Homemade Tinfoil Phonographs

Several years ago, we placed a set of instructions online for building a tinfoil phonograph, copied from the Scientific American Supplement of July 20, 1878.  We're delighted to learn that one of our site visitors, Charles Smith, has successfully used these instructions to build six tinfoil phonographs of his own, one of which is pictured above.  Here are some tips he's written up so that others looking to build machines based on the Scientific American plans can benefit from his experiences:

   It took more than 40+ hours for this one [i.e., the phonograph pictured above].  I even experimented with different diaphragms, and found five and one-half thousandths thick aluminium printing plate material works the best. Instead of sandwiching the diaphragm between (over) the diaphragm support and cover plate, I inserted the diaphragm into the cavity, and gasketed it both sides with the cover plate keeping it under more tension than if it was just held in place between the two.
   I couldn't find a porcelain mouthpiece anywhere, so I had the mouthpiece made from wood. I found a old hardware store that had the hook, new ones just didn't have the fit or character. The screw eye was hand made, another part unavailable anywhere.
   Making the plaster-of-Paris cylinder was an event in itself!! I made the mix way too thick the first time, and it had voids. So, out came my late Dad's brick hammer and I had to smash it off, what a time. I then made a new mould, and made the plaster very "soupy" and it worked great. No voids. Then there is another trick to fine tune the cylinder to the frame that has to be done before painting. There is so much to it beyond what the plans show that comes from experience making them.  I'd gladly help anyone through it if they need the help.  Anytime!

*    *    *    *    *

   There is a certain way to finish off the cylinder so it is correct and true.  First, the grooves need to be cut so the groove is truncated, and not a sharp point, so the foil sits flat.  If the groove is too pointed, the foil won't work as well because it doesn't have the support underneath. You need to go over the face with a very small very fine piece of sandpaper glued to a smooth strip of wood.  It goes between the diaphragm and cylinder, and you crank it through several or more times.  I can explain more fully later, but it's one of many "little things" I have incorporated into making these.

*    *    *    *    *

   For the plaster mandrel model: Quarter-sawed oak looks good for this machine, and is an appropriate thing to use.  If you can get a 1 13/16" thick piece, six feet long,and12" wide, and plane down what's needed, and save some of the 1 13/16" material for the crank knob, the machine will have been made from one piece of wood, and better looking!
   1-Blow up the side profile [with crank] on photocopier so the cylinder diameter is 4 1/2" to use as a template.
   2-Cut a 1" thick board 15" long and 4 13/16" wide, then cut that board exactly in half, yielding two 7 1/2" pieces.  Make sure all ends are perfectly square!! These are for the two upright standards.  Cutting it this way assures the two standards have the same grain pattern than if cut from separate stock.  Also, the front and back edges don't have to be scrollsawed, just the curves and area where the diaphragm support goes.  When tracing them out, be sure to keep the two pieces oriented so the grain is the same.  After tracing, put a pin point hole through the center shaft area of the print, and mark the wood there for a drilling reference.  After that, cut out the crank handle from the template and use that for marking out a crank.
  3-Before scrollsawing, drill the pilot holes in the bottom of each standard.  Secure them so the drill goes in straight and can't "walk".
  4-Drill the 3/4" shaft hole where marked.  Since you made the two pieces identically in height, you can make a "L" bracket as a guide and support from wood clamped to the drill press.  Drill one to a time, and place a scrap piece of wood underneath so the drill bit won't chip going through.  When you do the second one, make sure the area is clear of wood dust so the piece aligns as the first one did, exactly in the "L" guide.
  5-Before scrollsawing, use double sided scotch tape and attach the pieces together, so you'll be cutting them one time as identical.  Make sure the pieces are attached perfectly identical. This way there will be a lot less sanding to make them the same.  Cut out the two slots for the shaft as the final step.  Make sure the side [horizontal] cut is exactly in the center of the shaft area so the shaft is easy to install.

*    *    *    *    *

  For a lathe turned cylinder: Use .065" thick walled 3/4" O.D. brass tubing, available from McMaster-Carr of N.J.  Make it an even 16" long, and the threaded section 5 1/4" long.  It will work a lot smoother if the shaft is threaded as a square cut groove.
  1-Make the paper mould 5" in diameter, and 4 1/2" wide.
  2-Align paper mould so exactly 5" of the threaded end is out of the mould area.
  3-Make the plaster "soupy" and not stiff so it will pour without voids.
  4-When hardened up, remove mould and bake for two hours very gentle heat at 120 to 140 degrees.  Do not coat it with paraffin!
  5-True up the threaded side of the plaster cylinder until 5 3/16"of threaded shaft is exposed, then true up the opposite side until the width is exactly 4".
  6-Turn down the diameter to 4 1/2".Patch any air bubbles with plaster, sand VERY lightly with the circumference, and let dry.
  7-Cut a "V" groove into the surface, until it just forms a fully pointed edge groove.  Chamfer the two sides if desired.
  8-Place the assembly in the frame, and tighten the hold down caps securely.
  9-Fashion a 1/8" thick x 6" long x 1/4" wide board that has one of the 1/4" faces sanded smooth and square [no ripples or warps] for gluing a medium fine strip of sandpaper to.
  10-With the diaphragm holder off the machine, and the standards square, secure, and tight on the baseplate, place the sandpaper block [sandpaper against the plaster] between the cylinder and diaphragm support.
  11- Crank several or more times from left to right so every time a VERY LIGHT amount of plaster dust is removed.  Each time you will be truing the cylinder to the frame, a very important process.  Keep the diaphragm support tight to the block, and crank slowly.  Hold it so the block can be grasped near the bottom of the support.  When the sharp "V" groove is sanded into a truncated groove the process is complete.  You want the groove into the surface only.  The grooves will no longer have sharp points.  The foil will rest flat on the surface this way.
  12-The cylinder can be painted with oil based paint.  Two coats on the sides, and only one good coat on the face or the grooves will be filled in.  If the paint is thinned somewhat it will be easier to use.

*    *    *    *    *

For the diaphragm assembly:
   Make sure the recess in the diaphragm support is bored out true on a drill press, and not at an angle as a hand drill will do, so the diaphragm will be at the same parallel plane to the cylinder.  Drilling it about 1/8" should do fine.  The deeper it's drilled will take more gasket thickness.
  Put the diaphragm into the recess rather than over it, gasketed both sides, so when screwed down it will compress tight.  It can be made tighter this way than if the diaphragm is larger and placed over the recess.
   I use PVC shower liner for gaskets as the outside gaskets, with thin paper as the ones that mate to the actual diaphragm.  Make the thickness and amount of gaskets so when the diaphragm holding plate is resting on the support [flat and upright on a table] the diaphragm and gaskets are slightly sticking up before screwed down.  Five and one half thousandths thick aluminium printing plate material is the best I've used for the diaphragm.
   When making the stylus bar, drill the needle hole slightly larger than the needle you will be using, and use Elmer's glue, and not crazy glue, so it can be easily removed later if needed.  Make sure the bar is vertical and straight to the diaphragm, as the print shows, when installed.  Glue the gasket material directly to the diaphragm side of the bar, and use a very light coat of spray varnish, including the gasketed area so it will be smooth.  Too much or too thick of a coat will reduce the spring action of the wood.  When they say "A delicate wood spring" they really mean it.  If it's 3/16" wide across is good.  Try to limit the width to that.  Too wide and the sensitivity is lost, and it takes a lot more volume to record.  Make the "mallet" about 3/32" tall.  Use white pine, the kind that your fingernail will easily indent, and not yellow pine which isn't as flexible.

*    *    *    *    *

  Before the diaphragm support hinge is installed on the left side standard, make sure to find a good quality hinge that has a good fit, and doesn't have a lot of play. Try to find an old American made one, if possible!
  Place a thin shim stock [thick paper, or something similar] under both sides of the diaphragm support when it rests on the standards, and the same material behind both sides of the diaphragm support.  Then mark out the hinge area for the screws.  This way, there is a little clearance on all sides where the three pieces meet and it won't have a bind and the diaphragm support won't rub on the right side standard when closed.
  I have found that using brass flat headed screws makes a genuine authentic machine.  The brass screws match the shaft, and if flat headed screws are used, shows the hardware of 1878.  Phillips screws weren't invented then, and don't reflect the period when it was made.

*    *    *    *    *

   I think it really is neat that these still can be made.  The kids at school really marvel at it.  And when I first bring it in, the teachers give me a skeptical look like it can't work.  When it does, it reminds me of the looks of the "S-A" staff when they first heard it.
   Another very important part of this to me is the builder is making something authentic from actual 1878 plans.  There's really not many things that can be said like that nowadays!

The above images show the diaphragm and stylus bar and should help prospective phonograph builders with this important part of the apparatus.

Smith adds that he chose the deep red color for his mandrel based on a French Vital phonograph in René Rondeau's collection, and that he obtains good results using real tinfoil or heavy duty aluminum foil coated with Pledge before recording.  He's eager to correspond and swap ideas with others interested in building their own tinfoil phonographs; if that's you, drop him an email here.  (And we'd like to hear about your results as well!)

Smith has also shared the following images with us of a steel-mandrel model built by the Canada Science and Technology Museum (click on any image to enlarge):


Find the original instructions from 1878 here.

Original content copyright © 2009, Patrick Feaster.