Building a self looping "SMOT"

[maw2432]

http://www.youtube.com/watch?v=vINnB7tii8s

Happyfunball, 
Good find.   A self-looping marble run would be a good selling desktop toy. 
Bill

[maw2432]

I am not sure about the width of this track but maybe this would work?
See attached photo. 
Also if you decide to build like a marble run, take a look at this how to video.
http://www.squidoo.com/HowToBuildAMarbleMachine

Bill

[MeggerMan]

Hi Jouleseeker,


I have not got a closed loop running. 
All I have at the moment is some straight 1m aluminium channel, a large ferrite magnet and about 4 x 24mm ball bearings.


My experiment was a simple proof of concept of the idea about the ball being pulled up the track by the magnetic field and gravity pulling it back down again and the magnet pulling it back up again and so on.


But what I have found in this testing is that having a single pole facing the track is not as efficient as having the poles facing at right angles to the track (facing up/down).
The single pole facing the track is very "lumpy" even for a single 6" slab magnet.
So for a series of magnets, it too would appear very lumpy and this effects how evenly the ball is pulled up the track.


I will try to take a short video showing the difference in the two concepts.
With the right setup I was able to get the ball to move up and down the rail for about 38 seconds, with the tail end being just a few mm of movement up/down.
Elecar may be able to improve the working model by adopting a magnet array that has poles at right angles to the track.


@Maw2432,
The ball will probably ride up on the curve of the exit track and fall into the magnet array.
I think the switch points need to be around the mid point of the rails as per Elecar's drawing and the track banked away from the the magnet array slightly.
There may need to be a curved guide to push the ball across to the exit rails at the top of the climb - but you may be able to achieve this by an increased bank on the track to allow gravity to pull it over. Elecar may be able to comment on this.

I have some 19mm bearings on order and I will be getting some aluminium tube/rod to build the exit points and some more ferrite block magnets. Once the exit points work the rest should fall into place.


Thanks
Meggerman

[happyfunball]

You can just use wire hangers and epoxy or super glue the connecting piece to the outside of the rail. Easily bent /cheap/ low friction.

[lumen]

So now we have the same thing that always happens, we find several ideas as to why this impossible event can work.
If elecar has this working, then it's possible that he does not know exactly why it is working or the exact effect that enables it to operate.

It could be the ceramic magnets, the order of the sizes, the direction of the poles or many other reasons.

The only effect I know that may cause this is that any magnet attracts steel more when both poles are near as opposed to only one pole and especially the center of one pole.

So given that effect, when the ball approaches a large magnet, there is a greater attraction because it sees both poles. As it continues to the center, it is in effect shielded by it's single pole face and the attraction is greatly reduced.

If you place a large magnet under a sheet of glass and let a ball be pulled into it, it will gain enough energy to cross all the way to the other side of the magnet, moving through this weaker attraction area. Sometimes it will cross all the way to the other side and be pulled back with enough force to again cross the weaker area, back to the original pulling edge.

If the magnet was long enough and there was a track to keep it centered, it would cross the original pulling edge with acceleration and follow the track to a point of less attraction where it could be removed with less force.

I do not know if this is what's going on with elecar's device but it's another possible idea.