A simple cam

[Tarsier79]

Looks promising, though your spring tension will have to cause less friction than the positive effect of the gravity in the overbalance portion.

Look forward to the results.

[path_finder]

The idea of the cam is not fool at all.
The principle is simple. The corresponding animation is here:
http://www.besslerwheel.com/forum/viewt ... 6439#56439
The question is: how to implement on a practical realization (result), this theoretical animation (challenge).

The use of a fixed cam inside the wheel is facing to an important mechanical constraint: if the main wheel rotates, how to pass trough the wheel the needed axle.
Fortunately one of the possible solutions (passing the fixed shaft inside the main axis of the wheel) has been explained here:
http://www.besslerwheel.com/forum/download.php?id=6185

Now an important question is: what is the ideal shape for this cam.
The animation hereafter resumes the parameters of this design.
The dimensions of the cam are dependent from the radius of the weights, this is the reason why we must first setup the red weights in the right place.
During the return (left side if we assume a clockwise rotation) the shape is an half of circle where the weights are rolling on.
Then at 12:00 and 6:00 we can draw the two horizontal lines, wich give us the arc of the outer rim to be translated (indicated by the two black arrows).
At this step we have the cam, but we must install the axle for the springs connection: it shall be at the middle of the excursion.
The small green circle shall rotate on a bearing fixed on this cam, and on this small disk the eight springs will be attached for a connection with the corresponding weights.
The last problem to solve is the balance of this subassembly: the cam is asymmetrical and in addition is supporting the small disks for the springs on the same side .
This is the reason why we must include a counterweight on the other side (the grey circle). This requires a good adjustment.

The building now can be engaged, saying that the inner part of the main wheel will be separated in four volumes:
- a volume containing the cam (essentially in the middle of the wheel).
- a cylinder near the first side of the wheel, containing the shaft of the cam, the rigidity cage with the bearings, and the counterweight.
- a cylinder near the opposite side of the wheel, containing the small disk and the eight springs
- a volume (essentially around the inner rim of the wheel) containing the eight weights, the rods and the retaining wires.
The main wheel must have a bearing on the side where the cam's shaft has to pass trough, and on the other side has only an attached shaft for the power recuperation.

[sadiq attamish]

hi
path_finder try to make the movement of the weights in elliptic path (which he is right),but not by forcing them(which this is wrong).it should be the movement in arc(elliptic) by the nature or according to the physic.??,thanks

[path_finder]

After the animation, the real world.
The shot hereafter shows the side of the wheel in relation with the cam and its accessories, respecting the rules (see above).
The outer shape of the cam has been colored in dark red.
Note the double cage bearings, an elongated cage with the purpose to assume a perfect (almost?) orthogonal position of the axis for the shaft supporting the cam and others items.
The weight on the half-circle side of the cam, is the counterweight for the balance (almost?) of the cam and the springs disks versus the shaft axis.
I has been obliged to test some various diameters for this disk, in accordance with the length and traction force of the springs I had at my disposal.

[sadiq attamish]

hi

i dont believe it is just a matter of weights running around the helical path,we know from his note(bessler) he was putting together tow or three systems.

[path_finder]

Here is a shot of the final assembly.
I have been obliged to remove all the pins around the rim of the wheel, in view to pass my hands in the inner of the wheel for the adjustments.

In that state the original iron springs are replaced by some rubber straps: the previous strings were too much strong and not long enough.
The selection of a correct size/force for the strings remains an important problem.

This wheel (on the shot) had a big fault: the acting force of the springs being located inside an excentered plane, the weights are not still correctly aligned when the excursion reaches the maximum distance from the springs disk axis (between 2:00 and 4:00) like indicated by the red arrows on the shot.
In addition for that wheel I used some rods made with a electrical wire, so far the rigidity was not enough for a compensation of this misalignment.
I have been obliged to reconsider the mounting of the weights and the application way for the springs.

Nevertheless the torque seemed to be efficient in any position of the wheel, at the condition to fix the cam position at 3:00.
On the shot the wheel is locked to the ground with the white rod (see at 5:00).
In addition the nylon wires are not visibles, thus I colored them in violet on the shot.

Again, all these problems have been detected by the building.
I don't think any simulation software could be able to indicate such as disturbances.

[path_finder]

Returning back for a moment to the 'cam design', I made the modifications needed by the various misalignments.
The rubber springs are now attached very close of the central roller of the weights, and the traction force is therefore very close (about 2mm) from the cam's plane, allowing the weights frames to preserve their orthogonal geometry..
The torque is real and important (like already observed in the previous experiments).
The only remaining problem is the locking point coming from the shape of the cam.
But before to cut this cam for a new shape (half ellipse) I want to be sure that the main pendulum value is correct.
By the way I have observed that a correct motion is obtained when the cam has an alternative motion, improving the passage of the rolling weight at the upper corner.
The cam seems to have an additional role of pendulum, accelerating then slowing the rotation of the weights.
This could explain the presence of the two external pendulas in the first Bessler wheels (just a suggestion).
But I made a big error with the position of the counterweights on the cam: they must be at 9:00 and not at 6:00 as made up to now.
The first shot is correct and the second shot still includes this error.

[Trevor Lyn Whatford]

Here is a shot of the final assembly.
I have been obliged to remove all the pins around the rim of the wheel, in view to pass my hands in the inner of the wheel for the adjustments.

In that state the original iron springs are replaced by some rubber straps: the previous strings were too much strong and not long enough.
The selection of a correct size/force for the strings remains an important problem.

This wheel (on the shot) had a big fault: the acting force of the springs being located inside an excentered plane, the weights are not still correctly aligned when the excursion reaches the maximum distance from the springs disk axis (between 2:00 and 4:00) like indicated by the red arrows on the shot.
In addition for that wheel I used some rods made with a electrical wire, so far the rigidity was not enough for a compensation of this misalignment.
I have been obliged to reconsider the mounting of the weights and the application way for the springs.

Nevertheless the torque seemed to be efficient in any position of the wheel, at the condition to fix the cam position at 3:00.
On the shot the wheel is locked to the ground with the white rod (see at 5:00).
In addition the nylon wires are not visibles, thus I colored them in violet on the shot.

Again, all these problems have been detected by the building.
I don't think any simulation software could be able to indicate such as disturbances.

Hi Path_finder,

looking at your design it looks like it would work better in a anti clockwise rotation with the levers allowed to pivot freely (360 degrees movement), when the weights roll over the cam most of its weight will be taken off that side of the wheel as it is supported by the cam which is grounded and will be replaced by a pulled roll instead of a lift the more levers add to the wheel the more weight would be taken off the ascending side, so you get a imbalance by removing weight.

Because the top of the cam is level the pull over it will take less energy, just some thought on a different option of the cam use.

Edit, if you could also catch the weights swing after it leaves the cam the weighted levers could also be outward extended on the descending side of the wheel, all you would need is one way gates strategically placed and I think we could be onto a winner, see what you think.

Edit again I was hoping the one way gates would fall open under there own weight on the ascending side to allow access for the levers to move over the cam but it would not take a lot to sort it out if it needs sorting.

Edit, see, ideal_cam2.jpg above.

Regards Trevor

[path_finder]

Dear Trevor, many thanks for the suggestions.

Hereafter the new cam, including the two major above corrections (counterweights position and elliptical shape), before reassembling. Let cross your fingers.

[path_finder]

The first shot shows the main wheel (without the support frame second flange) in an horizontal position.
In this version the three counterweights are still here (three red crosses).
During the first tests it was obvious this was a wrong 'good idea': these counterweights generated a counter torque equal and opposed with the torque given by the eccentered part of the cam under the pressure of the weights. The cam was not only balanced but the torque also, as you can see on the second shot.
Therefore I removed them definitively.
This wheel rotates but stops rapidly. This is coming from the last problem I have to solve: the retention green wires are attached around the black rollers (for a mechanical reason in relation with the symmetry of the weights), and when these rollers rotates along the rim of the cam, the green wires enroll themselves around the roller and stop it. So far I'm now in way to reconsider the attachment of the wires, in view to leave the rollers totally free to rotate.
I must also install a lever in the main shaft for the manual control of the cam (with the nude shaft it's very hard to keep the position).
Rotating the main shaft allows to trigger the torque or to stop it, moving the cam from 6:00 (keeling position) to 3:00 (maximum torque).