DIRECT v3.0 - Thread 2

[clongton]

First, you shouldn't explicitly ignore ideas that make your problems go away.  If tugs help substantially with the cost of fueling, they should be developed.  This isn't Apollo, where the program was so short that development costs completely overshadowed operational costs - or at least, we hope it isn't...

Second, as we all know, the hope is that launch costs will be brought down substantially by the opportunity to launch huge amounts of very cheap stuff into space.  If SpaceX were to get their launch costs down to what they were hoping for originally, you could fill the depot with four F9H launches for less money than a single J-246.

This is why somebody should be paying for development of Skylon (or something comparable, naturally, if anyone has any good ideas).  If they hit their price and performance targets, they should be able to loft 100 mT of propellant for $20-50 million total (before anyone asks:  yes, the payload bay is large enough to max out the vehicle's payload capacity with liquid hydrogen, should you so desire).  And since the vehicle is basically an unmanned space shuttle, it should have no problem with automated prox ops at the depot, especially if the depot has a robotic arm.

Sure, near-term depot missions would have global cost greater than non-depot missions.  But we're trying to get the CATS ball rolling, and launching an SDLV twice as often doesn't help with that.

The DIRECT team *highly* recommends that you put as much of the expensive electronics and hardware on the depot itself and slave everything else to it. That will keep the cost of the supporting infrastructure down.

[Lobo]

First, you shouldn't explicitly ignore ideas that make your problems go away.  If tugs help substantially with the cost of fueling, they should be developed.  This isn't Apollo, where the program was so short that development costs completely overshadowed operational costs - or at least, we hope it isn't...

Second, as we all know, the hope is that launch costs will be brought down substantially by the opportunity to launch huge amounts of very cheap stuff into space.  If SpaceX were to get their launch costs down to what they were hoping for originally, you could fill the depot with four F9H launches for less money than a single J-246.

This is why somebody should be paying for development of Skylon (or something comparable, naturally, if anyone has any good ideas).  If they hit their price and performance targets, they should be able to loft 100 mT of propellant for $20-50 million total (before anyone asks:  yes, the payload bay is large enough to max out the vehicle's payload capacity with liquid hydrogen, should you so desire).  And since the vehicle is basically an unmanned space shuttle, it should have no problem with automated prox ops at the depot, especially if the depot has a robotic arm.

Sure, near-term depot missions would have global cost greater than non-depot missions.  But we're trying to get the CATS ball rolling, and launching an SDLV twice as often doesn't help with that.

I’m talking about pure costs, not seeding the commercial space industry by purchasing more expensive flight options.
Interestingly enough, in the past after I’ve stated my observations with costs, they always seem to fall back on how going EELV will seed the commercial rocket industry, even it if actually a fair bit more expensive.  That’s fine.  But that’s –not- the reasons that are presented for depots up front.  Those reasons are usually 1)  Fewer flights, and 2)  Cheaper.
My point is neither of those are really true.  It’s far more flights per lunar mission, and considerably more expensive in total dollars spent per mission.  Perhaps cheaper to the US if we get others to pick up those fuel flights, but more expensive and logistically complex overall.

So, if you want to assume depots and a high flight rate will eventually bring those costs down to less than 2 X J-246 launches, that’s fine.  But that wouldn’t be for some time.  Up front you are looking at significantly more cost.  We don’t have a depot yet, or a tug.  And you need fuel missions under probably $50 million to just be competitive with 2 X J-246…that’s not even cheaper, just similar, and doesn’t include the costs of developing, building, launching, and maintaining a tug and a depot.  And I wonder what sort of flight rate you’d need to get Atlas or Delta below $50 million each?  That’s what, 1/5 the current price?  So forgive me if I’m skeptical.  I have more faith SpaceX or the Ruskies could get the cost down closer to $50 with a high flight manifest.  But less?  Significantly less?  Hmmm….

If we want the depot/tug architecture to be basically a “Cash for Clunkers” specific industry stimulus program for commercial domestic rocket manufacturers instead of commercial domestic auto manufacturers, well then just be up front about that when advocating it, rather than talking about how you reduce the launches, when you really greatly increase the launches, and how it’s cheaper when it’s really not cheaper than 2 X J-246 architecture. 

See if “Cash for Rockets” flies on it’s own merits?

Not to mention, what happens to the J-246 costs when they have a high flight rate?  Wouldn’t they come down too just like EELV’s would?  So if you have J-246 flights well below $350million, that makes that math even harder to work for EELV fuel flights from a purely economic standpoint.  If you make enough of them to get the price down to around $250 million each, then you need EELV’s at more like $33 million each, or EELV-heavies at $50 million each to be just competitive.  Not cheaper, just competitive.  And I have a really hard time seeing those prices.
And again, why wouldn’t International partners rather barter for the J-246 rather than spending significantly more for those EELV flights?

Anyway, once again, let me iterate.  I’m not against depots, and many folks on this forum who are way smarter than and I respect advocate them.  I’ve just never quite understood the reasoning that’s often cited for them from an EELV fueling standpoint.

If you had a mission that required only enough propellant so that 1 or 2 EELV’s could supply it to a depot, and you didn’t need to launch a whole J-246, -then- the math starts working out.  Or if you needed 110mt of propellant  so 2 J-246 propellant launches would be a waste.  Things like that.  (although 2X J-246 fuel launches would still probably be cheaper than8-10 EELV flights) But current lunar mission requirements are about what one J-246 launch can do.
That’s all I’m saying.

[adamsmith]

...
Second, as we all know, the hope is that launch costs will be brought down substantially by the opportunity to launch huge amounts of very cheap stuff into space.  If SpaceX were to get their launch costs down to what they were hoping for originally, you could fill the depot with four F9H launches for less money than a single J-246.
...

...

So, if you want to assume depots and a high flight rate will eventually bring those costs down to less than 2 X J-246 launches, that’s fine.  But that wouldn’t be for some time.  Up front you are looking at significantly more cost.  We don’t have a depot yet, or a tug...

See if “Cash for Rockets” flies on it’s own merits?
...

I think there is an element of if you build it they will come when it comes to PD.  However the beauty of PD is it delinks the mission from a specific LV.  As you know when you launch a payload, you can send up propellant to take advantage of under-utilized lift capacity. Even simple H2O will work.  It is trivial to take H2O and make propellant if you have a cryo PD.

You are assuming that somehow every mission will fully utilize almost 100MT of payload.  I don't think so...

Anyhow... I think we are OT.

Stanley

EDIT:Typos and grammar

[kraisee]

Trying to stay a bit on topic, any word on when we'll see some artwork for DIRECT 3.1, or whatever version has the clustered engine design?

Here's a preview of the Series 42 vehicle which should keep you fairly happy for now

Ross.

[StuffOfInterest]

Here's a preview of teh Series 42 vehicle which should keep you fairly happy for now

That looks MUCH more efficient than the inline engine design.  The thrust structure appears much simpler and the airflow should stream around the back better without the ears sticking out of the sides of the tank for engines 1 and 4.

That second render would be a really good one for the Wikipedia page.

[kraisee]

With all the discussion recently, let me take a moment to try to clarify the Depot architecture decisions which we have made for DIRECT.

There is much debate about using an all-EELV-class approach.   What this would require, is approximately 9 launches for each mission.   Assuming a combination of 20mT and 25mT vehicles the following approach is hypothetically possible (though only if you choose to completely ignore the volume/diameter issues entirely):

1 Orion (fueled) -- Heavy
2 Lander Ascent Stage (fueled) -- Intermediate
3 Lander Descent Stage (dry) -- Heavy
4 EDS (mostly dry) -- Intermediate
5 Fuel for Descent Stage -- Intermediate
6 Fuel for EDS -- Intermediate
7 Fuel for EDS -- Intermediate
8 Fuel for EDS -- Intermediate
9 Fuel for EDS -- Intermediate

This architecture certainly requires the use of Propellant Transfer technologies and would almost-certainly require a full Depot to be deployed as part of the baseline Critical Path to success.

Of these 9 launches, the first 4 in that list are all mission critical and the loss of any one would result in an LOM situation.   The latter 5 launches are somewhat "interchangable" so there is "Partial Redundancy" possible there.   It's not too bad, but the logistics and the necessity to coordinate the launch of 4 of those vehicles perfectly in support of each mission, plus the constant fuel deliveries as well, makes it a very demanding logistical nightmare.



Comparatively, DIRECT chooses a three-step approach to getting to the final arrangement.

Firstly we deploy Jupiter-130 in order to preserve jobs & experience to secure the political backing we need in Congress.   This provides an incredibly capable system all by itself, but still isn't quite enough for Lunar use (although with a Delta Upper Stage, the Flyby missions are quite possible).

Step 2 is the deployment of the Jupiter-24x and the Altair lander, which opens up the Lunar capabilities without requiring any Propellant Transfer technologies at all.   This is an interim step designed to begin our new exploration efforts and to allow NASA to start the Exploration efforts in earnest while other important technologies continue to be developed -- without those technologies every appearing on the "Critical Path".

Step 3 is the ultimate goal though.   Here, every Jupiter launch represents a complete mission, supported by a constant stream of fuel deliveries going to a (one or more) Depot.   The mission Hardware all launches upon a single launcher which then rendezvous with the Depot, fills up all the tanks it needs to with whatever fuel load is required for that mission and then departs upon its mission without ever requiring any other docking events.

The purpose of this approach is to maximize the number of units in production, not just for the costly launch vehicles, but also for the even more expensive spacecraft as well.

This architecture opens the door not to just 2 Lunar-class missions per year, but to a possible 8 (or more) every year.   More importantly, this approach also enables all of the NEO and Mars missions as well without further investment in the basic infrastructure.   This approach is quite capable of sending hundreds of tons of useful payload material towards Mars -- or even Jupiter if required.

I am including the costs for the launches below.

Ross.

[clongton]

Here's a preview of teh Series 42 vehicle which should keep you fairly happy for now

That looks MUCH more efficient than the inline engine design.  The thrust structure appears much simpler and the airflow should stream around the back better without the ears sticking out of the sides of the tank for engines 1 and 4.

That second render would be a really good one for the Wikipedia page.

It's the SSME that makes it possible. The regenerative nozzle stands up unexpectedly well in the harsh thermal environment. It really is the best way to go.

[robertross]

Here's a preview of teh Series 42 vehicle which should keep you fairly happy for now

That looks MUCH more efficient than the inline engine design.  The thrust structure appears much simpler and the airflow should stream around the back better without the ears sticking out of the sides of the tank for engines 1 and 4.

That second render would be a really good one for the Wikipedia page.

It's the SSME that makes it possible. The regenerative nozzle stands up unexpectedly well in the harsh thermal environment. It really is the best way to go.

Umm...is that the way the propellant lines are being routed?? Doesn't like right to me...with gimballing and allowances for thermal effects & other forces, I would think they would have more bends (45s & 90s). I know straight-through is best to reduce losses, but you can't always have that.

[kraisee]

Robert,
Its just notional for now.   We are assuming flex joints throughout the plumbing structure, but until we actually get some high-fidelity loads modeling on the parts we aren't going to know what changes are required where.   That's all part of what NASA will have to do once they adopt the system and we're only taking the design 'so far'.

Ross.

[Namechange User]

Firstly we deploy Jupiter-130 in order to preserve jobs & experience to secure the political backing we need in Congress.   This provides an incredibly capable system all by itself, but still isn't quite enough for Lunar use (although with a Delta Upper Stage, the Flyby missions are quite possible).

This has always been a sticking point with me.  That's a tall order without any real detail behind it.  I don't disagree, and never have disagreed, that some jobs would be preserved.  However, many people there work in support of the orbiter....and there is still no orbiter. 

[robertross]

Robert,
Its just notional for now.   We have already included flex joints throughout the plumbing structure, but until we actually get some high-fidelity loads modeling on the parts we aren't going to know what changes are required where.   And that's part of what NASA will have to do once they adopt the system.

Ross.

Okay cool, thanks.

great models btw. Thanks.

[kraisee]

This has always been a sticking point with me.  That's a tall order without any real detail behind it.  I don't disagree, and never have disagreed, that some jobs would be preserved.  However, many people there work in support of the orbiter....and there is still no orbiter. 

The key is that it isn't about getting people to do precisely the same job as they are doing right now, its more to do with finding new positions which are similar enough that each person can apply their personal skillsets and knowledge to a similar task.

There's little question that a program which costs about $8bn per year is going to ultimately require more people working on it than the $5-7bn per year STS + ISS programs do now.

The only real question is whether to transition people across from one to the other or whether to get rid of all the experience we have now and start-over again in 5 years time.


How you do it is really difficult though -- logistically it is a complete 'mare.

There are some areas which don't require much change (Payload Processing and SRB processing for example).   There are other areas which will change, but which will change in a fashion whereby existing capabilities can be levered quite well (reusable SSME maintenance facilities > disposable SSME assembly line for example) and there are some areas which get completely altered (Orbiter structure maintenance for example).

Obviously, depending on the degree of change in each aspect, depends on the difficulty level for workforce retention too.   If the changes are minimal, you'll probably keep the workforce in that area quite stable.   If the changes are major, you're going to have a great deal of re-arranging to do.

Then there's the time factor.   Some parts of the whatever-the-new-program-will-be (Altair, Deep Space Hab Modules etc) are still many years away from being ready.   Yet do you really want to get rid of experienced staff and just assume you'll start with freshmen when those elements are ready?   There are some situations where even expediting things like Altair and closing the gap by stretching Shuttle out still leave you with a 'gap' for many hundreds of workers.   You can get rid of them and re-hire later -- that's certainly an option -- but if you want to retain all that experience it might be better to find some temporary tasks for them to do in the interim and just keep them on the pay-roll.

There are a variety of such 'temporary' tasks which we have come up with to allow those people to be productive during this "make work" period, so that they aren't actually a burden and are still actually contributing value to the program.   One such short-lived project which we have proposed is the SSPDM which would require skills from personnel with familiarity from Orbiter Structures, Orbiter RCS, Orbiter RMS, Orbiter TPS and a variety of other people for about 4-5 years.   This, together with a number of other such elements, can employ hundreds (perhaps thousands) of workers until such time as the main elements (Altair, Lunar Hab etc) start coming online and those people can all then transfer across to those projects permanently.


The key to saving the jobs though, is primarily about budget.   You have to have the budget available in order to pay the salaries and the contracts.   If you burn all the money elsewhere, you simply won't get the option.   And that's where DIRECT changed the dynamic.   The three factors of stretching Shuttle, reducing the development and deliberately budgeting in order to retain the workforce are what allows DIRECT to save most (though not all) of the current workforce.

One factor which does help quite a lot though, is that the NASA workforce has a relatively large population of people approaching retirement right now.   In total it represents about 15-20% of the total workforce who are ready to accept a retirement package within the next ~6 years.   There are a lot of cost savings to be had by simply not back-filling any of those workers.

It's not a perfect scenario, and frankly is a *&^%&$%%^$ to organize.   Many people will probably need to re-locate to other centers too.   But there are many ways to work this problem and allow that vast majority of people who want to stay in the program, to stay.

In summary:   DIRECT fundamentally creates a budget profile which ALLOWS the agency to keep the staff on the payroll.   How the agency chooses to make use of them is ultimately NASA's problem, but we have thought though many good uses taking staff from Shuttle thru the Interim period and eventually over into the Exploration Program later.   We have options which cover all those stepping-stones and we are more than willing to discuss them with the agency if asked.

The devil is in the details, as always, but I'd strongly recommend starting-out with a "maybe we can do this" attitude

Ross.

[Lobo]

Here's a preview of teh Series 42 vehicle which should keep you fairly happy for now

That looks MUCH more efficient than the inline engine design.  The thrust structure appears much simpler and the airflow should stream around the back better without the ears sticking out of the sides of the tank for engines 1 and 4.

That second render would be a really good one for the Wikipedia page.

It's the SSME that makes it possible. The regenerative nozzle stands up unexpectedly well in the harsh thermal environment. It really is the best way to go.

Yea, amazing what a coolant liquid at -423 deg. F can withstand. 

Other than that Chuck, briefly what if any advantages are there to this cluster configuration over the inline in terms of cost and performance.

[clongton]

Here's a preview of teh Series 42 vehicle which should keep you fairly happy for now

That looks MUCH more efficient than the inline engine design.  The thrust structure appears much simpler and the airflow should stream around the back better without the ears sticking out of the sides of the tank for engines 1 and 4.

That second render would be a really good one for the Wikipedia page.

It's the SSME that makes it possible. The regenerative nozzle stands up unexpectedly well in the harsh thermal environment. It really is the best way to go.

Yea, amazing what a coolant liquid at -423 deg. F can withstand. 

Other than that Chuck, briefly what if any advantages are there to this cluster configuration over the inline in terms of cost and performance.

Cost: Being a relatively standard design, it is less expensive to build.
Performance: Because this thrust structure would weigh less, performance is increased by that difference - not a lot, but nothing to sneeze at.

[Lobo]

>>>snip<<<

Of these 9 launches, the first 4 in that list are all mission critical and the loss of any one would result in an LOM situation.   The latter 5 launches are somewhat "interchangable" so there is "Partial Redundancy" possible there.   It's not too bad, but the logistics and the necessity to coordinate the launch of 4 of those vehicles perfectly in support of each mission, plus the constant fuel deliveries as well, makes it a very demanding logistical nightmare.

>>>snip<<<

Comparatively, DIRECT chooses a three-step approach to getting to the final arrangement.

Firstly we deploy Jupiter-130 in order to preserve jobs & experience to secure the political backing we need in Congress.   This provides an incredibly capable system all by itself, but still isn't quite enough for Lunar use (although with a Delta Upper Stage, the Flyby missions are quite possible).

Step 2 is the deployment of the Jupiter-24x and the Altair lander, which opens up the Lunar capabilities without requiring any Propellant Transfer technologies at all.   This is an interim step designed to begin our new exploration efforts and to allow NASA to start the Exploration efforts in earnest while other important technologies continue to be developed -- without those technologies every appearing on the "Critical Path".

Step 3 is the ultimate goal though.   Here, every Jupiter launch represents a complete mission, supported by a constant stream of fuel deliveries going to a (one or more) Depot.   The mission Hardware all launches upon a single launcher which then rendezvous with the Depot, fills up all the tanks it needs to with whatever fuel load is required for that mission and then departs upon its mission without ever requiring any other docking events.

The purpose of this approach is to maximize the number of units in production, not just for the costly launch vehicles, but also for the even more expensive spacecraft as well.

This architecture opens the door not to just 2 Lunar-class missions per year, but to a possible 8 (or more) every year.   More importantly, this approach also enables all of the NEO and Mars missions as well without further investment in the basic infrastructure.   This approach is quite capable of sending hundreds of tons of useful payload material towards Mars -- or even Jupiter if required.

I am including the costs for the launches below.

Ross.

Ross,

Thanks for the explanation.  Eloquent as always.

First, I think obviously the “all EELV” option is a mess and logistical nightmare like you said.  The discussion has been more your “Step 3”, where Orion and Altair are launched with the EDS on a single J-246 in a “single launch architecture”, to go up to an EELV filled  depot, tank up, and do the mission.

Myself, the point I was making is a lot of people call that a “single launch” approach, but really it takes several EELV launches to make that happen, so it’s more of a “6-8” launch approach, depending on the size of the EELV class rockets used.  So it adds logistics and complexity to a mission, rather than reducing it.  Those fuel flights can be interchangeable, if one is delayed or aborted, another could be sent up to replace it.  But a lunar mission cannot take place until X flights delivering Y kg’s of propellant to a depot were successfully launched.  That’s what I meant when I used the term “mission critical”.  The mission can’t happen without even one of a minimum number of flights.

As to your Step 3, two things.  First I think the concept of a constant flow of EELV rockets to a depot to fill it up is a victim of Direct’s own darned efficiency.  You’ve produced an HLV option that can do what 6-8 EELV class launches are needed to do.  The math just doesn’t pencil out where the EELV’s would be cheaper (or even competitive) with a single J-246 launch.  Your vehicle is too darn affordable for that role!  You are saying, “yea, I know it’d save a lot of time and money and logistics to use our new rocket, but we need to use a bunch of smaller rockets instead!”.   And I think people are looking at that going, “What?...”

2nd, you’ve made the case to me before that you didn’t want the Jupiters to be used for fuel (even though they’re the best way to do it), so they’d be free for more Lunar Missions, and put the fuel on EELV to help seed the commercial launch market (“Cash for Rocket stimulus program”). 
In reality, I’d be astounded if NASA would get to a place where they were doing more than 3 or 4 lunar missions a year anytime in our lifetimes.  They can barely get the Shuttle off the ground that often.    They were putting up 2-3 per year form 1969-1971 in their hay-day, and granted, we might be able to do it more efficiently now, but I just can’t see any more than that now.  (as much as I’d love more!).  I think you said MAF could pump out like 12 Jupiter cores per year maximum.  Enough for 6 missions with no depot.  Let’s say 2 of those were used for ISS crew rotations/repair/supply missions per year.  So we can do 10 cores and 5 Lunar Missions.  That’s still more than NASA is likely to do in this day and age.  I know you are optimistic that they could get up to max capacity and not have enough cores…and trust me brother, I want SOOOO bad to share that optimism, but I just don’t see it happening any time soon.  Maybe some day if we get a President and Congress that are excited about HSF and ramp up NASA in a way not seen since JFK and LBJ.  (but I’m not holding my breath for that from either party)  At such a time maybe there’s then perhaps a need for more cores than MAF could produce without a depot, and then NASA would have the resources and direction to design and build one, with a tug, and whatever other new technologies that would really make that work slick.  But even then, unless other parties pony up the money for all those EELV fueling flights, the Jupiter core is still the best way to get the fuel up there…so maybe MAF could be expanded to produce more cores for cheaper than could be done with commercial contracts for EELV class flights?

So let’s just say perhaps I’m realistically seeing Your Step #2 and saying that’s probably about as far as I see NASA going anytime soon.  Or anytime for a long, long time.  Mars, Jupiter, any place else I see as being so far out there that it would be a non factor for a couple of decades at least.  And I just don’t see the EELV fueled depot as making economic sense over Jupiter fueled.  That might seed the commercial launch industry, or free up Jupiter production capacity for other missions, but not financial efficiency by itself. 

That’s all…hopefully I’m wrong.

[kraisee]

Other than that Chuck, briefly what if any advantages are there to this cluster configuration over the inline in terms of cost and performance.

If I may...

We're seeing close to a ton improvement in overall structure mass, which is always a nice thing to have in your pocket

And from a cost perspective, this design can be implemented using many more "common" parts throughout the structure than the linear arrangement we had before.

Let me preface what I say below by stating that I'm not a designer myself, so my understanding of this design is fairly rudimentary and I might have mis-understood some of the nuances, so I apologize in advance for any such mistakes herein!

The basic thrust structure is made up from a conical arrangement made from:

4 Identical Thrust Panels (vaguely similar to the Interstage Thrust Panels)
4 Identical high-load Thrust Beams
2 I-Beam Bracing Members

This means that the primary structure can essentially be made from just 3 production parts -- which reduces the production costs significantly.

Yes, in addition to that there are Ring Frames on the inside (and probably even Longarons too, not shown in the image above) and non-load-bearing Aero-Panels which are added on the outside (again, many being common design so relatively high-volume production comes into effect), plus various forms of TPS, but you get the idea

Ross.

[Lobo]

Cost: Being a relatively standard design, it is less expensive to build.
Performance: Because this thrust structure would weigh less, performance is increased by that difference - not a lot, but nothing to sneeze at.

Cool,
Every little bit helps when you are talking about thousands of dollars per kg. 
Are the off-balance forces better or worse for the J-130 cluster v.s the J-130 inline?

[robertross]

As to your Step 3, two things.  First I think the concept of a constant flow of EELV rockets to a depot to fill it up is a victim of Direct’s own darned efficiency.  You’ve produced an HLV option that can do what 6-8 EELV class launches are needed to do.  The math just doesn’t pencil out where the EELV’s would be cheaper (or even competitive) with a single J-246 launch.  Your vehicle is too darn affordable for that role!  You are saying, “yea, I know it’d save a lot of time and money and logistics to use our new rocket, but we need to use a bunch of smaller rockets instead!”.   And I think people are looking at that going, “What?...”

But even then, unless other parties pony up the money for all those EELV fueling flights, the Jupiter core is still the best way to get the fuel up there…so maybe MAF could be expanded to produce more cores for cheaper than could be done with commercial contracts for EELV class flights?

If I may...

Considering the standing army required to process & launch all those 'NASA' vehicles, along with the facilities and such, I think it's still better to outsource all that. NASA has an important role to play, and that should be manned spaceflight & enabling technologies. For instance: maximize the upmass by partial-filling the depot; it needs cryos anyway for tank conditioning, the more the merrier.

Even if the costs eventually break even, it's better to not worry about the hassles of a propellant launch. Besides, it allows other players as well. You can't do that without outsourcing.

[Lobo]

Other than that Chuck, briefly what if any advantages are there to this cluster configuration over the inline in terms of cost and performance.

If I may...

We're seeing close to a ton improvement in overall structure mass, which is always a nice thing to have in your pocket

And from a cost perspective, this design can be implemented using many more "common" parts throughout the structure than the linear arrangement we had before.

Let me preface what I say below by stating that I'm not a designer myself, so my understanding of this design is fairly rudimentary and I might have mis-understood some of the nuances, so I apologize in advance for any such mistakes herein!

The basic thrust structure is made up from a conical arrangement made from:

4 Identical Thrust Panels (vaguely similar to the Interstage Thrust Panels)
4 Identical high-load Thrust Beams
2 I-Beam Bracing Members

This means that the primary structure can essentially be made from just 3 production parts -- which reduces the production costs significantly.

Yes, in addition to that there are Ring Frames on the inside (and probably even Longarons too, not shown in the image above) and non-load-bearing Aero-Panels which are added on the outside (again, many being common design so relatively high-volume production comes into effect), plus various forms of TPS, but you get the idea

Ross.

Cool, things just keep looking better and better as the design is refined for Jupiter.  I guess when you start out with a solid concept, things tend to fit together better, rather than spiraling apart like a certain other new rocket design I can think of.

Every kg and $ saved is another kg and $ you can put in orbit for your mission.  So that’s really good news.
Ideally, you want your booster to be a cheap and simple as possible as it just gets you where you are going, then you throw it away.  Like a home contractor who buys a van for their business.  You just want something that will get your tools to the jobsite.  The cheaper that is, the more and better tools you can afford and get to the jobsite to actually do the job you are getting paid for.
You don’t throw away the van, but I think you understand the analogy.

Now, if we can make the powers that be see that.  If it was a snake, it’d have bit them!

[clongton]

Myself, the point I was making is a lot of people call that a “single launch” approach, but really it takes several EELV launches to make that happen, so it’s more of a “6-8” launch approach, depending on the size of the EELV class rockets used.  So it adds logistics and complexity to a mission, rather than reducing it.

Although the scale may be wrong, consider taking a cross-country trip in your car. You load up your car with enough of everything you're going to need for the entire trip - except gasoline. For that you intend to stop along the way to refuel. Now everybody knows that the underground fuel tanks at the station are constantly being refilled by a fleet of tankers. So is your cross-country trip a one-vehicle trip or an 8-vehicle trip? It's the same with the Jupiter single launch profile. Because the propellant infrastructure has been removed from the launching agency's cognizance, the only thing the launching agency cares about is that the gas is there when needed, not how many tankers it took to make that happen.

Like I said, the scales are wrong, but the principle I'm expressing is exactly the same. Once you separate the fuel from the launch infrastructure, it doesn't matter to the launching agency how many tankers were needed to fill the depot. The tankers are somebody else's problem. The only thing that matters *to the launching agency* is that the needed propellant is available at the depot when you arrive to fill up, just like pulling your car into a gas station. When is the last time you went to a gas station and wondered how may tankers had visited the station in the last month?