Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Like gliding backwards through porridge.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



ring flywheel gimbal

a gimbal mechanism for ring flywheel
  (+4, -2)
(+4, -2)
  [vote for,

The idea of using a flywheel for storing energy in bikes is often suggested (see for example my idea 'bike hub with two motors') but has a significant flaw - the gyroscopic effect of the flywheel. The obvious solution is to mount the flywheel on a gimbal, but then what mechanism can be used to store and retrieve the energy to and from the flywheel?

I've come up with a flywheel gimbal that allows easy storage and retrieval of energy. See illustration first, then read on.

The ring flywheel (blue) sits on rails (grey). Cams (purple) are slidably mounted on an inside track in the flywheel, thus allowing each cam to slide a limited distance circumferentially. The flywheel and rails both spin together. The flywheel (not shown to scale) is heavier than the rails and thus has most of the rotational inertia.

When the bike tilts the ring flwheel stays vertical as it slides on the outside of the rails. The cams move circumferentially and their spacing changes - 'bunching up' where the edge of the ring is closest to the ends of the rails.

There are a few problems with this idea: the rails are adding 'ungimballed' rotational inertia to the system; lots of friction; the gimbal only allows limited range of rotation (fine for tilt, but not so much for turning).

This idea could also be used as a CV joint.

xaviergisz, Oct 28 2012

illustration http://imgur.com/a/9Svx9
[xaviergisz, Oct 28 2012]

"The Ringworld is unstable..." http://lifegivesmel...m/2011/11/endo1.jpg
[normzone, Oct 29 2012]

(?) An example of a flywheel electric load leveling/backup device http://www.beaconpo...smart-energy-25.asp
[CraigD, Oct 31 2012]

The Chrysler Patriot and possible F1 flywheel specs http://www.allpar.com/model/patriot.html
[CraigD, Oct 31 2012]


       I like this insofar as I understand it! +
csea, Oct 28 2012

       I'm wondering what prevents the flywheel from just falling off entirely, not turning axially, just sliding off to the side.You could add a larger element that rides inside the cage and prevents this.   

       Thus far it looks like a promising idea.
WcW, Oct 29 2012

       hmm... 'k, first off you could get away with only having three curved whatsises, but I think the main thing is that as the onion flourette turns it will slide up and down on the flywheel once each revolution... that and it will be transferring power at an angle so you havent lost precession so much as moved it around a bit.
FlyingToaster, Oct 29 2012

       Big, heavy outer ring. Three thin, stiff spokes connected to a ball. Ball clamped from above and below in a cage with 3 slots through which the spokes pass.   

       Probably not too hard to get +/- 45 degrees freedom about the axis.   

       Energy transfer is via the cage contact with the spokes, which have bearing surfaces - there could be roller bearings around the base of the spokes where they contact the cage to minimise frictional losses.   

       Much less machining. Simple. Fewer contact surfaces. Can be made arbitrarily large or small.
8th of 7, Oct 29 2012

       Considering the limited range of tilt, it seems like this idea is better if the flywheel is spining with a vertical axis (as in your illustrations). then oyu don't have to worry about turning, though you will have to worry about overcomming turning forces as you take energy out or put it back into your flywheel (unless you had a pair).   

       However I don't see any mechanism to keep the flywheel centered on the rails. When run as shown in your illustrations, it looks like the ring will just fall to the floor.
scad mientist, Oct 29 2012

       I agree WcW, a stop will be needed at the ends of the rails so it doesn't slip off. Also, there will need to be something to ensure the flywheel is constrained to the surface of the rails (the cams could be shaped to perform this function).   

       Scad, I don't think the flywheel needs to be 'biased' to the centre. Lets say the flywheel is stationary and tilting to one side; once the flywheel starts to rotate I think it will wobble for a bit, but then centre itself by centrifugal force.   

       FT, I'm not sure what effect precession would have on the flywheel. It's possible it could become unstable. I agree a smaller number of rails could be used.   

       I still think turning will be the biggest problem with this idea. The flywheel (mounted vertically) has two axes of freedom of rotation: tilt and turn. A bike can only tilt in a limited range, but obviously there is no limitation on turning. The solution to this problem is fairly rudimentary: dump energy (i.e. brake the flywheel) if turning outside a limited range is detected.   

       I had envisaged that the flywheel would only be used temporarily to store small amounts of energy - just enough to slow a rider from 20km/h to rest and then back up to speed again - thus the need for a large range of tilt/turn angle is lessened.   

       8th, can't visualise what you're talking about yet. I'll think about it for a while and try and illustrate it.
xaviergisz, Oct 29 2012

       there's also pitch as a 3rd axis, ie: going uphill or downhill. The precessional forces will come into play while you're loading or unloading the gyro when it and the mercator-ball aren't lined up.
FlyingToaster, Oct 29 2012

       //The obvious solution is to mount the flywheel on a gimbal, but then what mechanism can be used to store and retrieve the energy to and from the flywheel?//   

       An electrical motor/generator will allow a relatively frictionless system with full gymbalness.
Ling, Oct 29 2012

       Flywheel storage involving getting and putting energy into the flywheel using essentially the same technology as brushless electric motor/generators, is pretty well-baked technology, though mostly in large, heavy load leveling and backup power systems. (see link)   

       It’s had a long but checkered history in smaller flywheels mounted in vehicles, perhaps most famously in the Chrysler Patriot Le Mans racecar. The need to minimize mass necessitating reducing safety margins, while necessarily being close to people, and a catastrophic failure that involves lots of high-speed shrapnel, prevented the Patriot from being accepted when it was built and briefly raced in the early 1990s. (see link)   

       I’d not followed flywheel power for a couple of years, and see from the link that it may be up for a revival, again for racecars, this time F1. Rather than the big 67 kg flywheel used in the Patriot, the F1 proposal is said to be for a smaller, 5 kg, 400000 J one. At a glance, such a system seems nearly ideal for a bike or other human-powered vehicle.
CraigD, Oct 31 2012

       Could this keep a 'ring of cameras' (think multiple RasPi camera modules mounted on the outside of the ring) stable against the horizon?   

       Of course, RasPi camera modules need RasPi boards and power, so this won't work at the moment. But the future may find this useful for VR video recording. The entire setup could even be mounted on the outside of a helmet.
sonam, Jul 23 2015


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle