The Reaction Engines Skylon/SABRE Master Thread (6)

[Lar]

How much could the timeline be accelerated if they had Bezos class funding the way Blue does? Just interested in informed speculation, not official answers.

[t43562]

How much could the timeline be accelerated if they had Bezos class funding the way Blue does? Just interested in informed speculation, not official answers.

I'm not hugely informed but their current approach was formed after the money they had on offer to them became drastically less for some reason. I remember the talks of spending hundreds of millions on the engine development.  So they were planning to test bigger engines AFAICT that would have suited Skylon rather than a TSTO concept.  So I think that's the route they would have wanted to continue down perhaps. 

Nowadays with the scaled-back design money would probably have a big effect.  I'm sure money would accelerate the build of their two test facilities and allow them to attract experts in critical disciplines with high salaries. It would also encourage airframers to invest their own money, knowing that the engine was very well funded.   

[john smith 19]

Is that they consider TSTO being more viable, or obtaining development funding for a TSTO is more viable?

Good question. Probably a bit of both.

The trouble is, unless you have an existing fully ready to go upper stage to drop in your budget more than doubles. You're not just designing and testing 2 vehicles, you're testing their interactions with each other, which are likely to be complex.

Going Methane also (probably) means the US falling to higher density in the atmosphere before it generates enough lift to start decelerating. That suggests the skin temp will be higher, so they may have to abandon their preferred TPS options. Outside RCC European experience of TPS is limited, although the ARD capsule did use all European heatshield ceramics IIRC>

The upside is pump design should be simpler. LH2 is (AFAIK) the only compressible (cryogenic or not) at these pressures. One of the things (apparently the viscosity of LH2 is about that of air ) that make LH2 pump design such a massive PITA.

Application profiles with a cruise stage before higer speed air breathing and rocket transition would likely favour methane.

Having a customer and revenue stream would allow longer term development

True in both cases.

The LAPCAT programme goal was M8 with essentially Jet A fuel.  REL came closest with M5 and a very long range, but being over sea allowed them to go mostly full M5. The German M8 Kerosene powered SCramjet got nowhere close in range.

The passenger jet makers would still like liquid fuels but if it's an unattainable M8 Kerosene versus a viable M5 Methane powered vehicle they'll take what they can get. London to Sydney non stop goes from 17 hours to   about 3 1/4. Journeys of similar length likewise, but they will probably need to fly higher to ease the heating issues, the bit #2 problem once you've got a viable air breathing engine.

[edzieba]

How much could the timeline be accelerated if they had Bezos class funding the way Blue does? Just interested in informed speculation, not official answers.

The big benefits would be:
- Development in parallel. With low funding available, RE have been developing the HEX, and are now proving the HEX. Next, they will be developing and then proving turbomachinery. Then chambers and burners, then integration, then structure for the craft, etc. With enough funding to hire more staff and acquire more facilities, they could develop everything in parallel which should also ease integration (no need to accommodate parts designed with the assumptions of several years ago)
- Single-purpose development. No need to worry about the commercial viability of subsystems due to the need to generate revenue from them in order to fund further development. Focus can remain on the final integrated engine design.
- Less worry about ITAR. If you are freed from the concern over IP control and the need to sell commercially outside the US, then you now have a much larger selection of engineers to work with, and a larger selection of (somewhat) CoTS parts to use rather than develop in-house. Need a very mass-efficient Hydrolox upper stage? Buy a Centaur!

[Archibald]

How much could the timeline be accelerated if they had Bezos class funding the way Blue does? Just interested in informed speculation, not official answers.

Where is Bill Gates when you need him ? Surely, he has been badly burned by his Teledesic investment two decades ago, maybe that's the reason why he don't want to be a "space bilionaire" like the other four (Bezos Musk Allen Branson)

[ChrisWilson68]

Is that they consider TSTO being more viable, or obtaining development funding for a TSTO is more viable?

Good question. Probably a bit of both.

The trouble is, unless you have an existing fully ready to go upper stage to drop in your budget more than doubles.

No, it doesn't.

The real effect on budget depends on many factors.  You're just taking one factor and ignoring the rest.

If each vehicle has more margin, each can be far cheaper to develop than a single vehicle.  And the two can potentially have commonality.  And one or the other can re-use some existing technology.  It's not just a "fully ready to go upper stage to drop in" or nothing -- there's a whole range of possibilities.  And the combined size of the two vehicles can potentially be smaller for the same payload to orbit when staging is used.

[john smith 19]

No, it doesn't.

The real effect on budget depends on many factors.  You're just taking one factor and ignoring the rest.

If each vehicle has more margin, each can be far cheaper to develop than a single vehicle.  And the two can potentially have commonality.  And one or the other can re-use some existing technology.  It's not just a "fully ready to go upper stage to drop in" or nothing -- there's a whole range of possibilities.  And the combined size of the two vehicles can potentially be smaller for the same payload to orbit when staging is used.

I fear I have not been explicit enough. I was talking about the development budget. You seem to be talking about the manufacturing budget.

Sure the mfg of a common diameter, common propellant TSTO may have some economies of scale, like Titan II for example. The two together might (but that's not a guaranteed outcome) be a bit smaller

But the fact remains the DDT&E will be different as they undergo different trajectories. They don't fly over the same range of speeds, air pressures and altitudes. they experience different loads, if only because one is being carried by the other (VTO, HTO or VTHL). And DDT&E (or Non Recurring Engineering) costs for big LV's are  huge.

You're talking about the potential manufacturing cost of say SLS at $500k-1Bn while ignoring the $14Bn development budget (so far) , where the payoff is meant to be the lower $/lb to orbit (Earth or ultimately Mars)
Which incidentally is exactly the argument behind the BFR.

[Jim Davis]

I was talking about the development budget.

You remind me of something Zubrin once wrote in a completely different context:

"...I would argue that it would be much easier to develop both a 900s nuclear thermal rocket AND a 15% aerobrake AND a 5 MW nuclear electric propulsion system at 15 kg/kWjet, than EITHER a 1100s nuclear thermal rocket OR a 5% aerobrake OR a 100 MW nuclear electric propulsion system at 1 kg/kWjet."

I'm far from Zubrin's biggest fan but I thought he nailed it here.

[john smith 19]

You remind me of something Zubrin once wrote in a completely different context:

"...I would argue that it would be much easier to develop both a 900s nuclear thermal rocket AND a 15% aerobrake AND a 5 MW nuclear electric propulsion system at 15 kg/kWjet, than EITHER a 1100s nuclear thermal rocket OR a 5% aerobrake OR a 100 MW nuclear electric propulsion system at 1 kg/kWjet."

I'm far from Zubrin's biggest fan but I thought he nailed it here.

And he'd be correct provided you had the budget for all 3 items.
The issues are
Budget
Technical risk
Schedule risk.

But with reference to SkylonSABRE if you already have an engine that can give you an average 3000secs (3 1/3 bigger than the NTR and about 6 2/3 bigger than the best LO2/LH2 rocket) that means have a lot of payload you can put the second stage into.

But as an SSTO the first stage structure is also the second stage structure.

It's a cliche that "The rocket equation is very steep" but that turns the structural problem from virtually impossible (the bogeyman of all VTO SSTO) to very hard but possible.

The "compelling" reason for doing a SABRE based TSTO is you can avoid facing the re-entry TPS issue with an expendable US. However, unless the booster is (very carefully) designed to accept a reusable US at a later date you've essentially spent the budget and designed an evolutionary dead end.

Which is fine if you want to be yet another mfg/launch services operation like every other ELV/SRLV services supplier. 

[t43562]

The "compelling" reason for doing a SABRE based TSTO is you can avoid facing the re-entry TPS issue with an expendable US. However, unless the booster is (very carefully) designed to accept a reusable US at a later date you've essentially spent the budget and designed an evolutionary dead end.

Which is fine if you want to be yet another mfg/launch services operation like every other ELV/SRLV services supplier. 

I think the problem is that decades and (hundreds of?) billions have been spent on rockets but air-breathing hasn't had anything like that. Even a dead end would change the world in the sense that the next attempt would have far less to prove.   I think in a way that the strategy shouldn't  be to get to SSTO by the cheapest way but to create enough interest to make people start spending money in this area and then the economy with which one reaches the goal is not so important.

[john smith 19]

I think the problem is that decades and (hundreds of?) billions have been spent on rockets but air-breathing hasn't had anything like that.

In current year $, probably 100s of $Bn. For example the Titan ICBM had about 60 test flights, and Titan II had about 40 more. A number simply inconceivable in modern test programes (but fortunately also not as necessary).
In air breathing I'd say the US has spent > $10Bn over the last 6 decades. Unfortunately 99% of it has gone to full LACE systems (that periodically resurface from  people like Andrews Aerospace) and the ongoing chasing of a viable SCramjet. NASP alone swallowed close to $6Bn in current year $ values.

Even a dead end would change the world in the sense that the next attempt would have far less to prove.   I think in a way that the strategy shouldn't  be to get to SSTO by the cheapest way but to create enough interest to make people start spending money in this area and then the economy with which one reaches the goal is not so important.

Depends what your end game is.

If you want to lower the cost of space access radically IE by a factor of 10, you've still got the same problems. LH2 US gives you best Isp and hence largest largest available mass fraction but worst density.

At this stage I'd settle for REL getting to an actual flight vehicle.  I'd like it to be able to generate revenue and go to M6 but IRL if it can just get past the transition to full rocket mode and run the engine (in rocket mode) to full steady state conditions (I'm guessing 5-10secs) before it runs out of propellant and goes to a glide landing I'd be happy.

[Robotbeat]

Let's start with Methane: Yes, RE are actively looking at how performance would be changed by a switch to Methane, as well as confirming that it would simplify structure (prop density) and greatly simplify ground handling and safety particularly for manned flight and spaceflight. This also came with confirmation that while SSTO remains the 'holy grail' goal, RE are taking a much more pragmatic approach to development with TSTO being more viable.

I'm really glad to hear that they're being pragmatic about considering both methane and two-stage launch vehicles.  If the engine technology is good but hydrogen and single-stage-to-orbit would make the overall program impractical, it would be a shame that the core technology wouldn't get a chance to prove itself.

And, if they can make a go of it with a methane, two-stage vehicle, then they would be in a position to try hydrogen and single-stage as next steps in the future.

Methane makes the crazy structural mass assumptions go away and makes SSTO a lot easier (at least for pure rockets)

[john smith 19]

Methane makes the crazy structural mass assumptions go away and makes SSTO a lot easier (at least for pure rockets)

Which particular "crazy structural mass assumptions" did you have in mind?

[JCRM]

This month's JBIS has a paper by members the UK's National Space Technology Steering Group's (NSTSG) Access to Space Working Group - "The UK's Path to Launch: The National Space Technology Strategy for Launch Technologies, Vehicles and Services" (A Bradford et al) discussing an update of the 2014 Access to Space Roadmap undertaken in 2016. These roadmaps are, amongst other things, used when judging the NSTP funding. I assume the actual roadmap is available -- in order for organisations to prepare bids -- but haven't located it yet. I've included the parts that seem most relevant to REL (I will try to remember to link when an online version is available)

Disclaimer:

It should be noted that the NSTSG and it's [sic] working groups are not policy makers

According to this plan, after setting up launch services in the short term, and developing indigenous launch systems in the medium term, long term plans include

The development and introduction of game changing launch systems and technologies

In its discussion of archictectures and systems (which is a list "intended to capture any and all [ways of] getting into space" (i.e. not advocating them) it lists separately SSTO (both H and V TOL) and "Airline Service to Orbit: A space plane [sic] type approach, likely to take the form of a single stage to orbit (SSTO) orbital spaceplane"

In the Development Priorities, it includes the following

The roamap is not intended to set priorities itself per se; what it can do [... is indicate] which develoment activities should be prioritised and started/funded first in order to allow [things] to be reached in a certain timescale. This is especially relevant and important for the longer term, game changing launch systems that are in scope of the roadmap; even though readiness of those systems is not forecasted[sic]/anticipated before ~2025, development of some of the core, revolutionary, enabling technologies needs to be started now if the overall systems have a chance of bening ready in the desired timeframe

How many game changing concepts were there in 2016 with a projected fully funded development time of 8-9 years?

[john smith 19]

In its discussion of archictectures and systems (which is a list "intended to capture any and all [ways of] getting into space" (i.e. not advocating them) it lists separately SSTO (both H and V TOL) and "Airline Service to Orbit: A space plane [sic] type approach, likely to take the form of a single stage to orbit (SSTO) orbital spaceplane"

In the Development Priorities, it includes the following

The roamap is not intended to set priorities itself per se; what it can do [... is indicate] which develoment activities should be prioritised and started/funded first in order to allow [things] to be reached in a certain timescale. This is especially relevant and important for the longer term, game changing launch systems that are in scope of the roadmap; even though readiness of those systems is not forecasted[sic]/anticipated before ~2025, development of some of the core, revolutionary, enabling technologies needs to be started now if the overall systems have a chance of bening ready in the desired timeframe

How many game changing concepts were there in 2016 with a projected fully funded development time of 8-9 years?

"likely to take the form of a single stage to orbit (SSTO) orbital spaceplane"" is pretty much Skylon.

This suggests they are looking at ways identify long lead time technologies which would a)Be useful enablers of the identified future launch systems and b) Have a possible market outside that.

For some hypothetical examples of this with relevance to this thread title that could include.
1) Skylons strut technology is now baselined at SiC reinforced Titanium.  They will be needed in massive numbers (10s of 1000s) so price is an issue. Working out how to get production volume up and unit price down (ideally in a way that is not wholly dedicated to just making this part) would be a very useful development for the supplier.

2) Likewise developing a joint and joining technology for the nodes in the truss that can be machine assembled would also be a good investment. NASA has some very interesting papers for on-orbit welded truss structures (for making very large telescope reflectors) from roughly the mid 70's to the late 90's.  Turns out with large structures (and Skylon is definitely one of those) very slight deviations in nominal strut length can lock in a lot  of stress. They came up with a 2 piece telescoping strut using 3 welds, 2 at the ends, one to fix its installed length.  There are other options however. With the massive fall in computer costs a modern PC can drive a pretty complex assembly rig with  multiple degrees of freedom to high precision at reasonable speed. I think 1 strut/minute 24/7/365 is a conservative operating rate provided you can keep the machine supplied with struts.   I think 1 every 10 seconds could be possible, cutting airfram mfg time from months to days.

3)RCC is a small but significant part of Skylon. RCC is made in the UK for aircraft brakes but the direction of thermal conductivity is wrong.  Helping the relevant brake mfgs set up an "aerospace components" division would seem like a good idea. The Germans are very interested in this technology for their SHEFEX TPS concept of simple flat plates tessellated over the body, rather than each plate being individually curved.

4) Water cooled brakes were found to be critical to meeting the mass targets for Skylon. Apparently it's quite well known in heavy truck racing, where people race 18 wheeler cabs . Obviously Skylon is bigger so looking at ways to assure that technology in this size is available would be useful.
While passenger jet designers are very conservative about safety, landing gear is heavy. It's used about 5 minutes at each end of a flight but it's got to be carried everywhere. A brake system that lost 1/2 tonne of mass at the start of a flight could be very attractive to airline operators if they can be convinced it's reliable enough. With fuel prices only going one way not hauling 500Kg of structure per flight over a 30 year operating life of an A380 (which is about what Skylon's GTOW is) is not to be sniffed at.

5) UK Universities have done a lot to restore the ability to mfg something like the AEA Technology "System 2" SiC reinforced ceramic spec'd for the HOTOL and Skylon. REL have also been working with a French company in this respect. AFAIK small parts have been made but it would be good to either broaden the programme into the joining and mounting technology, or to scale up the size of panels, or at least the plant to make them.
With automated assembly I'm not sure small panel size  is an issue (as long as they are all pretty standardized), but the mounting tech has seen limited use. Likewise although the material seems to be quite impact tolerant inspection and repair techniques need to be worked out to avoid wholesale replacements, especially if 2 day turnaround at the IOC date is a goal. 

[EDIT late addition
6) Develop a premixed catalytic ignitor to allow the RCS thrusters to fire without need an HV system. Aerojet solved this in the mid 70's and since the rcs system is (IIRC) gaseous rather than liquid the reactants should be a bit warmer. Ideally a common basic size used on all the SABRESkylon combustion devices, with multiple copies on the bigger devices, like SABRE itself
]

Again a lot of this stuff is planning and scoping, or can be done by relatively small hardware testing. Small(ish) investments can make a big difference in viability further down the line.

Hypothetically if 1,2 and 5 were available by say the end of 2024 it might be possible to incorporate them in a SABRE test vehicle, which would significantly lower the technological risk of using these design approaches for larger vehicles, whatever their mission was.

OTOH I don't think they are absolutely necessary for such a vehicle, so a FTV could be built with more conventional materials to fly the basic zero-to-rocket transition-plus-a-bit-more for long enough for people to be convinced its stable in rocket combustion. Either could be designed with enough "stretch" to support other possible research programmes as well.

These are not predictions, merely possible areas the roadmap could point the UK government into investing in.
Wheather or not the will invest in any of them is frankly a complete crap shoot. 
They could just as easily decide to fund composite  LH2 tanks for fuel cell cars (because Skylon needs lightweight LH2 tanks. Some fuel cell cars need lightweight LH2 tanks. QED fund composite LH2 tank technology. Hmmm) .

[ChrisWilson68]

No, it doesn't.

The real effect on budget depends on many factors.  You're just taking one factor and ignoring the rest.

If each vehicle has more margin, each can be far cheaper to develop than a single vehicle.  And the two can potentially have commonality.  And one or the other can re-use some existing technology.  It's not just a "fully ready to go upper stage to drop in" or nothing -- there's a whole range of possibilities.  And the combined size of the two vehicles can potentially be smaller for the same payload to orbit when staging is used.

I fear I have not been explicit enough. I was talking about the development budget. You seem to be talking about the manufacturing budget.

No, I was talking about the development budget.

[john smith 19]

Methane makes the crazy structural mass assumptions go away and makes SSTO a lot easier (at least for pure rockets)

Except SABRE is not a pure rocket, is it?

And Skylon is not a VTO vehicle, which most (with the exceptions of the air launched ones) are.

So if REL designed a pure VTO rocket you'd probably be right.

[john smith 19]

No, it doesn't.

The real effect on budget depends on many factors.  You're just taking one factor and ignoring the rest.

If each vehicle has more margin, each can be far cheaper to develop than a single vehicle.  And the two can potentially have commonality.  And one or the other can re-use some existing technology.  It's not just a "fully ready to go upper stage to drop in" or nothing -- there's a whole range of possibilities.  And the combined size of the two vehicles can potentially be smaller for the same payload to orbit when staging is used.

I fear I have not been explicit enough. I was talking about the development budget. You seem to be talking about the manufacturing budget.

No, I was talking about the development budget.

So by your thinking the more alike the two (or three) stages are the cheaper they will be?
Because that's exactly the thinking behind the Bi or Triamese, as GD/Convair proposed for STS and BAC with MUSTARD.  Identical stages with identical engines.

Except no one has ever gone that way.  The upper stage really does not need to be that big, the lower stage can be easier to design with a denser propellant etc

So, little by little the designs diverge.
Then the conditions they are exposed to are different. But maybe you scale the US and just trust it's strong enough for all the forces?
That's not a good idea either. Now you're hitting the 1:1 structure/payload exchange rate of a US. If you want a decent payload you're going to want to lighten that US, so complete load analysis again.

Your problems get worse if you want US reusability as well. As SX have discovered the heat load from orbital velocity is a long way from what it was at stage separation.

And that's before we get into Coupled Loads Analysis. Where a 2 problem (SSTO and payload) has become a 3 body problem and all the interactons as well.

And let's not neglect engine divergence either. SX really like common parts designs yet the Merlin Vac is said to be 90% different to its Sea Level version. No one would make 2 engines that much different without expecting major benefits (I've no really clear idea why they are so mechanically different) and therefor very different operating characteristics. So that multiplies the number of different test variables you have to considers

IMHO the only way you get a TSTO with a lower development cost than an SSTO is you need a very well characterized booster or US and you build the other stage around those characteristics, accepting all their limitations. The only thing I can think of that came close to this is the shift in propellants from LOX/RP1 to hypergols in the TITAN II ICBM. Which still had about 40 development flights.

"It's cheaper to design a TSTO than an SSTO" [EDIT Note that word "cheaper," not "easier." It's about the budget, not the difficulty. IOW "Easier" is not a synonym for "cheaper ]is one of those pieces of industry "folklore," many dating from the late 50's and the US need to get something (anything) into orbit (or even a viable ICBM) ASAP.

It's like "Building a bigger rocket will cost more,"  or "Solids are the simplest rocket engines"* .
There are so many  qualifications and assumptions in such statements they should be treated with extreme caution.
So no I don't think designing a TSTO will be cheaper than an SSTO (again depending on what assumptions you make about the SSTO design). Design freedom multiples the analysis cases you need to test (both in the wind tunnel and computer simulation and in flight).

*until they undergo deflagration-to-detonation transition that is. Then you have several million lbs of high explosive going bang

[Jim Davis]

Note that word "cheaper," not "easier." It's about the budget, not the difficulty. IOW "Easier" is not a synonym for "cheaper...

But that is the normal usage in the context of large engineering projects. A problem that is "easier" takes fewer resources to solve and hence is also "cheaper". Time, labor, capital, etc can all be reduced to a monetary value at this scale.

[edzieba]

Due to the lack of existing SSTO launchers, designing a TSTO launcher is cheaper and easier. You're implementing solved problems rather than solving some new problems. Worse, you're trying to solve new problems that many people have already tried and failed to solve; the easy and cheap "just do X", "try Y", and "maybe Z will work out" solutions have all been found wanting, which is why we're even considering the brain-meltingly complicated regenerating-shifting-multi-cycle SABRE engine in the first place.
Designing an upper stage from scratch is expensive, but the point is you don't need to design one from scratch for a minimum viable launcher. Buy a STAR PAM, pay ULA to strengthen a Centaur for horizontal integration, etc. Once you have money coming in you can invest in an optimised upper stage if needed, but as we've seen with Falcon 9 an 'optimum' upper stage is not strictly necessary.