dougp:
But you could also count revolutions with an encoder and use the line thickness to calculate (the constantly changing) spool diameter and therefore how much line goes out for each revolution. Is there some way to insure the line is evenly laid across the spindle? If so, a distance sensor mounted to examine the spool could provide a number by which to calculate the amount of line on the spool à la figure H.6..02
dougp, this is definitely a clever solution. However, as you alluded to yourself, I think the main problem would be that there isn't a good way to ensure that the line is evenly laid across the spindle. The other problem I can see is the the line doesn't wrap around the circumference of the spool perfectly each time. Inherently, it will wrap at (at least) a slight angle, which means each wrap is larger than the circumference by some indeterminable amount.
Definitely an interesting idea though. Thank you for the suggestion. At this point I'm leaning towards wvmarle's suggestion to use a second spindle and monitor that somehow. I need to decide between a rotary encoder, some type of optical sensing setup, or a Hall effect sensor. My concern with the rotary encoder is that it could experience mechanical wear over time from being turned so much. So, I'm leaning towards the optical sensing setup, but I've never used a Hall effect sensor in this way.
I'd be curious to know people's opinion on an optical sensor vs. a Hall effect sensor for measuring the secondary spindle.