Ask Ethan: How do space telescopes stay focused?

In the era of space telescopes, we’re now used to clear, sharp images where everything is focused pristinely: reaching the theoretical maximum that’s possible based on the wavelength of light that’s begin observed and the diameter of the telescope’s primary mirror. First with Hubble (once the flaw in its primary mirror was corrected) and later with space telescopes such as Spitzer, GALEX, Herschel, and now JWST, the lack of an atmosphere to contend with means that all we need to do is make our optical systems as optimized as possible in the vacuum of space, and we’ll get those ideal images. Of course, that involves:

• keeping that telescope stable,

• keeping it pointed correctly,

• and keeping it focused correctly,

even as the telescope moves, rotates, and deals with external influences like radiation from the Sun.

How, then, do telescopes and telescope operators accomplish this? That’s what Jim Wilson wants to know, as he writes in to ask:

“How do space telescopes focus and stay focused? For example, on an object close by or an object far away, with near and far being relative, or on one object out of billions.”

We normally think about focusing as something we need to do differently for objects at different distances, but for space telescopes, that’s the least of our worries. Here’s why.

Whenever you have a mirror or a lens — two of the most important and essential focusing mechanisms for telescopes — it’s going to have a specific physical shape, and that shape determines a property of the mirror or lens known as the focal length. The focal length is a measure of how quickly that mirror or lens causes that light to be focused, and for a telescope, that means converged, down to a point. For our eyes, which act as our natural lenses, our lens shape changes dependent on whether we’re looking at an object that’s nearby, in which case we need more focusing power to converge that light down to a point on our retinas, or an object that’s far away, in which case we need less focusing power.

In the most extreme case, we can have an object that we treat as being infinitely far away. When we do, then the focal distance, from the mirror/lens to the point where light converges, actually equals the inherent focal length of the mirror/lens exactly. For a human eye, the focal length of our natural lenses is around 2 centimeters, but for a space telescope like JWST, its focal length is huge: at 131.4 meters (431 feet)! Its series of mirrors converges that light very quickly, as the light travels nowhere near that far from when it first strikes the primary mirror to when it arrives in the instruments, but the mirrors always have that same focal length; they never change shape substantially from that fixed configuration.

The reason the mirror shape doesn’t need to change is because even the closest objects that JWST images — things within our Solar System, like Jupiter, for example — are many millions of kilometers away. Jupiter, even at its closest, is still 587 million km away from the JWST, and because the ratio of the distance to the focal length (587 million km-to-0.1314 km, for an apples-to-apples comparison) is so large, we can treat it, as well as every other target JWST can image, as effectively being infinitely far away.

However, it remains a colossal challenge to keep JWST’s mirrors, or the mirrors of any space telescope, exactly and precisely focused when it’s in space. There are many reasons for this, but just imagining one example can be illustrative of all that can happen. Consider a single JWST observation, where a focused telescope acquires and observes a new target:

As the telescope moves, and it needs to move in order to change which target it points at, its internal components move, causing it to undergo vibrations. As its position and orientation change, thermal gradients arise. And as small, minuscule events happen, such as impacts from micrometeoroids, uneven heating, the flow of internal components (such as coolant for the MIRI instrument), and even light traveling through the telescope’s internal instruments and electric currents flowing, can all lead to the mirror segments to changing shape, to the pressures on the actuators behind the mirrors changing, and even to the different segments shifting position relative to one another.

There are entire systems, therefore, aboard space telescopes to keep them focused, to minimize and compensate for these tiny changes, and to prevent the telescope’s field-of-view from drifting over the course of an observation. The goal, remember, is to have all of the telescope’s mirror segments working together to behave as one single, perfect mirror that focuses all of this light — light acting as though it’s coming from infinitely far away — onto the remainder of the optical systems and then, ultimately, into the instruments. Overall, JWST seeks to have its 18 segments to its primary mirror act as one single focusing mirror: to a precision of around 20 nanometers. But that’s the hard part of the problem: keeping the segments of the primary mirror aligned and focused.

When JWST was first launched and deployed, the mirrors were not focused. The way they were initially focused was as follows.

• First, a single target — a guide star — was acquired, and the telescope’s optics were pointed in that direction.

• Next, the various segments were mapped onto each image of that guide star, identifying the segments.

• Then, you align the images, so that you get 18 images — one corresponding to each segment — which corresponds to coarsely aligning the mirrors.

• And after that, you stack the images together: producing one unified image of the field-of-view you’re seeking to learn about.

But that’s not the end of the story; that’s just the first step in focusing your new, state-of-the-art space telescope.

Image stacking is where all the light gets focused onto one point in your detector: an essential step. However, your telescope is still behaving as 18 small telescopes stitched together, not as one giant telescope. The next step is how you unify your various segments together, and that step is known as coarse phasing. Here’s an overview of how that works.

Imagine that you only had two mirrors, and you wanted to stitch the light from them together so that they behaved as one telescope. You’d need to make the two beams of light from the two mirrors overlap: so that the light from the multiple sources undergoes interference. Internal to the telescope, you’d have a wavefront camera: a camera that checks whether the light from the two mirrors aligns, and if the light is out of alignment, then the mirrors are out of alignment. You can then change how your mirrors are pointed, or, for more sensitive corrections, change the pressure on a variety of actuators located on each mirror’s backplane, until the alignment improves.

This was a very important part of JWST’s commissioning process, and that coarse phasing was followed by many iterations of refining the alignment of the mirrors, as JWST had:

• 18 mirrors,

• with six actuators on the backplane of each mirror,

• and an internal wavefront sensor aboard its NIRCam instrument.

At the end of that refinement process — known as fine-phasing — the wavefront errors are now very, very small. While JWST’s mirrors create a near-perfect mirror shape when stitched together, down to precisions of around 20-25 nanometers, the wavefront errors are still somewhat larger than that figure, but are much lower than were initially expected. Whereas the design specifications of JWST sought to reach and maintain wavefront errors at the 150 nanometer level, the achieved wavefront errors are only about 43% as large: at the 65 nanometer level.

Some of that is due to dirt, dust, and other impurities that made it onto the mirror despite all the efforts made to keep JWST as clean as possible. A cleaner-than-ever clean room helped minimize that. Some of that is due to tiny imperfections that appear in the mirrors, such as due to bombardment by micrometeoroids over time. JWST hasn’t gotten hit with very many large micrometeoroids (thankfully) thus far, so the mirrors are still in excellent shape. The net result, of a huge improvement in wavefront error, even over what JWST was designed to achieve, is part of why JWST observations are exceeding expectations as far as the telescope’s optical performance is concerned.

So that’s how the telescope gets focused, at least initially. However, we also noted that many events that occur, including simply the passage of time during routine telescope operation, can cause the telescope to drift out of focus.

For example, even though JWST doesn’t use thrusters to help it point, but rather reaction wheels, those reaction wheels still cause vibrations. The solar array, which provides power to JWST, also causes vibrations, as the array moves over time to maintain maximal power to the telescope. These moving parts are as physically isolated as possible from the telescope’s optics, but they still induce vibration effects, which can alter the positions of the mirrors. As metal, glass, ceramics, and other components experience thermal variations, they expand and/or shrink, leading to distortions.

A combination of software, mechanical mounts, and actuators behind each mirror segment all work together to compensate for these changes, as well as to ensure that the field-of-view doesn’t drift. However, even with all of this, errors will still appear, and if we do nothing about it, they’ll wind up compounding and growing over time.

That’s why, approximately once every 48 hours, the JWST team performs a check of the wavefront error. If necessary, iterations of improved phasing will then be conducted, bringing the errors down to the limits of the current condition of your equipment. In addition to the internal NIRCam camera, which monitors the wavefront error, there’s also an internal small camera that points at the mirror assembly, which allows astronomers to take a picture of JWST’s mirrors and evaluate their condition. In fact, there was an early, fairly significant micrometeoroid impact on one of JWST’s mirrors in the early stages of its operations, and this camera enabled us to find it and quantify the damage.

However, getting observing time on the space telescopes that we have are among the most competitive and sought-after endeavors in all of astronomy, and therefore, there’s tremendous pressure on the telescope operators to spend less time calibrating and focusing these telescopes: so that observers can spend more time observing. After all, time spent going through this iterative phasing process, and compensating for the vibrations and thermal variations induced simply by routine telescope operation, means less time for observers to actually conduct their science.

That’s why it’s so remarkable, as reported by Jennifer Lotz of the Space Telescope Science Institute (STScI) at the American Astronomical Society’s 248th meeting in June of 2026, that since JWST began science operations, a remarkably 85% of the time was actively spend conducting science programs. That mean’s only 15% of the time was spent on all other things combined, including:

• focusing and re-focusing the telescope,

• performing internal diagnostics,

• changing targets, moving, and pointing the telescope,

• finding and locking onto guide stars (which, remarkably, they got down to just ~2% of the total time),

• and adjusting the mirrors using the actuators.

It turns out that by formulating the wavefront problem as a nonlinear optimization problem, the computational resources (computational time, power use, and memory use) needed to determine which deformable mirror commands should be issued: a huge saver of energy and time both.

The more time we save as far as focusing the telescope and keeping it focused goes, the more time we can spend conducting science, learning about the Universe, and bringing astronomers’ research programs to fruition. It also leaves more time open to explore targets of opportunity, which arise when a transient signal unexpectedly arises, and if there’s time available, a telescope like JWST can follow it up and take observations even on short notice.

But one thing we don’t have to worry about with space telescopes — especially space telescopes like JWST, which are only used to observe objects at much greater distances than its focal length — is having to focus our telescope before each new observation based on the target’s distance. Every object, whether within our Solar System, outside of our Solar System within our own galaxy, an extragalactic object, or an object located at the farthest reaches of deep space, requires the same configuration of the telescope: where the mirrors act to focus a target located infinitely far away. It’s just a question of staying on target and not allowing the telescope’s optical configuration to slip away from a state where it’s adequately focused.

But this is an extraordinarily powerful fact: it means that we can image any object in the Universe with JWST — any target that we choose — and any other object that also shows up in that same field of view will also be perfectly focused! Whether you’re seeing a planet, star, galaxy, or something like an ultra-distant lensed object behind a massive galaxy cluster, or even if you’re seeing all of them in the same field-of-view, they all appear in focus, so long as the telescope’s optics are properly aligned and focused. By improving and optimizing how we keep and maintain a telescope’s focus, we can spend more and more time on the endeavor we’re all most excited about when it comes to space telescopes: conducting the science that teaches us what our Universe is like, what’s within it, how it grew up, and what was it like long ago, as far back as we can possibly observe.

Send in your Ask Ethan questions to startswithabang at gmail dot com!

This article Ask Ethan: How do space telescopes stay focused? is featured on Big Think.