March 29, 1996, Afternoon
KAREN ISHIZUKA: Good afternoon. My name is Karen Ishizuka, and by way of introduction, I'm a media producer and Curator of Moving Images at the Japanese American National Museum. My main interest is the significance, preservation and use of what is commonly known as "home movies," as providing visual documentation and hence a real critical link to the sort of intersection between popular art, memory and official history. Especially for people of color, and communities, and geographic communities and specialized interest groups in this country. We've been pretty much left out of the documentation that happened through newsreels at the beginning of this century, and hence really have to rely on the home mode to provide the visual documentation. I've been involved in transferring a lot of this home footage to video for re-presentation and in media productions. On behalf of all of you, I'd like to thank BAVC for bringing together such a diverse group of people. This afternoon, we have people from three very different areas. The first speaker and presenter is Dr. Peter Adelstein, Peter, would you come up, please? ...plastics, gelatin, photographic film and magnetic tape, and he's worked with areas of brittleness, dimensional stability, film base, chemical stability, and also with standards. He's been involved in the preparation of standards for permanence and physical properties of imaging materials, and he will speak on behalf of the working group on storage.
DR. PETER ADELSTEIN: Thank you, Karen, and it's a pleasure to be
here. There's an old saying amongst vaudevillians, those in
vaudeville, that you should never follow a dog act, and here I find
myself doing exactly that. May I have the first slide, please?
This is the working group which I am reporting for. I've attended a lot of symposiums in my years, and this one is unique. It's the first one that the speaker just doesn't give his own experience, but that of a working group, and I think it's a rather novel one, and one which benefits me. It means that I can speak to not only from my own experience, which is in the field of science, but also benefiting from these individuals whom you see on the slide, who have had experience in actual - being archivists or restorers. Amongst that group, Jim Lindner is here, sitting down on my right here, and if I misquote him, I'm sure he'll let me know. So this really has a dual function. It not only gives me a broader perspective from which to address you, but if I make outrageous statements, it's really not my fault. I'm just reporting what was fed to me.
Now, I think I'll go back. Can I go back? To get some perspective of what we're talking about, let me make a few remarks. We are only concerned with storage with materials that are not stable. Probably the oldest known recording material to man are clay tablets. And they have been around for 6000 years and have been stored in rock piles and underground, in Iraq and Crete, and nobody was very concerned with how they're stored, but when material is less stable, then storage becomes a very big problem. So if you consider banana cream pie, which seems - there was a coconut cream pie this morning, which seems to be the simile - then that is a material which you can't talk in terms of thousands of years or even years, you wouldn't leave on your counter top for six days. And storage for that material is very important. That's why you refrigerate it. And the sad fact is, and you may not like to hear this, is that magnetic tape is much more similar to coconut cream pie than it is to clay tablets, as far as its stability is concerned. So, on a scale of one to ten, if coconut cream pie is a one, and clay tablets is a ten, I would put magnetic tape at a three. So, we're trying to preserve material which basically is not a very stable material. That means that storage is extremely important. To get another perspective of storage and the keeping of recorded information, let's look at this slide here, where we see the materials that have been used by man over the years to store materials, starting with clay tablets, papyrus, parchment, then paper, and then we get into paper with rosin sizing, which is not very good paper, photographic materials, and some of the quick copy materials, and magnetic tape. And the sad fact is, that if we look at the stability of these materials, the clay tablets are extremely good, and then you go to papyrus and parchment, which are certainly not so great. Some of the early papers are excellent, and then you go to paper which is not so good, photographic materials, which can vary from fair to good, depending upon how they're stored and what the materials are. And now we come down to magnetic tape, which is poor to fair. And so, the history of society is such that information is being generated at an ever-increasing rate on materials that are generally less permanent. And that is the sad fact of life. We're producing more and more materials from material which is less permanent. And that is going to continue.
The magnetic - can we open the curtains a little bit more, because I can't see the slides and I really don't know what I'm talking about without them. You can't open the curtains? Thank you. Now I have a little better idea of what I'm talking about. The fact is, that magnetic tape is not an archival medium, and that it makes good economic sense to prolong tape life as long as possible. We heard a great deal about the fact that you're going to have problems with your hardware, you're going to have problems with your media. You do want your media to least at least as long as your hardware. You don't want it to be the weak link, and preservation is much cheaper, and there's no comparison to restoration. In fact, restoration, you can't always do, and if you do it, it's expensive.
The information that I'm going to be presenting comes from several sources. The first is an ANSI standard IT9.23. This is a standard that was just passed the ballot of ANSI. ANSI stands for the American National Standards Institute. It's the official standardizing body in this country, and there is a committee which is concerned with standards on electronic imaging materials, and after many years of work, they produced this particular standard on how to store magnetic tape. Peter Brothers and Jim Wheeler, who are both in the audience today, were contributing members of this group, as were many other archivists and representatives from some manufacturers.
The second source of information is a publication by John Van Bogart of the National Media Lab, it just came out last year, and it was published by the Commission on Preservation and Access, and if you want to get their address, Maxine Sitts is here in the audience, and she's sitting right over there, and she'll be glad to tell you how you can get this publication. It's an excellent publication, and John was also one of the members of our standards group, and the information in both the ANSI standard and in the Commission on Preservation and Access publication is in agreement; there are no conflicts. And, of course, the third source of information is the BAVC panel.
Now, let's go through the primary causes of tape failure. A lot of this was discussed this morning, and in a symposium of this nature, there's bound to be some duplication. So, if I repeat information which was said this morning, and I say the same thing, then you might have some assurance that maybe what I'm saying is true. And if I don't someone will let me know. Now, there are several sources of tape failure. One is the degradation of the binder, the loss of lubricant, distortion of the base, and the loss of magnetic signal. Let me go through each of these fairly quickly. With binder degradation, we heard this morning, Bruce Fellows talking about hydrolysis of the polyurethane binder that is used, and that gives the phenomenon know as sticky shed, where you actually lose the cohesiveness of the binder itself. It becomes sticky, and it results in clogs of the magnetic head, or else in dropouts. You can also cause a stip-slick friction, where the material does not go through the apparatus, the recording apparatus or playback apparatus, in a smooth fashion, or it can cause squealing. And, of course, any one of these things is very, very detrimental.
The second is loss of lubricant. You can have loss of lubricant from several sources. It can just evaporate. Also, every time you run tape over a stationary surface, you do lose a little bit of lubricant. Lubricant, some lubricants may degrade, and there are anecdotal reports that at very low temperatures that lubricants may separate out from the binder, and come to the surface and not be reabsorbed. The question was raised this morning: is there a lower temperature that you should not go? And within the standards committee, we had debates on both sides of this particular issue, and those who say you should set a minimum temperature to go to say there have been cases where lubricant has separated out. Those on the other side of the issue say we've never seen any hard data - produce some hard data. I still haven't seen hard data, but there is a recommendation that you don't go below 46 degrees Fahrenheit. This may be based on false assumptions; it may be absolutely correct.
Now we come into distortion of the base. It was mentioned this morning that some of the older backings - I guess in the photographic field we call it base, and in the magnetic field we call it the backing, but I'm referring to the plastic support, and when acetate was used as the backing material - this is cellulose triacetate - it broke down over time, and produced acetic acid, and this is the well-known vinegar syndrome, which is basically acetic acid, and which has that very nice Caesar salad smell to it, but is very detrimental to magnetic tape. Most people don't want to eat it. Another cause of base distortion is if you store the tape, and the wind is poor or loose. This also was mentioned this morning - or if you have excess tension, plastic is a material which flows. If you take material which is plastic and which your shirts come in, and you can stretch it, or sometimes when you try to tear it, you find you're stretching the material excessively. So, plastic flows under tension. If it's wound very tightly, you can cause distortions, particularly if the tensions are not uniform. And then you can damage by handling. Now, we have kind of a problem here. We want to get greater and greater compaction of data, and to get greater and greater compaction of data, the manufacturers put it on polyester backing, which is thinner and thinner. And the thinner the material is, the more fragile it is, the more subject it is to abuse. So, to get greater compaction and greater durability are exact opposites. You can't have both. It's like asking for a water soluble raincoat - there just is no such thing. And the current support which is used today is a polyester, and a polyester is really a very, very excellent plastic. It's by far the best plastic that can be used. It's been in existence about 45 years. It was first marketed in this country as Mylar, as a plastic sheeting material, and then was used subsequently as a photographic film base, and then as a backing forming a tape. It's strong and it's very stiff and it's tough. But that doesn't mean to say that it can't be abused.
Now, the causes of magnetic tape failure - this is kind of... this goes over some of the points I already made. By the way, most of these slides are in the handout, so that if you want to make notes, you don't have to copy the slide, you can just put little tidbits of information on the right hand column, such as "rubbish", or "I don't believe this," things which remind you how you feel when you get back home. High curl is one cause of failure. The friction or the stickiness, which I referred to, was another. Adhesion failure is a third. This is where the - actually is a result of poor manufacture, where the binder layer does not adhere to the polyester backing. And cohesion failure, which is due to the hydrolysis of the polyurethane binder, which I've already referred to and which was discussed this morning. When you have binder shedding, you have buildup of the heads.
Now, when you look at all these properties, the question was raised this morning about why can't - what do consumers like yourself do about specifying magnetic tape? Can't you just specify magnetic tape which is stable, or have a specification which tape must meet? And the standards committee tried to do just that, and ended in failure. The problem is that to evaluate magnetic tape, you would have to evaluate it for every one of these five characteristics. And you're not interested in, say, the characteristic of what is the cohesive failure today, but what will it be like in twenty years or fifty years? And there's only two ways of doing that: making the test itself extremely severe, or else incubating your tape before you test it. And so you have to arrive at a proper test method, a proper end point, and some hope that what you're doing to accelerate the stresses you're putting on the tape duplicate what's going to happen in the real world. The question was raised this morning about incubation tests and how do you know whether they represent what's really going to happen over a long period of time? So these are very, very difficult questions. Within the standards committee for about five or six years, there was a group working on this particular problem, and agreement was reached on some of these characteristics and how to test them, but not on others. And we had representation from Ampex and from IBM and from 3M. None of the Japanese companies agreed to participate in our activities despite repeated invitations on my part to do so, and after a number of years, the manufacturing companies just dropped out of this particular activity. And, so activity is stopped, because to pursue a specification, you really have to have laboratory facilities and you have to have scientists working on it. And the only scientific individual who still had facilities was the National Media Laband John Van Bogart, and he was just one person. And the manufacturers are no longer participating, and work had to be stopped. Now, you can say, well, those dirty manufacturers, you know, have been slammed a little this morning, but we have to understand their position as well. Magnetic tape is a commodity item. It's made as cheaply as possible, and competition is fierce. And companies only stay in business if their bottom line is in the black, and they are not going to make money by investigating the characteristic of their product which is the weak link. It just doesn't make any economic sense, and, let's face it, we all go by the economics. The hope was raised this morning that maybe consumers like yourselves can get together and beat on the doors of the manufacturers. They'll listen to you if you can say that if you come up with a high quality, permanent magnetic tape, we'll sell so many million dollars more of magnetic tape, then they'll have a competitive advantage over the other manufacturers. I don't know if that's the case, but that's the only way that they will listen. And you can't blame them, otherwise they're going to go under.
So right now, there is no specification for magnetic tape, and in answer to the question this morning, I agree entirely with what Al Lewis said: you have to buy your tape from a reputable manufacturer and avoid white boxes. I also agree that a lot of you who are curators and custodians have no control over what magnetic tape you're actually storing, so it's a serious problem. And there is no answer.
Now, I've talked about physical properties. Let me briefly talk about the magnetic characteristics, and the most stable - I'm only talking here about the magnetic characteristics, and from a magnetics point of view, how easy it is, when you put on the signal, how long is it going to stay there, how long will it last, how could it be demagnetized? The iron oxides and the cobalt, modified iron oxides are the most stable metal particle. Chromium dioxide, and metal evaporated are less stable. The statement was also made this morning - I think Jim Wheeler made it - that he'd stay away from metal particulate tapes. I think - oh, it was metal evaporated, you're right - and I think the reason there is because, physically, they're much more fragile. But here, I'm talking only about the magnetic characteristics.
Now, you have to worry somewhat about exposure to magnetic fields, but don't worry too much, because this is not the prime cause of tape failure. If you leave your tape close to motors or elevator installations, Jim Lindner told me a story about someone who had an excellent place to store their tape - it was on the roof of their building. There was a wall, and they built an enclosure beside the wall, and put all the tape up on the wall, and it was great, it was away from everything. It was away from everything except for the motor which ran the elevators on the other side of that wall, and you ran into a problem of that nature. You have to worry about headphones and speakers, microphones, magnetic cabinet latches have been known to cause problems, magnetized tools, TV monitors, but these are not... this is not really the main source of concern. The main source of concern is not the magnetic signal. The main source of concern are physical damages brought about by mishandling or by degradation of the binder.
Now, let's talk about storage. This is what I'm told I should talk about, and I've been talking for almost half and hour and haven't touched the subject yet. Let's talk about storage temperature and relative humidity and again, that was addressed this morning. And in the storage document, we define two types of storage: those which you want for the medium term, which is at least ten years, and those that you want for as long as you possibly can. In John Van Bogart's paper, he referred to this as access storage and archival storage. And there are different requirements for each. For medium-term storage, around 73 degrees and up to 50% RH, room comfort conditions are fine, and your tape should last ten years. But if you're talking about having tape last as long as it possibly can, those conditions are not good enough. If you go to a 68 degrees Fahrenheit, then you should have a maximum RH of 30%. It was stated over and over again this morning how high humidity is poor. And it's true: high humidity is poor, but you can go up in humidity if you go down in temperature. Tape life will be extended both at lower temperatures and lower humidities, so you can compensate for a higher humidity by going down to a lower temperature. This is shown in this particular graph here, which appears in the ANSI standard. And here we have what are called isoperm lines. These are lines in which the tape is expected to last, say, X years, and any temperature and humidity point along that isoperm line gives equal stability, and you can see that as you go down lower in temperature, you go higher in humidity. This is data that was obtained in the laboratory and is typical of many materials, including cellulose triacetate film base, stability of color dyes and photographic film, so it's established on very good scientific principles. And that is the reason why you have several options in the storage document as to what temperature and humidity you should use. Now, what are the concerns with high humidity? And again, I'm repeating what has been said this morning, which means that the speakers this morning were right on target, what is increased hydrolysis binder breakdown and then the nice green mold, and I won't discuss that any further. It was discussed this morning. What about low humidity concerns? And low humidity concerns are static, and static will also attract dirt, which of course we know is very detrimental. What about low temperature? As I already indicated, there are reports of lubricant separation if the temperature gets too low. Also, when you go down in temperature, if you start with a tight roll at room temperature, as you go down in temperature, the roll becomes loose. That's because when materials are cooled, they contract. This is true of all materials, but with the backing, the polyester backing used for magnetic tape, it doesn't contract the same in the circumference direction as it does in the radial direction, in other words, in the thickness of the polyester backing, it will contract about twenty times greater than in the linear direction in the circumference of the roll. So that means that when you go down in temperature, there's going to be a greater contraction in the radial direction, and you end up with a loose roll. If you don't handle your rolls at the low temperature, that doesn't create a problem, but if you take it out at low temperature and you're not careful how you handle it, you'll end up with a step roll when you get up to operating temperature. And the standard, as I indicated, specifies that the lowest temperature for storage is 46 degrees Fahrenheit. I personally am not completely sold on whether this is a real valid criteria or not. It's all based on anecdotal evidence, but the committee felt that it's better to be safe than sorry. I would like it if we could go even lower; I know with photographic film, you go down to minus 3 degrees Fahrenheit, and it's fine. And the lower you go in temperature, the greater the stability of the material. Explorers that have died in the attic - in the Arctic, at least [laughter] - I felt that I needed some water. I was hoping it would be gin. Explorers that have died in the Arctic are usually buried in the Arctic, and hundreds of years later they have dug up their remains and you can recognize who they are. So, in other words, those storage temperatures do retard degradation. This is very useful information for you to have. One of the problems, of course, in low temperature storage, is that you require warm-up. If you take a cold roll and bring it up to room temperature, you're going to have moisture condensation on the roll. You don't want it. It's not catastrophic; if it sits there for a short period of time, but it may get inside between the layers and eventually cause problems with sticking. And the way you do that is you go into the cold storage area, take your cold magnetic tape roll, stick it in a plastic container or a cardboard box, bring it out, and let it warm up, and then you take it out.
Also, humidity equilibrium may be required. Humidity equilibrium is much slower than temperature equilibrium, and I would not recommend it as a general way of doing things because the times required are very, very long, but in helical scan recording, you can run into dimensional problems if your humidities are off. And so what has been recommended is if you have helical scan recording, and you take the material from a low humidity, and it isn't operating the way it should, then you allow humidity conditioning. If it's operating fine, don't worry about it.
These are approximately times required for equilibration. For temperatures, it's very dependent upon the width of the tape, and if you get into 2 inch tape, you may require four hours, but much less for thinner tape. But as you can see, for humidity equilibration, you can talk up to very, very long times. And this is for almost complete humidity equilibrium. I'm not suggesting that you take 2 inch tape out and let it sit around for two months before you use it. But if you want to be a purist and get complete moisture equilibration, this might be required. I don't think it is.
In addition to storage temperature and storage humidity, you have to be concerned with what you're storing your material with. And your material should be as chemically stable as your tape. Obviously not debris producing, obviously not magnetic, should be free from distortion. You have to worry about fires, and the recommendation is that it will not spontaneously ignite below 150 degrees Centigrade, and shouldn't be more flammable than the tape itself. Let's go through some of the specific enclosures that you use - for hobs, obviously they should be cylindrical, and the larger the diameter, the less the distortion on the tape. For flanges, the aperture should be small. You should avoid padded flanges. Both at the University of Syracuse and at Vidipax, I was shown examples of tape in which there was neoprene or foam on the flanges, which degraded itself. And the degradation products leaked on to the tape. So, when you're storing material with tape, your tape is no better than the storage materials. If they go before the tape, they're going to cause problems. So the theory of having padded flanges is great, the materials used was not so great, so avoid them. And flanges should be removable and replaceable in case they become bent. We're talking about the cassette or cartridge shells. You would want impact resistant material, and one that can be disassembled. And the containers again - have impact resistant, moisture and dust resistant. The point was made this morning that the flanges should not be load-bearing; the tapes should be stored vertically, and that is correct. The two of us agree, I'm sure that more than two of us agree in this room, that the container shouldn't be deformable. Cardboard and paper, of course, create dirt, and you would want your container to be capable of being latched. And labels - people frequently cover their magnetic tapes with all kinds of labels, and they should be non-acidic themselves, not producing any agacious by-products or debris. With some labels, you can actually have migration of the adhesive beyond the confines of the label itself, with the possibility of getting it on the tape, and you should try to keep the number of labels to a minimum. The more material you store with your tape, the more you're asking for problems. And there are enough problems with the material itself without asking for more.
With storage housing, metal shelving is recommended. Again, store tapes vertically - since I said it twice, it must be right. You would want adequate airflow through the material in case there are any degradation products. You obviously want to keep it away from heat sources, sprinklers, and water pipes, and separate the material from any other degrading materials. You don't want to have magnetic tape beside a photographic film which has a vinegar odor smell, because when you smell it, it means the degradation products are in the air, and they can initiate degradation of adjacent materials.
The storage rooms obviously should be clean. You don't want nice fabric drapes, you don't want unsealed insulation, certainly not asbestos. You would like a positive air pressure in the room so that the air from the room goes to the outside, rather than the reverse. You don't want condensation, you don't want cold walls, in which you have moisture condensation on the walls. You would want floor drains, and you want to avoid basement levels, primarily for fear of flooding. There's an old saying that in any fight between man and water, water always wins. It's one of the things I have engraved on my pillow. And another thing, you should have your storage area only used for storage. It should not be a working area, because if it's a working area, it's going to create dust and dirt. You people are dirty, I hate to tell you this. And you should keep them in the dark. And, of course, agacious impurities can be a problem, so don't have your storage rooms, if possible, by your garage where you have all the exhaust fumes of your garage, your heavy industrial area if you have any choice at all.
What about tape preparation? It was stated this morning that your tape should be smooth, even tension, you shouldn't have step roll, you shouldn't have laps of tape which protrude beyond the edges of the roll. If it's a cassette, all the tape should be on one hob, and not on both; leafing stepbacks, spoking, spoking is like a tape which is tightly wound, and it can show radial spokes, like spokes of a wagon wheel. And you may have to have splices in the tape - sometimes you have no choice, but don't use 3M Scotch tape - use a tape which is recommended for the purpose.
There's been quite a controversy about should you take tape and should you exercise it every once in a while. Someone this morning almost gave real human characteristics to magnetic tape, it's a material with a life of its own. I don't really subscribe to that too much. I tend to look upon it more as coconut cream pie. That it is an inanimate object, and this question of exercising your tape - your tape has to be exercised, tape has to breathe. You hear all these human needs given to a roll of tape which I don't quite understand, but it sounds good. With respect to exercise, and this means every three to five years you take your tape out and rewind it. A lot of people say this is great, because if your tape is loose, it insures that you have a tight pack. And exercise is good for you. You see that Jane Fonda says that all the time. But there are also arguments against it. Jim Lindner says that facilities, archives that do rewind tape will usually not use their best machines. Their best machines are actually used for recording and playback. They use their old machines, and they probably end up with a worse wind than they had in the beginning, so this is a very practical consideration. It's also very labor intensive, and my concern is, that each time you wind tape - if it's a loose roll and you wind it, then you're putting more tension on, and every time you have tension on plastic, plastic flows. So you have a tight roll, and after three years or five years, the tape flows, the roll becomes a little looser, and then you rewind it, you're making it tight again, and you're getting more flow. And it makes more sense to me that it should be rewound just before you use it. But that's not necessarily a majority opinion.
I'll end up with a statement that's been made many times, that on preservation concerns for photography, you have three concerns: the material, photographic processing and storage conditions. And for magnetics, you have the material itself, the storage conditions, the hardware availability, and in the case of digital material, the software availability. So your concerns about magnetics are, one, is the stability of the tape, and two, is the hardware, the wear and the obsolescence, and the software retention for digital, and the solution is periodic refreshing or transcribing, which has been discussed at length this morning, and I don't think I need to discuss it any further. That completes my formal presentation, and if you have any questions, I'm sure members of my working group will be glad to answer them.
Q: [Unintelligible]
A: That would be - do it faster, but it's more labor intensive, so it depends on whether you want to take the Wheeler approach, which is more labor intensive, or the Adelstein approach, which is lazier and longer. Any other questions?
Q: [Unintelligible]
A: What is a relative humidity variation which is allowable? I forget what the storage document says; it's different for, I think, medium-term storage. I'm going by memory here, and my memory is getting poorer and poorer. I think for medium-term storage it's plus or minus 10, and for extended term, it's plus or minus 5. Is that correct, Jim? You don't really have to worry too much about rapid changes in humidity, because your tape isn't going to see it. By rapid, I mean a day or two. It's really the long term, which... Yes, sir?
Q: [Unintelligible]
A: Is grounding on metal shelving wise or not wise? I don't think it can do any harm, but I don't know if it's necessary. Maybe somebody else has more experience. I have never really heard any problem with metal shelving. I once heard a problem of a magnetic kind of a bookend on metal shelving. I don't think it's necessary myself. Yes?
Q: [Unintelligible]
A: The question was about rewinding. And what is the Adelstein position, and number two, will rewinding prevent or minimize print-through? On the first question, I would not recommend rewinding, but I'm one individual, and there may be others who are much more experienced than I that would recommend it. There are a number of people who don't. If you want to know what my opinion is, I would say don't. With respect to rewinding to prevent print-through, you're absolutely right. Rewinding will minimize print-through. Print-through is the problem with analog audio tape where a magnetic signal from one lap goes on to a mix. We recommend that you store your audiotape at tails out. So that means, even before you use it, you have to rewind it. Jim, you had a comment on that?
JIM LINDNER: Yes, tails out is good for audio master tapes. What happens is, the print-through actually starts occurring immediately, on an exponential curve. Within a few days, there's variable print-through. Rewinding is not for print-through, because it'll just start another print-through after you rewind. What happens is, if you want to eliminate the print-through, shuttle the tape through the machine two or three times. Each pass scratches the tape somewhat and drops the print-through level.