tinycnc

a belt-driven gantry-style cnc pen plotter running GRBL

CAD onshape v1

CAD onshape v2

devlog

bom

• rpi 5

• rpi level shifter

• rpi lcnc gpio <-> parallel port

• cnc breakout board https://www.amazon.com/dp/B07S5JSWC6

• 3 nema 17 steppers https://www.amazon.com/dp/B0B93J9QZY

• 3 a4988 stepper drivers https://www.amazon.com/gp/product/B07BND65C8/

• 6 guide rods https://www.amazon.com/gp/product/B09VDHKQLB/

• gt2 timing belts https://www.amazon.com/dp/B08R93QQ8Z

• gt2 pulley wheel

• linear rail mgn15h https://www.amazon.com/dp/B07QZ25F53

grbl stuff

C:\Users\anand\AppData\Local\Arduino15\packages\arduino\tools\avrdude\6.3.0-arduino17\etc\avrdude.conf

avrdude -p m328p -D -PCOM4 -c arduino -b 115200 -U flash:w:grbl_v1.1h.20190825.hex -C C:\Users\anand\AppData\Local\Arduino15\packages\arduino\tools\avrdude\6.3.0-arduino17\etc\avrdude.conf or use xloader2

grbl settings https://github-wiki-see.page/m/gnea/grbl/wiki/Grbl-v1.1-Configuration

1/21/25

• gopro mount on z
• redesign z to be rack and pinion + spring
• cloudflare site setup

todo:

• fix bom (30 mins)
• try to take a short video of it drawing using a rpi camera or gopro, I'm down to just hold it cuz I don't have a long enough cable
• replace xidler with the new tensioner
• microstepping
• load cell on pen, have control over force applied to page
• for the supported gantry, do the physics to see how much longer the y axis can be made
• look at how bed leveling works on a 3d printer, it would be cool to do that, but it would mainly be measuring deflection since that is way more of a change than how flat the wood board is
• find out what it would take to do closed loop with angular encoders
• improve the testing situation- it would be nice to log information on the page printed about what type of pen- what grbl settings and what cad settings were used to produce an image
• have some actual gcode written tests, like a bunch of lines spaced 1mm apart, and a big circle, stuff like that
• art- try a really thick pen (requires cad change) and try to draw some cool graffiti
• CAD tool changer
• far out but could use the z axis to control rolling paper across
• redesign to use 3 rails (2 500mm ones) increases the print area
• download all the timelapses on the SD card of printer. good for the video
• finally make a pcb to wire this thing, basically take 12v 5a barrel with a pretty good 5v regulator. then give power to shield but return the servo data pin.
  • then to unplug, just remove the barrel. no more 2 wall wart nonsense
• fix the limit switches to not be so dang loud
• https://developers.cloudflare.com/pages/framework-guides/deploy-anything/
• get a metric hex bit set for a drill so assembly doesnt suck so much
• board housing
• use a 775 spindle

shelved:

• try the quieter drivers (the pinout is different :/ )
• mount a time of flight on the z axis and do bed leveling or rather more a measurement of deflection
  • the precision of the TOF is 1mm so the picture wouldnt look that cool not worth

done:

• add little clamps to hold paper down instead of using tape
• make the pen holder even longer (closer to ground to improve deflection)
• finish assembly and hardware for the whole design in onshape
• fix tensioner to not limit travel in x
• minimize threaded insert usage and opt for tapping (particularly for thru holes, threaded inserts in blind is ok)
• instead of slotted nut put a cbore under rail, because the support removal for slot is a nightmare
• also for the cbore on the bolt that goes thru rail need to increase diam
• cbore on the bottom nuts for x axis motor and currently they are super inaccessible
• pegs for belt holder are too weak and can easily break (3mm to 4mm)
• slot for belt tensioning is UX nightmare terrible mechanism
• try 2 sets of linear bearings on Y see if deflection improves
• redesign the rail holder parts so stepperx is on other side, the screws can actually be put in, and the rail actually gets fully surrounded, on the idler side, make a slotted section to more precisely control belt tension, also surround rail and make more rigid and so screws can actually be installed
• cad changes for pen to be removable from the top
• tof mount on z
• redesign it to be a supported gantry
• beam bending physics, work out the expected deflection given the mass of the z axis
• make the pen holder way longer - requires making the z wider so that the pen can be taken out upwards