These are quick iPhone pictures of the almost finished product. I used printer paper (cream colored, for aged looks of the innards, just because I can and because I want it to be pretty inside as well as outside!), 0.05 mm enameled copper wire for the electric windings, a small rare-earth tube magnet (2 mm OD x 2 mm long x 1mm ID, from Supermagnetman's online store), a short length of 0.8 mm brass rod, quite a bit of thin viscosity CA glue, and another rare-earth magnet (tiny 1x0.5 mm disk) for "calibration" of the dial. Oh, and a repurposed wristwatch needle from a bag of old watch parts bought on eBay.
There's quite some prep work already done at this stage, I made a winding jig with a small gearmotor (also from ebay) and a Nylon bolt/nut/washers thingie. I first wrapped a strip of paper over a round bar of the appropriate inner diameter of the coil (here 6 mm) and use CA glue to make a rigid impregnated tube of the adequate length (4 mm here). Once the CA glue has cured, this gives a nice vintage-looking coil core. I then drill it across its diameter for a paper tube that will serve as bushings for the needle axle (the brass rod). Add two cardboard washers made in the same fashion (several layers of the same paper stacked and glued by impregnation with CA) to make the sides. I then wrap the needed length of copper wire (7,5 m here, I wanted to have a coil resistance of 75Ω give or take), solder a thicker wire, glue it to the coil body, check for electric continuity, and impregnate the coil with CA to protect it, trim the sides if needed.
Once everything is cured, I made the face with two layers of thick cardboard, glued my face template (made with Illustrator), drill the hole for the front bushing/axis, glue to the coil, check the brass rod fitting. The brass rod must be filed down a bit and tapered so that the magnet slides over it to a point: I want a friction fit here to avoid glueing, because at this scale I can assure you that capillarity does happen even if you don't want it to! I had to carefully force the rod to rotate with pliers and then force-turn it for a few seconds chucked in my Dremel because of CA glue seepin into the bushings... after spending 3 hours up to that stage, you definitely don't want a stuck coil! Anyway I managed to save it, but I won't be trying again to glue a tube magnet in such a tight space anytime soon!
Then I added the needle, centered the magnet to have a vertical needle at zero coil current, and started working out the amplitude of the scale by using a function generator on my computer and connecting the dial external wires to the phones output of the computer. And yay, it's moving!
After tinkering with weights glued to the bottom side of the magnet to get a decent zero (vertical needle) at zero current, I noticed that a tube magnet isn't the best option after all: I had to add a very small magnet to one side to have the needle returning to vertical when the coil is disconnected or the current is null. It's the tiny golden thing in the first picture. Also I had to add two small cardboard limiters to the sides of the needle to prevent it from shooting outside the zone at which the magnet can rotate in the magnetic field from the coil: this wouldn't happen with a flat disc magnet. Oh well, I had to try anyway. Then I made a big face ring with the same paper+CA glue carboard and glued it in place: this will act as a spacer between the face and the glass dome that will cover the gauge.
The following pictures show progress done on the dial indicator after all the steps described above but not pictured (building the coil, building the face, adding the magnet and needle, calibrating).
I designed this gauge for a 25mm glass dome (ebay rulez, I bought a bunch of different sizes for next to nothing). Pictures 4 to 6 show the dial with its outer cardboard protecting sleeve. Next step is to solder a brass tube of the right diameter and a retainer ring for the front lens, add a back with connectors and call it the day. That will wait for a few days since I don't have the tools for flaring brass here (holidays coming soon, I have all that stuff at my parents' house).
Oh, by the way, when designing gauges for domed face glass, take into account that these act as optical lenses that enlarge the contents, and remember that the outside portion of the innards won't be visible: I had to reduce my dial face design two millimeters from the outside diameter so everything would show up fine.
At the end of this preliminary article you can find a video of the dial being tested from my computer, using a function generator and the sound output: sine wave, 2 to 25 Hz, I tinkered with the amplitude a bit too. This dial can take up to 25-30Hz without losing track, which is far beyond what I will be using for it (it will oscillate in the 2-5 Hz range in its final application).
Well there you have it, a fully working steampunkish homemade dial indicator gauge contraption. Hope you like it.
Cheers from Paris, France.