on Thursday, January 22nd, Jon said

It shouldn't be a power fault, since the radio is still on transmitting, just it's transmitting pseudo-random noise, meaning the processor isn't currently sending data to the radio component.

Specifically, then it's likely that some poor architectural design (mostly including not enough redundancy) has led to the failure. There should be a complete offline component archive that has an independent radio, so if the system fails, the replacement can be kicked into life via a radio signal. The extra weight is insignificant.

If the system memory is corrupted, or the software is corrupted, where is the backup software, and where are the redundant memory banks.

If the command interruption caused the fault, why are commands not atomic.

This problem was unbelievably avoidable (assuming it hasn't been hijacked by Martians).

Jon

on Thursday, January 22nd, Phil said

I think the most believable explanation is the Martians.

One thing that has yet to be explained to me is why exactly must the mission expire in 90 days? Why should the progressive accumulation of dust on the solar panels lead to the batteries being unable to replenish themselves? Why let that happen? Why not wipe the dust off? Naive as it may sound, I have yet to come across an answer.

So I think like you, Jon, that some people have been a little overly optimistic about things here, and redundancy may contain a few syllables too many for some of our dear transatlantic friends...

But anyway, anybody posting good news here should the problem be solved will find himself invited by me to generous drinks!

on Tuesday, March 9th, Phil said

Mission duration of 90 days was a conservative estimate. New data suggests much longer likely duration due to more favourable environment: source news.bbc.co.uk

http://news.bbc.co.uk/2/hi/science/nature/3545747.stm