[00:00.000 --> 00:07.040]  ...discussion, that the command module and service module, the Apollo spacecraft, did not fly the Saturn V booster.
[00:07.320 --> 00:12.320]  The Saturn V booster had its own independent computer system that was developed by IBM.
[00:12.740 --> 00:18.520]  Nobody really talks about it because it was a very boring, conservative, you know, traditional computer.
[00:18.840 --> 00:27.260]  But it was the one that was in charge of getting the Saturn V aligned, tilt, pitch, roll, all that kind of good stuff, so it would get into orbit.
[00:27.260 --> 00:33.340]  So when you hear people talking about, oh, the command module is dead, they were worrying about, was the command module crippled?
[00:33.500 --> 00:40.040]  The problem, part of the problem was that the Saturn V computer was perfectly happy going to orbit.
[00:40.360 --> 00:43.780]  It didn't care if there was a big paperweight on the top of the rocket.
[00:43.940 --> 00:44.860]  It was going to orbit.
[00:45.080 --> 00:48.500]  And so there was the issue was, well, do we stay for the ride?
[00:49.180 --> 00:59.240]  So at Apollo 12 was the second moon landing, and it had a fairly, had a very nice start, and then something kind of weird happened to it.
[00:59.440 --> 01:05.140]  So this is the ground-to-Earth transmission, and this is what the world heard, if you were listening in on a NASA channel.
[01:05.260 --> 01:15.780]  This is the chat back between the astronauts, you'll hear the public affairs officials, you'll hear the CAPCOM, who is the only person on the ground who is allowed to talk to the astronauts.
[01:16.140 --> 01:27.300]  And you'll also hear some muttering at a certain point, because I believe this tape is from Honeysuckle Creek in Australia, where some of the engineers actually were speaking in the room, and the mic picked up the sound.
[01:27.460 --> 01:28.800]  So let's play this.
[01:29.020 --> 01:31.820]  And this whole incident only took about two, three minutes.
[01:33.760 --> 01:37.340]  Three, two, one, zero.
[01:37.600 --> 01:38.520]  All engines running.
[01:38.740 --> 01:39.080]  Commit.
[01:39.320 --> 01:41.060]  We have lift off.
[01:41.060 --> 01:42.800]  We have lift off at 11.22 a.m. Eastern Standard Time.
[01:44.920 --> 01:47.800]  Pete Conrad reports that your program is in.
[01:51.130 --> 01:52.310]  Power is clear.
[01:57.120 --> 01:59.120]  Pete Conrad reporting the roll-up.
[01:59.300 --> 02:00.120]  At the hull-ray have stopped.
[02:00.600 --> 02:01.160]  Not clear at all.
[02:06.110 --> 02:07.590]  Altitude at one-half mile.
[02:23.690 --> 02:25.130]  Altitude at a mile-and-a-half now.
[02:25.470 --> 02:28.210]  Velocity, 1,592 feet per second.
[02:36.270 --> 02:37.390]  Where's got your GDC?
[02:38.550 --> 02:40.350]  Okay, we just lost the platform.
[02:40.530 --> 02:41.710]  Gang, I don't know what happened here.
[02:41.790 --> 02:43.470]  We had everything in the world drop out.
[02:43.970 --> 02:44.390]  Roger.
[02:47.830 --> 02:48.350]  Plus one.
[02:48.610 --> 02:50.710]  Fuel cell lights and AC plus...
[03:11.650 --> 03:13.130]  Apollo 12, Houston.
[03:13.290 --> 03:15.090]  Try SCE to auxiliary.
[03:15.290 --> 03:15.350]  Over.
[03:16.350 --> 03:17.790]  SCE to auxiliary.
[03:19.410 --> 03:20.070]  SCE.
[03:20.070 --> 03:21.330]  SCE to auxiliary.
[03:27.720 --> 03:29.460]  If you can't report the readings back.
[03:31.080 --> 03:31.400]  Okay.
[03:32.600 --> 03:33.140]  And then they...
[03:33.140 --> 03:33.860]  Mark, watch Charlie.
[03:34.860 --> 03:40.720]  They pretty much go on to orbit then after that, and this goes on, and they do some clean-up, and they get a successful staging and everything.
[03:40.820 --> 03:41.960]  They teach them phonetics, huh?
[03:42.600 --> 03:42.920]  Sorry?
[03:43.040 --> 03:43.980]  They teach them phonetics.
[03:44.120 --> 03:44.680]  Well, yeah.
[03:44.880 --> 03:45.880]  There was a little bit of confusion.
[03:46.120 --> 03:47.680]  So that's what you heard.
[03:47.720 --> 03:49.580]  You heard that long silence.
[03:50.740 --> 03:51.220]  People hope.
[03:51.320 --> 03:53.620]  And this is, however, this is a ground-loop conversation.
[03:53.620 --> 03:59.280]  So it's a little harder to make out the audio, because what you're listening to is an overlay of the astronauts talking.
[03:59.560 --> 04:04.580]  You're also listening out for the flight director talking to the EECOM.
[04:04.760 --> 04:11.640]  And the EECOM was the mission controller in charge of the environmental... electronic environmental programs.
[04:11.720 --> 04:15.560]  At one point, he was also in charge of communications, but that was separated out, but they kept the name the same.
[04:16.040 --> 04:17.580]  And you hear him talk to EECOM.
[04:17.670 --> 04:21.800]  And most importantly, you hear him talking a little bit to his backroom boys.
[04:22.230 --> 04:28.500]  And so the way mission control was structured was, you see that very classic in all the guys in front of the rows of consoles.
[04:28.820 --> 04:34.620]  Each one of those, if you were to go out the door and turn right, there were many, many other rooms where you might have had...
[04:34.620 --> 04:42.850]  So you have the EECOM sitting as his console, and then he had three or four guys who also had very similar consoles looking at very specific subsystems.
[04:42.970 --> 04:48.080]  And oftentimes, they would chat themselves on the loop before they would maybe go back and talk to the flight director.
[04:48.080 --> 04:51.760]  So it's a little bit tangled, but I think you'll get the sense.
[04:52.740 --> 04:54.100]  So that's what the world heard.
[04:54.300 --> 04:55.020]  And now...
[04:55.020 --> 04:55.240]  Okay.
[04:55.560 --> 04:58.920]  So these are the actual transcripts from the mission controllers.
[04:59.600 --> 05:00.560]  Rules complete.
[05:00.850 --> 05:01.600]  Roger, Pete.
[05:01.880 --> 05:02.920]  Clockson 6, flight.
[05:03.240 --> 05:03.730]  Roger.
[05:08.420 --> 05:09.590]  Mark 1, bravo.
[05:09.920 --> 05:10.730]  Right here we go.
[05:11.360 --> 05:11.680]  Right here we go.
[05:20.240 --> 05:21.200]  Right here we go.
[05:21.820 --> 05:24.480]  Right here we go.
[05:38.590 --> 05:42.510]  I got three fuel cell lights, an AC bus light, a fuel cell
[05:45.690 --> 05:46.630]  bus AC out.
[05:46.650 --> 05:47.950]  How's it, how's it looking, EECOM?
[05:48.130 --> 05:49.610]  Is that the SCE?
[05:50.010 --> 05:52.010]  Boy, I don't know, John, it's sure what do you see.
[05:52.370 --> 05:52.490]  What do you see?
[05:52.710 --> 05:53.100]  I'm lost.
[05:53.570 --> 05:54.490]  Try SCE to AUX.
[05:54.830 --> 05:56.470]  Say again, SCE to AUX.
[05:56.690 --> 05:57.030]  AUX.
[05:59.030 --> 06:00.150]  SCE to AUX.
[06:00.150 --> 06:01.130]  Auxiliary flight.
[06:01.630 --> 06:03.070]  SCE to AUX, CAPCOM.
[06:03.850 --> 06:04.330]  What's that, EECOM?
[06:04.330 --> 06:05.150]  I'll follow us from Houston.
[06:05.530 --> 06:07.670]  Try SCE to auxiliary, over.
[06:08.950 --> 06:11.350]  SCE to auxiliary, S-C-E to auxiliary.
[06:11.790 --> 06:13.830]  SCE, SCE to auxiliary.
[06:13.990 --> 06:14.770]  We got a battle, EECOM.
[06:14.990 --> 06:15.270]  Looks good.
[06:15.550 --> 06:16.690]  We got a back flight, looks good.
[06:16.930 --> 06:17.130]  Okay.
[06:18.850 --> 06:19.630]  And that's it.
[06:19.770 --> 06:20.430]  They saved the mission.
[06:21.870 --> 06:26.450]  So what was happening there was the magic switch, SCE to AUX.
[06:26.610 --> 06:29.230]  Very clearly, that wasn't a common command.
[06:30.150 --> 06:33.410]  Nobody really knew what it was, but a couple of things went on here.
[06:33.890 --> 06:39.570]  One, the flight controller was a guy called, well, so let's talk about the SCE first.
[06:39.830 --> 06:43.690]  So the SCE is what's known as a signal conditioning equipment.
[06:45.390 --> 06:51.470]  And the Apollo missions were very heavily instrumented, and a lot of problems they had were due to instrumentation problems.
[06:51.670 --> 06:57.770]  But you get all these signals, you know, from the heater pump, from the radar, all of these instruments come through.
[06:57.890 --> 07:02.810]  They all have to be normalized into a standard range of voltages.
[07:03.390 --> 07:05.610]  So that one, they can be displayed inside the cockpit.
[07:06.050 --> 07:08.950]  And two, so that can be sent back down the radio feed.
[07:09.690 --> 07:18.570]  So what had happened was that when a lightning strike happened, all those fuel cells in the service modules that I mentioned tripped off.
[07:18.990 --> 07:27.310]  And so the command module was drawing power from a couple of batteries that were really only meant for reentry and a little bit of emergency situations.
[07:28.270 --> 07:36.350]  And if you read the Apollo 13 story, it became all about, not the section I'll be talking about, but it came very much about managing those battery reserves.
[07:36.750 --> 07:42.070]  But the problem was that when it went to those batteries, the voltage is dropped below what the SCE normally use.
[07:42.890 --> 07:48.290]  And there's a little switch down, a little panel that switched into auxiliary mode, which was not in the procedures.
[07:48.290 --> 07:53.390]  It was kind of there almost as a legacy issue from how they had built the equipment.
[07:53.630 --> 07:57.010]  And that said, whatever kind of crappy, shitty voltage you have, go with it.
[07:58.190 --> 08:11.350]  And when they did that, they got back telemetry and also the displays began, you know, reading meaningful results so they could see, not that they had the entire command module melting down, but that they had three disconnected fuel cells.
[08:12.610 --> 08:17.930]  So the man who made that call, though, who recognized that the SCE problem was this guy called John Aron.
[08:18.070 --> 08:20.830]  And he's also very important in Apollo 13 in different ways.
[08:21.150 --> 08:26.250]  But at the time, this incident made him like the original steely-eyed missile man.
[08:26.250 --> 08:35.550]  And the reason he was able to recognize that call that nobody else could have made was because a year earlier, at a PATH test, he had seen this result.
[08:35.550 --> 08:41.670]  And that test guy had made a mistake with the command module, the SCE had been giving out these screwy readings.
[08:41.970 --> 08:50.690]  And he knew they were screwy readings because, you know, there'd be a certain piece of equipment that he knew would only operate between 12 and 28 volts, and it was reporting 6 volts.
[08:50.890 --> 08:53.630]  It was a very distinct pattern of screwage.
[08:53.830 --> 08:56.370]  It's like, you know, sometimes if you look at your...
[08:56.370 --> 08:59.210]  You know when you get a blue screen of death and all those numbers appear?
[08:59.210 --> 09:04.710]  He was the guy who could actually look at those numbers and realize exactly what they meant.
[09:05.010 --> 09:09.810]  And so, in a sense, you might think, well, this was awfully lucky, and it really was lucky.
[09:09.830 --> 09:16.250]  But it does speak to the culture of mission control at that time, where these guys went through every system.
[09:16.390 --> 09:17.410]  They understood systems.
[09:17.530 --> 09:20.150]  They were deeply involved in developing the operations for them.
[09:20.210 --> 09:22.730]  They were deeply involved in talking with the engineers who built them.
[09:23.450 --> 09:27.270]  You know, each one of these guys had an immense depth of technical knowledge.
[09:27.270 --> 09:29.150]  It was built up over years in simulations.
[09:29.450 --> 09:33.090]  And then notice, when he actually made the call, first off, he checks his backroom guys.
[09:33.270 --> 09:34.550]  He says, well, do you think it's the SE?
[09:34.890 --> 09:36.250]  They don't know what the hell it is.
[09:36.890 --> 09:38.930]  And he comes back and says, flight, instantly.
[09:39.350 --> 09:40.550]  Flight, try SE to auxiliary.
[09:40.770 --> 09:45.530]  He has confidence to make that call that what he thinks he sees is what he sees.
[09:45.770 --> 09:50.010]  And I spoke to Gene Kranz about this, about how you take these flight controllers.
[09:50.130 --> 09:52.390]  Many of them are in their 20s, right out of college.
[09:52.550 --> 09:56.030]  And how do you get a guy to that particular point where he can make that call?
[09:56.030 --> 09:59.430]  And he says, well, it took about a year to 18 months to make a flight controller.
[09:59.530 --> 10:08.130]  And you start them off gathering information, doing very simple tasks, gathering information from manufacturers, helping the flight controllers build procedures and implementation.
[10:08.370 --> 10:10.890]  And then you put them through simulation after simulation after simulation.
[10:11.230 --> 10:15.730]  And the thing that would wipe out a flight controller was that they wouldn't have that level of confidence.
[10:16.010 --> 10:17.910]  It required a certain level of arrogance.
[10:18.590 --> 10:20.650]  It required a certain level of just go for it.
[10:20.750 --> 10:23.570]  And John Aron was possibly the epitome of that.
[10:23.570 --> 10:27.310]  He really understood all of his systems, and he had this unshakable confidence.
[10:27.650 --> 10:35.370]  And later on, again, it's an aspect of Apollo 13 I won't be talking about, but he devised this power-up sequence that the astronauts had to enter in without...
[10:35.370 --> 10:39.730]  In order to save power, they had to turn off the instrumentation that would tell them if they were making mistakes or not.
[10:40.030 --> 10:48.330]  So he had them radio up this checklist in which the astronauts had to enter in perfectly, first time, after they'd been awake for days, one of them had a urinary tract infection.
[10:48.690 --> 10:50.970]  And he made that call, and he had no problem with it at the time.
[10:50.970 --> 10:59.630]  Years later, as an older man, a little bit more reflective, he still has nightmares about it because of how risky that procedure was.
[11:00.170 --> 11:05.910]  But at the time, you know, you needed these cocky guys who were confident but not overconfident.
[11:07.050 --> 11:08.670]  So that was John Aron.
[11:08.710 --> 11:18.690]  The other thing I should say is that as soon as he made the call, notice how the flight controller, who has no idea what's going on, goes, well, okay, John, and he tells the CAPCOM, and it gets signaled up.
[11:18.770 --> 11:19.490]  They had trust.
[11:19.570 --> 11:20.270]  They believed each other.
[11:20.270 --> 11:21.370]  They trained with each other.
[11:21.510 --> 11:22.710]  They relied on each other.
[11:22.890 --> 11:26.250]  That's an engineering culture that is prime for success.
[11:26.450 --> 11:30.310]  I mean, recently in years, NASA has been dinged for its, you know, culture issues.
[11:30.330 --> 11:36.290]  But anybody who's involved in high-risk, high-tech enterprises could learn from what these guys did.
[11:36.570 --> 11:38.470]  So Apollo 13, which is a very famous...
[11:38.850 --> 11:40.570]  All kinds of hacks on Apollo 13.
[11:41.070 --> 11:47.990]  Some famous ones, you know, putting together carbon dioxide filters so they wouldn't choke, recharging the command module from the batteries.
[11:48.030 --> 11:49.730]  Now, I assume you've all seen the movie.
[11:50.470 --> 11:56.750]  That actual thing of how do we recharge the lunar module from the command module or the command module from the lunar module.
[11:56.870 --> 12:00.750]  That was actually something that they had been working on for several months previously.
[12:01.650 --> 12:03.370]  Using the LM engine for course corrections.
[12:03.370 --> 12:05.510]  Because again, they had been thinking about that for some time.
[12:05.970 --> 12:10.870]  Even the emergency activation checklist, they already had a kind of a dummy situation.
[12:11.130 --> 12:12.870]  So these are all great, pretty good hacks.
[12:13.150 --> 12:15.870]  And actually, the carbon dioxide filter issue, again, they had...
[12:15.870 --> 12:18.050]  They had maybe 24 hours to work on that.
[12:18.170 --> 12:18.830]  It was never...
[12:18.830 --> 12:20.870]  The astronauts never really broke a sweat about that.
[12:20.990 --> 12:22.150]  Because they knew people were working on it.
[12:22.210 --> 12:22.930]  They knew there was a solution.
[12:23.130 --> 12:25.930]  They actually were kind of a bit late in the day when they put it together.
[12:25.930 --> 12:29.350]  Because they had other things on their mind and they thought this was a lower priority item.
[12:30.030 --> 12:32.290]  So all of those were, you know, good hacks.
[12:32.590 --> 12:38.510]  But the hack I really want to talk about, and this is one that isn't terribly well known, was turning the LM on in the first place.
[12:39.150 --> 12:46.330]  And I'm not talking about the activation checklist, which is, you know, you turn on this bit of equipment in this sequence and you get, you know, your platform all lined up.
[12:46.530 --> 12:54.190]  The simple act of turning the lunar module from, again, a paperweight into a functioning spaceship requires power.
[12:54.190 --> 13:01.230]  There was no way to turn the lunar module on when the Apollo 13's fuel cells exploded.
[13:02.090 --> 13:05.550]  Actually, the oxygen tank exploded and that knocked off the fuel cells.
[13:07.410 --> 13:15.490]  So what had happened in that mission was the Apollo 13 service module, there was an explosion in an oxygen tank, the fuel cells died.
[13:15.770 --> 13:20.330]  Very quickly, the command module was running on those reserve backup batteries to power all its systems.
[13:20.570 --> 13:23.190]  Couldn't do that, so they had already started shutting down systems.
[13:24.530 --> 13:26.150]  in the command module.
[13:26.290 --> 13:28.050]  So the command module has very little power.
[13:28.270 --> 13:31.690]  And this turned to have this huge knock-on problem for the lunar module.
[13:31.870 --> 13:32.990]  And the reason is this.
[13:33.350 --> 13:37.170]  So these are the lunar module activation checklist.
[13:37.850 --> 13:44.350]  I got this from, if anybody really cares about the lunar module, there's the Cradle of Aviation up on Long Island has the Grumman archive.
[13:45.070 --> 13:48.490]  So if you wanted to build your own lunar module, most of it is in there.
[13:48.830 --> 13:50.910]  And you can go up and you can go through the manuals.
[13:50.910 --> 13:52.770]  This, actually, you can also get online.
[13:53.070 --> 13:58.230]  So this is the official activation checklist, which, again, was cut down and very much shortened.
[13:58.590 --> 14:00.890]  But notice the first assumption made.
[14:01.170 --> 14:06.110]  The CSM has initiated power transfer via docking tunnel electrical umbilicals.
[14:06.110 --> 14:08.670]  So the very first assumption was now invalid.
[14:08.850 --> 14:12.850]  So you could not proceed with anything on the checklist because there was a false assumption.
[14:13.190 --> 14:16.650]  And the reason for that was this simple little sketch.
[14:18.430 --> 14:23.070]  Actually, the full version of the thing is, actually, this came on, like, three gatefolds.
[14:23.130 --> 14:24.950]  The full version, like, this is the overview.
[14:24.950 --> 14:29.510]  The full detailed version is really beyond belief.
[14:29.930 --> 14:31.890]  But, so it's divided into three parts.
[14:32.050 --> 14:37.550]  You'll remember the lunar module had a descent stage and a descent stage on top.
[14:37.750 --> 14:40.570]  And then it was attached to the command module for most of the trip.
[14:40.670 --> 14:43.510]  Most of the trip, it was docked to the command module.
[14:46.270 --> 14:50.010]  So, the descent stage lunar module had great big batteries.
[14:50.230 --> 14:50.830]  Lovely batteries.
[14:50.990 --> 14:51.830]  Pull of lots of power.
[14:52.030 --> 14:53.350]  We'll happily run the lunar module.
[14:53.730 --> 14:59.650]  In order to get those batteries activated, it was controlled by a junction box, which was a powered relay box.
[14:59.810 --> 15:05.950]  So you could not get the descent batteries on unless you had power to a junction box.
[15:06.190 --> 15:11.610]  The junction box got its power from, ba-bam, through an umbilical from the command module.
[15:12.250 --> 15:18.610]  So, the only way you could power the junction box to turn on the descent batteries was via a switch in the command module.
[15:18.790 --> 15:20.370]  A switch which had lost power.
[15:22.470 --> 15:23.390]  There we go.
[15:24.150 --> 15:26.270]  So, why on earth would you design it this way?
[15:27.530 --> 15:30.490]  Why in God's name wouldn't you just have a simple on-off switch?
[15:30.690 --> 15:31.670]  An emergency switch.
[15:31.770 --> 15:32.710]  Perhaps you could hide it somewhere.
[15:32.830 --> 15:33.250]  It wouldn't matter.
[15:33.450 --> 15:35.190]  A little switch that you could turn on.
[15:35.470 --> 15:38.430]  Why in God's name would you design a system this way?
[15:40.470 --> 15:42.350]  So, the answer was because of the batteries.
[15:43.150 --> 15:50.950]  If you think battery technology sucks today, you go back to 1963 or, you know, 1972 even.
[15:51.170 --> 15:58.870]  So, the Apollo system used zinc, silver zinc batteries, which were chemically activated.
[15:58.950 --> 16:01.770]  The lunar module actually used non-rechargeable versions of these batteries.
[16:02.050 --> 16:04.210]  They were charged by... they were kept dry.
[16:04.210 --> 16:07.110]  When it was time for emission, you would fill them up with electrolyte.
[16:07.290 --> 16:08.730]  And then they would be charged up.
[16:08.950 --> 16:10.010]  There was no recharging.
[16:10.170 --> 16:12.290]  You had a set amount of hours.
[16:12.550 --> 16:13.870]  So, it became very much...
[16:13.870 --> 16:19.370]  A lot of design decisions became about conserving power in those batteries, using them as little as possible.
[16:21.910 --> 16:22.870]  And so...
[16:22.870 --> 16:23.330]  There was...
[16:23.330 --> 16:24.870]  But because they did...
[16:24.870 --> 16:28.910]  And they had used silver zinc batteries, because silver zinc batteries are very, very energy dense.
[16:29.190 --> 16:35.390]  There's actually some interest now in doing a type of silver zinc batteries to power laptops and so on as an alternative to lithium ion.
[16:35.650 --> 16:36.130]  But...
[16:36.130 --> 16:37.150]  So, very tremendously dense.
[16:37.370 --> 16:38.290]  Very, very heavy.
[16:39.250 --> 16:40.710]  But one shot only.
[16:40.710 --> 16:40.770]  Anyway.
[16:41.830 --> 16:43.590]  So, there was also another weirdness.
[16:43.710 --> 16:44.110]  So, this is...
[16:44.110 --> 16:46.130]  This is a schematic of the descent batteries.
[16:46.650 --> 16:48.490]  Again, I got this from Long Island.
[16:49.230 --> 16:49.670]  And...
[16:51.090 --> 16:51.970]  You'll see...
[16:52.530 --> 16:53.110]  If this works.
[16:53.230 --> 16:53.390]  Yay!
[16:53.610 --> 16:53.790]  Okay.
[16:54.310 --> 16:56.190]  So, there was actually two conductors.
[16:56.370 --> 16:58.790]  So, the silver zinc batteries had this weird...
[16:58.790 --> 17:00.950]  Where they did not give a constant voltage.
[17:01.590 --> 17:07.930]  And it was when you fully charged them up, they produced 38 volts when they were fully charged.
[17:09.070 --> 17:12.150]  Then, when the power went down, they dropped down to 32 volts.
[17:12.790 --> 17:20.610]  And so, what there was on the descent batteries was there was a tap at the 17 cells in, which would just give you the lower 32 voltage.
[17:20.910 --> 17:26.630]  And then, you know, as the battery was used a little bit, they would tap so the full 20 cells within that battery was used.
[17:26.930 --> 17:32.330]  And the reason for doing this was because there was certain equipment in the command modules...
[17:32.330 --> 17:35.710]  Sorry, in the lunar module that could not tolerate 38 volts.
[17:35.710 --> 17:38.670]  It was expected to operate at 32 volts or less.
[17:39.050 --> 17:46.030]  So, the descent batteries, again, controlled with that junction box, had this magic low voltage switch which you needed to turn the lunar module on.
[17:46.090 --> 17:50.890]  Otherwise, you would blow up important things that you would really want later on.
[17:51.450 --> 17:54.330]  Now, the ascent batteries did not have this tap.
[17:54.490 --> 17:55.810]  And this is going to become important in a little bit.
[17:56.010 --> 17:57.270]  They only had the full...
[17:57.270 --> 17:57.890]  You could only get...
[17:57.890 --> 18:00.110]  They started out at the full 38 volts and dropped to 32.
[18:00.110 --> 18:12.170]  The reason you could get away with that was because the way the missions were sequenced and the loads were designed, by the time you had the ascent batteries on, you had enough of a load that the lower...
[18:12.170 --> 18:14.530]  The voltage would drop enough not to damage some of the equipment.
[18:16.730 --> 18:20.870]  So, you had this terrible problem where you had an Apollo module that couldn't...
[18:20.870 --> 18:22.790]  Sorry, a lunar module that could not be turned on.
[18:23.730 --> 18:26.350]  And that was truly was nowhere in the procedures.
[18:26.970 --> 18:30.430]  There are two guys who are really unsung heroes about this.
[18:30.970 --> 18:32.430]  Bob Legler and Bill Reeves.
[18:32.530 --> 18:35.030]  Bob Legler sadly has passed away a couple of years ago.
[18:36.210 --> 18:37.430]  I think Bill Reeves...
[18:37.430 --> 18:39.370]  Bill Reeves eventually went on to become a flight director.
[18:39.890 --> 18:42.370]  And they had been sitting in their little back room.
[18:43.470 --> 18:45.950]  And they had noticed that...
[18:45.950 --> 18:51.430]  So, they had a little paper trace that was monitoring the five things in the lunar module that were on.
[18:52.450 --> 18:54.270]  When the lunar module normally...
[18:54.890 --> 18:55.310]  Sorry.
[18:56.690 --> 19:01.250]  For most of its flight while the lunar module was docked to the command module, it was in a very low power mode.
[19:01.430 --> 19:03.990]  You didn't have lights flashing because you didn't need the power.
[19:04.330 --> 19:08.790]  However, there were like a couple of things like heaters and glycol pumps and some instrumentation.
[19:08.870 --> 19:11.650]  A handful of equipment that you kept on throughout the flight.
[19:12.090 --> 19:18.290]  When the Apollo 13 explosion occurred, they saw all their little traces just go to zero.
[19:18.290 --> 19:20.810]  And they knew they had lost power in the command module.
[19:21.030 --> 19:28.890]  And at that moment, they knew they were going to have to find a way to turn on the lunar module without using any of the command module resources.
[19:29.210 --> 19:31.810]  So, this is, you can see in the background here, a more detailed map.
[19:32.010 --> 19:43.410]  They pulled these out in the back room and started going through them and working out if there was any way to turn the lunar module on without, you know, before the astronauts died.
[19:44.750 --> 19:50.110]  So, the solution they came up with was, again, so again, you have to send stage the junction box.
[19:50.650 --> 19:52.630]  I did mention you had these ascent batteries.
[19:52.830 --> 20:04.810]  But again, the problem was, if you had just connected the ascent batteries to the junction box, you would have fried all these pieces of equipment in the ascent module because the ascent batteries would have started pumping out 38 volts, which was just too high.
[20:04.810 --> 20:14.230]  So, what they did in about half an hour was work out a whole series of circuit breaker throws, which was to pull every piece of equipment that was going to melt down at 38 volts off the line.
[20:14.990 --> 20:16.350]  They got all that off the line.
[20:16.510 --> 20:22.990]  They got power teased through the system again with some more circuit pulls to get power to the junction box.
[20:23.650 --> 20:27.850]  They were able to power the junction box, get the junction box turned on.
[20:28.470 --> 20:30.970]  From there, then, they were able to get the ascent batteries turned on.
[20:31.290 --> 20:37.910]  And the ascent batteries set to the low voltage setting, turn off the ascent batteries, and then they could begin the full activation checklist.
[20:39.010 --> 20:50.930]  The reason it was sort of built this way was because when the Saturn V was sitting on the pad, it was connected by a launch umbilical, which was providing power to those handful of critical systems the lunar module always needed to be on.
[20:50.930 --> 20:59.050]  Just before launch, the umbilical would send the signal to the junction box, which would turn the lunar module onto its internal battery power.
[20:59.490 --> 21:00.490]  And it would take off.
[21:00.610 --> 21:06.070]  At some point, the command module would dock with the lunar module, and then they would attach these umbilicals.
[21:06.170 --> 21:13.270]  And the first thing you would do with the umbilicals then was turn off the ascent batteries to the junction box so the lunar module was being powered through the fuel cells.
[21:13.570 --> 21:18.050]  And so they were able to basically go back into this junction box and get the ascent batteries going.
[21:19.070 --> 21:20.750]  It took them about half an hour.
[21:21.090 --> 21:26.190]  This is really the one truly see-of-the-pants thing that happened on Apollo 13.
[21:26.470 --> 21:29.070]  And it very much is the thing that saved those astronauts.
[21:29.230 --> 21:38.730]  Because if they hadn't gotten the lunar module on, all the other fancy stuff, you know, waiting for them to come down to the atmosphere, none of that would have happened.
[21:38.830 --> 21:39.410]  They all would have died.
[21:40.590 --> 21:45.750]  So last great hack, Apollo 14th, February 5th, 1971.
[21:47.110 --> 21:51.890]  So now you have a situation where everything this time has gone pretty well.
[21:51.990 --> 21:53.050]  They've actually gotten to the moon.
[21:56.090 --> 22:00.850]  The astronauts have transferred over to the lunar module to descend to the lunar surface.
[22:00.870 --> 22:02.470]  They've separated from the command module.
[22:02.690 --> 22:07.130]  And now they're in kind of basically a parking orbit going around the moon.
[22:07.990 --> 22:10.810]  So then, however, mission control notices something funny.
[22:12.290 --> 22:15.550]  What happened was, so this is a little schematic of the lunar module.
[22:15.550 --> 22:21.610]  And there's a panel in front of the guys, a command panel, which looks like this.
[22:21.670 --> 22:30.070]  And there was a particular little worrying little incident with this particular switch, which is the abort switch, which is a very bad switch to have go wrong.
[22:30.150 --> 22:30.730]  You notice there's two.
[22:30.850 --> 22:31.370]  One is abort.
[22:31.470 --> 22:32.230]  There's abort stage.
[22:32.230 --> 22:43.730]  So the idea being that if you were, say, you were coming down towards the lunar surface, and you discovered your descent stage was crapping out for whatever reason, you could punch abort stage.
[22:43.830 --> 22:50.330]  It would sever the connections, immediately try and boost the ascent stage back up into orbit.
[22:51.570 --> 22:53.110]  Abort was just a general abort button.
[22:53.110 --> 22:54.490]  You could pretty much press it at any time.
[22:56.410 --> 23:03.010]  The problem was was that there was a little ball of solder had come loose inside that switch and was floating around in zero gravity.
[23:03.270 --> 23:06.030]  And every now and then, it would tip the contacts.
[23:06.270 --> 23:12.810]  And the guys who were paying attention to the register would see abort come up on their little screens.
[23:12.810 --> 23:18.730]  Now, fortunately, the lunar module's computer was in a mode that was not paying attention to abort modes.
[23:21.810 --> 23:24.670]  So they had a whole set of modes, and they had it set.
[23:24.670 --> 23:32.650]  So if you're just, like, flying along idly in orbit, you don't really need to pay attention to something that only is going to take place during a descent stage.
[23:33.530 --> 23:36.390]  And then they asked the astronauts, hmm, are you pressing the abort button?
[23:37.730 --> 23:38.770]  And they said, no.
[23:39.310 --> 23:41.350]  And then they said, hmm, could you try tapping it?
[23:41.350 --> 23:46.130]  And they tapped on it, and the abort little signal went away because they had knocked the solder ball.
[23:46.230 --> 23:48.550]  And there's this great line where it goes, well, you sure tapped nicely.
[23:50.310 --> 24:00.590]  But they had this problem in that at any moment during the descent, this little solder ball could move around and have triggered an abort, which was kind of bad news if you're trying to land on the Moon.
[24:03.650 --> 24:07.770]  So this all comes down to what's involved here is the Apollo guidance computer.
[24:10.590 --> 24:15.870]  For this audience, you cannot underestimate the significance of this computer.
[24:16.010 --> 24:21.530]  When people talk about, you know, spin-offs from the NASA program, this computer was essential.
[24:21.750 --> 24:24.590]  They did not invent integrated circuits for this computer.
[24:24.790 --> 24:30.550]  They did not invent, you know, the computer or even the small computer for this mission.
[24:30.550 --> 24:37.170]  But this mission, this computer had so much of an impact on computing after that.
[24:37.350 --> 24:50.130]  And part of the reason was because when integrated circuits were first introduced by Kilby and Noyce and all the rest of them, engineers hated them, despised them.
[24:50.130 --> 25:02.430]  But if you read the Nobel Prize lecture of Kilby, who got, you know, from inventing the integrated circuit of Texas Instruments, he says we, quote, provided the entertainment of technical conferences for several years.
[25:04.550 --> 25:07.150]  Engineers hated integrated circuits for two reasons.
[25:07.730 --> 25:09.510]  One, you have this black box.
[25:09.690 --> 25:18.730]  So previously their job was, if they were building, say, oh, I don't know, a B-52 autopilot, you would have resistors and capacitors and transistors.
[25:18.890 --> 25:25.210]  And you could measure the characteristics of all those devices individually and then get a fairly good sense of how it all worked together.
[25:25.450 --> 25:27.510]  So, and they understood how to build those circuits.
[25:28.090 --> 25:37.530]  And then you're asking them, instead of the situation where you could test each individual component separately and have confidence in the whole system as a result, here's a black box.
[25:37.890 --> 25:39.350]  It works, trust me.
[25:40.330 --> 25:43.430]  Really hard to get people to, like, just trust the damn things.
[25:43.430 --> 25:47.990]  The other issue with integrated circuits, why they hate it, was they were a little unreliable.
[25:49.030 --> 25:50.690]  You know, they did sometimes crap out.
[25:51.010 --> 25:51.670]  Yields were low.
[25:51.750 --> 25:52.510]  They were very expensive.
[25:52.950 --> 25:58.130]  Without the Apollo guidance computer, the only people willing to put up with them were people building nuclear missiles.
[25:58.330 --> 26:04.990]  Because they were willing to, you know, do whatever it took to build small missiles that you could put on a submarine, for example.
[26:05.230 --> 26:08.850]  But everybody else really, truly hated them.
[26:08.850 --> 26:17.450]  And what was unique about the Apollo guidance computer was they decided to make it one completely out of integrated circuits.
[26:18.750 --> 26:22.390]  The block two, this is it, but we'll talk about this AGC I have here.
[26:22.670 --> 26:26.430]  The block two had, like, something like 2800 NOR gates.
[26:26.610 --> 26:31.350]  And as you all know, if you have a NOR gate, you can build any other kind of gate by assembling enough of them together.
[26:31.350 --> 26:35.910]  So you get your AND and your NOT gates, all of your other logic gates.
[26:36.290 --> 26:41.570]  So basically, NASA was ordering a boatload of integrated circuits.
[26:41.830 --> 26:52.590]  In the early days, at one point, NASA was responsible for something like 40 to 50% of all the ICs made in the world were going into the Apollo program, going into the Apollo guidance program.
[26:53.110 --> 26:54.830]  What this did was it had two effects.
[26:54.990 --> 26:59.570]  One, NASA were not really... NASA were really anal about reliability issues.
[26:59.570 --> 27:03.610]  Every time there was a fault in one of these chips, NASA wanted to know why.
[27:03.770 --> 27:06.870]  So they would take them up, and they would look at them, and they would look at failures.
[27:07.070 --> 27:11.090]  And in doing that, it dramatically improved the manufacturing process.
[27:11.530 --> 27:14.870]  Two, as I mentioned, integrated circuits are very expensive.
[27:15.130 --> 27:16.490]  Very few people wanted to buy them.
[27:16.630 --> 27:26.450]  By providing a guaranteed customer, and for years, the Apollo system really underwrote an awful lot of the early integrated circuit manufacturer.
[27:27.150 --> 27:33.290]  The third point is, is that, again, people did not trust all digital, all integrated circuit computers.
[27:33.570 --> 27:43.110]  By going to the bloody moon and back, it kind of said, well, yeah, you can run your little chemical plant or your little thing, that's fine, because we went to the moon with this stuff.
[27:43.390 --> 27:45.310]  And it wasn't until... and it really was.
[27:45.470 --> 27:47.450]  That was a significant... that was a significant...
[27:47.450 --> 27:52.910]  proving you could trust digital integrated circuits on the most highest profile project ever.
[27:53.150 --> 27:56.150]  Then there were people, oh, okay, well, I guess we could use one or two.
[27:56.430 --> 28:01.850]  And that really was... that really was the birthplace of the commercial integrated industry.
[28:01.930 --> 28:07.070]  The reason why we have all of these super cheap processes around our necks owes a lot to this...
[28:07.070 --> 28:08.470]  to this particular device.
[28:08.470 --> 28:14.350]  And Frank O'Brien very, very kindly brought me in...
[28:14.350 --> 28:15.350]  Can you see this?
[28:15.590 --> 28:20.570]  This is a prototype Apollo guidance system.
[28:20.670 --> 28:22.130]  This is just the logic circuits.
[28:22.350 --> 28:24.970]  And if you look closely, you... I really...
[28:24.970 --> 28:27.510]  Frank is going to have this in the retro computing section.
[28:27.730 --> 28:29.390]  I really encourage you to look closely.
[28:29.390 --> 28:34.950]  You can see how the integrated circuits are potted in, how there's a little bit of hand wire wrap going on.
[28:34.950 --> 28:39.170]  In the very original days, it was thought that the astronauts would be...
[28:39.170 --> 28:43.370]  would be able to repair the Apollo guidance computer by slipping these modules in.
[28:43.450 --> 28:46.450]  You'll see it's all very much, you know, modules that plug in.
[28:46.730 --> 28:49.290]  Eventually it was decided that astronauts couldn't be trusted with a screwdriver.
[28:49.910 --> 28:50.650]  And so...
[28:51.370 --> 28:52.150]  Everything was...
[28:52.150 --> 28:52.790]  Well, that's a little unfair.
[28:52.870 --> 28:55.050]  But everything was designed so that it was...
[28:55.050 --> 28:57.810]  It was a completely sealed unit.
[28:58.510 --> 28:59.430]  So this is...
[28:59.430 --> 29:00.670]  Can you see that up on the screen?
[29:01.130 --> 29:01.830]  Okay, great.
[29:01.830 --> 29:08.190]  So I really strongly advise you to please go down, visit Frank, visit this unit.
[29:08.390 --> 29:12.510]  He will also have what's called a disk-y emulator running.
[29:14.290 --> 29:20.130]  Because you can pretty much get, you know, all this software and hardware running in emulation mode.
[29:20.210 --> 29:21.070]  And it's pretty cool.
[29:21.330 --> 29:22.190]  Let me go back.
[29:23.030 --> 29:24.530]  I think we can get an iPhone app for that.
[29:25.410 --> 29:26.670]  Yes, actually we should.
[29:27.290 --> 29:30.870]  It would be great to see an Apollo iPhone AGC application.
[29:31.510 --> 29:32.670]  Someone's written it for the phone.
[29:33.050 --> 29:33.450]  Sorry?
[29:33.690 --> 29:34.870]  Someone's written it for the phone.
[29:35.050 --> 29:35.930]  Oh, okay, great.
[29:36.090 --> 29:37.390]  Well, let's have a port.
[29:38.970 --> 29:44.090]  So one of the important things about it as well, just to go back, is that it used rope core memory.
[29:44.410 --> 29:45.970]  So it used, you know, you've all...
[29:45.970 --> 29:51.070]  So it uses basically little ferrite cores on top of long strings of copper wires.
[29:51.070 --> 29:53.670]  And this was programmed by hand.
[29:53.790 --> 29:58.570]  I don't mean by hand, by somebody flicking switches or very, you know, doing...
[29:58.570 --> 30:09.750]  I mean there were little old ladies in the Raytheon, middle-aged women in the Raytheon factory, who would either, if they passed a wire, very thin wire, through a ferrite core, that was a one.
[30:09.970 --> 30:12.070]  If they bypassed it, that was a zero.
[30:12.630 --> 30:15.750]  So it was literally, these were practically hand-woven memories.
[30:16.550 --> 30:19.850]  And each Apollo mission had...
[30:20.410 --> 30:21.030]  It had to...
[30:21.030 --> 30:24.630]  The program code had to be delivered six weeks before the mission was due to launch.
[30:24.930 --> 30:26.310]  You know, you could not change it.
[30:26.370 --> 30:27.690]  There was no patch.
[30:28.810 --> 30:31.370]  You know, you couldn't go, hey, you're in your mind, you know.
[30:31.490 --> 30:32.390]  Here's another little bit of rope.
[30:32.530 --> 30:33.430]  You just stick it in there.
[30:33.430 --> 30:33.690]  No.
[30:34.130 --> 30:40.110]  And I think a lot of the quality of the software was because they knew there couldn't be changes.
[30:40.410 --> 30:41.710]  You couldn't faff around.
[30:41.850 --> 30:43.890]  You couldn't go, oh, okay, this is 1.0.
[30:44.370 --> 30:45.150]  They'll be fine.
[30:45.390 --> 30:47.570]  That said, each version of the...
[30:47.570 --> 30:51.090]  Each version of the lunar modules went with a slightly different version.
[30:51.250 --> 30:54.050]  And in part, that's because emission parameters changed.
[30:54.130 --> 30:55.510]  A lot of issues were hard-coded.
[30:55.750 --> 30:59.630]  In part, because they did get better, they did try eliminate some bugs.
[31:00.830 --> 31:01.810]  But I've seen...
[31:01.810 --> 31:04.910]  If you go to the MIT Museum, they have a listing for what's called luminary.
[31:05.130 --> 31:06.230]  And it's about...
[31:06.230 --> 31:10.170]  It's a piece of fan-fold paper, which is about, like, six inches thick.
[31:10.490 --> 31:12.110]  Which is the complete program listing.
[31:12.370 --> 31:16.630]  All of which was fitted into about 72 kilobytes.
[31:17.430 --> 31:19.370]  I know it's 36 kilo words.
[31:19.510 --> 31:21.710]  72 kilobytes of ROM.
[31:21.890 --> 31:23.470]  It had about 2 kilobytes of ROM.
[31:24.310 --> 31:27.990]  So the way it worked was it had this verb-noun syntax.
[31:27.990 --> 31:32.890]  So it was a little bit like the early Activision text adventure games.
[31:33.130 --> 31:33.990]  Where you would basically...
[31:33.990 --> 31:38.630]  So you would press a verb, and then you would type a digit code for the verb.
[31:38.890 --> 31:39.950]  So the verb would mean do something.
[31:40.070 --> 31:41.310]  Noun would say what it was.
[31:41.450 --> 31:42.310]  So it might be verb.
[31:43.410 --> 31:50.050]  The verb might translate to load these numbers into memory, and then noun would specify exactly what the numbers were or what they were to do with them.
[31:50.130 --> 31:51.990]  Then you would type in a bunch of stuff.
[31:52.170 --> 31:55.590]  Oh, and by the way, a lot of this was an octal for extra fun and games.
[31:56.630 --> 31:58.930]  I should say also one final thing on the disk key.
[31:59.030 --> 32:00.050]  You'll see these green lights.
[32:00.270 --> 32:02.590]  A lot of people think these are like LED displays.
[32:03.770 --> 32:05.690]  LEDs weren't really around when this was built.
[32:05.770 --> 32:09.370]  These were electroluminescent displays, which rounded about 300 volts.
[32:10.330 --> 32:13.890]  So it was kind of a bit of serious engineering went into...
[32:13.890 --> 32:16.670]  So this display itself was also groundbreaking.
[32:18.890 --> 32:22.470]  So back to our poor buggers circling the moon with a dodgy abort button.
[32:23.150 --> 32:27.210]  The lunar module computer was locked out from acting on the problem for now.
[32:27.530 --> 32:33.170]  But as soon as the engine was fired, then the computer thinks, hey, it's time to pay attention to the abort button.
[32:33.730 --> 32:35.430]  And it would trigger one of two programs.
[32:35.550 --> 32:37.750]  P-70, which was abort with a descent engine.
[32:37.950 --> 32:40.590]  P-71, stage abort with a descent engine.
[32:40.590 --> 32:44.490]  The normal descent programs, I think, were called P-64, P-65.
[32:44.990 --> 32:46.110]  Frank will correct me.
[32:47.350 --> 32:49.290]  So enter Don Isles.
[32:49.570 --> 32:54.030]  You know, dashing young man who was working at MIT.
[32:54.910 --> 32:59.810]  Interesting enough, he says he came out with an incredibly undistinguished math degree.
[33:00.090 --> 33:03.690]  And basically Draper Laboratory was the only place that would employ him, more or less.
[33:03.690 --> 33:05.150]  He was involved...
[33:05.150 --> 33:07.350]  A lot of people wrote the lunar module software.
[33:07.650 --> 33:11.890]  But he really did a lot of the hard core work in taking very theoretical...
[33:11.890 --> 33:19.710]  taking the theoretical guidance equations, control equations, and converting those into the actual code that would run.
[33:19.770 --> 33:21.810]  He did a lot of the very critical flight...
[33:21.810 --> 33:24.690]  really the guts of the flight control system.
[33:24.830 --> 33:28.650]  You know, turning on the engine, turning the spacecraft, not crashing into the ground.
[33:29.670 --> 33:31.570]  So he did a lot of that work.
[33:34.030 --> 33:34.410]  So...
[33:34.410 --> 33:35.290]  What has gone wrong?
[33:38.510 --> 33:38.890]  No?
[33:39.010 --> 33:39.450]  Okay, there we go.
[33:40.730 --> 33:41.110]  So...
[33:41.110 --> 33:45.430]  Apollo 14, the astronauts really want to land, and they would like a solution, please.
[33:46.810 --> 33:48.530]  So the first idea that...
[33:48.530 --> 33:49.810]  So Don Isles, who...
[33:49.810 --> 33:51.390]  And they were always at...
[33:51.390 --> 33:53.470]  Don Isles says there's a story going around that the...
[33:53.470 --> 33:57.510]  You know, he was sitting at home and the Air Force pulls up in this big limousine and guys with soldiers.
[33:57.690 --> 34:01.590]  And I think somebody might be watching the Andromeda strain, because I think that's actually where I was from.
[34:01.590 --> 34:02.250]  But...
[34:02.250 --> 34:04.350]  And I whisked him away to like save the mission.
[34:04.470 --> 34:07.390]  In fact, he was in his office, as he always was, for a mission.
[34:07.610 --> 34:10.170]  And then one of the guys came in and says, hey, the abort discreet.
[34:10.330 --> 34:11.190]  So it was that bit as lit.
[34:11.250 --> 34:12.350]  He's like, oh, okay.
[34:12.550 --> 34:13.070]  Well, it's not too bad.
[34:13.230 --> 34:14.930]  And then it was obvious they weren't going to clear it up.
[34:15.150 --> 34:17.650]  So his first idea was, well, we'll just...
[34:18.610 --> 34:22.750]  Using basically the equivalent of a poke command, we'll just disable the abort monitor manually.
[34:23.050 --> 34:24.090]  The engine will fire up.
[34:24.350 --> 34:26.710]  The computer will start looking for the abort code.
[34:26.710 --> 34:28.610]  We'll just tell them to turn off the abort code.
[34:28.810 --> 34:32.070]  But that would mean that the astronaut was racing the computer.
[34:32.250 --> 34:35.010]  That little solder ball, you know, was being accelerated.
[34:35.170 --> 34:35.770]  It could have gone anywhere.
[34:36.050 --> 34:40.790]  If the astronaut, you know, with his pudgy astronaut fingers, couldn't get more...
[34:40.790 --> 34:43.050]  Because oftentimes they're in spacesuits, and so that was actually really...
[34:43.050 --> 34:52.470]  If the astronaut could not beat the time, the monitor would have checked that switch hundreds of times before the astronaut would have been able to lock it out.
[34:52.810 --> 34:55.010]  So it would have been a real race condition.
[34:55.530 --> 34:56.670]  And it would not have been a good one.
[34:57.350 --> 35:00.930]  So that was his first solution, which they radioed up, which nobody was particularly happy with.
[35:01.510 --> 35:03.350]  Alan Shepard was the commander of this module.
[35:03.610 --> 35:09.950]  Those of you who know about some of the astronauts will know that he was not noted for his general calm and good temper.
[35:11.990 --> 35:14.370]  So he was getting pretty grumpy about this.
[35:14.730 --> 35:15.790]  So the second solution...
[35:16.430 --> 35:17.990]  And this is where it gets a little complex.
[35:17.990 --> 35:18.950]  How am I doing for time?
[35:19.170 --> 35:20.230]  Because I need to...
[35:20.230 --> 35:20.970]  15 minutes.
[35:21.010 --> 35:21.710]  15 minutes.
[35:21.850 --> 35:22.530]  Okay, great.
[35:22.950 --> 35:23.730]  So step one.
[35:24.370 --> 35:28.850]  Once the computer went into ignition, so the computer would say, okay, I'm going to ignite.
[35:28.970 --> 35:32.570]  And there was a whole countdown process, which would happen because you would be turning on valves and stuff.
[35:33.770 --> 35:37.690]  So you would turn the AGC so that it would think it's already in the middle of an abort.
[35:38.610 --> 35:41.410]  So you had this little window of time, and you would say...
[35:41.410 --> 35:44.310]  You would program it so that it would already think it's abort.
[35:44.410 --> 35:48.910]  So this verb 21, noun 1 business, programmed what was called...
[35:48.910 --> 35:49.790]  Again, it was an octal.
[35:49.890 --> 35:51.490]  So you see that 107 at the end?
[35:52.570 --> 35:57.030]  So verb 21 was basically the equivalent of a poke statement.
[35:57.290 --> 36:00.310]  So this basically kind of resolves to...
[36:00.310 --> 36:01.990]  For those of you who remember basic...
[36:01.990 --> 36:03.850]  I'm portraying my Commodore 64 roots here.
[36:06.090 --> 36:07.710]  Verb 21 was basically...
[36:07.710 --> 36:18.190]  This command was equivalent of poke 1010 comma 107, which was put 107, which was the octal for 71, into memory location 1010.
[36:18.670 --> 36:27.850]  And so the way the computer was designed was, if it's already in the middle of an abort, there's no point in starting the abort monitor because you've already clearly done that.
[36:28.110 --> 36:28.970]  You must have.
[36:29.070 --> 36:30.490]  Otherwise, how would you be in the middle of an abort?
[36:31.890 --> 36:33.170]  So great thinking, computer.
[36:33.330 --> 36:34.730]  We're going to outfox you now.
[36:34.930 --> 36:36.250]  So they punched it in this one.
[36:37.770 --> 36:41.110]  And then, however, there is a bit of problem because now you've kind of...
[36:41.110 --> 36:45.910]  You can't put the computer in a bit of a weird place where it will continue its countdown, and then...
[36:45.910 --> 36:47.630]  But it won't throttle up all the way.
[36:47.730 --> 36:54.150]  And if you don't throttle up the engine all the way in the way that the mission designers have done, you will land somewhere else.
[36:54.430 --> 36:58.090]  God knows where, but it will do weird things to where you think you're going to land.
[36:59.670 --> 37:07.290]  So what you needed to do was, at 26 seconds after ignition, on their manual control, they would push the throttle up to maximum.
[37:07.590 --> 37:11.350]  And when I say manual, everything was still being flown through the Apollo guidance computer.
[37:11.450 --> 37:15.250]  There was no true manual in the lunar module, as you understand it.
[37:15.250 --> 37:16.990]  Everything was still done through.
[37:17.070 --> 37:17.530]  But there was...
[37:18.030 --> 37:26.130]  The computer didn't know that if the human being was pressing a certain control, then, you know, let the human being have his way and take it up.
[37:26.810 --> 37:30.410]  So they manually boosted the throttle to maximum.
[37:31.070 --> 37:38.650]  So now they're blazing through the sky with the engine rocket firing at maximum power as they're losing energy and they're kind of beginning to drop towards the surface.
[37:39.350 --> 37:41.050]  And now you have to turn on all...
[37:41.050 --> 37:43.930]  When they first did this, they forgot to the second bit, which I'm talking about now.
[37:44.250 --> 37:51.590]  And when they simulated, they realized the lunar module didn't do anything and just kind of stayed in its position all the way down, which was also considered bad.
[37:52.330 --> 37:54.570]  Because you need to, like, turn and land.
[37:55.570 --> 37:58.050]  So they now needed to turn on the landing software.
[37:58.270 --> 38:00.750]  So again, they did another...
[38:01.530 --> 38:03.590]  They did another hack.
[38:04.530 --> 38:05.650]  And this time they were setting a bit.
[38:05.850 --> 38:11.050]  So verb 20, 25, noun 7 is actually a bitwise operator.
[38:11.630 --> 38:14.550]  Again, 101 and 200 are some more information about that.
[38:14.690 --> 38:16.630]  And the bit that's being changed is one.
[38:16.830 --> 38:20.670]  So you're actually sending a single bit flag inside the memory with that little one there.
[38:22.450 --> 38:24.710]  When they did that, so now they had control.
[38:25.430 --> 38:28.410]  And now it was time to turn off the abort monitor.
[38:29.570 --> 38:33.870]  And so this, again, was another bitwise entry.
[38:34.130 --> 38:34.730]  Yes, bitwise.
[38:35.070 --> 38:37.990]  So verb 25, noun 7, you see it's both bitwise.
[38:38.310 --> 38:39.370]  Different memory address.
[38:39.370 --> 38:40.830]  And now they're setting it to zero.
[38:41.070 --> 38:44.490]  So they're clearing the bit, saying that that abort monitor should never start.
[38:46.410 --> 38:54.590]  And now, again, they want to do a little bit of cleanup, get the proper descent program running that, you know, handles the radar and all that kind of good stuff.
[38:54.790 --> 38:57.130]  And again, it's another poke statement.
[38:57.470 --> 38:58.970]  And so you see, it's the opposite.
[38:59.090 --> 39:01.350]  It's very similar to the original statement they first entered.
[39:01.650 --> 39:06.950]  See, first here they put in verb 21, noun 7, memory 1010, 107.
[39:06.950 --> 39:15.790]  And then they punched in verb 21, noun 1010, 77, which I think was 65, 64, 63.
[39:16.190 --> 39:16.890]  Sorry, yes.
[39:17.270 --> 39:19.550]  For the actual descent commandment.
[39:19.750 --> 39:26.170]  And then you had to drop the throttle back to minimum so that you, the computer no longer started doing the override.
[39:28.190 --> 39:32.930]  So it, you know, there were times when it needed, the computer program needed to go down to partial thrust.
[39:33.230 --> 39:34.570]  And that was when it did that.
[39:35.510 --> 39:40.710]  By setting the throttle to manual, you let the computer really do the full range of throttle commands.
[39:41.850 --> 39:43.350]  So, coming towards the end here.
[39:43.510 --> 39:43.990]  Further reading.
[39:44.450 --> 39:51.810]  For like a really great, you know, what it's like to be a flight controller in mission control, Cy Liebergoth, Apollo ECOM.
[39:51.890 --> 39:55.870]  He was the ECOM who was there when Apollo 13, when the fuel cells blew up.
[39:56.090 --> 39:57.770]  And he's just an amazing guy.
[39:57.770 --> 39:58.710]  And he's got a great story.
[39:58.790 --> 40:02.770]  And he does talk about where he thought about, at one point, of just getting up and walking away.
[40:02.770 --> 40:08.870]  And then you'll notice that the consoles in, in the mission control have these great big handles.
[40:09.330 --> 40:11.970]  And they're, you could, so you could lift them out and do servicing.
[40:12.170 --> 40:13.950]  But the controllers call them panic handles.
[40:14.230 --> 40:23.050]  Because in moments like this, and Liebergoth says, hey, you just, you just grab those handles, and you hold on, and you just, and then you get it together, and then you go on.
[40:23.270 --> 40:29.550]  But he's got, in his, in his, in his book, he's got like printouts from what you saw on his screens during the whole disaster.
[40:29.710 --> 40:30.350]  It's really great.
[40:30.450 --> 40:31.130]  But you can get it probably to come.
[40:31.490 --> 40:35.230]  Don Isles, who's actually an artist now, does installation art.
[40:35.690 --> 40:43.130]  And he's got a lot of detail about the Apollo, the AGC, and a lot of the squirrely things that happened on Apollo 11 and so on.
[40:43.230 --> 40:44.090]  So that's worth recommending.
[40:44.610 --> 40:49.710]  Kind of recommend enough, the Apollo Lunar Surface Journal, which Frank also contributes to.
[40:50.030 --> 40:54.350]  And just a huge, huge array of really great stuff.
[40:54.650 --> 40:59.850]  Another, again, because I said at the start that I'm really interested in engineering-centric histories of the space program.
[41:00.730 --> 41:04.630]  Apollo by Murray and Cox, is a very engineering-centric book.
[41:04.790 --> 41:05.910]  It was out of print.
[41:06.030 --> 41:07.630]  I think they had their own private printing.
[41:07.690 --> 41:09.450]  You can still get copies here and there.
[41:09.690 --> 41:10.770]  It really is awesome.
[41:10.870 --> 41:15.750]  It goes right through from Mercury, some of the Mercury days, through the Apollo program.
[41:16.270 --> 41:24.570]  Digital Apollo, which is a guy called David Mendel, which really goes into a lot of the tensions between the astronauts and some of the engineering and the decisions.
[41:24.630 --> 41:26.830]  And how much control should the astronauts have?
[41:26.930 --> 41:29.230]  Because the astronauts hated being spam at a camp.
[41:29.610 --> 41:32.490]  They wanted lots of control, as much control as possible.
[41:32.790 --> 41:38.950]  And for instance, there was a serious idea for a while that an astronaut would manually...
[41:39.330 --> 41:45.370]  So when you had these boosters, whether it be the Saturn V or the Atlas or Titan, whatever it was, that the astronaut would fly the Ascent program.
[41:46.690 --> 41:47.130]  Manually.
[41:47.590 --> 41:49.490]  in the middle of like a 12G launch.
[41:50.030 --> 41:52.210]  And they were dissuaded from this.
[41:52.430 --> 41:58.390]  But there's a huge tension in the system between what you automated and what you didn't automate.
[41:58.730 --> 42:01.350]  And David Mendel really talks about that.
[42:01.570 --> 42:03.830]  And secondly, I have to pimp a little bit of my own book.
[42:04.910 --> 42:14.450]  And I have a chapter in this book which talks a lot more about Apollo 13, which is called Footprints in the Dust, which just came out, which also covers a whole bunch of the other missions, and does talk about the astronauts, because they were awesome.
[42:14.570 --> 42:26.670]  And I think Apollo 13 showed that there really was a lot of wisdom in having test pilots, because those were guys who could, after days of no sleep, enter in command sequences furiously without any errors.
[42:27.470 --> 42:29.610]  And so I think that is about it.
[42:29.670 --> 42:31.830]  So I would love to take any questions.
[42:40.670 --> 42:41.630]  Good God.
[42:41.630 --> 42:42.310]  There's more of you.
[42:42.610 --> 42:42.850]  Okay.
[42:44.470 --> 42:45.630]  So any questions?
[42:46.350 --> 42:48.070]  I think there's a microphone.
[42:48.290 --> 42:48.370]  Yeah.
[42:48.870 --> 42:49.170]  Thank you.
[42:49.690 --> 42:55.510]  Do you know if a Dymo-label tape-making gun was part of the onboard flight equipment?
[42:55.850 --> 42:57.670]  I do not know the answer to that question.
[42:57.770 --> 42:58.350]  I'm very sorry.
[43:00.670 --> 43:03.850]  I noticed you had a little bit of a mix-up in the beginning as far as...
[43:03.850 --> 43:04.270]  What was it?
[43:04.470 --> 43:05.310]  SCE, was it?
[43:05.430 --> 43:05.790]  Or NCE?
[43:05.830 --> 43:06.530]  SCE to AUX.
[43:06.770 --> 43:09.990]  Weren't the astronauts, like, mostly most ex-military people?
[43:10.110 --> 43:13.730]  Why didn't they do something like, you know, Sierra Echo, you know, that sort of nomenclature?
[43:13.770 --> 43:15.810]  Well, some of them, they also came from different cultures.
[43:15.810 --> 43:18.410]  I mean, yes, most of them were military pilots.
[43:18.650 --> 43:19.650]  Why they didn't do Sierra Echo?
[43:19.870 --> 43:21.010]  Because a lot of it was in...
[43:21.010 --> 43:22.670]  Well, they had developed their own jargon.
[43:23.730 --> 43:26.250]  And so they tended not to use all the phonetic stuff.
[43:26.410 --> 43:27.270]  Usually there was some other words.
[43:27.370 --> 43:34.750]  So, for instance, the idle state of the Apollo guidance computer was program 00.
[43:34.750 --> 43:38.310]  So rather than saying Peter 00, they said, oh, the computer's in poo.
[43:39.230 --> 43:39.870]  So they...
[43:39.870 --> 43:41.430]  So the jargon was just...
[43:41.430 --> 43:42.430]  They developed their own jargon.
[43:42.470 --> 43:44.690]  I should say maybe that led to these kind of trip-ups.
[43:45.230 --> 43:45.710]  Yes, sir?
[43:45.970 --> 43:53.610]  Your talk highlights the value of being able to construct an expedient solution to an unexpected problem using the tools at hand and having somewhat discrete components.
[43:54.130 --> 44:01.670]  Do you think the systems they would put in aerospace and space mission hardware today would allow for that kind of thing as easily, or...
[44:01.670 --> 44:06.310]  I mean, do you think that they're thinking of that ahead of time that we might need to do things that we didn't know we would need to do?
[44:06.350 --> 44:06.910]  Well, I think...
[44:06.910 --> 44:11.270]  I think no, because this was the tension with the AGC in that it was basically a black box.
[44:11.430 --> 44:13.870]  And this was the complaint that maybe you couldn't do anything with it.
[44:14.190 --> 44:18.590]  And it turned out to be that the software is flexible enough that they were able to get around it.
[44:18.770 --> 44:20.690]  So I'm not...
[44:20.690 --> 44:26.530]  I think that, yes, you'll see stuff being gunged together that maybe you might have liked to have.
[44:27.990 --> 44:34.290]  In a previous generation, you might have tweaked it, but it will open up other opportunities for different types of solutions.
[44:34.610 --> 44:36.990]  So I don't think we'll lose a huge amount there.
[44:37.170 --> 44:37.450]  Sorry.
[44:37.710 --> 44:46.530]  I was wondering if the only channel for communicating, especially all the commands to what was voice, or if they also had a redundant data channel.
[44:46.630 --> 44:48.150]  No, they could have done it to the data channel.
[44:48.330 --> 44:50.010]  And David Mindell points out...
[44:50.010 --> 44:59.430]  David Mindell, I think, talks to a guy called Eldon Hall, and he was the real genius behind the design of the AGC.
[45:00.110 --> 45:06.990]  And Hall says this is proof that the astronauts never really trusted the computer because they could have sent up those commands on the data channel.
[45:07.230 --> 45:13.750]  But because the astronauts had read it out and were entering it manually, it made the... they felt that they were in control of the situation.
[45:14.050 --> 45:19.010]  And, you know, you kind of want people who are, you know, hurtling around the moon to feel psychologically comfortable.
[45:19.830 --> 45:21.510]  And so, you know, okay.
[45:21.990 --> 45:23.770]  I was also... I just can't quite remember.
[45:23.990 --> 45:26.570]  Is the moon approximately 10 light seconds away?
[45:26.810 --> 45:28.410]  Or how much of a delay is there?
[45:28.410 --> 45:29.250]  No, the moon isn't that long.
[45:29.370 --> 45:31.930]  I think it's more like a second and a half.
[45:32.770 --> 45:33.490]  It's like four.
[45:33.570 --> 45:35.490]  Yeah, it's not that long.
[45:35.850 --> 45:39.270]  Because I imagine saying, what, and having to wait for eight seconds.
[45:39.270 --> 45:39.350]  Right.
[45:39.350 --> 45:40.850]  You do get these little beeps sometimes.
[45:41.090 --> 45:42.570]  But it was close enough.
[45:42.770 --> 45:43.930]  Last thing we go into the moon, it was close enough.
[45:43.990 --> 45:45.810]  Those mission controllers were really...
[45:45.810 --> 45:47.350]  They were very tightly in control.
[45:47.350 --> 45:51.050]  They were always operating systems on the spaceships that the astronauts didn't know about.
[45:51.570 --> 45:59.990]  And so, obviously, when we think about going to Mars or other places, you're going to have a slightly different approach that's going to be teased out.
[46:00.150 --> 46:00.510]  Yes, sir.
[46:00.950 --> 46:01.310]  Morning.
[46:01.390 --> 46:01.950]  Excellent stuff.
[46:02.130 --> 46:02.550]  Thank you.
[46:02.670 --> 46:06.430]  One of the things I loved hearing, though, is that they take this stuff very, very seriously.
[46:06.570 --> 46:08.870]  But the engineers, I'm told, also had a bit of a sense of humor.
[46:08.990 --> 46:20.550]  So the Grumman engineers, after they had the Luna module had been rescued, had used to rescue, they sent a towing bill to Northrop Grumman, the manufacturer of the other component that had failed for $400,000.
[46:21.670 --> 46:22.430]  It's brilliant.
[46:22.430 --> 46:26.250]  I think it was North American.
[46:26.510 --> 46:34.190]  I think North American then sent back another bill saying, yeah, well, you owe us for, you know, six and like nine lunar missions or whatever it was.
[46:34.330 --> 46:35.790]  They had dueling memos for a while.
[46:35.910 --> 46:36.970]  And they did have a sense of humor.
[46:36.970 --> 46:44.050]  But there's this great, especially after Apollo 13, there's this great, you can get it in some places, where like gag reel, they had made for the big party.
[46:44.070 --> 46:44.890]  They had afterwards.
[46:45.270 --> 46:52.510]  And there's a very famous point early on where very early in the disaster, Cy Libra got the e-comm says, you know, people are a little concerned about what's happening.
[46:52.610 --> 46:54.550]  And he says, well, I think we may have had an instrumentation problem.
[46:57.090 --> 47:03.630]  Which sounded tremendously funny in retrospect, you know, afterwards when they knew that the whole thing had blown up to say it was an instrumentation problem.
[47:03.770 --> 47:07.450]  But you'll see from Apollo 12 when it really had been an instrumentation problem.
[47:08.010 --> 47:09.150]  So they were trained.
[47:09.390 --> 47:11.090]  I talked to Ernie Aldrich, who is Heather.
[47:11.210 --> 47:15.970]  And this is, they were trained to think first in terms of instrumentation issues, because you had a lot of ratty data.
[47:16.050 --> 47:23.010]  And you didn't want to like flush an entire mission because, you know, one dodgy signal on this very long transit route.
[47:23.010 --> 47:25.130]  But they played that loop over and over again.
[47:25.270 --> 47:28.550]  And later on, Gene Krantz was talking to somebody and said, well, I don't understand that Cy.
[47:28.710 --> 47:31.070]  And they would play that over and over and over again.
[47:32.070 --> 47:33.830]  So, yeah, they were pretty funny.
[47:34.570 --> 47:34.910]  Yes, sir.
[47:34.990 --> 47:39.190]  On the last one that you went over, I think Apollo 14, two questions.
[47:39.770 --> 47:43.150]  First off, what was the time period there?
[47:43.190 --> 47:44.710]  I think you mentioned it, but I forgot it.
[47:44.830 --> 47:46.930]  So how long did they have to come up with all those solutions?
[47:47.010 --> 47:50.490]  Oh, while they were floating around the moon, it was about 90 minutes, I think.
[47:50.750 --> 47:51.110]  Okay.
[47:51.110 --> 47:56.410]  And then second, how did they determine, or did they determine during the event that it was a solder ball that was floating around?
[47:56.410 --> 48:01.650]  Oh, they suspected it was a solder ball because of all the tapping, the magic tapping, which felt like a solder.
[48:01.750 --> 48:02.990]  But it really wasn't.
[48:03.330 --> 48:07.430]  And then they, because the ascent module did not come back, so they couldn't open that switch.
[48:08.350 --> 48:10.630]  But everybody agrees that that's pretty much what it was.
[48:10.850 --> 48:11.030]  Okay, thanks.
[48:12.030 --> 48:14.550]  Okay, I think I am out of time.
[48:14.670 --> 48:16.110]  But thank you all very much for coming.
[48:16.770 --> 48:21.310]  And I say, please go downstairs to the retrocomputing area, check out the AGC.
[48:21.510 --> 48:23.510]  It deserves your respect and love.
[48:24.330 --> 48:24.750]  Thanks.
[48:25.770 --> 48:26.590]  Wait for all afterwards, yes.
[48:26.590 --> 48:27.190]  Wait for all afterwards, yes.
[48:27.890 --> 48:28.170]  Thank you.
[48:28.430 --> 48:28.490]  Thank you.