HST @ STScI Update

R. Osten (osten[at]stsci.edu)


The Hubble presence at the recent winter AAS meeting in Seattle consisted of the STScI booth with informational pamphlets, opportunity for interaction with instrument experts, and posters on calibration performance. In addition, a special session entitled "A Hubble Space Telescope for the 2020s: Capabilities and Opportunities" occurred on Thursday, January 10. This special session highlighted new results from the observatory in some of the forefront science areas in which Hubble excels. The session was forward-looking and intended as the start of a dialogue between Hubble’s science operations center and the scientific community about the role Hubble should play in the science of the 2020s.  The format consisted of a few short talks to describe new results and motivate future observing strategies, followed by a panel discussion with input from the community.

Panel discussion afterward touched on lifetime-limiting aspects to observatory operations, plans and capabilities in reduced gyro mode, and interest in the ultra-flexible observing programs (described in the 2016 STScI Newsletter Article Volume 33, issue 2). The oral session was accompanied by a poster session with numerous submissions, ranging from updates on new observing modes and tools (some described in the 2018 STScI Newsletter Article in Volume 35, issue 3) to specific science topics that could be addressed with Hubble in the next decade. Collected talks are available at this link.

The winter AAS meeting also marked the second instance of an STScI Town Hall at the AAS meeting. This is a recognition by the society of the valuable role the institute plays in enabling science. The agenda for the Town Hall included general updates on the status of STScI, a description of public engagement opportunities for astronomers, as well as how the institute is leveraging work on Hubble and Webb for WFIRST science operations. A series of short talks highlighted: Hubble research and results in Solar System and exoplanet science; work on the exoplanet characterization toolkit (ExoCTK); laboratory work exploring planetary atmospheric conditions; and technology development for high-contrast imaging with segmented telescopes.  A description of some of the processes implemented at STScI to build a safe and inclusive workplace in astronomy concluded the session. The charts presented at the meeting can be found at this link.

Observatory Status

Default Image
Figure 1: This image of a portion of the Carina Nebula, taken from program 15238, PI A. Kraus, was the first WFC3 image captured by Hubble after anomaly recovery on Jan. 13, 2019. It was taken with the IR channel using the F139M filter.

Instrument Anomalies

On January 8, 2019 the Wide Field Camera 3 on Hubble suspended operations due to an anomaly on the instrument's UVIS channel, and was restored roughly a week later, with a return to science observations a few days after that date. The anomaly occurred within the CCD Electronics Box (CEB) on the UVIS channel. Analysis indicated that it was safe to reset the electronics and recover the instrument to an operational state. Those steps were performed without incident. An investigation into the cause of the corrupted telemetry that triggered the suspension of operations concluded that it was likely caused by a single event upset within the analog to digital collection electronics for CEB telemetry. Since that recovery, WFC3 UVIS has been operating nominally. Figure 1 shows an image obtained shortly after recovery.

ACS experienced a temporary suspension of operations from February 29 to March 6. An error arose during a check of the ACS memory, which occurred while the instrument was transitioning between operational states. Several tests performed to troubleshoot the anomaly—involving different ways of writing to and reading from the ACS memory—executed successfully. Recovery of the instrument proceeded following a detailed review of schematics and anomaly signature characterization. The on-board memory check was expanded to include the affected chip area.  Observations with the remaining instruments continued unimpeded during the ACS suspension. The search for the most likely cause continues; it does not appear to be due to a failure in a memory chip (for which ACS retains significant redundancy). The instrument experienced a second temporary suspension on the morning of April 3. The instrument was recovered that afternoon and troubleshooting is happening in parallel. In both events, ACS was in the process of returning to operations after its monthly anneal and reboot, with the suspension triggered by erroneous values in an unused portion of instrument memory. The repeat indicates that the problem is unlikely to be  a random event. There is consensus that it is safe to operate the instrument. Changes to the monthly maintenance procedures will mitigate or prevent future problems. Since that recovery, ACS has been operating nominally; see Figure 2.

Default Image
Figure 2: ACS image of an outer region of the globular cluster Omega Centauri in the B, V, and I filters, taken on March 14, 2019, after the return to science following recovery from the first anomaly. Image credit: K. Platais, J. Ryon.

Recent Gyro Behavior

Last year saw the failure of two of Hubble's gyroscopes, used for pointing and control (STScI Newsletter Volume 35, issues 1 and 3). During the time before Gyro-2 failed, the observatory experienced an elevated level of jitter which resulted in an increase in the failure rate of acquisitions and re-acquisitions. Operations with Gyro-3, brought up after the failure of Gyro-2, indicate that while the jitter levels are stable at values consistent with those seen in prior years, this gyroscope does exhibit large rate bias shifts which impact acquisitions and re-acquisitions.

There continue to be short periods of time where there are an increased number of failures in acquisitions and re-acquisitions. There does not appear to be one single cause. The large rate bias shifts of Gyro-3 can result in the Fine Guidance Sensors requiring a larger search radius during acquisitions and re-acquisitions; this can extend the time to find the guide stars, and delay or lose the ability to take data for science. Performing an On-Board Attitude Determination before most re-acquisitions is minimizing the attitude error and limiting failures. Additionally, adding extra time (~60 seconds) for re-acquisitions ameliorates the situation in some instances, without significant loss of science. Changes to the magnitude limit used to select guide stars and the sampling rate for those stars are also expected to mitigate some of the acquisition issues. On-board bias updates are now done routinely and frequently, particularly for Solar System observations. This results in increased difficulty scheduling programs and crafting the weekly observation plan.

Sun Angle Constraints

The nominal Sun Angle limit for science observations with Hubble has historically been 50 degrees.  As noted above, Gyro-3 experiences large rate bias shifts; having a large uncompensated bias while on gyro control can lead to large undetected attitude errors. This poses a number of risks to the spacecraft. Since the return to science using Gyro-3 on October 26, a more conservative Sun Angle limit has been in place to ensure observatory protection against thermal concerns in the event of unexpected Gyro-3 behavior. It is currently 54.3 degrees, which is likely the limit for the foreseeable future in three-gyro mode.

Longevity of Hubble, Reduced-gyro plans

This April marks the 29th anniversary of the start of Hubble’s on-orbit science operations, and plans are already underway for next year's 30th anniversary since launch. In addition, in May a full decade will have passed since the last servicing mission. With the exception of a few minor anomalies with instruments, the observatory health is generally very good. The instruments are aging gracefully, and scientists' and engineers' understanding of and compensation for instrument degradation generally outpaces the rate of actual declines. NASA's Engineering and Safety Center routinely re-examines assumptions driving the probability lifetimes of Hubble's subsystems. Current estimates give an 80% reliability or better for all subsystems in 2025, including having at least one gyro for pointing and control.

Gyroscopes have long been a limiting factor to Hubble's performance. A complement of three gyroscopes is normally used for regular science operations. The current gyroscope configuration uses Gyro-3 together with Gyro-4 and Gyro-6. All three have enhanced flex leads, with an anti-corrosive coating to overcome a vulnerability discovered on earlier generations of gyros. The gyros which have failed, Gyro-1 and Gyro-2 last year, and Gyro-5 earlier, were all non-enhanced and lasted as expected, for about 50,000 hours. Gyros 3, 4, and 6 are the remaining gyroscopes on the observatory. The current plans are to proceed using all three, until there is another failure, at which point Hubble will enter either one-gyro mode or two-gyro mode (depending on the particular condition of the remaining gyros at that time). On-orbit experience with reduced gyro mode occurred in Cycles 15 and 16, with two gyros for pointing and control, in addition to a one-week period when tests of one-gyro operations occurred. There is very little difference operationally between the two reduced-gyro modes, and the longevity is extended by keeping one gyro in reserve. The science capabilities of Hubble in reduced-gyro mode are also only modestly decreased from that in three-gyro mode operations, with expectations of science productivity at roughly 75% that obtainable previously, and most observing modes still realizable.

Instrument teams are proactively examining calibration plans for their sensitivity to reduced-gyro mode in monitoring of external targets, and modifying as appropriate to be robust against the impact of any changes to the number of gyroscopes on target availability. In addition, user documentation has been updated with reduced-gyro information to be prepared when this is needed.