I take your point, and Ten's.
But what about the things that can go wrong after launch that the designers
can't control?
Obviously things like the possibility your probe might be destroyed by, say,
a meteor are risks you have to take?
However, as cool as the work is, it can't be easy to put your heart and soul
into a project when, for example, you know there are no gaurantees that
funding will be available to process (or even collect) data sent back when
(and if) the probe reaches its destination.
I tend to look more at the longer term issues than many people (in my
experience) do. If I was a rocket scientist and was asked to bust my gut
working say 60 to 80 hours a week to get a probe ready by a particular
deadline, I'd like a little certainty that when it arrives at its
destination the information it gathers will be collected and analysed. If
sufficient certainty is not there I'd rather not build the probe until I had
developed the technology to get it to its destination while I was still
around to ensure my time was not wasted..
Katipo
Well, I can include a few personal observations and opinions below. Consider
this from the perspective of an engineer at JPL, where I was employed at the
time. I no longer work there.
>I take your point, and Ten's.
>
>But what about the things that can go wrong after launch that the designers
>can't control?
>Obviously things like the possibility your probe might be destroyed by, say,
>a meteor are risks you have to take?
Of course, there are no sure things. Getting hit by an asteroid. Or as
Henry has pointed out, there were thruster issues. Or any one of a
bajillion things can go wrong. But that doesn't mean one never leaves the
harbor. For spacecraft systems the design rule was no single point
failures. That gives one some margin for a random failure but not in
general a systemic or common mode failure. I don't believe the instruments
had full redundancy. Now, is dual redundant with cross strapping good
enough? It depends is the only answer you can come up with, how much you
are willing to pay, how you trade off spacecraft vs. science, and how much
risk is acceptable. Of course, that is a high level policy decision. For
ths space shuttle (and this is sci.space.shuttle) they have a higher level
of fault tolerance. How much is enough?
>However, as cool as the work is, it can't be easy to put your heart and soul
>into a project when, for example, you know there are no gaurantees that
>funding will be available to process (or even collect) data sent back when
>(and if) the probe reaches its destination.
I never regarded that as a factor.
Since there are no guarantees for any of these missions it would not make
sense to require a guarantee of some sort. For example, consider just the
risk in the launch vehicle.
>I tend to look more at the longer term issues than many people (in my
>experience) do. If I was a rocket scientist and was asked to bust my gut
[quoted text clipped - 4 lines]
>developed the technology to get it to its destination while I was still
>around to ensure my time was not wasted..
Again, no guarantees, particularly in exploration misssions such as this.
I asked above how much is enough. Do you put two computers on? Or three?
Or four? Or five? It simply doesn't matter, you will never reduce the risk
to the point of having a guarantee. Consider the control electronics for a
digital fly-by-wire aircraft such as some Airbus models or the Boeing 777.
There are no guarantees and they design to a failure rate that is deemed
"acceptable risk" for the flying public.
>Katipo
>
[quoted text clipped - 16 lines]
>>>Katipo
>

Signature
rk, Just an OldEngineer
"The number of people having any connection with the project must be
restricted in an almost vicious manner. Use a small number of good people."
-- Kelly Johnson in Skunk Works