In Apollo-11, during preparations for docking with Command and Service Module in lunar orbit, Armstrong and Aldrin managed to put Lunar Module Primary Guidance System Inertial Measurement Unit (IMU) into gimbal lock.Source

127:53:21 Armstrong (onboard): That roll's pretty far. I don't know just how much - so that's - Oh, it's going to go gimbal lock! 127:53:28 Aldrin (onboard): That's it - going to AGS? 127:53:32 Armstrong (onboard): Yes, [garble] Att Deadband.

Armstrong, from 1969 Technical debrief:  "... No doubt, we were firmly ensconced in gimbal lock. We had all the lights on, the DAP was not operating anymore, we had no control outputs, clearly no CDU outputs were being processed, so we just put it in AGS and completed the docking in AGS."

In LM, apparently, there was a back-up guidance system : Abort Guidance System (AGS)

This system was not as accurate as the primary IMU but allowed to perform basic attitude control for docking with CSM.

What was the back-up plan in case the CSM Primary Guidance System IMU was put into gimbal lock (especially on the trip back from the moon, after separation from LM)?


1 Answer 1


The Apollo attitude control system starts from a known orientation established by taking star sightings, then uses gyroscopes to track the change in orientation over time. The problem with gimbal lock is that the system can no longer distinguish the axes of rotation properly, so can't tell how the orientation is changing.

The remedy for gimbal lock is just to start over -- reset the physical positions of the gimbals to take it out of lock, take a new set of star sightings, and tell the computer which way you're pointed in a process called "realigning" the IMU. I believe this was program P52 in the AGC, and the process is described in another QA here.

This process does take a while, so you want to avoid it if you can, and it can be a big problem if you're in the middle of something time-sensitive like a docking approach, which is why Apollo 11 switched over to AGS instead of realigning.

Since the gyros do drift slightly over time, the IMU needs to be realigned periodically even if you don't go into gimbal lock, so the procedure is fairly routine.

  • 2
    $\begingroup$ The document hq.nasa.gov/alsj/e-1344.htm describes a quick "coarse align" procedure that could be done if lock occurred. But this document's pretty old and I don't know if that approach survived into operations. "Once this is done he can operate the "coarse align" mode control provided at the pilot's AGC keyboard. This would quickly slew the IMU to the spacecraft attitude described by the CDUs. " $\endgroup$ Commented Nov 22, 2019 at 17:38
  • $\begingroup$ How did they suppose to "reset the physical positions of the gimbals to take it out of lock"? If I understand correctly, gimbal lock means two initially orthogonal frames (gimbals) coincide in one plane therefore start to move together. Was there a special mechanism to force an individual gimbal to move relative to the others? $\endgroup$ Commented Nov 22, 2019 at 21:29
  • 1
    $\begingroup$ @OrganicMarble But there should be a difference between a standard IMU drift alingnment without a previous gimbal lock and the IMU alignment after a gimbal lock. The standard IMU alignment using P52 was very comfortable, the astronaut entered a star number, the guidance computer rotated the CSM to view the star with the sextant and the astronaut measured the error. Repeating the procedure with another star and all 3 angle drifts are corrected. But after a gimbal lock, the astronaut had to rotate the CSM manually to the star. $\endgroup$
    – Uwe
    Commented Nov 22, 2019 at 21:36
  • $\begingroup$ @LeoS I don't know the mechanism by which the coincided gimbals were de-coincided, but it seems evident they had to do so somehow. I'm guessing it was done by the computer as part of the P52 process. $\endgroup$ Commented Nov 23, 2019 at 0:25
  • 1
    $\begingroup$ @LeoS you can read about the IMU (and the torquers) in section of this document hq.nasa.gov/alsj/SM2A-03-BK-II-%281%29.pdf $\endgroup$ Commented Nov 23, 2019 at 23:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.