On Mon, 21 Mar 2005 09:08:33 -0500, Ned Simmons
wrote:


There's one more piece of information you could throw in to
narrow things down. Presumably the input and output of the
gear train are coaxial as on a BP, so if the pitch on both
gear sets is the same

a+b = c+d

If the pitches are different

(a+b)/dp1 = (c+d)/dp2

where dp1 and dp2 are standard diametral pitches in a
reasonable range, say 8,10,12,16,20. An estimate of the
center distance between the two shafts would limit the
range of possibilities further.

Interesting observation, Ned!

Assuming that the ratio for such a speed reduction is somewhat
arbitrary, (so why not 6.25: 1?) and they wouldn't use overly-large
gears if they could help it, I widened the tolerance to .001 and
limited the tooth counts to 50,30, 50, 30, then computed the
difference you mention (a+b -(c+d)) for each set that made the cut.

Results: 39, 20, 44 and 14 give a ratio of 6.128571 with a difference
of 1 tooth in the sums, while 44, 19, 45 and 17 give a ratio of
6.130031 with a difference of 1 tooth.

The best I could do with DP variation was 22,10, 39,14 with DP's of
12 and 20, ratio of 6.128571 -- not

Now, Karl, it's time to count teeth and see what the *real* story is!