Simulate or Fabricate?

I moved house two years ago and took the opportunity to throw out most of my old bits and pieces of wheel experiments.  I had hoarded them thinking that I might want to go back at some point to revisit an earlier idea.  But then I realised that after almost 40 years of fabrication I had progressed to the point where I knew instinctively what would or wouldn’t be useful in a new design.  So I bought in new material and once I had my workshop up and running, earlier this year, I set to with renewed enthusiasm to build my latesy version of Bessler's wheel..

And that is my point; I have tinkered with simulations from time to time but have never had the same feedback that I get from actually handling the piecwes, and have dismissed them as a waste of time. Making parts, manipulating them, finding that sometimes they don’t work as you expected, or discovering new movements in the process, or  they inspire new thoughts and designs - you cannot beat actually building the wheels.  It generates enthusiasm for new designs - well it does for me.  BTW, I cannot imagine anyone who is not an incurable optimist ever solving this puzzle.  You need a 100 per cent positive attitude to find the answer.

The design I’m building has been more or less complete in my mind for a couple of years and I am sure that it would be difficult to build a simulation so an actual build is the best thing and anyway you would need to build it eventually just to prove you were right.  Maybe the simulation might indicate that it would not work because of some simple error in the input or the settings.

I’ve seen any number of comments about problems with a simulation and I simply don’t trust them. I’m sure that an expensive simulation software could predict the correct solution, but these off-the-shelf versions seem to me to be full of glitches and are not to be trusted.

Returning to my own build, I set out the design parameters and copied them onto the backboard (the wooden disc which will hold all the parts) and drilled the necessary holes in exactly the right places. I fitted the levers on their swivel posts.  The levers were precisely the correct length.....but when I manually rotated them, two of them touched two adjoining levers, interfering with their actions. Admittedly space was tight and I could have planned for them to have more room, but it is surprising how hard you try for extreme accuracy you can still overlook some small discrepancy in the position of for instance a swivel post.

In fact the accuracy of the positioning of the posts is not vital to success.  A millimetre in any direction would not affect the viability of the design.  So I drilled two new holes and corrected the problem.

It has been often stated that there is only one design that will work,  but is this true? I ask this for two reasons, firstly Bessler said he had several designs which would work, and secondly I am aware of at least one design which apparently works. I have seen the design on paper and I cannot say for sure if it will work, but it seems that a working model was made.  What I can say is that this design is different to mine and I do know that my design is the same as the one Bessler intended to pass on to us. This suggests that there is more than one way to use gravity as an enabler of continuous rotation.

I can see from my own design that it looks possible to create the same effect usng differenr mechanism designs.

So my preference is definitely for fabrication but I know many people swear by simulations and animations. 

JC