The summary of my latest studies

[path_finder]

This is also in relation with the jimmyjj old suggestion:
http://www.besslerwheel.com/forum/files ... _work2.jpg
http://www.besslerwheel.com/forum/download.php?id=4050


edited:
For eccentrically2,
Another suggestion from ovyyus and ken_behrendt:
http://www.besslerwheel.com/forum/files ... al_175.jpg
http://www.besslerwheel.com/forum/files ... xlepin.jpg
http://www.besslerwheel.com/forum/files ... arnest.jpg
http://www.besslerwheel.com/forum/files ... attach.jpg

[path_finder]

Returned back to the most promising experimented designs.

One of them was here: http://www.besslerwheel.com/forum/viewt ... 1761#81761
A previous animation (here: http://www.besslerwheel.com/forum/files ... swing1.gif) attempted to summarize the process.
But I suspect now a not fully completed analysis. Some ways have not been explored in particular these with a pendulum of this kind attached to the inner rim.
I made some building using this idea, like here: http://www.besslerwheel.com/forum/files ... ample1.jpg, but never with a length close to the diameter.
This is the reason why I decided to remake it, before to build a new experiment. The animation is hereafter.

Some remarks can be done:
1. The starting position is recovered after 720 grades (two turns)
2. Due to the length of the pendulum (a little bit shorter than the diameter) there are two locations where a jump occurs:
- one when the pendulum is suspended (named 'position A')
- one when the pendulum is inverted (named 'position B')
3. when the rotation speed of the wheel is very slow (about null) the duration of the jump for the both positions A and B is equal.
But when the wheel rotates, due to the relative speed between the bob and the rim, and also due to the centrifugal force, the duration becomes very different.
4. Looking carefully the animation you will see a very strange behavior for the weight located at 6:00.
It looks like the bob will be transferred from one pendulum to the next. Is this in relation with the 'escamotage'?

The idea is to start from a perfectly balanced situation at the rest position.
This can be done very easily: all two opposite pendula must be aligned, showing a perfect cross.
Then, using the rotation speed and the centrifugal force, to create an overbalance coming from the differential duration of the jumps.

[paul1]

Hello path_finder,
I have read a lot of your and many other posts,but until now I had trouble registering.
After seeing various designed it seemed that what most people wanted to achieve was a centre of gravity permanently to one side of the centre .

So I spent a lot of time messing with bits of tube and bike wheels and came up with a configuration that does that.
I have been out of work on medical grounds for a long time so I don't have the money to build a proper version with the right parts .
But, if you or anyone who posts on this forum have the resources I would love to see a precise engineered prototype made.
I would hate to think it will never be built due to my lack of funds.

I made one from copper pipe and bike wheels and fabricated some joints ,but they were not strong enough and as i say i cant afford the proper materials.
The centre of gravity moves from the centre of the wheel along about 75% of the radius and back again ,it does 2 oval rotations on the radius for every one rotation of the rim/circumference.
The radius you want to travel on can be any of the 360 available it doesnt have to be vertical or horizontal.
the extent to which the weights can be pushed out to overbalance can be multiplied on the heavy side by various ways.
I just did 2:1 overbalance .
So the weights on one side are always twice as far out as the opposite weights ,but this can be varied.

If you are interested and can show me some easy to understand software to show you some drawings .
I could show you what i made but i had to take it out of the frame until I can make one out of suitable materials.

[path_finder]

Dear Paul1,
Many thanks for the proposal.
If you have some difficulties for whatever reasons, there is a simple way to go forward in your research.
Let's do like me: publish everything in this forum, and then I'm pretty sure if your design is confirmed, there will be several members happy to replicate it.

Now IMHO the best way to make some simple animations is gifsicle, working either on Windows or Linux. It can be downloaded here: http://www.lcdf.org/gifsicle
Just prepare the individual pictures in gif format (using paint p.e.) and then assemble them with gifsicle.

[paul1]

Thankyou for your reply ,
I did do a quick drawing in paint as you suggested.
I would need to study that gif software a bit more as I can't understand it .
I'am more of a practical person.
What I made in my garage worked but was too flimsy to carry the weights and pressure ,the wheels I used were only bike wheels and not made to take sideways pressure.
I only used copper tube again this needs to be solid bar .
It needs bearings and ball joints which I can't afford.

How to you attach an image to these replies ?
[/img]

[path_finder]

http://www.besslerwheel.com/forum/viewt ... 5521#75521

[jim_mich]

Click here for instructions.

[paul1]

Thankyou for your instructions ,
Ok this is first time using paint so don't laugh.
The image should give you the general configuration to allow the pivot to pass as I say along just the radius .
It looks very simple but I had a few scratching my head moments .
Love to hear your comments Thankyou.

This is the basic principal .

If you like I can draw the red circle configuration.

Then there is the third circle which determines were the weights are in relation to the red circle.
Would be a bit cluttered to draw it in yet.

I Have search all over and I can find anyone who has converted circular motion into linear in this fashion.All the ones I have seen have got to travel around the central pivot.

[AB Hammer]

paul1

First welcome to the forum.

Several of us have done some similar wheel approaches. Here is one of mine.

http://www.besslerwheel.com/forum/downl ... er=user_id

[paul1]

Thankyou for the welcome AB,
I cant fully make out how yours works but at first look, it doesn't seem very overbalanced and the weights seem fixed is there any leverage involved?
What is the overbalance ratio of that configuration?

[AB Hammer]

paul1

It is the angled stress/friction that is what has to be overcome with these approaches. You may want to start a new string. This one is path_finder's. To add a bit here or there is OK but we don't want it to be even accidentally highjacked.

Alan

[paul1]

I have seen a few posts that have wheels set at an angle .
But I havent seen any so far that has the cog following this path .
If anyone has seen the cog follow this path then Thats fine,could you tell me what the issue was that prevented it working?
Or just point me in the direction of the author.
At least I can convince myself this has issues that I have not come across so far .Due to me not having the correct materials to take the forces involved.
Thankyou.

[paul1]

I have seen a few posts that have wheels set at an angle .
But I havent seen any so far that has the cog following this path .
If anyone has seen the cog follow this path then Thats fine,could you tell me what the issue was that prevented it working?
Or just point me in the direction of the author.
At least I can convince myself this has issues that I have not come across so far .Due to me not having the correct materials to take the forces involved.
Thankyou.

[paul1]

No problem AB ,
I wanted specifically to present to path_finder because His models of his experiments are like a work of art and very well done .If anyone could put my idea into a practical form he would be the man.

But can I say that with the permenant overbalance and large leverage friction and the angled stress is a non issue.

Just configuring as I have shown without any weights attached it will spin as freely as a single wheel on a bike would.

Adding 6 weights of the same value and exactly the same distance from the cog it continues to run as freely ,but I need wheels rods and bearing that can take the weights involved.

Thankyou for your time guys I will only post further information if asked .

[path_finder]

Dear Paul1,
Many thanks again for sharing your idea.
Nevertheless and without any further details I have nothing particular to suggest.
IMHO your red circle is nothing else than a 'double cardan' (or 'homokinetic join' see here: http://en.wikipedia.org/wiki/Constant-v ... ble_Cardan)
So far instead to use a sliding ring and two rods, I recommend to you a more efficient and modern mechanism, allowing a significative reduction of the friction.

This kind of design, using two secant (NOT parallel) planes is difficult to build because if the principle is attractive, the realization is fronted to a big amount of friction.
Several threads in this forum have released this point in the past, like here:
http://www.besslerwheel.com/forum/viewt ... 4742#74742 and following posts
http://www.besslerwheel.com/forum/viewt ... 0049#80049
The principle was interesting enough to me for building a demonstrator, looking at the related problems (see here: http://www.besslerwheel.com/forum/files ... shape2.jpg), without any significant success.
The principle is attractive, like explained in the drawing below, because as long the wheel rotates all weights (linked to the outer rim of the wheel by some rigid legs) will follow the eccentered path (in green).
This can also be completed by a spring beetween two opposite weights.
But the constraints generated on the axle are enormous if you want to save the angle made by the two shafts at the cardan location.