The Insane Engineering of James Webb Telescope
This episode of Real Engineering is brought to you by the CuriosityStream & Nebula bundle, now just 11.59 for an entire year until December 24th. “I don’t think it’s very useful to speculate on what god might or might not be able to do, rather we should examine what he actually does with the universe we live in, all our observations suggest that it operates according to well defined laws. These laws may have been ordained by god, but it seems he does not intervene in the universe to break the laws, at least once he set the universe going” In the cauldron of the early universe no light could escape the dense opach fog of primordial gas, but as this cosmic soup of atomic particles began to cool down hydrogen atoms began to form, leading to the universe’s first bright, violent new starts burning through the fog that once blocked all light from escaping the expanding universe. Some of these early photons have travelled unhindered through the vast empty expanse of space for 13.5 billion years, and will reach their destination, here on the man-made detectors of the James Webb telescope. A space odyssey coming to an end because of the curiosity of humans. The James Webb telescope is going to give us our first detailed glimpse of this early universe from which we and everything we know was born.  The James Webb telescope is a 10 billion dollar endeavor. An endeavor that has eaten into NASA’s limited budget, consuming one quarter of NASA’s entire astronomy budget for years, and in the early hours of a tenuous launch date of December 24th, this 10 billion dollar gamble will launch aboard the Ariane 5 rocket, a European heavy lift launch vehicle, from the European SpacePort in Kourou, French Guiana  Astronomers, physicists, and enthusiasts alike will look on with nervous excitement as this rocket carries the next generation in human curiosity. This is the insane engineering of the James Webb Telescope. The combination of technologies required to make the James Webb telescope possible are unique to this time period in human history. The launch vehicle, the image processing, the electromechanical systems, the cooling systems, the mirror, and the sun shield. This endeavor is the culmination of not just the decades of work from the engineers and scientists at NASA, but thousands of years of work of our ancestors. The materials and engineering required to peer back 13.5 billion years into the reionization epoch are a punctuation point in human history, that we, the human race, should be celebrating and watching with bated breath together. The launch will take place here, in French Guiana, a space port ideally located on the Earth’s equator to give the James Webb Telescope an extra push towards its final destination. The James Webb Telescope will not be in orbit around Earth like Hubble, it will be launching to a destination 1.5 million kilometres from Earth, Lagrange point 2. Lagrange points are special points in space where small objects, like satellites, can stay more or less in the same position relative to the gravitational bodies that they are traveling with. This happens because the gravitational pull from two bodies precisely equals the centripetal force required for the object to move with the gravitational bodies. Like little parking spots in space that allow satellites to sit in a relatively stable position while using a minimal amount of fuel to stay there. There are 5 lagrange points between the Sun and Earth. L1 lies between the Sun and Earth. It’s extremely useful for Sun observation satellites. However, the nature of the James Webb telescope’s job wants it to avoid the light from the Sun as much as possible. It is an infrared telescope, infrared is heat, and the heat emanating from the Sun would completely saturate it’s sensors and make observing the distant past impossible. So, it will be launching to L2, located about here. Here the telescope can turn it’s back to the Sun, Earth and Moon, which will stay in the same position, nicely lined up behind the telescope thanks to Lagrange point 2’s unique physics. In order to operate correctly the dark side of the telescope needs to operate at minus 233 degrees celsius (-388 fahrenheit). Without a way to block out the heat from the Sun and Earth the telescope would be scorched at 83 degrees celsius, nearly hot enough to boil water.  This is a huge amount of heat to block and to do this the James Webb Telescope will carry a massive shield on it’s back, like a tortoise. “And making such a device, is a very very tough problem.” That’s Mike Menzel, Missions Systems Engineer for the James Webb Telescope. “We had to map every heat flow to make sure that we do not let any leak through from the hot side to the cold side. to make sure that that sunlight, which is dumping approximately 200,000 watts of power in our direction – we only want less than a watt of that to get through to the telescope and passively cools the telescope. “ Preventing that heat transfer is, as Mike said, a very tough problem. Heat can transfer in 3 ways. Conduction where heat is transferred from atom to atom in direct contact with each other, like heat travelling down a copper pipe. Convection, where heat is transferred from the physical movement of atoms, and radiation heat is transferred by electromagnetic waves. In the vacuum of space convection isn’t a concern. So that leaves conduction and radiation as methods for heat transfer, let’s see how the James Webb Telescope is managing these. First, material choice. The sunshield needs to be light, strong, resistant to degradation from solar radiation ,dimensionally stable across a range of temperatures, and reflective. That’s a long shopping list of requirements, and Kapton, a type of high performance plastic, manages to check all the boxes. Each layer of the kapton sunshield is incredibly thin. Layer 1, the layer closest to the sun is the thickest at just 0.05 millimeters, while the next 4 layers are just 0.025 millimeters thick.  Kapton by itself is actually mostly transparent, which isn’t a fantastic trait for a sun blocking heat shield. Thankfully, the wonder material that is kapton can also be easily coated in other materials. Each layer is coated in a 100 nanometer thick coating of aluminium, giving the sunshield its reflective appearance. This reflective quality helps prevent heat transfer through radiation, by simply bouncing that radiation back to space, and with gaps between each layer, the heat that is absorbed can’t easily transfer through conduction or convection, taking advantage of the highly insulating vacuum of space between each layer. Heat could still transfer between each layer through radiation. The outermost layer will gain heat and start glowing with infrared radiation, just as we see through an infrared camera. In order to prevent this the sunshield has some clever engineering designs. The layers are angled relative to each other to ensure the reflected radiation between each layer is funneled outwards to space. Ensuring that each layer gradually reduces the temperature as it gets closer to the critical components in the instrument bay.  The layers gradually get smaller in area from layer 1 through 5, ensuring the mirror only has a direct line of sight with the coldest layer at all times. Layer 1 itself is also coated in a special silicon coating 50 nanometers thick, giving it this pink appearance.  Silicon was used because it has high emissivity. Simply meaning it emits a lot of the energy it absorbs out as thermal radiation. Meaning, the material will not hold onto its heat, which would give it time to conduct through the structure of the spacecraft to areas we want cool. This high emissivity silicon coating is applied to layer 1 and 2, the two hottest layers, helping them send their heat back out to space, away from the spacecraft, as fast as possible. These design choices are what allow the heat shield to maintain the massive heat differential between the hot and cold side, but blocking heat is just one challenge. “That’s one of the bigger challenges, along with just designing a deployment system that does this complicated and necessary unfolding- reliably and correctly. “ In order to fit into the fairing of the Ariane 5 rocket, the sunshield has to be folded and stowed before launch, leading to some incredibly complicated mechanics to ensure it unfolds correctly when gametime arrives. “deploying things in space is always difficult. But when you’re deploying a rigid structure – that’s generally what engineers call deterministic… that’s relatively easy….Membranes and cables are almost inherently non-deterministic. And if you want to have or illustrate what that means – try pushing on a string. The string will move. If i ask you to determine the shape that it will assume you will have a very very hard time doing it. So to control these almost non-deterministic things takes a great deal of effort, takes a great deal of trial and error. And even after we’re done getting the design right, the one thing about the sunshield is it’s almost like a parachute or similar to a parachute. You know the parachute will work, but it’s also only as good as the very very last time you fold it. And you’re going to find out whether you folded it correctly or not when you use it. “ The unfolding process will begin a few days after launch, not too far from Earth. Starting with relatively simple mechanisms with the solar panels and communications antenna deploying. The truly nerve wrecking process begins on Day 7, as the satellite is coasting towards L2. There are over 300 single points of failure in this unfolding sequence. 300 chances for a 10 billion dollar, 25 year project to end. 107 pins,  holding the sunshield together, have to be released on queue, to allow the system of pulleys, motors, cables, bearings and springs to begin unfurling the sunshield into its precise complete shape. This process will take 3 days, and once complete the optical components will unfold and lock into place. Completing the transformation process, but we are most certainly not in the clear. The likelihood of the tennis court sized sunshield being struck by micrometeorites is fairly high, and because this a thin layer of plastic stretched out under tension, a small tear caused by an impact could cause a runaway tear ripping through the whole sunshield. To prevent this, rip stop seams have been molded into the sunshield, which will arrest tears and keep them confined to a single portion of the shield without compromising structural integrity. The film has also been carefully moulded with corrugations and other shapes to stiffen and shape the shield as needed.  This passive cooling system helps tremendously, ensuring the dark side of the telescope is shielded from the sun’s heat keeping it’s sensitive heat detecting instruments at 40 degree Kelvin, about -233 degrees celsius. But parts of the telescope, specifically the mid-infrared detection instrument, located here, needs to be even colder to work correctly. It needs to be 7 degrees Kelvin, just 7 degrees off the absolute minimum temperature of the universe of zero degrees Kelvin, and for this we need active cooling.  The James Webb telescope includes an innovative cryocooler for this purpose. The challenge in developing this cryocooler alone was immense, costing 150 million dollars. Getting cold temperatures is just one small part of the design. Vibration has to be eliminated, as the tiniest movement at the telescope would cause massive blurs in the image as it attempts to focus on objects billions of lightyears away. That means eliminating moving parts where possible, and when that can’t be done incredibly precise machining and movement is needed to balance weights as they move. The cooler also needs to use a tiny amount of electricity, as the telescope only has 2000 watts of power provided by it’s solar array, and it needs to run reliably for years. That means a closed loop cycle, with our refrigerant being continually reused. I found this explanation of the cryocooler on NASA’s site. The precooler features a two-cylinder horizontally-opposed pump and cools helium gas using pulse tubes, which exchange heat with a regenerator acoustically. Okay, horizontally opposed pumps, with carefully balanced pistons that will cut vibrations as the weights balance each other out, but the rest of that explanation sounds like it came straight out of a sci-fi novel. A sound wave is just a pressure wave and pressure and temperature are directly proportional. Higher pressure will cause higher temperature. One way we can take advantage of this is by creating a standing wave, where the peaks and troughs of the wave are stationary. We can do this in a closed tube where the resonant frequency of the tube is determined by the tube’s length. Here the sound wave will bounce off the closed end and create a region of compression and high pressure, and therefore high temperature. This alone isn’t terribly useful. The energy and temperature in this system will stay relatively stable, left on its own, but what if we could extract some of this heat with each cycle? Then, on each cycle, we could gradually cool the overall system. To do this, we need a way to pass energy out of the system, This is done with a stack, a porous material with air gaps that allow sound to pass through it, which is placed so that it smoothly spans both the hot region at the end of the tube and the cold region in the centre, like this.  A heat exchanger is then placed on either end of the stack, one for the hot side and one for the cold. The hot heat exchange will conduct its heat to the centre of the sunshield, where it can radiate out to space, while the cold portion will conduct its heat, or lack thereof, to a copper plate attached to the back of the infrared sensors to cool them to 6.2 degrees kelvin.  This is an extreme over simplification of the actual operation of the pulse tube cryocooler. This is just a basic explanation of the physical phenomenon that allows it to work. The pulse tube cryocooler is quite possibly the most fascinating part of this spacecraft to me, utilizing a simple physical phenomenon with extreme precision. Allowing those infrared sensors, located in the centre of the telescope’s beautiful golden mirror to work. The golden mirrors are the most striking part of the telescope. Made of 18 hexagonal segments 6.5 metres in diameter. So, what’s the deal with design? It’s unlike any telescope mirror I have ever seen. The mirror surface itself is beryllium plated in gold. That’s a unique and expensive material choice. We need the structure of these mirrors to remain in an extremely precise shape to reflect light as desired. They can’t bend and they can’t warp with temperature changes, and they also need to be extremely lightweight to reduce launch costs. Beryllium is a lightweight metal, with an atomic weight of just 4 it’s much lighter than silica glass, a more traditional mirror subsurface material, while being far more capable in dealing with the cryogenic temperatures the mirror will operate in. Keeping its shape and not contracting so much that it ruins the carefully shaped curves of the mirror. While nowhere near as strong as steel, beryllium is much stiffer with a young’s modulus of 300 Gigapascals . This means, while the beryllium is easier to break than steel, it’s harder to deform before it actually breaks. Giving it excellent dimensional stability. On a pound for pound basis, beryllium is 6 times stiffer than steel.  Making it the perfect subsurface material for a mirror. However, it is not reflective, and for that we turn to gold. Gold is not the best reflector of visible light, being particularly poor reflecting the lower frequencies of the visible spectrum, giving it its distinctive golden hue, but critically it is an excellent reflector of the infrared spectrum, while being very unreactive, ensuring the mirror surface will not tarnish and lose its shine during its operation.  To reflect that light a very thin coat, just 0.1 micron in thickness, is coated over the polished beryllium subsurface. Taking just 48.2 grams of gold, about the same weight as a gold ball. A surprisingly small amount for the huge mirror, which has a collecting area of about 25 m2, 5.5 times larger than Hubble’s 4.5 metre squared circular glass mirror.  The mirror needs to be massive, and to explain why, I asked Mike Menzel. So, Mike, why is this mirror so big? “Well I can tell you it collecting, first we’re looking for uh, stars or stellar objects that are approximately going to be a nanojansky. And to explain what a nanojansky is, its units of brightness, very very dim. “ Okay gotcha, could you put that in practical terms? “If I were to put a child’s night light, that’s about 5 watts, put it on the surface of the moon and look at it from the Earth, that source would appear to be 20 nanojanskies. So we;re looking for objects that are 1/20th as bright as that. To do that, you need a big telescope. Picture light as rain coming in, if you want to collect a lot of rain you take a big wide bucket. Well even at the size of our bucket, 6 meters across, we’re only collecting about 1 photon per second. 1 particle of light per second. And to put that into perspective, i;ll go out tonight or any night and look at the brightest star there is in the sky. Your eye is probably collecting 1 million photons per second from that star. So to see these very dim things, the dimmest things there are to see in the universe you need a light bucket that’s at least 6 meters in diameter. “ 1 photon per second really puts things into perspective. Mike and the rest of the team working on the James Webb Telescope actually wanted the mirror to be bigger, but the cost of launching a mirror that size, between the increase in weight and limited space available inside the Ariane 5 fairing, was not cost effective. They maximized the size with the resources available, and incredibly, even though the mirrors collecting surface is 5.5 times larger than hubbles, the James Webb mirror is 62% lighter than Hubble’s massive solid glass mirror. That is an astounding weight saving, driven by launch weight requirements to get the telescope to L2. And the mirror is even programmable. When Hubble first began transmitting images back to earth it became clear that there was something wrong with the telescope’s optics. Instead of the crisp awe inspiring we are familiar with today, the early images came back blurred. The mirror had been ground down too flat, by a mere 2000 nanometers 1/50th the thickness of a human hair, but that was enough to cause the light to be focused incorrectly on the telescope’s sensors. Replacing the mirror was not an option, but Hubble was designed to be serviced throughout its lifetime, featuring modular equipment bays that allowed older equipment to be removed and replaced. In order to correct the issue, corrective optics were installed into one of these equipment bays, like a giant pair of glasses for the 1.5 billion dollar telescope. James Webb will not be serviceable. It’s simply too far away from earth, beyond the range of any space vehicles capable of carrying humans to service it. If there was a problem with the mirrors, that would be game over, but the engineers were not taking any chances this time, and have engineered a system capable of adjusting it’s focus by itself. Each of the 18 separate mirrors can contort its shape and adjust its position relative to the secondary mirror located in the main mirror’s focal point. The weight saving isogrid rear side of the beryllium mirrors are assembled with a system of back plates, struts and motors that can not only adjust the mirrors rotation, but with the centre motor and these struts, the mirrors can actually change their curvature to adjust the focal point of the mirrors, a feature that could have corrected Hubble’s issues from Earth. Once fully deployed the telescope will begin it’s calibration phase, with each mirror adjusting itself until each of the 18 segments have aligned correctly with the secondary mirror, a 0.74 metre convex mirror, which itself has 6 motors to adjust its position. These motors and control systems are so precise that the mirrors can adjust their position in steps on the scale of wavelengths of light, creeping closer to alignment by increments 1/10000th the size of a human hair. That is an astoundingly accurate electromechanical system. The engineers of the James Webb telescope performed this calibration test here on earth with an absolutely massive vacuum chamber that can be cooled to the same temperature that the telescope will operate at, ensuring proper focus can be achieved. But the job to get a clear image isn’t done with primary and secondary mirror alignment. They focus the light onto the cassegrain focus, which is located inside the aft optics subsystem. This black protrusion in the middle of the primary mirror, which blocks stray light from entering the aperture.  In the darkness there are two more mirrors, one of them being the fine steering mirror, and this thing is the world’s most expensive image stabilisation tool. It is controlled by the fine guiding system. The fine guiding system is locked onto a guide star and it’s job is to keep that star in the centre of its field of view. Every 64 milliseconds  the fine guiding system will send signals to the attitude control system to make adjustments to ensure the telescope stays on target. This attitude control is done with a combination of 6 reaction wheels, located inside the spacecraft bus, below the heat shield, and with the fine steering mirror. This mirror will constantly be adjusting itself to ensure the target of the telescope stays steady on the sensors, minimizing blur. The telescope also has thrusters for larger position maintenance.  191 litre (42 gallons) of hydrazine and 95.5 litres (21 gallons) of it’s oxidizer dinitrogen tetroxide will be stored Inside the spacecraft bus that will feed 20 different rocket thrusters scattered around the telescope. There are 8 thruster modules, 2 on each corner of the spacecraft bus, to aid the reaction wheels in spinning the telescope to point towards points of interest. These 16 engines will be fed with hydrazine only, a monopropellant reaction where the hydrazine is passed over a catalyst, causing a highly exothermic reaction, breaking the hydrazine down into nitrogen, hydrogen and ammonia. The other 4 motors are for orbital and positional control, and require more power. They will be fed with both hydrazine and dinitrogen tetroxide. This fuel and oxidizer mixture react hypergolically to form nitrogen and water. Hypergolic meaning they do not need an igniter, they simply ignite on contact with each other. Hydrazine is an excellent choice for a long lasting mission like this. The hypergolic reaction means the motors can repeatedly and reliably fire without a point of failure causing issues, like an ignitor breaking. Hydrazine is also stable for long periods at room temperature. Allowing it to be stored over the expected 10 year life cycle of the James Webb telescope. Unfortunately that life cycle is limited to 10 years precisely because of the fuel. Between pointing and orbital maintenance, we will run out of fuel at the same point and we currently has no way of refueling the telescope, but rumors are, behind the scenes, NASA is looking to develop the technologies required to refuel the James Webb telescope before it’s demise 10 years from now. Robots capable of refueling spacecraft far from earth is an exciting concept. The James Webb Telescope could end up teaching us many more fascinating things, beyond the early stages of the universe. It’s my hope as an engineer, after being 25 years on this job, that eventually telescopes, the really really big ones of the future, will be built in space. Testing James Webb – a telescope that’s designed to work in space, has been a very difficult thing to do on the ground. And I’m hoping that someday we’ll be building these things in space, testing them in space, tweaking them in space, and then deploying them in space. We are on the frontier of a new space age, and the James Webb telescope is a milestone on our journey towards being a more capable space faring society. This is just one of many milestones in our brief time as a species capable of escaping our planet’s gravity. From building our incredible global position network and sending satellites to the far reaches of our solar system to visiting the moon and building reusable rockets. It’s never been more exciting to be an aerospace enthusiast, especially when you can get access to the CuriosityStream and Nebula bundle deal for just 1159 a year. This holiday deal gets you access to incredible space documentaries on CuriosityStream for less than a dollar a month, and with yearly subscription you will get full access to our Battle of Britain series launching in the New Year, and our 9 part Logistics of D-Day series that is ready for you to watch right now. This discount will end on the 24th of December, so you have less than a week to avail of this deal. A great stocking filler gift for the documentary fan in your life, while supporting this channel in the most effective way possible. Nebula gives us a place to experiment in safety, without any worries of alienating our current audience on YouTube or being demonetized for covering sensitive topics like war. Your support there has helped this channel evolve and grow, and we are able to give you exclusive content like this as our thanks. Many of you will be happy to hear that we have finally released a Roku app for Nebula, giving you more options on how to enjoy our ad free videos. The best way to support this channel is by signing up to Nebula through the Nebula and CuriosityStream bundle deal for just 11.59 a year. With access to both Nebula, and all the amazing award documentaries on CuriosityStream, like this amazing series called “Trajectory”, that tracks the scientific milestones of space science over time. You can sign up to this amazing deal by clicking on this button on screen right now, or if you are looking for something else to watch right now you could watch our last video on the future of supersonic travel, or you could watch Real Science’s latest video on the insane biology of Orcas.