Sound recording and playback have come a long way in the last century or so, but it’s fair to say there’s still a lot of interesting stuff locked away on old recordings. Not having a way to play it back is partly to blame; finding an antique phonograph that plays old-timey cylinder recordings is pretty hard. But even then, how do you digitize the output of these fragile, scratchy old recordings?
As it happens, [Jan Derogee] is in a position to answer these questions, with an antique phonograph and a bunch of Edison-style wax cylinders with voices and music from a bygone era locked away on them. It would be easy enough to just use the “reproducer” he previously built and set up a microphone to record the sound directly from the phonograph’s trumpet, but [Jan] decided to engineer a better solution. By adding the piezo element from an electronic greeting card to his reproducer, potted with liberal quantities of epoxy and padded with cotton, the piezo pickup was attached to the phonograph arm in place of the original stylus and trumpet. The signal from the piezo element was strong enough to require a shunt resistor, allowing it to be plugged directly into the audio input jack on a computer. From there it’s just an Audacity exercise, plus dealing with the occasional skipped groove.
We appreciate [Jan]’s effort to preserve these recordings, as well as the chance to hear some voices from the past. We’re actually surprised the recording sound as good as they do after all this time — they must have been well cared for.
This is awesome ! I have in my DJ library a collection of wax cylinder recordings. They were obviously made with the microphone in the horn method. There is so much history that can be saved by converting them. If you have ever seen the movie “Songcatcher” (2000 indie) you will understand that there are one off copies that should be preserved.. Would love to add more “wax tracks” to my collection.
Better off using a laser to read them without contact if the recordings are important rather than bodging together things which are just going to wear out and destroy them.
You could also probably make a mechanism to rotate the cylinder under a cellphone camera looking through a magnifier. You could then make a video of the surface, and with some appropriate image processing, you could digitize the audio.
That would be even better because you could reject dust.
I wonder if you could somehow use confocal microscopy in some limited way to 3D image a d actually see through the dust
AFAIK, that would work with the old Berliner disks because they recorded the data laterally; but the Edison cylinders stored the data vertically, in the depth, so you would need some kind of extremely precise 3D scanner.
Increase the magnification enough, and you’ll get there.
yes, the Edison discs held onto the original style he used, called “the vertical cut” whereas the resonator was aligned perpendicular to the disc, to catch the audio moving up and down… it was a very involved process, and Edison himself was such an incredible audiophile, he went to great pains to make those recordings the best of the acoustic era, removing any and all ambient surfaces in his studios… so any room resonance or anything in addition to the actual sound of the instrument or person was inaudible, so much that he called his recordings “recreations” and somewhat accurately.
I appreciate the attempt to make a piezo pickup but this isn’t the way to play the cylinders.
I have an Edison model B 2 minute cylinder player whose motor I rebuilt and now works like new. I don’t know what that wind up mechanism is in the video, but in the Edison players the pickup was driven along the length of the cylinder by a screw and half-nut. That kept the stylus (a sapphire ball) in the groove (no skipping) and always along a line parallel to the cylinder’s axis. The “tone arm” in the video means the pickup is tracking an arc on the cylinder, probably damaging it as it plays, and probably not sounding right, either.
Did you actually watch the video, and ignore that the original horn of that player pivoted at the same point that his piezo pickup arm now pivots on?
https://youtu.be/eoG_11cioC0?t=626
Much like the record players that were made many years later, there are methods of playing with a pivoting arm as well as with a linear motion arm.
He also makes the disclaimer, about 10 seconds later, that this is not a high end preservation tool, and if you have something of particular value, you should probably contact a professional in order to prevent damage.
Sitting at a computer and calling the designer out as a hack isn’t productive.
Show us how it’s done, possibly a better way, or possibly just a different way…
I have one also and would like to capture the audio on a computer, this could fitted to my player.
That’s true, but in my experience, it’s good to have an arsenal of slightly varying playback styli, as the precision either which they were made is sometimes a little varied. Especially true on gramophone, where I have a ton of semi precious stone styli to find that perfect unfettered part of the groove that was cut, but not touched by the playback stylus. And ideally you should record these with a stereo cartridge, which will split the mono phase, so you can process independently and then mix together at the end.
The question that springs to mind, is so;
Is it possible to 3D print a phonograph cylinder that would work as well as those ancient ones, with a modern msla printer like the sonic mini 4k?
I was under the impression that recording on wax cylinders is no more complicated than playing them (Edison’s first one, which I think used tinfoil, could both play and record) so you could just make a blank wax phonograph cylinder and play your audio file through the piezo pick up and it should, in theory, record the audio to the cylinder. (Or at least that’s what I think, I could be TOTALLY wrong.)
You are completely right there. But the difficulty lies in the material you record on, correct pressure cutting angle the proper cutting stylus, etc. It would be a while project in itself. But I,m pretty sure that this approach, using a piezo could be used. Although you might require some frequency correction / equalizer in order to achieve optimum results.
Oh I’m sure it would take a lot of trial and error but once the details would be worked out it would be nice to make a copy of a century old cylinder and reproduce it on a new cylinder so you could have your cylinder and play it too. I don’t have the skills to do it but you do, maybe if it strikes your fancy and you have the time you could give it a shot.
You mean leveraging the Lippmann effect but you won’t get significant excursion of the needle.
Did somebody mention Larry Lippman?
Have you hugged your kitty today?
Huh?
A piezo pickup doesn’t have a Pi Zero?
B^)
I tried a Pi Ezo once, but it must have been defective, since it only beeped.
B^)
But I did read (somewhere) that the Pi Enzo will be the fastest model.
B^)
Someone remarked on lasers being an alternative… If that were really true then we would be using lasers to listen to all of our vinyl and grooved recordings. The fact of the matter is lasers are just not selective enough and tend to reproduce things that a stylus would basically shrug off. And since they’ve already been played a myriad of times you might as well get that 1 last good play out of it and have it sent to a hard disk directly to edit and filter. There are ways that a stylus can bias certain parts of the groove that a laser would not be capable of and with little tricks like splitting the phase and canceling everything out of the mono says you can actually cancel app we can actually cancel out the majority of the noise that a laser pick up would simply Bring to an equal level with the audio. They just haven’t made a laser pick up that is sophisticated enough to squelch out the noise and bring the actual original program material further to the fore. I’ve been building pizza pickups for these old machines for years because a lot of them did have contact pickups with crystal elements in them originally but if you’ve done what I’ve done and opened up those pickups to see why they’re not working anymore you don’t really have a great idea because the internal components of that pick up in the ensuing 80 plus years have literally turned to a fine dust with nothing left but the small Braces and Connectors that originally made contact with the crystal pickup which has since been reduced to base elements!! These recordings weren’t great to begin with but couple that with devastatingly devastatingly deep gouges from the playing stylists that became progressively more Blunt as the years went by and you have a recipe for something that is almost unlistenable. But you can find that 1 little magic position with a lower weight and better tracking pizza pick up and a collection of variable wit’s variable width and depth stylists from various semi precious stones and even tungsten to achieve that ideal sound. From that point on you are left with something you can edit and manipulate with the various filters on your computer. The best way to truly get at that original source material though is to record the grooves instirio with one channel being either side of the mono phase and being able to pick and choose and bring the actual original signal as close to the listenable level as possible.
” I’ve been building pizza pickups for these old machines for years”
Like this one? B^)
https://barnfinds.com/toy-story-tribute-1979-toyota-pickup/
IT WAS THE SPELL CHECK! LOL
If you produced a detailed 3D scan of the groove (whether using a laser or by other means), you could then use a physical simulation to “play” the recording by calculating what happens to a virtual stylus as it moves through the groove. You’d have complete control over the stylus shape and weight and attachment parameters, so you could model all kinds of playbacks without causing any physical damage to the original media.
Sure, when that becomes a reality… and making a 3-d scan could certainly be done in anticipation of that, but at the moment, at least from the vantage of the typical audiophile, those options aren’t there yet. I’ve not seen one LASER based player that has come close to sounding better or equal to a high quality analog cartridge.
I know that several years ago I heard of a secretive group that was developing a very expensive laser based method of reading and digitizing wax cylinders. Then, using some very expensive proprietary software, they could actually reproduce some of the damaged music. Unfortunately the expense was financially way beyond anything even above average people could afford.
I have been searching a long time now for a digitized copy of “How Could I Forget Thee” by the Neapolitan Instrumental Quartet on Edison Blue Amberol 1630. Any help would be appreciated.