I couldn't find an answer to this question.

James Webb Space Telescope will align primary mirror segments and secondary mirror in space.

Question is whether these will need re-alignment during the course of operation? And how often if so?

  • $\begingroup$ This is a great question! With such a large structure and even with low coefficient of thermal expansion materials, it's possible that even small changes of temperature (or its gradient) due to orientation changes might require some retuning. It's also possible that after some period of data collection that might become predictable and even preemptive. $\endgroup$
    – uhoh
    Jan 4, 2022 at 22:41
  • $\begingroup$ I'm pretty sure it would, though "how often" is much more difficult to guess at. (Presumably this is something that will be regularly monitored and adjusted as needed.) $\endgroup$ Jan 5, 2022 at 9:59

2 Answers 2


Each of James Webb's mirror segments are adjustable through seven small motors, called actuators. Due to thermal variations, the segments do indeed need to be re-adjusted occasionally.

Exactly how often is a compromise between desired image quality and valuable science observation time, as argued by Park et al. (2017) who develop a thermal model, but don't seem to actually recommend any interval between re-adjustments.

An older proceeding by Mather (2010) says that

it is […] expected the JWST primary mirror may be readjusted on the order of every two weeks.

which is backed up by this recent SPIE article.

  • $\begingroup$ I suspect that 14-day "sampling period" is a tradeoff between the bandwidth (calculated) of all thermal effects, the amount of power neede to issue a correction, and the image degradation due to loss of primary figure. By comparison, ground-based adaptive optics, with the same control loop, need to run on the order of a kHz to compensate for atmospheric bandwidths up to 100 Hz or so. $\endgroup$ Jan 5, 2022 at 16:13
  • 1
    $\begingroup$ Amazing answer, thank you, pela. @CarlWitthoft, isn't ground based adaptive optics working in a very different way? $\endgroup$ Jan 5, 2022 at 22:27
  • $\begingroup$ Good point with the power needed, @CarlWitthoft, which isn't a problem for ground-based or LEO telescopes. My un-educated guess would be that the power needed to adjust the mirrors (typically by fractions of a wavelength I think) is negligible compared to slewing the telescope and positioning instruments, filters, coronagraphs, slits, etc., but I don't actually know. $\endgroup$
    – pela
    Jan 6, 2022 at 7:32
  • $\begingroup$ @akostadinov yes and no :-) . Both involve analyzing sources of phase error and manipulating a mirror (bendable or segmented) to remove that error. $\endgroup$ Jan 6, 2022 at 13:22

Welcome "akostadinov", and welcome to me, too! Indeed, a great question.. In an attempt to start an answering discussion, understand each of the Webb's 18 primary mirrors have I believe 5 motorized adjustment points (correction - 6 actuator motors!), for just what you're asking! Yes, once everything is locked down and deployed, there will be an alignment/collimation period to get all the optics in the right position, something essential in a multiple mirror primary system. I can imagine over time, these mirrors may get occasional tweaking, but I don't know with what regularity. With the multi-mirror Keck observatory here on Earth, they actually "warp" the primary segments based on what they're observing! Hope this helps! Cheers..twg.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .