Helium-filled balloons are a poor man's way of reaching altitudes where conditions are almost space-like. With them, you can transport multi-ton instruments to altitudes of 100,000 feet or more at a cost of as little as $100,000. These altitudes are a quarter of the way into space, the stars are pinpoints of light that don't twinkle, and 99% of the atmosphere is below your payload. You can't launch these balloons just anywhere, however, which leads you to some unlikely corners of the world.
Palestine, Texas is a small town, population 15,948, tucked away in the eastern part of the state about 116 miles southeast of Dallas-Fort Worth. It's the most unlikely spot in the country to expect to find an astronomical observatory, only a few hundred feet above sea level on hilly countryside dotted here and there by cattle ranches. Assuming they're not there for the annual Dogwood Festival, a casual visitor would not have the slightest idea what draws hundreds of astronomers and physicists to this small community each year. It's really not that much different from Tyler to the north, or Nacogdoches to the east, except for one thing: Neither of these towns harbor the National Scientific Balloon Facility - the " B'loon Base " as the natives call it. Here, NASA and the National Center for Atmospheric Research, NCAR, have put together a permanent installation for launching scientific payloads of all types using enormous, helium-filled balloons.
One of my first observing trips as a graduate student in 1979 was an expedition to Palestine to help launch an infrared telescope to study astronomical objects. The multi-million dollar instrument was a 1.02-meter, cassegrain telescope built by engineers and astronomers at the Center for Astrophysics and the ,A href="http://www.as.arizona.edu/"> University of Arizona under the direction of Dr.Giovanni Fazio. Built between 1973 and 1975, the early years of infrared astronomy, it continues to be the largest such telescope in use for studying the far-infrared sky. It's a highly versatile instrument. For each flight (there have now been 18) as new detector systems become available, they can be installed at the focus of the telescope, providing astronomers with constantly increasing sensitivity. The telescope itself operates like a camera lens; though we upgrade the speed of the 'film' that we use each time, the 'camera body' remains the same. The 'film' consists of a 1 millimeter square chip germanium crystal, cooled to almost Absolute Zero and located inside a cryogenic thermos bottle called a dewar. A slight change in the infrared radiation reaching its supercooled surface causes a billionth of a volt change in the current flowing through it, which is amplified and read out.
No two trips to Palestine were ever quite the same because each was filled with its own special combination of people, equipment, scientific goals and, of course, technical problems. Giovanni Fazio is a cosmic ray physicist who came to the Smithsonian Astrophysical Observatory during the late '60s to set up an infrared program there. By the time I joined the Balloon Group it had grown to include two engineers, Frank Licata and George Nystrom, and a variety of Visiting Scientists along with their specialized instruments. Dr. Brian McBreen a visiting astronomer from the University of Dublin had also started out as a cosmic ray physicist at Cornell, but as for Giovanni, his interests turned to infrared astronomy once the heyday of cosmic ray research came to an end in the early '60s. In recent years, Giovanni's telescope has also been used by atmospheric scientists like Dr. Wesley Traub whose spectrometer who uses it to study certain atmospheric compounds as well as molecules in extraterrestrial objects.
The goal for our September 1979 flight was to launch the telescope with a pair of sensitive infrared detectors at its focus, one built at Harvard and one built in France by Dr. Gary Chanin and his graduate student. We were going to use the Harvard system with its improved sensitivity to map a dozen far-infrared sources and to see how well the telescope could be made to track stars with its new, upgraded servo control system. The French detector was a new, untested system and we wanted to find out if it was sensitive enough to see anything at all.
Thursday, September 4 Usually the trip to Palestine would begin with an early-morning flight from Boston's Logan International airport to Dallas-Fort Worth, arriving sometime in the afternoon with no lost baggage and a rental car waiting. Brian McBreen and I left for Texas on the 11:00 AM flight out of Logan arriving at DFW at 6:00 PM. Our bags did not arrive until 10:00. After a monotonous two hour drive we took the left turn onto West Palestine Avenue, arriving at 12:30 AM at the TexAnn Motel. West Palestine Avenue is a 'main drag' and has a variety of motels and restaurants to choose from. The natives speak in the unhurried inflections typical of 'these parts'. Pick-up trucks are not uncommon. You can find fast-food restaurants offering fried chicken gizzards if you know where to look, or care to.
Friday, September 5 The next morning after a late breakfast at the 'Pitt Grill', we drove to the Balloon Base. The 5-story hangar building housing all the scientific payloads is enormous, with four large doorways some 40 feet high through which the payloads are moved when their launch time comes. Six groups at a time can be accommodated, though rarely so, because it can get rather tense when thirty people crowd into a small space, especially with each group trying to get their payloads working while jockeying for the earliest launch date.
The payloads themselves take a variety of shapes and sizes, usually very large. Gamma ray and cosmic ray physicists are the champions of the heavy payloads since their detectors have to be shielded with heavy steel or lead plates to protect them from unwanted radiation or charged particles. They also need to fly their payloads the highest, in some cases 120,000 to 150,000 feet, to see through their 'windows' more clearly. Even in comparison to these heavyweights, Giovanni's infrared telescope is one of the largest payloads standing twenty feet tall and weighing two tons. That's right, two tons. When I first laid eyes on it I was astounded. Its mirror is three feet across and the telescope is mounted inside of an aluminum frame made of three inch diameter pipe. In fact, most of the payload's weight is in this superstructure which shows the nicks and scratches from all the landings its been through. After every landing, some pipes have to be replaced, so that the current frame resembles a crazy quilt of old and new aluminum support members welded together.
The entire payload is suspended from the retaining chains of the overhead crane by a single, one-inch diameter steel bolt! Eventually, it will be suspended by the same bolt from the shroud lines of a parachute whose crown will be attached via a second set of pins to the balloon. For the present, I just stand and gawk at this aluminum and glass leviathan. Weighing as much as a car, when we finally get all of its control systems working, we will be able to move it by radio control with the precision of a Swiss watch, from two hundred miles away! During the next week or two, we will finish refurbishing the 'gondola' electronics, replacing defective 'chips' and resoldering connections. The control room will be set up with our equipment and we will run through checks of all the major payload systems using the telemetry link. We will even operate the telescope remotely at night to check out the command link and see if the telescope performs well enough to track bright stars. We have a long way to go before we are ready to see 'first light' on our reborn payload. The road is usually filled with unexpected twists and turns.
It's not that we can't anticipate potential problems, it's just that the payload is such a complex interweaving of electronics, servos and mechanical devices that on one day, it could be working just fine and on the next, a weak solder joint could give out, causing some vital circuit to mysteriously stop working. After working on the system and running it long enough, this refurbishment phase eventually ends, and the payload is deemed flight-ready.
As a novice in this program with no engineering skills, my responsibilities were to assemble the sky charts for each of the objects we were going to observe, and to figure out in what order we should observe them during the payload's 10 hours aloft. Preparation of the observing schedule is a challenge. About 15 to 20 objects located in different parts of the sky have been selected that we want to observe for 15 minutes to an hour. Since some of these objects will be rising while others are setting, we have to decide on the order in which to observe them. Fortunately, the time of year and other constraints leave you with only a handful of viable possibilities. After talking these over with my more experienced collegues, we would usually agree on a schedule that looked optimal. At this point starts the long process of determining the best way to move the telescope to target the objects in the viewing field of the telescope. A TV camera bolted to the side of the telescope enebles us to see the stars near each of the objects of interest. Since infrared sources are invisible to the naked eye, we first have to select a bright star that's easily identifiable on the TV screen, and then offset from it to the location of the infrared source. My job was to use a star atlas to anticipate exactly where the stars would be on the TV screen when we are correctly pointed at the target. Since the orientation of the stars changes from minute to minute, these charts have to be updated each time the launch date is delayed.
Saturday, September 6 I continued working on the observing schedule much of the day and read a few technical articles describing each of the targets. Instead of joining the others for lunch, I decided to go out jogging around the perimeter of the launch pad down the road. It was a warm day, 85o F, and since I was wearing my shorts and sneakers anyway, running seemed like a good idea, especially after sitting at a desk all morning. The launch pad is a circular, gravel covered area about 1/2 mile in diameter. Together with the 1/4 mile jog from the building, it's an ideal 2 miles of nearly flat roadway. My goal for the end of this campaign, about 3 weeks hence, was to run 5 miles in under 40 minutes.
The remainder of the afternoon was taken up by unpacking equipment, and moving it into the control room on the 3rd floor. That evening after dinner, some of us went to the Western Club, had a few beers, and played pool until 1:00 am.
Monday, September 8 We began our checks of the telemetry link, the CIP electronics, and the electronics aboard the payload to make sure everything was working properly. The Consolidated Instrument Package or 'CIP' contains all the telemetry equipment, the data encoders and command translators. If it were to fail in any one of a thousand ways, we could lose communication with our payload. It's got to work, and it's got to work perfectly every time we use it.
Testing the CIP and its interface with the payload is a tedious process. I have anxious memories of lying on my back under the 2- ton payload suspended inches above my face. With a walkie-talkie in one hand and an electrical probe in the other I would systematically touch the probe to each of the 200 pins in the cable connector, Brian McBreen would talk to me from the control room to let me know if my test signal was getting through the telemetry link or not. If not, we would have to call an NCAR technician to test the CIP and fix it. We don't like people mucking around with the CIP once it's installed because even if they can fix it, that means we have to do all the tests over again to make sure some other problem was not produced while the one at hand was fixed. We had no problems this time.
Monday, Sepember 15 Ten days have passed. We can't get the telemetry computer in the control room to synchronize with the binary data being transmitted from the CIP, but we're also unable to issue commands to the back-up command system on the payload. Last night I inadvertently exposed the test film we took with the starfield camera, so we'll have to shoot it again tonight. Our electrical engineer Frank says that due to the misloading of one of the circuit board cards, the time-of-flight clock inside the camera may not be working. A few test exposures will tell us if this is so. The workday did not end so dismally though. We eventually discovered that the synchronization and command problems we were having in the control room could be traced to a faulty cable in the control room. And then there was a bad experience with a cow.
It was 1 AM. As we were returning to the motel after a long day at the base the car ahead of us came to an abrupt stop and pulled over to the side of the road. We got out to offer our assistance and an obviously shaken woman stepped out of the car. " I hit something" she said, "but it was too dark to see what it was". Stepping around to the front of her car, we saw that the grill was totally smashed in. Our curiosity piqued, we walked up and down the side of the road until 200 yards behind us we found the culprit. A large Black Angus cow was lying on its back with all four legs pointed straight up in the air. It was quite dead from the impact. Within a short while the Highway Patrol was on the scene, and the rest is local history.
Saturday, September 20 On Friday we tried to get to the bottom of why the momentum dump ( we call it the P-dump ) system had mysteriously crapped out that afternoon after running well all week. Without it, the telescope would not be able to counteract the slow, side-to-side twisting motion of the balloon, or its own motion after it finished moving to a new object in the sky. Many tests were performed but to no avail. I'm not an engineer, so I kept right on working on the sky charts, breaking my concentration only long enough to continue my lunchtime jogging ritual.
Then, last night, the servo amplifier integrated circuit was accidentally shorted out. This $300.00 item can only be bought in Arizona, prompting Frank to suggest while ordering the new one by phone, that when the payload circuits are re-built next year, all components should be available from Radio Shack. If we were lucky, it would arrive in two days. Later that week, with the new servo amplifier installed, the P-dump system died completely; it wasn't until late the next day that we found the problem. The motor was jamming the reduction gears in the tachometer. George Nystrom tossed it into his car and drove to Dallas to get it repaired.
Aside from the ills of our own payload, it's been a rather uneventful stay so far. Only one or two groups have had their payloads launched and they went up without a hitch. A few years back, a group of physicists from Holland had a much more memorable experience. Shortly after their balloon was launched, a huge rip appeared in the delicate plastic of their balloon. Within seconds, the enormous stresses had completely torn the balloon to pieces and it fell back to the ground. Their payload had not been released so it wasn't damaged; in fact, they had a successful launch a week later with a new balloon. A local furniture store, thinking to cash in on the novelty of such an event, put up a sign announcing a 'Balloon Bust Sale'. This was the shortest sale in history because once the Balloon Base caught wind of it, they decided it was really tacky and told them to please take the sign down.
Saturday, September 27 The payload is finally up and running and it looks like we've successfully exorcised the last gremlin from it. We have been here for 3 weeks now. Our only source of delay is the current spell of bad weather. Because we had long since passed the old tentative launch date, I spent today recalculating the target positions and parameters for a possible launch on Monday or Tuesday.
At Monday's weather briefing, the report was for continuing cloudiness with rain likely. Tuesday looked good for a launch, but not ours! First, the Maryland group would fly, then Chicago, followed by Goddard, but not a word about us. Giovanni had several talks with Al Shipley and Bob Kubara who ran the Balloon Base, but they wouldn't budge from their scheduling decision. Since our flight was shut out for the next 10 days we decided that it was actually cheaper for us all to go home for 10 days and return, rather than stay in our motel rooms for the same period of time. Our French collaborators were pessimistic about being able to get more travel money from the French space agency, but they decided to leave their instruments here anyway, just in case they could find the money to return. Brian would not be able to come here a second time because of his teaching commitments. But, it wasn't just our schedules that were upset by this delay. Giovanni called Wes Traub at Harvard to tell him that he would have to wait a few months before he could have his spectrometer flown on the gondola. Wes was not at all happy about hearing this bad news. While we were here in Palestine, he and his two graduate students, Jan Vrtilek and John Brasunas, were back at Harvard packing the spectrometer for shipment here. It's a shame. All of this could have been avoided if the P-dump hadn't been so insufferably flaky.
We stored the loose gear and cleaned up our bay. The gondola would stay where it is until we returned next month to try again. Following a gathering for our 'last supper', tomorrow we'll depart for Boston, France and Ireland. Whether we assemble here in mid-October for this flight remains very unclear.
Friday, October 17 Back again! Got the payload working properly in record time. I updated the targeting data files for tonight's observing of the sky in preparation for a launch possibly tomorrow. The weather is very clear and we are first up for launch in an almost empty hangar.
Monday, October 20 The 10:00 AM weather briefing predicted good launch conditions and we got the go-ahead for a launch this evening! Now comes the next phase: How do you carry a delicate, 2-ton, 20 foot tall payload, from the Hangar Building to the launch pad a half mile down the road? The answer: With a 4- wheeled machine that weighs 52 tons, is 35 feet tall, has tires 8 feet high and is nicknamed Tiny Tim. It will arrive at 1:30 PM and at the pace of a brisk walk, transport our payload to the launch area. We will begin our final flight check out at 2:00 PM, and if all goes smoothly, we'll be ready to launch at 6:00 PM.
After our checkout was complete at 4:00 PM, the NCAR technicians put their tracking beacons and transponders on the payload. We watched them with a wary eye because last year someone accidently put 48 volts across the input power terminals of the CIP and blew out many vital circuits. The accident not only forced us to abort the launch which was to have happened within three hours, but we couldn't get another flight opportunity until 6 months later. Everything went well this time, however, and we were still 'go' for launch.
Two hours before launch, a second crew of technicians drove up in a helium tank truck carrying the balloon. The parachute was attached via long shroud lines to a bolt on top of the gondola and the crown of the parachute was attached with explosive release bolts to the balloon coupling. Using kid gloves so as not to damage its thin plastic, the balloon was carefully unfolded and about 30 minutes before launch, inflated with helium. The balloon itself weighs about 2 tons and is made of a polyethelene plastic only a hairs-thickness wide. Nylon cables called gores run from the balloon's crown to the coupling collar where the parachute is attached. When the balloon reaches its floating altitude, the helium inside will have expanded the balloon into a 350' diameter ball.
The launch of Flight 1226 at 5:55 PM was flawless. By 8:00 PM the payload was 91,000 feet above our heads and 50 miles down range. We commanded the gondola to unstow the telescope and a few commands later, the telescope was pointing at our first target, M-17. As our fifth magnetic tape came on-line to record the telemetry data, the ink pens on our stripchart recorder sprang to life telling us that our infrared detectors had just scanned across this bright source. Unfortunately, Gary Chanin's infrared system was silent. Following 30 minutes of troubleshooting, Gary concluded that their experiment had completely lost its sensitivity because their cryogen had evaporated. Once the one- millimeter-square detector 'chip' heats up a few degrees above absolute zero, it loses nearly all of its ability to detect infrared photons from distant astronomical objects striking its surface. Sadly, all their effort since September was in vain and, quietly, Gary and his student sat back in their chairs watching the flat traces from their dead instrument. Meanwhile, our next target was W-51; no sooner had we offset from the star Zeta Aquili that all four of our infrared detectors leaped into action as one infrared source after another passed into view. We counted nearly a dozen of them, each possibly a luminous dust-enshrouded star being born.
While observing the Perseus OB-2 star group at 11 PM, we lost both telemetry synchronization and the TV images for a mind- numbing 10 minutes! Just as suddenly, they came back. This was no time to ask questions; we just kept right on observing hoping that this new gremlin would vanish and not bother us again. The infrared signals from Perseus OB-2 looked very odd. The wavy lines tracing the rise and fall of the infrared signal were not at all what we ordinarily see when scanning across an infrared source. The telescope was at an elevation angle of 65 degrees. Could it be that the detectors were seeing a bit of the warm balloon? Running out of time for this object, we moved on to TMC- 1 at a lower elevation. Almost immediately, a very strong signal was picked up which we then proceeded to map out. After 15 minutes the telescope was pointed at the Crab nebula in the constellation Taurus, but there was no evidence of a strong infrared source there.
We are now 8 hours into the flight at 2:00 AM. We commanded the telescope to move the short distance to the bright star Betelgeuse in Orion, but we could see no trace of it as an infrared source. FU Orionis, a very young star, was next, but we saw nothing there either. Then, at 9 1/2 hours into the flight we got a strong signal from the star forming cloud NGC 2024 in Orion. For the next 30 minutes we scanned the telescope back and forth in an automatically executing raster pattern to map out the emission from this object. The galaxy M-82 came last on our program, but after nearly one hour of searching we could not locate it to obtain a measurement.
With one hour to go before sunrise blinded our TV camera, we quickly pointed the telescope at Jupiter to obtain our all- important calibration measurement. It took only a minute to find this bright infrared source, and we all breathed a sigh of relief. Without this measurement, all of our mapping work would have been rendered useless. Jupiter is a well-studied infrared source, so by comparing its signal with those of the unknown sources, we would be able to find out just how bright the unknown sources are.
Sunrise at 7:00 AM signaled the end of our long, arduous night. The gondola was commanded to stow the telescope in the vertical position and the preparations for 'cut-down' commenced. We began gathering our papers, notebooks and 25 precious computer tapes just as the sun rose on the horizon. The safety of our telescope now rested in the hands of the recovery crew. The ground crew was already in position just outside of Shreveport, Louisiana to rush to the payload once the parachute had placed it on the ground. This can be a very touch-and-go procedure because if the balloon doesn't separate from the parachute properly, or if the parachute fails to unfurl correctly, a freefall from 30,000 feet up can reduce a delicate, costly experiment to a crumpled mess. I have heard of grown men rendered mute by a payload dragged through a swamp and returned in cardboard boxes on the back of a flatbed truck. Fortunately, our payload has survived over 18 flights with only minor damage due to the occasional rough landings by parachute.
Payloads can come down in strange places. One landed a few dozen feet away from a $250,000 house, another, with its expensive telemetry and control electronics located near its top, came down near a lake. It had tipped over so that the only part in the water was, you guessed it, the expensive telemetry and control electronics. Payloads sometimes pass across the Mexican border. No 'extradition' treaty exists with Mexico, so the payloads are never heard from again. Louisiana is only a few hundred miles to the east and, yes, ground crews have recovered scientific equipment from swamps. One landed in a field and when the recovery crew got there they found it riddled with bullet holes. Someone had mistaken it for a flying saucer and reduced it to just so much scrap metal. With all these precedents, we could only cross our fingers and hope for the best.
The recovery and storage of the payload would be carried out by the NCAR staff and by our own engineers, so that there was little else for us to do but to check out of the hotels and catch the next available flight home. After a long relaxing weekend, we would make our ways back to the lab and begin the difficult process of analyzing the data and making sense of what we had observed. Little did we suspect as we were leaving Palestine this time, that we would never see Gary Chanin again. On January 8, 1985, Gary died of cancer.