Gravity wheels with a fundamental difference

[Georg Künstler]

Hi Robinhood46,
you made a great progress with simulating models with the simulation program.

But as you know my approach is different to yours,
I prefer like a walking mechanism in a Hamster cage.
So my internal construction is standing in the wheel, it can freely fall forward.
The free fall is limited, because the rim of hamster cage is the limit.
A collapsing window frame is an moveable structure construction.
The collapsing window frame is making a combination of a move, it is moving downwards and sideways in one go.

What we need for the indirect impact is the sideways move, the horizontal move of the upper part of the collapsing window.
The collapsing window frame is a carrier of pendulums. This equates Besslers words, the children are playing on the pillars.

My test wheel, the Hamster cage has an diameter of 50 cm and the collapsing window frame has 32 cm.
Every side of the window frame is carrying 2 pendulums. So in total I have 8 pendulums.
The suspension point of the pendulums is 11 cm from each side of the frame.
The pendulum length is 5 cm. Every pendulum has about 50 gr.

For understanding I have sketched it up with 4 pendulums so you can detect the movement.

[Sam Peppiatt]

George,
Questions: Does the window rotate? Is it part of the Storks bill? The big question for me is, how do you drive the storks bill-------------Sam
PS Dam! Wish I could draw like you do.

[Robinhood46]

I have started a sim to be sure we are on the same page.
I can't see how you hope to get the frame "over" early enough to have it in the position in your sketch. Is the sketch for anti clockwise rotation? I don't think it is, but i would like to be sure. I added some springs which gained a few degrees, but it is till a long way short.
There are 25 points on the rim, and each section of the frame steps 5. Changing the numbers will move the swinging top part nearer and further from the central axis of the "hamster cage". Are there a different number of points around the rim you prefer?
https://youtu.be/-Lf9NdGJ9Pg

[Tarsier79]

RH, Your sim is what I would expect to happen. I think he is hoping the swinging pendulums will help. This lag will always occur with a symmetrical design.

[WaltzCee]

Latency is a huge issue compounded by the loss of GPE.

Da Vinci wheels used that same combination.

. .. .. .
The collapsing window frame is making a combination of a move, it is moving downwards and sideways in one go.
. .. .. .

[Georg Künstler]

Hi Robinhood46,
in my test wheel I have 80 dowels on the rim, but for the simulation and the movement 25 is also good.
The dimension of the internal window frame is not big enough.
If you enlarge the window frame size, the impact point on the rim is also on an higher position.
So far I see, that you are on the same page.
You will see in the simulation that we will get an impact on the down going side.
Now add for the testing 4 Pendulums on a rope 1 on each side of the window frame.
Variate the pendulum length and you will see the upswing of the pendulum.
Every impact on the rim of the window frame will produce an simultaneously upswing of the pendulums.

[Robinhood46]

Georg,
I put 80 points around the rim (your dowels).
There are only three possibilities, in my opinion for the configuration of the frame. Any bigger and there is no lateral movement at all, and any smaller it becomes too low. I don't think the third is likely, myself, but i am happy to have a go with whichever you prefer.
Next step, I'll try with different rope lengths and even solid rods for the swinging weights.
https://youtu.be/CRAe-iWSEEE

[Robinhood46]

The swinging of the weights can't get the window frame to shift over, at the appropriate time. But the swing of the weights in this manner is interesting.
Here the weights are swinging somewhat similar, but on a finger spinner type set up. As the weight swings to its lowest point, it applies a pull to the arm and swings back up, to then apply it's weight to the wheel/frame at a distance further from the centre than on the ascending side. Which, if i understand correctly, is what Georg is trying to say, at least something along those lines.
This doesn't have the fundamental difference, but if applied to a curved crossbar, that has only two mechanisms, this could maybe be a way of shifting the top of near vertical over from 10.30 ish to 1.30 ish. Which, i think would be more likely than shifting the top of the window frame.
https://youtu.be/6Ra_mrI5MbE

[Georg Künstler]

Hi Robinhood46,
the relative size from the window frame to the Hamster cage is nearly that what I have in my buildt.
I hope that you have seen that the pendulums are swinging up in a flash during the impact of the window frame.
A swinging up will create also an downforce.
The downforce is applied on the window frame and from there to the wheel.
As I have not seen what you have simulated, I cannot give you more suggestions for improvement.
I think you have experimented with the pendulum length, and there you must have detected that as shorter the length of the rope is, as more will be the upswing.

Another point is the mass of the pendulums.
The pendulums are in the shape of a cylindrical weights.
So we have a concentrated mass point.
The cylindrical weight has a hole, and we can thread a cord through it.
On this cord we can adjust the pendulum length when we connect the cord to the suspension point.
So the swinging mass is relativ high against the carrier itself, the window frame.

[WaltzCee]

and Bessler's is his worked and mine doesn't!!

crap.

. .. .. .
another point is the mass of the pendulums.
The pendulums are in the shape of a cylindrical weights.
So we have a concentrated mass point.
. .. .. .

A bunch of masses circling on conservative circuits adding up to zero or just one mass.

What's the difference?
I'm really struggling with this point.
please help.

[Robinhood46]

I thought i would post this bad attempt at getting the weights to stay on the descending side at all times, which could explain the "one side empty and one side full".
As can be seen in the video, the bloody thing doesn't work. The opposite would have been more of a surprise. I mean function correctly, not be a runner.
When i test run with only one "section", the ratio 1;4 works fine, and if i move manually the heavy weight the wheel shows over unity.
Wishful thinking is probably playing a big role, but it does look promising and i would have liked to see the thing function correctly. When i displace the weight manually, i try to be as close as poss to real life, but this is no easy task. When i multiply everything by three, there are obviously 3 heavy weights, which looks very good, but the raising lighter weights accumulate, which is not so good. 9 knocks per revolution.
I can really get the heavy weight to fly up, it is more like thrown up than lifted. In the video there is a spring fixed to the centre, which is pushing to try and hold the weight up against the stop, just for the purpose of sharing my progress.
https://youtu.be/gW7_mp4FjH4

[Robinhood46]

Well, I'm actually getting nearer to achieving the sort of movement that I've been trying very hard to do for a while now.
My thoughts are that one exterior light weight should be able to action the hammer-men who pass the hammer head from one to the other (the heavy weight). this would allow the swinging back of the lighter weights to take it in turns nudging the heavy weight laterally, to get it away from the centre at a higher position. I think if multiplied, this would give the peacock feathers opening and closing. I do like actually having a single crossbar that is able to complete a full cycle, which can be multiplied at will, which is not something most of my attempts have had.
The linking rod passes through holes in the central axle. Adjustable bolts would be needed to align the arms correctly in both positions. The R's of his logo can be seen also.
This is my first attempt, that actually functions, so i thought I'd share it, even if there is a lot that needs doing to get it to work, if it will work.
https://youtu.be/Z4Kdd0vBDzs

[Tarsier79]

I am not familiar with Algodoo. Does it have a function like WM2d Where you can view the C.O.M? (Centre of Mass)

Even though this looks inviting, lifting the large weight from 7, then dropping it to around 4:30, the small weights are probably making a bigger impact than you know.

[thx4]

@robinhood46,
You start the year very strong, it's true a lot of small obscure points, but what I really like is the mobility, with the sphere. it's a very interesting way.
The best for the future.

A++

[Robinhood46]

Merci thx4, à toi aussi.

Tarsier, The weights that are larger, are lighter than the central moving weight.
0.25 kg central and 0.13 for each of the exterior. The weights of the arms also comes into play. The black "heavy weight" is small and dense, to get it in near to the central axle and reduce the negative effects of the mechanisms. Passing a proportionately larger weight, is even more of a pain in the butt than a small one.
I only slapped the weights on, at the exterior, so as to have a demonstration of it functioning. There are many things to play around with, and this can only be done, once the desired effect functions, even if it does do it very badly. As i often say "it's early days".
It may be possible to have two heavy weights, connected one each end of a small rod, so as to move one horizontally from the centre to 3 ish as the other is raised from 6-7 ish to the centre. This, if possible, would reduce the need for the heavy weights to be, as heavy, because of the improved path taken by each weight when it takes it's turn to apply force to the wheel.
The falling back of the lighter weight, on the ascending side, i think, needs to be avoided, and just having it swing back lightly at the bottom is preferred. This means increasing the fall of the lighter weight on the descending side, which hopefully, shouldn't be too much of a problem. Some of my other attempts do this just fine. Getting each lighter weight to cause the movement of both hammer men, independent of the other lighter weight, would be relatively simple if i could get ropes to work, unfortunately my computor doesn't want none of it.
Wm2d would probably be better than Algodoo for many reasons, particularly the ability to turn off and on some functions at different places. Algodoo does give some information with regard the forces in play, but i can't get graphs like Wm2d, but it is possible that it's me who is completely useless and not the programme itself. I should be able to share scenes, so that anyone, who does know what they are doing, can share their observations and understanding of everything that is going on.