_Augustus_

NASA SLS/Orion/Payloads

Recommended Posts

I feel like NASA should be facilitating other entities' rocket launches. Mainly infrastructure. More pads? Maybe get an electromagnetic launch assist system? A... skyhook (even a relatively low capacity one)? Might be useful in the long run...?

Not to say that I don't like big rockets, but having infrastructure this spread out is just... impractical.

Share this post


Link to post
Share on other sites
On 17.5.2017 at 3:11 PM, tater said:

I actually meant from a design, scheduling, transportation, and stacking standpoint. The teams are in different countries, the hardware needs to be shipped to be tested, then finalized and shipped to be launched. 

Regarding the use of US Customary Units---I agree it's appalling, and while NASA has been officially metric for decades now, the aerospace contractors have not all changed, hence the lack of universality in that regard. Remember that these same contractors are still making hardware for keeping things like the B-52 flying (in service since 1955!). It's tough issue since even a brand new rocket (say F9) needs to integrate payloads that might be from a contractor with legacy tooling.

Yes, the adapter is just an metal plate after all. 
One issue is standards. Russia and SpaceX uses horizontal assembly and integration, US, China, and ESA uses vertical, you also have load requirements during launch. 
Second is fairing size, You design you satelite to be launched by one rocket but its offline and the replacement is either 20cm to small or an serious overkill. 
You also have data buses but you can patch around them but far more expensive than the adapter. 

Share this post


Link to post
Share on other sites

So DARPA selected Boeing's XS-1 "Phantom Express" design, which will use 2 RS-25s for propulsion.

Could this help to reduce SLS engine costs?

Share this post


Link to post
Share on other sites

Only if they use those strapped onto the SLS tank instead of the SRB's (assuming it can match the launch TWR and dV of the boosters).  The problem with SLS isn't just the low volume, it's the lack of re-use.  

A few more engines used on a re-usable space plane won't make a difference unless DARPA orders a lot of them.  Based on the shuttle and X-37b, I'd suspect they'd end up building no more than 6 XS-1's.

 

Share this post


Link to post
Share on other sites
3 minutes ago, sjbuggs said:

Only if they use those strapped onto the SLS tank instead of the SRB's (assuming it can match the launch TWR and dV of the boosters).  The problem with SLS isn't just the low volume, it's the lack of re-use.  

 

IMO they should make the "advanced boosters" liquid flyback boosters, and replace the RS-25s with cheaper alternatives/have them fly back like Vulcan's engines.

But SLS is a government pork project and is being built solely to keep Shuttle jobs in various states......

Share this post


Link to post
Share on other sites
2 hours ago, _Augustus_ said:

I love how SLS/Orion is so delayed and whatnot that we get excited over welding fuel tanks and testing of tiny SRMs.

To be fair a rocket like this hasn't flown in a long time. Not to mention that it's a far cry from the shuttle in many respects. Almost complete redesign of the SRBs as well as the core.

Share this post


Link to post
Share on other sites
16 minutes ago, Bill Phil said:

Almost complete redesign of the SRBs as well as the core.

Despite the fact that it was touted as being able to use Shuttle parts off-the-shelf with few modifications.....

Share this post


Link to post
Share on other sites

I still say that the new vehicle should be the largest scale booster available lifting 2x as much as the heaviest boosters corporations can launch and just start from there. If it needs additional TWR or DV, NASA should outsource to SpaceX or other groups as to strap on Falcon 9s to outreach to other businesses to help them continue with NASA's vision of a heavy lifter capable of a Mars mission.

Share this post


Link to post
Share on other sites
7 hours ago, ZooNamedGames said:

I still say that the new vehicle should be the largest scale booster available lifting 2x as much as the heaviest boosters corporations can launch and just start from there. If it needs additional TWR or DV, NASA should outsource to SpaceX or other groups as to strap on Falcon 9s to outreach to other businesses to help them continue with NASA's vision of a heavy lifter capable of a Mars mission.

But a heavy lifter isn't necessary for a Mars mission... 

Share this post


Link to post
Share on other sites
6 hours ago, kerbiloid said:

It will press the rocket to the ground preventing the launch whil the crew gently gets out through the door and walks away?
So-o Kerbal...

They could just have launched it up in the air, thinking of the insane video there the escape system escaped and the parachute popped some seconds after. 

Share this post


Link to post
Share on other sites
3 hours ago, Bill Phil said:

But a heavy lifter isn't necessary for a Mars mission... 

So we launch a lander, a self sustained space station and crew using using a single Soyuz launch instead?

Share this post


Link to post
Share on other sites

Why limit everything to a single launch?

Share this post


Link to post
Share on other sites
3 minutes ago, razark said:

Why limit everything to a single launch?

MOR forever. 

Share this post


Link to post
Share on other sites
33 minutes ago, ZooNamedGames said:

So we launch a lander, a self sustained space station and crew using using a single Soyuz launch instead?

Uh, no. We launch already existing rockets. If you need a few hundred tonnes of stuff in orbit, say, 300 tonnes, you shouldn't develop a whole new launcher for it.

Akin's 39th Law of Spacecraft Design:

Quote

39. (alternate formulation) The three keys to keeping a new manned space program affordable and on schedule:
       1)  No new launch vehicles.
       2)  No new launch vehicles.
       3)  Whatever you do, don't develop any new launch vehicles.

An Atlas V 551 could launch 20 tonnes into a 28 degree Low Earth Orbit (according to Astronautix) at 153 million dollars. That's only 2.3 billion dollars to orbit 300 tonnes. There are cheaper launchers, probably. But still, that's doable. Most of that mass is propellant, so if we could somehow get some for cheap using some other method (space guns like Project HARP, from an asteroid, from the Moon, whatever) then we'd be golden. Heck, that orbital launch cost is still less than the cost per lunar mission for Apollo (over ten billion per mission). Most of our cost will be R&D and construction for the final space hardware. If we don't waste anything developing new launchers, we can keep the cost down. But it's probably too late for that...

Point is that we can get away with EOR, which would still be necessary for a Mars Mission anyways, even with a heavy lifter.

Share this post


Link to post
Share on other sites
1 hour ago, Bill Phil said:

An Atlas V 551 could launch 20 tonnes into a 28 degree Low Earth Orbit (according to Astronautix) at 153 million dollars. That's only 2.3 billion dollars to orbit 300 tonnes.

The problem with such simple minded math is that it fails to significantly reflect reality.  First, because it fails to address total programmatic risk.  More launches, more rendezvous and prox ops, more interfaces that must be connected, all of these things increase the risk that something will go wrong.  (Not to mention the schedule pressure and headaches.)  Second, because it fails to address parasitic mass.  The more chunks you break your final vehicle into, the higher the proportion of parasitic mass (stuff needed to get the chunks into position, but not required for the final vehicle) grows.  Basically, the higher the proportion of parasitic mass, the higher your overhead costs.  (That is, the "obvious" step of reducing costs per launch doesn't mean your costs go down in proportion to the launch cost reduction.)

The latter can significantly alter your total costs...  An Altas V 551 doesn't launch 20 tonnes of final payload mass - it launches (as a SWAG) around 15-18 tonnes of final payload mass and 2-5 tonnes of one-time-use support and delivery mass.  Even using the optimistic end of the spectrum, you end up shelling out 2.5 billion dollars (an increase of almost 10%).  And that doesn't take into account the increased costs of engineering and integrating a finely divided system, nor the increased operational costs, nor the engineering and hardware costs of the one-time-use support and delivery systems.   (That's why so many orbital assembly schemes use independent reusable tugs and fuel depots, trading increased complexity and risk for reduced costs.)

Share this post


Link to post
Share on other sites
53 minutes ago, DerekL1963 said:

The problem with such simple minded math is that it fails to significantly reflect reality.  First, because it fails to address total programmatic risk.  More launches, more rendezvous and prox ops, more interfaces that must be connected, all of these things increase the risk that something will go wrong.  (Not to mention the schedule pressure and headaches.)  Second, because it fails to address parasitic mass.  The more chunks you break your final vehicle into, the higher the proportion of parasitic mass (stuff needed to get the chunks into position, but not required for the final vehicle) grows.  Basically, the higher the proportion of parasitic mass, the higher your overhead costs.  (That is, the "obvious" step of reducing costs per launch doesn't mean your costs go down in proportion to the launch cost reduction.)

The latter can significantly alter your total costs...  An Altas V 551 doesn't launch 20 tonnes of final payload mass - it launches (as a SWAG) around 15-18 tonnes of final payload mass and 2-5 tonnes of one-time-use support and delivery mass.  Even using the optimistic end of the spectrum, you end up shelling out 2.5 billion dollars (an increase of almost 10%).  And that doesn't take into account the increased costs of engineering and integrating a finely divided system, nor the increased operational costs, nor the engineering and hardware costs of the one-time-use support and delivery systems.   (That's why so many orbital assembly schemes use independent reusable tugs and fuel depots, trading increased complexity and risk for reduced costs.)

The problem with the "total programmatic risk" approach to assessing multiple types of missions is that it fails to take into account parasite launch vehicles. Which can easily add tens of billions to total project cost. Not to mention the nightmares of having just a launch per year, if not less. There's a reason I said "could", and that's because it "can't", at least not in the 551 version. The 552 version might be able to put more up there. Heck, the Atlas V has had a pretty significant cost reduction in recent years, not to mention its high reliability. I'd rather rely on a launcher with very few failures than one with no launches. Basically, if a parasite launch vehicle shows up, expect an increase in program cost by many billions of dollars.

Yes, there are more chances for something to go wrong. But you're trading an increase in the occurrence of failures for an increase in redundancy for those failures. More things will go wrong, but the system as a whole is more capable of handling those events than a heavy lift architecture.

The arguments you make are true, yes, but there's a reason I didn't take that into account. Because the small increase in cost from those concerns is still less than the cost of developing and then launching six SLS rockets. Far, far less. And considering that most of the engineering and cost will be for the final assembled spacecraft, launch costs are unimportant. A few billion is less than NASA's annual budget, what's going to be the vast majority is developing the spacecraft and the cost of the spacecraft itself. In any case, it should be similar in cost to the ISS, except that it can actually go to Mars. And it may cost even less, considering that most of the mass is propellant, and that no Shuttle flights would be involved. We wouldn't really know until we do a detailed analysis.

The point I'm making here is that the effort spent to develop SLS would've been far better suited to developing a spacecraft using many smaller launchers. But instead we had to wait for a heavy lifter when it's totally unnecessary. Sure, you may get reduced costs with an already existing heavy launcher, but we don't have one. Heck, we technically had one for 30 years, but it had to waste 80% of its payload capacity. While there are plenty of added complications to EOR, they are possible to overcome. Heavy lift is not necessary for Mars. Even the Moon. What is necessary is money. A lot of it. But if you don't already have a heavy lift system, then developing a new one just increases the cost.

Edited by Bill Phil

Share this post


Link to post
Share on other sites

The other issue is volume.

The large volume bump for SLS is non-trivial, though to be fair, Delta IV Heavy is 5m with a flush fairing. Fairings typically go to what, about 125% of stage diameter? That would be a 6.25m fairing on D IV Heavy, which certainly allows for some larger components. 

Still, there is benefit in integrating components terrestrially, vs on orbit.

Share this post


Link to post
Share on other sites

What if the SLS would use Falcon 9 core stages (2 on each side) instead of the Solid rocket boosters? Do you think this can increase the performance of the SLS ?

Share this post


Link to post
Share on other sites
1 minute ago, Julien Kerman said:

What if the SLS would use Falcon 9 core stages (2 on each side) instead of the Solid rocket boosters? Do you think this can increase the performance of the SLS ?

Not ever going to happen, not even worth thinking about.

That's also not nearly enough thrust for SLS. Most of the lifting for SLS is the SRBs, the RS-25s are sustainers that burn the whole way.

Share this post


Link to post
Share on other sites
2 hours ago, tater said:

The other issue is volume.

The large volume bump for SLS is non-trivial, though to be fair, Delta IV Heavy is 5m with a flush fairing. Fairings typically go to what, about 125% of stage diameter? That would be a 6.25m fairing on D IV Heavy, which certainly allows for some larger components. 

Still, there is benefit in integrating components terrestrially, vs on orbit.

It's a fun balancing act, many small launches mean the boost is simpler at the cost of more inefficiencies and complexity once in orbit.  A few big launches mean more and more of a corner case the booster has to be built around.  

Somewhere in between should be the balance of doable orbital assembly and not breaking the bank with the launches.

Using the 300T metric suggested before, that'd take about 5 falcon heavy launches vs the 15 Atlas's.  And the Falcons should cost less than a billion (even with a 100% gov markup).

You'd be limited to the 5.2m fairing but that seems reasonable.

Share this post


Link to post
Share on other sites

Offtop about space stations. Just because discussion already took place.

Spoiler

On the first hand, the smaller the modules - the more parasite mass (as mentioned above).

Say, Mir consists (in past) of 6 modules, 20-23 t each. 
But as we now know, its base module was overweighted for several tonnes, so they had to redistribute several tons from it between both Kvants.
Every other module included full orbital tug based on TKS, with engines, RCS, solar panels, fuel tanks, antennas and so on.
Just the Kvant tug has separated after docking, while other tugs stayed inside their modules.

So, as a tug itself weights several tons, if remove all equipment required only for approaching and docking, 
20..23*6 = 125 t of Mir modules total launch mass would be 1.5..2 times less. 
About 80 t, i.e. like a single-module Skylab, though much more cramped inside, without Skylab's hotel apartments.
In this case it would also require much smaller truss. So, I guess, instead of ~150 t it would weight about 100 t. My fault, used extended mass value.

A similar situation with ISS. If read its wiki-description, the modules themselves are the lesser half of ISS.
Mostly it consists of trusses, hubs and so on. 
Every additional hub is a spit into faces of both mass efficiency and functional decomposition.
They launch numerous sheds, put anything here and there in any free space.

Those numerous notebooks sticked along the ISS walls here and there obviously tell us that the ISS IT network is planned by throwing spaghetti against the wall.
Any sysadmin trying to do this on the Earth, would be utilized into home pets' food in a week.
Obviously, such "architecture" is caused just by the unpredictable multi-module chaos, when nobody knows where they will mount a telescope, and where put pack of towels.

And you can take, say, Mir, and try to extend it with several modules more. Every second module will be a hub. 
Then you will use every of them as a shed just to use its space.
Then you realize that most of your equipment is spread around between the modules, and your "main" modules look exotic.
As a result you have a countryside lego station instead of a wisely planned orchestra.

Also, the way to deliver 20t modules with a shuttle (i.e. launch 105 t instead of 20) is beyond the good and evil.
105 t = 5 Mir-style modules.

So, any multi-module station is by definition a sanctuary of Chaos.


On the second hand, the bigger the modules, the less of them you need.
So, the less you need to launch. And the less rockets you need.
When you need just 10 launches, you get very expensive rocket infrastructure and follow the way of Saturn and Energy.
Your station gets golden. 
Unless your heavy rocket consists of parts being produced by the same equipment as for 20 t rockets.

On the third hand, the bigger modules - the harder to change/replace them even if you have money.


So. Mir. 
Let's presume we have a rocket enough capable to launch it's whole originally planned base module. ~30 t.
Let's presume there are no shuttles, all other modules should be self-propelled, as Mir modules were.
Kvant tug weighted about 10 t. 

So, if use the same rocket like for the heavy base module, ever module would weight ~20 t, i.e. the same as all ISS modules.
And they will be delivered without any shuttle. 

Instead of launching a 105 t shuttle with one 20 t module, they can deliver 3 full-featured 20 t modules, then separate and deorbit the tugs.
In this case you can replace 5 additional modules of Mir with two Harmony-style modules. And still have enough room for 3 modules more.
If add a docking module on the rear side, you can mount a whole ISS in Mir style, without multiple ugly auxilliary hubs and joints.

Now this is extended Mir and optimized ISS two-in-one.

Crew.
Retrospectively (I took a look through the Mir and ISS statistics), there are usually 3-up-to-6 spacemen onboard.
Average spaceflight lasts for ~3 months, per every 3 humans they launch 2 Progresses or equivalent mass of HTV/ATV/etc.

Looks funny, but this exactly corresponds to the original Almaz planning: crew of 3, duration of 3 months, 20 t of supply ships.
Just in case of Almaz all this stuff would be launched in one launch, while now requires 3 (1 Soyuz + 2 Progress). 
Shuttle delivered up to 20 t (usually less), so doesn't change the statistics much.

This means that also a 20 t rocket is required for supply ships.

I.e. the combination of 20 t and 35 t rockets is the best, the golden mean of two monstruosities: all-in-one and lego-station.
They also can easily use the same equipment to be produced. Every time that one which is currently required.
Also, their stages are enough big to combine several of them into a 100 t asparagus once per decade if absolutely needed.

As we can see, this corresponds to 20 t Proton, 35 t (planned by Chelomei, but rejected) and 100 t Proton-based UR-700.

 

Edited by kerbiloid

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.