EPISODE 99: WEBB SPACE TELESCOPE’S FIRST YEAR

Transcript

Mike Menzel: We’ve never, ever deployed anything this big or to this extent on-orbit.

Every time you put a new telescope in space that has an order of magnitude increase in performance, you see things that you didn’t even think to ask.

I really, really hope that Webb sees something totally new, totally unexpected.

Deana Nunley (Host): Welcome back to Small Steps, Giant Leaps, a NASA APPEL Knowledge Services podcast where we tap into project experiences to share best practices, lessons learned, and novel ideas.

I’m Deana Nunley.

The James Webb Space Telescope is the world’s premier space science observatory. The orbiting infrared observatory was launched almost a year ago — December 25, 2021 — and it’s already providing new views of the universe and uncovering secrets that were previously inaccessible.

The Webb Telescope is an engineering marvel with its design pushing the boundaries of space telescope capabilities to solve cosmic mysteries.

Mike Menzel is the NASA Mission Systems Engineer for the Webb Telescope, and he joins us now to discuss highlights of this first year. Mike, thanks for being on the podcast.

Menzel: Oh, thank you for having me.

Host: It’s been a remarkable first year for the James Webb Space Telescope. What was your initial reaction when you saw the first images from Webb?

Menzel: Myself and my colleagues were overwhelmed. I’ve used the word ‘giddy,’ which I would never use for myself or my engineering colleagues, but we were giddy. The deep field that we saw, and we saw these images maybe about a week before, a couple days before they were released. And when I was looking at that deep field image, the gravitational lensing image, I turned to one of the lead astronomers and I asked him, ‘What was the deepest thing in that image? What was the magnitude, the brightness of it?’ When he told me, the hairs on the back of my neck stood up. The next question I asked was, ‘How long did we expose for?’ And he said, ‘12 hours,’ and I, like I say, the hairs on the back of my neck stood up. I was overwhelmed.

See, that image may look like a Hubble image, but that Hubble image that looks similar to it took 14 days. We did that exposure in 12 hours. So when I heard that, I realized that everything that we had strived for for the past 25 years, we’ve achieved. We wanted to build a machine capable of seeing the first galaxies and stars that turned on in the universe, and when I saw that image and he told me it was a 12-hour exposure, I knew that we had done it. So whatever is out there, we’re going to see it.

Host: Incredible. What’s made the biggest impression on you during the first year?

Menzel: Well, my favorite image at first was that deep field image, right? Because I knew that, ‘Hey, we built the First Light Machine. We achieved it.’ But then shortly after that, maybe about a month after that, they took a picture of Neptune, and I was yelling down the hall to my friends saying, ‘You got to check this out. Look at this. Oh, my God.’

See, we’ve known that Neptune has had rings for many, many years, but they’re actually very pathetic when it comes to what they look like. And when I saw that image of the rings of Neptune, I was wild with enthusiasm, and I was calling out to my friends, and we all, all of us engineers took a look at that, and it was all — we had the same kind of wonder that you see when you see Saturn for the first time through a small telescope. And if you’ve ever looked at Saturn, the rings, even in a small telescope, are vivid, they’re overwhelmingly beautiful, and when I saw that, I just went wild.

Host: Let’s talk about the triumphs and the challenges of the project. This was the most challenging and complex on-orbit deployment sequence ever attempted. What are your thoughts as you reflect on the flawless performance on orbit and what it took to make that happen?

Menzel: Well, first, it was not flawless. It may have seemed that way, but that’s a tribute to the team. In particular, the Deployment, Design, And Operations Team. We had a couple of problems, and thanks to the preparation of that team and the design work that they did to that, all the problems we had, which numbered about I’d say two or three different deployment anomalies, we solved them. We knew what they were, and we solved them and worked around them in a matter of hours. And when we were practicing all the horror scenarios for this, believe me, some of those horror scenarios that we practiced would’ve taken days, but thanks to the preparation of this team, we countered and addressed all the minor issues that we had within, like I say, two or three hours.

We also, for the deployments, the on-orbit deployments come with a cost. And the cost is these things called single-point failures. When you put a lot of mechanisms on board, no matter what, these mechanisms, the release mechanisms, some of the motors and things like that, can never be fully redundant, so they are counted as what we call single-point failures, and we took them very, very, very seriously. We found out how they could fail, why they could fail, and for the most part, the things that would make these release mechanisms fail were in fact the way they would be installed for their very last time.

It’s almost like a parachute. You can practice and design the parachute and practice folding it and folding it and folding it, but it’s always going to be as good as that last fold. And you never know whether you folded it well until you jump out of the plane. Well, that was a situation we were in, and we took these very seriously. We photographed the last installation. We photographed and videoed the way they were handled. We monitored them very closely, and because of that, because of that attention that our reliability engineers and our deployment engineers took on that, that was another reason why this went very smoothly.

And finally, we did concentrate on testing. And one of the things we did, and this was a tribute to Jim Flynn and the Northrop guys, they built a full-scale model of the sunshield. We practiced deploying it, folding it, deploying it, folding it, finding out all the different mistakes we could make doing that on that full-scale model. You see, you couldn’t do all those tests on the real thing because every time you tested and refolded and tested and refolded the flight unit, you were putting more wear and tear on it. So we had to find a balance between testing the flight unit enough and testing it too much. So, the fact that we made this full-scale model of the sunshield that we could beat the hell out of, and we did, was a tribute to the, once again, to the Deployment Team and how seriously they took this.

Host: What was it like for you and the team as the telescope’s primary mirror segments were brought into precise alignment and formed the largest and most sensitive mirror system ever launched to space?

Menzel: Well, that was very exciting. And once again, that had its anomalies, but they were relatively minor. I know that Lee Feinberg and the OTE team was monitoring that very closely. Once again, they’ve practiced it over and over again on a testbed that was built out at Ball Aerospace. And I remember we were just through some of the early steps in our OTE alignment, and Lee called me into a room, and we were looking at the images, very early images, images that we never expected to be very good. And Lee turned to me, he says, ‘Mike, this thing’s going to work great.’

Very early in the alignment process, the images were indicating that those mirrors really didn’t move that much during launch, and that we could put them and place them back in almost the exact position we wanted to, even before we really started to focus it. There were one or two moments of concern, but they went away, they went away very quickly. There was one point where there was one segment that was not exactly what they had expected, and after a night of pondering, Lee and the team came back and said, ‘No, no, no, that’s a software glitch, look.’ We fixed that up, and bingo, even that one mirror was working really, really good. So, it was very exciting, and actually, I actually think that the alignment of the telescope was more smooth than some of the deployments we had.

Host: Really? From your perspective as the lead mission systems engineer, what were the toughest challenges of this project?

Menzel: Well, obviously the deployments, right? We’ve never, ever deployed anything this big or to this extent on orbit, and it would be fair to say that we literally rebuilt and realigned the observatory on orbit. That’s never been done before to this extent.

And the next thing is the cryogenic design. Three metric tons of hardware, over 3,000 kilograms had to be cooled down to 55 degrees Kelvin or less. And that, you can’t build a refrigerator to do that, so we had to do that passively. We had to build that big sunshield that shadowed the telescope and the instruments, and they had to cool down to those temperatures. Now, if you want to cool down to something that’s around 55 degrees above absolute zero, you had to, we had to keep careful track of every watt of heat, every way a watt or even milliwatts could leak from the hot side of the observatory to the cold side.

The next thing was we had to be passively stable, meaning as our observatory moves around the sky and looks at one point of the sky and then moves to another, we want to make sure that temperatures didn’t change so much that we had to refocus every time we moved the telescope. We call that passive stability. That was a bit of a challenge for us, but in the end, we did something where the observatory is passively stable.

And finally, the fourth big challenge for me was to verify the observatory, primarily by analysis. Now, verification to an engineer, to a spacecraft engineer, is the process where you prove you’re flightworthy. You prove, ‘Hey, what we built is capable of fulfilling the mission and is ready to launch.’ So, most of this observatory, as an observatory, is not testable. We couldn’t test it, so we had to prove that the observatory worked by doing tests at a lower level of assembly, like the test on the telescope, the test on the spacecraft, a separate test on the sunshield, and build analytic models based on those separate tests, put the analytic models together to form an observatory model, and then predict our performance on-orbit by using that observatory model, by doing it analytically.

That’s a tough job for a lot of us who are used to, who like testing, because one, you have to consider how much margin do you need for your performance? In other words, how off or how accurate can these models be? And finally, the models themselves do not consider workmanship. If somebody doesn’t tighten a bolt right, or something like that, the math model doesn’t know any better. It assumes that everything’s put together the way you said it would be put together. So, to complement this analysis process, we had to do fundamental tests just to make sure that everything was put together correctly, that the workmanship was correct. And that verification process was pretty long and pretty convoluted and had some risks to it. So those are the four things that challenged, from a system engineering perspective, challenged me. The deployments, the cryogenic design, passive stability, and the verification.

Host: Could you explain how the science goals drove the engineering?

Menzel: Oh, well, the engineering was definitely driven by the science goals. First, we’re building a telescope that wants to see the first light in the universe, the first stars and galaxies that turned on. By definition, they’re the farthest things there are to see, and therefore they’re some of the dimmest. They’re of the order of a nanojansky, which is 10 to the minus 35 watts per meter squared per hertz. And if you want to know what that looks like, if you want to understand what that is, take a child’s nightlight, puts out about five watts, put it on the Moon and look at it from the Earth. That would appear to be 20 nanojanskys. So, we’re looking at things fainter than that. So if you’re going to see anything that faint, well, you’re going to have to build a big telescope, six meters, at least six meters.

But that’s only half the problem. The other half is that you don’t want these faint objects to be overwhelmed by stray light. So, a lot of our design, a lot of our attention to this design was driven by don’t let stray light overwhelm these dim things that you’re looking for. And because of that, one of the chief contributors to stray light would be the glow of the hardware itself. So our telescope had to be 55 degrees Kelvin or less to keep it from glowing brighter than the very stars we’re looking at, and building anything that’s three metric tons and cryogenic, that is not a, that’s a challenge. In fact, one of the obvious things in our telescope that people point out is we don’t have a barrel around it. Most telescopes that go into orbit have a barrel around it. Well, we couldn’t have a barrel, because the very barrel would act as a blanket that would keep us from cooling down to those temperatures.

The other thing we also, as we slew around the skies, we look from one star to the other star to another star, we don’t want our image quality changing. We don’t want to have to refocus the telescope every time we moved it. So that’s what I alluded to earlier, we had to make it passively stable. And what I mean by passively stable is we didn’t want to put control algorithms on board to readjust the focus every time we turned. Future telescopes, by the way, are probably going to need that, but we wanted to get away without it.

And finally, the data system by itself was a bit of a challenge. These pretty images that we take are all corrupted by things like cosmic rays and things like that, so we had to give some attention to our data system, and in the end, we decided that we would much rather do our data correction on the ground rather than onboard, but we gave that some serious thought. Those are the ways the science really drove this. We were called the First Light Machine, and we designed a machine that was sensitive enough, radiometrically sensitive enough to see these very, very faint objects.

Host: After all the hard work on Webb, how does it feel to actually deliver on these ambitious science goals?

Menzel: Oh, it feels great. However, to be fair, we’re still waiting on the final delivery. Part of our delivery, part of our engineering requirements, was to last for five years, and to get over 1,000 galactic spectra, and things like that, so we’re still working on the final delivery, but our initial performance on orbit, I tell people, is two times better than we had spec’d.

And for those that are a little interested in optics, what I mean by that is we are supposed to be diffraction-limited at a wavelength of two microns, meaning at a wavelength of two microns, we couldn’t do any better. We can’t make a better image. Well, we’re diffraction-limited at one micron, which is even harder. What it’s saying is, our telescope is working twice as good as we expected. And because of that, even the radiometric sensitivity that we’re seeing, the time that we have to expose to see these faint objects is much less, and we’re doing much better, so I have every confidence that we’re going to make the five years right now.

And in terms of some of the resources on board, I think some of your listeners may have heard that we put enough fuel on board for even under the worst situation, to last for 10 years. Well, because of the great launch that Ariane gave us, and because our Flight Dynamics Team executed our mid-course corrections right on time, we have well over 20 years’ worth of fuel on board. So we not only have great scientific performance margin, but even if something goes wrong, we have enough time. We have about 20 years’ worth of fuel life, so if we have to make up for something by exposing longer, we can do it. And right now, just to sum up, as of right now, it feels great, and we’re well on our way to delivering at least five years’ worth of science data, and hopefully much more, hopefully more than 10 years of science data.

Host: What are you looking forward to in the Webb Telescope’s second year and beyond?

Menzel: Well, let’s see. As a kind of amateur astronomer, my degrees are in physics, so I wanted to be part of a team that would build a telescope that would detect the first stars and galaxies, so I want to actually be part of the detection of the first light, first stars and galaxies in our universe.

Next, I’d like to be able to say that I was part of a team that discovered the first biomarkers on an exoplanet. An exoplanet is a planet that orbits another star, and we know of about, right now, 5,000 of these exoplanets, and we’re well on our way to identifying in the spectra compounds or elements that are either produced by life or necessary for life. And if you see these biomarkers like water, carbon dioxide, methane, atomic oxygen, in the right proportions, that says, ‘Hey, this is a good candidate that there’s life here,’ so that would be the second thing.

And the third thing is I really, really hope that Webb sees something totally new, totally unexpected. And every time you put a new telescope in space that has an order of magnitude increase in performance, you see things that you didn’t even think to ask, and I’d really love to say that I worked on a telescope that discovered this new phenomenon. Those are the astronomical things.

As an engineer, I’m really interested in the relatively and hopefully mundane monitoring of how the observatory performs over its life, how it trends. I’d like to be able to say to the next generation of engineers, ‘Hey, here’s how much margin you really need. If you do it right, here’s how long it should last. Here’s what you can expect of the next generation of telescopes if you design it the way we designed this.’ So, I’m hoping that that is relatively uneventful and mundane. I don’t like a lot of excitement in my life in that department.

Host: This has been so much fun, Mike. Thank you so much for joining us on this podcast.

Menzel: My pleasure.

Host: Do you have any closing thoughts?

Menzel: I guess my closing thought is I consider myself to be extremely lucky to be part of this team and to work with the engineers and the scientists that I did over these 25 years. I’ve really lucked out. This is the job I’ve probably wanted since I was six years old, and I just consider myself extremely lucky to have been a part of it.

Host: Mike’s bio is available on our website at appel.nasa.gov/podcast along with links to topics discussed during our conversation and a show transcript.

The Small Steps, Giant Leaps podcast is on holiday break the next few weeks and returns with the first episode of the new year Wednesday, January 25.

Before we close for the year, I want to express appreciation to members of our podcast team: Steve Angelillo, Masha Berger and Kevin Wilcox. Thanks for everything you do to keep the podcast moving forward.

And Steve, Masha, Kevin, and I want to thank you for listening and being part of another successful year for the show. We appreciate you taking time to engage with the podcast, offering positive comments along with suggestions for guests and topics, and sharing the podcast with your friends and your colleagues.

On behalf of the APPEL Knowledge Services team, we wish you Happy Holidays and again thank you for listening to Small Steps, Giant Leaps.