The Reaction Engines Skylon Master Thread (1)

[93143]

Your pessimistic assessment rests entirely on historical examples.  It is thus manifestly invalid when dealing with an approach that specifically claims to solve the problems encountered in those examples.

No, until it flies all they've shown is a potentially successful new approach. It might work, it might not. I didn't dispute the former, I merely pointed out the latter.

I think you'd better read that paragraph again.

I'm not accusing you of claiming it could never work, and I'm not claiming it definitely will.  I'm specifically accusing you of bad reasoning.  Your argument is invalid, and the paragraph you quoted explains why.

I'll try to rephrase:  You've criticized Skylon for reasons that don't apply to it, in an evident attempt to make general knowledge substitute for specific knowledge of the subject in question.

Earlier approaches too have all had their own secret sauce, their own reasons to believe they would make it.

None of them had the kind of margin Skylon has.  We're talking about a mass ratio of four and a half, with a payload fraction of 4.6% after structural and performance margins, on a vehicle in its fourth major design revision after two decades of engineering including lessons learned from other attempts.  And the only radical technological challenge is (was?) the precooler.

SABRE really is a game-changer if it works.  And right now it looks like it will probably work, at least technically.  The 'acid test' (full-depth cryo) remains, but the precooler is not the big unknown it once was, and everything else about the engine is fairly well understood.

It appears that you have no new information to add to the discussion, which means your pronouncements basically amount to a variant of "contempt prior to investigation".

I feel no contempt of Skylon at all, I said I found the concept very interesting.

I said "a variant" of a commonly-referenced intellectual blunder.  Basically the idea is that you seem to be more skeptical than is warranted by your level of knowledge, and your skepticism seems to be based not on the knowledge that you do have, but on general principle.

Most people here know it might not work.  Most people here are at least passingly familiar with the history of attempts at rSSTO, and as you don't seem to have any special command of the thread topic, I'm not sure your wet blanket approach (spiced up with the suggestion that you know how REL should be spending their money and effort better than they do) has added a whole lot to the discussion.

It will be expensive.  It's an airplane the size of an A-380, and it will probably cost a similar amount to develop, SSTO or not.

It will cost a lot more, because it will include development of a new engine, and not just any old engine, an engine of a type that has never been flown before.

Do you have a good reason to flatly contradict REL on this point, or were you simply unaware of their estimate?

That said, I'd love to see them succeed.

It's good to remember that we're all basically on the same side here.  I'm sure there are people in the world who wouldn't be ecstatic if Skylon were to be successful, but they probably don't post on this forum...  (Without prejudice, I exclude competitors such as SpaceX and Blue Origin from this statement on the technicality of conflict of interest.)

[zt]

Why don't they expect the first two prototypes to reach orbit? In what ways will those prototypes differ from the planned production variant?

[bolun]

Move to open sky for Skylon spaceplane

11 July 2012

http://www.bbc.co.uk/news/science-environment-18784866

[mmeijeri]

I'll try to rephrase:  You've criticized Skylon for reasons that don't apply to it,

Rephrasing is good, but I still don't think it's true my arguments don't apply to Skylon and I'd be happy to discuss it further.

in an evident attempt to make general knowledge substitute for specific knowledge of the subject in question.

I don't know why you keep suggesting that as I've said several times now that I'm quite familiar with the public documentation about SABRE and Skylon. I know what makes Skylon different from earlier approaches, why RE think they will be more successful and what they think they need for a breakthrough in specific launch costs.

None of them had the kind of margin Skylon has.  We're talking about a mass ratio of four and a half, with a payload fraction of 4.6% after structural and performance margins, on a vehicle in its fourth major design revision after two decades of engineering including lessons learned from other attempts.

And without any flying hardware.

  And the only radical technological challenge is (was?) the precooler.

Which is a very serious challenge. Are you confident they will achieve the effective Isp and T/W for SABRE? If so, what do you base that on given that nothing has ever flown?

SABRE really is a game-changer if it works.  And right now it looks like it will probably work, at least technically.

I don't think this confidence is justified by the known facts. Perhaps you have access to information I haven't seen?

The 'acid test' (full-depth cryo) remains, but the precooler is not the big unknown it once was, and everything else about the engine is fairly well understood.

Just to clarify, I'm not disputing that.

I said "a variant" of a commonly-referenced intellectual blunder.  Basically the idea is that you seem to be more skeptical than is warranted by your level of knowledge, and your skepticism seems to be based not on the knowledge that you do have, but on general principle.

I honestly don't understand why it seems that way to you. And as I said above I don't understand why you seem to be assuming my skepticism stems from ignorance.

Most people here know it might not work.  Most people here are at least passingly familiar with the history of attempts at rSSTO, and as you don't seem to have any special command of the thread topic

For the nth time I know damn well what I'm talking about, thank you very much. I find your repeated suggestion to the contrary presumptuous and offensive.

, I'm not sure your wet blanket approach (spiced up with the suggestion that you know how REL should be spending their money and effort better than they do) has added a whole lot to the discussion.

I said I wasn't confident they would achieve SSTO, and you seemed to be agreeing it wasn't certain. Given that, I don't see how you can see my comments as a wet blanket.

Do you have a good reason to flatly contradict REL on this point, or were you simply unaware of their estimate?

Both, and I gave you the reasons. I don't understand how the development cost of Skylon could be comparable to that of A-380, even if it excludes the cost of SABRE development, but I'd interested to hear why others find it plausible.

It's good to remember that we're all basically on the same side here.

Yes.

[simonbp]

Why don't they expect the first two prototypes to reach orbit? In what ways will those prototypes differ from the planned production variant?

They are really engine testbeds, rather than operational vehicles. The impression I got was that the prototypes don't need to go orbital to complete the test program, but REL might try it anyway once that program is done and enough risk is retired.

[QuantumG]

Are you confident they will achieve the effective Isp and T/W for SABRE? If so, what do you base that on given that nothing has ever flown?

That's my biggest concern too.. from a practical standpoint they've yet to demonstrate that they've got all the technology in-hand to make this vehicle. Which is fine, in time they will or they'll downgrade their plans.

There's other concerns, including purely theoretical ones, but there should be no debate that RE are currently in the technology development phase of their program and the outcome of this phase is not guaranteed.

[john smith 19]

Looks like a deeply precooled air turborocket. I'm not convinced it will give them SSTO though and if you go with TSTO you could go with a less ambitious "undeeply" cooled hydrocarbon air turborocket, perhaps using LOX/methanol or methanol/peroxide.

It is fair to describe SABRE as an air turbo rocket. However it may be the bits you recognize are causing you to overlook the bits you don't.

 ATR descriptions usually have the fan being driven either by a conventional shaft from a high temperature turbine or by a “gas generator.” The former makes for a pretty heavy installation while the latter is very far from the trivial engineering task its name implies. In all cases you need to allow for propellant pump drive as well as turbine drive.

SABRE's actual cycle has elements of both the RL10's expander cycle and the SSME's staged combustion. It could also be said that the use of an inert fluid power transfer loop echos the Silverbird power plant of Eugene Sanger. REL's senior design staff have long memories.

Like the RL10 the “high temperature” drive turbine can be made of Aluminum because it only operates at a few 100 degrees. By heating indirectly, rather than using the combustion products you avoid having to design turbines that can cope with 2x the chamber pressure while being bathed in superheated steam and hot GH2 moving at about Mach 0.5. A tough problem as SSME design engineers found.

This also means you can ramp up engine pressure and flows without forward motion by raising the temperature in the preburner/HX3 combo. As it picks up speed more heat will be extracted from the higher air flow.

If you have a model you'll need to factor those features in to see what difference that makes.

REL have stated that a successful test series will release c£200m in development funding. Judging by their press release this series has been successful. I'd strongly doubt that would happen if there were serious concerns over the tests validity.

If you have anything more than a hunch I'd be interested in hearing it, especially what assumptions you've made on the implementation. Where SSTO is concerned hidden assumptions can prove lethal. The background to the X30 project in "Surviving the Heat Barrier" should serve as a model of how *not* to do it.

As for their cost modelling they discussed this during the 90's. IIRC John Scott Scott said there's was the only flight vehicle cost estimate that Lockheed engineers had agreed with.

They also stated on one of the sci.space.* newsgroups the cost estimating equation they used to give the cost of SABRE development.

It's risky but if you build another one of what you've always built (TSTO) you'll get pretty much what you've always got.

Spacex have gone with this path of conservatism. The VTO TSTO transitioning (hopefully) to VTOL TSTO. They believe there is substantial cost to be squeezed out of the process. Time will tell how much they can squeeze out. But it's still a TSTO and that 1st stage has to come down somewhere in between the launch pad and where ever the 2nd stage makes orbit.

One of Skylons surprise (at least to me) selling has been it offers ELV payload fraction with SSTO re usability. All VTOL concepts seem to be *substantially* less than this (1% of GTOW seems to be the rule of thumb) which also seems to bother customers a lot.

[mmeijeri]

It is fair to describe SABRE as an air turbo rocket. However it may be the bits you recognize are causing you to overlook the bits you don't.

What's with the assumption I don't know what I'm talking about?

SABRE's actual cycle has elements of both the RL10's expander cycle and the SSME's staged combustion.

Yes. As well as an unusual counterrotating turbine and an unusual LOX cooled combustion chamber. I'm not aware of any production engine that uses LOX cooling, but there have been experimental ones and apparently it's not as risky as it might sound at first, even if there are leaks.

It could also be said that the use of an inert fluid power transfer loop echos the Silverbird power plant of Eugene Sanger. REL's senior design staff have long memories.

And they're also aware of patent issues.

Like the RL10 the “high temperature” drive turbine can be made of Aluminum because it only operates at a few 100 degrees.

That's true of RL-10, IIRC its turbine inlet temperature is ~0C or lower, but are you sure about SABRE? According to a document that has disappeared from the RE website since its recent redesign SABRE's turbine inlet temperature is 1320 K.

By heating indirectly, rather than using the combustion products you avoid having to design turbines that can cope with 2x the chamber pressure while being bathed in superheated steam and hot GH2 moving at about Mach 0.5. A tough problem as SSME design engineers found.

Heating indirectly (or at least in a separate combustor that sees no air) is of course standard for a turborocket and is what sets it apart from turbojets and turbofans.

If you have a model you'll need to factor those features in to see what difference that makes.

Well, I certainly don't have my own models, I wonder where people's confidence in models comes from and what models they've seen.

REL have stated that a successful test series will release c£200m in development funding. Judging by their press release this series has been successful. I'd strongly doubt that would happen if there were serious concerns over the tests validity.

Let's hope so. Any information on whether the recent tests included the HX4?

If you have anything more than a hunch I'd be interested in hearing it, especially what assumptions you've made on the implementation.

It seems to be the question is how much confidence we can have in the models presented by RE.

It's risky but if you build another one of what you've always built (TSTO) you'll get pretty much what you've always got.

There is a plausible case for why you might need SSTO (as with so many RLV design heuristics, there's an interesting paper listing a whole bunch of such mutually contradictory heuristics), but I don't think it's fair to say TSTO is the same thing you've always got. It is less ambitious, but could be expected to have higher operating costs than a SSTO. No one really knows for certain whether SSTO is needed for a breakthrough.

[MP99]

Spacex have gone with this path of conservatism. The VTO TSTO transitioning (hopefully) to VTOL TSTO. They believe there is substantial cost to be squeezed out of the process. Time will tell how much they can squeeze out. But it's still a TSTO and that 1st stage has to come down somewhere in between the launch pad and where ever the 2nd stage makes orbit.

One of Skylons surprise (at least to me) selling has been it offers ELV payload fraction with SSTO re usability. All VTOL concepts seem to be *substantially* less than this (1% of GTOW seems to be the rule of thumb) which also seems to bother customers a lot.

IIUC Skylon will require multiple $billions to bring to fruition. If you amortise those development costs in a commercial fashion, how many launches do you need before Skylon surpasses SOTA cheap ELV (or rTSTO) in price?

cheers, Martin

[john smith 19]

IIUC Skylon will require multiple $billions to bring to fruition. If you amortise those development costs in a commercial fashion, how many launches do you need before Skylon surpasses SOTA cheap ELV (or rTSTO) in price?

cheers, Martin

The flippant answer is REL don't care.

They are manufacturers *not* operators.

It's how many of them do they need to *sell* to make a profit.

A rough guide on launch *price* is that the total propellant bill amounts to $10/Kg to LEO. What the servicing costs would be is another matter but there's a lot of scope to set pricing. It depends how much you'd have to charge to meet the interest payments you need to make to buy one in the first place.

[zt]

They said in the interview linked here[1] that their investors will recoup the investment if they sell 30 Skylons. How will the operators make money is the operator's own problem. How much does Boeing and Airbus help airlines in planning how to make money off airliners they bought?

1 - http://forum.nasaspaceflight.com/index.php?topic=24621.msg925525#msg925525

[krytek]

REL's site go an overhaul and an update. Awesome.
http://www.reactionengines.co.uk/

[DLR]

I think it's good that RE focuses on developing and selling Skylon, rather than operating it. With RLVs, you can get away from the artillery paradigm and separate production and operation. You don't see Boeing or Airbus running their own airline, even though they could.

That's why I always found it odd that SpaceX is planning to operate its reusable Falcon by itself, rather than selling it to dedicated "spacelines" like Virgin Galactic.

[MP99]

IIUC Skylon will require multiple $billions to bring to fruition. If you amortise those development costs in a commercial fashion, how many launches do you need before Skylon surpasses SOTA cheap ELV (or rTSTO) in price?

cheers, Martin

The flippant answer is REL don't care.

They are manufacturers *not* operators.

It's how many of them do they need to *sell* to make a profit.

A rough guide on launch *price* is that the total propellant bill amounts to $10/Kg to LEO. What the servicing costs would be is another matter but there's a lot of scope to set pricing. It depends how much you'd have to charge to meet the interest payments you need to make to buy one in the first place.

They said in the interview linked here[1] that their investors will recoup the investment if they sell 30 Skylons. How will the operators make money is the operator's own problem.

ISTM that a single Skylon could take over all payloads currently launched on ELVs in it's payload range.

Both F9r and Skylon assume they can expand the market by making launch cheaper. However, until that larger market does appear, massive up-front investments just mean massive costs spread across a trickle of launches, making each launch *more expensive*.

The F9r approach seems to be to make a small up-front investment, which can pay back over time at current flight rates, then allowing them to offer bulk discounts which will grow the market if it is as elastic as they hope.

Assuming each Skylon needs to fly dozens of times per year to pay back, ISTM that needs launch rates in the thousands to support a fleet of thirty.

It's possible that F9r would start to expand the market, making the very demand that a fleet of Skylons could then come in and expand / make yet cheaper again.

cheers, Martin

[sfjcody_]

I'd like to know what the upgrade path for this vehicle is, assuming they can get the baseline version to operate as planned. I'd also like to know whether it can be theoretically scaled up to carry much larger payloads. What would a 100 tonne to LEO Skylon be like? What kind of runway would it need?

[zt]

In this lecture (), Alan Bond said their planned Mars architecture, which they did to make sure their vehicle design was good enough for any need, is to use an ungodly amount of launches using regular Skylons. I don't think they have a plan for a bigger vehicle.

[john smith 19]

What's with the assumption I don't know what I'm talking about?

Simply that you will not state exactly what your concerns are. I'm suggesting you may have focused on the bits of the design you are familiar with while under rating other aspects that give it its edge.

SABRE is *like* an ATR as it is *like* a LACE.
It is *neither* an ATR nor a LACE.

Yes. As well as an unusual counterrotating turbine and an unusual LOX cooled combustion chamber. I'm not aware of any production engine that uses LOX cooling, but there have been experimental ones and apparently it's not as risky as it might sound at first, even if there are leaks.

True. Rotary Rocket tested O2 cooling as has NASA in a 40 000lb test rocket and the air/O2 cooled main combustion for Skylon was tested at DLR in Germany.
HMX has discussed this in the context of Rotary Rocket and Doug Jones could no doubt add a few words as well. It was perceived as very scary. In reality it turned out to be completely controllable.

Contra rotating shafts were used in the Pegasus which powered the Harrier and AV8b. The world's only active service VTOL jet. I beleive REL staff include people with experience of Pegasus design.

So they are uncommon but well within the state of the art.
That kind of sounds like a straw man argument to increase Fear Uncertainty & Doubt so far.

And they're also aware of patent issues.

No doubt. They are not keen to discuss the frost control system for example.

Heating indirectly (or at least in a separate combustor that sees no air) is of course standard for a turborocket and is what sets it apart from turbojets and turbofans.

And what sets SABRE apart from an ATR (apart from the fact that what is being heated is *not* the inlet air) is where that air is going. Not to large combustion chamber which in a jet engine might be up to say 14atm above atmospheric, but to a rocket combustion chamber running 100-120Atm above atmospheric.

Well, I certainly don't have my own models, I wonder where people's confidence in models comes from and what models they've seen.

I was thinking more in terms of a set of configuration settings for one of the propulsion modelling codes professional engineers use in this field.

REL have stated that a successful test series will release c£200m in development funding. Judging by their press release this series has been successful. I'd strongly doubt that would happen if there were serious concerns over the tests validity.

Let's hope so. Any information on whether the recent tests included the HX4?

It seems to be the question is how much confidence we can have in the models presented by RE.

Well the UK Space Agency funded ESA to review the underlying technology and REL also hosted the 2 day "Requirements Review" on 20-21st of Sept 2010 where over a 100 guests had a chance to critique the design and the economics (perhaps you were there?) and both it and the ESA report were clear they saw no show stoppers, a process which AFAIK is also *unique* to REL, but I hope will be used by others in the future.

but I don't think it's fair to say TSTO is the same thing you've always got.

Every orbital launch vehicle I'm aware of has at least 2 stage to achieve orbit and is expendable. The Pegasus is the odd man out at being horizontally launched and needs 3 stages (4 if you include its carrier aircraft, which you probably should) to achieve orbit.
Most of the rest were and are 2 stage liquid fueled VTO ELV's, complete with similar failure rates over long enough launch histories.

[sfjcody_]

I like almost everything about the new website, but I'm not enjoying the music choice in their videos. They need something a little more grandiose. Ethereal sounding electronica or one or two carefully chosen classical pieces would do a better job of selling the technology.

[Carreidas 160]

I like almost everything about the new website, but I'm not enjoying the music choice in their videos. They need something a little more grandiose. Ethereal sounding electronica or one or two carefully chosen classical pieces would do a better job of selling the technology.

Hm, well at least it's not Muse (as used by SpaceX in their reusable F9 video) which is really out of place. I cringed a bit with that.

Then again, this is about showcasing space technology 10 years into the future, not selling the latest smartphone

[mmeijeri]

Simply that you will not state exactly what your concerns are.

I did state my concerns, but you don't find them convincing, which is fair enough.

But let me give some specific scenarios: what if rocket mode Isp turns out to be slightly disappointing (as it was with RS-68) and if that has structural implications (as it did with Delta)? What if the precooler only works efficiently up to Mach 4.5 instead of 5.5? What if it uses more LH2 than expected? What if lower than expected T/W leads to a slower ascent, more drag losses and a need for beefed up TPS? In my opinion all of these could reasonably happen. They could also reasonably not happen. Enough reason for me not to be confident Skylon will work as an SSTO. It could, but I'm not confident.

I am however confident that it could work well as a TSTO, and not too worried about the economic implications. In fact I believe a TSTO RLV with much less ambitious engines than SABRE could be possible (and even the much less ambitious VTVL TSTO RLVs could still be very economical IMO).

I contend this is a reasonable (and not even unusual) opinion to hold. I don't contend other opinions could not also be reasonably held.

I'm suggesting you may have focused on the bits of the design you are familiar with while under rating other aspects that give it its edge.

You are implying I'm not familiar with the aspects that give it its edge. I see no reason to believe that, given that I studied the available documents closely and given that I had to correct you on the turbine inlet temperature. Which aspects do you think I've missed?

SABRE is *like* an ATR as it is *like* a LACE.
It is *neither* an ATR nor a LACE.

I think high pressure deeply precooled ATR with a ramjet bypass duct is a fair characterisation.

So they are uncommon but well within the state of the art.

Agreed, that was my point.

That kind of sounds like a straw man argument to increase Fear Uncertainty & Doubt so far.

Nope, not my intention at all. Remember that I find Skylon very interesting, one of the most interesting airbreathing concepts. Most of the attention goes to scramjets, which strike me as overly ambitious. And I like airbreathers. I'm just pointing out that there's more about Skylon that is unusual than just the precooler and that there's no reason to assume what happened to RS-68 couldn't happen to Skylon, with possibly greater impact since Skylon is intended to be an SSTO.

And they're also aware of patent issues.
No doubt. They are not keen to discuss the frost control system for example.

I meant the helium loop. Hydrogen embrittlement was not the only reason for it.

And what sets SABRE apart from an ATR (apart from the fact that what is being heated is *not* the inlet air) is where that air is going.

No, one of the defining characteristics of an ATR is that the inlet air is not used to drive the turbine, just as with SABRE. There's no difference between SABRE and an ATR in that respect.

Not to large combustion chamber which in a jet engine might be up to say 14atm above atmospheric, but to a rocket combustion chamber running 100-120Atm above atmospheric.

The term ATR is used for both Air Turbo Rocket and Air Turbo Ramjet and the meaning of the two terms even varies from publication to publication. The term air turbo rocket has been used for engines with such high pressure combustion chambers.

I was thinking more in terms of a set of configuration settings for one of the propulsion modelling codes professional engineers use in this field.

If you know of publicly available models, I'd love to hear more.

Well the UK Space Agency funded ESA to review the underlying technology and REL also hosted the 2 day "Requirements Review" on 20-21st of Sept 2010 where over a 100 guests had a chance to critique the design and the economics (perhaps you were there?) and both it and the ESA report were clear they saw no show stoppers, a process which AFAIK is also *unique* to REL, but I hope will be used by others in the future.

An interesting endorsement, but they're still bickering about Ariane 5 ME vs Ariane 6, not advocating a switch to Skylon.

Every orbital launch vehicle I'm aware of has at least 2 stage to achieve orbit and is expendable.

I meant there are many TSTO RLV concepts out there, and it is not clear SSTO will be required for a breakthrough in specific launch prices, nor whether SSTO is even possible without including non-chemical power sources. SSTO could indeed be necessary, but I hope it's not, since I'm not confident a chemical SSTO RLV is possible. Hopeful yes, with Skylon a leading contender, but not confident.