The venerable space telescope spent weeks out of action this summer. Melissa Brobby looks at how NASA brought it back to life and asks how long Hubble has left

Scientists around the world breathed a sigh of relief on 15 July. After a nervous few weeks, NASA had returned the Hubble Space Telescope’s science instruments to operational status, and science data collection was able to resume for the first time since the space telescope stopped working in mid-June.

The nerves were understandable: since deployment in April 1990, Hubble has been one of the most productive scientific instruments ever built. It has made over 1.3 million observations, contributing data for more than 18,000 published scientific papers on a broad range of topics from black holes to planet formation. Astronomy textbooks include contributions from the orbiting space observatory, and Hubble’s discoveries and memorable images have ignited the public’s fascination with space over its 31 years in orbit. But is Hubble now starting to show its age? Are its components just too old? And is the launch of the James Webb Space Telescope this November coming at just the right time?

First, let’s go back and look at what Hubble’s issue was. The short answer is its payload computer went wrong. This sits in the Science Instrument Command and Data Handling (SI C&DH) unit on board the telescope and controls, monitors and coordinates Hubble’s science instruments. On 13 June, the payload computer suddenly stopped working and it was unable to send its regular signal to the main computer, telling it all was well. Because of this, the main computer placed all science instruments in safe mode and suspended the telescope’s observations.

The Hubble team had to move swiftly to work out the cause of the halt and fix it. But fixing a telescope built in the 1980s and orbiting 550km above Earth needed the help of those who have worked on it over its 30-plus year history, and original paperwork dating back up to four decades.

Hubble has backup modules but switching to these did not resolve the issue. So a series of tests – including attempts to restart and reconfigure the main and backup computers – were carried out and, although these were also unsuccessful, results from them showed that the possible cause of the issue was the Power Control Unit (PCU), designed to supply a steady voltage to the payload computer’s hardware. Addressing these issues would add additional risk, as switching to these components’ backup units would mean switching several other hardware boxes too.

More than 50 people went through the procedures to switch to backup hardware, testing them on a simulator. On 15 July, the team planned the switch to the backup SI C&DH unit, which contains a backup of the PCU. Around 15:30 UT, the team declared the switch successful and the science instruments were bought back to operational status, allowing Hubble to begin taking scientific data again.

Early problems

Alongside the discoveries and breathtaking images, Hubble has not been free from problems since it started circling Earth. Designed to be serviced in orbit, the space telescope was built with modular components that astronauts could handle and replace easily, and Hubble could be fitted with new science instruments and other equipment during servicing missions, which lasted from 1993 to 2009.

But these couldn’t help with the major fault that was discovered just after launch in 1990, when it was immediately apparent that its images were blurred. The fault was traced to Hubble’s primary mirror which had been ground precisely, but to the wrong shape. Astronauts were sent on the first servicing mission in December 1993 with replacement instruments that fixed the flaw, allowing Hubble to capture such famous images as ‘The Pillars of Creation’.

The second service mission took place in February 1997 to replace degrading spacecraft components and install the Space Telescope Imaging Spectrograph (STIS), and Near Infrared Camera and Multi-Object Spectrometer (NICMOS) instruments. The STIS separated light taken in by the scope and spilt it so the composition, temperature, motion and other properties could be analysed; while the NICMOS allowed scientists to see clear views of the Universe at near-infrared wavelengths for the first time.

Then in November 1999, the fourth of Hubble’s six gyroscopes failed. The gyros measure Hubble’s rate of motion and help the telescope point towards its observation targets, and it needs three to be working to make observations. This led to Hubble going to sleep while a fix was awaited. So the third servicing mission, originally a maintenance mission, was spilt into two parts: Servicing Mission 3A flew in December 1999 to replace all six gyroscopes; and in March 2002 Servicing Mission 3B replaced solar panels and installed the Advanced Camera for Surveys (ACS), taking the place of the Faint Object Camera, which was the telescope’s last remaining original instrument.

During Hubble’s fourth servicing mission in May 2009, astronauts performed the first ever in-orbit repairs on the STIS and ACS, which had both failed. Along with new batteries, new gyroscopes and a new science computer, these replacements played an instrumental part in prolonging Hubble’s life.

The problems this June are not the first time that backup hardware has come to Hubble’s rescue. In 2008, the Hubble team performed a switch similar to the one that took place recently, when another part of the SI C&DH unit failed. The unit was replaced during the final servicing mission on 11–24 May 2009.

Since the Space Shuttle’s retirement, servicing Hubble from the ground was always going to be a challenge, especially as it continued to run into problems after its last servicing mission. In March 2014, one of Hubble’s gyroscopes failed, followed by another in April 2018 and the last in October 2018.

The operations team activated a backup gyro but, according to NASA, the backup “…incorrectly returned rotation rates that were far in excess of the actual rates,” so Hubble was once again placed into safe mode while the problem was fixed on the ground.

The operations team instructed Hubble to perform a series of manoeuvres and switched the gyro between different operational modes to ensure it was stable. Once the team had tested pointing accuracy and target activities without issue, Hubble’s scientific instruments were restored to normal operational mode with three functioning gyros.

However, Hubble entered partial safe mode in January 2019 due to suspected hardware issues in its Wide Field Camera 3 (WFC3) unit – the telescope’s last and most technically advanced instrument, installed during Hubble’s fourth Servicing Mission in 2009. The cause of the failure was found to be a software problem, so the telemetry circuits were reset and the (WFC3) was returned to normal operation.

Now that Hubble’s latest issue is fixed, what’s next for it? There may not be any in-orbit repairs scheduled, but there is a dedicated team of engineers and scientists working hard to keep Hubble operating for as long as possible. There are currently no set plans for Hubble’s retirement; Nzinga Tull, Hubble Systems Anomaly Response Manager at NASA’s Goddard Space Flight Center, says the telescope is expected “…to be at or near full functionality through the mid-2020s, at least, and we are considering alternative operational modes so that we can continue a reduced science mission in the future.”

While Hubble continues on, the James Webb Space Telescope is getting ready to begin its mission, with a target launch date this year on 31 October. JWST is often referred to as Hubble’s replacement, but NASA sees it more as its scientific successor because its capabilities are different. While Hubble studies the Universe at optical and ultraviolet wavelengths, JWST will primarily observe in the infrared.

The JWST has a much larger 6.5m-diameter primary mirror, which is huge compared to Hubble’s 2.4m mirror, and it also has a much larger field of view than Hubble’s NICMOS camera, covering more than 15 times the area. The JWST will be able to look further back in time than Hubble, giving a glimpse of the birth of the first galaxies. It will look inside dust clouds where stars and planetary systems are forming, and study the atmospheres of exoplanets to see where else the building blocks of life are developing in the Universe.

Lessons learned

Despite a multitude of launch delays, NASA is confident that the lessons learnt from Hubble will prevent the JWST from experiencing the same issues. Destined for a position 1.5 million km from Earth (at the L2 Lagrange point – where the gravitational forces from the Sun and Earth exactly balance the centrifugal force) the JWST is not designed to be serviced by spacecraft. Rigorous testing on the JWST’s mirror has taken place before launch, including an alignment check of its 18 gold-plated, hexagonal primary mirror segments to ensure they act like a single mirror – testing that NASA had not done on Hubble before its launch.

The JWST is the biggest and most technologically advanced space telescope built and a lot is riding on its success, especially as its costs have reached about $10bn. Whatever challenges it throws at its team on the ground, it’s set to build on Hubble’s record and open up a new side to the Universe that will delight astronomers and the public for years to come.

---

Keeping an eye on Hubble

Meet Nzinga Tull, Hubble Systems Anomaly Response Manager

What does your role at NASA involve and how long have you been in it for?

One of my primary roles is as an Anomaly Response Manager who uses systems engineering expertise to coordinate anomaly investigations and lead anomaly response and recovery efforts. After 13 years on the Hubble mission operations team earlier in my career, I rejoined the team in November 2018.

How long did the problem with Hubble take to diagnose and how serious was the issue?

Hubble has an extensive, automatic data-monitoring system that sends system engineers ‘alert’ signals when something goes wrong; so we were alerted immediately when the payload computer first halted on 13 June. There were about two weeks of dedicated anomaly investigations before we began planning to recover the payload by activating back-up components. The meticulous development and testing of the recovery procedures took about 2.5 weeks to complete in parallel with on-going root cause analysis before the on-orbit recovery was attempted.

The issue with the Power Control Unit (PCU) was serious enough to halt the science mission but we still don’t know for sure if the issue is permanent.

Is it possible that the same issue with the payload computer could happen again and is there anything that can be done to prevent the problem from recurring?

If it is a transient problem that can be improved, say, if the Power Control Unit sits cold for some time, we may still have some redundant functionality down the road. But, for now, we are prioritising the continuity of the science mission and we will continue to operate Hubble using the backup components.

What is next for Hubble now the issue is fixed? How long is the space telescope expected to be in operation for?

Now that payload computer is working again, we have re-started the science instruments and they are back to collecting the amazing images that make Hubble a beloved resource with astronomers and the stargazing public. The systems engineering team will continue to monitor and analyse telescope data to get ideas on how to optimise operations and extend mission life for years to come.

Computer says no

The unit that led to Hubble’s outage in June was a vintage computer component

Hubble’s payload computer is a NASA Standard Spacecraft Computer-1 (NSSC-1) system, which sits in the Science Instrument Command and Data Handling (SI C&DH) unit. Hubble’s NSSC-1 system was built by IBM in the 1980s but the original design, which was developed at the Goddard Space Flight Center as a standard component for the MultiMission Modular Spacecraft, dates back to 1974. It was developed as part of a set of modules and components that would help reduce costs, and was used on the Landsat 4 and Landsat 5 missions, the Solar Maximum Mission and other missions mainly limited to the Solar System, including the Gamma Ray Observatory and the Upper Atmosphere Research Satellite.

NSSC-1’s purpose is to control and coordinate the science instruments onboard Hubble, monitoring them for health and safety. Hubble has two payload computers, one of which serves as a backup that can take over in the event of a problem. Both computers have access to any of the four independent memory modules, which each contain 64K of core memory. The payload computer uses only one memory module at a time while the other three serve as backups.

IMAGES: ESA, ILLUSTRATION