Symmetrically Balanced Systems – are they able to develop useable torque ?

[Wheeler]

Fletcher
Will the vessel expand at any point?
I was looking at the leverage system and wondered if tidal action could also be useful.

[Fletcher]

Paul & Wheeler .. both of you are close to how I see two modes of operation, as I mentioned in a previous post.

1. The ball float is actually free to move (slide) up & down its shaft (but has a latching point). The shaft (T rod) extends down close to the vessels bottom so that there is a small clearance maintained & it floats the ghost weight, in the stable position.

Release (unlatch) the float so that it rises in the fluid & the shaft moves downwards to make contact with the bottom of the reservoir. Then transmission of the weight is allowed & the ghost becomes corporeal, causing imbalance.

The pendulum then swings. The fluid has inertia but due to its viscosity the fluid 'lags' behind as the mech swings causing the unlatched float to rise further up the shaft. Then baffles in the vessel (& perhaps flap valves in a partially inclosed vessel that allow some back seep, or air lock valves) pull the fluid with the vessel. As the fluid is forced to move sideways with the swing it sets up a surge (tidal rocking motion) which then lets the float slide back down the shaft to be latched in its correct position again, to lift the ghost weight on the mechs return swing.

A self pumping pendulum. Obviously you don't need a wheel. One central mech could be connected to a crank to turn a wheel.

2. This is slightly different mode of operation. The red ghost weight is attached at the end of a horizontal scissor mech on the T rod. The scissor is 'fired' (extended) sending the red weight sideways to the right. This increases the leverage on the float (there is a larger clearance to the bottom of the vessel this time allowing reasonable down movement of the float which is fixed to the shaft). The shaft (still protruding from the bottom of the float) contacts the vessels bottom & is latched there until ready to be released.

The contact allows some transmission of the weight to cause some imbalance but this time the CoM of the fluid is raised upwards so that it is closer to the central pivot & h1 > h2. This makes the mech top heavy causing it to imbalance & swing.

At the right time the shaft & float is unlatched causing it to rise in the fluid, retracting the scissor & pulling the red weight closer to the pivot from whence it came. The CoM of the fluid moves down again to its original distance from the central pivot, ready to repeat the cycle.

Paul .. your thoughts on a double float system raise interesting prospects for a double pump system.

[Fletcher]

This is a continuation of the hydro theme, for now.

As many of you know Rainer [Tinhead] & I work as a team. We have complimentary & overlapping skills & besides we have become good friends from this entire exercise.

This thread is used by us as a repository of ideas that may have had potential to be 'different, unusual, out of the box' to some degree or other, as seen by us.

We have explored many many ideas & concepts & these are just a few that were a little more interesting or went a little further than normal perhaps. Mainly the ones that at the time had question marks about their exact performance or were initially a little difficult to build accurately on WM. They needed quite a bit of thought to work within the limitations of the program.

Some required an actual real world build to fully test the concept but this was undertaken mainly to check back against the sim model for accuracy, where we had one. We are always hopeful that WM might be wrong 'this time' but so far its predictive ability has been spot on "sigh". That means that in our experience it has been entirely reliable, so far. This is both comforting & disquieting at the same time.

As you will see in the following posts we have had to be a little creative with certain formula's & functions that otherwise could not be easily simulated otherwise. WM entirely mimicked the real world testing. All thanks to Rainer for his ability to take an idea, break it down into its simplest parts & construct a WM analogue that does the same thing as you want, without compromising the idea.

[wikiwheel]

""""As you will see in the following posts we have had to be a little creative with certain formula's & functions that otherwise could not be easily simulated otherwise. WM entirely mimicked the real world testing. All thanks to Rainer for his ability to take an idea, break it down into its simplest parts & construct a WM analogue that does the same thing as you want, without compromising the idea.""""""

Is their a way to save the picture machines in Working Model program demo?? Im tired of starting once over each time.
Any helps will be appreciateted.

Mik

[Wheeler]

Fletcher
Thanks for the update and it is good to know you guys are working together with confidence and friendship.
I look forward to your posts.

wikiwheel, I think you may want to ask Jim_Mich about how to set up a program that you can save, but I think you may want to invest in the full use program from Working Model 2D when you can.
I did not go deep into working model 2d, but others have.
The thread about wm2d is also on this forum, and you may wish to address your thoughts on that thread also.

10 to 15 inchs tonight in the eastern US
cheers

[Fletcher]

Although Bessler said that 'all things must rotate with the wheel' & seemed to specifically rule out 'anything hanging from the axle' it does not necessarily completely rule out Counter Weights, as you will see.

For instance, there could be multiple counter weights located on their own stub axles, that all rotate round with the wheel. This scenario would neatly side step the center axle issue.

The problem with counter weights is that as soon as you use them as an artificial gravity by attaching a mech to them the torque from the mech causes the counter weight to be displaced left or right [no matter how massive it is] & this creates counter torques.

See attachment for a simple counter weight & its action after the small weight applies its torque. The CoG will always find its lowest point [position of least potential energy] which is usually vertical below the axle pivot.

So .. is there another way to use counter weights that don't have all the undesirable characteristics of a mechanical artificial gravity.

There might be .. see the second attachment. Use a container [bucket] that contains a liquid. Have it symmetrically hanging below a pivot. To that same pivot attach a right angle bracket supporting an ordinary old floatation device [float]. At the corner of the right angle bracket attach a secondary pivot from which can hang a free moving driver.

Now we can experiment with applying different weighted drivers to see if anything interesting happens.

The driver causes the float to descend into the liquid [it displaces its own weight of water, in this example]. A few interesting things happen.

As the float displaces the liquid the water level in the bucket compensatorally rises so that the CoG of the entire mech cannot find a lower point than it started at. So, as you apply more weight to the float [or force] the CoG stays at the same horizontal level. Also the bucket always hangs vertically [always - so it has no back torque issues acting on its pivot attachment].

As the buoyancy force resists the weight of the driver [when they are in equilibrium] it increases the depth of the liquid which in turn increases the water pressure force on the bottom of the bucket. The same happens when an external force is applied, say pressing down on the float. The pressure at the bottom of the container increases creating an unequal force to an identical bucket located opposite side of the wheel, without the applied force.

Also this arrangement could potentially provide a very handy mechanism to create an offset position for mechs to leverage off within a wheel. This might be useful.

[Fletcher]

So we had what appeared to be an interesting counter weight mechanism that used ordinary, readily available materials & was very common knowledge. A combination of Archimedes & Pascal's Laws.

But was there a way to use it ? So far it looked liked a torque free counter weight that essentially acted as a spring.

The attachment below is the actual bucket simulation as built by Rainer.

Note the buoyancy force is equaled by the water pressure force acting on the bottom of the container. Although the formula's do not account for viscosity or dampening of movement they were sufficiently complex, yet simple enough, to closely approximate the real world situation.

The CoG is always directly below the bucket pivot & therefore the bucket always stays horizontal providing the driver does not touch the bucket.

[Fletcher]

I decided I had to build this basic device to test the sims against the real world. The sims were adjusted to accommodate the dimensions, lengths & weights of parts in the real build, so we were comparing apples with apples.

Besides, I didn't have a better idea to build at the time.

This gave us a few basic scenario's to follow up on.

The first we tried was a rail guide system that made the driver follow an eccentric path around the wheel axle. Pics attached. Rainer also simmed it (screen shot included).

Disregard the secondary upturned float that you see. It was the simplest way to balance the weight & torque of the working float around its pivot, so that it was counterbalanced & taken out of the equation so to speak.

[Wheeler]

Beautiful
A real build with pre testing on wm2d.
Thats refreshing.
Any positive action on the build?

[Fletcher]

We learnt a lot more practical detail about Archimedes Theorem & it was about this time that Ralph started asking questions about rocks dropped from barges ;)

I will give the results & conclusions about this concept in the next post. Got to get some painting done while its fine.

After that we changed tack slightly & our hearts missed more than a few beats because the sim showed a slight torque advantage that meant that the wheel accelerated away. It was to be relatively short lived euphoria as consternation & natural caution took us back to the sim buoyancy formula's to poor over them again with a fine tooth comb.

[coylo]

Take Rainer's sim, at the 6 and 12 o'clock positions the water is displaced to the left as the float is forced down bearing the load, yet the guided load does not move in your favour to counter this......that's a problem - it's a countertorque.
It should balance the system, cancelling the gains you made at 3 and 9 o'clock.

I've been there before with something similar. I think I made the assumption that the displacement was going to be a neutral reaction, but it wasn't.

[Fletcher]

How true, so you might be surprised what I tried next to crack that particular nut. I look forward to your analysis & comments Coylo. It's good to think others have been here also because then I don't have to explain so much detail.

P.S. the float bearing down displaces the its own weight & the weight of the driver arrangement. Because of the particular properties of liquids the average density [once the float is neutral buoyant] up to the water line means that the CoG is still directly below the bucket axle (no torque from the bucket).

So the torque or counter torque as the case may be is from the unsupported weight of the driver leaning on its particular guide rail.

[coylo]

I think I tried changing the design of the bucket/container from a rectangular cuboid as you have, to a parallelogram one leaning to the right, to solve that particular problem. So that when the float was submerged the water would raise/displace to a neutral position.
The float design itself was a cuboid.

[winkle]

it's probably a good thing that mentally pissing into the wind don't get anybody wet cause after three hundred years this business could have got really soppy 8)

[KAS]

Fletch,

This is a very complex and impressive system that looks as though it could and should work.
I have racked my brains to discover way to offset the centre of gravity on rotating weights but your idea is a real kick yourself one.

From my observation, It appears that the mass of the weights (scaled to the wheel) is very small. Yet given the buoyancy value of the floating drums, it looks as though you could afford to increase the mass value of the weights considerably.

This is only an observation, and no doubt you have already done the homework with this on WM2D. What was the max mass value that you could use on this setup before the drum buoyancy was lost?

Kas