Huygens News Thread, News as and when we find it Options

[djellison]

Nasa TV will be starting coverage in about 3 mins - but I'm watching multiple TV channels to see if any carry coverage - and will post any news thru the day as it happens

Doug

[bobik]

15 years after the event, what do we actually know about the root causes of the loss of Channel A. As far as I know, the results of the investigation announced at that time were never openly released.

Incredibly, this still seems to be one of the most detailed accounts of the incident available in the public domain, many questions remain unanswered:

Arriving at the ESOC main control room full of cautiously hopeful scientists monitoring the incoming data, one couldn’t help notice the unexpectedly stern expressions on the faces of the DWE co-investigators. The cause of their uneasiness was one single bit in the data stream indicating the status of the DWE ultra-stable oscillator in the Channel A receiver. In contrast to all previous checkouts and tests, that bit was not the expected "1" (power on), but rather an incredible "0" (off). Even worse, the ultra-stable oscillator had been intentionally "selected". This had been enabled by a separate toggle switch to designate the DWE unit, rather than the powered internal oscillator, as the receiver’s active local oscillator. Without power to its local oscillator, the receiver was incapable of locking onto the incoming signal from Huygens. All of the Channel A data, including the DWE Doppler measurements, about 350 images from the Descent Imager/Spectral Radiometer (Huygens camera), and smaller pieces of data from other investigations, were lost. As later determined, the command to switch on the ultra-stable oscillator had been omitted from the critical command sequence for the probe relay event. Channel B, which functioned perfectly from beginning to end, saved the Huygens mission. On the Huygens probe side, thankfully, the matching DWE ultra-stable oscillator designated to drive the Channel A radio link was not forgotten. It was selected and powered. Indeed, it performed flawlessly and provided the ultra-stable radio signal necessary for the Earth-based detections.

[mcaplinger]

15 years after the event, what do we actually know about the root causes of the loss of Channel A.

http://emits.sso.esa.int/emits-doc/ESTEC/A...l-Functions.pdf

It's not a big mystery, it was just human error, no doubt exacerbated by the complex cross-organizational, international nature of the mission and the fact that the commanding was somewhat arcane and error-prone.

[rlorenz]

it was just human error, no doubt exacerbated by the complex cross-organizational, international nature of the mission and the fact that the commanding was somewhat arcane and error-prone.

Yes, but Bobik is right, there should have been a more complete and open description of the circumstances and procedures to avoid similar occurrences in future (and IIRC David Southwood, ESA D-SCI at the time said in front of the cameras that the investigation would be published, but it never was). There are a couple of SpaceOPS papers by some of the relevant individuals 2005-2006, but these remarkably fail to mention the commanding error at all. One does note some useful background, setting the context for the missing command, but doesnt discuss the failure as such :

"The Probe Relay critical sequence only required 38 commands to set the proper spacecraft configuration and
enforce the required tracking attitude. However, in order to provide the necessary critical sequence infrastructure
and provide enforcement of the spacecraft state in response to a fault, 107 more commands were added. The
distinction of the Probe Relay critical sequence is its integration of changes to various fault protection algorithms to
achieve full autonomy and complete its objectives even in a fault scenario. The critical sequence was designed with
a “mark and rollback” logic that supported these objectives. This capability allowed the critical sequence to interact
with fault protection and adapt to the different hardware configurations and states deemed necessary by fault
protection."

(Allestad et al., Systems Overview of the Cassini-Huygens Probe Relay Critical Sequence, AIAA 2005-6388)


The guy whose job it was to assemble the PSA commands once told me it was 'headslap' moment, as soon as the telemetry started coming in, he knew what had gone wrong. And on the documentary 'Destination Titan', you can hear Robin Dutta-Roy of the Doppler Wind Experiment asking on the voicenet 'what time was the RUSO to be powered on', so the DWE team (hurt most by the failure) knew pretty quickly too.

As you note above, though, there were lots of interfaces and steps here. When there are singular events like this, it is not always possible to test things in the configuration they are going to be in. So it's easy for such things to be missed, which is why such failures should be documented for the benefit of all. At this point, there's no need for ESA to be bashful, no careers are at stake at this point, Huygens was a success and will always be considered such, warts and all.

Ralph