Zeroing in on Bessler's wheel

[Fletcher]

If you can explain exactly what you have tried then it is possible that Shadow can build it in Algodoo - then he can analyze why it does not continue rotation and perhaps someone here can suggest a workaround to solve the issue ( if it is a solvable problem ) ..

FWIW - what we all try to do is invent a certain combination of mechanics that has more torque on one side of the wheel than the other side - usually, torque ( turning force ), can only be manifested if the system Center of Mass ( COM ) can lose some vertical height ( GPE ) - and then it must be recovered again to repeat the torque imbalance conditions .. if the COM doesn't change then to my way of thinking it is like a Roberval Balance or Ramelli Balance - they don't rotate because they can't lose system COM and then recover it, imo ..

[Fletcher]

... they don't rotate because they can't lose system COM and then recover it, imo ..

Should have written "can't lose system GPE" ( Gravity Potential Energy ) etc ...

[SHADOW]

Bonjour à tous,
Il n'est pas besoin de me demander de publier ici ou ailleurs ce que je partage!
Le concept est a attribuer à Sam, lui seul décide.
Le centre des barres tranversales est toujours aligné, j'ai donc aligné les roulements de barre sur la goupille équipée d'un galet.
Si on sort l'ensemble des barres équipé de leur rouleaux on voit un seul degré de liberté indépendant en rotation.
Autrement dit les barres peuvent se rejoindre.
Quand elles sont remises en place, on a l'impression qu'elles font bloc.
Le principe de fonctionnement est comme un pédalier équipé de sa pédale mais dans ce cas la pédale a 6 branches.

Mon avis si la construction correspond à la demande de Sam, cela illustre mon hypothèse qu'une trajectoire circulaire excentrée ne peut pas convenir.
Ceci ne clos pas le sujet sur la pertinence du concept de Sam!

Hello everyone,
There is no need to ask me to publish here or else what I share!
The concept is to be attributed to Sam, he alone decides.
The center of the tranversal bars is always aligned, so I aligned the bar bearings on the pin equipped with a roller.
If we take out all the bars equipped with their rollers we see a single degree of independent freedom in rotation.
In other words, the bars can join.
When they are put back in place, it feels like they are working together.
The principle of operation is like a pedalboard equipped with its pedal but in this case the pedal has 6 spokes.

My opinion if the construction corresponds to Sam’s request, it illustrates my hypothesis that an eccentric circular trajectory cannot be suitable.
This does not close the topic on the relevance of Sam’s concept!

[SHADOW]

J'ai remplacé les roulements de barres, j'ai mis une goupille pivotante et la goupille passe dans chaque trou de barre.
Il n'y a plus d'erreur possible d'assemblage Algodoo. le résultat reste identique sous Algodoo!

I replaced the bar bearings, put a swivel pin and the pin goes through each bar hole.
There is no more possible Algodoo assembly error. the result remains the same under Algodoo!

[Sam Peppiatt]

Hi SHADOW,
I concur. The rollers seam to have little or no effect on rotation, in either direction. Before they did now they don't. It has to be the crank. With it not on board, not rotating with the wheel, it is effectively grounded. Where by, all forces cancel each other out.

It's a little hard to understand but, that's the way of it--------------Sam

It's the crank all right, if I roll them to one side with an onboard string,(without the crank), it works just fine.

[Robinhood46]

I made a video of the previous one, but i can't put it on Youtube because it's raining and my internet is crap.I should have clear skies this afternoon.
The new version does make it easier to see the independence of each crossbar, but it shows the movement i thought you were looking for isn't happening.
This obviously means i haven't understood what it is you are hoping to achieve.
The last version needs the "contacts" set to "pas d'auto contact", on the crossbars, they are the reason they all vibrate.

I'll make a video of the most recent version this afternoon, weather permitting.

[SHADOW]

Bien vue Robinhood, on reconnait l'expert!

Well seen Robinhood, we recognize the expert!

[Sam Peppiatt]

Robinhood,
The crank doesn't work, I thought it would but it doesn't. For that reason there isn't a real need for a video. No use making one if it doesn't work-------------------------Sam

[Robinhood46]

OK.

[Sam Peppiatt]

Fletcher,
I missed your earlier post some how. I think the problem is or was, with the crank on the center line / not rotating / grounded; it, the crank, introduces a back torque or reverse torque equal to the forward torque of the rollers. And they all balance out. There wasn't any thing wrong with the rollers then selves.

Maybe you are right it's like, or at least similar, to the Roberval balance. If there is any way to shift the rollers, it has to be done on board of the wheel / rotating with it.

Not sure what can be done with it, if any thing---------------------Sam

[Fletcher]

IMO Sam .. here's how my own experiences colour my thinking and analysis ..

The rollers were never the problem - they do exactly what they are designed to do - inside their rings they seek and find the position of lowest GPE - this results in the System COM being directly below the wheel axle, with zero torque i.e. no ability to turn the wheel in either direction ..

To turn the wheel underneath the rollers the rollers must be displaced to one side or other - because the rollers are housed inside of rings ( circles ) to move them sideways means they must be lifted upwards a small amount ( working against the ring [ wheel inertia ] ( a small amount of work/energy must be put in ) ) when also shifted sideways .. then they can attempt to again roll downhill and find the System position of lowest GPE beneath the axle - and the losing height transition will also turn the wheel beneath the rollers ..

So imo there is a need to eliminate some roller shifting techniques to see if the above assumptions are indeed true or not ..

The simplest testing method is just the weighted string, or spring, I talked about earlier - these will have some feedback into the wheel ( like the non-working pivoted pendulum driver ) .. if they do not work to reinforce the feedback loop so the imbalance grows, instead of dies, then some other workaround is required for this design to move closer to being a "runner" ..

Roberval and Ramelli Balances don't work because altho there is an obvious looking physical imbalance going on, opposing weights ( in any linear position from center ) both gain and lose the exact same vertical height - NET zero GPE change - i.e. no torque !

** Whatever shifting device you ultimately employ to side-shift and lift rollers it would have to build/gain the feedback loop to grow the imbalance, IMO .. iow's the energy cost of work/energy input into the driver/perturber has to be less than the wheel output Rotational KE, for the two systems to work together and grow the imbalance, and gain, and accelerate the wheel ..

ETA .. fwiw - the fixed pivot pendulum & crank does cause an oscillation of both pendulum and wheel if you position it initially to do so - but it quickly dies down after release until both are at the system position of least PE possible ..

[SHADOW]

Bonjour Sam, j'ai essayé le principe avec une manivelle indépendante pour chaque couple de rouleaux.
cela ne change rien, la roue s'arrête!

Hello Sam, I tried the principle with an independent crank for each couple of rollers.
it doesn’t matter, the wheel stops!

[Sam Peppiatt]

Hi Fletcher, SHADOW,
FWEIW, it, the wheel, really likes these rollers. I installed a one way clutch / bearing, (that the crank rotates on), on the axis of the wheel. Rotating it, the crank, displaces the rollers to one side, (the clutch holds it temporarily), and the wheel takes off like a shot, briefly. Then I have to do it again.

SHADOW, I probably shouldn't ask, but could you put a big pendulum directly onto the crank shaft, with the 'throw' of it, rotated by some amount to one side or the other? To shift the rollers; they don't have to move very far. If it would work, it would hold them to one side continuously.

I have sworn off pendulums but, it was the only thing I could think of to try-------------Sam

[SHADOW]

Bonjour Sam,
Le poste précedent avait 3 manivelles équipées d'un levier contre poids pour le décalage des rouleaux, ce qui correspond à un pendule pour chaque manivelle.
j'ai testé ce qui suit sous Algodoo, cela doit être validé vue la fiabilité du logiciel!

Hi Sam,
The previous station had 3 cranks equipped with a lever against weight for the shift of the rollers, which corresponds to a pendulum for each crank.
I tested the following under Algodoo, it must be validated given the reliability of the software!

[Sam Peppiatt]

SHADOW,
Great! Thanks, thanks for giving it a shot. All you can do is try things. The rollers produce constant torque, no lifting or resetting. So, it's worth the effort to try and find a way to shift them to one side or the other-------------------Sam