OT: Failure to communicate? Genesis

[Paul Hovnanian P.E.]

Product developer wrote:

"Paul Hovnanian P.E." <Paul@Hovnanian.com> wrote in message news:<4179ECE7.34D4099F@Hovnanian.com>...
Tim Shoppa wrote:

Fred Bloggs <nospam@nospam.com> wrote in message news:<4170B00B.2090206@nospam.com>...
That package isn't symmetric- only one end is flanged. They don't show
the circuit board- but it looks like a simple component outline in
silkscreen should have minimized the chances of reverse assembly.

Pieced together from several sources, some of them not yet in the media:

* It wasn't an assembly issue, it was actually designed incorrectly. The part
was installed "as designed".

* The subassembly was tested, but only on a sine-wave shake table.
Evidently they verified that the output of the accelerometer was a sine
wave but they didn't check the polarity...

That makes sense. If the accelerometer was a linear device, mounting it
backwards would make no difference. It would put out a signal of the
opposite polarity, but since mounting it backwards would reverse the
acceleration vector, the two effects would cancel out. If so, there's no
need to key the sensor to ensure 'correct' mounting.

One has to wonder why they used an axial leaded part like this in such
a mission critical function. I would have specifed an Endevco Piezo
device that is hard mounted. No one has more experience in shock,
vibration, and acceration in this arena than Endevco. They are the
recognized pioneers of accelerometers going back to the early 60's.

If the device is linear for both positive and negative acelerations,
mounting it backwards won't make a difference. That's why a desgin error
seems to make sense.

--
Paul Hovnanian mailtoaul@Hovnanian.com
------------------------------------------------------------------
Don't hate yourself in the morning -- sleep till noon.

 

[Rich Grise]

On Sat, 23 Oct 2004 19:41:06 -0700, Tim Shoppa wrote:

jdurban@vorel.com (Product developer) wrote in message news:<118afaeb.0410231052.67a69b1b@posting.google.com>...
One has to wonder why they used an axial leaded part like this in such
a mission critical function. I would have specifed an Endevco Piezo
device that is hard mounted. No one has more experience in shock,
vibration, and acceration in this arena than Endevco. They are the
recognized pioneers of accelerometers going back to the early 60's.

What I've read indicates that components were chosen exactly to match
what's used in nuclear bombs to prevent them from accidentally blowing
up in storage or on a plane. One component is an accelerometer to make
sure that the bomb had really been dropped or launched or was about to
hit. (This is in addition to all the fusing etc. stuff they must have).
A returning space probe probably sees accelerations very similar to a
end-of-flight ICBM, after all.

There are lots of moralistic arguments about nuclear testing etc., but
from a stewardship standpoint, as long as you've got the bombs you've got
to make sure they're safe. What always bugged me about ICBM's is that
while we'd launched ICBM's with dummy warheads at ourselves just to make
sure they could navigate, we never did it with a real warhead that
was expected to go boom at the end, and we never tried flying them over the
north pole. I'm guessing that we did do it with dummy warheads that had
maybe everything *except* the fissile fun stuff.

I wonder if the A-bomb guys are going back over decades-old drawings
and test procedures since the revelation of the Genesis boo-boo. After
all, if it was copied verbatim from an ICBM, maybe...

Maybe after Kerry wins the election, George the Loony will decide to

go out in a blaze of "If I can't have it, nobody can" glory, and push the
button. America's nuclear arsenal is launched, the missiles take their
majestic arching ballistic flight, and ten thousand multiple independently
targetable reentry vehicles go "plop." ;-)

Cheers!
Rich

 

[David Lesher]

"Paul Hovnanian P.E." <Paul@Hovnanian.com> writes:

They're G switches for goodness sakes. They *can't* be one-time deals.
What do you think they are subjected to during launch? You get three guesses
and the first two don't count.

I was wondering about how many of these kind of components would survive
the UPS guy dropping the package, never mind the launch.

http://groups.google.com/groups?q=Wanttaja+accelerometer&hl=en&lr=&ie=UTF-8&selm=airliners.1996.277%40ohare.Chicago.COM&rnum=1

--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433