Countdown to new smallsat launchers

[Kosmos2001]

Wow!  Those companies are growing like mushrooms in autumn after a rainy summer.

And most will fade away just as quickly.

I also think so, yes. There's no market yet for so many launcher companies. At least experienced engineers could recycle and join the surviving companies.

[gongora]

[Kitsap Sun] Secretive aerospace firm to test rocket engines in Bremerton

A Renton-based aerospace firm will begin testing rocket engines next year in a facility under construction at the Port of Bremerton.

Radian Aerospace is involved in research and development of “aerospace hardware to serve a variety of customers,” according to a company representative.
...
Incorporated in 2016, Radian Aerospace shares leadership with Holder Aerospace, a Renton company headed by former astronaut Livingston Holder and aerospace executive Curtis Gifford.

Anyone know what's up with this company?

[TrevorMonty]

[Kitsap Sun] Secretive aerospace firm to test rocket engines in Bremerton

A Renton-based aerospace firm will begin testing rocket engines next year in a facility under construction at the Port of Bremerton.

Radian Aerospace is involved in research and development of “aerospace hardware to serve a variety of customers,” according to a company representative.
...
Incorporated in 2016, Radian Aerospace shares leadership with Holder Aerospace, a Renton company headed by former astronaut Livingston Holder and aerospace executive Curtis Gifford.

Anyone know what's up with this company?

The engine is RP1/LOX.

A 15,500-pound mount will brace engines during testing. Liquid oxygen and jet fuel will be stored in stainless steel tanks shielded by walls. 

[Kryten]

 That location is pretty close to the Blue HQ in Kent.

[TrevorMonty]

RelativitySpace to 3D print complete LV.

relativity-space-aims-to-3d-print-entire-launch-vehicles/?utm_content=buffer45f27&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer

This is not that new, their competitors (Vector, RL, Firefly, Launcheone) are also aiming to do this. Everybody 3D prints engines and composite fuel tanks are  additively/robotically manufactured. Human labour involved in final assembly may vary between companies but all will try to reduce manhours required over time.

Relativity LV is 9+1 x15klbs Methane engines. Not stated but probably pressure feed as cheaper to build. 1250kg to LEO?.

At this payload range they would also be up against Firefly and Boeing XS1. If all US cost same to build it will be interesting competition against reuseable XS1 and cheap expendable boosters.

By 2021 RL should also be flying Electron successor, whatever that will be.

[TrevorMonty]

Indian are looking at developing 500kg launcher.

newsclick.in/isro-develop-smaller-rockets-carry-satellites-weighing-500-kg

[john smith 19]

I also think so, yes. There's no market yet for so many launcher companies. At least experienced engineers could recycle and join the surviving companies.

Well the resurgence in actual paid-for jobs in rocket engineering is somewhat encouraging although you do wonder how much of it is is the result of VC investors with "spare" cash to invest and a need to invest it somewhere, hoping (somehow) they will be funding the next SpaceX. 

Obviously not all of them are going to make it, although of course if you knew exactly which ones would get to at least a first launch and which won't you'd probably be able to make quite a lot of money.

What we do know is it took SX about $200m to go from a flat lot to the first F9 launch, regardless of what industry cost models said it would cost.

So scale down the payload on your VTO ELV and you scale down the startup costs, because that's the lowest risk plan.

Because no one has ever thought of this plan before, right? 

Meanwhile, since most of the staff of most of the companies will have zero actual experience in rocket design and build they will be needing to make exactly the same class of mistakes that SX (and Masten and Armadillo, although XCOR did theirs at Rotary Rocket) had to make as they went up the learning curve.

So what's the win in this? Best case is they make an ELV that's as good as the Pegasus XL? The most expensive ELV (in terms of $/lb to orbit) on the planet?

Or maybe slightly better?

What I'm waiting to see is a startup that worked the problem backward from the cost of a launch and said "OK we want to get to this level where people are going to be queuing up to launch with us. What does that mean for our whole development budget and plan? How many flights a year do we have to hit?"

I'm not seeing that so far.

AFAIK all of these mfg will also operate the whole LV, so they will carry the whole development budget on every launch, with more or less numbers of payloads needing to be launched before they break even. If they break even.

The paradox of all VTO ELV businesses stems from the technologies historical basis in  ICBM work which begins "First, get someone else to fund the development project as a cost plus project, effectively writing off the development cost."

[TrevorMonty]

Developing a small LV that will fly is only part of costs for these companies, building launch facilities and gearing up for large scale low cost manufacturing are the lions share of costs. RL latest round of investment was for something like $50M and that was for large manufacturing plus some rainy day money.

Both LauncherOne and Vector are investing heavily in their production facilities, launch facilities are different again for these two companies, plane and mobile launcher.

[john smith 19]

Developing a small LV that will fly is only part of costs for these companies, building launch facilities and gearing up for large scale low cost manufacturing are the lions share of costs. RL latest round of investment was for something like $50M and that was for large manufacturing plus some rainy day money.

Both LauncherOne and Vector are investing heavily in their production facilities, launch facilities are different again for these two companies, plane and mobile launcher.

Indeed.

To really cut development costs you have to find a way to launch from an unmodified model, so you don't have to buy (and support and maintain) your own plane.

The best options for this seem to be planes that can either carry one of their own engines on a wing pylon, or that can open a cargo door in flight.

The key benefits are the flexibility in launch site (in principle right down to the Equator) and the lack of range costs, which (7 decades on from Sputnik) still don't seem to scale with the size of the vehicle and which disproportionately hit small sat launchers, unless  you build your own range as well. 

[john smith 19]

RelativitySpace to 3D print complete LV.

relativity-space-aims-to-3d-print-entire-launch-vehicles/?utm_content=buffer45f27&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer

This is not that new, their competitors (Vector, RL, Firefly, Launcheone) are also aiming to do this. Everybody 3D prints engines and composite fuel tanks are  additively/robotically manufactured. Human labour involved in final assembly may vary between companies but all will try to reduce manhours required over time.

Relativity LV is 9+1 x15klbs Methane engines. Not stated but probably pressure feed as cheaper to build. 1250kg to LEO?.

At this payload range they would also be up against Firefly and Boeing XS1. If all US cost same to build it will be interesting competition against reuseable XS1 and cheap expendable boosters.

By 2021 RL should also be flying Electron successor, whatever that will be.

The problem all of these vehicles will have is that that every dollar spent on them will have to be directly paid off by the launches they provide.

The only case where this did not apply (Sea Launch and Arianespace Soyuz) provide mixed evidence of cost spreading for various reasons, starting with how much the version supplied by Sea Launch and Arianspace diverged from the "stock" version launched normally.

People say launch costs are high because it's like buying a jumbo jet and using it once (so does that mean SX is like getting the wings and engines back?) but actually it's worse than that.

Imagine if different aircraft types needed separate airports  as well from which to operate from, with their own runways, taxiways, fuel handling etc.  Yet with the exception of the EELV designs (AFAIK Atlas V and Delta IV can use the same pads now) all infrastructure is also dedicated to specific LV's, just as the purchase costs for the recovery ships have to be carried by SX launches.


And then of course there is the process of building up a track record of successful launches.

I'm simply not hearing anything radical enough to change the market, when you factor in the development costs of your new hardware

If you've built an ELV that can launch someone else's 1U cubesat for $100k. That's about $75K/Kg.  Whoopee.

A secondary payload on an Ariane 5 could get you about 100Kg of payload for the same money. Other launch providers can supply similar deals and NASA built a whole Moon probe out of (basically) stringing the secondary payloads together.

On a personal level the ability to design, develop, test and deploy a new ELV and its engines even once is an enormous achievement and teams (and make no mistake team management is just as important a skill here as being able to run a lathe or program a flight computer in this) who does so can take considerable pride in having done so.

But from a business perspective....

Can you launch within 1 week of my supplying you with my payload?
How many coupled loads analysis (and associated redesigns) will I have to do before your LV does not shake my payload to bits?
Is your LV characterized well enough that you can answer that question?
As maiden launch customer can you offer me odds of better than 50/50 it won't blow up?
What's your orbital range in altitude and inclination? DoD EELV reference range (9 of them) or "Anything as long as it's the launch site inclination and about 200Km"?
Can you put your LV on a "something" (from a 40 foot shipping container to a C17) that allows you to launch my payload from the Equator,or someplace else that's at a different inclination (possibly not the US)?
Can you recover any part of your LV?
If you can will it lower your prices?
Do you know how many launches you will have to make to cover your development costs?
Do you know what portion of your market that is?
Do you what your marginal costs will be then and hence your gross profit margin?

I don't expect the answer will be "yes" to all of those questions but AFAIK the answer will be yes to hardly  any of them.

Let me suggest a few things for anyone else who wants to play the LV game in 2018.
1) Don't try to reduce a cost, try to eliminate it. OTRAG realized that a "tank" is basically a pipe with 2 end caps 4 decades ago. Despite this it seems everyone still wants to build their own "tanks"  .
2) Partial recovery and reuse is a fact, not a theory anymore, and it does deliver benefits. What's your excuse for not doing so?
3) Solids are for weapons systems unless you plan to be a solids mfg. Do you plan to be an SRB mfg?
4) The SRB/RP1/LOX/LH2/LOX Atlas has theoretically excellent performance, but 3 propellant systems is a massive PITA (whose costs for ULA are already sunk already). Pick a common propellant set for your stages.
5) Unless you're using solids or hypergols propellant (especially LO2) is cheap. Get the one that gives you best (affordable) Isp.
6) LN2 is even cheaper than LO2. Sub cooling anything (provided it's target temperature is at least 2-3K higher than the NBP of LN2) is fairly easy, and worthwhile if planned in from the start (or the engines can be upgraded to use the additional propellant mass). It means you don't have to build as much to carry the same propellant. It's not the propellant, it's the tankage you didn't build.
7) Up to the limit of your (or your sub contractors) mfg machinery LV production costs scale sub linearly.
That's why there is no "Falcon 5" or "Falcon 7." Small is good for development. Big is good for cost effectiveness.
8 ) 2 level mixture ratio change on the 2nd and 3rd stages of Saturn V increased payload to Lunar orbit 2.5% (would have been more if the 1st stage had supported it). The paper by Longsdon & Africano is essential reading for this and other tricks to use before you start cutting metal.
9) An "austere" launch pad is literally a flat pad of concrete.  The V2 and the Apollo 1 managed with a metal plate with 4 concave curved sides as a flame deflector, the former launched 9 rockets in 1 week while under Allied air attack with no warning. 
10) It is very much easier to precisely calculate the liftoff point than the landing point, making the landing for any reusable stage much bigger than the takeoff area, unless the takeoff area is a runway.
11) Recovered stages are 1/10 or less the T/O mass, so the recovery area concrete can be much thinner.
12)Clustering is a really good idea.
13) Transport and logistics costs rise a lot once you can't be carried by a regular truck. Anything up to what fits in a 40 foot shipping container is good. Anything below the point where the load is "wide" or "long" (or both) is still OK.
14) The studies by Whitehead more than 2 decades ago showed reciprocating pumps were simpler and much cheaper and just as efficient below about 5000lbs of thrust. Most of the patents have run out.
15) The expander or dual expander cycle is attractive because it a)Allows  you to use a clean, pure drive gas rather than the complex mixture of part burned hydrocarbons from a GG cycle and b) operates at a much lower temperature, allowing the use of much lower temperature (and possibly lighter) materials, like Aluminum. However it retains the potential issue of hot Hydrogen getting through seals to the LO2 flow.
16) Keep the expendable parts as simple and cheap as possible.
A modern cell phone (even a "dumb" one) has a processor, clock, camera, USB interface, RF interface, GPS (not LV grade of course) and an SDK to access them. Cell phone batteries have enough life to fly a mission. So do star or (even Earth limb or Sun) tracking on the camera and drive the rest of the system through the USB? How complex is the attitude control? How about a set of Sodastream cartridges and some vac formed nozzles, rather than the cast body, with spun metal nozzles thruster set designed to use MMH? 
17) Design the support, don't support the design IOW make sure you've got all the unique bits of the system inside the system boundary. That includes the launch pad, any special GSE etc.
18) vibration isolation systems (dampers) have been available for decades, but no one seems to use them. They'd rather hard bolt their payloads and have them be shaken to bits (or trigger resonance in the LV and have it shake itself to bits) than risk the payload hitting the fairing, despite CLA's sinking a lot of time and cash.
19) People use what they know but an electronic answer may not be the best option. Resonance ignition does not need a spark plug, just a cryogenic propellant with a tank pressure of about 2 bar gauge.  "Pressure ladder" systems controlled GG cycle engines without a computer for a decade. Modern CNC can deliver very high precision mechanical parts much faster (and cheaper) than historical manual machining once set up, with much tighter production tolerances. 
20) The annual NASA "Space Mechanisms" conference proceedings is a massive back catalog of ways to do tricky tasks for LV's and payloads, such as deploying the Trident missiles drag reducing "Aerodisk" without a computer signal or electrical power.  It's also full of the problems such development programmes have and the ways they can bite you when you "fail to plan." 

Just a few points to consider.

What I haven't heard anyone is looking at.

1) Airbreathing launch vehicle (no I don't mean a carrier plane).
2) Selling the whole package. The customer does their own launches, but that's probably because
3) No one seems to be talking about reuse (except in the sense of "Big specially built plane") all of them are expendable.
4) Hybrids with the same Isp as solids. Still have the same stacking issues IE Need a really big crane, but much safer to move and mfg. 

[Katana]

Airbreathing sucks.
My turbopump project turns out to be ironically easier than my duct fan / ramjet projet up to now.

Hybrids already have more isp than solids for decades, real problems are mass ratio and endless of bugs.

[gongora]

What we do know is it took SX about $200m to go from a flat lot to the first F9 launch, regardless of what industry cost models said it would cost.

More like $400M, and that's not for the current F9.

[john smith 19]

What we do know is it took SX about $200m to go from a flat lot to the first F9 launch, regardless of what industry cost models said it would cost.

More like $400M, and that's not for the current F9.

Which is why I specified the first F9 launch in 2010. Obviously F9 has had a lot of mods since them but up to then SX had launche 5 F1's, 3 went bang, the 4th got to orbit with a mass simulator and only the 5th had an actual customers payload who got a working satellite on orbit.

It was once F9 started launching successfully they started getting substantial revenue, but to that point most of the cash was from the "Bank of Elon."

That's based on NASA having a deep look at SX's spending.  Using standard Aerospace industry cost models they expected F9 would have cost about $2Bn, IE 10x higher as a standard govt project.  The numbers stuck in my head because the difference was so damm big.

IIRC Rand Simberg and Jon Goff commented on the massive discrepancy at the time.

[TrevorMonty]

Not another SpaceX please, we have enough of them. Small LV only, F1 is no more so SpaceX is totally off topic.

[john smith 19]

Airbreathing sucks.
My turbopump project turns out to be ironically easier than my duct fan / ramjet projet up to now.

Not really a big surprise to anyone who knows anything about ramjets or pump design.

Hybrids already have more isp than solids for decades, real problems are mass ratio and endless of bugs.

That's good to know. Perhaps you supply a reference where I can read more on them? What propellant combination did you have in mind?

[Katana]

Airbreathing sucks.
My turbopump project turns out to be ironically easier than my duct fan / ramjet projet up to now.

Not really a big surprise to anyone who knows anything about ramjets or pump design.

Hybrids already have more isp than solids for decades, real problems are mass ratio and endless of bugs.

That's good to know. Perhaps you supply a reference where I can read more on them? What propellant combination did you have in mind?

When turbomachinery is not mature (corresponding to 1940s technology), ramjets looks promising. That's why Navaho and Burya are developed.

But when you have some decent turbomachinery, they have MUCH more bang per development effort than sluggish ramjets, though starting at a bit higher price.

LOX hydrocarbon (paraffin or PE) hybrids have theoretical isp same to kerosene, some test engines could even reach theoretical performance. However  hybrids can't scale. This caused enormous trouble to SS2.

https://web.stanford.edu/~cantwell/Recent_publications/Cantwell_IJEMCP_9_(4)_305-326_2010.pdf

[Humuku]

Just a few points to consider.

Great post, Thank you!

And now you have to find people who really think like that to get a team which gets things done. From time to time such a thing happens and the world is changed.

[john smith 19]

Just a few points to consider.

Great post, Thank you!

And now you have to find people who really think like that to get a team which gets things done. From time to time such a thing happens and the world is changed.

That's why team formation and management is a serious challenge, up there with managing or coaching any top level team in any major sport. 

Getting a LV through design to first launch to orbit is a very serious challenge and anyone who does so should be proud of the effort but it has changed.
In 1940 this was SF. By 1950 (in the US) it was (sort of) on the radar. By 1960 (IE post Sputnik) there were multiple programmes and it was known to be possible. By 1970 building a rocket to put people on the Moon was a solved problem, albeit at very great expense.

So building a rocket is no longer a blind jump into the unknown where most of your peers will scoff at the absurdity of the notion to begin with.

But that very "normalization" of the process means that it no longer "adds value" in the same way.  The first overnight courier service had  the market to itself. Today....

So anyone looking to get into this needs to ask themselves (apart from the sense of satisfaction) "What does my system deliver that any other can't?" My instinct is this is not  just a technology issue, but a process and mfg issue. Unfortunately the quality of those (both yours and your competitors) is quite difficult to judge inside a startup.

[john smith 19]

But when you have some decent turbomachinery, they have MUCH more bang per development effort than sluggish ramjets, though starting at a bit higher price.

You seem to think "air breathing" begins and ends with ramjets.

LOX hydrocarbon (paraffin or PE) hybrids have theoretical isp same to kerosene, some test engines could even reach theoretical performance. However  hybrids can't scale. This caused enormous trouble to SS2.

https://web.stanford.edu/~cantwell/Recent_publications/Cantwell_IJEMCP_9_(4)_305-326_2010.pdf

Interesting report. The contractor on SS2 inherited their tech from Amroc, using HPTB and multiple ports.
The work at Stamford and SPG suggests a newer design would have the regression rate of a solid without needing multiple ports, hence having higher fuel loading and a stronger structure, so they would scale up.

[Katana]

But when you have some decent turbomachinery, they have MUCH more bang per development effort than sluggish ramjets, though starting at a bit higher price.

You seem to think "air breathing" begins and ends with ramjets.

LOX hydrocarbon (paraffin or PE) hybrids have theoretical isp same to kerosene, some test engines could even reach theoretical performance. However  hybrids can't scale. This caused enormous trouble to SS2.

https://web.stanford.edu/~cantwell/Recent_publications/Cantwell_IJEMCP_9_(4)_305-326_2010.pdf

Interesting report. The contractor on SS2 inherited their tech from Amroc, using HPTB and multiple ports.
The work at Stamford and SPG suggests a newer design would have the regression rate of a solid without needing multiple ports, hence having higher fuel loading and a stronger structure, so they would scale up.

For high mach operation, all airbreathers are limited by fuel energy density v.s. incoming air energy density. Even scramjets can't be much better than ramjets (gaining better mach range but loose lots of T/W ratio). Skylon is also limited to mach 5.

For low mach operation,  RBCC/TBCC/skylon are much more complex than either ramjets or pure rockets.

If ramjets can't compete on R&D COST v.s. pure rockets (F9R style vtvl), nothing else can.