Part Three is the Charm

[Tarsier79]

OK. One more go then I will leave it. I recommend building some real world" tethered" simple builds with a single mechanism(2 weights) and a counterweight. See how the weights feel, work and act.

Below is a close up of my edited diagram. At 3:00, the large weight force is straight down, while the spring is pushing across. The weight will not compress the spring in this position. You can see the angle of force of the weight compared to the spring as it rotates down. I suspect the 4:00 will not be strong enough either. Eventually the weight will be in a position to compress the spring.

The other diagram shows the back torque compared to what you want.

[mryy]

OK. One more go then I will leave it. I recommend building some real world" tethered" simple builds with a single mechanism(2 weights) and a counterweight. See how the weights feel, work and act.

Below is a close up of my edited diagram. At 3:00, the large weight force is straight down, while the spring is pushing across. The weight will not compress the spring in this position. You can see the angle of force of the weight compared to the spring as it rotates down. I suspect the 4:00 will not be strong enough either. Eventually the weight will be in a position to compress the spring.

The other diagram shows the back torque compared to what you want.

Okay I didn't understand what you meant earlier with the edited diagrams. Duh. Basically you *believe* the levers will not swing down fully until around 5:00. And I thought you were suggesting some unknown counter-torque effect...

The lever will swing out at 2:00 as I expect. I use spring at the lever's tip and the farther-out yellow weight compartment to facilitate this. At the fulcrum of the lever the torsion spring constant will be optimized. The constant will neither be so large to prevent the weight and tip spring from doing their jobs at 2:00 nor so small to lessen the necessary projectile action near 6:00. It's a balancing act of all the players in the system.

And do note that I am offering here a wheel concept that depicts more or less idealized, yet plausible movement and position of the weights.

[Tarsier79]

The lever will swing out at 2:00 as I expect. I use spring at the lever's tip and the farther-out yellow weight compartment to facilitate this. At the fulcrum of the lever the torsion spring constant will be optimized. The constant will neither be so large to prevent the weight and tip spring from doing their jobs at 2:00 nor so small to lessen the necessary projectile action near 6:00. It's a balancing act of all the players in the system.

I think you are too close to the design. Take a step back and try to look at it without bias.

At 2:00: the Yellow weight pushes the lever inward. The spring pushes the lever inward. The red flying weight tries to push the lever outward against 2 forces. The red weight started off with the same energy as the spring (not quite), but when it lands at 2 it has lost a significant amount of energy, so will not compress the spring......

[mryy]

The lever will swing out at 2:00 as I expect. I use spring at the lever's tip and the farther-out yellow weight compartment to facilitate this. At the fulcrum of the lever the torsion spring constant will be optimized. The constant will neither be so large to prevent the weight and tip spring from doing their jobs at 2:00 nor so small to lessen the necessary projectile action near 6:00. It's a balancing act of all the players in the system.

I think you are too close to the design. Take a step back and try to look at it without bias.

At 2:00: the Yellow weight pushes the lever inward. The spring pushes the lever inward. The red flying weight tries to push the lever outward against 2 forces. The red weight started off with the same energy as the spring (not quite), but when it lands at 2 it has lost a significant amount of energy, so will not compress the spring......

You forgot about the tip spring (which was mentioned in the OP and in my previous post). Its purpose is to counterbalance the inward forces of the weight and the fulcrum spring. IOW the 2:00 lever is held in a state of equilibrium between two sets of opposing forces. The flying weight striking the lever adds just enough energy to the side of the tip spring, triggering it to unwind and swing out the the lever. Once the lever enters a certain position, the heavy weight reverses course and begins cocking (via gravity) the lever toward the other end of the blue guide. By 3:00 the lever is expected to be fully cocked.

[Fletcher]

Here's mryy's wheel drawing again for reference ..

[Tarsier79]

Thanks Fletcher. The 2:00 lever has been drawn as if it has already been pushed forward.

The tip spring: That is one magical spring. Mryy, you should build a mockup to help comprehend the forces you are working with.

[mryy]

Gee, I wasn't expecting every detail in the conceptual diagram to be scrutinized. That's what the written descripton in the OP was for. The diagram -- neither drawn to minute detail nor kinetic accuracy -- was intended to illustrate the way the various components of the wheel worked together. In fact it is the 1:00 lever as it descends into the 2:00 position that receives the flying weight. I mentioned it in another topic "Introducing ??theOne$$: A Breakthrough Concept?"

True, a mockup will definitely aid in the study of the forces inside the wheel.

[johannesbender]

The 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'.

Rethinking this , I'm speculating that it may be that this could have been a weight transition time (lag or such) causing it.

[Tarsier79]

Mryy, I do hope you progress to testing/mockups. Simulations are a good starting point that requires only time to learn.

http://www.algodoo.com/download/

Algodoo is not the best, but it can be useful for testing movements etc. and it is free.
Just remember, in sim, if something doesn't look like a natural movement, it probably isn't. Most of the time, if something starts to self revolve, it is a computation error. Again, use the tool, understand its limitations.

Good luck.

[Fletcher]

I think most of us follow your logic and apparatus reasonably well mryy .. fwiw, you are getting mainly questions about anticipated practical implementation aspects of the idea. Things you can reasonably answer with a mock-up, sim, or real-world build, or all three as you choose.

Speaking from my experiences seldom does a concept drawing not morph in the simming process (often seeking simplification, accuracy, efficiency, and flexibility to test input ranges), and again in the practical real-world when we hone the design some more.

Previously I asked about whether your concept could be adapted to something MT13ish i.e. half-full half-empty concept - requiring lever-weights to be lifted at the top. Personally I can't see an easy way to adapt it. Part of that question coming to me was a flash back to May 2005 when a member here called jim_mich published his Roller-Ball concept (I had kept a pic of it in my computer files) and I was thinking maybe your flung weight could drop onto his lever system, or something like it, rather than ride the wheel around etc.

Anyways, for a blast from the past I post up jim_mich's Roller-Ball gif. Also, I just did a forum search and found an animation of it he did at around the same time and saved it so we could see how he thought it might operate, or perhaps how it could be adapted. He animated in on his visual basic program of the time IIRC i.e. not a sim program.

Good luck.

[Fletcher]

The 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'.

Rethinking this , I'm speculating that it may be that this could have been a weight transition time (lag or such) causing it.

Or .. if it can overcome dissipative energy losses, like frictions, windage etc (i.e. minor Work), and still accelerate up to operating rpm, and do substantial Work on top of that, and still accelerate etc. Then it was gaining in momentum/RKE each rotation.

We normally think that everything inside must be "accurate as can be" otherwise we reduce the chance of our wheels being successful.

What if it is as it sounds - accurately made, inaccurately made (within reason), it makes no difference to the momentum gains. IOW's the gain mechanics has a huge tolerance for workmanship and accuracy of parts and weights etc. That's what I think he meant, imo.

[Sam Peppiatt]

To @,
Old Bessler makes his wheel sound pretty darn cool; too bad he makes no mention on how to do it-----------------Sam

Live your days inspired anew, LYDIA

[Tarsier79]

I did like Jim.

I don't specifically remember that animation, but I like it. (obviously not as a runner).

I suspect the lever system would need tweaking to operate with equal force over its length and angle.

This animation shows the chicken and egg scenario that often happens: is the lever being pushed down by the ball, or is the wheel forcing the lever under the ball? Most people will say 1. I say 2. The ball could be replaced with a static roller. I see the connection to MT13.

[MrTim]

JC's AP wrote:
The 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'.

We normally think that everything inside must be "accurate as can be" otherwise we reduce the chance of our wheels being successful.

What if it is as it sounds - accurately made, inaccurately made (within reason), it makes no difference to the momentum gains. IOW's the gain mechanics has a huge tolerance for workmanship and accuracy of parts and weights etc. That's what I think he meant, imo.

If the mech prototype I'm currently building actually moves as it's supposed to, the design will be very forgiving as in the above. My crafting skills aren't the greatest (e.g. trying to drill centered holes) so I'm figuring on 1% to 3% of mechanical 'slop' in the final build. (One 'improvement' I'm making as I go is trying to round measurements to the nearest half inch, instead of having fractions like 3/10 & 4/10 inch. If the finished prototype works, it becomes a pattern. ;-) As to weights, the ones I plan to use average 4.06 oz each (+/- a smidge) so close enough... ;-)

[Fletcher]

I liked and had a lot of respect for him also Tarsier, for many years. Not so much at the end when he had his health problems etc and became a little irrational, but that was somewhat understandable under the circumstances.

This animation shows the chicken and egg scenario that often happens: is the lever being pushed down by the ball, or is the wheel forcing the lever under the ball? Most people will say 1. I say 2. The ball could be replaced with a static roller. I see the connection to MT13.

Yes, some of us discussed it couldn't be a runner, tho it looked impressive art. Part of that argument is as you say. The ball would ebb and flow (hunt back and forth and up and down) as it attempted to depress the lever, and the lever attempted to move under the ball lifting and moving it etc. In the end all torques balanced IIRC, and with frictions it would come to a stop.

And also as you say the ball could be replaced by a pinned roller that the lever must pass beneath. Usually when people imagine it like that then it becomes easier to see why it can never be a runner (not enough CCW torque and it stops) - but somehow the thought was that if the ball could 'hunt' then it might be able to be a runner - unfortunately it makes no difference. All forces balance out at PQ position.

ETA : P.S. with a fixed position roller it would effectively be just like MT13 - even in MT13 the central hanging counter-weight with roller above could have been replaced by a fixed position roller independent of the 'wheel'. It wasn't, and the counter-weight and roller could 'rock' slightly (depending on mass) i.e. absorb momentum and then give it back again on the back-swing thru the top mounted roller etc. These examples suggest how complex it can get trying to anticipate each and every movement, and action / reaction forces. Sims do a great job of doing that for you (if used properly, and it can be simmed), but nothing beats the real-world build as it can't fool you and you can't fool it.