I did promise you pictures, didn't I? :) Here are some photos taken by the authorized photographers at White Sands. In addition, Steven's got some pictures of the cracked detector system parts on his FOXSI blog, along with some other shots he took outside the range.
Sunday, March 25, 2012
The wrapup and the rebuild...
I last left off with the unhappy news that we're pushing the launch. But I don't want you to think that's where the White Sands work ended...in truth we had much, much more to do that week! Here's what happened:
When we decided to delay the flight we also decided to go ahead with the next steps of integration and vibration, as a practice run. We'll have to redo these when we come back for the real deal, but doing them now gives us a chance to identify and fix problems early. Integration is the process of connecting our experiment (the payload) with the rest of the rocket (systems for data telemetry, pointing control, and recovery, among others). These parts were made by a rocket team from Wallops Island, Virginia. Not only did we physically connect everything and make sure it fits (remember, we had some troubles earlier when we tried this with the first piece...) but we also tried running the power, data, and communication through their systems just like it will be during the flight. We did several sequence tests, where we run through everything in order as if we are doing the launch...except of course we don't launch anything. :) Most of these tests were easy as our system is relatively simple to integrate, but we did have to work out a couple more problems in mating various skins and sections together.
The next step was vibration! As I told you before, a vibration test is done on a shake table that simulates the shaking the rocket will feel during launch. Passing this test is absolutely essential to the health of the mission because it means that nothing will break or come loose during the launch. Many things could go wrong in a vibe test: parts could break, connectors could detach, bolts can come out as if an invisible hand is unscrewing them! We've never vibrated the electronics side of our payload, so this test served an important purpose. (Imagine if we came back in August and THEN found that we couldn't pass a shake test!) Fortunately, the test was a success! All the detectors (well, the ones that were still working after the accident...) showed no damage in the vibe test and still worked just as well afterward. What's more, there was no change in the alignment of the optics, despite the shims and regluing. So I'm happy to say that once we rebuild the detector system, all major obstacles to launch will have been cleared! (Yes, there are always more things that can go wrong, but this takes care of all the obvious areas for concern.)
There was one final step for our work in White Sands, and that was to ship everything back to Berkeley. Things were a little melancholy last Tuesday as we packed everything back up and loaded up the Fedex truck. A day later, we unloaded the truck in Berkeley and FOXSI was officially back home! So what's next? Well, we have to start over by rebuilding the detector system, but we can use everything we learned along the way, so the process will be much quicker this time. We first need to get more detectors from our colleagues at ISAS in Japan; as soon as we get those I will begin gluing them onto new boards. After the boards are loaded and wirebonded, we will start the process of testing and calibrating them...and then...we will bring FOXSI back to the desert and we will fly it!!! Right now we're looking at an August launch date; I will keep this blog going in the meantime so you can keep tabs on the rebuilding. :)
When we decided to delay the flight we also decided to go ahead with the next steps of integration and vibration, as a practice run. We'll have to redo these when we come back for the real deal, but doing them now gives us a chance to identify and fix problems early. Integration is the process of connecting our experiment (the payload) with the rest of the rocket (systems for data telemetry, pointing control, and recovery, among others). These parts were made by a rocket team from Wallops Island, Virginia. Not only did we physically connect everything and make sure it fits (remember, we had some troubles earlier when we tried this with the first piece...) but we also tried running the power, data, and communication through their systems just like it will be during the flight. We did several sequence tests, where we run through everything in order as if we are doing the launch...except of course we don't launch anything. :) Most of these tests were easy as our system is relatively simple to integrate, but we did have to work out a couple more problems in mating various skins and sections together.
The next step was vibration! As I told you before, a vibration test is done on a shake table that simulates the shaking the rocket will feel during launch. Passing this test is absolutely essential to the health of the mission because it means that nothing will break or come loose during the launch. Many things could go wrong in a vibe test: parts could break, connectors could detach, bolts can come out as if an invisible hand is unscrewing them! We've never vibrated the electronics side of our payload, so this test served an important purpose. (Imagine if we came back in August and THEN found that we couldn't pass a shake test!) Fortunately, the test was a success! All the detectors (well, the ones that were still working after the accident...) showed no damage in the vibe test and still worked just as well afterward. What's more, there was no change in the alignment of the optics, despite the shims and regluing. So I'm happy to say that once we rebuild the detector system, all major obstacles to launch will have been cleared! (Yes, there are always more things that can go wrong, but this takes care of all the obvious areas for concern.)
There was one final step for our work in White Sands, and that was to ship everything back to Berkeley. Things were a little melancholy last Tuesday as we packed everything back up and loaded up the Fedex truck. A day later, we unloaded the truck in Berkeley and FOXSI was officially back home! So what's next? Well, we have to start over by rebuilding the detector system, but we can use everything we learned along the way, so the process will be much quicker this time. We first need to get more detectors from our colleagues at ISAS in Japan; as soon as we get those I will begin gluing them onto new boards. After the boards are loaded and wirebonded, we will start the process of testing and calibrating them...and then...we will bring FOXSI back to the desert and we will fly it!!! Right now we're looking at an August launch date; I will keep this blog going in the meantime so you can keep tabs on the rebuilding. :)
Thursday, March 15, 2012
Day 24 (Thursday, March 15)
Yesterday we made the difficult decision to cancel our launch. Our condition: we have 4-5 working detectors out of seven. Of these, three survived the cooling accident, and there are concerns that there could be cracks in the readout electronics that may get worse under the vibration of the rocket launch. There are two new detectors, however, our calibration data for these is minimal (limited to what I could measure in half a day!) Taking these things into account, it was decided to start over building our detector system from scratch. This will probably take 3-6 months, so we will try for a new launch in the summer or fall.
I've gotten a lot of questions, so here's some detail of what went wrong in our cooling accident. We have a cooling regulator that controls the flow of liquid nitrogen to our system. You can program it to cool to whatever temperature you want. It knows what the actual temperature is from a temperature sensor inside our system. Here's the nice part: based on the difference between the actual temperature and the programmed temperature it decides how much nitrogen to allow into the system. So you can essentially dial in the temperature you want and walk away; it takes care of the rest. Except, as I found out, you shouldn't do this because it can make a mistake! It turns out that if the temperature sensor is temporarily disconnected (i.e. if you unplug it and plug it back in) then things go a little haywire and the cooler does not behave properly. In this case it got confused about what the temperature was and started dumping way too much nitrogen in our system, resulting in much lower temperatures than our system can tolerate. This actually cracked a couple of components on our detector board, which are not replaceable. Long story short, we're starting over by building new boards.
Yesterday we made the difficult decision to cancel our launch. Our condition: we have 4-5 working detectors out of seven. Of these, three survived the cooling accident, and there are concerns that there could be cracks in the readout electronics that may get worse under the vibration of the rocket launch. There are two new detectors, however, our calibration data for these is minimal (limited to what I could measure in half a day!) Taking these things into account, it was decided to start over building our detector system from scratch. This will probably take 3-6 months, so we will try for a new launch in the summer or fall.
I've gotten a lot of questions, so here's some detail of what went wrong in our cooling accident. We have a cooling regulator that controls the flow of liquid nitrogen to our system. You can program it to cool to whatever temperature you want. It knows what the actual temperature is from a temperature sensor inside our system. Here's the nice part: based on the difference between the actual temperature and the programmed temperature it decides how much nitrogen to allow into the system. So you can essentially dial in the temperature you want and walk away; it takes care of the rest. Except, as I found out, you shouldn't do this because it can make a mistake! It turns out that if the temperature sensor is temporarily disconnected (i.e. if you unplug it and plug it back in) then things go a little haywire and the cooler does not behave properly. In this case it got confused about what the temperature was and started dumping way too much nitrogen in our system, resulting in much lower temperatures than our system can tolerate. This actually cracked a couple of components on our detector board, which are not replaceable. Long story short, we're starting over by building new boards.
Monday, March 12, 2012
Day 20 (Sunday March 11) Hurting but alive...
More bad news, folks. We broke almost all the detectors in the cooling accident. Two of them looked ok on data readout, two looked completely broken, and the other three were somewhere in the middle, with parts working and parts not. The situation's pretty dire because we didn't think we brought any *good* working spares! We did bring five extras, but they never worked particularly well in the lab -- that's why they didn't make it into the final buildup. You better believe there were some long faces around the lab today...one of the worst FOXSI days ever!
But it's not all bad news. After many attempts to figure out a way to revive the broken detectors we ended up putting in our spare detectors...and found out they work better than we thought! Two of the original detectors survived the cooling catastrophe, and of the five extras we put in two are working pretty well and we might be able to revive one or two more. So that gives us at least 4 out of 7 working detectors, with the possibility of more. Getting these detectors ready will be a ton of work. They have to undergo a full testing and calibration process...something that took us several weeks at the lab, but here we will need to do it in a few days! We also have to prepare these detectors for flight by staking down parts and gluing temperature sensors on them. And add onto that the time we lost (at least a day) in troubleshooting this problem...the race is on!! It's gonna be a lot of full days and late nights but I am confident we can do it. FOXSI lives!!
More bad news, folks. We broke almost all the detectors in the cooling accident. Two of them looked ok on data readout, two looked completely broken, and the other three were somewhere in the middle, with parts working and parts not. The situation's pretty dire because we didn't think we brought any *good* working spares! We did bring five extras, but they never worked particularly well in the lab -- that's why they didn't make it into the final buildup. You better believe there were some long faces around the lab today...one of the worst FOXSI days ever!
But it's not all bad news. After many attempts to figure out a way to revive the broken detectors we ended up putting in our spare detectors...and found out they work better than we thought! Two of the original detectors survived the cooling catastrophe, and of the five extras we put in two are working pretty well and we might be able to revive one or two more. So that gives us at least 4 out of 7 working detectors, with the possibility of more. Getting these detectors ready will be a ton of work. They have to undergo a full testing and calibration process...something that took us several weeks at the lab, but here we will need to do it in a few days! We also have to prepare these detectors for flight by staking down parts and gluing temperature sensors on them. And add onto that the time we lost (at least a day) in troubleshooting this problem...the race is on!! It's gonna be a lot of full days and late nights but I am confident we can do it. FOXSI lives!!
Saturday, March 10, 2012
Day 19 (Sat March 10)
Bad news, folks. Today I learned the hard way not to leave a cooling system unattended. Our detectors (the parts that make up the camera of the telescope) are meant to be used at -25 degrees Celsius. To get that temperature we have a liquid nitrogen cooling system that supplies the detectors with just enough nitrogen to keep them at the temperature we want. This is a great system if it's properly set! Unfortunately after I programmed the settings I did not double-check that these settings had properly registered, and the cooling system started the cooldown and just kept going...all the way to -190 degrees!!! As soon as I discovered this I stopped the cooling and shut down all power. We put a bag around the detector system and flooded it with dry nitrogen so that as the temperature comes back up water won't condense on the electronics. Now there's nothing to do but give it time to warm up...and tomorrow morning check to see if the detectors still work. :( I have the feeling I won't be sleeping well tonight...everybody wish me luck on this one!!
Bad news, folks. Today I learned the hard way not to leave a cooling system unattended. Our detectors (the parts that make up the camera of the telescope) are meant to be used at -25 degrees Celsius. To get that temperature we have a liquid nitrogen cooling system that supplies the detectors with just enough nitrogen to keep them at the temperature we want. This is a great system if it's properly set! Unfortunately after I programmed the settings I did not double-check that these settings had properly registered, and the cooling system started the cooldown and just kept going...all the way to -190 degrees!!! As soon as I discovered this I stopped the cooling and shut down all power. We put a bag around the detector system and flooded it with dry nitrogen so that as the temperature comes back up water won't condense on the electronics. Now there's nothing to do but give it time to warm up...and tomorrow morning check to see if the detectors still work. :( I have the feeling I won't be sleeping well tonight...everybody wish me luck on this one!!
Friday, March 9, 2012
Day 18 (Friday, March 9)
FOXSI just cannot catch a break! We came in today with the intention of doing a quick and easy operation in the morning. Guess what? Not so easy! The operation was to place the experiment in the hollow metal cylinders that form the outer shell of the rocket, called the "skins" of the rocket. These skins are just a little larger than our experiment and get quite hot during the launch. To do this fitting, you have to use a crane to hoist the experiment into the skin, then bolt the skin to a special fitting ring to join the two together. Well, when we tried this there was a big old gap on one side between the ring and the skin! (By "big," we're talking millimeters, but that's big in engineering terms.) It took well into the afternoon to come up with a solution. The best solution is to ship it all the way back to the facility that made it and have them machine it down. But this would cost us a few days in the schedule, and we have no room for that! However, there is a machine shop at a NASA facility nearby that could do the job. After some frantic attempts to get in touch with them they generously agreed to take care of it right away (late in the day on a Friday afternoon/evening!) and had the job done in a few hours. (Never met those guys, but I'll gladly buy them a beer if I ever do.) We tried the procedure again and everything fit like a glove. Yet another problem arising...yet another problem solved...! :)
FOXSI just cannot catch a break! We came in today with the intention of doing a quick and easy operation in the morning. Guess what? Not so easy! The operation was to place the experiment in the hollow metal cylinders that form the outer shell of the rocket, called the "skins" of the rocket. These skins are just a little larger than our experiment and get quite hot during the launch. To do this fitting, you have to use a crane to hoist the experiment into the skin, then bolt the skin to a special fitting ring to join the two together. Well, when we tried this there was a big old gap on one side between the ring and the skin! (By "big," we're talking millimeters, but that's big in engineering terms.) It took well into the afternoon to come up with a solution. The best solution is to ship it all the way back to the facility that made it and have them machine it down. But this would cost us a few days in the schedule, and we have no room for that! However, there is a machine shop at a NASA facility nearby that could do the job. After some frantic attempts to get in touch with them they generously agreed to take care of it right away (late in the day on a Friday afternoon/evening!) and had the job done in a few hours. (Never met those guys, but I'll gladly buy them a beer if I ever do.) We tried the procedure again and everything fit like a glove. Yet another problem arising...yet another problem solved...! :)
Thursday, March 8, 2012
Day 16 (Wednesday, March 7)
Back in the saddle after two days off this weekend in SF. (Being surrounded by water and people felt so strange after weeks in the desert!) The last couple days have been occupied with blanketing the entire payload. The blankets keep everything at their correct temperature...the optics at room temp...the detectors at -25 degrees Celsius. Keep in mind that outside this environment a fantastic fireball of an explosion is taking place, so thermal isolation is really important!
Too sleepy to write more now...good night, everyone!
Back in the saddle after two days off this weekend in SF. (Being surrounded by water and people felt so strange after weeks in the desert!) The last couple days have been occupied with blanketing the entire payload. The blankets keep everything at their correct temperature...the optics at room temp...the detectors at -25 degrees Celsius. Keep in mind that outside this environment a fantastic fireball of an explosion is taking place, so thermal isolation is really important!
Too sleepy to write more now...good night, everyone!
Subscribe to:
Posts (Atom)