For Reaction Engines, cool is the key

What a difference a generation makes. Thirty years ago a British-led effort to develop an uncrewed, fully-reusable runway take-off and landing spaceplane that promised to revolutionise access to space with a radical single-stage-to-orbit capability – was cancelled. Today, a British-led effort to develop an uncrewed, fully-reusable runway take-off and landing spaceplane that promises make trips to Earth orbit as easy as booking an airline ticket thanks to a radical single-stage-to-orbit capability – looks decidedly plausible.

The 1980s HOTOL programme – for HOrizontal Take-Off and Landing – saw British Aerospace (today’s BAE Systems) and Rolls-Royce eyeing the possibility of a winged spaceplane. The scheme ultimately ran into the sand as expectations of cost and development time spiralled while question marks hanging over the viability of the spaceplane’s design led to concern that it would not outperform much simpler conventional rockets. Equally devastating, the UK’s partners in the European Space Agency (ESA) were not enthusiastic, preferring instead to develop what became the hugely successful Ariane 5 heavy launcher. Meanwhile, the USA showed some interest but London feared that it would end up a junior partner in a transatlantic HOTOL effort – while alienating its European partners.

Today, different paradigms prevail on all counts. ESA is a declared enthusiast and provides technical assistance. UK government money is matched by private-sector support, including from BAE, Rolls-Royce and Boeing. Testing in the USA has been successful – with help from DARPA and the US Air Force – and post-Brexit London no doubt sees the successor programme as an opportunity to play its desired role as a rising space power, to maintain influence in Europe via ESA and to court American favour with participation in a stunning technology.

But the biggest change between then and now is the maturation of the technology at, literally, the core of this project. HOTOL failed in part because it would have relied on a propulsion concept that in the 1980s was alluring but technically very distant. In contrast, today’s sleek and sexy Skylon spaceplane is little more than artists’ impressions of a possible configuration built around an engine born of the radical concept that might have powered HOTOL. In Skylon’s case, though, the key enabling technology has been proven to work.

Over the course of this year and into 2021 engineers expect to demonstrate other critical components and, ultimately a test-bench article. Those engineers work in an office at the UK Atomic Weapons Establishment’s Culham science park at Abingdon, near Oxford, for a company called – intentionally or incidentally suggesting that old rule of aircraft design, to start with propulsion – Reaction Engines.

HOTOL’s 1989 demise led Alan Bond, the engineer whose work on pre-cooled jet engines inspired the project’s propulsion concept, to found Reaction Engines. Persistence appears to have paid off; by the time he retired in 2017, the company had demonstrated pre-cooling technology – convincing ESA, the UK government and the US Air Force Research Laboratory that his hybrid air-breathing rocket engine concept was a goer.

Bond’s Synergistic Air-Breathing Rocket Engine, or SABRE, relies on a pre-cooler capable of turning intake air into the liquid oxygen needed to fuel a rocket engine. That pre-cooler or heat exchanger is essentially a radiator made of many kilometres of tubing the size of fine spaghetti with very thin walls; pumped through those tubes is cryogenic liquid helium, and so air flowing past them is cooled, if necessary to cryogenic temperatures.

Rockets are essentially big tanks of fuel and oxydiser, with suitable mixers at the nozzle. One typical combination is hydrogen and oxygen, which under pressure mix back into water with huge energy release. But fuel is heavy – an Ariane 5 rocket can weigh nearly 800t on the launch pad but merely 20t or so a couple minutes after lift-off – and hence to generate adequate and sustained thrust efficiently, rockets work in stages, spectacularly discarded when empty. However, if pre-cooling can run a rocket on intake air, the vehicle it propels may be able to do away staging – carrying only enough liquid oxygen, which is much heavier than hydrogen, for flight at very high altitudes or in space. A SABRE-powered spaceplane would get its thrust for take-off and early ascent by mixing tanked hydrogen and oxygen from the air until reaching about M5.5 at some 25km altitude, when tanked oxygen takes over and SABRE becomes a normal, self-contained rocket engine for a journey to low-Earth orbit.

UK tests in 2012 showed the pre-cooler concept was viable. At the time, technical director Richard Varvill described the “magic” piece of the puzzle as the ability to prevent the cooler from being completely closed by frost, which it would do in seconds without a technique which, he stressed, is being kept absolutely secret. However Reaction Engines does it, frost-free operation is both critical and very, very impressive, as video of those tests shows; imagine leaving the your kitchen freezer with the door open and coming back the next day to find it clean rather than full of hard snow. At the time, Varvill spoke of 10min steady state runs.

All together, those 2012 tests convinced the European Space Agency that SABRE could actually work. The agency’s head of propulsion engineering Mark Ford called it a “potentially disruptive” technology, and the UK government’s vote of confidence came in the form of £60m (about $80m at today’s exchange rates), to be released in milestone awards pending the company’s ability to raise matching private funds. BAE Systems joined three years later with £20 million, in exchange for a one-fifth share of the company; Rolls-Royce and Boeing followed; programmes director Shaun Driscoll says milestone payments are starting to flow.

In 2014, Reaction Engines signed a co-operative R&D agreement with the US Air Force Research Laboratory. The company formed a USA subsidiary in Colorado, where it achieved a major success in late 2019 by cooling air from 1,000°C to ambient temperature in less than 1/20sec – a “blink of an eye” says Driscoll, who adds that going further, to the -150°C or so of a cryogenic liquid, is possible if rocket engine design ultimately requires it. Significantly, 1,000°C represents air at M5, indicating success at the hypersonic speeds needed to reach orbit.

Those Colorado tests at so-called TF2 – test facility 2, with TF1 still under construction at Westcott, not far from Oxford – also demonstrated Reaction Engine’s ability to work internationally with partners. As Driscoll notes, the head of Reaction Engines Inc. is Adam Dissel, ex-Lockheed Martin and running a small team with an “amazing” culture. TF2, says Driscoll, went from desert to operating in 18 months on a site at Colorado Air and Space Port near Denver; DARPA offered to fund the facility and use existing equipment in the form of an ex-USAF General Electric J79 jet engine with afterburners, as used in F-4 Phantoms, and bolt-on heaters to replicate hypersonic conditions.

The “HTX” test pre-cooler and associated rig was built in Culham and shipped to Colorado at the end of 2018; hot tests started in March 2019 and, says Driscoll, the cooler performed beyond expectations, working just as planned from the start. TF2 will eventually be made available to industry, technology developers and universities who may need its particular capabilities; he describes the facility as “a mini success within the story”.

Not to be overlooked is another piece of magic: Reaction Engines’ ability to manufacture these pre-coolers. The unit used in the 2012 tests, displayed on a mezzanine overlooking the company’s Culham workshop, is about 1m in diametre but packs in some 40km of those fine tubes; in full-scale form a pre-cooler will have around a million brazed joints and must be leak-free – no surprise, the techniques and machinery used to achieve such fine work are another closely guarded trade secret. Driscoll notes that, now, the company is working on automating the process for speed and volume, but meanwhile is winning contracts to do some work with the machinery it has at Culham: “We have actually put things into space.”

 

NEXT STAGE

The Colorado tests opened the path to a demonstration of the air-breathing rocket engine core in 2021. Engineers at Culham working toward that milestone are building key components for testing in the second half, working up to a full-core test next year. While the pre-cooler is the most critical part of the system, there are other innovations in the thermodynamic cycle of the air-breathing rocket engine. One, says Driscoll, is the use of cryogenic helium as the working fluid to cool incoming air; the heat taken up by that helium will then drive the compressors that feed hydrogen and oxygen into the combustion chamber at pressure (an earlier idea was to tap tanked liquid hydrogen for cooling).

With much of the technical development behind it, says Driscoll, Reaction Engines’ focus now is to demonstrate integrated systems. Ground testing of the core can go to a very high level of technology readiness, he adds, because the core will operate the same whether it is flying or not – the core is in that sense insulated from incoming air speed by the pre-cooler, whose performance in a jet of hypersonic air has been demonstrated. Milestone payments from the UK government, via the UK Space Agency, and technical oversight from ESA, particularly at design review points, will help keep the programme moving.

If Westcott is not ready in time to test the complete air-breathing core, Driscoll says there are facilities in Europe to validate the hardware.

Either in air-breathing or liquid oxygen mode – when SABRE will become a normal rocket engine – the subsequent phase of development should be flight testing. Reaction Engines is holding to a longstanding roadmap that puts a cost of around $500 million on having ready a suitable vehicle for trials in the mid-2020s. One key difference between Skylon concepts and HOTOL is the shift from a rear engine – sort of a rocket lying down – to SABRE units mounted on wing pylons, airliner-style to facilitate maintenance and improve the centre of gravity.

While there is clearly much work to be done – and much money yet to be spent – SABRE has the potential to disrupt access to space, as the advantages of runway take-off and landing over normal rocketry would be compelling. Whether such a system could replace big heavy-lift vertical launch systems is ultimately a question of payload capacity, but broadly speaking a SABRE-powered spaceplane offers a flexibility that even air-launch systems like Virgin Orbit’s, which should make its maiden launch imminently, can unlikely match.

Like air-launching a conventional rocket, a SABRE system could operate from any suitable runway with minimal ground infrastructure, to satisfy almost any mission plan. And, once airborne, either can reroute if necessary before launching and even return to base with payload still attached if there is need to abort. Conventional rockets, by contrast, cannot be recovered after lift-off – though payload recovery systems may one day be feasible in the event of catastrophic failure.

Where a SABRE system could shine, though, is in pushing its payload – either a spacecraft or small rocket – all the way to orbit at hypersonic speed. Driscoll sees this inherent feature as another route to reusability which, he stresses, is about saving money but perhaps more importantly about increasing the launch rate and speed of operations. SpaceX has done a good job with conventional technology, but it “can only go so far”, he says: “That’s our view – there’s a plateau.” SABRE, he stresses, is a technology shift.

And, he adds, Reaction Engines talks about achieving a “magnitude” reduction in launch costs – not 10-20% but much, much more, enough to justify the investment in new technology.

But while space is sexy, pre-cooling technology has much wider applications. At the tests in Colorado, notes Driscoll, the HTX text rig took 3.8MW of heat out of that M5 jet exhaust – enough to power 1,000 homes. Power plants generate vast amounts of waste heat, sometimes recovering some of it with huge infrastructure, so the energy market is rich with applications for compact, efficient heat management units.

Indeed, Reaction Engines has established an Applied Technologies business unit looking at spinoffs in thermal management. At Culham, Driscoll showed Flightglobal a “pre-cooler” unit about the size of a coffee jar, built for motorsport and to use water as a cooling fluid. Ordinary road-use electric vehicles also have heat challenges that could be addressed with such gadgets.

Civil and defence aerospace are also prime candidates for Reaction Engines magic. A SABRE-style heat exchanger could feed cool air to hot-section compressor blades, which in current jetliner gas turbines run at temperatures above the melting point of the metals they are made from and so have to be cooled by pumping less-hot air through internal holes, the formation of which greatly complicates blade manufacture. Better cooling with colder air may permit even hotter combustion.

Another technique would be to use a heat exchanger to take heat from the exhaust and feed it back into the combustion chamber, to get work from energy that would otherwise be wasted. Varvill previously told Flightglobal that these approaches might cut fuel burn by 5-10%, though would require new engine architecture rather than any bolt-on approach.

The other obvious aviation application of SABRE technology is in high-speed flight in the atmosphere, either supersonic or M5-plus hypersonic. In July 2019, Rolls-Royce announced a two-year study for the UK Ministry of Defence into hypersonic propulsion systems for the Project Tempest future air combat system project. The plan is to adapt a Eurojet EJ200 turbofan, as used on Eurofighter Typhoons, with a Reaction Engines pre-cooler to test the effect on engine performance of reduced inlet temperatures.

Driscoll adds that DARPA is also interested in pre-coolers with conventional engines. And, Reaction Engines is looking at civil supersonic flight, where it might improve thermal management of both the engine and the whole aircraft – especially as these become more electric. The company believes, he says, that it could help enable a step-change in supersonic cost and emissions.

 

SUSTAINABLY BRITISH

For the moment, Reaction Engines is broadly keeping to promised development timetables and delivering on performance promises. Chief executive Mark Thomas, a former Rolls-Royce engineer who joined the company in 2015, has even talked of SABRE in terms of a “Whittle moment” – referring, of course, to the UK’s jet engine pioneer.

Such huge potential corresponds to huge demands on money and talent, raising the obvious question of whether or not the company can remain wholly or even majority British. Driscoll says that Reaction Engines sees itself as having a “robust, sustainable business model”, with some commercial work now coming through to exploit its proprietary cooler manufacturing technology at Culham. As for what comes next – say, does Reaction Engines eventually become a listed company? – he simply observes that there are many options and nothing is spelled out yet, but big ambitions need a partnering arrangement.

In any case, he adds, deciding what to do next is in part about hanging on to culture and hence talent; Reaction Engines works at pace and with agility to do novel things, learn from failure and keep moving forward.

And, whatever lines of business it ultimately pursues, space “is an exciting industry to be in”; Driscoll sees Reaction Engines hiring young graduates who could instead choose to go to “brand-name” companies; investors, too, understand the journey and don’t expect too much too soon – but horizons are broad; work on rockets lends itself to helping solve big problems.