Most motorcycle turn signal cancelling cicuits use a timer. Because motorcycles must lean to turn, installing a tilt switch in the turn signal control circuit would allow the blinker to turn off automatically after a turn instead of a predetermined time limit.
-- ato_de, Feb 03 2006

Comus international http://www.comus-in...om/comustilttip.asp
Everything you never wanted to know about tilt switches. [ato_de, Feb 03 2006]

Useful, except for the times you want to leave the turn signal on because you're just about to make another turn, or when you start to rely on the auto-turn-off feature and your turn signal fails to cancel when you make a turn onto a road with a long, continuous turn, giving the false impression you're about to turn down a side-road.
-- hippo, Feb 03 2006


Problem: even when turning, the apparent acceleration experienced by the motorcycle is approximately through the motorcyle's vertical axis. That's why you have to lean in the first place. If the rider consistently places his/her centre of gravity on the inside of the curve relative to the motorcycle's vertical axis, as far as the tilt sensor is concerned the motorcycle will be leaning towards the _outside_ of the bend. The movement of the tyre contact patches towards the side of the tyres on the inside of the curve has the opposite effect. So, the behaviour of the tilt switch would be hard to predict.
Solution: Use an angular accelerometer to detect the _change_ in lean as the motorcycle begins to stand up after the turn. This still might not be able to distinguish between the end of the turn and small adjustments in lean during the turn, but then again it might. [edit] No, wait, I've got it. It just needs a gyroscope. I'm sure you can figure the rest out.
-- spidermother, Feb 03 2006


And a periscope.
-- normzone, Feb 03 2006


Our car switches off the turn signal when the wheels straighten out (not always useful on a complicated intersection).
-- DrCurry, Feb 03 2006


I can't think how this could work reliably, because a slow lane shift requires very little leaning, and sometimes you might want to turn left on a road that is bearing to the right, and so on.
-- Ling, Feb 03 2006


How about balloon based cancellation?  On the back of the bike, behind the rider, is an enclosed chamber in which a helium balloon is tethered yet free to find its position within the chamber.  Machine vision is used to determine the relationship between the balloon orientation and the chamber.  After a turn signal command is issued the system interpreting the chamber video looks for a chamber lean threshold value in relation to the balloon which, presumably, remains truly upright, to indicate that a turn has been initiated.  The system then looks for the chamber orientation to return to an upright orientation measured against the balloon and once that is achieved the system issues a signal cancellation command.

Alternatively, in place of the machine vision system, the balloon tether could be attached to a very freely moving joystick that is used to determine when the chamber and the balloon orientations match.
-- bristolz, Feb 03 2006


I hate to say it, but this one has almost been baked. Harley-Davidson has been doing this since 2003, there is a little computer that measures how long the turn signal has been left on and also how far the bike traveled since the signal has been on. I'm not sure how they programmed, it but I was impressed the first time I rode one, at first I thought it detected lean because it was so accurate, but as it turns out, it just calculate distance and time. As an added bonus to this system, you never leave your turn signals on too long when after you've changed lanes and forget to click them off. (In a lane change situation, lean is minimal because of the higher speed. Even with a gyroscope it would have to be pretty accurate.) I'll still give it a crossaint because-

1- As far as I know, nobody has tried it. (Probably because of the problems [Spidermother] pointed out.

2- I love bikes and welcome any idea that could remotely make my rides more enjoyable.
-- Pac-man, Feb 03 2006


[Bristolz] I wish I could give you a bun for that annotation!
-- ato_de, Feb 03 2006


[brstolz] For the balloon to be self-righting it would need to be bouyancy-polarised (denser at the bottom). But then its uprightness would be compromised by the redistribution of air pressure in the chamber during acceleration. So the balloon still needs a gyroscope. Oh, and [normzone] a periscope to peer into the chamber. (Or is it a Peer-O-Scope?) (edit) Sorry, your balloon is tethered. Same problem though. Hold a tethered helium baloon in a moving vehicle and watch what happens when the vehicle accelerates. Some good physics in that.
-- spidermother, Feb 04 2006


I know, the balloon goes the opposite way from what you'd think but, we'd just have to fix that.
-- bristolz, Feb 04 2006


A conveyor belt system that frequently replaces each balloon with a freshly-inflated, correctly oriented one? ;-)
-- spidermother, Feb 04 2006


Fully baked.

You can buy after-market turn signals with a gyroscope that self-cancels after a turn signal is used, and the turn is completed. My friend has on on his GSXR1000.

Sorry to burst the bubble :o).
-- DeliveranceRiverBoatCaptain, Feb 04 2006


Cool. Do you get an artificial horizon display?
-- spidermother, Feb 05 2006


I guess I didn't get the joke, [spidermother]. :o)
-- DeliveranceRiverBoatCaptain, Feb 05 2006


Huh? Which joke? Can you explain it to me, please?
-- spidermother, Feb 05 2006


[DRBC] First; Your friend probably doesn't understand how his turn signal works. Second; even if it does work using a gyroscope, it is outside the scope of this idea. Third; second hand hearsay is not evidence, post a link.
-- ato_de, Feb 05 2006



random, halfbakery