WEBVTT FILE

00:00:00.000,00:00:04.004
>>KATRINA: One of the Hubble
Space Telescope’s most

00:00:04.004.00:00:06.607
memorable moments was observing
the fragments of Comet

00:00:06.607.00:00:11.712
Shoemaker-Levy 9 impact Jupiter
in July, 1994. This was a huge

00:00:11.712.00:00:14.381
astronomical event that captured
the attention of the public

00:00:14.381.00:00:18.285
across the globe. But what most
people don’t know is that Hubble

00:00:18.285.00:00:21.588
came close to not getting those
observations. The telescope

00:00:21.588.00:00:24.091
encountered some problems that
might normally take over a month

00:00:24.091.00:00:27.661
to solve just days before the
comet impact.

00:00:27.661.00:00:34.067
[music]

00:00:34.067.00:00:38.939
>>HEIDI: Shoemaker-Levy 9 was
a comet that was discovered by

00:00:38.939.00:00:42.609
David Levy, Caroline Shoemaker,
and Gene Shoemaker. We had

00:00:42.609.00:00:46.880
about nine months of warning
that this comet was going to hit

00:00:46.880.00:00:51.551
Jupiter. I knew from the
calculations we had done
beforehand that if

00:00:51.551.00:00:55.522
Hubble was working at the peak
of its game, that we would get

00:00:55.522.00:00:59.559
images that far surpassed images
that could be taken by any other

00:00:59.559.00:01:03.697
telescope on the Earth. >>BRIAN
V: So, so we knew that Comet

00:01:03.697.00:01:07.935
Shoemaker-Levy was going to
impact Jupiter on July 16th, so

00:01:07.935.00:01:10.237
we wanted to make sure that
Hubble was ready for that.

00:01:10.237.00:01:13.907
>>BRIAN R: July 5th, 1994
started out as an ordinary day

00:01:13.907.00:01:16.877
until we were called and told
that the Hubble had been

00:01:16.877.00:01:19.947
commanded into inertial hold
safemode because it appeared

00:01:19.947.00:01:22.950
that it was misbehaving and not
executing the proper commands.

00:01:22.950.00:01:25.319
>>BRIAN V: When they command
inertial hold, all those

00:01:25.319.00:01:29.022
commands that were loaded stop
executing. We tell it to don’t

00:01:29.022.00:01:32.225
look at those any more and just
hold steady. >>HEIDI: You know

00:01:32.225.00:01:34.962
your first response is sort of
an empty feeling in the pit of

00:01:34.962.00:01:39.633
your stomach, like oh no, we
have this huge event coming up,

00:01:39.633.00:01:43.337
and the telescope’s not working.

00:01:43.337.00:01:44.037
>>CHARLIE: We had about a week

00:01:44.037.00:01:48.775
to resolve the vehicle issue and
bring it back to normal mode.

00:01:48.775.00:01:51.044
>>BRIAN R: The spacecraft was
not executing the commands in

00:01:51.044.00:01:53.413
the stored command sequence it
was supposed to be, it looked

00:01:53.413.00:01:56.817
like it was doing something
completely random. So we

00:01:56.817.00:01:59.219
immediately suspected that there
was a memory unit problem.

00:01:59.219.00:02:03.623
>>CHARLIE: We were lucky
because, the previous mission we

00:02:03.623.00:02:08.095
had done we installed a
coprocessor.

00:02:08.095.00:02:09.496
>>HARRY: During the first
servicing mission

00:02:09.496.00:02:12.299
in December of ’93, the
coprocessor

00:02:12.299.00:02:17.170
was basically a memory upgrade,
an additional memory to augment

00:02:17.170.00:02:21.575
the DF-224 flight computer.
>>BRIAN V: Now this coprocessor

00:02:21.575.00:02:27.948
has what we call a shared memory
that both computers can use. So

00:02:27.948.00:02:32.853
the shared memory wasn’t really
configured yet to be useable,

00:02:32.853.00:02:38.225
but we knew that it was there.
>>CHARLIE: We verified different

00:02:38.225.00:02:42.162
configurations and different
architectures that we could test

00:02:42.162.00:02:47.534
out, and we found that it was no
issue to swap out the memory.

00:02:47.534.00:02:50.037
>>BRIAN R: So we started doing
that, and things were going

00:02:50.037.00:02:53.273
extremely well. And suddenly
while we were in the middle of

00:02:53.273.00:02:56.643
reconfiguring it, we were told
that the spacecraft had entered

00:02:56.643.00:02:59.846
zero-gyro-sunpoint, which is a
more serious version of

00:02:59.846.00:03:04.584
safemode. What it looked like is
that we had just lost two gyros

00:03:04.584.00:03:08.221
simultaneously, which that can’t
happen. We had just replaced all

00:03:08.221.00:03:10.957
six of them in the first
servicing mission along with the

00:03:10.957.00:03:13.760
coprocessor; this just couldn’t
be happening. >>BRIAN V: I

00:03:13.760.00:03:16.163
remember just going, “What’s
going on?” You know, I didn’t

00:03:16.163.00:03:18.598
know what was happening. Because
the hardware guys looked at the

00:03:18.598.00:03:22.302
gyros — they were fine. But the
software said that they had a

00:03:22.302.00:03:24.471
problem with two of them.
>>BRIAN R: Here we are halfway

00:03:24.471.00:03:28.108
through the reconfiguration, and
we were in deeper trouble than

00:03:28.108.00:03:31.611
we had thought. >>BRIAN V: I
started looking at the time

00:03:31.611.00:03:34.214
between the two events. Started
figuring out, well okay, it’s

00:03:34.214.00:03:37.517
two and a half days, what is
that in hours, minutes, and

00:03:37.517.00:03:42.022
seconds. I saw the number, and
it was, you know it was really

00:03:42.022.00:03:44.524
obvious at that point what had
happened. There was an overflow

00:03:44.524.00:03:49.362
in the computer. Finally
discovered that we had this

00:03:49.362.00:03:53.867
thing called a “gyro good count”
that counts up how many seconds

00:03:53.867.00:03:57.270
has there been since there was
the last, you know, “bad gyro

00:03:57.270.00:04:01.041
count” if you will. >>BRIAN R:
When the gyros change mode, this

00:04:01.041.00:04:04.111
counter gets reset. Typically
this will happen several times a

00:04:04.111.00:04:07.547
day. Well because of the work we
had been doing for the past two,

00:04:07.547.00:04:10.884
two and a half days, this
counter never got reset. And

00:04:10.884.00:04:14.121
when it overflowed, or when the
counter value went past the

00:04:14.121.00:04:17.791
highest value it could count to,
bad things happened in the code.

00:04:17.791.00:04:20.060
>>BRIAN V: Once we discovered
that, and knew there was an

00:04:20.060.00:04:22.162
overflow in the software, we
knew we didn’t have our computer

00:04:22.162.00:04:24.631
problem. It was easy; now it was
just, now we just have to

00:04:24.631.00:04:29.169
recover. So by the time we
finished that and got back up

00:04:29.169.00:04:33.273
into science operating mode, it
was probably the middle of, end

00:04:33.273.00:04:36.510
of July 9th. So in plenty of
time for the observations.

00:04:36.510.00:04:39.012
>>HEIDI: Well as soon as we
heard that Hubble was back on

00:04:39.012.00:04:42.349
track, we were like, “Yes! Ready
to go!” >>HARRY: It allowed the

00:04:42.349.00:04:46.419
Shoemaker-Levy campaign to go
forward, and I’d hate to think

00:04:46.419.00:04:48.989
what would happen if we hadn’t
gotten the coprocessor in and

00:04:48.989.00:04:52.359
got it checked out. >>BRIAN R:
We retreated to our offices and

00:04:52.359.00:04:54.928
waited for the impact, wanted to
see the pictures just like

00:04:54.928.00:04:57.764
everybody else did, so we were
watching it on NASA TV just like

00:04:57.764.00:05:02.035
the rest of the world was. And I
remember seeing a press

00:05:02.035.00:05:04.404
conference at the Science
Institute announcing it and they

00:05:04.404.00:05:08.308
were waiting for the pictures,
when Heidi Hammel came in waving

00:05:08.308.00:05:11.178
the picture, the first picture
from Hubble. >>OFF-CAMERA VOICE:

00:05:11.178.00:05:14.114
Prepper shot, prepper shot,
prepper shot! She’s going in!

00:05:14.114.00:05:16.650
Camera 3! She’s bursting in,
she’s bursting in! Go, Heidi!

00:05:16.650.00:05:20.520
[noise of press conference]

00:05:20.520.00:05:22.155
>>GENE: ...that probably
indicates

00:05:22.155.00:05:25.892
we’re dealing with larger
objects than was concluded by

00:05:25.892.00:05:29.329
Asphaug and Benz. And I think we
may have some up-to-date

00:05:29.329.00:05:32.399
information from Heidi Hammel!
[laughter] >>HEIDI: Eugene

00:05:32.399.00:05:35.702
Shoemaker said he would be
personally astonished if we saw

00:05:35.702.00:05:39.906
nothing. Well he’s not going to
be astonished. We actually saw

00:05:39.906.00:05:43.343
some amazing things. The comet
delivered. It delivered

00:05:43.343.00:05:47.280
big-time. It had big black
spots, and if you looked at

00:05:47.280.00:05:50.150
certain colors of light it had
white spots, and it had the

00:05:50.150.00:05:54.154
rings, and it had plumes, and it
had big giant storms! And I was

00:05:54.154.00:05:56.923
really proud to be a part of
that. And I was really proud

00:05:56.923.00:05:59.059
that all the engineers and
scientists could pull together

00:05:59.059.00:06:00.961
and make that happen.

00:06:00.961.00:06:05.298
[music]

00:06:05.298.00:06:06.800
>>KATRINA: Comet
Shoemaker-Levy 9

00:06:06.800.00:06:09.903
added to our knowledge
of Jupiter’s atmosphere
and the

00:06:09.903.00:06:12.405
physics of atmospheric
explosions, and even influenced

00:06:12.405.00:06:15.108
some space policies on finding
near-Earth objects that could

00:06:15.108.00:06:18.879
impact us. Thanks to the entire
Hubble team, the telescope was

00:06:18.879.00:06:22.282
able to make these impactful
observations, and perform all of

00:06:22.282.00:06:25.819
its incredible science over the
past 25-plus years. The stories

00:06:25.819.00:06:29.256
don’t end here. Stay tuned for
more Hubble Memorable Moments.

00:06:29.256.00:06:31.758
[music ends]

00:06:31.758.00:06:33.460
[spacecraft swooshes by]

00:06:33.460.00:06:35.428
[spacecraft beeps]

00:06:35.428.00:00:00.000
[beep beep... beep beep...]