New Horizons Hibernation and Cruise to Pluto Options

[nprev]

Actually, Gordon, the odds are that you're probably right about it being software. It really is impossible to test every possible case (or even foresee them all), so hopefully this is just an easy to find & fix bug.

[stevesliva]

I do hope this turns out to be a weird software issue rather than a hardware problem even though extensive ground testing probably makes software bugs less likely.

I hope it's neither. I hope it was an uncorrectable, transient, radiation-induced error. What will be most interesting is how deep the safe mode is, and how graceful the failure is.

[ugordan]

I hope it was an uncorrectable, transient, radiation-induced error.

That wouldn't be too good, either. It would mean the spacecraft is too sensitive to random, unpredictable cosmic-ray events even while in the state of hibernation when most of the electronics are powered off. Imagine a trip like that during Pluto closest approach. I'd actually put those kinds of faults in the hardware "fault" category.

[stevesliva]

That wouldn't be too good, either. It would mean the spacecraft is too sensitive to random, unpredictable cosmic-ray events even while in the state of hibernation when most of the electronics are powered off. Imagine a trip like that during Pluto closest approach. I'd actually put those kinds of faults in the hardware "fault" category.

I think it was reading the GP-B updates with all the talk of multi-bit errors that makes me think they're something you have to live with... GP-B dealt with them far more often than 1 in 2 years.

i.e. http://einstein.stanford.edu/highlights/hl_040105.html
"Investigations into the root causes of the recent spate of MBEs suggests the possibility that the scrub routine which is checking the memory cells of the on-board Command & Control Computer Assembly (CCCA) could trigger a safemode response on a single MBE if that error occurs in certain locations. As a result, the team has re-programmed the safemode response for the MBE test to automatically stop the mission timeline, rather than rebooting the computer."

[ugordan]

Oh, they're something you have to live with, alright. Statistics are hell, but it just doesn't seem right for the spacecraft to run into a red alert mode only some 2 months (?) into hibernation. Statistically, that can mean anything from an unlucky fluke to a spacecraft that has to be babysitted every month. It could be a combination of the two for all we know, a cosmic ray trip triggers a software corrective action that's buggy and you wind up in a reset.

Take Cassini for comparison - it has a couple of trips of a certain power-related circuit (I forget) per year due to cosmic rays and that's with a spacecraft doing stuff basically 24/7. The fact NH had a mostly trouble-free flight up until hibernation does seem to suggest it's in some way related to that mode, but all this is wild speculation. Hopefully, Alan and co. will find out the real sequence of events soon enough.

[Alan Stern]

Status: Recovering. DSN Allocating Additional Shifts to Assist in Recovery.

Root Cause of C&DH Reset: Remains TBD.

Team: Working Long Hours; Need to Pace.

Prospect: Back in Hibernation in <1 Week.

[nprev]

Thanks for the update, Alan. Best of luck to the team, and sympathy for the long, long hours...please tell them we REALLY appreciate their efforts!!!

[ChrisP]

Thanks for the update Alan. I've lived through all the first visits to the Moon and other planets and I do so want New Horizons to make it to Pluto. Best of luck to you all!

Fingers Crossed,
Chris.

[ElkGroveDan]

Prospect: Back in Hibernation in <1 Week.

You or the spacecraft?

[Alan Stern]

Recovery complete! New Horizons is back to nominal hibernation.

[djellison]

Wake up, scratch, roll over, go back to sleep.

We've all been there.

Thanks for the updates Alan.

Doug

[nprev]

Excellent. Alan, if you get a chance, could you please tell us what seemed to be the most probable cause of this event? Interesting.

[Alan Stern]

Nprev-- Forensics continue down at the 1's and 0's level to learn why the reboot occurred. Attention is focusing on a possible (and subtle) C&DH software interaction with autonomy rules.

-Alan

[nprev]

Yep; that's exactly what I meant earlier about it not being possible to test every case when it comes to software. Thanks, Alan!

[ed_lomeli]

Hello Alan,

I read somewhere in a response to the media that the code for the Pluto encounter will be designed and written this year and in '08. Then, in '09, there would be a first rehearsal. Does this mean a major wakeup in '09? Any details of what that will entail? Thanks.