I am tired. It is 4 am California time and I have been awake since early the previous morning, having just arrived in the US from Europe. I am jet-lagged. I am sitting in an old but comfy business-class seat close to the nose of a Boeing 747SP. The SP version of the venerable Jumbo Jet has never been widely used and is even rarer today; thirteen are still in active use at the time of this writing. But the one I am flying on is unique: it is the SOFIA airborne observatory. We are on a ten-hour flight across the US to observe the night sky, and I am on board to record interviews for my omega tau podcast. I have been recording on and off for several hours, and now I'm taking a short break, drinking Coke Zero to try to stay awake for the next few hours. Our landing in Palmdale is scheduled for around 7 am. SOFIA missions are usually planned to return before daylight, because the rising sun might damage the delicate optics if the telescope's door fails to close because of a technical malfunction. There's a plan B -- flying a track that points the telescope, located on the left side of the fuselage, away from the sun. But why take a risk, better to land before sunrise. If the mission requirements dictate landing after sunrise, the mission planners have a software program that tells them which ground heading is safe for the telescope based on the date and time. And which (alternate) runways are available to land with a suitable heading.

Chapter Contents

• The Making of SOFIA
• Telescope Architecture [Preview ]
• Controlling the Telescope
• Flying SOFIA
• Instruments
• Science
• A SOFIA Mission

I don't like being hungry, and I suspect Kyle here feels the same way. We've been standing in front of a deserted terminal building in the harbor of Limassol, Cyprus, for close to thirty minutes. We are trying to get into -- or rather, through -- it so that we can board HMS Enterprise, which is berthed in the harbor. For Kyle getting through won't be a problem, since he is an officer in Enterprise and has the paperwork to prove it. I however am a civilian, and the lady we are talking to cannot quite square a civilian with a military ship. Here's the problem: if we have to wait any longer, we'll miss dinner!

It is mid-December 2017 and I have just flown into Larnaca from Munich. Kyle O'Regan, Navigation Officer Two in Enterprise, met me at the airport and we took a taxi to Limassol. I'm scheduled to spend a week in Enterprise on her trip from Cyprus to Malta to record interviews for omega tau. Even though the Enterprise's crew had filed all the paperwork for their civilian visitor with the Cypriot authorities, things didn't go quite as smoothly as expected.

Kyle, confirming my suspicion that he was also hungry, called the ship on the phone to make sure they would put two portions of turkey and fries aside, ensuring we'd get something whenever we arrived. They promised they would, and we were both relieved. About forty-five minutes and several phone calls by the lady later we were allowed to pass through the terminal building. We crossed the jetty towards the ship and walked up the gangway. A friendly, smiling guard with a rifle welcomed us to the ship. We made our way to the mess straight away, where we were promptly handed our dinner.

Afterwards Kyle showed me where I would live for the week: a two-person cabin I would share with him. He assigned the top bunk to me. His clearly showed that he was an experienced sailor, because it sported his personal feather bed instead of the standard issue bedclothes. It looked very comfy!

After making myself at home Kyle took me to the captain's quarters, where Commander Philip Harper, Captain of HMS Enterprise, welcomed me to the ship. Phil has been in the Royal Navy for 27 years and has served on everything from submarines to both large and small surface vessels. He's the first non-hydrographer to command a Royal Navy hydrographic ship in 220 years. He was kind enough to invite me to the Enterprise to record a podcast episode after an amusing Twitter exchange initiated by a mutual friend, Rainer Kresken. It was great to finally meet him in person.

I had arrived in Enterprise, and I was massively looking forward to the week!

Chapter Contents

• On the Bridge
• Surveying Equipment
• How Multibeam Sonars works
• The Ship's Systems [Preview ]
• Life on the Ship
• Safety and Exercises
• Driving into Valetta

Ahead of me I can see a beautiful cloud street. I'm flying along the street, at 2.600 meters altitude, with almost 200 km/h, racing “down” the Schwäbische Alb, a low mountain range in the South of Germany. Making very good progress towards my first turn point Titisee-Neustadt on today's cross-country flight, I can see the Black Forrest from almost 100 km away. Visibility is exceptionally good today. On this day in June, the thermals are so powerful, and perfectly stringed along the cloud street, there is no need to circle or even just to slow down to maintain altitude. I am busy, even slightly stressed, looking out the cockpit to see and avoid all those small and hard-to-see oncoming gliders. Everybody is flying the Alb today because of these perfect conditions created by a convergence line. My FLARM, a collision warning system based on peer-to-peer communication between the gliders, is beeping regularly to warn me of gliders I may not be able to see yet. A small display indicates the bearing and elevation of the most “dangerous” ones.

The cockpit is a weird place. It is a pilot's perch, his outlook onto the 100s of square kilometers that serve as a glider pilot's playing field. The view is magnificent, at least when atmospheric humidity is limited and visibility is good, like today. On the other hand it's rather tight: the cockpit walls on the left and right are a few centimeters from my body and the wings extend right out from my shoulders, figuratively. My head is two centimeters below the canopy, my feet, down from my knees, are stuck below the instrument panel. It's almost like a cocoon. You don't get into a glider, you “put one on”. But to me it doesn't really “feel” tight or claustrophobic because of the vast open space right beyond the acrylic glass that is the canopy. The cockpit is also the command center of your aircraft: the soaring computer presents information about terrain, weather and airspace, the radio connects you to fellow pilots and occasionally, to air traffic control. An acoustic variometer fills the air with the ever-important information about climb vs. descent, which you consume subconsciously. The vario's beeping is mixed with the wind noise which, through its intensity, tells you roughly about airspeed. And of course the cockpit has the stick, rudders, the flap handle and various other levers which you use, often also subconsciously, to control and guide the glider. The cockpit is your resting place and picnic area: while you cannot simply stop to rest, there are phases of the routinely many hours-long cross-country flights when you can rest your mind somewhat. Or to eat an apple, a bun or a cereal bar. Or to pee into a bag. After thousands of hours of gliding, and many hundreds of them in this ASG-29, I really feel safe and sheltered in the cockpit. Even though objectively, it's just this small cabin soaring through the heavens.

Chapter Contents

• Cross-country Gliding
• Glider Physics [Preview ]
• The Future of Gliders
• Wave Flying
• Higher and Faster: the U-2 and SR-71
• Stability and Control
• Flight Control Systems: F-16 and A-320 [Preview ]
• (Flying) Simulators
• Helicopters
• Flying in an F-16

If you read the introduction to this book, you will know that I had a relatively untypical time at school. In particular, the physics teachers let us do lots of cool stuff. At one point we were supposed to measure the speed of light. How do you do that? You let light travel a distance, as long as possible, and measure the time. Obviously. The specific experiment involves a laser, a mirror at the end of a long corridor, and a photodiode to detect the reflected laser, plus a way to measure time. We used an oscilloscope to do that: when the laser was triggered, the beam of the oscilloscope started, and the detection of the laser in the diode showed as a peak on the screen. You needed a very high resolution, because the time would be on the order of 0.3 microseconds. The school had one oscilloscope that could do this. And we were allowed to use it. Not a very good idea as it turned out. We cleverly plugged the power chord into an outlet that was right outside the door of a classroom. And it was the slot right before lunch. In order to be at the front of the lunch queue and maximise the usable time during lunch break, everybody sprinted through the school building to the cafeteria when the bell rang. It was actually dangerous to be in the corridors at that time. And a tiny little power chord wouldn't stop anybody. And it didn't. But the oscilloscope was on the ground. And then sent for repair. My “reputation” among the teachers took a hit that day. Also, we didn't get to measuring the speed of light. But no worries, we knew it anyway: c = 299,792,458 m/s in vacuum, and for our purposes also in air.

Chapter Contents

• Special and General relativity
• Detector Basics [Preview ]
• Passive Seismic Damping
• Vacuum
• Thermal Noise and the Q Factor
• Active Damping
• Shot Noise
• Squeezed Laser Light
• Power Recycling Cavities
• Control and Detection
• Five Sigma
• Why Do We Care?
• Future Detectors

The Large Binocular Telescope is at 3.200 meters, on Mt. Graham, a three-hour drive from Tucson, Arizona. Once in the enclosure, the dimensions of the telescope became clear: two giant 8.4 m mirrors mounted next to each other on a common frame. Everything is extremely rigid to avoid bending and vibrations as the telescope moves. Speaking of moving: when it moves, you hear nothing and you feel no rumbling. The telescope is mounted on oil films in azimuth and elevation and is so well balanced that it can be moved with 3 hp electrical motors!

Every large machine or scientific instrument I've visited has this magical moment when you are awestruck by its sheer impressiveness. With SOFIA, the moment came when I first entered the hangar, walking up close to the beautifully polished 747. I literally had to suppress a tear or two, not least because I had been looking forward to this moment for a long time -- - the trip had been in the planning phase for over six months. The opening of the LBT enclosure was another such moment.

It is completely dark in the enclosure (important for calibrating the telescope), only a few dim red lights indicate the status of some some kind of electronic device. We used small flashlights to find our way down to the base of the telescope and forward towards the big sliding doors. It was quite cold, almost freezing. A slight rumble. We are standing in complete darkness, and as I look up, the doors move sideways, revealing a clear, starry night sky like I have never seen before. Turning around, I can see the sky reflected in the two 8.4 meter mirrors. Amazing! The the huge telescope start moving above/behind us with absolutely no sound or vibration. The stars reflected on the mirrors move slowly, and so does the dark shape of the landscape of the planet below us. The whole event was deeply impressive and and clearly drove home the point that this was another masterpiece of engineering and a “spectacular instrument” for science, as Richard called it. He has been involved for more than a decade, from design through construction, first light and -- at the time of recording the podcast -- validation of all of its features.

• Introduction
• The large Binocular Telescope
• LBT Optics
• LBT Instruments [Preview ]
• Other Optical Interferometers
• The Next Generation: the ELT
• Controlling the ELT [Preview ]
• Radio Astronomy
• The 100 meter dish in Effelsberg
• The ALMA Array
• The Event Horizon Telescope

Let's start with model railways. This might be the first time in a child's life where the word “model” is used. In this context, a child probably understands the term to mean “small”: it's a smaller version of a real railway. And being smaller is often a characterising feature of a model because smaller usually means cheaper and faster to build. Of course there are exceptions: models of atoms or other microscopic phenomena are (very much) bigger than the real thing.

But for a model to be a good one it has to be realistic. As a child I was very annoyed if the proportions of a H0 locomotive such as the Deutsche Bahn BR 103 didn't quite resemble those of the real thing. Apart from being a toy, a model that looks like the real thing can be used to visually illustrate something to people who have never seen the original. That is useful, but not very much.

So why do we build models in the first place? We build them because they help us understand something about the real thing. So what else can we do with a model train? We can drive it into a train station and stop, wait a little bit, continue on to the next train station, stop again, and so on. So we can replicate what the real thing is used for. But only to a degree. For example, the proverbial aliens would probably ask themselves why the train stops periodically; after all, the model railway doesn't include people that enter and exit the train when it stops at a station. So the purpose of a real train is not apparent from the model. Unless you know the real thing, like most children, because then you can use your imagination to fill in the missing details such as entering and exiting passengers. The whole thing makes sense then.

Can we use a model of the Stuttgart train station to come up with a scheme for positioning the switches in a way that guarantees that trains never collide? This is actually a non-trivial problem. Train tracks are divided into block sections, and only one train can be in each block. Signals enforce this rule by requesting trains to stop before any block that is already occupied by another train. Any such red signal must be preceded by an advance signal that tells the driver to expect a red signal, so he can slow down. If the driver ignores a red signal, magnets signal this to sensors on the locomotive and it stops automatically. Sensors on the tracks count axles as they enter and leave blocks to detect carriages that are separated from a train and stay back in a block. Flank protection ensures that no switch from a parallel track leads into a protected block from the side, independent of whether a train is allowed onto that parallel track. An interlocking scheme prevents combinations of switch configurations and signal indications that would be dangerous. It's really a complex system, and using models to figure out configurations that are safe and at the same time maximise utilisation of a given set of tracks is a good idea. So could we do this with our model? Assuming we have a faithful replica (same tracks, switches and signals) and we are only interested in the algorithmic aspect of the switching scheme, then yes, the model could be used.

But do we need a physical model to do this? Not really. We can perform the same task with a pen-and-paper model. Or more realistically, with one expressed in software. Software models are great, because they allow us to tune their fidelity: we can model a switch as a binary variable with two states and an immediate change of position. Alternatively we can model the position of a switch as a real number between zero and one, and when we command a change, we can make it change its position slowly from zero to one, simulating the fact that a real switch takes time to change its position. We can add a random element that prevents a position change when commanded to simulate faults. We can simulate system faults (such as icing) by relating the simulated faults of multiple switches that are closely near each other.

So whenever we use a model to understand or predict aspects of the real world we have to be sure that our model considers that particular aspect, that it is able to make a meaningful statement about what we are interested in. This is the most important caveat when using models. In the rest of this section we'll explore models and their use further.

Chapter Contents

• Mathematical Models
• Applicability
• Numerical Models
• Precision
• Prediction vs. Explanation
• Statistical Models
• Agents and Emergence
• Model and Experiment
• Meta Models and Languages
• Descriptive vs. Prescriptive Models
• Multi-Level Models
• Implementing Models
• Your own Languages
• Other Terms

Here is how I’d start if I were to write a cold war novel. It was a cold and wet fall day, dark clouds in the sky, fog hiding the ground-bound world. I had just landed at the city’s airport after a three hour flight and when I came landside, I was met by my guide for the day. We exited the airport and got into a VW Golf. We left the airport premises, and drove through the city, eventually arriving at a grey office building. We parked the car, entered the building and checked in at the counter. A slightly annoyed-looking clerk checked my passport, verified my name on a list of pre-announced guests and then handed me a visitor pass: “Make sure you wear this all day, and ensure it’s visible all the time.” Of course.

We left the building and got into the car again to begin a confusing drive: out of the office-style parking lot, through more urban areas, into a more rural landscape, from one village to another. In what turned out to be the last of the chain of villages we drove into a housing area, just to leave that area again via a narrow, curving road. That road abruptly ended at a barrier with a guard shack. An official-looking guy peeked into our car and checked out the credentials of my guide as well as my visitor pass. He opened the gate and we drove through. We parked the car a few dozen meters inside the premises, next to a non-descript building. Exiting the car, we walked towards the building. A guy was standing in front of it having a smoke, probably annoyed at not being allowed to smoke inside on this damp fall day. My guide used her pass to open the door, and we entered the building. From the inside it looked just as boring as from the outside: a cross between a warehouse and a factory building, except that it was almost empty apart from a bunch of technical-looking equipment at the western wall and a huge crane across the top.

After talking for a little while with some of her colleagues in the building, we eventually headed towards and then through another door; there we came against what looked like a gridded cage. My guide and I picked up a bunch of equipment that would allow us to go proceed further into the installation: a helmet and a radiation detector. Both were mandatory for everybody. We also removed two tokens from a registry that kept track of how many people were inside at any time. The facility cannot be started up as long as people are inside. The cage-style section of the room had a rotating door that allowed one single person to pass at a time. To activate the door, my guide needed her passport, but a sensor also scanned her retina to ensure she’s really her―a classical two factor authorization, where one factor is based on a token you possess and the other one uses a non-stealable biological marker. Once through the door we entered a relatively big industrial elevator cabin. My guide selected the lowest floor, and we started our trip down towards a secret world of big machines 50 m below the surface. I was looking forward a lot.

This is how my trip to ALICE started, one of the experiments at the Large Hadron Collider in Geneva, Switzerland. Everything in the story above is accurate except that it was a beautiful sunny day in May, I arrived by train and tram instead of by aircraft, and the building that housed the lift and the crane was actually very “descript”: it had a picture of the ALICE experiment painted on the outside, at 1:1 scale. My guide was Despina Hatzifotiadou, a physicist who is responsible for public outreach at ALICE. We were scheduled to record a podcast episode and to visit the experiment; like the whole LHC, ALICE was offline, undergoing a major upgrade in the 2019-2020 timeframe.

As you probably know, the LHC is the world’s largest particle accelerator. It accelerates heavy particles on a 26 km circular racetrack under Switzerland and France and then collides those particles to investigate the subatomic structure of matter. Four large detectors, or experiments, perform the analysis. In this chapter we’ll look at how the LHC works. But before that we’ll recap the structure of matter.

Chapter Contents

• The Standard Model
• Particle Colliders
• CERN and the LHC
• Producing the Beam
• The Superconducting Magnets
• Discovering the Higgs
• The other main experiments
• Whats Are we Looking for today
• Future Colliders