Monday, November 12, 2012

FOXSI launch day, part III


Now that the rocket flight was over it was time for another adventurous flight, this time via helicopter.  

Rockets launched from White Sands land far uprange, about 40-60 miles (60-90 km) from the launch rail.  The payload sheds all its accessories during the flight: the two motor stages that give the payload its boost into space fall off after burnout and the nose cone (the tip of the rocket) falls away during the descent to allow the parachute to open.  These components (or what's left of them) also land somewhere in the desert.

Our mission was to find the payload, split it into two sections, and bring these back via two helicopters.  Besides the Navy helicopter pilots and photographer, there was room for four of us.  The team ended up as Ed (Wallops electrical engineer), Paul, Steven, and me (after Säm graciously turned down a seat in order to let Steven or I go).  We were warned that this could be quite strenuous work -- each of the payload sections weighs 200-300 pounds, and the landing site could be a small ravine or a difficult incline.  Luckily, FOXSI made this easy for us by picking a nice flat stretch of desert to settle on!

Even without the rollercoaster emotions of the flight and the anticipation of recovering FOXSI, the helicopter ride (my first ever!) would have been memorable.  It took about a half hour to reach the landing site, during which we were treated to a spectacular view of the desert.  200 feet below us the sand dunes and vegetation created strange geometrical patterns on the desert floor.  I was able to see for the first time how vast the famous white sands really are, the peculiar fine white gypsum powder giving the illusion of a giant salt basin.  I saw tracks that looked like lake/river beds and creeks, obviously long dried but perhaps occasionally bursting back to life during rare desert rainstorms.

And then we found FOXSI.

I am sure that the mesquite bush our payload chose to land in is the thorniest one in the desert.  It looked innocent enough, until we dove in to clip the parachute cords, and then its spines cut us at every move.  I like to think of it as sneaky little FOXSI getting in a few last jabs!

But the landing was perfect!  After FOXSI was finished observing the Sun, a shutter door closed over the telescope.  The door carries a metal honeycomb structure that acts as a crush bumper, and it is this structure that the payload lands on.  FOXSI landed on end on its crush bumper and then fell over on one side, snuggling itself into the thorny mesquite bush for us to find.  (If the parachute had not opened as expected we would have found a hole in the ground instead of a happily resting payload.)

The work was straightforward: Ed vented the compressed gas carried in the payload for fine pointing maneuvers, we unbolted the two halves of the payload and broke all the electrical connections between them, and one by one we loaded the halves onto the helicopters using the large rocket cradles we had brought with us.  We covered the open ends with large tarps taped to the rocket skins, and then we were on our way!

Except we needed a pit stop.  The payload sections were so long that they stuck out the sides of the helicopter.  The high air turbulence outside the helicopter ripped one of our protective tarps to shreds, exposing the detector end of our telescope and causing electrical connectors to fly around wildly.  We landed again in the desert and used an entire roll of heavy-duty tape to tape directly over the end of the payload, creating a protective cover.  This proved good enough to make it back to the launch complex with no further damage.

Back at the vehicle assembly building, the triage began.  We were jubilant about the success of our payload, but now was the time (with all equipment, people, and memories present and intact) to figure out what things had gone wrong.  We checked our alignment with the rocket pointing system to see if we could have been horribly off target, and found that this was unlikely.

We examined our telescopes and electronics for damage…and found a little.  One of the seven optics modules stuck out a little farther than the others.  This happened because a glue joint holding that optic in place had failed, allowing the module to move.  Of course, with any payload damage a big question is whether it was the launch or the landing (a huge shock!) that caused it.  In this case we *think* the shift happened at launch, although it's difficult to be sure.

The other optics modules all looked just fine, and the next day we did one last X-ray alignment test to check that everything was still functioning.  With the exception of the one shifted module, everything looked exactly the same as it had before the flight.  This is fantastic news, because it means we have a mostly intact payload which could be used again for future flights!

…which brings us to the big question of what's next!  FOXSI has a second flight funded, tentatively scheduled for 2014.  For that flight we have planned some upgrades to the instrument, but those plans did rely on recovering the payload in one piece (which it mostly is).

As we begin to look at the data I'll attempt to post some updates about our progress.  First, though, is a big push to finish my dissertation in time to graduate this semester!  And one more piece of exciting news -- I'm told that the official launch photos and videos are on their way to us soon.  I will post these as soon as I get them!

Thanks again for all the warm wishes and support from all our FOXSI friends!

Friday, November 9, 2012

FOXSI launch day, part II


The countdown continued.  I felt strangely detached.  The countdown hit zero.  On the TV screen there were smoke and flames and the rocket leapt up off the ground.  Before that moment FOXSI had always seemed kind of small on the launch rail, at least compared to the gigantic rockets you see in pictures of spacecraft launches.  But at that moment FOXSI was huge and powerful and dangerous, an object visible from all over the desert.

The sound hit us later, having taken several seconds to travel the miles between the launch rail and our building.

FOXSI was in the air, climbing slowly.  The second stage motor ignited, there was another bright flash, and FOXSI shot off unbelievably fast, now truly on its way to space!

I was floored by what I had just witnessed, but now there was work to do!  The most important step was to send the uplink command to turn on the voltage to our detectors.  If this command were to fail for any reason then our X-ray cameras would not work.  It was for this reason that I was inside, watching the launch on a TV, rather than outside with the others -- I wanted to be focused and ready for this crucial step.   At 30 seconds after the launch, I sent the command and watched the data to see that it had registered.  It did, and the detectors immediately began to respond.

There were about 70 seconds more to wait for FOXSI to get to space, the shutter door to open and reveal our telescope, and the pointing system to find the Sun.  We would start observing when the rocket was about 150 kilometers (100 miles) high, well above the dense parts of the Earth's atmosphere.  The other team members ran back in to take their places by my side and we eagerly watched our screen to see the data streaming in…

…and we saw nothing.

Well, not exactly nothing.  There were a few X-ray hits here and there on our detectors, but the rate was quite small and they did not come from a concentrated spot on the Sun as we had expected.  We looked at each other and calmly made the decision to switch to a backup target.  Säm (our principal investigator and the "driver" for our payload) sent the command to change targets.

The second target showed no improvement.  Again, no concentrated spot appeared, but only random hits here and there.  At this point it looked as if the mission might be a failure; with no recognizable image at all it would be difficult to prove we had even been looking at the Sun.  Despite this, we moved on to a third target. 

 This target showed nothing much as well.  Now about three minutes had gone by, which is half our observation time, and all of our best prospects for finding an X-ray source were gone.  The rocket reached its peak height of 300 kilometers and then started on its way down.

We switched to a fourth target, one of two we had programmed in as last resorts.  (We had appropriately nicknamed these "Hail" and "Mary.")  We did not expect much, if any, emission from these targets, but they were regions on the Sun we hadn't looked at yet.

And suddenly there it was!  The count rate jumped up significantly, and an X-ray image started to build up.  Excitedly, we yelled at Säm to center the image!  He sent the command to do this, and the image disappeared.

That's right, it disappeared.  We had gone the wrong way!  If you ever needed an "undo" button, this was the time!  And luckily we *did* have an undo button.  Säm steered us back to the target and we found our image again.  (Later, we found that a problem in the way we had written our software was responsible for our wrong turn.)  We were in instant agreement not to touch anything more; we stayed right where we were for the last minute of the observation time.

And then it was done.  The rocket was descending back into the atmosphere.  The shutter door closed over our optics.  I sent the command to begin the shutdown of our detectors.  The parachute opened and FOXSI slowly continued its descent back to the desert sands, so much more carefree and calmly than the manner in which it had left.

Steven gave me a high-five that nearly broke my hand, and that began the celebration…we had done it!  FOXSI had done it!  We had made a focused image of X-rays from the Sun, the first ever made in this particular way in this energy range.  At that point we did not really understand what it was we had seen, but the important thing was that the instrument had worked.

Later we would learn that we had been extremely lucky.  The solar object we imaged was not a sunspot, as we had expected, but a solar flare in progress!  It was a small flare, but it appeared at just the right time and in just the right place for FOXSI to see it.  Without this flare, it is possible that we may have had no recognizable signal from the Sun.  Hail Mary, indeed…

So what to make of our other targets?  Why did we see no, or very little, X-ray emission?  There are several possibilities.  The active region targets could have been either cooler or fainter than we expected.  Our sensitivity could have been reduced for an unknown reason.  The quiet-Sun signal could have been extremely faint or even nonexistent.

But here is where the solar flare we caught really comes into play.  With this flare we can do a cross-calibration of our instrument with an existing spacecraft, RHESSI.  With this proof of a working instrument we can then make a case for what the X-ray flux actually was from the quiet Sun (and from the active regions).  If we find that there was no detectable emission, this is in itself an important science result.  But there may be emission after all.  As I mentioned, there was a very small rate of X-rays even before we saw the flare.  At this point it is not clear whether this signal was from the Sun or from the background X-ray flux that persists throughout the solar system.  It will take a tremendous amount of work to properly make that distinction!

But I digress…at this point there was a helicopter about to arrive to take us to the landing site and recover our payload.  We grabbed a couple bags of gear for the ride and took off for the helipad.

Stay tuned…part 3 will describe the journey into the desert to retrieve FOXSI!

Thursday, November 8, 2012

FOXSI launch day, Part I


Normally it would be a challenge to get me out of bed at 4:45 am.  But not on this day -- today we were ready to launch our rocket!

Everyone assembled at the missile range at 6:15 am.  This included the FOXSI team, the NSROC team -- some from Virginia and some from White Sands --who had built our rocket systems and managed the flight, engineers from the local university who provide telemetry, and Navy representatives, providing range and safety support.  Most of these team members would stay in the block house during the flight.  This is a sturdy building about a quarter mile from the launch rail where mission control takes place.  With 10-foot-thick walls and even thicker ceilings, this bunker is well insulated from the fiery blast about to take place.

But most of the FOXSI team would move to a building a couple miles away where data is received and sent during the flight.  From here, we can watch the data streaming in real time using software that we had written and send uplink commands to the experiment.  The best thing about this building is that it's far enough away that you can actually watch the launch outside!  (Not the case for the block house.)

The reason for arriving so early was that there were two more sequence tests to get through -- one horizontal (with the launch rail lying on its side on the ground) and one vertical (launch rail up, ready to launch).  After the launch rail "went vertical,", the rocket motors were armed (i.e. ready to fire).  After this, nobody is allowed to the rail except in case of emergency.

The other reason for getting an early start is that a few hours before the launch roadblocks went up all around the base.  The main road, highway 70, which leads from Las Cruces to Alamogordo and cuts through White Sands, was blocked.  The White Sands National Monument was closed for the morning.  Roads to the uprange section of the missile range were cut off.  All this was to make sure that FOXSI could not hurt anybody when it launched or when it landed!

The sequence tests were passed without incident.  During each of these tests, as before, we practiced the countdown and the commands we would send during the flight.  For the last time, we ran through scenarios for decision-making during the observation.  In the downtime between tests, we checked solar activity, took some team photos, ate a lot of food, and watched videos of previous launches.  The mood was excitedly nervous, but calm.

Twenty minutes before the flight a final check-in was called.  We put on our headsets in order to communicate with the rest of the team, got our software up and running, and got ready for the real deal.

Ten minutes before the flight the countdown began.  At this point we filled out a checklist of various temperatures, voltages, and tests to make sure that our experiment was working and ready.  If any of these checklists items looked incorrect, we would have the option to call a "hold" (five minutes at a time) or call a postponement of the launch.  There was a flurry of discussion as a few of our temperatures looked marginal, with one out of range.  We debated whether to call a hold, but decided that the deviation was acceptable.  (This proved to be the right decision, as all temperatures came within range in the several minutes before launch time.)

We were asked for our "go" status and we responded…GO!

In the block house, Lupe continued the countdown, her voice broadcast over loudspeakers in our building.  On one wall of the building, a garage door was rolled up, giving everyone on that side a clear view of the launch rail.  Most of our team gathered there to watch the launch.  There would be plenty of time to run back in before the observation interval began.  I, however, stayed inside, and I'll tell you why later.

I'll leave off at this cliffhanger for now…stay tuned for part 2, a description of the launch and flight!

Saturday, November 3, 2012

FOXSI's successful flight!


I've now spent almost a month in the desert, and I've never seen it from this perspective before.

"This perspective" is 200 feet above the ground, staring at the white desert sands from the jump seat on an open-sided helicopter as we fly to the landing site of the FOXSI rocket to recover the payload.

FOXSI flew today, and it was one of the most exhilarating experiences I've known.  Considering nothing else, witnessing the launch of a space-capable rocket would be enough to make a day exceptional.  Participating in the flight, turning on the instrument, and making decisions about the flight in real time makes the day one to never forget.

To say there were tense times would be an understatement.  Data coming back mid-flight showed discouraging results and we had to switch to all of our backup emergency targets before seeing an unambiguous X-ray image that we could interpret.  To understand this harrowing process, see my previous post on the challenges of making split-second decisions!

Once we finally saw it, the X-ray image was unmistakable, and we knew our mission was a success.

More details will follow (I promise!) telling you all about the experience, but for now I'd like to thank all the coworkers, family, and friends who gave such crucial support during some very tough times in putting this mission together.  In particular, without Andy Sybilrud I am sure this flight would not have had such success, and the FOXSI experiment owes much to him!

Thursday, November 1, 2012

Mission almost accomplished!


We have worked for four years to get to this point and today we got to see our rocket standing up on the rail, ready to launch.

It feels good.  It feels good to have passed so many hurdles and have gotten to the point we are now.

A few people have asked me what we expect for tomorrow, whether we expect to have a successful flight.  The answer is yes.  Everything we've seen from our testing here indicates that our systems are ready to go.  The caveat: there is still plenty that can go wrong.  Sounding rockets are by nature a risky business, and we've heard several stories of failure.  On one flight, the data downlink didn't work and they couldn't record any data.  On another, the alignment between the experiment and the guidance system shifted and they weren't able to look at the Sun.  On another, a grating shook loose and no useful data could be taken.  Some rockets have to be terminated shortly after launch.  Despite all these possibilities, we remain very optimistic that the rocket and experiment will work as planned.

Whatever happens tomorrow, I'm very proud we got this far.  I work with a wonderful team, and we built a wonderful instrument.  Tomorrow I hope to show you screenshots of never-before-seen solar data!  

Launch time is 11:55 Mountain Daylight Time.  Recovery via helicopter will take a few more hours, and after that I will try to update you all as soon as possible.