Spacecraft number five (FM5) was to be named Phoenix, after a mythical Arabian bird which was burnt on a funeral pile and then rose from the ashes to live again.

Phoenix was given the go-ahead by the ESA Science Programme Committee in July 1996. It was later agreed that the potential science return from a full Cluster reflight was so important that a further three near-replicas of the original spacecraft would also be built.

Today, the fruits of this persistence and vision are visible for all to see. Phoenix, the last of the four Cluster II satellites to be integrated, is undergoing a thorough post-assembly examination in the test facilities of IABG near Munich.

Just before Christmas the spacecraft has completed the acoustic tests which ensure that it can survive the shaking endured during launch next summer. Blasted with 144 decibels of sound (not just sufficient to deafen, but to kill anyone foolish enough to enter the room) from a giant horn-shaped loudspeaker fitted into a wall -- Phoenix has been subjected to vibrations across the entire frequency range from 30 to 8000 Hz.

Having successfully come through this traumatic trial, the satellite is now ready to be baked and frozen during a series of thermal-vacuum tests beginning in the second week of January.

Though outwardly identical to the other Cluster II spacecraft, the present- day Phoenix is an unusual combination of the old and the new. At its heart is the first spacecraft structure ever manufactured during the original Cluster programme back in 1992. Although never intended to fly in space, this main body was used for a variety of vibration and shock tests and eventually grabbed some limelight at the 1995 Paris Air Show.

There are also important differences between Phoenix and its Cluster I predecessors. Under the first Cluster revival plan, Phoenix was to have carried spare experiments, but most of its scientific payload is now composed of new instruments.

"We decided that, since we were having to make three new units for each experiment, it would be just as easy to make four," said John Ellwood, ESA Project Manager for Cluster II.

Significant modifications made to its design since the days of Cluster I include the addition of a solid state data recorder with a larger memory; two new computer boxes, a new high power amplifier, digital transponder and experiment booms which have been slightly shortened to fit inside the protective fairing on the Soyuz rocket. Various other components which are no longer manufactured have also been replaced.

"Nothing from the four Cluster spacecraft that we lost has been used again," commented John Ellwood. "To all intents and purposes, we regard Phoenix as a new spacecraft. However, it will always remind us of that tragic day in June 1996, and reinforce our determination to succeed the second time around."