This script helps gearbox design by finding the optimal gears for a gearbox that
fits inside a particular range of ratios. In other words: If you need a D220-D231
gearbox, enter 220 an 231 as the range below, and it will tell you what gears you
need to get a final ratio of 220% to 231%.
The output format is specified as an equation: Each division indicates gears with
that many teeth 'meshed', and each multiplication indicates a gear on top of another.
It's simpler than it sounds, here's an example for a gearbox that has a ratio of 42%:
| Formula |
Cost |
(3/6) * (5/6) = 42% |
4 medium gears |
This indicates that you should have a 3 tooth gear driving a 6 tooth gear, which is stacked with
a 5 tooth gear driving a 6 tooth gear. Invisible in here, due to the shorthand is another 5 tooth
gear, as a 3 and a 6 cannot be directly connected. Here's the shorthand ratios used:
| Shorthand |
Expansion |
| (3/6) | (3/5) * (5/6) |
| (6/3) | (6/5) * (5/3) |
| (4/5) | (4/7) * (7/5) |
| (5/4) | (5/7) * (7/4) |
The cost column gives the cost in medium gears. Small gears are costed at nothing, and large gears
are equivalent to 6 medium gears.
A few points:
- The first result will always be the absolute cheapest way to create that ratio, assuming
iron is your primary cost. Other results are listed in case the optimum
result has a set of gears you do not have, or you are building a multiple-output gearbox and
need several results to work with.
- This tool does not design the gearbox for you. You still have to take the proposed gears and arrange
them on the design board, and you still have to make it reach the output peg (adding
gears as needed)
- The cost value includes only iron for medium and large gears.
- This tool does not handle multiple output gearboxes. If you're designing one,
the tool can be useful in determining the best set for any one output and for determining
the best set to get from one ratio to another, but you still have to do the work of trying
to combine these for the best multiple-output box.
- This tool does take into account what gears can be positioned by what other gears.
It should never generate a pair of gears that are not possible to mesh.
For the existence of this tool, I owe my everlasting gratitude to Isomer, for hacking up in
python the tool I originally used to search for the best gearboxes, Tamutnefret for upstaging
me by having already created a list (that list now drives this script), Zubon, for his
post containing
price estimates for the gears, possible gear arrangements, and other information, to
FaceAnkh, for making the sed one-liner to convert the original data into a format more usable for the
database, Brant|SN and TheDude[OWRH] for their brief participation in the Distributed Optimal
Gearbox Effort, and to Garan for his assorted contributions in IRC.
I've now put up the source and a dump of the DB for anyone that wants to mirror this or create their own.