This is my wheel concept 2017, not Bessler's 1717...

[Sam Peppiatt]

Hi raj,

Please forgive me, I haven't been following too closely, (no good reason).

Your October 21st post was the easiest for me to understand. Where you show 12 position of the weight. I think position 10 is the key to it. With the hinge arm to the left and the spring to the right.

This is where the weight would fall the farthest, the hinge arm would tend to pull the wheel CCW, while the spring would be pulling CW. I think this is the farthest it would go. It would stop at 10 and not continue on to 11 and 12. At this position it would form a toggle and I think would drop down quit a bit, do to the MA of a toggle.

I don't mean to say I'm right, but I think that's what would happen. You just have to try it and see. Please take care, Sam

[Fletcher]

For all those who have followed the full presentation on this page, this is the first PROOF OF CONCEPT of my AUTO WHEEL invention.

The 12 weights reset their starting positions every 30 degrees rotation of drum wheel.

This video with David M explanation above, shows drum wheel rotating over 45 degrees. They prove that there is net torque well past the resetting positions of the weights, most sufficient to overcome frictional loss and air resistance.

More than convincing to go for a full blown working prototype build, starting New Year!!!

https://www.youtube.com/watch?v=50s1Ur5XVnc

Raj

Seasons greetings to you Raj .. having computer and internet connection problems but a window is open this morning it appears, so I'll comment on your presentation so far.

I think you will find that many are reading and following your thread here with interest. Mainly because you have taken steps to independently verify what you believe will happen with the shifting lever weights in your AutoWheel. And you now believe there is enough positive support from those same people to start a real world build. Good luck with your build as always.

I make the point that IMO not many will give you an opinion or further input on your AutoWheel design because you appear to not take any comment that is not supportive of your beliefs very well. Silver Tiger is an example where he actually built a WM2D simulation of your design (with or without ropes).

For my part I will offer only the following, given that you are preparing to build such an AutoWheel. Firstly as I understand it there are 12 lever weights even spaced around the rim (inward facing), with the weight located at the CoR/axle and thus the lever length is the radius. Attached to the centered weight is a spring with attachment point diametrically opposite to the lever pivot (say 6 for lever pivot and 12 o'cl for spring attachment to rim). 90 degrees to this is the rope attachment at the rim, say at 3 o'cl, and it also connects to the center weight and its length is also that of the radius. So far so good I think.

All springs have the same length and the same spring constant (n.b. springs are considered conservative).

You have David M give the sim he is building a small push start, just to get it moving in one direction. As you would do for a real world build where the wheel is at its lowest PE position.

Without building this description of your AutoWheel I can see that it will turn, for a while IMO. I think I can see where David M gets his 45 degrees of positive torque, even tho each sector is 30 degrees (for 12 mechs). I think that is because there appears only a 90 degree sector where the lever weight can shift position and is not constrained by the limiting rope. That is when the lever is horizontal and then vertical 90 degrees later. Thus 45 degrees of positive torque and I predict 45 degrees of equal negative torque (+45 + -45 = 0 surplus torque). That will be because the lever weights when not constrained are able to move downwards. When the lever is horizontal the weight will move down and away from the vertical line below the axle, creating positive torque. When the lever is past a further 45 degrees the weight will be on the opposite side of that same vertical line creating negative torque, as the spring tries to restore its original position, IMO.

IF this is what you find when David completes his simulation, and if the physics profs don't give any other suggestions, then you might consider changing the ratios, pivot points, and/or having one falling lever weight pull on the spring of a forward positioned one needing restitution for example. I'm sure you looked at these alternatives already.

But I put the cart before the horse. Let's see what David M delivers for you and what his AutoWheel sim shows.

As usual if I my understanding of your AutoWheel design and its likely actions is flawed then take my predictions of its likely behaviour with a grain of salt. No offence intended.

[raj]

Thanks a lot, Fletcher.

I am not near anywhere I can claim I have a working wheel.

My gut's instinct, based on what I have seen for the past three months, is forcing me to believe that I MAY be on to something.

I have learnt a lot more about my Auto Wheel concept in the last few days, that is reinforcing my belief in the concept.

Raj

[Fletcher]

Then perhaps this little pic representation of a Single Mech will help the understanding some more for you, or at least others watching this thread.

I have taken a single mech and rotated the wheel in 45 degree increments. So there are 8 representations reading left to right and top to bottom. I started the first one with a 1 degree offset CCW and they all follow suit. That was to show visually when the lever can fall and lose GPE.

The lever is a Green massless rod, the weight a circular Yellow mass of 1 kg, the massless restitution spring is Red, and the massless limiting flexible rope is blue.

Included for ease of visualization is a Grey Arc to show the potential lever-weight path. You can reference that to an imaginary vertical line below the axle and see which side of this line the path goes and what effect it would have on the CoM and hence Torque.

FYI .. 5 thru 8 can move downwards and upwards again along the Arc path. This movement of the lever-weight is not shown in this representational pic.

It appears in this representation that because the spring is conservative and linear that it can not lift the lever weight early enough (just a guess till you build it to find out). This is why I suggested that perhaps a falling lever-weight (losing GPE) could be linked to a forward spring element via a pulley system etc to cause an lift to center earlier than usual. It might also come to nothing but probably worth a think about if trying to find asymmetric torque conditions. Just a suggestion if you run out of ideas or options.

So let's see what David M's sim predicts eh ?

[Fletcher]

FWIW - I think this is important !

Often over the years, and I have thought and said so myself, it has been said that adding additional mechs will not enhance or augment the result favourably. That you are better to test the action of just one mech (I think Chris HM's was the latest to mention this). That if one mech can't enable full restitution (aka Stevin's Principle) i.e. refund in full plus some, then multiple mechs won't change that unfortunate fact.

But if you could find a one mech system that could cause full restitution, then some (to cover system energy losses and do Work), then there would be benefit in adding additional mechs internally to multiply the effect, or add additional mechs to the same axle etc.

But .. no-one has managed to find a single mech combination that can fully restore GPE each full cycle, and have residual KE of velocity (aka momentum) that can be bled off to do Work. And that is easily explained by scientists who say that gravity force is conservative etc etc. IOW's that a weights path downwards and upwards again makes no difference to the net Torque sums and there can be no net gain.

BUT .. Bessler said early in JC's MT (MT's 9 and the very next 10 in fact) that "nothing is to be accomplished with his thing unless one acts out of my connectedness principle" .. "and indicate the correct handle-construction".

So IMO Bessler is telling us unequivocally that a Single Mech system has ZERO chance of replenishing GPE and having residual KE/Momentum. IOW's, it can't work ON IT'S OWN ! Just as scientists say and the math predicts.

It MUST be connected to another mech in some manner to bring forth a favourable result i.e. Bessler's PM Principle is dependent on one mech influencing at least one other, IMO.

Perhaps this resonates loudly for some reading this thread.

[raj]

Hello Fletcher,
Thanks again for your help. Your drawings are accurate, showing the potential paths of the levers/weights.

But your assumption that the weights won't be lifted in time, is wrong.

The first secret in my Auto Wheel concept, if any, is in the equilibrium LENGTH (L units) of the springs and the EXTENDED equilibrium LENGTH (L+E units) of the spring due to the hanging mass of the weight.

Use the Extended equilibrium LENGTH (L+E units) of the spring, as RADIUS of the wheel.

Then you will find, that LIFTING of the weights can be done in time.

The second secret in my Auto wheel design is that I am using THREE (3) CROSS(ES) to hang my levers, springs, cords and weights.

While you are at it, why don't you do a proper simulation for me.
I shall be ever grateful to you.

Raj

[Fletcher]

1. The first secret in my Auto Wheel concept, if any, is in the equilibrium LENGTH (L units) of the springs and the EXTENDED equilibrium LENGTH (L+E units) of the spring due to the hanging mass of the weight.

Use the Extended equilibrium LENGTH (L+E units) of the spring, as RADIUS of the wheel. Then you will find, that LIFTING of the weights can be done in time.

You've got me Raj. I just don't understand what you are saying in these two quotes from you.

The springs (in Red) I used are placed between the centered weight and the rim at radius 'r'. They already have length 'r' and this is their resting length, or natural length, where they do NOT apply any extension or compression forces. And they have a spring constant value 'K' of in this case 50 N/m. This determines the power of the spring when they do contract or expand in length. BTW, these are 'perfect' spring elements with NO losses whatsoever, which is not realistic but good for ease of initial simming. The length of the spring thereafter is determined by the lever-weight mass acting at a certain angle to the lever rim pivot, IINM.

If you pretension the springs by shortening the natural/resting length then the lever-weights will be delayed in moving and loss of GPE. Conversely they will restore the lever-weight earlier as you desire. But because the K value is constant and the length predetermined I believe it will still be a zero sum game.

2. The second secret in my Auto wheel design is that I am using THREE (3) CROSS(ES) to hang my levers, springs, cords and weights.

I looked at your drawings and I just don't see the relevance of the CROSS(ES) ? A wheel feels the Torque solely determined by the weight masses position relative to the vertical line below the axle i.e. individual CW and CCW turning moments.

I could sim just one mech in action when/if we are speaking the same language but that will have to wait till tomorrow. Beer o'clock here.

[raj]

Cheers Fletcher, I am joining you with a pint from here!

I cannot explain with physics or mechanics.

I am using paint drawings to explain. I hope it helps.

Drawing on the left shows a spring of natural equilibrium length ( L units) connected to the rim of a wheel at 12 o'clock position and connected to a weight on its other lower end.

The drawing on the right shows the weight having moved vertically down by the force of gravity pulling it downwards. By so doing, the weight has lost some potential energy, but the spring has extended to its MAXIMUM achievable Equilibrium Length ( L+E) units = radius of wheel but gained some potential energy.

This gained potential energy by the springs would help in asymmetric orbital path of the weights motions and lifting of weights in TIME.

There is a continuous change in potential energy in the springs, depending on their positions that pulls with or against gravity, forcing continuous asymmetric motion of weights, and providing continuous asymmetric torques to the wheel.

This is what I SEE.

I don't know if others see the same.

Raj

[Fletcher]

Hmm .. so we have a vertical spring on the left at its natural length, with no stored elastic potential energy (EPE1 = 0). A weight is pinned to the background and also attached to the lower spring end. Therefore the weight can not change distance to the center. It has GPE1.

Total Energy = EPE1 + GPE1 = 0 + GPE1 = GPE1

In the second drawing on the right the weight is unpinned from the background and so the spring is extended storing EPE2 whilst the weight reduces the distance to the center and its has GPE2.

Total Energy = EPE2 + GPE2 = GPE1

or GPE2 = GPE1 - EPE2

https://en.wikipedia.org/wiki/Hooke%27s_law

http://hyperphysics.phy-astr.gsu.edu/hbase/pespr.html

[Fletcher]

Anyways .. here is the sim of a single mech starting with a small torque push start. There are 3 pics (Start, Mid Point Approx, and End i.e. rpm = 0).

Note the weight leaves its track in the 3rd pic once it starts moving from center. It does not reach original GPE (when wheel rpm = zero), then it reverses direction to CW, until finding position of least Energy (PQ point).

Included is the sim for others to play with. I used a Dampened Spring Element this time with Air Resistance on Low. This site doesn't support avi files so I can't show a video of the wheel rotating etc. Someone with a YouTube account may like to post it for you Raj. You'll just have to follow the screen grabs.

[raj]

@ Fletcher.
A BIG BIG thank you.
From your drawings, it's exactly what I expected.
It seems to have answered a long standing headache question.

Raj

[ME]

Fletcher,
From image "drawings for physics tutor 291117.jpg" I would say that the "K" seems good (perhaps set it to 49.05 N/m), but the 'Length" should be set to 0.3 for a wheel radius of 0.5.
From image "Spring extended equilibrium length...291217.png" I would say the length should be 0.1; so with a mass it drops to about 0.3, and the lever stretch it to 0.5.

In any case the elliptical path of the weight --which remains relative smooth while the wheel goes at half the resonance of mass-lever-spring combi-- points towards the ascending side.

With the reported simulation effort, I just hope for Raj that David-the-Rolls-Roys-Engineer provides his help mostly for the fun of it and not charging by the hour with a company-title-matching hourly rate.
On the other hand, when some energy gain is truly lurking somewhere hidden in this design and waiting for discovery, then with such guy onboard, it should almost be a safe bet this gain will be found... and then charge accordingly :-)

Anyhow, goodluck Raj...

[Fletcher]

OK ME .. I understand now after your explanation that Raj wants the weight to be at center when it's hanging from the bottom of a spring i.e. that the spring is extended and has elastic PE. So not the natural length 'r' that I used. My bad.

His notes of 1. and 2. giving 3. now makes more sense to me. I'll take another look at the single mech sim with those changes - thanx.

I think it's a good idea that Raj employ outside skills to get an independent validation by way of sim. David is using a program that is not WM2D for starters, and presumably reliably used in his industry. There are others around up to the task such as Solid Works Cosmos but nobody talks about using them here. It's good to have comparisons and discussions of sim build details and what the predictions of behaviour are likely to be. Then we all get on the same page, so excuse me if I seem pedantic at times, but it can save a lot of time ultimately.

For my part, I still don't know if WM2D has an override function that makes all sims conform to CoE and CoM ? That is, that it iterates top down (will never show OU) or bottom up starting with brick placed upon brick (possibly will show OU). I always felt that if a path to mechanical OU is there then it should become visible and understandable, once the actions are seen and understood. That may mean throwing out a sim result because it blatantly makes no sense. So if we have multiple approaches and sims to compare them maybe that path will become visible.

[Fletcher]

Got to it .. here's the Raj AutoWheel Single Mech sim again with changes.

I altered the Spring Length to 0.8 meters (from radius 1.0 meters), but kept the K value at 50 and dampening at 1.0.

This length meant that the 1 kg weight used in the sim did not move up or down when suspended from the spring. IOWs the spring has Elastic PE and the weight has GPE. There is tension in the system.

Then I ran the new sim and plotted the weight track again. It doesn't go so deep beneath the axle and the wheel gets to 354 degrees of rotation now (previously 332 degrees) before reversing direction to CW.

Sim included as attachment.

[Fletcher]

While I'm about it .. I ran the sim again without the push start.

I also rotated the wheel background around to 180.1 degrees CCW as the Starting Position. Just enough to start the lever-weight falling downwards and backwards under its own steam.

The wheel rotated to 348.9 degrees. That is is started at 180.1 and went to 348.9 degrees before going to zero rpm and then reversing direction. The track shape changed to be more circular because there is no extra impetus at the start.

That was 168.8 degrees of a minimum 180 degrees required for reset.