Eyez is a servo controlled pair of 3D printed eyes for the Halloween pumpkins.  Ideally they will BlueTooth controlled.

Eyez is a servo controlled pair of 3D printed eyes for the Halloween pumpkins.  Ideally they will BlueTooth controlled.

Eyez is a servo controlled pair of 3D printed eyes for the Halloween pumpkins.  Ideally they will BlueTooth controlled.

I like to make my own PCBs using my milling machine (MIllie).  I'm able to make two sided PCBs with unplated through holes.  The boards are designed using KiCad v6 and plotted as DXF files and DRL drill files.  The DXF files are opened with CarbideCreate which creates gcode files. The DRL files are converted to gcode using drl2gcode.py python script.  I'm using a free version of all the software.

KiCad:

• Set the PCB lower left corner at coordinates 0,0

• Trace should be .35mm or .45mm is even better

• Pads with through holes should be square

• Use Place->Drill/Place file origin to set the origin to lower left corner

• Plot - Plot format DXF, check Use drill/place file origin

• Drill Files - set units to mm

Carbide Create:

• Open the F_Cu dxf file.  It should be set in the boundary box. 

• Go to the Toolpaths tab

• Select Contour

• Change the tool: diameter .4mm, length 12.7mm, angle 20, flutes 2, feed rate 254mm/min, depth .15mm, rpm 10000

• Set cutting Depth Max to 0.006in

• Offset Direction: if the generated tool path ends up inside the pads, change it.

• Click OK.  Tool path gets generated.  Make sure path lies outside the pads.   Also check that paths around runs and pads are complete.  If not either change the tool diameter to .3mm or use kicad to change pad sizes or move runs.

• Save GCode.  Makes a .nc file, rename to .ngc for linuxcnc.

• Repeat for B_Cu, with extra steps:

  • On the design tab, select all and click Mirror Horizontal

  • adjust the position horizontally so that the center of a drill hole is at the origin.  This will be used later to position the pcb on the mill to get the drill hole to line up.

drl2gcode.py:

• convert the drl file to gcode using the python script.  execute:

py drl2gcode.py --spindle-speed 10000 --drill-depth 2 Eyez-PTH.drl --safe-height 10

Milling:

• Align the PCB so the top edge parallels the X axis.

• Attach the raw PCB with double sticky carpet tape.

• Run the drill file.  I got some old dental drill bits from my dentist.  They worked great for through holes.

• Run the F_Cu gcode file using a 20 degree engraving bit with .1mm or .2mm tip.  If the board was positioned properly, the cutting path will match the drill holes correctly.

• Flip PCB left to right.  

• Load the B_Cu gcode file

• Align the PCB so the top edge parallels the X axis.

• Set the origin to the center of the through hole the was used in creating the B_Cu gcode.

• Run the B_Cu gcode.

This is a skill I want to learn.  I want to make an app I can talk to a bluetooth device.  The long term goal is to make an app for Kelvix using Flutter.

Todo:

• Try the flutter reactive ble package

It worked when I first tried it, but Flutter has gotten a major update and it doesn't work now

Words with Gringos is a game I would like to write in Flutter.  It is similar to the Monkey Wrench game with the twist that the clues are in English and the words are in Spanish.

The main screen is a playing area is a hexagon grid.  The clues page shows the words to be found with options to give hints for the answers.  The settings page defines the hints language and the solution language and other configuration data.

Progress:

12/14 - The game is taking shape. I can display the hexagon grid and randomly place the words on the tiles.  The tiles can be clicked on.  If it runs out of space it quits the word but still works.

Todo:

• Repack the grid to make space for words.

• Look into state managers.

Our motto is Hair'll be back!  It's got it's own page.

Ever since Ally went through chemo, her hair has been a bit thin.  One of the deals I found was a Hair Restoring device from Costco for $380.00.  It didn't seem like there was a lot to it, but many similar models sell for $1000 with 80 lasers instead of just 40.   I did some web surfing to see if there was any more "info" about it, and there certainly is a lot of talk.  There are lots of comments regarding how well it works and also how it does nothing.  There are even a few "studies" on it's effectiveness, and what I deduced is that maybe it works or maybe it doesn't.  But probably not likely enough to throw $400 at it to find out.  It doesn't really look like there is that much to it and I found a YouTube video  where a guy made it a DIY project: Laser Hat.  His website HowToLou has lots of other fun projects worth checking out too.

I found out I could get the 120 of the lasers for about $32 on ebay.  Just the entertainment value is worth that much even if it does nothing.  So thus began the Hairminator Project.

 HowToLou had a how to build it video.  I should be able to get 100 lasers for $30.

The first decision is 5v or 3v lasers.  5v works easy with an external supply, but 3v would would work good with 18650 batteries.

• 5v lasers require 20ma per device, so 100 diodes would need a 2A 5v supply.  They could be connected directly.  I could use a USB battery to drive it.  This seems like the easier option.

• 3v lasers require 40ma per device or 4A. the 18650 battery are 3.7v and probably around 3000mah (even though the batteries say 9800 mah).  I probably shouldn't pull more than 3A, so I could do 100 lasers at 2x 2A which would last 1.5 hours.  I would need to drop the voltage down to .7 volts, but I could have it  processor controlled and use a pwm and put in safety features.

The HowToLou version just pushed the lasers into some foam and glued it into a hat, then twisted the wires together.  A fancier way would be to 3d print a helmet or make pcb holders and solder the lasers on.  The overheating problem would be better with the pcb.  But I think I will do an improved version of the foam using 1/4" think foam sheets and molding into the shape of a helmet.

The next decision is how many lasers and what voltage.  5v seems the easiest.  Also, they are so cheap, so 120.

Safety:  I should add a momentary switch that clicks on when the hat is on your head.

12/30: I checked the order tracking and it looks like the lasers should be here soon, maybe today.  Time to start thinking about it more.  I ordered some glasses to filter the red light.  It looks like I should be able to control the lasers with a few mosfets and a processor.