By | August 15, 2001 | Feature

Chris Bulloch Interspace

The Arianespace Enquiry Board investigating the AR.5 (Flight 142) upper stage malfunction of July 12 reported back on August 1 as promised, though its findings were not made public until six days later. As a result of the findings further modification work will need to be carried out on the Ariane 5 series of launchers causing a delay of some two months to flights.

Arianespace says it hopes to resume flight by the end of November. As to the cause of the AR5 problems, the Board identified a “combustion instability” during the ignition of the Aestus upper-stage or EPS (storable propellant) engine, which is powered by monomethyl hydrazine (MMH) burned with nitrogen textroxide (N2O4) oxidizer. This instability, or fluctuation of the flame, was attributed to an “unfavourable dynamic coupling” between the propellant feed system and the internal fluid cavities of the combustion chamber.

A result of this instability was degraded engine combustion, leading to a lower-than-normal thrust output (20 per cent down on nominal value). The on-board guidance computer attempted to compensate for this by increasing the burn time and postponing a number of planned flight events in order to achieve the nominal trajectory.

However, the upper storable-propellant stage shut itself down some 80 seconds before it should have done, due to exhaustion of one of the two propellants. According to Arianespace, this was the nitrogen tetroxide needed for burning the hydrazine fuel. At that point, the stage’s velocity was some 500 metres per second slower than the 9065m/sec required for correct orbital injection. As a result, the two payloads were put into excessively low orbits (17,528 x 592km instead of the 35,853 x 858km required), and at an inclination to the Equator of 2.9 degrees instead of two degrees.

According to Eduard Perez, Arianespace senior VP of Engineering, the Enquiry Board determined that all processing and countdown operations for Flight 142 were normal, as was the actual flight prior to ignition of the upper-stage engine. The Board concluded that nothing (such as excessive vibration) could have affected the stage before this point. Also, the Board determined that there was no connection between this AR.5 anomaly and the Ariane 4 launcher, since there is little if any commonality between the two vehicles. The next Ariane 4 launch – of Intelsat 902/Flight 142 – will proceed on its planned end-August lift-off.

The Board recommended that:

  • Hydraulic conditions occurring during the Aestus engine ignition should be modelled mathematically
  • The upper stage ignition should be improved to make it steadier and smoother
  • Qualification criteria should be adapted to the modified ignition phase,and test benches adapted to duplicate flight conditions more closely
  • In-flight operating margins should be demonstrated by a test programme, and the next flight engines qualified according to these new criteria.

Arianespace and its contractors will adopt these and other recommendations. In addition, the companies will update the dynamic modelling of the propellant supply system, with particular attention to the hydrazine fuel control system. This gives colour to the assumption that hydrazine supply was defective on Flight 142. Certain combustion chamber parts will also need an update of their mechanical modelling.

If this action plan is applied in a timely fashion, Arianespace expects the next Ariane 5 mission to take place by late November. This means as little as two months’ slippage from the original schedule. But any further delays in the remedial work will push A5 launch schedules into 2002.

Effect on Arianespace

Meanwhile, the effect on Arianespace’s mission planning cannot easily be determined. Much depends on the availability of satellites to launch. The fall-back option is of course Ariane 4; there will remain 11 of these in inventory after the Intelsat 902 launch later this month; it is now planned to launch three more this year AR.4 had already been earmarked as a single-payload vehicle. There are no plans to build more.

The only near-term payload awaiting launch for which Ariane 5 is mandatory is ESA’s Envisat, which involves lofting 8.3t into sun-synchronous orbit, under a new long fairing. Apart from this, the next ‘really big’ payload, too heavy to fly on AR.4, is probably Telesat Canada’s Anik F-2 at 5.9t, but this is not due for launch until December 2002. Most if not all prospective payloads can be accommodated on AR.4, but there are more than eleven of them. An early return of AR.5 — and no further problems – is therefore vital.

Frequently payload availability creates just as much of an impact on Ariane’s finances as the current technical set-back; it is already apparent that ISRO’s Insat-3C may be late. But to balance this (as a single launch), Eutelsat wants Atlantic Bird-2 into orbit as soon as possible. In the near term, Arianespace still hopes to break even this year (following a E242 loss in 2000, due largely to late satellite delivery). But revenue yields from single-payload AR.4 missions will damage its income stream.

The company expects to fly its first post-failure AR.5 mission with a paying payload. Whether ESA will be prepared to take a chance with its expensive Envisat (total cost around E2.4 billion) remains to be seen, especially since the agency’s Artemis was a victim of the Flight 142 near-failure.

Backstage murmurings

There are also some criticisms of the Enquiry Board from people who think it is “too French”. Its president was Roger Vignelles, originally from the French space agency CNES and then until his retirement president of the former SEP division of Snecma Moteurs. Three other members of the seven-member Board were also French.

Payload progress

As previously reported (Interspace 742) one of the Flight 142 payloads (ESA’s Artemis) is now established in an 18-hour sub-geostationary circular orbit at some 31,000km altitude. This was accomplished by successive firings of its bi-propellant liquid apogee motor, thus largely depleting its fuel supply.

Starting around end-September, it will be spiralled-out to its nominal24-hour orbit at just under 36,000km by means of its ion engines.

Live chat by BoldChat