The summary of my latest studies

[path_finder]

Dear rasselasss,
On the prime reflexion, I'm afraid your mechanism will very soon KEEL like shown below.
The big question is: how to oblige any mechanism to rest in place at 3:00 or 9:00?

[rasselasss]

Thanks for that Path Finder,as usual i did'nt express my intentions very well............i'm suggesting the other side with spokes,weights rods etc tied exactly in the same manner..but the spokes ...positioned separately on the same hub but timed ...to take up so when one side is about to keel the other side is as the sketch taking it past that moment and vice versa .......As i said its only a thought ....you have an excellent grasp of the movement involved ...i do appreciate your knowledgable imput ..Thanks.

[rasselasss]

Apologies,Path finder,ii'm not thinking of the practicality ...it would have to be another wheel based on the mirror image of the sketch at the other axle end with the timing" keyed"to do what i described because of the "Pivot"for the weight rods.............i can't see the woods for the trees today....would it work?...thanks.

[path_finder]

As promised, and returned back home, I can now restart my experiments.
Based on the earlier drawing (remember here: http://www.besslerwheel.com/forum/download.php?id=10598), I have rebuilt a new wheel but now much more stronger.
The frame (second shot) has been made the lightest as possible (1,4 kg) in view do not compromise the rotation momentum of the swinging weights.
The next step is now to include the five freewheels and the chain (see the first shot) coming from a shop for bike's accessories.
The four pendula will be attached to the freewheels.
Then I have to build a stand with a large size: due to the excursion of the pendula at the bottom, I suspect the top of the wheel will be close from the roof (standard height here: 2,45m). The next week-end will be interesting.

[path_finder]

An attempt based on the same concept:
http://www.youtube.com/watch?v=YIDOD1jDDEo
IMHO the mechanism is to much complex.

edited:
Another one based on the same concept, of order three, but this time with an electronic positioning control:
http://www.youtube.com/watch?v=PfjT5Sn5apU
Please don't pollute the gravity religion with some iconoclastic artifacts!

[Mark]

... the gravity religion ...

OMG, I'm religious !!

:D

[path_finder]

Hereafter are some shots from the current building, which was not easy: the most highest difficulty was to find a way for attaching the bike freewheels and most important : how to lock them.
By the way they have a big hole (35 mm) with a screw path inside the inner rim, it was not possible to find a wired part which fits exactly.
On the first shot you can see the two flanges for each side of the freewheel.
Note the two pins which must be inserted in the two holes of the freewheel in view to get everything well linked (instead the torque could let rotate the parts, what I had as trouble in my previous building).
In the third shot you can see how the pendulum is fixed to the two flanges (and associated with the central ring of the freewheel), and how the axle is attached to the cross (frame of the wheel).
If you look well at the second shot you will see two bearings located inside the previous assembly: this is a mandatory for a correct actuation of the mechanism. Don't forget during the main wheel rotation, each pendulum axle rotates counterclockwise due to the mutual motion of the wheel and the suspended pendulum. Therefore some bearings are requested between the pendulum and the support axle.
This represents a long time work and I'm still in way to build all other pendulum assemblies.

[path_finder]

Some news from the building of the wheel in relation with the earlier drawing here:
http://www.besslerwheel.com/forum/viewt ... 6229#96229

I had several difficulties.
First the size of the wheel required a stand of great size. So far large size means strong frame (instead everything will dance like the washmachine). I shall abandon this idea and decided to use another way: the pillars like Bessler did it.
As you can see on the both shots below, I got two PVC tubes of 50mm diameter and 2m long, and also the corresponding screwed valves. By rotating this PVC valves I can increase the length of the pillar until a strong fixation (the same principle can be found on the market for some halogen bulbs holders).

Then it was very difficult to double the size of the chain (bike's model), to cut it at the right length, and specially to pull hard the two terminator together and to insert the pin. Alone it's not easy.

Now the concept seems conform to the wished behavior, even without the springs (which are not installed yet), but I have to make some important mechanical changes because the pendula are to much close from the freewheel cogs and I have some shocks between the weights and the freewheel axles.

I shall take care of myself, because a shock of a single weight in my head could drive me directly to the hospital.
This way to install the wheel (with the two pillars) has a big advantage: I can restore the whole assembly and remove it very quickly, during the absence of my wife, so far kept in the ignorance of what happens really in this flat.
Nevertheless I'm far from Bernard PALISSY which burnt all the home floorboards and furnitures in the oven for completing his experiments on the china porcelain process. For those interested by a brillant inventor, see here: http://www.palissy.com/NEWhistory.html

[path_finder]

As said by me several times earlier, there are several ways to get two wheels, rotating in the same direction but with two speeds in a 2:1 ratio.
This old post shows one: http://www.besslerwheel.com/forum/viewt ... 9635#69635
and specially below in the page, this drawing:
http://www.besslerwheel.com/forum/files ... build1.png

Sometime the old ideas are useful and need to be completed:
http://www.besslerwheel.com/forum/viewt ... 9222#79222
Combining these both ideas, I found recently the mechanism we need for the Raj concept.

[path_finder]

Some (bad) news from my latest building: I was unable to let run this wheel more than a half turn.

I made the important changes I identified previously, in view to resolve the problem of the collision between the pendula and the chain.
You can see on the shot below the details: now the pendula and the chain are in two different parallel planes and are separated with an acceptable clearance.

This point has been solved, but another trouble appeared: the chain jumps over the remote teeth.
Because of the distance between two consecutive freewheels, and due to the bad parallelism of each of the freewheel, there is no way at all even to start the experiment.
The swing (in yellow) is important even at a slow speed.

The solution could be coming from the installation of some guide rollers, but this seems to me to much complex.
In fact the best way is to concentrate the five freewheels in a small area around the main axle, and duplicate the position of each freewheel thanks a rigid rod linked to the remote axle of the pendula. I have to think about.

Once again only the practical building can educate us for a real working wheel, whatever the concept used.

[path_finder]

Hereafter an animation showing how the freewheels can be replaced by some simple locking pins.
They are located on a excentered circle rotating at the same speed than the main wheel.
These pins cancel the centrifugal force on the left side of the wheel, but leave free the pendula to swing on the right side thanks the centrifugal force.
(The animation is slow for a better understanding, but the real speed should be much more high for an acceptable torque given by the CF)

The springs are not represented (the purpose is to accelerate the extraction of the pendulum from the 12:00 position for a more efficient centrifugal force).

Note this point: the centrifugal force is active during the left side path: there is a lever effect transmitting the locked CF into a torque in the clockwise direction.

Just a concept for the moment. A building is needed for the discovery of the mechanical troubles (in particular the relative position of the parts).

[jim_mich]

Path_finder,

Your animation shows the weight moving (by CF) from the wheel center straight out to 3:00. Unfortunately this is not how CF works. Such a weight always moves at a trajectory path that is somewhere between a straight line tangent to its motion before being released (Path 'A') and its circular path when held and not released (Path 'B').

CF is an outward force but it never causes a weight to move straight outward.

[path_finder]

Dear Jim_Mich,
Many thanks for your comment.

This is the purpose of the string.
As soon the weight has acquired a sufficient distance from the center, the CF acts.
So far the CF acts on the right side by the created unbalance and on the left side by the reaction torque on the pin.

[KAS]

Path,

If you move the bobs so that the elbow pivots are on the periphery, the tangential CF force that Jim is rightly talking about could act on its mass sending it horizontally to the 2 o'clock position.

IMO, I don't think that there is any milage in this design however as it relies on a second wheel to drive it; and the same CF reaction force would put huge negative pressure on the levers on the ascent; counteracting any advantage.

Kas

[path_finder]

There are some springs but not used as you think

My latest experiments let me discover an important (but very simple) concept.
If you remember above I discovered that a freewheel was not sufficient for limiting the swing of the pendula (this is due to the fact the axle itself rotates within the rotation of the main wheel, instead to be fully constant in the polar position).
Therefore I removed all accessories (including the chain of the latest version) and decided to spend some time observing the motion of the four pendulas just hanged on their axle.
Immediately the solution was obvious: what do we want? a permanent unbalance? there is a trivial way to do that, without any complex mechanisms: to keep the four pendulas in the same obliquely position.
Now the practical solution is easy: imagine a big spiral spring on each pendulum axle, and let suppose this spiral spring has been preset for a certain angle at the keeling position: you get now this reference position, because each time the pendulum will make an excursion the spiral spring will compensate.
We must now complete the system with a freewheel, in view to limit the swing of the pendulum during the ascending left side (assuming a clockwise rotation of the wheel).
So far the combination of the spiral spring and of the freewheel will give us the wished effect (extension on the right side under the CF, restriction of the swing on the left side).
But what is much more interesting is the fact in some cases the freewheel is not needed: the spiral spring can give the same effect when fully tied.
In the drawing below I resume the principle. The spiral springs are in violet (expanded on the right side, and compressed on the left side).


IMHO the Prince Karl did see absolutely nothing. He just saw the four pendulas ('so simple that a simple carpenter apprentice...etc') but never detected the tricky way these pendulas were linked. At that level Bessler had no danger to show him the inner of the wheel.
There are two Bessler's drawings where a spiral is represented (MT94 and MT95) but used with some water (what is obviously an intellectual fraud, everything being balanced).
IMHO Bessler want not to discover to much on his solution, so far showing a detour.

my signature seems to be pertinent...