.  If we assume that it uses a hydrogen/oxygen ratio of 1:6 (by mass), and stores them at 3 bar pressure at their respective boiling points at that pressure (24 K and 102 K), that has an average density of 339 g/l, and its tank volume must then be 61.4 m³, and the stage thus has an empty mass to tank volume ratio of about 33 kg/m³.

Starship, on the other hand, uses methane and oxygen with a ratio of 1:3.6 by mass, stores them at 6 bar, and significantly below their boiling points.  If we for a start assume they use 111 K and 90 K, we get an average propellant density of 835 g/l.  1200 tonnes methalox then uses 1437 m³, and the ship has an empty mass to tank volume ratio of about 83 kg/m³ (assuming an empty mass of 120 tonnes).

So Centaur has a much better dry-mass-to-tank-volume ratio than Starship, by a factor 2.5.

But since Centaur then fillls its tanks with a very low density propellant, its dry-mass-to-propellant-mass ratio suffers.  (On the other hand, hydrolox gives you a much better specific impulse, which evens out things a bit.)  If SpaceX had chosen to fill Starship's tanks with the same hydrolox mixture that Centaur uses, its structural mass ratio would become 20% instead of 9%.  But they didn't.

(Densities of fluids at various temperatures and pressures taken from the NIST Chemistry WebBook.)

Where Musk claims that Starship is already around 100 tonnes? I hope he do not such claims in front of his investors.

In part two of Tim Dodd's interview with Elon Musk / walkthrough of Starbase on July 30th, at around 24 minutes in, Elon said "hopefully not much more than a hundred tonnes" about the dry mass of SN20.  Links to all three parts of it on Youtube are in the first post of this thread.


(EDIT: Word choice.)

[envy887]

Where Musk claims that Starship is already around 100 tonnes? I hope he do not such claims in front of his investors.

The Tim Dodd interview.

If you want to claim that Ship 20 weighs significantly more than Musk says it does, you should provide more evidence than your own uninformed speculation.

[Robotbeat]

"The Centaur is around 9%, the EPS on Ariane 5 is around 11%, Falcon 9 second stage is about 4-6%, and Starship using your numbers is 8.5%." Exactly. The simple question is: how starship will achieve the same EPS as Centaur having in mind an additional weight of ~1000 m³ payload bay, engine bay, thermal protection, flaps with powerful mechanism and batteries, a propellant for landing, additional tanks and plumbing for landing propellant, landing gear etc.?

Starship is built by virtually the same SS 301/304 as Centaur. The structural design and detailing are very common. Even bullheads have the same shape as early Centaur. So from where comes the extreme structural efficiency. From twits?

http://www.b14643.de/Spacerockets_2/United_States_3/Centaur/Centaur%20A.jpg

Using a propellant that’s *much* denser, for one.

[AC in NC]

Where Musk claims that Starship is already around 100 tonnes? I hope he do not such claims in front of his investors.

The Tim Dodd interview.

If you want to claim that Ship 20 weighs significantly more than Musk says it does, you should provide more evidence than your own uninformed speculation.

Do your recall which of the 3 parts of the interview and perhaps roughly which section?  Didn't recall that he said that and wanted to listen again.

[envy887]

Where Musk claims that Starship is already around 100 tonnes? I hope he do not such claims in front of his investors.

The Tim Dodd interview.

If you want to claim that Ship 20 weighs significantly more than Musk says it does, you should provide more evidence than your own uninformed speculation.

Do your recall which of the 3 parts of the interview and perhaps roughly which section?  Didn't recall that he said that and wanted to listen again.

I think around 45 minutes into Part 2, during the TPS discussion. Tim included in this summary, too:
https://everydayastronaut.com/starbase-tour-and-interview-with-elon-musk/

[equiserre]

To satisfy my engineering curiosity, I started keeping track of every piece of information  that comes up on weight, just to check if weights are possible, making assumptions when lacking information. 120ton SS and 200ton SH do seem possible. Spreadsheet attached.

edit: all this should go in the Starship thread

[johntobin]

Interesting piece of work.
The tanks are using 4mm SS, but it sounds like the plan is to move to 3mm thick stainless steel. Presume they will accomplish by further work hardening the SS,  or work hardening the welds by plannishing.
Remembering that the strength of the 304 SS is also enhanced at cryogenic temperatures.