Braille Column Rods PCB

Braille-Display-CAD

Experimental design for a braille display using build123d CAD in Python

Interesting Parts

• 4mm x 8mm DC Motors: https://www.aliexpress.com/item/1005004370822475.html
• 2.8mm x 3.9mm DC Motors: https://www.aliexpress.com/item/1005006897190118.html

Explored Ideas

Pogo Spools

• Laser printer fuser-like roller mask.
  • Branch: archive/pogo-spools
  • Files: ideas_pogo_spool_mask/
• Made PCB with small stepper motors, tiny CAD parts (spools).

RF Inductors as Electromagnets

• Use Mouser's "RF Inductors" selection, mounting an inductor vertically.

  • Branches: archive/inductor-smd-vertical, archive/inductor-smd-flat
  • Eaton makes good-looking ones. 3225 metric (1210 imp) is too big; 2520 metric (1008 imp) might be okay. 2012 metric (0805 imp) is great!
  • "Power Inductors - Leaded" are never small enough.
  • Filter to >= 20uH inductance. Filter to >=70 mA.
  • Examples (Vertical-Mount):
    • BWLD00241715 -> Not stocked in smaller quantities, though.
      • 2.4 mm x 1.72 mm x 1.52 mm
    • LBR2012T470M -> 0805 imp (2012 metric). 47uH. 75mA rated max.
  • Examples (Horizontal-Mount):
    • VLS201612CX-220M-1 -> 22uF, very good, but shielded.

• Test results:

  • 4mm x 4mm x 3mm 33uH coil-wound SMD power inductor (ASPI-4030S-330M-T) can point a magnet. Magnets stick to the inductor always though.
  • Several inductors, up to 6mm diameter (and 1-2 MH), with 24V get very hot, but don't levitate a 1mm or 2mm neodymium magnet.
    • Good at pointing pointing the magnet, if it starts far away.
    • Cannot remove the magnet from the inductor once it's stuck.

• Conclusion: RF inductors don't have enough magnetic force to levitate 1mm or 2mm neodymium magnets.

Nut/Bolt Grid

• Turn a bolt in a threaded hole/nut. Each bolt is a dot. User directly touches the dot.
• Movement can be done with one motor per dot (very tiny motors, maybe stacked), or motors can be moved between dots (CNC plotter style). Multiple heads can be used.
• Branch: archive/cnc-plotter
• Files: cad/nut_bolt_grid.py, cad/vertical_motor_layers.py

Silicone Sheet

• Use CNC plotter to manipulate a silicone sheet which can pop up and down.
  • Branch: archive/silicone-sheet
  • Files: cad/silicone_sheet_mold/

Unexplored/Unrealized Ideas

• Investigate bi-metallic strip with actuator.

• Investigate shape memory alloy (SMA) approach. Benefit: Would be very quiet.

• Should buy some 4mm solenoids.

• Key and lock mechanism pushes up on the pins

• Use GT2 belt (or similar) to press up on pins.

  • T2.5 belt is probably best - https://www.aliexpress.com/item/1005004668423141.html
  • Could try to TPU 3D print belt with the perfect braille pattern, in a closed loop. Teeth on the inside, and pattern on the outside.
  • Spring pushing on pogo pin flange, on solenoid-relieved cell frame.

• Use CNC plotter to turn pins/bolts for placement. Use hex drive, plus solenoid to mate/un-mate. Build the plotter on a PCB with 3mm linear rails, belts, etc.

  • Branch: cnc-plotter
  • Bolt options - Search "M2.5 plunger"
  • Best option: M1.6 Grub Screw, probably.
  • Create top plate with Z-stacked nuts on it.
  • Screw driver or nut will move up and down and turn screws.
  • Search "Micro Linear Motor" for gantry.
  • Focus on M1.6 and M2.5 designs.
  • To move the screw:
    • Use a spring to push the bit up, and a solenoid to pull it down.
    • Use a hex bit maybe, or use rubber friction.
  • BOM:
    • LM3UU linear bearings (ID=3mm, OD=7mm, L=10mm)
    • 3mm Rod x 100mm
    • PM15S Stepper Motors: https://www.aliexpress.com/item/1005005152436475.html
    • GT2 belt
    • GT2 pulley, custom to mate with stepper motor's gear.

• Motors mounted on orthogonal PCBs, with rotating slanted plate to push on the pins.

  • Use magnets as the mating, somehow.

• Microfluidics as valves.

• MEMS on-chip stepper motor:

  • https://www.youtube.com/watch?v=n3YMjgbhvTA
  • Could be useful to drive the screws on the CNC Plotter style.

• Double screw with balls and weird geometry.

• Cams on 2.4mm diameter rods that run along each column of dots.

  • Each cam would have 8 positions, so 45 degrees per position.
  • Would probably need to move table up-and-down independently of the cams.
  • One motor per cam, probably.