ARCA Expendable Smallsat SSTO

[Exoscientist]

On 4/5/2017 at 3:34 AM, Streetwind said:

.

Rocket Lab is about to launch a full composite orbital rocket, possibly as soon as this month even.

 

  Hadn't heard that. Hope they are successful.

 

     Bob Clark

[tater]

Their non-aerospike is sort of interesting, actually. Might have to drive down to Cruces and poke around...

[Kryten]

There's no way this is happening. Sorry to be so pessimistic about it, but the last however many years or ARCA activities leaves no other plausible option.

[sevenperforce]

27 minutes ago, Kryten said:

There's no way this is happening. Sorry to be so pessimistic about it, but the last however many years or ARCA activities leaves no other plausible option.

Yeah, ARCA doesn't seem particularly reliable, but their design is promising even if their track record isn't.

[kerbiloid]

Their video list on youtube is not very space-related.

[Firemetal]

Looks nice but what good is an SSTO if it isn't reusable? The whole point of making it single stage is to make it easier to reuse.

Fire

[Shpaget]

2 hours ago, Firemetal said:

The whole point of making it single stage is to make it easier to reuse.

I'd say the main goal is to get stuff up there as cheaply as possible. Number of stages and reusability are not the goal in itself. It's a method of reducing cost.

If it's cheaper to make a new rocket than to reuse one, then any sane businessman will make new rockets, not waste money reusing old ones.

[kerbiloid]

Optimists are having the inner struggle between "Wow! Reusable rockets!" and "Wow! 3d printers make everything cheap!"

[wumpus]

6 hours ago, Shpaget said:

I'd say the main goal is to get stuff up there as cheaply as possible. Number of stages and reusability are not the goal in itself. It's a method of reducing cost.

If it's cheaper to make a new rocket than to reuse one, then any sane businessman will make new rockets, not waste money reusing old ones.

While reusability isn't a sure thing, number of stages needed to get to orbit is known pretty well to be >1.  A company with a poor record of getting things done combined with a plan that ignores the rocket equation implies they are looking for gullible and uninformed businessmen.  Sanity is purely optional.

As a side note, the "sane businessman" will only do this as a private launcher.  A government contractor (like those who built the shuttle) likely understood (along with NASA) that having an expensive reusable shuttle meant that Congress would keep the thing flying due to the sunk cost fallacy.  This might even work with stock shareholders, although it is somewhat harder.  If your goal is to get to orbit more times on somebody else's money (only Bezos is capable of entirely paying his own way), reusability has some hidden attractive features.

[sevenperforce]

9 hours ago, Firemetal said:

Looks nice but what good is an SSTO if it isn't reusable? The whole point of making it single stage is to make it easier to reuse.

As others have stated, the point is getting your desired payload into your desired orbit as cheaply as possible. This changes things. For example, that's why cost per kilogram isn't always the most accurate determiner. If Company A charges $10M to put up to 500 kg into LEO and Company B charges $6M to put up to 150 kg into LEO, then Company B is twice as expensive per kilogram...but if my bird is only 120 kg, I'm sure as heck not spending an extra four million bucks just for the fun of it.

Same with expendable vs reusable launchers. An RLV is only cheaper if the cost of recovery, refurbishment, and recertification, plus manufacturing costs amortized over the expected lifetime, is less than the cost of a completely new ELV. Recovery, refurbishment, and recertification costs all get passed on to the launch purchaser.

An RLV must reserve payload margin for recovery (TPS and legs/gear, and either parachutes or additional landing propellant). So, if you assume the same engine performance and tankage ratio, a single-stage RLV will be larger than a single-stage ELV for the same net payload. This drives up vehicle cost and fixed operating costs.

For example, consider an ELV that can put 100 kg into orbit for $1M. Let's say half that ($500K) is fixed operating costs, profit margin, and fuel, while the other half is the cost of the expendable vehicle.

Then consider an RLV with the same performance and structural ratios. If it needs to reserve 30% of its payload capacity for reuse, then it needs to be 50% larger. This means the manufacturing costs and operating costs also go up by 50%. So now it costs $750K to operate and $750K to construct.

Let's estimate refurbishment, recovery, and recertification at 20% of the cost of a new LV, or $150K. Now the fixed operating costs for each launch are up to $900K. In order to beat the ELV, the $750K sticker price to construct the LV will have to be amortized over at least eight launches with no major part replacements in order to break even. However, an RLV program is more expensive to develop than an ELV program, so the launch company needs additional profit margin to recoup its investment.

That's how an ELV SSTO can beat out an RLV SSTO.

Now, ARCA seems to think that the complexity of staging drives up the costs of an ELV, and they're partly right. There are fixed manufacturing costs associated with building a stage that are more or less independent of the stage's size, up to a point, which is why it is cheaper to build a single stage with larger engines and larger tanks than it is to build two distinct stages. You only need one engine, one avionics package, and so forth.

Doesn't give me any confidence in ARCA's ability to pull it off, but the configuration seems sound...if they can meet their structural mass ratios, that is.

[sevenperforce]

1 hour ago, wumpus said:

While reusability isn't a sure thing, number of stages needed to get to orbit is known pretty well to be >1.  A company with a poor record of getting things done combined with a plan that ignores the rocket equation implies they are looking for gullible and uninformed businessmen.  Sanity is purely optional.

Eh, the rocket equation is just math, and the math checks out. If ARCA is ignoring anything, it's the demonstrated limitations of structural mass fractions. I'm skeptical of their ability to fit a linear aerospike engine, a pressurant system, fairings, and pressure-fed tanks for a 16.3-tonne-GLOW rocket into 550 kg of dry mass.

Unrelated note: I wonder if it would be possible to use liquid hydrogen or liquid methane as a pressurant. Obviously you'd still need liquid helium for your oxidizer, but if you could get the same performance out of LH2 or CH4, and your engine could handle changes in fuel flow density, you could end up with a de facto tripropellant rocket that converted to a higher-isp fuel once you were nearing orbit.

Edited April 7, 2017 by sevenperforce

[Elukka]

On 4/7/2017 at 7:08 AM, Firemetal said:

Looks nice but what good is an SSTO if it isn't reusable? The whole point of making it single stage is to make it easier to reuse.

Fire

I think SSTO and reusability are two goals that make each other harder. SSTO is a performance penalty and reusability is another performance penalty on top of that. You're basically adding two sets of razor thin margins together. If you have two stages you do have two different recovery systems and events, but you have so much more margin to spend on it it's a lot more practical.

[sevenperforce]

4 hours ago, Elukka said:

I think SSTO and reusability are two goals that make each other harder. SSTO is a performance penalty and reusability is another performance penalty on top of that. You're basically adding two sets of razor thin margins together. If you have two stages you do have two different recovery systems and events, but you have so much more margin to spend on it it's a lot more practical.

Yeah. Recovery and reuse is easier if your vehicle is physically smaller, but SSTO can only get positive payload fractions when it is very, very large.

[sevenperforce]

I was thinking about this, and for a proposed expendable SSTO, the question is really whether the simplicity of an SSTO actually saves you money. If a TSTO can deliver the same payload for a lower expense than an SSTO, then SSTO makes no sense. If not, it makes sense.

So all I need to do is compare ARCA's SSTO to a hypothetical TSTO using the same performance specs and margins.

First, I'll verify ARCA's specs. Since we are dealing with altitude compensation I will take the dV to orbit as 9,150 m/s and just use the vacuum ISP of the aerospike engine (314 seconds) for the entire trip. They claim a GLOW of 16,290 kg, a payload of 100 kg, and a dry mass of 550 kg. Plugging everything into the rocket equation, we get 833 kg to orbit, so their numbers (at least) aren't unreasonable. If we use their payload numbers and solve for dV, we get 9,913 m/s which will be our baseline for gravity drag, aerodynamic drag, and impulse losses.

So what can we get for a balanced TSTO delivering the same payload? We don't know the claimed TWR or mass of their engine, which will make our estimate a little more guessy, but we can just adopt their overall GLOW/dry-mass (29.62) and hope for the best. Let's split that 9,913 m/s up equally between the two stages: 4957 on the second stage and 4956 on the first stage.

Second stage wet to dry ratio needs to be 5:1, so if our payload is still 100 kg, then total second-stage mass at staging needs to be 601.7 kg (481.4 kg of propellant, 100 kg of payload, and 20.3 kg of dry mass). This 601.7 kg becomes the payload on the first stage.

First stage wet to dry ratio also needs to be 5:1 (remember, balanced TSTO), so our GLOW needs to be 3,620 kg (2,896 kg of propellant and 122.4 kg of dry mass).

This is ridiculous. Propellant costs for a TSTO using the same performance and margins as ARCA's would be a scant 22% of the SSTO, and dry mass would be just 26%. Even if there were substantial square-cube losses, that's not anywhere close. There's no way it would cost more to do a TSTO than it would to do an SSTO.

Why do I get the feeling that ARCA simply set the performance numbers up and then back-calculated to get the mass ratio that they needed?

Their claimed GLOW/M_dry is just a hair under 30. By comparison, the ITS Tanker proposed by SpaceX comes in at a GLOW/M_dry of 28.8, and that's with the highest-TWR turbopump-driven liquid-fuelled engines in the world, non-pressure-fed tanks, and massive square-cube advantages.

Not gonna happen.

[Steel]

15 minutes ago, sevenperforce said:

There's no way it would cost more to do a TSTO than it would to do an SSTO...

Except for developmental costs and cost to manufacture (which, I think, is where ARCA are coming at this from). IMO their thinking is that a big chunk of cost will be saved by not having to develop and manufacture two engine types, two different tank and plumbing set-ups and the associated systems required for stage separation e.t.c as well as not having to do so much work to make sure the mission flies successfully (i.e if there's only one stage, they don't need to spend time and resources working out how to separate successfully, light the second stage engine, make sure that the fuel settles e.t.c ).

Edited April 12, 2017 by Steel

[sevenperforce]

36 minutes ago, Steel said:

Except for developmental costs and cost to manufacture (which, I think, is where ARCA are coming at this from). IMO their thinking is that a big chunk of cost will be saved by not having to develop and manufacture two engine types, two different tank and plumbing set-ups and the associated systems required for stage separation e.t.c as well as not having to do so much work to make sure the mission flies successfully (i.e if there's only one stage, they don't need to spend time and resources working out how to separate successfully, light the second stage engine, make sure that the fuel settles e.t.c ).

ARCA projects a million bucks per launch. If we assume a 10% profit margin, and a bit of quick number-crunching gives me a propellant cost of $135,000 per launch, then we arrive at an operating-and-manufacturing cost of $765,000 per launch.

I find it hard to believe that a company which can build and launch a single-engine 550-kg rocket for $765,000 could not build and launch a two-engine 143-kg rocket for less than $870,000 (the number to beat, given far lower propellant costs for the TSTO). Even if the first stage still cost the same $765,000 as the whole SSTO, the second stage would only cost $28,000 if you use the same price per kilogram for construction and operation.

Edited April 12, 2017 by sevenperforce

[monophonic]

18 minutes ago, sevenperforce said:

ARCA projects a million bucks per launch.

I emphasised the imperative word for you. Fixed costs like range safety, mission control or facilities maintenance don't depend on the number of stages. Or if they do they go up with the stage count. For a relatively rare use like SLV launches those costs really do add up.

Not to mention needing to pump up the profit margin to cover for the additional development costs or the associated risk with that develoment. You are probably right that a two stager could be more efficient, but for a smaller player with limited funding taking the safe route can be a good move. Safe here being defined as "not developing any technology that can be done without." Such as staging. Problems with getting that to work could bring them to bankruptcy before they reach orbit at all.

[sevenperforce]

1 hour ago, monophonic said:

I emphasised the imperative word for you. Fixed costs like range safety, mission control or facilities maintenance don't depend on the number of stages. Or if they do they go up with the stage count. For a relatively rare use like SLV launches those costs really do add up.

Not to mention needing to pump up the profit margin to cover for the additional development costs or the associated risk with that develoment. You are probably right that a two stager could be more efficient, but for a smaller player with limited funding taking the safe route can be a good move. Safe here being defined as "not developing any technology that can be done without." Such as staging. Problems with getting that to work could bring them to bankruptcy before they reach orbit at all.

I get that, but we aren't talking about a 15-20% margin difference; we're talking about a 74% decrease in vehicle dry mass. That's insane.

Even if you add hefty square-cube losses, you'd only increase total TSTO dry mass to 40% of the SSTO variant.

I think the more likely culprit is that they are simply far overestimating how tight they can get the dry mass ratio. 

[monophonic]

15 hours ago, sevenperforce said:

I get that, but we aren't talking about a 15-20% margin difference; we're talking about a 74% decrease in vehicle dry mass. That's insane.

Even if you add hefty square-cube losses, you'd only increase total TSTO dry mass to 40% of the SSTO variant.

I think the more likely culprit is that they are simply far overestimating how tight they can get the dry mass ratio. 

That is an insane difference - I doubt anything like that would be possible. Remember in addition to square-cube-losses in two stage design you have the interstage, which is basically launch clamps for the second stage. Also it has to carry the full weight of the second stage against the maximum acceleration from the first stage. The clamps on the pad only need to hold the entire assembly against a measly 1g. That is a non-negligible mass we haven't yet accounted for. Oh, and not only the interstage, but the first stage as a whole has to carry that weight - fuel, another engine, the interstage itself - while the SSTO tank/stage only has the payload + shroud + adapters burdening it. So it needs to be beefier than the lower half of the SSTO has to be.

That said ARCA's numbers do seem awfully tight to me too. I'm not trying to convince you that their plan is in any way inherently better than a two stage approach - just that the opposite isn't necessarily true either.

[magnemoe]

15 hours ago, monophonic said:

That is an insane difference - I doubt anything like that would be possible. Remember in addition to square-cube-losses in two stage design you have the interstage, which is basically launch clamps for the second stage. Also it has to carry the full weight of the second stage against the maximum acceleration from the first stage. The clamps on the pad only need to hold the entire assembly against a measly 1g. That is a non-negligible mass we haven't yet accounted for. Oh, and not only the interstage, but the first stage as a whole has to carry that weight - fuel, another engine, the interstage itself - while the SSTO tank/stage only has the payload + shroud + adapters burdening it. So it needs to be beefier than the lower half of the SSTO has to be.

That said ARCA's numbers do seem awfully tight to me too. I'm not trying to convince you that their plan is in any way inherently better than a two stage approach - just that the opposite isn't necessarily true either.

True, however lower stage mass don't hurt payload capacity much and after staging it has no effect. 

Using an solid first stage has the benefit that you can optimize upper stage for vacuum isp and don't need much twr. 
 

[Shpaget]

20 hours ago, monophonic said:

That said ARCA's numbers do seem awfully tight to me too. I'm not trying to convince you that their plan is in any way inherently better than a two stage approach - just that the opposite isn't necessarily true either.

But it is.

Using the best of our current and near future realistic materials, SSTO is just marginally possible. Multistage will always give better payload fraction. That's just math.

[sevenperforce]

4 hours ago, Shpaget said:

But it is.

Using the best of our current and near future realistic materials, SSTO is just marginally possible. Multistage will always give better payload fraction. That's just math.

Indeed.

In theory, there might be a size threshold where the vehicle is simply too small to stage efficiently. But at that point, square-cube losses would dominate anyway.

[magnemoe]

1 hour ago, sevenperforce said:

Indeed.

In theory, there might be a size threshold where the vehicle is simply too small to stage efficiently. But at that point, square-cube losses would dominate anyway.

Think its other factors to, some solid fuel icbm have lots of stages, even the abm version of the standard missile has 4 stages including warhead. 
Yes its probably to improve performance then size is fixed. Also it cost little to add stages to solid fuel rockets. 
Liquid fuel engines are expensive, bulky and having more stages increase the error rate. 

[monophonic]

8 hours ago, Shpaget said:

But it is.

Using the best of our current and near future realistic materials, SSTO is just marginally possible. Multistage will always give better payload fraction. That's just math.

That is true. But cows aren't spherical and you can't build a rocket out of just math. If a marginal system costs less than you can afford and a superior system costs more than you can afford, you pick the marginal system. That's just math.

You can't pick just one side of the problem and decide what's best from that angle is best overall. You have to consider all the other sides too. For another company two stage design could certainly be best. For ARCA it evidently wasn't or they have a very incompetent CEO.

Going SSTO is saving them design costs on a second engine design and staging system. That may be the difference between them getting into launcher business or not. It could also be an attempt to cut on development time to get the income flow started sooner. It might even be just a risk control measure, less money lost if the business doesn't fly. That is their business secret though so we are unlikely to know.

I know this is a science subforum, but we are discussing a commercial venture. I find it unfit to focus solely on the physical inefficiency of their design without giving any consideration to any other aspects of it.

[sevenperforce]

37 minutes ago, monophonic said:

For another company two stage design could certainly be best. For ARCA it evidently wasn't or they have a very incompetent CEO.

Well I think there is a very good chance that they have a very incompetent CEO...

Hint: if you can't decide whether your company manufactures hoverboards, skateboards, or launch vehicles, you might be an incompetent CEO.

39 minutes ago, monophonic said:

I know this is a science subforum, but we are discussing a commercial venture. I find it unfit to focus solely on the physical inefficiency of their design without giving any consideration to any other aspects of it.

Naturally. But you'll note that I did address the cost-benefit analysis above; I just don't see any way doing a TSTO could possibly be more expensive if their mass fractions really are as stellar as they claim.