July 1999, Cape Canaveral: This isn't my first rocket launch, but somehow seeing your entire career ride on a pillar of flame into the sky is an entirely different experience.
Flying into Orlando, I picked up a rental car and drove across the flat eastern Florida landscape to the coast. To reach the Cape Canaveral and Cocoa Beach communities to the south of the space center, you drive on long causeways across the inlets known as Indian River and Banana River, catching sight of the enormous VAB (Vehicle Assembly Building) on your left. Turning south on route A1A, the main street of the beachfront towns, you pass a low building which looks like a car dealership but is actually Boeing's Shuttle Logistics Depot, where they keep spare parts for spaceships. We checked in to the Holiday Inn in Cocoa Beach, one of the traditional launch hotels, and made our obligatory visit to Ron Jon's Surf Shop for NASA souvenirs and T-shirts.
The trip to the base itself takes about half an hour - everything is really spread out. NASA's John F. Kennedy Space Center is located on Merritt Island, to the north of Cape Canaveral Air Force Station where the Deltas and Atlases fly. I was lucky enough to be going to the press site, closer to the pad than most of the astronomers present managed to get. The site is right next to the VAB, a building designed for the simultaneous assembly of four Saturn moon rockets, and in front of a grassy field leading down to the water between us and pad 39B, where Columbia stood easily visible in the distance about three miles away, illuminated by floodlights. At the water's edge is the famous countdown clock that I'd seen on TV so many times. T minus 3 hours 10 minutes and counting... it's eerie standing out in the open, seeing the Shuttle with your own eyes, and hearing the voices from the control center echoing over the powerful loudspeakers in the background. There's a lot of cryptic information that goes over the speakers during the countdown, and it helps to know a little of the background if you ever watch a launch on NASA TV or in person.
The countdown clock doesn't include planned holds. So, T-25 minutes is actually 75 minutes before launch. There are two standard built in holds just before launch: one at T-20 minutes, for 10 minutes, and one at T-9 minutes, which is usually short but which is very long - 40 minutes - on this flight, dunno why. During the early period of the countdown, you'll hear a lot of `comm checks'. People are referred to by initials. On the Shuttle, there is CDR (``Cee Dee Arr", no relation to Cee Threepio) - commander Eileen Collins, PLT (Pee Ell Tee, the pilot Jeff Ashby), MS1 (Mission Specialist One) Cady Coleman, MS2 Steve Hawley, MS3 Michel Tognini. On the ground, there's OTC (Orbiter Test Conductor), NTD (NASA Test Director), and RSO (Range Safety Officer) among others.
If anything is iffy, the T-9 hold can be extended further. There's also a possibility - e.g. if the weather is bad - that they might add a T-5minute hold. Window is from 1236 am to 0122 am, total of 46 minutes. It moves 4 minutes earlier each day (sidereal rate, since we are trying to keep RA of our orbital plane constant).
The crew access tunnel (`white room') swings back about T-5min. At 2 minutes to launch, the vent hood on top of the orange tank (the ET) swings back, and you see the liquid hydrogen boiling off. At this point the crew are told: `close and lock your visors, initiate O2 flow'.
At T-31 seconds, the launch control computer takes charge and everything is automatic from then on. If there's a problem between T-31s and T-6s, that's an `RSLS hold'; RSLS = Redundant Set Launch Sequencer, = the launch control computer. At T-6s the SSME (main engines) light up. If one of the three doesn't start quickly enough, the computer will order shutdown. That's an `RSLS abort'. Abort is a nasty word; this is the most benign kind, and there have been several, but it's still scary when the engines light and then you don't go.
That's what almost happened this first night - July 20 at 0436 GMT, just after midnight local time. After hours of building tension, the moment of truth is arriving. It looks like we're going to fly! Any moment now...
``T-31: computer has control. T-20... minus eight, minus seven... minus seven... hold, we have an RSLS hold." Our hearts stop.
During the final count, a launch controller saw a sudden spike in the sensor checking for hydrogen leaks near the main engines. We don't like highly flammable materials in the wrong place when we're about to light the blue touch paper... (Ed: This is a Britishism-but you get the point.) The controller overrode the computers and hit the hold button at T minus 7 seconds, a moment before the engines would have ignited. A close call; with an RSLS hold we could try again two days later. A second later, it would have been an abort and we'd have been looking at weeks of delay.
In our day off, I wangle a tour of the Air Force launch base at the Cape with a group of space history buffs. We found the slab of burnt concrete where the first Canaveral launch, a modified World War II V-2, took off fifty years ago, and made a pilgrimage to the abandoned pedestal of pad 34, where the first Apollo crew died in the January 1967 fire, and the pads where Al Shepard's Mercury and the first Explorer satellite were launched. In the distance, we saw the construction rising at pad 37, destined for the new Delta 4 rocket. After many years in the 80s when this part of the Cape was moribund, activity has picked up and commercial launches are flying monthly.
Back to the complex 39 area for our second launch try. Wednesday night should be fine - the USAF weather officer predicts 100 percent chance of good weather. But as we reach liftoff time, there are clouds nearby and we must hold at minus 5 minutes. Then we see lightning near the press site. Managers convince themselves that a longer launch window won't endanger the mission, and we hang on hoping for the weather to clear. But the sky lights up again and again. Ever since a lightning bolt momentarily knocked out the electrical systems on Apollo 12, launching in lightning has been considered just a bit too exciting for comfort. In the end, NASA admits what we had all expected: the launch is off for today - which we thought was our last chance. A deal with the Delta rocket folks buys us just one more day. I agonize, and change my flight, finding a new hotel room to crash in. If you want to see a launch, you have to be prepared to change your plans.
Night falls on Thursday and I'm driving back towards the VAB. Suddenly the KSC cops block off the road... what's going on? A motorcade drives past with flashing lights, surrounding the famous blue astronaut van. Commander Collins and her crew are on the way to the launch pad. OK, I guess I'm happy to wait a few minutes to let them get past, I don't want them to miss their flight...
The routine is familiar now. We know the spokespeople for NASA, for Boeing, and so on; the weather officer on duty is new and highly embarrassed about yesterday's miscall. After two scrubs, it seems a little unreal, and my total lack of sleep isn't helping. Will this time do it, or will we have to slink back to Cambridge with our tails between our legs and try again later in the year? It's a smaller band now watching the launch; Miles O'Brien from CNN, Bill Harwood from CBS, Al Bunner from NASA HQ... The weather's great, but what about hydrogen leaks? T minus 31 seconds. We're down by the water's edge. The insects are feasting on our ankles. The Shuttle looks beautiful in the floodlights. T-6 seconds. Ignition! The main engines on the orbiter ignite, but the Shuttle is still fixed to the pad. If anything goes wrong in the next few seconds, they can order a cutoff. But this time there is no last minute letdown - Columbia and Chandra are heading for space.
At T=0, 12:31 am Eastern Daylight Time on Jul 23, the solids light and you are off, with the folks in Houston controlling the flight. The first call is `tower clear' as they pass the top of the launch tower. We don't like to hit the launch tower. The next call is `roll program', as the Shuttle turns to face the direction it needs to fly (due east in this case for a 28.5 deg orbit). Now something unexpected happens. As I watch the Shuttle arc into the sky, I hear Eileen Collins's voice report a problem with a main engine controller. Oh, my. This is NOT what I want to hear five seconds after launch. My heart sinks and I wait for disaster. Losing one of the three controllers is not fatal - each engine is cross wired to a backup controller. But we may now be a single electrical fault away from losing the vehicle, and later inspections will show more frayed wires in the Orbiter. Unknown to me, there is already a second problem in the engine system: a small hydrogen leak in one engine is spilling fuel. This is a nightmare scenario: two separate failures in the same general system is exactly the sort of thing that causes what NASA calls `having a bad day'. We are closer to a bad day than at any time since the loss of Challenger. Collins' voice continues reporting calmly as half her control panel lights up red - the first woman spaceship commander, I hear she now has the nickname `Janeway' for her coolness in a crisis.
The next big thing is `max q', maximum dynamic pressure. As your speed V increases and the atmospheric density rho decreases (with altitude), the product rho V squared (ram pressure) goes through a max. Lots of rockets with not strong enough nose cones have come to serious grief at this point. The SSME's throttle down to go more gently through max q. Then they throttle up again. The next call is `Columbia, go at throttle up'. Now you're back at 104 percent rated thrust. At T+2min 3sec, the SRBs peter out and are jettisoned, falling back to Earth on parachutes. Breathe deeply. The next call is `performance nominal'. This means that the SRBs gave the expected amount of thrust and everything is ok for that first stage. Now you have 6 more minutes to MECO (main engine cutoff). CNN often stops covering it at this point, but there's plenty of chances for trouble. In fact, every radio call from now on is about what to do if things go wrong. You'll hear:
`negative return' (you're far enough downrange, don't bother turning round coming back to KSC if things get bad, go transatlantic instead) `two engine Banjul' (you can make the fastest ever trip from Florida to the Gambia if one engine fails) `single engine Banjul 104' (if two engines fail, ramp the third one up to 104 percent thrust and you might make it to Africa) `select Zaragosa' (turn the abort switch so that if things go bad, you fly toward Spain instead of Africa) `single engine press to MECO' (if two engines fail, carry on and try and make it to orbit)
There are some things you really don't want to hear. Anything with the word `abort' in, for instance. Examples: `Abort! Abort ATO!' This is the best case. It means we lost an engine, but we're going to try and make it into a low orbit anyhow (``Abort To Orbit") by dumping some fuel and some other tricks. Then check you are still in one piece, and if you're real lucky maybe still carry out the mission. `Abort AOA!' (note that these are *always* said with exclamation marks). Abort once around, doing a suborbital skim around the world and landing in California. `TAL abort!' Transatlantic. Very quick frequent flyer miles to Africa.
`Abort RTLS' Very bad news. Only between T+2min and T+4min. Try and turn the orbiter around in a 180 WHILE STILL FIRING THE ENGINES, to slow it down and head it back to Florida. (Take a *big* curve. Cross your fingers when you go through velocity zero.). Flip the orbiter over (initially you're flying upside down). Hope that by some miracle the wings don't fall off the orbiter. Then turn off the engines, dump the tank, and glide back to the runway at KSC. The astronauts are not fond of simulating this one.
`Contingency abort'. This is NASA officialese for `you are screwed. We'll do what we can for your families'. For instance, the SRB just burned through. You wait until the SRBs burn out at T+2min, since you can't dump them. If you are still alive, turn off the engines, dump the tank, and try and glide down to a low altitude above the ocean. If you are still alive and the orbiter is stable with wings level at Mach something (as opposed to the more likely spinning and tumbling around all axes), open the side hatch and jump out. If you manage to miss hitting the wings and your parachute manages to open, glide down and watch as a two billion dollar spaceship plunges gracefully into the Atlantic. In the event, we made it into orbit without any of those bad things happening. You hear the call `MECO' (main engine cutoff). 10 seconds later, the tank separates and falls away. They say `Go for the ET Photo DTO', telling the crew to try and get a picture of the tank as it falls. The SSME engines are now useless weight. You are now meant to be in an orbit of about 78 km by 276 km. In fact, Columbia ran out of fuel a second early because of the hydrogen leak, leaving you in a slightly lower orbit of 78 by 269 km. If you do nothing else, you will reenter the atmosphere in one hour. At 41 minutes after launch. i.e. 1:12 am, you reach apogee and fire the smaller OMS engines to increase speed. Now you are in a 267 x 290 km orbit, a little lower than planned but not a huge deal. In an orbit like this, you'll stay up for weeks, so you can relax a bit and get down to work. At 1hr 30min (a little after 2am) the Chandra powerup begins and the payload bay doors are opened.
A few hours later I'm sitting on the sandy shore back in Cocoa Beach, watching the stars and drifting in and out of sleep. As dawn breaks, Chandra is raised out of the payload bay, attached to the large firework called the IUS. At 1147 GMT on Jul 23, Columbia cast free the Chandra Observatory, and the astronauts film it floating out of the bay. The crucial rocket burns follow soon afterwards, and by 10 am local time (1400 GMT) we know Chandra is free of its rocket stage and has opened its solar panels - it's time to hop in the rental car and head back to Cambridge for my first duty science shift in the control center. There are a lot of milestones still to come and plenty of opportunity for problems - the IPS burns to reach our final orbit, the instrument and sunshade door openings, and of course the first light image - but the waiting is over at last. The long saga of AXAF/Chandra, the years of preparation, the months of delay, the tense week in Florida, are all history now. It's time to start doing science.
- Jonathan McDowell