Trainset

Generate all thread pitches your lathe is capable of using some inputs, with the set of gear trains needed on the banjo!

Building

cargo build --release --features yaml

Running:

Trainset has two modes:

• generate
• lookup

Generate will generate a database file, and a plain text file of all the pitches the lathe can produce along with the gears needed to produce that pitch.

Lookup will allow a thread pitch to be entered (after the database is generated), and trainset will search the database for the desired threadpitch showing all values within 2 decimal places of the target. (on the TODO list)

Running a generate:

./trainset generate CQ6125.lib.txt CQ6125.gb.txt 40 60 2 CQ6125.out.txt

These settings are for my lathe, a Chinese "CQ6125". The Grizzly G0602 will be similar, though the library and leadscrew pitch will have different values. The gearbox will be the same however.

• CQ6125.lib.txt is a library file containing all the gears
• CQ6125.gb.txt is a list of the gearbox ratios (or gears if known inside the gearbox).
• 40 is a gear attached to the spindle
• 60 is a gear driven by the spindle. Typically this will have the "A" gear on the same shaft.
• 2 is the pitch (in mm) of the leadscrew
• CQ6125.out.txt is the output text file.

Running a lookup

... on the TODO list. But it'll be something like this:

./trainset lookup CQ6125.db 2.5

to get a listing of all combinations close to 2.50mm pitch.

TODO list items

Some of the TODO stuffs:

• lookups
• write a binary file for storing gear combos with fast lookups
• write the text file to an output rather than to STDOUT (screen)
• merge the different configuration files (.lib.txt, .gb.txt, and the three magic numbers into a single configuration file)
• less .clone()s in the code
• optimise CPU
• too many .unwrap()s in the code... can I fix this?
• sqlite is kinda sluggish (about 10x slower) compared to just writing out a text file

Thanks

Many thanks to Mitchell Hunter [mitchicus] for putting up with my troubleshooting of Rust.