Mikes Wheel Idea

[Mike]

Hello all,

I have been thinking alot about Georg Künstler's ideas. While looking at his wheel drawings I thought maby it would be better to add a one way bearing to the weighted wheels so it would have a smoother rotation without swinging. To me this looks like the way to go. I will try to make a test wheel this weekend and I will post any progress I have. I can order these bearings from a R/C Car catalog, they are used on the small gas engines they have. you can find them on the crank shaft where the pull starter is located, the bearing locks when you pull the pull starter, and free wheels when the engine is running.
Please let me know if any of you have allready tried this wheel idea before and found that it doesn't work.

Thank you

Mike

[snpssaini]

sorry, It will not work.

[bluesgtr44]

I agree...unless maybe it latches the weights somehow...even then I have my doubts.


Steve

[jim_mich]

McMaster-Carr http://www.mcmaster.com sells one-way clutch bearings on page 1024 of their current online catalog. If you can get your hands on one of their physical catalogs it is about two inches thick, over 3200 pages full of everything you can imagine for factories and manufacturing.

They take orders from anyone with no minumum order size. Just go online and use your credit card. They usually ship the same day, many times within hours of receiving your order.

I've ordered from them many times. Their prices are fair, but not bargains. Local hardware stores are at times cheaper, but local stores don't have the huge selection and don't deliver to your door.

[ken_behrendt]

Mike...

I assume that the direction of rotation is CCW. Since a weight is being raised at the 12:00 position, the remaining weights must experience a drop in their gravitational potential energy the magnitude of which exceeds the increase of gravitational potential energy by the weight being raised at 12:00.

It is not possible to just look at this design and immediately tell what will happen during its rotation. Also, trying to build a CAD model of it would be very challenging due to the clutch type bearings and the requirement for the weights at 12:00 to rotate after contact with the surface.

Building such a design as a physical prototype would also be a daunting task.

The simplest approach to "testing" this design before you decide to build is to just use graphical techniques to determine what is happening to the CG between rotational segments where the pattern of weights is repeated. That is, using geometry, you can calculate, roughly, where the CG is, say, after each 5° of CW rotation.

Unless the CG of the weights stays on the left side of the wheel, then there is little hope that a physical model would work...

ken

[Fletcher]

Hi Mike,

Using friction to "roll" the eccentric weight to its top most position (at about tdc) will rob the system of loads of energy. It is I suppose an alternative to using a geared lifting arrangement or simple ramp to lift the weight into position, though perhaps the drawbacks are not so obvious.

Because the eccentric weight follows an orbital path the rubber contact track must be spring loaded in some way to provide sufficient non slip to do its job otherwise the eccentric weight won't rotate. Thats a lot of pressure & back torque.

Therefore the resulting unbalanced arrangement must be able to overcome the spring tension so the track can move out of the way. You might be envisaging using just its weight alone rather than spring loading but then the rubber tracks own weight must be lifted.

Either way the the rubber track must be moved to a higher position whilst maintaining its pressure contact with the eccentric weight.

This is just like a upside down ramp but doing the same thing with the same problems imo.

[Mike]

Hello,

Thank you Jim, the McMaster-Carr site is great and the prices on the one way bearings (roller clutch) are very reasonable.

Thank you Ken, I will find the center of gravity at each 5° of rotation. I now have the WM2D software, and I ran a simulation of my wheel idea.
Instead of the one way bearings I simply just pinned the three down going drive wheels to the main wheel. The simulation shows that the three down going drive wheels are more than capable of rotating the 12:00 drive wheel into position. I realize building this wheel won't be easy, and I will take your advice on testing before I get my hopes up.

I have a video clip of my WM2D test but it is to large to attach I could E-Mail it to anybody interested.

Thank you all.

Mike

[Mike]

This is my WM2D test setup.

[Fletcher]

That looks interesting Mike. So how did it perform ?

I take it the orange rubber track is reasonably counter balanced & the blue steel weight is 'adjusted as needed' to create enough friction coefficient to roll the eccentric weights without slippage.

The process of rolling the eccentric weight thru 180 degrees to the top will happen over an arc of the wheel circumference so I guess the plan is to start it b4 tdc & finish just after tdc to maximise the leverage gained. Starting to early or to late will be counter productive ?!

Catches or latches to hold & release at the appropriate positions are notoriously difficult to do on wm2d so an actual build might be your best 'test option' as you are no doubt planning.

[Mike]

Hello Fletcher,

The WM2D test wheel performed just as I was hoping it would.

The idea:
Drive wheels are rotated into position starting at 12:00 and reach the desirable position (eccentric weight at outer most edge) shortly after. The trick is to use a one way roller bearing or sometimes called a roller clutch. this one way bearing will hold the eccentric weight in position until 6:00 then the eccentric weight is able to dangle down again do to gravity and the one way bearing un-locking. See the diagram of the bearing I posted to understand this type of hold & release, the latch is built into this bearing.
I have put togther the main wheel and now I must wait for the special bearings. I think others should try this Idea, to me it looks promosing, although I might be wrong. I appreciate your comments and hope to hear more.

thank you

Mike

[Georg Künstler]

Hi Mike,

I like to say, free energy for free minds. Don't forgett that you can also rotate the wheel on the top(second system), with the trouble force. Not only hold it.


Fletcher wrote:

Starting to early or to late will be counter productive ?!

yes it will be counter productive, you can make moves that are gravity supported. An eccentric weigth is heavy, when you turn it up, and light, when you turn it down. So timing is essential. Doing the lifting always on one side of the wheel, will pass torque. So easy.

the future has begun

Georg

[Mike]

Yes Georg,

"The future has begun" and soon you theories will be proven.

Thank you for inspiring me.

Mike

"Free energy for free minds"

[ken_behrendt]

Mike...

I tried making a quick, but crude, WM2D model for your "Eccentric Roller Weight Wheel" and the results are attached below.

My version consists of a transparent 6 ft diameter wheel that weighs only 0.5 lb and carries 8 rubber rollers whose eccentric weights each weigh 2 lbs. They are supposed to be vertically raised at the 12:00 position after they make contact with the odd shaped rubber piece (orange) that weighs 20 lbs.

I just rigidly pinned the eccentric weight rollers on the wheel's descending or left side into place, then I ran the simulation.

Well, the CG of the 8 eccentric weights is definitely to the left of the larger wheel's axle and this design wanted to rotate CCW. However, as soon as the roller came in contact with the 20 lb rubber piece at the 12:00 position and its eccentric weight began to be raised up, a torque would come into existence that tended to make the larger transparent wheel rotate CW.

No matter how I positioned the 20 lb rubber piece or what I made its weight, I could not seem to prevent the CW rotation of the large transparent wheel. And, in every case the wheel would stop rotating after it had rotated CW a little bit.

Consequently, I do not think that this design is workable...but, then again, my WM2D model is very crude...

ken

[Mike]

Hello Ken,

You are right I had the same problems for a while, but after adjusting the contact point and period just right it works. I can now do it easily there is a sweet spot where this happens also with your model, you are missing a essential piece the rest for the rubber track if there is no rest the weight of the rubber track will push the 12:00 eccentric back the other way. Look at the WM2D test wheel I posted and you will see the rest point. otherwise your version looks good. If you want I can e mail you a video clip of the WM2D simulation I made.

Thanks

Mike

[Mike]

Ken,

I found that if you right click on my WM2D test wheel post and select copy you can then paste it to your WM2D work area, then you can bulid right over it this way you will have an exact copy to test.

pretty cool

Mike