Discussion Elon Musk: the F9 first stage can reach orbit as an SSTO.

RGClark

Mathematician
Joined
Jan 27, 2010
Messages
1,635
Reaction score
1
Points
36
Location
Philadelphia
Website
exoscientist.blogspot.com

Hlynkacg

Aspiring rocket scientist
Addon Developer
Tutorial Publisher
Donator
Joined
Dec 27, 2010
Messages
1,870
Reaction score
3
Points
0
Location
San Diego
What's is the point of being SSTO if you can't carry payload or reenter?

For the umpteenth time, SSTO isn't the challenge useful SSTO is the challenge.
 
Last edited:

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
For the umpteenth time, SSTO isn't the challenge useful SSTO is the challenge.

Yes, it is pretty academic. We know that we can already build one since 1960-something. We just have no reason to do so. It would be way more expensive than a much smaller TSTO at the same technology.

A reusable SSTO with reliable technology would be the turning point - if even launching a fully reusable TSTO (like Kistler K1) would be less economic over a full year of launches than a SSTO, that could require less ground infrastructure.
 

RGClark

Mathematician
Joined
Jan 27, 2010
Messages
1,635
Reaction score
1
Points
36
Location
Philadelphia
Website
exoscientist.blogspot.com
What's is the point of being SSTO if you can't carry payload or reenter?

For the umpteenth time, SSTO isn't the challenge useful SSTO is the challenge.


The tweet suggests the payload will be less, not zero. The upper stage is a significant portion of the cost. If a small payload can be launched by the first stage only, why incur the extra cost of using the upper stage?

Note also as the last launch of the Falcon 9 demonstrated, a second stage introduces new possibilities for failure.


Bob Clark
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
The tweet suggests the payload will be less, not zero. The upper stage is a significant portion of the cost. If a small payload can be launched by the first stage only, why incur the extra cost of using the upper stage?

Because we are talking about a lot less than 1 ton of payload there. At 3200 m/s specific impulse (rather optimistic), you still get only a mass fraction of 5.6% - without reuse. Launch and gone.


And now calculate: Will the SSTO be cheaper than a VEGA for example? Without reuse? A VEGA can launch 1,900 kg to elliptic orbit and is rather expensive at 22 million € during full-scale operations. Even just the first stage of a Falcon 9 1.1 FullThrust will cost around 42 million € if you trust the marketing department of SpaceX. NASA pays $133 million (125 million €) per launch to the ISS.

Remember: As SSTO, you are competing with multistaged launchers, that are orders of magnitude smaller than you, and which are likely also a lot cheaper than you.

The air launched Pegasus rocket did just cost $6 million initially, without liquid upper stage. And still launched 450 kg to orbit.
 

RGClark

Mathematician
Joined
Jan 27, 2010
Messages
1,635
Reaction score
1
Points
36
Location
Philadelphia
Website
exoscientist.blogspot.com
Because we are talking about a lot less than 1 ton of payload there. At 3200 m/s specific impulse (rather optimistic), you still get only a mass fraction of 5.6% - without reuse. Launch and gone.
And now calculate: Will the SSTO be cheaper than a VEGA for example? Without reuse? A VEGA can launch 1,900 kg to elliptic orbit and is rather expensive at 22 million € during full-scale operations. Even just the first stage of a Falcon 9 1.1 FullThrust will cost around 42 million € if you trust the marketing department of SpaceX. NASA pays $133 million (125 million €) per launch to the ISS.
Remember: As SSTO, you are competing with multistaged launchers, that are orders of magnitude smaller than you, and which are likely also a lot cheaper than you.
The air launched Pegasus rocket did just cost $6 million initially, without liquid upper stage. And still launched 450 kg to orbit.

Why do you say it will be less than 1 ton payload? Elon hasn't said how much the payload would be.

Bob Clark
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
Why do you say it will be less than 1 ton payload? Elon hasn't said how much the payload would be.

Bob Clark

Try it with math.

The launcher weights about 500 tons at launch, the specific impulse of both stages is known well enough for a first order estimate. It does not need to be excessively accurate. We do know also the maximum payload to LEO, so we can calculate the payload mass fraction / lambda of the whole rocket and now only have to optimize the lambdas for the stages. By that you can estimate the mass of the first stage, despite SpaceX keeping the mass secret. In that process you also already get a coarse estimate on the dry masses of the stage.

It is no accurate number - but to +/-5 tons accurate is enough.

Now subtract the dry mass fraction from the 5.6% and multiply it with the first stage mass estimate. Yes, the result is almost zero and even if you include all errors possible by the estimate, you still get a range from zero (high dry mass fraction) to about 900 kg (dry mass be extremely low and lots of first stage fuel)
 

steph

Well-known member
Joined
Mar 22, 2008
Messages
1,395
Reaction score
716
Points
113
Location
Vendee, France
SSTO with landing capabilities or it just stays up there, unable to return? Would a heat shield exceed the payload capability? :blush:
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
SSTO with landing capabilities or it just stays up there, unable to return? Would a heat shield exceed the payload capability? :blush:

As much as I can tell: Yes, it would have to stay there. It would not even have enough fuel for deorbit.
 

steph

Well-known member
Joined
Mar 22, 2008
Messages
1,395
Reaction score
716
Points
113
Location
Vendee, France
Might be ok-ish for a science mission. Slap a few instruments on it, some solar panels (IF you manage to get those within the accepted payload), launch to LEO and call it a satellite :p
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
Might be ok-ish for a science mission. Slap a few instruments on it, some solar panels (IF you manage to get those within the accepted payload), launch to LEO and call it a satellite :p

Take the instruments, some solar panels, an attitude control system, that the first stage lacks, a commercial F9 launch to LEO that does not use the full payload mass budget and just piggyback the science mission on the commercial payload.

Use the remaining money for buying a dozen Bugatti Veyrons.
 

RGClark

Mathematician
Joined
Jan 27, 2010
Messages
1,635
Reaction score
1
Points
36
Location
Philadelphia
Website
exoscientist.blogspot.com
Try it with math.
The launcher weights about 500 tons at launch, the specific impulse of both stages is known well enough for a first order estimate. It does not need to be excessively accurate. We do know also the maximum payload to LEO, so we can calculate the payload mass fraction / lambda of the whole rocket and now only have to optimize the lambdas for the stages. By that you can estimate the mass of the first stage, despite SpaceX keeping the mass secret. In that process you also already get a coarse estimate on the dry masses of the stage.
It is no accurate number - but to +/-5 tons accurate is enough.
Now subtract the dry mass fraction from the 5.6% and multiply it with the first stage mass estimate. Yes, the result is almost zero and even if you include all errors possible by the estimate, you still get a range from zero (high dry mass fraction) to about 900 kg (dry mass be extremely low and lots of first stage fuel)

Are you saying the dry mass estimate is accurate to within 5 tons? In that case the payload could be 5 tons more than your estimate.


Bob Clark
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
Are you saying the dry mass estimate is accurate to within 5 tons? In that case the payload could be 5 tons more than your estimate.


Bob Clark

No, the total first stage mass estimate, which I had on about 380 tons during my calculation. The dry mass fraction of the first stage works out between 5.4% and 5.6%. Higher would no longer reach orbit and would mean that the second stage has provide much more DV. Lower would mean that the geostationary missions without reuse would no longer work, as the payload mass fraction would no longer fit to the SpaceX data.

In the best case, the stage reaches a payload of 760 kg. In the worst case only a few kg to nothing. Even at 385 tons, the payload would only be 10 kg more. Include that a payload fairing can be jettisoned already at 85 km altitude in the best case, and you get a few 50 kgs payload mass more.
 

RGClark

Mathematician
Joined
Jan 27, 2010
Messages
1,635
Reaction score
1
Points
36
Location
Philadelphia
Website
exoscientist.blogspot.com
No, the total first stage mass estimate, which I had on about 380 tons during my calculation. The dry mass fraction of the first stage works out between 5.4% and 5.6%. Higher would no longer reach orbit and would mean that the second stage has provide much more DV. Lower would mean that the geostationary missions without reuse would no longer work, as the payload mass fraction would no longer fit to the SpaceX data.
In the best case, the stage reaches a payload of 760 kg. In the worst case only a few kg to nothing. Even at 385 tons, the payload would only be 10 kg more. Include that a payload fairing can be jettisoned already at 85 km altitude in the best case, and you get a few 50 kgs payload mass more.

I was using the propellant mass fraction Elon cited in this lecture at the Royal Aeronautical Society at about the 30 minute mark:


He gave the propellant fraction of the new Falcon 9 v1.1 first stage as around 96%, or perhaps 95.5%. The 96% propellant fraction number gives a 25 to 1 gross mass to dry mass ratio, which means the propellant mass is 24 times the dry mass. At a propellant mass of 380 metric tons (mT), the dry mass would be 16 mT.

Taking the delta-v to orbit as the commonly cited 30,000 feet per second, 9,100 m/s, and the Merlin 1D vacuum Isp as 311 s, then by the rocket equation this allows a payload of 4 mT:
311*9.81ln(1 + 380/(16 + 4)) = 9,140 m/s.

If the propellant fraction is 95.5%, the dry mass would be 18 mT, reducing the payload to 2 mT.

However, as I have discussed to maximize performance for a SSTO altitude compensation is essential. This would allow the Merlin 1D to have comparable vacuum Isp as the Merlin Vacuum, 340 s. Then this would allow a payload of 10 mT if the dry mass is 16 mT:

340*9.81ln(1 + 380/(16 + 10)) = 9,160 m/s.

If the dry mass is 18 mT, then the payload would be 8 mT with altitude compensation.


Bob Clark
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
This would allow the Merlin 1D to have comparable vacuum Isp as the Merlin Vacuum, 340 s

Never happens. Regardless how good your altitude compensation is, you only control how much specific impulse you lose. Which is easily understandable: The exhaust still has to perform work to counter the atmospheric pressure - it only goes away in vacuum.

Also you generally never reach a theoretical full vacuum specific impulse in vacuum with any altitude compensation - you are always less good than an optimized nozzle. Even in the best case in vacuum with a compensating nozzle, you only reach 95% of the performance of a comparable optimized nozzle.

Also the dry mass figure of claimed by Musk can't work out. The Falcon 9 has the same performance to LEO has a Zenit-2 launcher. They have the same propellants (but the Zenit has a higher propellant density because of using stronger cooled propellants). The Zenit-2 launches from much higher latitudes. The Zenit-2 has a much higher specific impulse in its first stage (30 seconds better in both SL and vacuum) and similar ISP for its second stage. The Zenit-2 first stage was not even optimized for optimal performance because of its history.

The Zenit-2 is over 50 tons lighter and smaller than the Falcon 9.

If I use the data by SpaceX, the maximum specific impulse of the Merlin 1D is 311 seconds in vacuum - the 320 seconds that I assumed in my calculations are already extremely favorable.
 

RGClark

Mathematician
Joined
Jan 27, 2010
Messages
1,635
Reaction score
1
Points
36
Location
Philadelphia
Website
exoscientist.blogspot.com
Never happens. Regardless how good your altitude compensation is, you only control how much specific impulse you lose. Which is easily understandable: The exhaust still has to perform work to counter the atmospheric pressure - it only goes away in vacuum.
Also you generally never reach a theoretical full vacuum specific impulse in vacuum with any altitude compensation - you are always less good than an optimized nozzle. Even in the best case in vacuum with a compensating nozzle, you only reach 95% of the performance of a comparable optimized nozzle.
Also the dry mass figure of claimed by Musk can't work out. The Falcon 9 has the same performance to LEO has a Zenit-2 launcher. They have the same propellants (but the Zenit has a higher propellant density because of using stronger cooled propellants). The Zenit-2 launches from much higher latitudes. The Zenit-2 has a much higher specific impulse in its first stage (30 seconds better in both SL and vacuum) and similar ISP for its second stage. The Zenit-2 first stage was not even optimized for optimal performance because of its history.
The Zenit-2 is over 50 tons lighter and smaller than the Falcon 9.
If I use the data by SpaceX, the maximum specific impulse of the Merlin 1D is 311 seconds in vacuum - the 320 seconds that I assumed in my calculations are already extremely favorable.

I'm sure you're aware the Russian launchers such as the Zenit have poor weight optimization compared American launchers and especially compared to SpaceX's launchers. SpaceX couldn't match the Isp of the Russian engines so they opted to optimize weight instead.

It should also be noted that the 13 metric ton payload to LEO for the F9, similarly to Zenit-2, that is cited by SpaceX is actually assuming F9 first stage reusability. A fully expendable version of the F9 actually has a ca. 16 metric ton payload to LEO.

This was indicated by Gwynne Shotwell in an interview where she said without reusability the F9 payload is about 30% higher than the cited amount which would put it at about 16 metric tons:

NASA, CNES Warn SpaceX of Challenges in Flying Reusable Falcon 9 Rocket
May 5, 2014 by Amy Svitak in On Space
SpaceX President Gwynne Shotwell says Falcon 9's reusability is already designed into the rocket's first stage, including the weight of the landing legs that would otherwise detract from the rocket's performance. She also said Falcon 9 retains 30% performance margin over the company's advertised mass-to-orbit capability of 4,850 kg to GTO – margin SpaceX is using to conduct operational trials of a reusable Falcon 9 first stage.
http://aviationweek.com/blog/nasa-cnes-warn-spacex-challenges-flying-reusable-falcon-9-rocket

The newly introduced increase in thrust of the Merlin 1D should increase the payload as an SSTO even further.

Bob Clark
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,665
Reaction score
2,387
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
I'm sure you're aware the Russian launchers such as the Zenit have poor weight optimization compared American launchers and especially compared to SpaceX's launchers. SpaceX couldn't match the Isp of the Russian engines so they opted to optimize weight instead.

You can't build a rocket stage from love and air alone. The 9 engines alone already weight 4200 kg. 5% of the 380 tons is 19 tons - so you have less than 15 tons left for tanks, propellant utilization and structure. The engines need 18 propellant lines with 10 cm diameter. Strong enough to survive not just the tank head pressure, but also the inevitable water hammer, when you close the valves for shut down or if you just throttle the engines.



A tiny soda can is just 0.07 mm thick at the top and 0.15 mm thick at the bottom - but keeps about the same pressure inside as a much larger Falcon 9 tank. The wall thickness of a tank capable of holding the same pressure and being made of the same material scales up linear by inside diameter: So without any special optimizations about the welds used for the tank or internal stiffeners, the tank would already need 3 mm thickness on the average - with some optimizations you can get this down to tapering between 1.5 and 2.5 mm.

The rocket itself is quite long. If you would use 2 mm thin aluminum for the tanks, a 40 meter long tube with 3.66 m diameter would already weight 5 tons. Now add anti-sloshing devices, propellant utilization, valves, the bulkheads for the tanks, stiffeners for getting the loads of the engines transfered to the second stage (or payload)... and you will be around 8 tons, and have only 7 ton left for all the rest. Thrust structure is harder to estimate, because it depends on many factors. But you can be sure that it is likely almost as heavy as the tanks themselves, because it has to transform the forces of the engines into forces that don't break rocket, especially the vibrations of the engines and the vibrations by the hydraulic TVC system have some effect there. Also the more engines you have, the heavier this structure will be. So, lets be nice and say it weights just 4 tons. 3 tons left.

Now you need to get the propellant into the engines. So, you will have 18 short pipes with 10 cm diameter, one short 30 cm pipe for the Kerosene tank and a long 30 cm pipe inside the Kerosene tank for the LOX (30 cm means the same cross section as nine 10 cm diameter pipes), include some bellows to allow the pipes to survive the vibrations... About 2 tons of mass, likely more (The pipes have to withstand some stronger changes of static pressure than the tanks). Now you have only one ton of mass free for all the electric power subsystem equipment (batteries, power controls), all the GNC electronics that you need, all the pressurant bottles and all the radio antennas, receivers, data handling, you name it. All fits barely into the 5% budget and I had been pretty optimistic in the numbers.

Now, where do you think you can reduce the mass to get below 5%? I have already calculated pretty optimistic, the second stage of the Saturn V did also just barely reach the 5% limit back then (No important electronics and much lower thrust to weight)

The second stage of the Falcon 9 could get below the 5% easily and even with very conservative calculations. But for getting even slightly below 5% for the first stage, you would need to assume perfect magic welds (Structure acts like it was made in one piece) and no safety margins too often. 5.5% (2.5 tons more dry mass) would still be a great lightweight structure, but far more realistic in its assumptions.

And in case you think about it: You can't make a thrust structure for a rocket purely of composites. The vibrations make it very hard to use composites there, only the few static parts of it allow it. Otherwise, you risk that fatigue shredders your rocket during launch - nobody has yet managed to find a way to make composites vibration-resistant and light at the same time, despite the huge demand for such a magic material.

Composites work great for interstages - but the Falcon 9 SSTO does not have many of those.
 

Col_Klonk

Member
Joined
Aug 29, 2015
Messages
470
Reaction score
0
Points
16
Location
This here small Dot
Slightly off topic..
Pizza and beer, says that if the falcon gets to final test flight.. it'll fall apart on the way up, due to weak structure.

It looks like it needs a redesign if it's going to work as proposed.
:facepalm:
 
Top