KickSat Sprites aboard CRS 3

[Prober]

bump this info from another thread...

Bad news about Kicksat.  The timer for the Sprite deployment reset on 30th April, apparently due to a cosmic ray hit.  This means that that deployment is postponed from tomorrow to May 16th.  The current atmospheric decay estimate for the satellite is May 14th.
https://www.kickstarter.com/projects/zacinaction/kicksat-your-personal-spacecraft-in-space/posts

[Bob Shaw]

Latest news on KickSat at 15:33 BST today (and, most importantly, *my* satellite) - and it's not entirely good: the clock has been reset aboard the KickSat carrier and the whole shebang *may* re-enter before the Sprites are ejected. Poo!

"Project Update #72: KickSat's Current Status
Posted by Zachary Manchester

Hi Everyone,

First off, I'd like to sincerely thank all of you for your support over the past two years. KickSat has been a success up to this point because of you.

As those who've been keeping up with the telemetry data coming in from KickSat on our mailing list may have noticed, the packets we've been receiving have changed in the last couple of days. This was due to a hard reset of the "watchdog" microcontroller on KickSat - the sort of "reptile brain" of the satellite that manages turning on and off the rest of the subsystems and keeps the master clock. It appears the reset happened some time in the morning of Wednesday, April 30th. The reset doesn't seem to be the result of power issues (the watchdog should run until the batteries reach 5.5 volts, and they've been holding steady around 6.5 volts). Instead, it seems the likely culprit was radiation.

One consequence of the watchdog reset on KickSat is that the spacecraft's master clock was reset, thus also setting the deployment countdown for KickSat back to 16 days. That would put the deployment some time in the morning of May 16th. Unfortunately, it looks like KickSat will most likely reenter and burn up before the 16th. We've spent the last couple of days here at Cornell trying to think of every possible contingency, but it seems there aren't very many options right now. KickSat's uplink radio, which we could use to command the deployment, can't turn on unless the batteries reach 8 volts, and it doesn't look like they'll reach that level in time.

While the situation looks a little bleak, there is still some hope that the batteries may recharge sufficiently to command the satellite. There is also a small chance that KickSat could remain in orbit until the 16th, at which point the timer would set off the deployment as originally planned. We'll continue tracking KickSat over the next few days with the help of the ham community, so that we can keep track of its battery voltage and the Sprite deployment status. I'll post updates here, as usual, but you can also see the latest data as it comes in on our mailing list.

Thank you again for your support. I promise that this won't be the end of the KickSat project.

- Zac"

[Elmar Moelzer]

Man, this sucks! Radiation is a bitch!
I hope that they can get it to work! Best of luck!

[mlindner]

Man, this sucks! Radiation is a bitch!
I hope that they can get it to work! Best of luck!

Happens on cubesats all the time. We had a watchdog timer to reset the spacecraft, a watchdog timer to reset the watchdog timer and a backup processor inside the radio that had rudimentary control of the spacecraft power bus. We regularly saw the spacecraft reset from radiation hits, just a fact of life of cubesats. You need to design the system with constant resets in mind.

Really unfortunate about this though.

[Bob Shaw]

Mmmm. It does seem to be an unfortunate design choice. I'm reminded of the Soviet spacecraft of the 1960s, and the way that so many were nobbled by timers. In fact, not just the 1960s, now that I think about it. And not just the Soviets...

Timers for 'Help! I can't communicate!' backup seem perfectly sensible, and could provide a straightforward form of insurance in a situation where you're already dealing with an unhappy spacecraft, but an unprogramable timer which is fully responsible for mission success and a spacecraft design which has rendered said timer unprogramable does seem like a design flaw.

I was recently able to have a good look at (and handle) ClydeSpace CubeSat components (battery packs and solar panels) and was *very* impressed with their general quality and detailed design; I hope to speak to them again, and will ask about CubeSat design methodology when I do. Zac Manchester's KickSat was, however, built at a small fraction of the cost of a ClydeSpace spacecraft, and the available level of funding may well have been reflected in a series of necessary compromises...

Here are some links:

http://www.spacecraftresearch.com/blog/

http://www.clyde-space.com

[Prober]

Man, this sucks! Radiation is a bitch!
I hope that they can get it to work! Best of luck!

that's why I asked about Radiation.

http://forum.nasaspaceflight.com/index.php?topic=34414.msg1182300#msg1182300

[Bob Shaw]

Man, this sucks! Radiation is a bitch!
I hope that they can get it to work! Best of luck!

that's why I asked about Radiation.

http://forum.nasaspaceflight.com/index.php?topic=34414.msg1182300#msg1182300

Ah, it was *your* fault, was it...

[mlindner]

Man, this sucks! Radiation is a bitch!
I hope that they can get it to work! Best of luck!

that's why I asked about Radiation.

http://forum.nasaspaceflight.com/index.php?topic=34414.msg1182300#msg1182300

Radiation hardening is not needed on cubesats, nor should it be used. It's out of the scope of their design IMO. There is an interplanetary cubesat coming, that may be the first cubesat that may need radiation hardening.

Given that, there is part selection you can do among COTS things that allows for natural radiation hardening. Like not using SD Cards on orbit (these failed constantly for us) and using FRAM based processors.

Zac Manchester's KickSat was, however, built at a small fraction of the cost of a ClydeSpace spacecraft, and the available level of funding may well have been reflected in a series of necessary compromises...

Building them with unpaid undergrad and graduate students worked for us. (I was one of them. Was paid during the summer though.) I seem to remember our 3U cubesat was somewhere around $45k in parts including engineering units and development units (not counting labor).

[Bob Shaw]

I'd heard of FRAM memory, but didn't know it was used in microprocessors themselves - it's interesting to also note the longevity of PowerPC CPUs in space applications, long after Apple gave up on them!

[CuddlyRocket]

Radiation hardening is not needed on cubesats, nor should it be used. It's out of the scope of their design IMO. There is an interplanetary cubesat coming, that may be the first cubesat that may need radiation hardening.

Given that, there is part selection you can do among COTS things that allows for natural radiation hardening. Like not using SD Cards on orbit (these failed constantly for us) and using FRAM based processors.

The other alternative - like SpaceX uses on Dragon - is redundancy including enabling resetting; though that brings its own obvious problems in such small satellites.

[Bob Shaw]

The way I read it, the ability to do a reset is there, but the battery power required to get the signal isn't.

[Llian Rhydderch]

Radiation hardening is not needed on cubesats, nor should it be used. It's out of the scope of their design IMO. There is an interplanetary cubesat coming, that may be the first cubesat that may need radiation hardening.

Given that, there is part selection you can do among COTS things that allows for natural radiation hardening. Like not using SD Cards on orbit (these failed constantly for us) and using FRAM based processors.

The other alternative - like SpaceX uses on Dragon - is redundancy including enabling resetting; though that brings its own obvious problems in such small satellites.

Correct.  A fault-tolerant architecture can be used rather than the rad-hardened processors that have been typically used.

With continuous improvement in the level of integration in integrated circuits, and dramatically-lowered costs year-on-year for many decades now (see Moore's Law), there is no inherent reason the electronic hardware won't be at a size, and sufficiently integrated for the multiprocessor Byzantine fault-tolerance scheme that SpaceX uses (both on Falcon rockets and on Dragon capsules) to be available in cubesat form-factors and costs in the next decade.

And if you ask, what would drive a chip maker or system-integrator make such a system, I have two answers.  Both provide potential drivers for movement in this direction:

 * with the architecture SpaceX has been using now flight-tested, shown to be reliable, and further data on in-space performance coming in rapidly, it could be that SpaceX and other companies that could use this capability in space applications may help drive the forward progress on the technology.

 * the demand might come from the consumer market.  While fault-tolerant high-availability systems are generally available only in the higher-end "enterprise class" of computing systems (servers, data storage controllers, etc.), at many times the price, the consumer market is large enough to drive change itself.  We could see a future, and possibly a near future, where consumer demand for better quality PCs/tablet/smartphones (fewer resets, need to reboot in the middle of important use of the device, etc.) may incent chip makers to package hardware in more-capable and more highly integrated chips/memory/storage specifically aimed at a consumer fault-tolerant market.  If this happens, expect rapid adoption of the same tech in cubesats, and migration into avionics architectures like SpaceX is using, and other launch providers are certainly free to emulate once market forces take over for the top-down government driven requirements of previous space systems.

[gwiz]

Kicksat has apparently decayed, two days before the Sprites were to be released.

[mlindner]

Kicksat has apparently decayed, two days before the Sprites were to be released.

That's too bad. Hopefully they can do a second one so the funders can have a second shot at their pico-sats. Much of the development cost is already paid for so they just need to re-buy components (along with possibly some minor code/hardware changes to remove this fault mechanism).

[Bob Shaw]

Oh, poo.

KickSat II please, Zac!

[Elmar Moelzer]

Sorry to see the hard work of so many people go down because of this
Hope there will be another try, Zac!

[mlindner]

KickSat is a good example of innovation through cubesats though. Innovate fast, fail fast, innovate again. The mechanism to deploy the kicksats is honestly rather innovative and ingenious. It's not something that could be tried and depended on first time for a larger mission. I hope to see this trend continue and the ideas of cubesats get more adoption through to microsats.

[Llian Rhydderch]

KickSat is a good example of innovation through cubesats though. Innovate fast, fail fast, innovate again. The mechanism to deploy the kicksats is honestly rather innovative and ingenious. It's not something that could be tried and depended on first time for a larger mission. I hope to see this trend continue and the ideas of cubesats get more adoption through to microsats.

Indeed, mlindner, "Innovate fast, fail fast, innovate again." 

I suspect this team, and what re-assembles after the post-mortem review is complete, will learn some good lessons in electronic hardware and embedded software architecture, and that they'll come back with an even better CubeSat offering of more, and more capable, Sprite femtosats on the next go 'round.

[Bob Shaw]

Here's Zac Manchester's message to the KickSat backers:

" Project Update #73: KickSat Has Reentered
Posted by Zachary Manchester

Hi Everyone,

KickSat reentered the atmosphere and burned up last night some time around 9:30 PM EDT (01:30 UTC). Unfortunately, we were not able to command the Sprite deployment in time. While we are certainly disappointed that things did not go as planned, I think we still have a lot to be proud of.

Over 300 people from all over the world came together to make KickSat happen. We built a spacecraft, tested it, and launched it. Hundreds of people had their names flown in space, more than a dozen radio amateurs were able to receive signals from KickSat's beacon radio, and volunteers collected and processed telemetry data and predicted KickSat's orbit and reentry. This kind of participation is exactly what KickSat is all about and I'm glad we all got to share in this experience.

We've learned a lot from KickSat, and I plan to take those lessons and build an even better KickSat-2. This is only the beginning! Thank you all for your amazing support over the past two years. I hope you'll stick with us as we continue to try to make space something everyone can take part in.

- Zac "

[jumpjack]

Did I unerstand correctly?
This "satellite" consists of a bare PCB with some components & a solar panel mounted on it, and that's all?
And it should have survived space environment?
And it costed 30,000 %?

Before sending a circuit to the space... isn't a designer supposed to know something about space? Something like -100°C/+100°C range , ionizing radiation, cosmic rays and kilorads??

Why didn't anybody explained these things to the designer (and to backers) in last 4 years??

I'm astonished.