This is a collection of code for exploring delta printer calibration. The idea is to use a proper computer to send commands to the printer, telling it how to move and where to probe, returning the results to the computer. Then on the machine where this runs you can use the full tools of modern programming to work with the results: visualization with matplotlib, numerical optimization with scipy, additional constraints obtained from measuring test prints specified by hand, and interactive use with ipython notebooks. The resulting calibration data can then be transferred to the machine and saved for use in normal printing.
This is not (yet?) a user-friendly calibration tool.
As currently configured, this tool runs on a user-facing machine and communicates with the printer by making a telnet (arguably RFC2217) connection to a port on a printer-control machine that forwards the serial data back and forth. The code could be modified to communicate directly over serial with a printer, but I haven't done this because that would be inconvenient with my setup. In order to make this work, on the machine directly connected to the printer, run
tcp_serial_redirect.py /dev/ttyACM0 250000or whatever your serial port and baud rate are. This will also echo
all communication to standard output, so you can see what's going
on. You can run this under dtach, tmux, or screen, but it often
needs to be restarted anyway. Currently this opens port 7777 to all
connections. If you don't want the whole world to be able to take
your printer for a spin, run this only on an internal network. A
little jiggery-pokery would allow connections only from localhost,
which would allow ssh forwarding to provide secure access.
So far, the ipython notebooks primarily provide demonstrations and
testing for the tools in the Python files. In particular,
probing.ipynb demonstrates what can currently be done; simply
clicking on it in Github should allow you to see a rendered version
of the notebook with graphics.