CAPE CANAVERAL, Fla. — NASA engineers are preparing a plan to return the planet-hunting Kepler space telescope to service following a reaction wheel failure that shut down the four-year-old observatory in May.

“I think the general feeling is that the odds are not good. We might see a wheel spin, but I suspect that it will not spin freely, that there will  be noise on it — vibrations — which would not make the science happy,” Charlie Sobeck, deputy project manager at NASA’s Ames Research Center, told SpaceNews.

The May 15 failure of Kepler’s reaction wheel, needed to keep the telescope’s laser-like focus on its target stars, was the second of four wheels to shut down. The telescope needs three wheels to keep its gaze steady enough to catch the slight eclipses of starlight caused by orbiting planet passing by, relative to Kepler’s line of sight. 

Since its launch in March 2009, Kepler has added 132 confirmed planets beyond the solar system to a list that now numbers 723, plus another 3,216 candidate planets awaiting confirmation. The telescope is the cornerstone of a $650 million program to find Earth-sized planets orbiting sun-like stars at the right distances for liquid water to exist on their surfaces. Water is believed to be necessary for life.

The telescope completed its primary four-year mission and had just started an extended stint when the second wheel failure halted science operations.

Sobeck, who is overseeing the rescue effort, said his team is preparing to send commands to Kepler in mid- to late-July, first to characterize the condition of the wheels, and then to try to get one moving again.

“We’re taking this in steps, and that will inform future activities,” Sobeck said.

When engineers last checked, the first wheel to experience problems could still move, but only with a great deal of friction. 

“We would expect that it might very well seize up as soon as we try to spin it up again,” Sobeck said.

The second wheel to fail was completely seized and Sobeck expects no change in that status. 

The next step would be to see if the wheels can be moved again. 

“The options are pretty limited, but we’ll do some things, like warm them up to provide as much lubrication as we can, and then we’ll command them in both directions, in sort of a ‘rocker test’ as we call it, rocking first clockwise and then counter-clockwise to see if we can’t move whatever is binding the thing up out of the way,” Sobeck said.

The team expects to work on the last wheel to fail first — Wheel 4 — and then move on to Wheel 2, which shut down last year. Wheel 2 has a somewhat better chance of being brought back into service. 

“We’ll learn any lesson on the wheel that’s less likely to respond, and then well move on to the better bet,” Sobeck said.

“If either of those wheels starts to spin and spins freely then we’ll stop whatever additional testing we had planned because the goal would be to unstick the wheel, and if it’s spinning then it’s unstuck,” he added.

Scientists have some preferences for studies if the telescope can be revived, but overall they’re looking for any observation period that can last for three months or longer in the same orientation.

“They would like to have a data set that is one full-quarter long. They don’t care particularly which quarter — we look at the same stars each time — it’s just a matter of getting a complete quarter-set of data,” Sobeck said.

“I don’t think that the odds are real great,” he added, “but we could be surprised. It could be that when we command Wheel 2, which hasn’t been commanded for a year, that having sat there for a year soaking in its own lubrication that it’s going to start right up and give us another 10 months of operations. We don’t know, but we’ve got to at least try.”