Wire Recorder Project

This web site describes how recordings made on steel wire by my Grandfather from about 1949 were recovered to be enjoyed again by his descendants.

To play these wires I needed to get hold of a vintage wire recorder. These old recorders are very rare in Britain, and by an extraordinary co-incidence the only ones I could find for sale in the whole of the UK were from a dealer just a few kilometres from my home in Edinburgh, Scotland. He had two "Webster Chcago" (or "Webcor") recorders in unknown condition and let me have them both for a reasonable price. The two recorders are:

• Webster Chicago model 180: popular domestic model from about 1948. The one I bought had been modified for the UK market by the addition of a 240V-120V step-down transformer.
• Webster Chicago model 288: slightly updated version of the model 180, from about 1950. Slightly bigger cabinet than the 180 with 'improved audio response', and improved anti-spill mechanism. The one I bought has an ominous smell of 'wood smoke' and I haven't had time to open it up yet!

There are some excellent resources on the web describing the history of Webster Chicago, and just about everything you need to know if you own a wire recorder. I'm very grateful to the authors of these sites.

Using information gathered from these and similar sites, we now knew what we had to do in order to play back the wires. We chose to work on the model 180 as it seemed to be the cleanest (and, as it turned out, was adapted for UK mains voltage). The steps were:

1. Restore the mechanical operation of the model 180:
   • Clean and lubricate all mechanical parts, especially the 'bailer' mechanism which ensures the wire is correctly bailed onto the spools. The failure of this mechanism results in disastrous tangles!
   • Replace felt pads and adjust the mechanism that is used to 'brake' the rotating spools. See pictures below. Applying just the right pressure to the spools ensures that they are packed tightly after play or re-wind.
2. Isolate the original electronics so that it is not powered-up, and take a feed directly from the playback head. This minimises any 'hum' on the audio, and gives the highest quality audio output.
3. Take the audio output to some kind of amplifier and find a way of feeding it to an audio recording application on a PC.
4. Once the original Webster spools can be played back in this configuration, build a separate playback spindle to hold my Grandfather's 'non-standard' spools so that they can be played back onto the take-up spool of the Webster machine.

With my Uncle's help we quickly got the set playing back the original Webster spools through a guitar amplifier. My Uncle found an old mechanical counter that I anchored to a bench and this provided a perfect spindle to hold my Grandfather's spools. It was necessary to completely remove the Webster's own playback plinth because it got in the way.

I wanted to digitise the audio from the playback head at the earliest possible stage. This was done by feeding the output from the head straight into a guitar pre-amplifier unit. The guitar pre-amp immediately digitises the sound and then offers a wealth of pre-processing before the digitised sound is fed to a laptop via USB. I used the pre-amp to get the levels optimised for the laptop, and to 'equalise' the sound to get the right balance of treble and bass.

On the laptop I ran Audacity: a freely available audio recorder and editor from those wonderful people at Sourceforge. This fantastic piece of software enabled me to do the following:

• Record all the wires
• Post-process the recordings to:
  • Reverse any wires that were spooled 'backwards' (several of them were!)
  • Adjust the levels of any sections of the recordings that needed it
  • Edit out bad noises when things went wrong!
  • Speed up the recordings to account for differences in recording and playback speed
  • Convert to MP3 format for efficient storage on hard disk and CD.

The ability to change the speed of the recordings was crucial. It seemed that my Grandfather's recordings were not made at a consistent speed. It was difficult to judge what was the correct speed for any given recording, but I mainly used the pitch of musical instruments (along with a guess at what key they were playing in!!) in order to work it out. In some cases I could use the frequency of mains hum (which is a very constant 50Hz - and probably was even in the 1940s and 1950s), measured using Audacity. I needed to speed up the recordings by an average of 7%, and sometimes as much as 10%. This showed that my Grandfather's machine recorded faster than the Webster, probably to achieve a better fidelity with lower quality components.

The playback process, miraculously, went without a single problem. My only job was to keep an eye on everything and to listen as the sounds played back for the first time in decades, bringing back so many voices and sounds from my past. There were some tearful moments as I heard my Grandfather playing the piano again as I remember him doing when I was a kid. There were lots a laughs as I heard great grandparents telling stories in accents that don't exist today!

Strangely, some of the wires had deteriorated and provided a very poor quality sound, but others sounded fresh and as if they were recorded yesterday. There was no correlation with the age of the recording. The earliest recordings were among the best.

So playback turned out to be easy. The one small snag was: I needed to re-wind the wires back from the take-up spool onto their original spools. Because I was playing them off a 'passive' spindle, I could not use the Webster's own re-wind mechanism. After some experimentation the solution turned out to be an electric drill with a large rubber grommet!! This had to be held manually against the spool to turn it, dragging the wire back from the take-up spool (which, luckily, drives the bailer mechanism) at about 2 x playback speed. This was tiring, and the slightest 'wobble' meant a big 'spill' of wire which would take hours to untangle. I only had a couple of big spills that luckily affected less important passages on the wires....

Now that the contents of the wires are safely captured, the next project is to fully restore at least one of the Webster recorders. With some carefully placed bits of connector block inside the model 180, it will still be easy to isolate the old electronics and take a feed directly from the playback head for any future audio recovery projects. Feel free to contact me if you have any wire spools that I may be able to help with.