Fundamental physics applied to a body of fluid (or a container of bb balls).
The above is the same principle behind capillary action.
Water molecule is attracted to the tube by electrostatic charge. However, hydrogen bonding creates surface tension that makes extracting energy from capillary action difficult.
Solution: bend over a capillary tube. A meniscus will form. Arrange 20-30 tubes together and touch the meniscus' together. The hydrogen bonding will overcome the electrostatic charge to the tube and a water droplet will form. With enough tubes arranged together, the droplet will be large enough for gravity to get a hold of. How does water get into a coconut?
The above is the same principle on a larger scale that can be used to extract useful work.
IMPORTANT: the 2nd Law of Thermodynamics discusses entropy. Mathematically, entropy is q/T (heat/Temperature) and applies to heat engines. So 2nd Law of Thermodynamics is inapplicable to the above.
this is a diagram of an experiment i did. if you imagine a flat rubber sheet and the magnets as being dents that allow the ferro fluid to puddle. any place on the surface of the FF around the neo magnets, is EXACTLY like the level of a pond or lake. the small magnets are like a trough leading to a large pool, [big magnet] all of it having the same level, the pool being deeper and wider. there are only 2 forces, attraction/repulsion. i DID this experiment and it absolutely works. a constant flow machine. this flow must be started, not self starting. the ferro fluid will easily climb 2, 5, 10 feet, up a neo magnet chain....david
http://img822.imageshack.us/img822/5953/continiousflow.png this is the diagram.....david
Here ya go Dave,
hopefully this will help get rid of the mess, and make that a little easier.
select a tube slightly larger than the magnets so theres room to flow
angled opening
now, install the magnets in pairs, one inside the tube and one outside to hold it. and slide them up the tube into position.
another funnell at the top for it to drip down from.
your dish wont have to be much more than a couple "tubefulls" of fluid.
and only needs to be deep enough to get the first "inside" magnet close to the surface.
you want the first "outside" magnet to be up and around that funnell-inlet, so the fluid doesnt creep up the outsides of the tube.
then maybe you can move it around and get it on video