I work as an Audio Preservation Engineer at Manchester Central Library, where we are providing the digitisation element of the Royal Northern College of Music’s National Lottery Heritage Fund-supported project Throwing Open The Concert Doors.
Throwing Open the Concert Doors is celebrating the RNCM’s 50th anniversary by preserving the college’s archive collections and engaging local people with its recordings and the stories behind them. This means I get to listen to unique local recordings of classical music, jazz and opera every single working day. It’s a tough gig but somebody’s got to do it!
It was plain sailing for a good long while – the first 900 reel-to-reel tapes transferred very easily. All I needed to do was clean the machine, check the condition of the tapes by running them through the machine, and then set the azimuth (the angle of the tape head) to give nice clear high-frequency playback.
The analogue sound goes through an analogue-to-digital converter and pops out as 96kHz 24 bit .wav files, which are checked, saved and backed up. The amount of data in each file is huge – a stereo recording of an hour’s concert can be about 1GB in size. The quality is about double that of CD audio.
The concerts and masterclasses between 1973 and 1985 were extremely well recorded and have been so well looked-after that there were very few issues. Until, that is, we got to 1986.
Suddenly, the tapes consistently started leaving what I can only describe as a brownish, sticky goo on the tape heads and rollers. The tape had always lived together in the same storage conditions, and there was nothing visually different between the tapes from 1985 and 1986. Or was there?
It was only when I looked closer that I noticed the text ‘AGFA-Gavaert PEM-369’ printed on the green ‘leader’ tape at the start of the sticky 1986 reels. A bit of online searching confirmed my worst fears: this type of tape suffers from the dreaded STICKY SHED SYNDROME!
Confusingly, this has nothing to do with garden sheds (although storing tapes in sheds is definitely not ideal!) The ‘sticky shedding’ refers to the stuff that sheds away from the tape as it passes across the tape machine rollers and magnetic heads.
This *stuff* (technical term) mucks up the tape head which reads the signal, stopping high frequency audio signals being picked up. The result is a muddy-sounding, muffled playback which can sound like it the microphone was sitting in a cardboard box. Not ideal. Typically the start of a tape sound nice and clear but the signal deteriorates gradually through the recording as the playback head is covered in gunk.
Seeing applause
Here are two samples from a composer’s workshop of soprano Ida-Marie Turri singing and some applause. The first is before baking; the second is after 100 hours in the oven. The applause is particularly useful as it is a high frequency sound which is badly affected by sticky residue.
Still, I thought, how many of these AGFA PEM-369 tapes can the college have used? Surely the manufacturer must have done a prompt recall. Well, yes. The manufacturer did offer to transfer recordings onto alternative tape types at their West German factory in 1987. But this message didn’t get through to the college. And the AGFA tapes continued to be used for most RNCM recordings until 1992.
On the plus side, audio technicians at the college did pick up on the problem and transferred many reels to DAT tape (at 48kHz) and some onto CDs (at 44.1kHz) in the mid-1990s. Although these transfers were performed extremely well by professionals, using good kit, they are at a lower sample rate than our analogue transfers of today.
It’s therefore a shame that these sticky reels were thrown away in the 90s – but then again, they would be more sticky now anyway. We’d never do this now but it was a very common practice back then. At least we as a profession have since learned not to throw away the original carriers – not matter how messed up they are!
They might one day be useful to correct my mistakes, or they might even be able to be transferred at a higher quality by a high-tech machine that hasn’t even been invented yet. Who knows. But we still have a mountain of sticky tapes that need urgent transfer before they deteriorate further – 800 to be precise!
Baking Tapes
Treatment of sticky shed syndrome involves drying tapes in a gentle low heat for several hours, removing any accumulated moisture that can make certain tape types sticky. All tape types need slightly different treatments, and even the same tape types produced at different times or stored in different conditions can respond in different ways.
It’s a real learning curve with every sticky collection even for an experienced audio engineer. For me, as a relatively inexperienced engineer with only four years of practice with mainly good condition tapes, it has meant relying on kind colleagues in the sound archives at the National Library of Scotland, the British Library and Leicester University for advice and tips. The go-to online resource on sticky tapes is Richard Hess’s Audio Tape Restoration website.
Thanks to the National Lottery Heritage Fund, we have been able to buy an industrial food dehydrator with project contingency – you might more often find these machines in big hotel kitchens or restaurants. I can fit over 60 7” reel-to-reel tapes and leave them to bake for 100 hours at 55°C. This means I can bake tapes at scale and make the process of transferring them a bit more efficient than I could with our little 12-hour, 8 reel-to-reel domestic dehydrator.
Some of the tapes don’t seem to respond well even to bakes of 100 hours. If they are still sticky after this point, I whack them back in and see if they are any better after 150 hours. They seem to respond best while still warm, like a pizza still hot from the oven. If that doesn’t work, or if I hear any squealing or high-frequency scraping distortion on playback, then it can require a different (and even more niche) treatment in the form of alcohol. The booze is not for the engineer (although by this stage with a tape can be tempting!) No, it’s for the tape.
You’ll be able to hear the point at which I pressed the alcohol-soaked cotton bud against this tape – it was about half-way through the sample. The sound suddenly loses the far-away feel, and the high-frequency distortion disappears, leaving a clear signal.
We apply rubbing alcohol (99.9% isopropyl alcohol) directly to the playback side of the tape before it reaches the playback head. This is done either manually by an alcohol-soaked cotton bud (which is fiddly, not ideal for the engineer who has to breathe in alcohol fumes and anyway not practical for longer tapes) or by an IV drip suspended over the tape machine, just like in a hospital. This constant application of alcohol can help remove what is often described as a ‘loss of lubricant’ or LOL (although it’s no laughing matter), and smooth the jittery scraping of the tape across the head.
Now that I’m up to speed with these techniques and have the required sticky shed kit, we are making good progress through the sticky batch of tapes. We should be able to do most of them as part of this funded project. But who know what other surprises might be waiting for us on the sound archive shelves. So, yeah. I do sit and listen to lovely music all day. But it’s not quite as simple as that!