Big Dumb Boosters- and why we're overthinking this whole rocketry business

[Bill Phil]

How does that have anything to do with the BDB concept? It's not even Big, that's basically just Vega, CZ-11 or Minotaur-IV.

It's big compared to a lot of things. Big is relative.

The BDB concept has more to do with cheaper designs and sloppier engineering than size.

You're also forgetting that there have been very large solid stages. Some many meters in diameter. Like the AJ-260-2.

Edited October 1, 2015 by Bill Phil

[78stonewobble]

Ah this thread is still here... Without wanting to reread the whole thread again.

I do have a question though...

What is the biggest cylinder size, that is currently able to be mass produced, to "near rocket use quality"?

Sure, there are rockets, but these aren't exactly mass produced per say and aircraft and submarines, while having cylindrical sections are composed of many sections. So I'm thinking some kind of industrial pressure vessel?

[Kryten]

It's big compared to a lot of things. Big is relative.

The BDB concept has more to do with cheaper designs and sloppier engineering than size.

You're also forgetting that there have been very large solid stages. Some many meters in diameter. Like the AJ-260-2.

Big solids are extremely difficult to move, requiring either complex segmenting hugely expensive infrastructure. You'll note the AJ-260-2 was fired in the pit it was cast in, because they simply could not move it.

[Motokid600]

Not to mention how deep do the engines have to throttle to keep burn out twr down?

[Bill Phil]

Big solids are extremely difficult to move, requiring either complex segmenting hugely expensive infrastructure. You'll note the AJ-260-2 was fired in the pit it was cast in, because they simply could not move it.

And that's exactly why I kept the size small.

- - - Updated - - -

Not to mention how deep do the engines have to throttle to keep burn out twr down?

I don't know, but solids do have grain geometries that can give specific thrust curves.

[Kryten]

And that's exactly why I kept the size small.

SO how is this any different from Vega or PK-SLV?

[sgt_flyer]

Managing the TWR with solid boosters is not really a problem - depends on the shape of the hole left for the solid fuel to burn. Basically, SRBs TWR change with how much surface area burns at the same time (more fuel burning at the same time - more gases for the exhaust - more thrust). (A star shaped hole has much more surface area than a cylinder shaped hole) - once the wedges around the star's branches burned completely however, the remaining surface area is much closer to a cylinder, so you have less fuel burning at the same time - and less thrust.

There's a lot of various shapes used in solid fuels to achieve various thrust curves during the whole SRB burn

[Bill Phil]

SO how is this any different from Vega or PK-SLV?

It's designed to use common materials and be much easier to build. Sloppier design.

Steel casing, common propellant(s), optimized for cost, huge margins, things like that.

Engineering for rockets of most types is often very precise. BDBs do away with that, but still use liquids, which are fairly complex. Solids are (relatively) simple, and (relatively) cheaper. So, using solids would allow it to be even more simple. The size limit is mostly for practicality's sake. It's also much easier on the logistics end.

[Kryten]

It's designed to use common materials and be much easier to build. Sloppier design.

Steel casing, common propellant(s), optimized for cost, huge margins, things like that.

So how is it supposed to precisely put sats into sun-synchronous orbit, which is almost the entire demand for this class of rocket?

[PB666]

One reason not to use Larger boosters is the problem with speed control. If your rockrt is shaped like an SR71 this is not a problem.

The space shuttle coupled SSME to boosters and could control force/mass during the first 90 seconds, largely because the coonections could not handle the stress of breaking mach at low altitude. Remember that main engine throttle down at maximum dynamic stress, that dynamic comes from drag, and the stress was largely created and maintained by the boosters.

At least in the game you could use fat boosters just make you parts more aero or they will want to flip. Then of course lower the CoD for the leading part. One way to get around this is to have big smart booster in which the available solid fuel over time is structured to decline so that thrust begins to fall off as maximum dynamic pressure is reached.

This is how i think about the problem. I have all these nice efficient engines that work well in space, and the i have a rocket sitting on a launch pad. How can i get my rocket high enough so that i can use thos light weight but efficient rockets. The solution is to modulate the booster phase

This basically means boosters are a composite, so we close the game and enter RL 'well it worked in KSP' :^}

Since th perfect launch to high altitude booster is a composite ..... drumroll.........you don't have to transport an interstate crusher to the launchpad or build the entire thing on site. You can compose it on site from managable pieces.

[sgt_flyer]

you have less complexity in SRBs for sure, but you have to manage the increased risks associated with moving loaded SRBs

and the increase in weight also mean you need logistic systems capable of handling this weight - NASA had to reinforce the crawlers / crawlerways when they converted the Saturn V pads to shuttle pads to handle the weight of the loaded SRBs - whereas Saturn V was empty when moving it to the pad

One way to minimize the SRBs logistics problems is to cast the solid fuel in the SRBs next to the assembly facilities - but then if you have to multiply casting facilities, you're losing on bulk production capabilities if the customer want to launch from different launch sites (ex, he needs to do polar orbits / or normal ones)

@ kryten - i agree on that - you need some form of precise control at least for the upper stage - (even solids could achieve that with gimbaled nozzles and blowout panels to stop the thrust at a specific point - but that's not BDB anymore at this point - requires very precise engineering to get this kind of precision from SRBs) - no wonder VEGA went with a liquid upperstage here )

Edited October 1, 2015 by sgt_flyer

[SuperFastJellyfish]

Managing the TWR with solid boosters is not really a problem - depends on the shape of the hole left for the solid fuel to burn. Basically, SRBs TWR change with how much surface area burns at the same time (more fuel burning at the same time - more gases for the exhaust - more thrust). (A star shaped hole has much more surface area than a cylinder shaped hole) - once the wedges around the star's branches burned completely however, the remaining surface area is much closer to a cylinder, so you have less fuel burning at the same time - and less thrust.

There's a lot of various shapes used in solid fuels to achieve various thrust curves during the whole SRB burn

Visualization!

[kStrout]

The big dumb booster concept is only really useful if what your launching is really cheap, but needed in large quantities(ex water). No one cares if they lose kilotons of water on launch, they'll just demand a refund and fly on the next launch. If NASA did this on a regular basis by selling out to private companies I expect we would see serious development in the heavy lift vehicle department.

[Bill Phil]

So how is it supposed to precisely put sats into sun-synchronous orbit, which is almost the entire demand for this class of rocket?

Either a small pressure fed stage or a small hybrid engine on the last stage. With about 100m/s Dv with payload. Not counted as operational payload.

Edited October 1, 2015 by Bill Phil

[Kryten]

And some very expensive avionics, right? So what's the point of the rest of it, and how is it different Vega?

[Finox]

...So, why do we waste so much effort on ultra-engineered rocket designs, made out of fragile composites and high-tech alloys, when it would be a lot cheaper to slap something together out of corrugated steel and just launch the payload to orbit on THAT?...

Having been interested in WW2 and Cold War aviation history my whole life I can tell you that the first rockets were designed not for cost but for range and payload. The military didn't care about cost, just that the rocket (or missile really) in question could deliver its warhead the required distance with a certain modicum of accuracy. The early ICBM's did not have very good readiness rates, they often expected only around 50% of their missiles to really work should they be called upon, but that was enough for the war-planners. Since the early space programs of the U.S. and USSR utilized military rockets at the beginning, they just carried over the habits they inherited from the missile makers. It wasn't until SDI came along that many saw a real need for cheaper access to space, lookup DC-X, it's the vehicle that really started this discussion.

Basically we make high-tech expensive rockets because of an old habit inherited from the military that few cared to question until relatively recently.

Edited October 2, 2015 by Finox

[SomeGuy12]

I think the reason we make high-tech rockets so expensive is because the agencies who launch them lose face - they look bad when their funding comes up for review, they are a less reputable institution - when they fail. "Better, Faster, Cheaper" Supposedly launched 10 missions for the price of one, and succeeded about 50% of the time? That's an enormous boost in productivity, but people are still making jokes about that unit conversion bug that took out a mars probe.

[GoSlash27]

I think the reason we make high-tech rockets so expensive is because the agencies who launch them lose face - they look bad when their funding comes up for review, they are a less reputable institution - when they fail. "Better, Faster, Cheaper" Supposedly launched 10 missions for the price of one, and succeeded about 50% of the time? That's an enormous boost in productivity, but people are still making jokes about that unit conversion bug that took out a mars probe.

I disagree. If it were that simple, pure market forces would hold sway and we'd have dirt- cheap launchers with a decent chance of success. Companies pay these agencies to "deliver" their packages to orbit. Insurance is expensive and the whole business model hinges on an agency's ability to deliver the payload 1) reliably and 2) cheaply. In that order.

Reliability is many times more important when human lives rely on the success of the mission.

Optimizing for cost while skimping on reliability is theoretically good, but nobody's willing to pay to take that risk.

This holds true even in the more mundane world of ground shipping. For inconsequential packages, it's probably fine to go with a cut- rate method with an iffy chance of success, but people will pay a premium when they want to be sure of delivery.

Best,

-Slashy

[peadar1987]

I disagree. If it were that simple, pure market forces would hold sway and we'd have dirt- cheap launchers with a decent chance of success. Companies pay these agencies to "deliver" their packages to orbit. Insurance is expensive and the whole business model hinges on an agency's ability to deliver the payload 1) reliably and 2) cheaply. In that order.

Reliability is many times more important when human lives rely on the success of the mission.

Optimizing for cost while skimping on reliability is theoretically good, but nobody's willing to pay to take that risk.

This holds true even in the more mundane world of ground shipping. For inconsequential packages, it's probably fine to go with a cut- rate method with an iffy chance of success, but people will pay a premium when they want to be sure of delivery.

Best,

-Slashy

Nail on the head. For most space launches, the most expensive part is actually the payload. It's probably some of the most sophisticated technology we have, designed to operate in the harshest of environments for possibly decades.

For things like supplies to the ISS, which could conceivably be a few thousand pounds worth of food, water, hammers and duck tape, this concept could really make sense. However, such a niche market does not currently justify the costs of getting an entire Big Dumb Booster industry off the ground.

[Northstar1989]

Nail on the head. For most space launches, the most expensive part is actually the payload. It's probably some of the most sophisticated technology we have, designed to operate in the harshest of environments for possibly decades.

For things like supplies to the ISS, which could conceivably be a few thousand pounds worth of food, water, hammers and duck tape, this concept could really make sense. However, such a niche market does not currently justify the costs of getting an entire Big Dumb Booster industry off the ground.

I'm amazed that this thread was revived last month, after such a long time. But, since it's active again, I guess I'll re-interate the key point I made earlier...

One word- fuel.

The vast majority of the mass of any mission is fuel. That expensive, 10 kg probe might be 20% of the mission budget, but it's sitting on 90 kg of fuel that has to get it from Low Earth Orbit to its destination- and that's not cheap. In fact, part of the reason that the payloads are so expensive in the first place is to lower the required fuel mass. Ion thrusters, high-ISP resistojets; why do you think anybody uses these? Because the additional cost of the more complex and high-tech engines is much less than the reduced launch-costs from not having to lift as much fuel to orbit, and thus being able to use a lighter (and therefore, cheaper) lifter.

With the low-reliability variety of Big Dumb Booster (REMEMBER- Big Dumb Boosters come in two varieties, the high-reliability ones like Sea Dragon, which are highly-reliable but just have terrible mass-fractions, and the low-reliability variants like Aquarius which have both terrible mass-fractions AND reliability, but *even lower* per-kg cost), you can tell the engineers: "hey, you've got 20 kg's dry mass to play with on the probe instead of 10 kg, and can do it with lower ISP engines too" and lift 300 kg of fuel on separate Big Dumb Boosters (which can easily achieve 1/4th the per-kg launch costs or less), and *STILL* do the whole thing cheaper than with a smart rocket that can launch 100 kg to orbit in a single go.

THAT is why that sub-type of Big Dumb Boosters make sense. It's not to launch the expensive parts of interplanetary or even trans-lunar missions, it's to launch the cheap parts: namely the fuel, food & water (for manned missions), extra coolant, spare parts (currently useful only for manned missions or ones operating close to manned stations- but DARPA is working on simple robotic repair probes that could perform some simple repairs and maintenance without the need for humans nearby...) etc. separate from the rest of the payload. You stock up more than what you need, and if the BDB fails to make orbit with the cheap payload, you just load another one up and try again (NOTE: you launch this stuff BEFORE the main mission, in case you need to try more than once...)

Regards,

Northstar

P.S. I think there's been a lot of confusion about this issue, since there are actually two *entirely different* types of Big Dumb Boosters discussed in this thread. Low-reliability variants, like Aquarius, that achieve the lowest cost-per-kg of payload to orbit (even after accounting for a high rate of launch-failures) ; and high-reliability Big Dumb Boosters like Sea Dragon, which aren't as cheap as rockets like Aquarius, but are still much cheaper than conventional "smart" rockets. High-reliability Big Dumb Boosters like Sea Dragon are perfectly safe to load extremely expensive payloads on. They don't necessarily fail more often than "smart" rockets, in fact if anything with their more generous engineering margins they should fail LESS often- they just are much, much larger (and cheaper) rockets on the launchpad compared to the size of the payload they're launching...

P.P.S. Sea Dragon was designed to launch larger payloads than Saturn V because it was designed with single-launch Mars missions in mind (to avoid the need for orbital docking of multiple components, like in the modern-day "Constellation" mission-plan). It was also designed to carry heavier payloads because the designers (correctly) assumed that if there was less price-pressure on the payload to shave every possible ounce off the mass, a heavier payload mass would result, as the optimal balance of launch costs vs. costs of shaving payload mass would be different with lower launch-costs. Let me emphasize that there is no reason you can't design a Big Dumb Booster to carry a payload in the size-range of Saturn V, SLS, Ares V, Falcon 9, or even the Falcon 1 if you wanted. Big Dumb Boosters of this type are so-named not because they are necessarily large (in absolute terms), but because they are much larger compared to the size of their payload than conventional "smart" rockets with the same payload-capacity. Thus, a Big Dumb Booster with the lifting-capacity of Saturn V would be smaller than Sea Dragon, but bigger than Saturn V was, and a Big Dumb Booster the size of Saturn V would have a much smaller payload-capacity than Saturn V, but a much lower per-kg launch cost...

Edited October 3, 2015 by Northstar1989

[Kryten]

Almost all customers want a large payload placed in geostationary transfer orbit with a high level of reliability; how much you can place into LEO with your fictional low mass-fraction booster is irrelevant in this case, because trying to boost further to relatively high-energy orbits like GTO with a low mass-fraction stage is a fool's errand.

[Northstar1989]

Almost all customers want a large payload placed in geostationary transfer orbit with a high level of reliability; how much you can place into LEO with your fictional low mass-fraction booster is irrelevant in this case, because trying to boost further to relatively high-energy orbits like GTO with a low mass-fraction stage is a fool's errand.

Kryten, I think you miss two fundamental distinctions here:

(1) The Big Dumb Booster's payload isn't necessarily large. Once more, the Big Dumb Booster derives its size from the relative size of the rocket to the payload. So, you can still launch your little 100 kg payload to LEO, you just might need a 12-ton Big Dumb Booster with a 0.75% mass-fraction to do it instead of a 3-ton "smart" booster with a 3% mass-fraction. The Big Dumb Booster will still be much cheaper despite its larger size, though, due to the much less precise construction/manufacturing techniques that the wider engineering-margins allow you to safely utilize without sacrificing reliability (if you are willing to take a hit to reliability as well, you can get even lower per-kg costs, with better mass fractions, though, like with Aquarius: but that is only suitable for cheap payloads like the fuel to get from LEO to GSO...)

(2) Anything that you launch to LEO is payload, by this definition. The "Big Dumb Booster" only encompasses the launch-stage and sustainer/upper stages themselves. Once the payload is delivered to orbit (or, to avoid generating debris, just shy of orbit) it separates from the orbital stage- which is the same, high-cost kind of engineering you're used to. Nobody is suggesting using these low mass-fraction construction techniques for the orbital stages, that would just be stupid.

Let me re-iterate, the satellite and its GSO transfer-stage (remember, two burns are required to travel from LEO to GSO: one to reach GTO, and a second to circularize thhe orbit once GSO altitude is reached) are considered part of the payload. What we are comparing here is the cost of getting a 100 kg satellite+transfer stage to LEO with a Big Dumb Booster vs. getting that same satellite + transfer stage to LEO with a "smart" booster. For the high-reliability variant of Big Dumb Booster, nothing differs after reaching orbit whatsoever.

(For low-reliability Big Dumb Boosters like the Aquarius, the transfer-stage is launched "dry", and must be fueled in LEO from an orbital fuel depot- which in turn is fueled with Aquarius-style launches. You *do not* launch the transfer stage or satellite on an Aquarius-style BDB, although you can launch them on a high-reliability, Sea Dragon style BDB, while launching the fuel separately and ahead-of-need Aquarius-style.)

Regards,

Northstar

[Kryten]

But you're not talking about a 100kg satellite and transfer stage, you're talking upwards of 25 tons satellite and transfer stage. You can only get good figures for loss mass fraction boosters for that if you ignore infrastructure and handling costs, which you are doing.

EDIT: You're also ignoring that the largest portion of the cost of a booster is the engines. If you want to combine low-ISP engines with low mass-fraction everything else, the rocket equation is going to mess you up badly.

Edited October 3, 2015 by Kryten

[Northstar1989]

Here is what a sample mission plan might look like, using the lowest-cost, BDB-based solution. Let's say that the payload is a 50 kg satellite with a 50 kg transfer-stage to GSO that only weighs 10 kg when launched "dry"... (so that's 50 kg of satellite, 10 kg of transfer stage, and 40 kg of fuel for the GSO-transfer, all to LEO as the payload)

Mission Vehicles:

(ROCKET-TYPE #1) Aquarius-style rocket with 20 kg payload-capacity, 0.75% mass-fraction, 66% success-rate, and $1250/kg to LEO

(ROCKET-TYPE #2) Sea Dragon style rocket with 60 kg payload-capacity, 0.75% mass-fraction, 99.8% success-rate, and $2500/kg to LEO

Launch #1: 20 kg of fuel to LEO depot on Rocket-Type #1. Mission Success.

Launch #2: 20 kg of fuel to LEO depot on Rocket-Type #1. Launch Failure.

Launch #3: 20 kg of fuel to LEO depot on Rocket-Type #1. Mission Success.

Launch #4: 60 kg payload of satellite + transfer stage to LEO depot on Rocket #2. Payload refuels at depot and proceeds to GSO. Mission Success.

Total Cost: $25,000 x 3 + $150,000 = $225,000

Compare this to a "traditional" mission:

Mission Vehicle:

(ROCKET-TYPE #3) Conventional "smart booster" with 100 kg payload-capacity, 3% mass-fraction, 99.7% success-rate, and $10,000/kg to LEO

Launch #1: 100 kg of satellite + transfer stage + fuel to LEO. Payload separates in LEO and proceeds to GSO. Mission Success.

Total Cost: $1,000,000

Note the following:

- Using the "smart" booster the total mission cost is more than 4x as high.

- Progressively larger "smart" boosters require larger launchpads. Both styles of Big Dumb Booster are designed to launch at-sea, so you can scale them as large as you want without having to build/upgrade a bigger/better launchpad.

- The total chance of mission-success is actually HIGHER with the Big Dumb Boosters. This is because Rocket-Type #2 is actually MORE reliable than Rocket-Type #3, in the same way that Sea Dragon type Big Dumb Boosters would have been MORE reliable than "smart" boosters with a comparable payload-capacity, due to having very wide engineering-margins...

- The LEO Depot required for Rocket-Type #1 (the Aquarius-style rocket) is actually factored into cost-per-kg estimates, the same way that launchpad costs are factored into the cost of a typical "smart" booster. However, an Aquarius-style rocket assumes that the LEO fuel depot was purpose-built for storing fuel launched to LEO on low-reliability rockets (specifically, it assumes an annual launch-mass of cheap consumables equal to the average annual mass of cheap consumables such as food/water used by the ISS) and serves no other purpose. The LEO depot is a piece of orbital infrastructure capable of storing both fuel and consumables, however, and could also be useful for storing propellant harvested from Near-Earth Objects (specifically, ice and hydrate-rich asteroids), Lunar ice-mining, etc. Any shared utilization of this depot with other projects, or any mass launched on it in excess of the annual consumables-budget of the ISS would amortize the setup-costs of the depot (which comprise nearly half the $2500/kg cost-estimate of the Aquarius-style rocket, as it was assumed by the engineers that the depot would be launched on a $10,000/kg "smart booster") over more missions, resulting in a further cost-per-kg decrease for Rocket-Type #1. Also, if a Sea Dragon style booster were used to launch the depot in the first place (which is expensive, and therefore must be launched on a high-reliability booster) the cost/kg of Rocket-Type #1 would be lower to begin with...

Regards,

Northstar

- - - Updated - - -

But you're not talking about a 100kg satellite and transfer stage,

Yes, yes you are.

you're talking upwards of 25 tons satellite and transfer stage.

No, not necessarily. We can launch 100 kg satellites if we want, and I prefer to use them as illustrative examples to make it clear that Big Dumb Boosters can be scaled to any requisite size. Sea Dragon and Aquarius were only prototypes for entire classes of rockets, they were not the only sizes their respective styles of Big Dumb Booster could be scaled to.

You can only get good figures for loss mass fraction boosters for that if you ignore infrastructure and handling costs, which you are doing.

No. The cost-figures I gave pages back are the actual total-cost estimates for Sea Dragon and Aquarius (and the #'s in the scenario above are based on even more conservative #'s). Both rockets launched from the sea to avoid the requirement for launchpad-infrastructure (Aquarius because of its poor range-safety, Sea Dragon because of its massive size which was larger than any existing launchpad). All other costs, including crawlers, barges, etc. were already factored in.

EDIT: You're also ignoring that the largest portion of the cost of a booster is the engines. If you want to combine low-ISP engines with low mass-fraction everything else, the rocket equation is going to mess you up badly.

Mass-fraction, by definition, is the fraction of the total rocket that is payload. This figure takes into account the ISP of the engines, mass of the fuel-tanks, and everything else.

Big Dumb Boosters had lower mass-fractions, but not insanely so. Their designers knew that the Rocket Equation still applied, and designed with it in mind. They just cut down mass-fractions by a factor of about 4 (hence the 3% vs. 0.75% earlier) and used the wider engineering-margins this allowed to utilize lower-cost manufacturing methods and materials for the rockets. These were still very expensive pieces of equipment, even if they were 1/5th, 1/20th, or less the cost of their "smart booster" brethren (the cost-per-kg for Sea Dragon and Aquarius still had to deal with infrastructure costs, and in the case of Aquarius, the cost of a LEO fuel/consumables depot. This was *all* factored into the final cost-per-kg given for each rocket of 1/4th and 1/8th the cost-per-kg of a "smart" booster...)

Regards,

Northstar

Edited October 3, 2015 by Northstar1989

[Kryten]

You can't prove anything with numbers you've just pulled out of your arse, especially for something like sea dragon; it's full of 60s-era assumptions about big hydrolox engines that we know don't true today. You might remember people actually tried making a rocket with a relatively cheap hydrolox engine and poor mass fraction; it's called Delta IV. You might also remember it's one of the most expensive rockets available, because they fudged both the infrastructure and handling costs as well as demand, exactly like you're doing.