[time-nuts] Space mission comes to an end becuase of a "computer time tagging" problem

Jim Lux jimlux at earthlink.net
Sun Sep 22 18:57:07 EDT 2013


On 9/22/13 10:07 AM, Bill Hawkins wrote:
> Well, since this thread has recurred and there seem to be people with
> good knowledge here, I've got to ask how a spacecraft that has lost
> its ability to aim antennas at Earth or align solar panels with the
> sun could possibly be diagnosed with having a time tagging problem?

Most likely because the telemetry (which is time tagged in several ways) 
didn't make sense.  If you got frames down and the "spacecraft time" 
wasn't monotonically increasing, you're pretty sure you've had a reset.

For that matter, there's a sequence number in the transfer frames, and 
if that reset, you'd see it.

  I don't know what the format of the DI telemetry frame is (it's 
probably published.. in general, telemetry isn't export controlled) but 
a typical scheme has the SCLK (spacecraft clock) in every frame, or 
every N frames.  The question is whether they have all the bits of sclk 
all the time, or just the LSBs all the time, and the MSBs occasionally. 
  DI is a fairly old design, so the telemetry format probably isn't very 
complex (e.g. it's not like they have dynamically allocated data fields, 
etc.)

> Jim Lux pretty well demolished the idea that time had anything to do
> with it, IMHO.
>
> Seems to me that the DSN doesn't listen to Deep Impact continuously
> because it has other spacecraft to track.

Yes.  There's probably 50 spacecraft that DSN is tracking.  As I recall, 
they track Voyager once a week or maybe every other week.  There's 
actually a published schedule on a regularly updated website.


So there's no growing
> anomaly to indicate a future problem. One time communication is
> fine, the next time there is no answer. Pretty difficult to diagnose
> a problem from those symptoms.

Well.. if the s/c is in safe mode, it's sending data at something like 
10bps or 7-8 bps.  And that data will have a sequence number or clock 
time in it, and you might be able to infer what's going on from that.

Or, they have a "trickle back" scheme that sends a few words of memory 
in every frame, so that after enough frames, you have an image of memory.

>
> Used silver cell batteries for an upper atmosphere density probe in
> 1958. They're still around, but not suitable for long missions. What
> kind of battery (not RTG) would a deep space probe use?

for rechargeable?

NiCd (in the past)
Lithium Ion (now)

A lot also depends on temperature range expected.  Most orbiters even 
around Mars, stay nice and toasty warm without much work, as long as the 
thermal design was decent.  A shiny metal ball in orbit around Mars will 
get quite warm.  I don't know if it would get to 60-78C, but it 
definitely will get to 30C. It's all about alpha/epsilon ratios.

But different battery chemistry has different temperature range 
tolerances.  And then, there's the whole "will it have charge after it 
thaws" question.  I have some supercaps that I've been meaning to freeze 
to try it out (or at least to see if they'll survive)



More information about the time-nuts mailing list