PBS's "Nova" and MER

Our local PBS station ran a special tonight on the MER spacecraft and
their rovers design, testing, and launch (as well as the successful
landing of #1)
The show was chock-full of info on the rovers and what they can do;Which
if they work right is very impressive- but in my opinion showed a major
flaw in the program timeline that led up to their launch...in the last
few months before the launch, the MER team runs into no less than _5_
unexpected problems, any one of which is a complete show-stopper if not
fixed:
1.) An attempt to use a Pathfinder type parachute for the descent of the
rovers fails after a helicopter drop, as it is not structurally strong
enough to take the extra weight of the MER probes.
2.) A redesigned chute fails to open properly in wind tunnel tests. Its
central hole is too large and this makes it "squid" as it attempts to
open, due to too much airflow out of the top. The test that fixes this
flaw is done by sewing in a constraint around the top hole; and holding
the chute reefed via two guys in the wind tunnel with a cord at the apex
of its shroud lines; as there is neither time nor money to do another
test ejection from its canister. The chute is found to have been built
to the wrong specifications, and a redesigned one has to be made in
short order to correct the flaw.
3.) One of the landing airbags ruptures during a bounce test. Again
unexpected...again fixed with a few months to go.
4.) A failure of a circuit board in both of the rovers to come up to
spec requires that it be replaced _after_ they have been sealed in the
landers; this requires firing the pyros on the landers to open the
petals so that they can be reached.
5.) The possibility that firing the pyros leaves them in an electrically
shorted configuration that will burn out the resistors in their firing
system- so as to make them non-functional on the two spacecraft- has to
be examined...this means digging up the fired and replaced pyros with
no paper trail on them.. they are finally tracked down one-by-one after
a exhaustive search; which is good, as there is no way to replace the
resistors in question inside of the launch window.
This to me STRONGLY suggest that such probes need more time and funding
thrown at them in the future; if even for a few million dollars and a
extra month or so leading up to launch; looking at this, one can
certainly see a distinct downside to BFC in that these were all serious
problems that were being addressed at the last moment via lots of hard
work, sleepless nights, and just plain dumb luck.
I will not even comment on the idea to launch the probes before their
landing software was completed; and then uploading it to them in-flight-
or the extremely young average age (by engineering standards) of the
people involved in the MER program as shown in the special, after Dan
Goldin's scythe cut down all the old pros at NASA.
In another ten years or so, this team will be old pros...but it seems
stupid to have to re-learn how to do interplanetary flight all over again.
We got away with it this time; but this is one hell of a poor way to run
a planetary exploration program, in my opinion.

Pat

Pat Flannery wrote:

This to me STRONGLY suggest that such probes need more time and funding
thrown at them in the future; if even for a few million dollars and a
extra month or so leading up to launch; looking at this, one can
certainly see a distinct downside to BFC in that these were all serious
problems that were being addressed at the last moment via lots of hard
work, sleepless nights, and just plain dumb luck.

That's just the way real space projects go. I've never seen one that
didn't. The technology and missions are so unforgiving that even tiny
problems have to be addressed, and beaten to death. That takes time and
money - there's no getting around it. It should be taken into account
up front. You are correct that cheaping out or funding or schedule, or
simply ignoring this essential fact, is exactly why there are so many
failures in all sorts of aerospace projects.

This is also why I fear the day when the "cheap access to
space"/X-Prize types actually try to launch. It's not just a trade-off,
it's that most "space buff" types involved actively sneer at the idea of
spending the time it actually takes, or don't know enough to be worried
about it. Those "dinosaur" aerospace companies actually know some things
about how to conduct projects that lead to their "exorbitant" bids.
Whenever you ignore the rules, you get in trouble - FIA, MPL, continual
pressure on the shuttle to cheapen up operations, all lead to the sorts
of results that have been demonstrated. Most of the proposed budgets for
the "stunts" like Roton aren't enough to cover even the contingencies
that will arise, much less the entire program. That's why they fail.

Brett

On Mon, 05 Jan 2004 06:34:10 GMT, Brett Buck
wrote:

Pat Flannery wrote:

This to me STRONGLY suggest that such probes need more time and funding
thrown at them in the future; if even for a few million dollars and a
extra month or so leading up to launch; looking at this, one can
certainly see a distinct downside to BFC in that these were all serious
problems that were being addressed at the last moment via lots of hard
work, sleepless nights, and just plain dumb luck.

That's just the way real space projects go. I've never seen one that
didn't. The technology and missions are so unforgiving that even tiny
problems have to be addressed, and beaten to death. That takes time and
money - there's no getting around it. It should be taken into account
up front. You are correct that cheaping out or funding or schedule, or
simply ignoring this essential fact, is exactly why there are so many
failures in all sorts of aerospace projects.

This is also why I fear the day when the "cheap access to
space"/X-Prize types actually try to launch.


I imagine the big companies do too. They'll look pretty stupid if
Rutan succeeds.

Scott Ferrin wrote:


This is also why I fear the day when the "cheap access to
space"/X-Prize types actually try to launch.



I imagine the big companies do too. They'll look pretty stupid if
Rutan succeeds.

Succeeds in winning the X-Prize? Who the hell cares? You guys are
really something. The X-prize is not even on the radar screen to
Lockheed, TRW, or Boeing, and whether or not anyone wins or not, it
won't make a whit of difference to the majority of the industry. It
could have been done in 1960 if anyone had cared to, and you took out
the "no government money" clause. Nobody wanted to, because then as now,
it doesn't really lead anywhere or advance the state of the art in any
way. Far more capable plans were well on their way to succeeding in that
era, but were derailed, once again based on "lack of need" that persists
to this day.

Rutan has a backer with money, I applaud him for taking on the task,
and I expect that eventually he will make the X-prize requirements
(before or after the deadline). Coefficient of relevence to space
technology is negligible.

I will grant that it will make sci.space.policy "true beleivers"
ecstatic, but that's really not much of a goal.

Brett

On Mon, 05 Jan 2004 07:43:54 GMT, Brett Buck
wrote:

Scott Ferrin wrote:


This is also why I fear the day when the "cheap access to
space"/X-Prize types actually try to launch.



I imagine the big companies do too. They'll look pretty stupid if
Rutan succeeds.

Succeeds in winning the X-Prize? Who the hell cares? You guys are
really something.

Maybe you could explain to us why they need the "levelest and flatest
floor in the world" just to line up the friggin' boosters on the Delta
IV. Hell the sections of a drive shaft on a Nimitz class carrier need
to be lined up just as well and they use a CRANE in a shipyard. Then
maybe you could explain why the ATF program was able to produce two
types of fighters (YF-22 and YF-23) both of which had far more new
technology at a FAR cheaper price than this "new technology" little
spaceplane NASA wants to build. The fact of the matter is that NASA
does go way down the diminishing returns curve when comes to
precision, complication, and materials all for CYA and because they're
stuck in this mentality of "we're NASA so it all needs to be gold
plated".




The X-prize is not even on the radar screen to
Lockheed, TRW, or Boeing, and whether or not anyone wins or not, it
won't make a whit of difference to the majority of the industry. It
could have been done in 1960 if anyone had cared to, and you took out
the "no government money" clause. Nobody wanted to, because then as now,
it doesn't really lead anywhere or advance the state of the art in any
way. Far more capable plans were well on their way to succeeding in that
era, but were derailed, once again based on "lack of need" that persists
to this day.


If there is a lack of need then why the X-34, 37, 38, and 40? Why the
tiny spaceplane NASA wants to build? The X-20 didn't get cancelled
because of no need, it was cancelled because of $$$$$.





Rutan has a backer with money, I applaud him for taking on the task,
and I expect that eventually he will make the X-prize requirements
(before or after the deadline). Coefficient of relevence to space
technology is negligible.

A newly designed launch vehicle and manned rocket for far cheaper than
a "lifeboat" that all it needs to do is glide back to earth?
Something of which they've already been doing for DECADES?. First
manned use of hybrid rockets? Totally new way of reducing speed when
returning to earth? Yep, you're right, no relevence whatsoever.





I will grant that it will make sci.space.policy "true beleivers"
ecstatic, but that's really not much of a goal.

Brett

Spoken like a true believer in unnecessary complexity, precision, and
gold plating.

Scott Ferrin wrote:

Maybe you could explain to us why they need the "levelest and flatest
floor in the world" just to line up the friggin' boosters on the Delta
IV. Hell the sections of a drive shaft on a Nimitz class carrier need
to be lined up just as well and they use a CRANE in a shipyard.

Well, you are utterly and completely *wrong*.

The crane is used to install the segments into the hull. Once in the
hull all of the bearings have adjustments to align and center the
drive shaft, and lasers and micrometers are used to ensure that the
alignment is spot-on perfect.

D.
--
The STS-107 Columbia Loss FAQ can be found
at the following URLs:

Text-Only Version:
http://www.io.com/~o_m/columbia_loss_faq.html

Enhanced HTML Version:
http://www.io.com/~o_m/columbia_loss_faq_x.html

Corrections, comments, and additions should be
e-mailed to , as well as posted to
sci.space.history and sci.space.shuttle for
discussion.

Brett Buck wrote:

That takes time and money - there's no getting around it. It should
be taken into account up front. You are correct that cheaping out or
funding or schedule, or simply ignoring this essential fact, is
exactly why there are so many failures in all sorts of aerospace
projects.

About the time they started talking about sending the finished landing
software to the spacecraft _after_ they launched, a feeling of deja-vu
hit me; the first thing that came to mind are all those wonderful
Microsoft products that get released before they are debugged, knowing
that people are expected to promptly start downloading the fixes after
getting them...but another bell went off in regard to that and also the
whole frenzied pace of the last few weeks before launch as shown in the
Nova show...the MER program seemed to be operating under Deming's "Just
In Time" technique of manufacturing; thing were only finished at the
last minute...this may work fine in say the manufacture of automobiles;
where the company has decades of experience building basically the same
machine, and can take a very good guess at how long things will take to
be completed- but to extrapolate that concept to building spacecraft
that are produced in ones and twos is to me an unsound idea.
One of the instruments on the rovers is a infrared imager intended to
observe the IR signature of rocks so that the ones that are possibly
water-related can be picked for closer study by the rover; the program
showed the MER team in a real quandary- they had exactly two of these
instruments, and they wanted to do a stress test on one to assure that
it could take the force of the landing impact...but do they dare do it?
Both of the instruments are to be put on the rovers, and if they break
one, they don't have the time or money to replace it...so maybe its
first all-up test should be the landing itself. That is a horrible
situation to put the team in...zero back-ups. They do test it, and it
doesn't break- but again, is this anywhere near how things should be
done in the future?

Pat

i think we have to cut some slack to the developement for the hasty work
created by the astonomy (once every 26 month mars/earch minimum approach).
It was not a case of FBC, just a deadline that was impossible to work
around.

I was not worried about the youth of the crew: it was unavoidable given the
all at once surge of the 60's (it takes several generations of more even
funding that NASA has ever had to even things out, and organizational
cultural transmission of experience is the best we could hope for. Given the
level of change in technology, this is probably not as important as we
think.

Finally, i worried not at all about loading software after the hardware went
up. Software is just that, soft. They had the device here, software can be
tested. If the hardware is designed correctly, software follows the form. I
design software for a living, and I know who's boss: reality, not me.

bob

In article m,
com says...
Finally, i worried not at all about loading software after the hardware went
up. Software is just that, soft. They had the device here, software can be
tested. If the hardware is designed correctly, software follows the form. I
design software for a living, and I know who's boss: reality, not me.

Dijkstra was perhaps the first to realize that we build computers to
execute our programs, not the other way around. Software is soft in the
sense it is easy to change. Yet experience shows that it is *hard* to
make the software correct. It is usually better to get the software
right and let the hardware guys deal with the implications of that
decision.

Defining the hardware and then leaving all the other issues to be fixed
as "a small matter of programing" is a recipe for disaster. There are
many examples of this...
--
Kevin Willoughby lid

Imagine that, a FROG ON-OFF switch, hardly the work
for test pilots. -- Mike Collins

Pat Flannery wrote:
About the time they started talking about sending the finished landing
software to the spacecraft _after_ they launched, a feeling of deja-vu
hit me;

Perhaps because this isn't all that uncommon for planetary spacecraft.
I believe Magellan did the same thing for one.

- Matt