A suggestion for the unidirectional Gera wheel

[path_finder]

Dear greendoor,

Hereafter a serie of shoots extracted from a video made during my experiments.
(The size of attached files is limited here, so far I made some jpg compressed pictures with an acceptable size and quality)
I don't know from where the energy is coming from, but I suspect this is from the simple gravity.

These pictures are directly related with the design I suggested few days earlier for the Gera wheel
(it can be seen here: http://www.besslerwheel.com/forum/downl ... ef4cfa6dd9)
The shown sequence is this one where we need to lift-up the outer weight between 4:00 and 6:00.
In fact there are some differences with the original animation.
The number of the weights is kept to EIGHT, wich drives to some important problems of promiscuity (conducting to the special shape used for the inner pendulum).
In addition the lever has a ratio of 2/1 instead to 1/1 (wich appeared not strong enough).
The small locker at the right side down (I'm removing by hand) is needed to avoid the rotation of the wheel to the right (due to the imbalance).
In that first state the inner pendulum is still vertical so long I don't remove this locker.
Immediately after releasing of this locker you see the inner pendulum falling down, and its opposite side (where is attached the spring) lifts up the outer pendulum.
Nothing avoid you to build one similar assembly, in view to confirm or deny my explanations.

[DrWhat]

Donda este el video pour favor?

[path_finder]

DrWhat: many thanks.
The videos are not usefull at all (just use the cut/paste and loop functions, and a turtle will run like a rocket, what demonstrates nothing).
Hereafter just a picture showing the corresponding building (practical application of the altern5.gif animation above).
Build it and check yourself. We have here to much theoricians and so less builders...

[path_finder]

Sometime we forget to improve some old design wich could be a way to the solution.
After review of the last animation on the first page I observed that this idea was not stupid at all.
The suggested design was not perhaps explicit enough, some parts being not represented in the animation.
So far I rebuilt another animation, using the same principle, but this time including all the parts.
It seems to me much more demonstrative than previously.
The springs are in violet.
There are two sets of weights.

The first set (in yellow) is acting in the respect of a very old principle (see my avatar) explained in one of my first posts.
Remember: the main question at that time was 'how to lift-up the weights between 4:00 and 6:00?'.

The second set of weights (in red) solves the problem: arriving at 3:00 the red weight flips the seesaw blue rod, wich itself lifts up the previous yellow weight.
At that time the yellow weight is retained by its black rod within an angle of 30 grades, therefore the couple is greater from the side of the red weight, assuming an effective rotation (instead normally a balanced situation).

The linkage distance being variable we need a spring as link (violet).

Regarding the comment of AB_hammer (many thanks), there is no contention between the weights and the rods if they are in different planes.
This design appearing as serious enough, I will try to build one example (recycling the previous harware of the above shot).

[KAS]

Path-finder,

I have always enjoyed watching your simulations which only proves that you are a very able mobilist.

Your designs are always interesting and thought provoking.

In my opinion however, this particular design will not produce the results that your sim cleverly animates.

The reason for this is that your concept blinds the observer into thinking that a clear overbalance is present; but if you stop the sim at any point, you will see that more weights are present on the ascent than on the descent.

I have fallen foul to this illusion many times over the years. It appears that if a lever moves any distance on the descent side to form an axial gap, that same gap ensures the balance occurs by increasing the mass on the opposite side.

Interesting though this design is, I think it will fail.
I very much hope I am wrong however, because I think we are now in a desperate position with regard to finding a solution to this problem in time to make a difference.

Kas

[path_finder]

Dear KAS,
If you have really the time to control your assumption, i will send to you (on personal e-mail) the sixteen elementar images of the animation where you can calculate the COG position. You will be surprised.
Many thanks for your comment.

[path_finder]

Dear KAS,
You put the trouble in my mind, and I made a small work in view to confirm or deny your assumption.
You know that the calculation of the COG is difficult.
The device of Desaguliers is not sufficient for detecting all tricky positions.
For me the best way is to separate the position where the weight is sticked to the frame from those where the weight is in motion around a pivot.
The result of my calculation for the red weights is in the drawing hereafter.
It is based on the following facts:
- between the points C and A the weight is leaned on the pin, therefore the COG applies at the pin position (the half inner circle).
- at the point A the weight abandons the pin and get free (it is the small horizontal segment)
- then under the gravity the free weight rotates (the small half circle between the points A and B)
- then on first approximation the weight is in the keeling position, hung at the pivot by the blue rod (half circle between the points B and C)
On first view there is well an unbalance:
- the half circle between B and C can be removed (symmetry with the 6:00 vertical line)
- the excentricity between A and B is obviously a major contribution

For the yellow weight there is no doubt at all (see the old previous thread on this design).
For sure this calculation do not take care of the dynamic motion of the main wheel, and in fact the curve between A and B could be an ellipse (or a cycloid?) in reality.
Can anyone say if this calculation is correct?

[KAS]

Path-finder,

Thank you for explaining your reasoning.

I too have battled with this in the past, and even built a test wheel the prove the illusion.

I don't know if you have WM2D but attached is a similar system where axial movement of a weighted arm levers another weight to perform a radial moment.

I could not succeed with this through simulation or in reality.

KAS

[WaltzCee]

Animations simply show a sequential movement. They don't show the energy input required to power that movement.

This is an interesting point and also an excellent one. It's been made before yet is always worth repeating.

The sequential movements of most animations lack the energy to cause them from what I've observed.

There is a difference between animated motion and real motion.
.

[DrWhat]

So much quoting....!!!

[KAS]

Now having had more time to study this ingenious design, I have concluded the following:

The yellow weights are following a good OOB path. The dream path I like to call it. As you probably know, it has been discussed many times on this forum, where the radial transference occurs naturally to enable it to take up the outer position at 2 o’clock, but help is required to return it to its inner radius horizontally at 4 o’clock. In the past, all sorts of objects have been used to do this including wheels and ramps etc; all of which caused more friction than was gained by the overbalance.
In path-finder’s design, red weighted levers are employed to achieve this. However IMO, I don’t think it will work for the following reasons:-

1.The red weights.
The red weight will not fall from the vertical position as the animation suggests. When it is vertical, it is at rest. To achieve the required velocity, it will at least need to reach its 1 o’clock position to fall.

2.The Yellow weights.
This means that the wheel will need to rotate a little further. This additional semi-rotation means that you will lose the horizontal path required to return the yellow weight to the inner radius. The path will now be up hill.
I believe there will not be enough energy in the system to achieve this.

3. Time.
If the weights on the red levers are just enough to move their respective yellow weights, the system will not work fast enough. If the red mass is in excess of that required to move their weights, the red weight will be weightless for a brief moment as it falls. In my experience, it is not a good idea to impose a weightless situation on the descent side of the wheel. Ah! I hear you say; what about the impact?
It is true that the impact will cancel out the effect of the weightlessness (Newton’s 3RD Law); but the time gap means that the wheel will counter rotate during this process (see below). This will result in no net gain.
4. The ‘Y’ factor
Whenever a weighted arm moves axially on a wheel, the gap it creates means that the COG moves to the opposite side. A good way of visualising this is to draw an imaginary ‘Y’ tilted to where the gap is at its widest. If the fall gap occurs say, between 1 and 3 o’clock, the tale of the ‘Y’ on the opposite side will point to the 8; and that is where the COG has moved to. This of course means that this point will want to keel at 6 o’clock position (Counter torque).

I am sorry to throw water on your fire but if my views are worth anything, I think you will be wasting your time and money. God knows, I’ve wasted enough of mine finding this out!

I am impressed by the thought behind the design however.

[path_finder]

Dear KAS,

Many thanks for your attention.
Instead the systematic negation, a positive discussion even with criticism and with justified arguments, is the best way for sharing.
May I add some remarks at your pertinent comments:

Point 1:
On the animation the red weight starts to rotate as soon it reaches the vertical position (here where its axle is at 3:00).
Practically if any delay is observed there is a very easy solution: shift clockwise the leaning pin a little bit in view to compensate the delay.
Another point I cannot precise more further the role of the centrifugal force applied on this red weight.

Point 2:
You can adjust the different lengths (rods and string) in view to obtain an effective horizontal travel of the yellow weight at 4:00, also playing on the relative positions of the axles. But if we accept anyway do not follow exactly this theoretical horizontal path, do not forget also that in the same time the coupled red weight has a very excentrical position and this extra torque could be sufficient for compensate the loss due to the yellow weight detour.

Point 3:
I'm not sure the COG will pass to the left side during the semi-circle A-B: the red weight at that time is not fully flying. because the centrifugal force there is a link with it's rotation axle, maintaining the COG on the right side.
In fact the difficulty in the calculation of the COG is coming from the repartition of the forces inside the seesaw made by the couple of weights.
I agree with you this design cannot work slowly.

In any case a building will be the only answer to all these questions.
Hereafter a shot of the device under construction.

[KAS]

I think you're right path-finder,

As doubtful as I am, I think the proof of this one will be in the building, and what an impressive build it is!

I would have to admit; I'd love to see that baby spin.

Kas

[path_finder]

Another phase of the construction (without the links).
The red weights pivots have been relocated in accordance with the point 1 of the last post. Now, as soon the yellow weight arrives at 4:00 the red weight is already in motion. But the most important is the implementation of the 'curved planks' with the purpose to lift-up the yellow weights like a lever, allowing the famous ration 4:1 and giving a much more smooth action. A direct rod between the couple of parts seems to be not efficient enough.
The violet crosses on the shot are the posts limiting the course of the planks, wich by the way follow the main circle of the outer rim. This is intended do not disturb the keeling position of the falling yellow weights on the right side.
Now the question is: what shall be the dimension/ratio/location of the links and in particular the force of the springs.
N.B. when all the parts are set at their correct position the wheel is very instable and seems to rotate counterclockwise. Strange?

[path_finder]

Two shots of the building showing the parts location on the both wheel plates.
Note the relative path of the pendulum arm and the two curved planks (like the jacob ladder in MT138).
The four links (crossbar?) are not installed yet.
This model has only four couples of weights, therefore the 'crebs tail' is not completed.