“Houston We Have a Podcast” is the official podcast of the NASA Johnson Space Center, the home of human spaceflight, stationed in Houston, Texas. We bring space right to you! On this podcast, you’ll learn from some of the brightest minds of America’s space agency as they discuss topics in engineering, science, technology and more. You’ll hear firsthand from astronauts what it’s like to launch atop a rocket, live in space and re-enter the Earth’s atmosphere. And you’ll listen in to the more human side of space as our guests tell stories of behind-the-scenes moments never heard before.
Episode 89 is the second part in a special edition of the podcast. John Hansen and Greg Wiseman take the first crack at repairing the only machine capable of playing back the voice recordings of Apollo 11 flight controllers, and they soon find out they’re going to need some help…
If you’re interested in hearing more of this historic audio from the Apollo 11 Mission Control Center check out the Explore Apollo website. For more of our Apollo Podcasts, including part one of this series, check out the “Houston We Have a Podcast: Apollo 50th Anniversary” webpage!
Transcript
Pat Ryan: Houston, we have a podcast. Welcome to the official podcast of the NASA Johnson Space Center. Episode 89, “The Heroes Behind the Heroes, Part 2.” I’m Pat Ryan, and I’m pleased you’ve joined us to pick up on the story of the rescue of an artifact of American history. The audio recordings of NASA’s mission control center during Apollo 11, the first landing of men on the moon. In part one we introduced the story of a professor from the University of Texas at Dallas who studies speech processing and language technology. And wants to move the field beyond a smartphone being able to understand simple questions. In 2012, he wanted to study language as used by a large group of people interacting with one another on a common project or during an emergency. And he found out that NASA’s Johnson Space Center in Houston had recordings of the conversations of the mission control team during the Apollo missions. That would be perfect for his research. And he got a grant from the National Science Foundation to fund the work. But when he got to Houston, he discovered some seemingly impenetrable obstacles to his plan. And that’s where we start part 2 of “The Heroes Behind the Heroes.” Here we go.
[ Music ]
NASA’s Apollo program was created to achieve an outrageous goal. The one set by President John Kennedy in 1961, to land men on the moon by the end of that decade. And return them safely to the Earth. The agency’s scientists and engineers not only had to develop the rockets and the spaceships that would execute the missions, but the systems on the ground that would support those flights. And that included the development of a mission control center where experts could monitor the systems on a spacecraft. And collaborate to resolve any issues that arose. The conversations among the mission control team members during Apollo 11 and the other missions were recorded. Thirty separate channels of them, simultaneously on big reels of one-inch wide audio tape. And then the tapes were pretty much forgotten about for more than 40 years. Until John Hansen asked if he could borrow them. John Hansen is a professor at the University of Texas at Dallas who studies speech processing and communications and biomedical engineering. And he’s a leader in the field of speech recognition technology. He wanted to use those tapes for his study of the speech of groups of people working collaboratively. And he won a federal grant to fund the work. When he approached NASA with this proposal, he got a big surprise.
[ Music ]
John Hansen: Right after we got the grant, we came down to NASA. And I naively said, well, can you show us where you have the audio. And Greg Wiseman pointed us to some boxes of tapes. And I said, we’re in trouble here. Because we thought this was all digitized. And so that was 2013 or so. And at that point we started the process of figuring out how do we actually go about digitizing these things? So that was an engineering problem by itself.
Pat Ryan: Well, but even before that I guess at some point you’ve got NASA who, somebody high up the food chain who says, yes, we’ll be happy to share this with you.
John Hansen: Well, so we talked actually with several folks. There’s one colleague of mine at the University of Maryland who was also involved on this project. And he’s kind of a NASA, or at least Apollo, I don’t know, maybe a closet trivia expert on this. And he kept telling me, John, it’s probably going to take you a year to two years to get permission from NASA to actually go in and digitize this stuff. And so we were, actually I was really nervous. Because we got the proposal, and I thought, well, if we’re going to have to go through a lot of bureaucratic processes here, we weren’t going to be able to actually advance the engineering algorithms and so forth. But I will say we had the best support that we could ever have possibly hoped to imagine here with NASA. Greg Wiseman just opened the door and really gave us support that we could never imagine would have been so helpful in developing this process. Because it wouldn’t have happened without that infrastructure here at NASA.
[ Music ]
Greg Wiseman: My name is Greg Wiseman. I’m an audio engineer here at the Johnson Space Center, supporting the Public Affairs Office. I work in multimedia production and live television for dynamic broadcast coverage.
Pat Ryan: You’ve worked here for 15 years, 20 years?
Greg Wiseman: I worked here since February of 1999.
Pat Ryan: Seen a lot of weird stuff come through here in that time.
Greg Wiseman: I have, I started working around the beginning of the International Space Station. I worked through a big part of the shuttle program. I was here during the Columbia accident. And then worked through return to flight. And now building towards Orion and commercial crew.
Pat Ryan: How did you first become aware of John Hansen and his project dealing with the Apollo 11 mission control tape recordings?
Greg Wiseman: Hansen had the project with the National Science Foundation. And he had reached out to the JSC history program who eventually directed him to our department. And I think, if I remember right, Hansen had seen an article about our department in “Mix Magazine.” They had done a short story about some of the things that we do here in the audio control room. And they had included a picture of some of my colleagues standing in front of the SoundScriber. So he became aware of who we were, and he reached out. Made contact with my supervisor John Stoll, who then handed the project over to me.
Pat Ryan: When you met with Hansen then, what was your sense of what he was after?
Greg Wiseman: So he shared with me his proposal he had submitted to the NSF. And so I had a basic understanding of the academic research that he was pursuing with the project. Which was interesting. But what really interested me was to, was that he had funding that he was willing to apply towards solving some of the problems we had in being able to do this. Part of that was getting the SoundScriber back in working order.
Pat Ryan: Now, let me stop you there. And let’s do a little clarification. You’ve mentioned SoundScriber. And we hear the word “SoundScriber” in this story before. Can you describe, characterize, explain that machine. What is the SoundScriber?
Greg Wiseman: The SoundScriber is an analog audio tape machine. It plays back one-inch 30 track tape. It has two heads for playback. One channel each head. Originally they did have a dedicated head for just reading IRIG time codes so that you could queue up a particular section of tape that you needed based on a timeframe. But the other two playback heads were, they sat on a little moveable block. And the way that you would select which track that you wanted to listen to is you would turn this little crank. And it would physically move the head up or down the width of the tape to line the head up over the track. And it had a little number counter that would tell you which track you were on, approximately. I mean, it wasn’t always super precise. So you dial up which track you wanted. You’d hit play. You’d wait to hear some audio. And then you might need to tweak the crank a little to better position the head over the tape track.
Pat Ryan: So the SoundScriber machine was a playback unit, not the, didn’t record these tapes?
Greg Wiseman: Right. So the way that these were used back in the 60s and the early 70s, it was part of a larger system. They had two 30 track recorders that were recording 24/7, at least during a mission. Recording 60 discrete channels of audio. And they had a couple units right underneath them that, they would kick in once the top one was near the end of its tape. So it would get near the end of the tape, the bottom one would kick in. And this allowed for uninterrupted recordings. So they would roll off that tape. Then they would put it on a shelf. And then later on, if somebody wanted to request a copy of a particular channel of audio in a particular timeframe, then they would, the operator would take that tape over to the SoundScriber that would load it up. They could queue it up to whatever time that they were looking for. And then they could make a copy. They could roll that off probably onto 1/4-inch tape.
Pat Ryan: This is, to call it a playback, call it a tape recorder that plays back audio is, doesn’t really give you the sense of what this enormous machine is.
Greg Wiseman: The SoundScriber is huge. It’s this pale green, painted monster. I mean, it would be tricky to fit this thing in the back of a pickup truck. It’s about 4.5 feet tall. Another four-feet wide. And maybe 2.5-feet deep. Very heavy. It’s certainly not a compact piece of equipment.
Pat Ryan: That’s the machine. You also had earlier made reference to the tapes, as if you’re really aware that these old recordings exist and have worked with them before.
Greg Wiseman: We have a half dozen of these 30 track tapes in our collection from other missions. Such as Gemini. And we have a few from Apollo-Soyuz as well. Back when I started in 1999, that was one of my first projects was to try and digitize these tapes. Back in ’99 super cheap hard drive space wasn’t a thing yet. So we were going to archive these onto digital data tapes. Which was our common archive format back then. But I really didn’t have a lot of success because the SoundScriber just wasn’t fully functional at that point. I would load up a 15-hour tape when I left at the end of the day, intending for it to roll all night into our digital archive system. But when I would come in in the morning, the tape would be stuck about halfway through. There just wasn’t enough friction left on the pinch roller to pull the tape across the head. So that was a problem.
Pat Ryan: Wiseman’s problems with the SoundScriber ended up being mentioned in an engineering periodical. And that’s where it caught the eye of Larry Vrooman, who had worked on these very machines at the Johnson Space Center in the late 1970s and early 80s.
Larry Vrooman: I tripped over an article in “Mix Magazine.” It’s a professional audio magazine, professional musician audio magazine. Really more about studio support personnel. And I tripped over an article in there about Greg. They had interviewed Greg about his project. And back at that time, he was struggling with, for example, the pinch roller.
Pat Ryan: Wait a minute, there’s that phrase again. He was struggling with the what?
Larry Vrooman: Pinch roller.
Pat Ryan: I feel compelled to pause here. What is a pinch roller? And why does it matter?
Larry Vrooman: Well, the rubber is a compliant tacky surface. And the capstan, the metal machine, the capstan is not. So the capstan is rotated at a speed that the surface contact with the tape achieves the desired tape speed. And then the pinch roller simply ensures intimate contact. And the pinch roller actually does the pulling of the tape.
Pat Ryan: And so, if anybody has got or remembers a tape recorder, there’s a black rubber roller that comes down against that slender metal, silver metal thing. The capstan that comes out. And when the two come together, that’s the actual force that pulls the tape through.
Larry Vrooman: Yes. Yes, that pulls the tape through. And then both reels, which have motors behind them, apply back tension so that the tape is pulled across the heads with relatively even tension.
Pat Ryan: And that’s in both the case of when a recording is made and when it’s played back.
Larry Vrooman: Yes. The actual tape movement in those two instances should be identical.
Pat Ryan: And you want that so that it sounds like, in the playback, it sounds like what it did when it happened.
Larry Vrooman: Yeah, you want an accurate reproduction of what was recorded.
Pat Ryan: And, hence, why having a pinch roller and the machines, the motors that turn those things that were in good shaped and worked as they were supposed to, was critical in order to get the playback.
Larry Vrooman: Absolutely.
Pat Ryan: Okay, that’s what a pinch roller is. And now, on with the countdown.
Larry Vrooman: I e-mailed him. I just did some research and e-mailed him. And I, it occurred to me that I needed to e-mail him in such a way that I would get his attention. So he wouldn’t think I was just anybody. And basically I said that, you know, if that thing hasn’t been thoroughly cleaned, it’s still got my fingerprints on it.
Pat Ryan: Wow.
Larry Vrooman: And I also offered my services to help work on the machine. Because, in fact, I was, in fact, at one time responsible for that particular machine.
Pat Ryan: So—
Larry Vrooman: That’s how I got started. I found an article in “Mix Magazine” about his project.
Pat Ryan: Tell me how, about your involvement then. Tell me from that point, what were you doing?
Larry Vrooman: Well, I contacted Greg about what he needed to, if he needed help. And he said he did. Because he didn’t have the training and the experience to do that, and I do. So we arranged a trip for me to come to Houston to take a look and analyze and evaluate the machine. And it was pretty much the way I envisioned. That it needed a remanufactured pinch roller. And it had a couple other minor issues that we worked through. And we swapped some parts from the old, the two, he had two machines.
Pat Ryan: Two of these SoundScriber playback machines.
Larry Vrooman: Two of the decks.
Pat Ryan: Right.
Larry Vrooman: I had never seen the second one. Because in operation, when it was mounted on a second floor of mission control center, it was in its own cabinet. And I had, and it always worked. And I don’t recall ever fixing anything on it. And it never needed any service. So I’ve never seen the second machine. Turns out it’s not fully complete. I don’t know if they, if it had ever been in service. Maybe in the early days maybe it was. But we, in fact, swapped a few parts back and forth between the two machines. And got the one we want to work actually working once we replaced the pinch roller.
Greg Wiseman: The pinch roller is essentially just a little rubber wheel. And this one, the rubber over years had broken down. It had become slick. And it couldn’t, you kind of depend on that tackiness of the rubber to get the right grip on the tape so that it can pull it across the heads. And this one couldn’t do it anymore. So Larry is an analog tape enthusiast. He has a lot of experience and contacts who have expertise. And he couldn’t replace the rubber, but he had a friend that could. So we disassembled the pinch roller from the SoundScriber. And sent it off to his friend in Michigan. A guy named Terry Whitt. And it was Terry who replaced the old rubber with newly vulcanized rubber. And then he had it machined down to just the right size for the SoundScriber. And once we got it back, we could tell there was a big difference. It was tacky. And once we got it back installed, it could easily maintain the proper friction to pull the tape across the head. Then we sort of tore down some of the machine. Lubricated any kind of moving parts. I think we replaced a belt. Which we were able to do that by cannibalizing from another broke down machine that we had. I think we also swapped out one of the supply reel motors. And then Larry just kind of went over the machine, making little adjustments. Tightening screws. And just getting it back into full working order.
Pat Ryan: How long did it take to get it back into operating condition?
Greg Wiseman: I think it took about three days. I think Larry visited three times. He made pretty short work of it because he knew what he was doing. He worked on these machines for years. So he was able to quickly diagnosis problems and knew what to do to get it functioning again. And luckily we didn’t have any major part breakdowns. If we had, getting replacement parts would have been really difficult or next to impossible. The few things that we did need, we were able to scavenge off the other SoundScriber that we had that was basically just pieces in a box.
Pat Ryan: Did you ever think of contacting the manufacturer and say, hey, help us out?
Greg Wiseman: Based on some Internet research, it looked like to me that the SoundScriber was discontinued in the late 70s. So parts would have been hard to come by. And Dr. Hansen had done some research on his end. And from what he could tell, we may have one of the only machines left in existence. So we figured we were going to be on our own.
Pat Ryan: So the SoundScriber is now operating. It’s playing back tapes. Although it is doing that one channel of audio at a time. And there are still some hurdles up ahead. But Wiseman was jazzed about the situation which he found himself.
Greg Wiseman: This was something that I had wanted to do back in 1999. This was one of my early projects. So finding someone who had some funding and could help solve some of these problems, yeah. I mean, it got me really excited because I’d been sitting on this project for a while.
Pat Ryan: So what was your first step? How did you go about figuring out how to refurbish the machine?
Greg Wiseman: So once we knew that this was something that we were going to take on, after some initial meetings with Dr. Hansen, we discussed some of the challenges that we were facing. And one of which was the playback head. As it was, the SoundScriber could only play back two channels at a time. The other problem was getting the actual Apollo tapes on hand. Because that mission was specifically identified in his NSF grant. We didn’t have those tapes NARA did. And the process for requesting those tapes involves pouring through handwritten transfer documents from the late 70s that have been scanned and stored electronically. And even if you’re able to narrow it down and find a document that might describe the correct box that these tapes would have been stored in. It was a little cryptic because the transfer document will just say something like “historical audio tapes.” But nothing as to what’s on the tapes or even what format of tape. I mean, I was looking for one-inch 30 track specifically. I didn’t want 1/4-inch. I mean, that’s, that wouldn’t have helped me. So I was looking for anything that was identified in that way. But, yeah, it was very much like a needle in a haystack. Dan Rooney at NARA was able to give a big assist in helping us find specifically what we were looking for. But in any event that was my first action item.
Pat Ryan: I assume that that worked fine. National Archives said, sure, we’d be happy, and they delivered. What kind of condition were these tapes in when they’re, you know, 45 plus years old?
Greg Wiseman: Yeah. Actually, surprisingly, they were in great condition. And we were able to play them right away. Which isn’t always the case with older tapes. At some point in the mid-70s, the tape manufacturers changed the formulation that they used for making tape. And they began using a different kind of glue. And the glue is what binds the magnetic oxide to the physical plastic tape. And over time that binder breaks down. And so whenever you load up one of these older tapes and you try to play it back, it will start shedding that magnetic oxide. And the tape will start to stick inside the machine. It won’t pull smoothly across the heads. We’ve actually got a lot of 1/4-inch tape that suffers from that. So the way to fix it is to bake them.
Pat Ryan: Excuse me, bake them?
Greg Wiseman: Yeah. Literally, we throw it in a special little oven. And we bake them at around 135 degrees. And that sort of reactivates that binder. And reforms that bond between the magnetic oxide and the tape so that you can play the tape back. It’s not a permanent fix. But it’s good enough so that you can at least get the tape to play back one more time and digitize it.
Pat Ryan: And it would seem to me that that would have been a concern too. Even though the tapes were in good shape, they’re still pretty old. And you don’t want to have to risk playing them more times than you would need to.
Greg Wiseman: Absolutely. They’re in good condition. But they’re still 40 plus years old. So we really just wanted to play these back once to digitize them.
Pat Ryan: And to remind myself, we’re talking about, you’re talking about playing them back once so we can get everything off of them. But there’s 30 tracks of audio on that tape. In order to get all 30 tracks, you would have to have a head that can read 30 different tracks all at one time. And you didn’t have that.
Greg Wiseman: Right. The SoundScriber wasn’t designed to play back more than two channels at a time. So we would have had to play back each tape 15 times to capture all 30 tracks. And we didn’t want to put that kind of stress on the tapes. So we had to figure out a way to digitize all 30 tracks in one pass.
[ Music ]
John Hansen: And so once we got, we knew the tapes were there. I had one of my students who came down for about a four-day, five-day type week, to do some testing. We thought, well, let’s put the tapes onto the playback system called SoundScriber. And let’s see what we can kind of get out of this. And the playback system, actually we, again, were just expecting that you would play this back. And there’s 30 tracks. So you would be able to get all 30 tracks off it once. But that was not the case. You could only get one audio track at a time. And this was a challenge for us. Because one of the things we wanted to do is we wanted to make sure that all 30 tracks were digitized at the same time. It was very important to synchronize all those tracks. So this way we could start to look across the tracks at the same instant of time.
Pat Ryan: Let me stop you for a second. The tapes that we’re talking about, physical audio tapes that were recorded in 1969, in the case of Apollo 11. And 30 tracks, there are 30 different sources being recorded at the same time? Actually recorded at the same time? So they’re already synced up in time because they were recorded that way?
John Hansen: Exactly. That is correct. And so—
Pat Ryan: What were the 30 sources? Did you know? Were they always the same?
John Hansen: No. Interestingly enough, NASA, when they recorded this, they had two historical recorders called HR1, HR2. Historical recorder one, historical recorder two. For each of these two recorders there’s an upper and a lower recorder or a tape system; right? And the reason for that is, you know, if you’re kind of recording on one historical recorder, you can’t tell everyone, hey, stop. We got to switch the tapes right now. So they have two tapes. And so they’ll record on the lower system. And then, when the tape’s starting to run out, they’ll start running tape on the other.
Pat Ryan: An overlap.
John Hansen: So there’s a good overlap there. The two recorders, actually, there’s some overlap in them. But there’s some differences. In one of the historical recorders there’s, I think three Department of Defense channels. And in the other one, there’s about five Department of Defense channels, the Air Force or—
Pat Ryan: So all together maybe more than 30 channels.
John Hansen: Actually, collectively, you could look, I mean, there could be as many as 60 channels of recording of—
Pat Ryan: But not so in this case.
John Hansen: Well, so the tapes, each tape actually only has 30 tracks. Of those 30 tracks, track one is called IRIG. It’s the time code. So that’s the actual time that tells you, you know, for mission elapse time, exactly when things are happening. If you try listening to track one, it sounds like a squirrel is running around in your attic or something like this. It doesn’t have any speech in there. So there’s really 29 tracks of audio.
Pat Ryan: And did you know what those 29 sources were?
John Hansen: Yeah.
Pat Ryan: How would you characterize?
John Hansen: Yeah, so the thing that helped us a lot was because NASA’s very good in documentation. And one of things they had is, on every tape, you know, there was a decal or a handwritten, you know, sticker. That basically said when this tape started. When it stopped. And then on the actual box, there’s something called a heat sheet. It actually lists all the tracks. And it tells what is the code for each of those tracks. And it tells you what position the recording was done. So the position typically means the type of handset. Whether it’s a head mounted headset or kind of a telephone type handset. To be honest, I don’t know all of the positions of every handset but.
Pat Ryan: What I’m trying to get at I think is, did you know that, was it, for example, the crew on the moon talking to Earth on one channel? Was it the flight director talking to, was it the recording of the flight loop, the flight director’s loop on another channel? Or was it people talking to each other in that control room? Or people talking to people in other rooms?
John Hansen: So each track basically is, basically you think of it as a loop; right? Now, there’s typically one person that, I’ll say it this way, that owns that loop; right? So there could be, for example, there’s one track that’s PAO. Public Affairs Office.
DOD: This is DOD; PAO.
PAO: Hi, this is Bill O’Donnell over in the news center. The name of the artist that’s on the carrier is Jamieson. Is that not right?
DOD: I have to check that, sir. I believe you’re right though.
PAO: Would you check it and give me call back.
John Hansen: And so that person owns that loop. And so, at least the way I say it. I don’t know if you say own it. But so you’ll have other audio that’s there. Another one is EECOM.
Flight Director: EECOM, Flight.
EECOM: Go Flight.
Flight Director: If we don’t dump any water at all, would we be over 100 percent by entry?
EECOM: That’s confirmed flight. We’d hit at 185 hours.
John Hansen: EECOM, there’s communication that’s going on that loop. But while the communication is going on in that loop, there’s typically other pieces that are folded into that loop. So you might have, for example, Air to Ground that’s actually folded in there.
Capcom: Copy all that, Neil. And we’ve got a Entry PAD if you’re ready to copy, over.
Buzz Aldrin: Houston, Apollo 11. Ready to copy.
Capcom:Thanks, Buzz. It’s the Entry PAD, MPL…
John Hansen: That’s means, when you’re talking to someone in EECOM, you’re communicating with people there. But you can also hear, Capsule communicator, Capcom, talking to the astronauts. They can’t hear you, but you can hear them. And so that’s important because there may be something going on that you have to pay attention to. Now, that’s important for the loop. It’s really, really difficult for the speech technology. Because if you’re attempting to try and do speech recognition and you’ve got Neil Armstrong, Buzz Aldrin and Collins in the background talking to the flight—
Pat Ryan: To the Capcom.
John Hansen: To Capcom, then you got to figure out, you know, who is talking to who. And where the things are. Now, fortunately, the air to ground is typically at a lower volume than the primary speakers in that loop. And so for that reason we tend to do a better job of getting just that core speech.
Pat Ryan: So I understand then, the sources, the 29 sources which might have been on either one of the two machines are the loops that are owned by the operators at the different positions in the flight control room?
John Hansen: That’s correct.
Pat Ryan: Is there a conversation involved, so that would be, to use your example. The Public Affairs Officer could be talking to other public affairs people.
John Hansen: That’s correct, yeah.
Pat Ryan: In their background, what they call the back room. The same thing is going on from each of the other positions. And you’re recording each one of all of those positions.
John Hansen: They’re all being recorded, yes.
Pat Ryan: And they’re all, may have the air to ground on top of them.
John Hansen: That’s correct.
Pat Ryan: They may all have the flight director loop on top of them.
John Hansen: That’s correct.
Pat Ryan: The, okay.
John Hansen: So there’s a lot going on. I will say that, when we first played some of the first tapes. What kind of surprised me, we picked up, I thought one that was probably PAO. And so I was a young kid at the time. But we actually picked up ABC, NBC or CBS news feed on these tapes. I didn’t think that you would actually have heard that. And I heard some commercials that I remembered from 1969, you know, there. And I thought, well, this was kind of odd.
Pat Ryan: Breakfast cereals.
John Hansen: Yeah, yeah, exactly.
Pat Ryan: Dishwashing soap.
John Hansen: Smoking cigarettes.
Pat Ryan: Yeah.
John Hansen: They actually had advertisements for that.
Commercial 1: I’d walk a mile for a Camel.
Commercial 2: If you don’t buy RCA, you may be buying an obsolete color TV.
Commercial 3: [ Singing ] Take me along with you!
I love you little cutie but the office is my duty!
Commercial 4: Featuring special savings on specially equipped Fury. Valiant. Barracuda. And Belvedere models.
Commercial 5: When you’re out of Schlitz, you’re out of beer.
Pat Ryan: Well, as a PAO commentator, I’ll tell you that the audio from news channels is among the different loops that people in their can listen to. So that would explain why they were there.
John Hansen: Okay.
Pat Ryan: Not making your surprise at hearing it or finding it any better. So the 30 tracks on each of these tapes are the conversations that are going on within the flight control room. As well as conversations involving flight controllers and other people who are in different rooms.
John Hansen: Yeah, back room support centers.
Pat Ryan: All that is going on at the same time.
John Hansen: Yep.
Pat Ryan: And you’re finding that these reels of audio tape that have 30 tracks on them is being presented to you as, okay, you wanted it. Here it is.
John Hansen: Yeah, we got, so these were all analog tapes. And we had to figure out a way to digitize them. Once we digitized them, then we had to develop technologies that we could use to try and do what we call diarization. So I won’t cover that just yet but we can talk about it.
Pat Ryan: Yeah, because I want get you to explain, to say, well, we figured out a way to digitize them and, you know, story’s over. But there was a lot more to it than that.
John Hansen: Yeah. The digitizing process was something that I have to say was an unexpected challenge. Because when that came up, the fear I had was there’s only one SoundScriber system that I know of. And it sits in one building, number two, here at NASA. And we didn’t know if another one had existed. And so I had contacts and friends that worked in other audio spaces. The CIA actually maintains a library of all types of recording and storage systems that have existed from the beginning of storage.
Pat Ryan: Because of course they do.
John Hansen: Yeah. Well, yeah. And so I actually contacted their archive to see if they had a SoundScriber system. My thought was, well, if they had a SoundScriber system, could we rent it? Borrow it? Or somehow get it to University of Texas at Dallas. To see if we could use that to digitize them, you know, on our space. CIA doesn’t have one. And so I did do a little research on SoundScriber. SoundScriber actually was an audio logging system, a company that basically did audio logging. It was primarily for police and for emergency. Like calling 911, rescue squad and so forth in the 60s. But tapes were expensive at that time. So when it got recorded, they would recycle and reuse those tapes over and over again. So it was basically a way to do a logging of the audio. NASA used it, obviously, for different purposes. It was, obviously, to capture the audio for Apollo. But what kind of, you know, really changed my perspective of things is when I went to use the SoundScriber the first time. And I said, well, how do you switch between tracks? And Greg said, well, you have this little handle. And you rotate this little wheel. And what the wheel does is it manually moves the read head up and down against the tape. So, you know, and when you rotate it, a little counter actually tells you which track the head is supposed to be on. And it mechanically changes from track two to three to four. And so you thought, okay, this was built in 1967, ’68. So what are the challenges here? Well, are the motors that actually move these tapes, are they the original ones from the 1960s? What about the actual read head, is that from the 1960s? There’s a little piece of rubber like a wheel that’s supposed to put friction, tension on the tape.
Pat Ryan: The pinch roller.
[ding sound effect]
John Hansen: Yeah, pinch roller. And was that the original one? Because rubber oxidizes over years. So these were small little things where, you know, if you are thinking you’re just going to have to develop speech technology. You don’t think that you’re actually going to have to roll up your sleeves and be an engineer and try to solve these things. We were concerned about digitizing only one track at a time. This was just not an option. Because I start doing some calculations. And assuming that you could digitize, you know, for a good portion of the day. Looking at, you know, maybe 30 tracks. You know, probably 10,000 hours or so. It would probably take you over 100 to maybe 150, 170 years to digitize, like, the Apollo program.
Pat Ryan: The entire Apollo program.
John Hansen: The Apollo program. And I thought, well, we don’t have 170 years. We got to do this faster so.
Pat Ryan: Our grant’s only for three years.
John Hansen: Yeah, it was only for three years, so.
Pat Ryan: How many tapes are we talking about?
John Hansen: So for Apollo 11, my understanding, if I remember correctly. And it’s been a number of years. I think it was about 22 or 24 tapes.
Pat Ryan: And each tape had, not how much tape, but how many minutes?
John Hansen: Each tape has about 14 hours. And when you digitize, you must digitize in real-time. So that means you literally play back the tape. It takes 14 hours to play it back. If you now have—
Pat Ryan: On what, track one.
John Hansen: On track one, yeah. So now you’re going to have to take that same tape and literally play it back 14 hours. So, basically, if you could digitize 14 hours in a day, it would take you 29 days to digitize one tape.
Pat Ryan: Because you don’t want to digitize IRIG.
John Hansen: That’s right. Right. Now, and also the thing that you have to worry about is, once you digitize it, now you’ve got to go back and figure out how do you actually align all those tracks? Because if you could digitize them at the same time, well, now they’re automatically aligned. If you digitize them one at a time, they’re not necessarily going to be aligned.
Pat Ryan: You’d have to find something that happened in each track in order to coordinate the playback.
John Hansen: Yeah. Actually would be the obvious solution that you.
Pat Ryan: Nothing, but if not the master of the obvious.
John Hansen: And one of challenges also is that it’s not just the fact that you’re just playing back a magnetic tape. The issue is that, over that 14 hours, you can kind of think the tape actually expands and contracts. So the playback is not necessarily exactly at the same rate. So when you, when we digitize, you know.
Pat Ryan: So that wouldn’t work.
John Hansen: Yeah. So what will happen is that, if you play this back, you know, 29 times. At the end each one is going to be slightly longer or shorter than the rest.
Pat Ryan: I’m thinking that you got a bigger problem than that. You’ve got 50-year-old audio tapes that you’re, think are going to hold up to 30 14-hour playbacks.
John Hansen: And it’s quite common, I’ve done some work with audio archiving. We had an NSF project about ten years earlier that was called the National Gallery of the Spoken Word. So we had spent five years on that project. And it was looking at audio that had been captured from 1890s up into the, 1995 or so. And so in those types of tapes, you literally have an oven. And that was one of the things I was actually familiar with in the space that Greg Wiseman was working in where there’s an oven. So if you have an analog tape, sometimes you have to bake the tape. Because there could be moisture or some other types of impurities. Or maybe the tape might stick together. So baking it actually, hopefully, allows the tape to travel more smoothly through the playback system.
Pat Ryan: So confronted with the option of 170 years worth of playback in order to digitize, you had to find another solution.
John Hansen: Yeah. And so the thing we looked for is we wanted to come up with a way of maybe designing a new read head. It was a risk we figured, well, how could we do this? So we were pretty lucky because, when we looked at the read head that was actually sitting in SoundScriber, it had a manufacturer’s name on that. So from that I ended up, it was like CSI type of thing. I ended up investigating where I could go and try to find the manufacturer of that. And we kind of went through about four or five different paths. I ended up finding a company in New Jersey that actually specializes in customized recording equipment.
[ Music ]
John French: My name is John French. And our company name is JRF Magnetic Sciences, Inc. And we started this little company back around 1979. And we started off just manufacturing little components for tape recorders. And also refurbishing the tape heads.
Pat Ryan: And your business is still focused on analog recording.
John French: Almost exclusively. We service, let’s see. We service recording studios. Both commercial big recording studios and also literally small personal studios that musicians run and things like that. We do a lot of service for archive facilities. One of our bigger customers is Library of Congress. We do a lot of work for, oh, colleges, institutions that have large libraries that they are archiving. Film. That’s in that end of things. Then we do a lot of work with the government. We built some special machines for the FBI and the military. And custom head assemblies for film. Let’s see, what else?
Pat Ryan: It’s quite a range of things that you’ve got. You’re touching in a lot of different businesses.
John French: I know. And it’s basically who we are. I’m mean, we’re a very small company. But, boy, we are busy.
Pat Ryan: Tell me how you and your company became involved or even aware of this deal with the Apollo mission control tapes in Houston.
John French: Well, that’s, it was interesting. We received a call from a student. And I believe he was a student of John Hansen’s there at the University of Texas. And when he first called, he just told me that he was trying to play back some channels on a 30-channel piece of recorded tape. And I said, well, you know, I’m going to need a little more than that. Can you tell me what the machine is? What, and so we went back and forth. And I think it probably was three or four phone calls that I finally found out that what he was really, he never did mention it in the beginning. But he told me that he was working with NASA on the Apollo tapes. Which, you know, obviously piqued my interest pretty significantly.
Pat Ryan: The tape that they were trying to deal with had 30 tracks of audio. Well, actually one was for IRIG.
John French: Right.
Pat Ryan: And their problem was that they were trying to find a way to get it to play back more than one or if not all of those tracks simultaneously?
John French: Correct.
Pat Ryan: Is that an issue that made sense to you that you got that’s something I can get my teeth into?
John French: Well, so many projects, when we first start talking about them, scare the daylights out of me. But feel, because you have to consider that I don’t have the machine. I have no way of testing anything. You know, projects usually don’t have the budget that we can fly out to the facility and do testing and that kind of thing. So we are literally, we have to take the information that we get. And then come up with whatever solution we can work out to offer. Now, on this particular project, the initial discussion was that we were going to build a single-channel head that could play back. But they already had that, you know. I mean, that wasn’t adding anything, any kind of an improvement. So what I had suggested at that time was we’re probably going to want to go with a 30-channel head. I did research on other 30-channel systems. And, as I recall, I couldn’t find any. So the first, the next problem was it wasn’t going to be a standard head that we can pick off of, let’s say a Dictaphone or one of those other systems. Because we’re looking for 30 channels. And I work very closely with a head manufacturer. And I had discussed this with John about making up a three-channel head that was going to be interlaced channels. See, a 30-channel head is on a one-inch is very, very difficult. Channels are very close together. Shielding is a problem. Very, very seldom do you see a 30-channel head. Normally it would be an interlaced with 15 channels on one head being all odd channels. And 15 channels on the playback head which would be all even channels. So that was our initial idea. And, of course, we had the one track head. But we didn’t know what the center-to-center spacings were. There really wasn’t any information that we could sink our teeth into that, you know, we could build a 30-channel head to. I mean, that, again, would be a very expensive proposition. So what we did to start off with was we built a three-track head. Which I don’t remember. I’m pretty sure it was channels 13, 15, and 17. And built that prototype head. Sent it down to NASA. And they were able to hook it up. They were able to use it. They were able to play back three channels. Which, you know, was really a breakthrough. So we had that part of it done. Next, I went back to the manufacturer. And I told them that the concept worked. And that we were going to need two 15-channel heads. Because he refused to try and build a 30-channel head. And spent about two weeks negotiating back and forth with him because there was another big expense involved with this. And that is the tooling. You need a mandrel with 15 slots for the tracks. And then 17 slots for the shields. And he said to put together a piece of tooling like that, it was many thousands of dollars. And I went back to John and told him that, you know, that I really felt like this was going to be a difficult thing to do for any kind of reasonable price. And he said that, well, you know, get me all the pricing. And let’s see where it falls. So I went back to the manufacturer and told him that they wanted to know what the total cost was going to be. And the head manufacturer said, well, on second thought, I really don’t want to build that head.
Pat Ryan: Oop.
John French: I said, what? He said, John, he said, it’s just too much work. There’s huge risk. What happens if one of these 15-channel heads fails in production? You know, I don’t want to take the loss. I don’t so he basically bailed out on me. And I went back to John, I said, well, we got a problem. Houston, we have a problem.
Pat Ryan: Right.
John French: And he said, well, is there anything else you can come up with? And so I told him, well, I was going to do some research. And I would let him know. And we had worked with a company over in Belgium that manufactured heads for Dictaphone. It was Applied Magnetics. And so I went to Dictaphone first. And there wasn’t anybody there that could give me any information on the track placement. You know, the spacings and all that. But I went to the people in Belgium, and I told them the situation. I told them that we needed either an interlaced 30-channel head or a 30-channel head all in one. And it just so happened that he said, well, John, he said, we built a 60-channel with 30 channels on each head. They were interlaced. I said, well, what’s the track format on that? And he gave me the, sent me a drawing actually.
Pat Ryan: Yeah.
John French: And the spacing on the, each channel center to center was extremely close to what we had calculated out the NASA tapes to be. So, anyway, the cost on that head, what happened was, he said, well, you’re not going to believe this. But we only have one 30-channel head in stock. It’s the last one. We’re not building any more. And we only have one. So it’s, doesn’t have its mate to make it a 60 channel. So we’ve been, you know, he told us they’d been keeping it in case there was a system failure. And they could support it. But it had been a number of years, and the thing was still in stock. So they gave us a very good price on it. I went back to John, explained what we had found. Of course, everybody was really concerned that the tracking was going to be correct. That, you know, the crosstalk. And there were a number of conversations regarding whether or not this head would actually work. So the long and the short of it is we decided to bite the bullet. And I ordered the head. The head came in. We made up a mounting base for it. It actually came with cables and connectors. And it really a real find to actually have as much of the head that we needed, you know.
Pat Ryan: Right.
John French: So, anyway, it came in. We mounted it up. I set it up, alignments, you know. So that the azimuth and all was correct in wrap and everything. Sent it down to them. And as I understand it, it just worked beautifully.
Pat Ryan: The new 30-track playback head found and retrieved from a warehouse shelf in Belgium did work beautifully. Just not right away. Hansen’s team was up against another new problem in trying to integrate this piece of hardware into the SoundScriber. That’s where we’ll pick up next time.
[ Music ]
Yes, Virginia, there will be a next time. Next time, here at the same coordinates, nasa.gov/podcasts. Where you can also hear part one of this series. And all the other episodes of our podcast. And when you go there, please look around at the other cool NASA podcasts you can find. Including “Welcome to the Rocket Ranch” from KSC. JPL’s “On a Mission” podcast. NASA in Silicon Valley, from the Ames Research Center. There’s also “Gravity Assist.” “The Invisible Network.” “Small Steps, Giant Leaps.” They are all available right there at the same spot where you can find us. nasa.gov/podcasts. “The Heroes Behind the Heroes” episodes of “Houston We Have a Podcast” are produced by Greg Wiseman and me, with editing and audio engineering by Greg with help from Alex Perryman. Thanks to our guests, John Hansen, Greg Wiseman, Larry Vrooman, and John French, and to Norah Moran and Gary Jordan for helping us pull it altogether. See you next week.