Loss of ASTRO-H Satellite

[Glaran K'erman]

As uncommon as it is, it's not unheard of to lose communications or functionality of a space based instrument - it is another thing entirely to have it break into pieces. ASTRO-H was launched last month by JAXA (in slight cooperation with NASA) and was intended to observe the X-ray emissions of the universe. Though earlier today JAXA reported the spacecraft had experienced a loss of communications, followed by an orbital change, and further backed up by radar detection which showed the spacecraft in 5 pieces. Details are still coming in but regardless something went horribly wrong, a disappointing loss for science.

http://global.jaxa.jp/press/2016/03/20160327_hitomi.html

https://twitter.com/jointspaceops/status/714103414225960960

http://www.nature.com/news/japanese-x-ray-satellite-loses-communication-with-earth-1.19642

[fredinno]

9 minutes ago, Glaran K'erman said:

As uncommon as it is, it's not unheard of to lose communications or functionality of a space based instrument - it is another thing entirely to have it break into pieces. ASTRO-H was launched last month by JAXA (in slight cooperation with NASA) and was intended to observe the X-ray emissions of the universe. Though earlier today JAXA reported the spacecraft had experienced a loss of communications, followed by an orbital change, and further backed up by radar detection which showed the spacecraft in 5 pieces. Details are still coming in but regardless something went horribly wrong, a disappointing loss for science.

http://global.jaxa.jp/press/2016/03/20160327_hitomi.html

https://twitter.com/jointspaceops/status/714103414225960960

http://www.nature.com/news/japanese-x-ray-satellite-loses-communication-with-earth-1.19642

Already?

Was it a space junk collision?

[Glaran K'erman]

33 minutes ago, fredinno said:

Already?

Was it a space junk collision?

Could very well be, too early to tell for sure. A collision is much more concerning than say some catastrophic systems failure though, which is why I definitely am interested to see what they find in their investigations.

Edited March 28, 2016 by Glaran K'erman

[Scotius]

If this satellite (or rather its remains) changed orbit it means either collision or rapid unplanned disassembly. Ruptured fuel tank? Maybe it carried a pressurised substance onboard? Hopefully investigation will explain it.

[YNM]

Yet another collision ? Can someone give some data on how it changed ?

[Shania_L]

I would actually favour a hardware malfunction as a more likely cause given the close proximity of spacecraft activation and this accident.

It was launched back in Febuary and has been sat in that orbit for a while (being calibrated) the day after calibration is completed this happens? I think it most likely that something broke, or ruptured in one way or another as a result of the startup procedures.

Complete speculation on my part ofcourse, it may as well have been a Kraken attack for all the evidence I have seen, hope they find out what happened. Shame that this (or any) mission ends before it got a chance to do anything

[Frida Space]

Yes, JSpOC ruled out the possibility of a collision (of course, there could have been some undetected debris, but it's extremely unlikely). The main other possibility is an event of explosive or propulsive nature. The following is a timeline updated this morning with the latest info:

(all the times are in UT, however because of summer time change, UT = British time only for March 25th-26th. Starting from March 27th, the UK is one hour ahead of UT, so if you're in the UK you should add an hour to the times from March 27th onwards on this infographic)

Edited April 4, 2016 by Frida Space

[fredinno]

1 hour ago, Frida Space said:

Yes, JSpOC ruled out the possibility of a collision (of course, there could have been some undetected debris, but it's extremely unlikely). The main other possibility is an event of explosive or propulsive nature. The following is a timeline updated this morning with the latest info:

(all the times are in UT, however because of summer time change, UT = British time only for March 25th-26th. Starting from March 27th, the UK is one hour ahead of UT, so if you're in the UK you should add an hour to the times from March 27th onwards on this infographic)

Yeah, it looks dead. Possibly a tank or battery explosion.

[Spaceception]

Will there be a replacement?

[fredinno]

1 hour ago, Spaceception said:

Will there be a replacement?

Probably not, I can't find any mention of a back-up, and in any case, X-ray telescopes are expensive.

 

[Frida Space]

When the attitude control anomaly occured, ASTRO-H had just completed an attitude manoeuvre and had  begun observing Markarian 205. The attitude control anomaly was visible as a lack of solar illumination on the panels, as low power in the systems and as a variation in the temperature distribution. 6.5 hours later, the breakup.

[Scotius]

If there was radio contact after RUD, then probably batteries survived - or maybe probe was still getting some power from remaining solar panel. "Remaining, because i have a suspicion Object J might be  one of solar panels that broke away.

[Frida Space]

1 hour ago, Scotius said:

If there was radio contact after RUD, then probably batteries survived - or maybe probe was still getting some power from remaining solar panel. "Remaining, because i have a suspicion Object J might be  one of solar panels that broke away.

Yea, at least one of the >10 fragments is definitely a solar panel, I believe.

[fredinno]

4 hours ago, Frida Space said:

Yea, at least one of the >10 fragments is definitely a solar panel, I believe.

Is there any way the probe could run on only one solar panels and no batteries (if the batteries exploded)?

[Frida Space]

Okay, we have a lot more info now. A preliminary investigation into the failure yielded a chain of events that were likely responsible for the event.

I won't repeat all the times and dates of the various events, communication windows etc, because they remain largely unchanged from the ones posted above. The only major change is that the 6-second radio signal received at Santiago at 1533 UT on March 29th wasn't actually from Hitomi, or what was left of it at that time anyway.

The light curve of the main piece of debris (presumably the main spacecraft bus) suggests a rotation period of 5.2 seconds.

The telescope started an attitude manoeuvre to switch from the Crab nebula to Markarian 205. The manoeuvre took place from 18:01 to 18:22 UTC on March 25th. At around 19:10, the Inertial Reference Unit (IRU) detected a non-existent roll about the Z-axis. Usually, the star tracker data has the priority over the IRU data, which in this case was faulty, but star trackes were not being used at the time of the failure.

The IRU data indicated a 21.7 degree/hour roll. The flight control system tried to counter the roll by entering a roll of 20 degree per hour in the opposite direction. The torque was induced by the reaction wheels.

The momentum continued to build up from 20:49 to 01:04 UTC. The magnetic torquers attempted to dump momentum, but were ineffective. At 112 newton-meter-second, the spacecraft was near its saturation momentum of 120 Nms. With the IRU faulty data still showing a large roll rate, Hitomi switched to Safe Mode to use its sun sensor to dermine the solar vector and thus the attitude. The system should have then used the thrusters to point the solar arrays to the sun.

At this point, another error was introduced. The software uploaded after the deployment of the Extensible Optical Bench prevented the sun sensor from working correctly. Furthermore, the thruster algorithms were not correctly updated to match the new center of mass. The thrusters fired and increased even more the roll of the spacecraft.

Consequently, at 1:31-1:53, the spacecraft began shedding at least 10 pieces. Likely pieces are the extended optical bench and the outer portions of the solar panels.

Sources: press briefing handout, spaceflight101

[Scotius]

So, basically Hitomi lost the orientation in space and rotated itself to death? What an ignoble end

[Camacha]

Why does a certain movie clip of a washing machine spring to mind?

[PB666]

3 hours ago, Frida Space said:

Okay, we have a lot more info now. A preliminary investigation into the failure yielded a chain of events that were likely responsible for the event.

I won't repeat all the times and dates of the various events, communication windows etc, because they remain largely unchanged from the ones posted above. The only major change is that the 6-second radio signal received at Santiago at 1533 UT on March 29th wasn't actually from Hitomi, or what was left of it at that time anyway.

The light curve of the main piece of debris (presumably the main spacecraft bus) suggests a rotation period of 5.2 seconds.

The telescope started an attitude manoeuvre to switch from the Crab nebula to Markarian 205. The manoeuvre took place from 18:01 to 18:22 UTC on March 25th. At around 19:10, the Inertial Reference Unit (IRU) detected a non-existent roll about the Z-axis. Usually, the star tracker data has the priority over the IRU data, which in this case was faulty, but star trackes were not being used at the time of the failure.

The IRU data indicated a 21.7 degree/hour roll. The flight control system tried to counter the roll by entering a roll of 20 degree per hour in the opposite direction. The torque was induced by the reaction wheels.

The momentum continued to build up from 20:49 to 01:04 UTC. The magnetic torquers attempted to dump momentum, but were ineffective. At 112 newton-meter-second, the spacecraft was near its saturation momentum of 120 Nms. With the IRU faulty data still showing a large roll rate, Hitomi switched to Safe Mode to use its sun sensor to dermine the solar vector and thus the attitude. The system should have then used the thrusters to point the solar arrays to the sun.

At this point, another error was introduced. The software uploaded after the deployment of the Extensible Optical Bench prevented the sun sensor from working correctly. Furthermore, the thruster algorithms were not correctly updated to match the new center of mass. The thrusters fired and increased even more the roll of the spacecraft.

Consequently, at 1:31-1:53, the spacecraft began shedding at least 10 pieces. Likely pieces are the extended optical bench and the outer portions of the solar panels.

Sources: press briefing handout, spaceflight101

Gyro malfunction. What about backup gyro, XY inertial censor. Or how about a 'things aren't getting any better, just stop and try to communicate with home plate'.

I don't think I would want to be a software engineer on that project, I'de be looking at my accrued vacation time before i did programmed this:

Quote

The momentum continued to build up from 20:49 to 01:04 UTC. The magnetic torquers attempted to dump momentum, but were ineffective. At 112 newton-meter-second, the spacecraft was near its saturation momentum of 120 Nms. With the IRU faulty data still showing a large roll rate, Hitomi switched to Safe Mode to use its sun sensor to dermine the solar vector and thus the attitude. The system should have then used the thrusters to point the solar arrays to the sun.

During a Z axis roll a third of the thrusters become useless. The best thing to do is to reverse what was done step by step and see if the inertial sensors calm down. If you are adding angular momentum around the Y axis and its not responding stop, reverse, if reversing has no effect either change plan.

Theres a saying in electronics that one of our engineers once told me, when your high-tech stuff begins to fail, look for a low tech work around (as he put it everything inh is house is repaired old 60's and 70's equipment) - euphemistically known as 'stone knives and bear skins'. Some of the new electronics tend to have tighter tolerances and shorter lifespans than the old stuff. They work great as long as temperature remain inside a certain range, not so good if you roll them into a sun exposure with no radiator potential. Even some of the new-production 80386 knockoffs will overheat if they cannot radiate off some of their heat. There are probably redundant sensors on the space craft that can act as angular gyros (with a little calculus).

1. Basic data (alternative sources)
a. are the solar panels open, true / false
b. if true then are the solar panes pulsing amperage in an unexpected light cycle true/false
c. look at panels, are panels actively tracking, (check amperage to tracking motors) - true/false
d1. if false confirm z-axis stable (see 2), true/false.
d2. if true then roll sensor is correct other sensors have malfunctioned. Find out which sensors.

2. Asses control.
a. make a controlled turn on y axis of 90' using reaction wheels, panel trackers should turn (amp up) then stabilize (amp zero). light should be constant on at least one panel. Roll Y back to original, confirm.
b. if z-axis was in roll then panels would track, if panels did not cyclically track after the roll it means no axes were rolling.
c.  block sensor.
d. if d2 above, a rolling 90's on y should take more energy if craft was rolling Z

3. Assess attitude, stable?
a. using panel trakers from Y-roll and Z-axis positions to determine the Z-axis relative to the sun.
b. get an attitude check from USAF
c. rotate 90' on any axis (avoid pointing the scope at the sun)
d. get another attitude check from USAF
e. rotate 90' on another axis (avoid pointing the scope at the sun)
f. get another attitude check from USAF
g. rotate back to original position.
h. do some astronomy, find out where target is roll to that approximate position. confirm star-tracker agreement, true/false.
i. if there is small disagreement thank your stars that you did not roll your craft into 8 pieces.

4. Repair
a. Call NASA have them reinstitute the shuttle program and request a repair. (heh-heh)
b. after they laugh you off the phone, call Elon and beg for a proof of abilitiy mission (free) so he can put his expertise to work on your repair. Maybe haul your scope to the ISS and drop it off for a repair.
c. after that fails. limp around like Kepler K2.

 

 

 

 

 

[Glaran K'erman]

Ahhhh, that's definitely very little bueno. Seems like a gut-punch to lose it that way.

[fredinno]

11 hours ago, Frida Space said:

Okay, we have a lot more info now. A preliminary investigation into the failure yielded a chain of events that were likely responsible for the event.

I won't repeat all the times and dates of the various events, communication windows etc, because they remain largely unchanged from the ones posted above. The only major change is that the 6-second radio signal received at Santiago at 1533 UT on March 29th wasn't actually from Hitomi, or what was left of it at that time anyway.

The light curve of the main piece of debris (presumably the main spacecraft bus) suggests a rotation period of 5.2 seconds.

The telescope started an attitude manoeuvre to switch from the Crab nebula to Markarian 205. The manoeuvre took place from 18:01 to 18:22 UTC on March 25th. At around 19:10, the Inertial Reference Unit (IRU) detected a non-existent roll about the Z-axis. Usually, the star tracker data has the priority over the IRU data, which in this case was faulty, but star trackes were not being used at the time of the failure.

The IRU data indicated a 21.7 degree/hour roll. The flight control system tried to counter the roll by entering a roll of 20 degree per hour in the opposite direction. The torque was induced by the reaction wheels.

The momentum continued to build up from 20:49 to 01:04 UTC. The magnetic torquers attempted to dump momentum, but were ineffective. At 112 newton-meter-second, the spacecraft was near its saturation momentum of 120 Nms. With the IRU faulty data still showing a large roll rate, Hitomi switched to Safe Mode to use its sun sensor to dermine the solar vector and thus the attitude. The system should have then used the thrusters to point the solar arrays to the sun.

At this point, another error was introduced. The software uploaded after the deployment of the Extensible Optical Bench prevented the sun sensor from working correctly. Furthermore, the thruster algorithms were not correctly updated to match the new center of mass. The thrusters fired and increased even more the roll of the spacecraft.

Consequently, at 1:31-1:53, the spacecraft began shedding at least 10 pieces. Likely pieces are the extended optical bench and the outer portions of the solar panels.

Sources: press briefing handout, spaceflight101

So, is ASTRO-H dead now?

7 hours ago, PB666 said:

c. after that fails. limp around like Kepler K2.

 

 

 

 

 

I don't think that ASTRO-H will be able to limp around, one of its panels fell off, and probably a lot else.

[Frida Space]

2 hours ago, fredinno said:

So, is ASTRO-H dead now?

I don't think that ASTRO-H will be able to limp around, one of its panels fell off, and probably a lot else.

Apparently there haven't been any new radio contacts, so I'm assuming so. JAXA said it will keep trying to salvage the mission for at least a few months. I don't want to call it dead, I'd rather refer to it as an 11-pieced, non-transmitting satellite

[Camacha]

1 hour ago, Frida Space said:

Apparently there haven't been any new radio contacts, so I'm assuming so. JAXA said it will keep trying to salvage the mission for at least a few months. I don't want to call it dead, I'd rather refer to it as an 11-pieced, non-transmitting satellite

I guess we should rename the ASTRO-H to "The Some Assembly Required 2000".

 

10 hours ago, PB666 said:

[long post]

That is a lot of could haves and should haves. In hindsight, most disasters can be prevented with some minor care and precautions. It is always the little things that add up.

 

[PB666]

18 hours ago, fredinno said:

So, is ASTRO-H dead now?

I don't think that ASTRO-H will be able to limp around, one of its panels fell off, and probably a lot else.

Read carefully, before they did the spin up they should have stopped and evaluated their sensors, after they did 2 sessions of 20h spin up where they are at 118/120 tolerable, they are doomed.

Had they looked at the solar panels before allowing the spin up they would have seen what was going on.

ASTRO-H is in 8 pieces, it may still communicate, but its function is ended, its >8 pieces of unwanted space junk.

 

[PB666]

15 hours ago, Camacha said:

I guess we should rename the ASTRO-H to "The Some Assembly Required 2000".

 

That is a lot of could haves and should haves. In hindsight, most disasters can be prevented with some minor care and precautions. It is always the little things that add up.

 

If you walk into a mine field, the best thing is not to start jumping around to find out where the mines are. Turn around and step your previous steps backwards until you are out.

[Frida Space]

3 hours ago, PB666 said:

Read carefully, before they did the spin up they should have stopped and evaluated their sensors, after they did 2 sessions of 20h spin up where they are at 118/120 tolerable, they are doomed.

Had they looked at the solar panels before allowing the spin up they would have seen what was going on.

ASTRO-H is in 8 pieces, it may still communicate, but its function is ended, its >8 pieces of unwanted space junk.

True, but it was the computer who made the mess. At the time of the anomaly, the satellite wasn't communicating with Earth, so the only thing they could have done to prevent the anomaly (given what we know so far) is have uploaded correct software for the IRU and the thrusters and activated the star trackers.

Edited April 18, 2016 by Frida Space