Excess Torque Hypothesis : Mechanical Sustainable Imbalance Possibilities ( <>> ) ?

[eccentrically1]

Looks good Fletcher. I don't have wm2d. My question is does the program automatically calculate the pivot, joint, chain, etc., frictions and account for those in the output? Or does the user have to do those calculations and subtract them manually? I can't tell from the screen shots which is the case. Those are the real energy sappers imo.

Air resistance probably doesn't slow it down that much, if you've tried it with it turned off to see the difference. I watched a YouTube video of a newton's cradle in and out of a vacuum and the running time was only a second or two longer in the vacuum. The reason it stops is mostly from the string pivot friction and the impact losses.

[Fletcher]

Actually I build sims initially with Air Frictions OFF. Then I add them in if something looks interesting so there are more realistic system wide energy losses. They are a good quick and dirty way to account for many types of internal and external frictions that otherwise would have to be added separately. The thing about Air Friction is that it is a non-conservative force (drag) which is exponential i.e. goes up by the square of v. And you can change the multiplier any way you want, triple or quadruple the pre-set high value for example to really load the sim (e.g. to simulate a Working Load). At the High setting and a decent RPM it's like swimming thru a viscous fluid. So sims top-out quickly because of the v^2 relationship and the coefficient of drag selected. I'll run the sim with AF's OFF (as when I built it) and I'd expect a small improvement in RPM and System KE at the same time stamp simply because those energy losses are not occurring (it takes me about 3 hours to run 10 seconds of sim atm). Other losses are occurring tho. I'll edit this post when I have the result. ETA : trending to a slightly higher RPM curve (12 up to 14) at the same time; is similar so will stop the sim.

As you comment and I said in an earlier post anytime objects move within a wheel they have to be accelerated and decelerated. The deceleration usually involves an impact or impulse situation. And usually that KE of transition is wasted as material deformation, heat, and sound energy losses. Or pivot frictions etc. In the sim all the rlw and chain link objects have elasticity set to 0.5, and static and dynamic frictions set to 0.3 (usual settings for those simulated materials - range is 0.00 to 1.00 with steel being 0.95 elasticity for example). So there are already energy losses built into those objects that have contact and the gross AF's is on top of that to catch anything else. That means pivots and slots are frictionless and covered-off in the High setting for AF's.

The upshot is that the sim is set up for continuous imbalance - this is achieved by the addition of counter-torque (torque matching). Then the Chain provides the system asymmetric torque. The rlws can be as massive as you want, which means more KE and Momentum for a given RPM. We can add many more ganged arms. I used 8 coz I'm keeping the complexity as low as feasible for testing purposes. I would say it would be better performance wise to have as many as you could stuff in there - say up to 24 (42) but that requires a really large diameter wheel. The Chain path would then be very stable.

[ME]

I'm sticking with the dual axle setup for now but ultimately we'd want to migrate to a single axle setup if successful. At least to be similar to B's wheels, tho its entirely unnecessary.

I simplified the system to MT009 (only to make it complex again).
It's still able to pull a weight upwards at 12- or 1-oçlock by the combined pull from 12- till 7-o'clock.

So I tried to put the Leverweights (LWs) on one single axle, and then synchronize them by means of a scissor mechanism.
Attached is a possible setup. The red weights are for the anti-clockwise direction, the orange ones for clockwise direction.
One red is connected via a half-scissor to the opposite orange by a pivot on rails.
With all the connections it looks a bit messy (design-state a), and it's looks worse when we let it go (relaxed-state b).

Next post: an animation.

[ME]

Here's the animation. For the first half second it was sped up with a motor, then it freewheeled without air-friction.
I actually didn't anticipate that the pivots would go that far out.. those who are familiar with WM2D understand that they are still on those center rails :-)
We can observe that the upstroke of the red weights comes a bit early because it's coupled to the orange weight at the opposite side.
Lucky the outer-left and the outer-right are the only important ones for a hanging chain. That is still offset to the left.

Meanwhile, the computer is extremely busy with a chain-implementation.

[Fletcher]

Nice one ME .. even I am having trouble following your animation lol.

Is it possible to put up a simple picture to show the connections and mechanics etc in a simplified schematic form ? Even just some rod connections black ? Perhaps with the system CoM ON ?

I was going to ask you today what you were trying to show with the sim - torque neutralizing I think whilst still having useful movement for a chain ?

And was there a chain to be involved ?

I guess I'll wait for your computer to finish being busy. I know how that is ;7)

[ME]

Nice one ME .. even I am having trouble following your animation lol.

At least it gives something to meditate on.

Is it possible to put up a simple picture to show the connections and mechanics etc in a simplified schematic form ? Even just some rod connections black ? Perhaps with the system CoM ON ?

It's is simply MT009 in red weights, nothing really special.
The orange weights are also MT009, but they go in reverse.
Because I want them to behave symmetrical, I connected the Red with the Orange in the simplest way I could come up with.

To follow the connection- for example - see as designed in [CounterM009-a].
The Green pivot at 9-o'clock has a Red LW pointing upwards.
The Green pivot at 3-oçlock has an Orange LW pointing upwards.
They both connect to a vertical sliding pivot at 12-o'clock (at the center circle).
The others LW's do the same, but 30° rotated.

When I let them react to gravity it settles like [CounterM009-b] - a total mess and hard to follow.
A new attachment [ME_CounterMT009_b_Lines] has enhanced lines - black for one red-orange pair from 9 to 3, and blue for 6 to 12.
While I'm at it anyway, also attached [ME_CounterMT009_c_blurred-path] a blurred result of the animation to show the effect (perhaps) a bit better. The chain will hang on the horizontal extremes of the red-path.

For those who ponder about these things, we could clue-glue by thinking that this pair now moves like a crab. and if you really want you can almost see MT137 in design image [CounterM009-a].
...

[ME]

I was going to ask you today what you were trying to show with the sim - torque neutralizing I think whilst still having useful movement for a chain ?

The path of the Center of Mass: [ME_CounterMT009_c_CoM]
The Center of Mass starts balanced with the Red on the left-side, and the Orange on the right.
I find it interesting that when I speed it up for half a second CoM of the Orange LW's rise. Actually I find the path of the Orange CoM to be rotated about 135° compared to the path of the Red CoM.
As expected, the combined CoM is on the right side: Only averaging to about 0.0025m for a shown wheel radius of 0.55 m. About 1/6th inch for wheel with a diameter of 3.6 feet - IOW, not much.

For the fun of it I can estimate the slowdown: When the weights go from an outer radial distance of 0.5m to an inner 0.35m then say it averages on 0.43m.
When I neglect the MoI of the wheel then the deceleration will be about 0.0025 · 9.81 / 0.43² = 0.133 rad/s² about 7.6 deg/s², which should be about that downwards slope as shown in the earlier animation.

... now we have to figure out how much a counter-MT009 actually saves in lost torque.

[ME]

And was there a chain to be involved ?

Nope.

I guess I'll wait for your computer to finish being busy. I know how that is ;7)

And then I wanted to move controls around for a better animation. And then it crashed. And now it's calculating the same thing again :-)

NB. These kinds of simulations can take a long time. In that case a build could predict the outcome of the simulation.

Still valid I guess ;7

[Fletcher]

Thanks ME .. that's clearer. And yep .. it's tricky stuff to work out. And yes, there are always going to be energy losses to ANY system from transitioning levers (repositioning) etc as they accelerate and decelerate - unavoidable and usual. All false principles that rely on internal repositioning fail at that very first hurdle of energy losses, without even considering anything else such as full GPE recovery and friction energy losses. Yet for B's. new approach REPOSITIONING ENERGY LOSSES WAS NOT A SHOW STOPPER ! There was a Preponderance of Imbalance Torque to overcome ALL energy losses and fully restore GPE, imo.

And a build might be quicker but then B. didn't have a sim package like some of us do so had little choice but to experiment thru builds, and gain insight. He set to work and it rotated a little. Then he improved it (N.B. the word spontaneously was added by the translator).

"For I put together the very first device which could spontaneously revolve a little. I saw that I had finally made the right choice, and why the earlier ones had been wrong. My heart leapt for joy at the sight of this genuine Mobile." - DT pg 271

"design has, in fact, progressed to the point where there is nothing supercritical about the exact disposition of the weights - an ounce more or less, here or there, makes not a scrap of difference to the Wheel, which will hold its course serenely without 'turning a hair'." AP pg 308

What he had to have was a new thought on how to solve the problem of sustained imbalance (a True Mechanical Perpetual Motion Principle), from the usual false and done to death principles. He looked at the problem from a different vantage point, imo.

While your sim tests your patience (good on you for trying) I'll probably tomorrow run thru the MT11 cue of simple doubling the systems and show those interested the inherent problems this approach has to nulling torque that you and I discussed earlier in the thread. The good, the bad, and the ugly.

But B. might have given us a pointer which limits the options by half if everything he says is to be completely believed.

[Fletcher]

So far I've attempted to show that a MT9 (hung together) like rlws rig like my single secondary (carrier) wheel will be unbalanced. It has a tendency to rotate in the opposite direction to the direction the rlws open and "fall".

I have not found any better arrangement than the Jack system using the Gaffles and Pulleys and Ropes to generate a significant lateral force to potentially use. A normal MT9 arrangement of in-series rope connected weights on levers appears much weaker and to have much less usable lateral force generating capability than the former.

When I add a looped Chain around the Carrier Wheel rlws the Jack fully deploys and creates a force, able to lift and move the Chain sideways on the side of the opening and falling rlws Jack. This changes the CoM position of the Chain giving it a tendency to rotate the same direction as the Jack operates. So the Chains positive CoM displacement partially offsets the Carrier Wheels CoM negative displacement. Ratios can be manipulated to make the two almost balanced. Almost but not quite complete balance at all times. The addition of the Chain alone can not make the combined CoM positive in the direction of the Jack !

This means that the Carrier Wheel plus Chain Drive has a tendency to rotate in the opposite direction of the Jack deployment. And it needs to rotate in the same direction as the opening and falling Jack rlws so that the next rlw can opening on cue and thus shift the Chain etc etc, perpetuating the actions and an Excess Torque.

What is missing is a third component that generates a further Torque to the dual arrangement. And its Torque ultimately creates a System Wide Net Positive Torque in the direction required for the Jack to operate effectively.

The sum of the Torques can be viewed in two main ways.

A near or partially balanced secondary Carrier Wheel plus Chain, plus a second secondary wheel.

Or two secondary wheels with equal or near equal opposite Torques so that when combined they are near to a perfectly balanced condition (the flywheel analogue). Then the Chain Driver is added to give a System Wide Net Positive Torque to perpetuate motion and asymmetric Torque. Thus the Chain Driver is in this sense the Prime Mover altho the Jack does the work of first displacement.

That is the Hypothesis behind my Theory of a True Mechanical Perpetual Motion Principle.


If we accept that the addition of the Chain Drive does indeed provide continuous Positive Torque then we can dispense with it for the time being and concentrate on other matters. It can be added back later. It's temporary sidelining should make simming of the optimal secondary systems a lot easier without it crashing the sims due to complexity issues.


Next I will look at doubling the secondary systems (in various forms) and what effects this can have in our quest for secondary system "balance".

[FunWithGravity2]

Have you tried adding a motor and watching reactions at different speeds or is the system to complicated to run reliably?

Crazy Dave

[Fletcher]

Yes Dave .. I can add a motor. But instead I've added a massless force to turn the RHS Carrier Wheel CW. The force is an arbitrary amount of 20 N so that the sim moves along.

I've refocused the problem of a single wheel with lws etc repositioning under gravity, and pulling and pushing, or raising either another or something else internally, to supposedly cause constant Asymmetric Torque conditions. As we know with these closed path regimes the Torques are always Symmetric and this defeats us and it finds balance and rest.

I've created imbalance by using a Secondary Carrier Wheel with ganged rlws (the Jack) holding out a Primary Chain. This combination is near to balanced. Then I theorize that adding a second Secondary Wheel with opposite Torques to the first will near to Null Torque the Secondary Dual System. Thus the Chain Drive Primary will provide the overbalancing element for continued Asymmetric Torque conditions.

This hypothetical near balanced Dual Secondary System will then act somewhat like a flywheel, however its CoM may move up and down. That has to be factored in the permutations in case the Chain Torque can not overcome it in some circumstances.

See below the templates I created that use Dual Wheels - one is a Mirrored System and the other is a Same Orientation System. Both use the MT9 analogues I've been working with here. Sims included. Readers without WM will have to imagine and interpolate what the pics suggest the old fashioned way. The sims can be altered at will as they have inputs and two gearing systems, one of which must be selected, and which determines which way the wheels turn etc. N.B. one wheel always has its CoM to the side with the rlws hanging below.

In some instances the Torques are Doubled. In others the Torques are near Equalized. The problem with using a secondary functionally identical Secondary Wheel is getting the up-going rlw to "fall' back into the rim ! So imo another alternative for the second Wheel is most likely required.

If B. can always be believed then he perhaps gives us a clue when he says ...

"In a true Perpetuum Mobile everything must, necessarily, go round together. There can be nothing involved in it which remains stationary on the axle." - AP pg 361

"by making the true claim - that no weights hang from the axle of my wheel." - AP pg 281

Using an Artificial Horizon and Gears/Cogs opens up lots of opportunities for reverse direction Secondary Wheels.

However if what B. says is true (of his wheels) then this theory requires that one Secondary Wheel be mounted inside an outer. And they both turn in the same direction on the same axle.

That limits our options to half available, if following his directive. As per MT11 suggests imo.

[FunWithGravity2]

Hey Fletcher

What i was thinking with adding a motor was so that you can begin to look at what i believe to be the underlying item of most importance with the wheel. Timing, at what RPM do things become futile. I know that we are not trying in some case to replicate B's wheels but like you i do believe that many things can be taken at face value and some others we can set as fairly concrete. I have always tried to replicate size and speed because those items we can assume from the witness statements are non negotiable(closely). I have always taken any design i had and forced it to rotate at a given speed to determine its reaction in a rotating reference frame.

In a design such as the one you have shared with us all things are affected by at least two forces (cringing as i say force) that we can say should account for the bulk of what we hope to manipulate. Ignoring heat and friction we must know how the acceleration of gravity and CF react in our rotating reference frame. Since all must go around with the wheel the time that each component is impacted by these forces either in opposition or harmonically becomes very important. The wheel is IMHO driven by the most basic gravitational frequency that we don't seem to understand yet, and that is TIME, or our concept of it.

Sorry for the ramble.

One day hopefully all my rambling will make sense.

Crazy Dave

[daxwc]

Hi Fletcher, interesting idea. I still don't understand what the inner red string/chain does though. Sorry if you have addressed the question already; I've been busy and don't have time to wade through it all.

[Fletcher]

Hi dax .. the inner red ropes are part of a pulley system between lws. There are 8 pulley systems in total, each linked to the next in series.

The bottom quarter rlws thru the pulley system and Gaffles (SB's) apply a force to the one above when the pulley rope is taut. Each rlw opens an incremental distance from the proceeding.

And so their combined effort forms a Jack !

See pic below.

1. Red arrows point to some Red Pulley Ropes.

2. Black circles indicate where one Pulley System joins the next in series.

3. The Purple circles indicate the actual Pulleys.