Hypothesis .. Raising GPE without using Law of Levers ? ..

[Sam Peppiatt]

Try some thing; you'll learn some thing----------------Sam

[Fletcher]

Should have summarized .. I believe the WM kinetic program mimics real-world Newtonian Mechanics about as good as it gets - enough to be reliable and predict movements and motion with accuracy ..

@ fletcher, Best wishes for success and health. Hey thx4, thanks ..

Just a detail 🙂, you could have the best software in the universe that it could only show you what it assumes based on possible calculations on the KNOWN, .. the best simulation software in the universe is called the real-world experiment - sims make predictions based on the Laws of Physics - and they are Laws until an exception to them can be demonstrated and repeated ( scientific principle ), and then they are demoted - so far no exceptions have been found at the Classical Physics and Newtonian Mechanics level to knock any off their perches ..

In other words, it's BLIND and will never see what's really happening in the PRESENT. If B had been born today, in our time, he would never have found it. I disagree - he could still have found it - he didn't have sim software, so that is not a requirement but it certainly speeds up the investigation - IMO he simply found a novel way to maintain and regenerate his OOB that is within the known Laws framework and anybody else could do it too using those same Laws - we know that because an educated Karl said a carpenters boy could make it - there was nothing that a carpenters boy had not seen or could fabricate ..

It's a fact that needs to be introduced once and for all, and that's why even the dumbest experiments can surprise us. ( I'm a specialist 🙂) There is no doubt that even the dumbest experiments can sometimes surprise us, no one is immune to that sometimes humbling experience - but we learn from it so we don't have to repeat the same experience ..

Have you ever found any empirical ( real-world ) experiment to create asymmetric torque in an OOB wheel that was repeatable giving the same result each time that couldn't be reasoned using the Classical Laws framework of Physics thx4 ? - I've never seen one, but I heard about one in 1712 and none ever since ;7) - due a second time eh ..

[Fletcher]

@Fletcher , typically kinematic software simulate physics by using the formulas for motion and forces and energy etc , where bodies have physics properties like mass elasticity etc which is all variables to be plugged to the formulas (newtons laws etc) , conservation laws etc , then there's also an algorithm model used to process/calculate or approximate the differential equations of all the numerical data (incrementally) such as the runge kutta or euler methods ( you would have to research them to have a clear and error free answer) , which after all calculations are processed the visuals on your screen is finally updated (positions etc) , of course there are all other types of formulas processed like collision detection and constraints etc.

Thanks, background and modus-operandi understood ..

Many software use these formulas to simulate motion and physics , but not all aim to have a high level of accuracy , some software aim to have an acceptable level of speed and accuracy such as physics engines and games , wm2d and other simulator software aims to have a much higher and customizable level of accuracy , but all in all they all must use the formulas we know to adhere to what we know in the real world of motion and physics .

I set my wm2d on Kutta-Merson ( Accurate ) Accuracy ..

I know this is an unstructured summation and pretty much a vain answer , but the facts are , whether one is a coder or not , in the end you cannot really fully answer these questions without inside knowledge of the code or structure of the software in question , and the reason is that coders don't have to stick to a certain particular blueprint for their software , there are different ways of getting from A to B , but in general this is more or less how its done and as much information i can conjure up without knowing whats really under the hood , the best one could do is ask the developers of a particular software for accurate information .

It is as I expected jb - sometimes 2 codes achieve the same result, but one is shorter than the other - but ironically the longer written code can be quicker and more efficient than the shorter ( all roads lead to Rome, just some are quicker ) - thanks for taking the time to lay it out .. more about it when I come back ..

[Fletcher]

... but all in all they all must use the formulas we know to adhere to what we know in the real world of motion and physics .

I am counting on it jb !!!

Background .. for decades I've wondered about and asked the question whether kinematic sim programs like wm2d were bottom up or top down driven .. we know they adhere to the Laws of Physics and Motion - I devised various sim experiments to test the boundaries to try and determine exactly how the coding was structured .. as an aside any online momentum calculator will give you Conservation of Momentum if you know the velocities before and after contact ( no brainer ) ..

These latest very simple sim experiments show that the Conservation of Momentum is the paramount consideration, from bottom up .. because ropes in wm2d can't deform and change length etc etc, and because the sim program can't identify and simulate each and every dissipative energy loss, so it handles this 'constraint' by keeping to momentum conservation - it finds the velocities, and since velocities are common to both mv and KEt then it gives remaining KEt and assigns all other energy losses ( summed ) to the ether as part of COE as a secondary accounting consideration ..

Question .. so can we all agree that for these simple experiments that if we know the mass/inertia of 2 objects set to collide, and we know their start velocities, and we know the COR, we can easily calculate what the after collision velocities for both objects will be ? ( we could use the SUVAT equations but my method is very very simple, and accurate ) - and hence also their individual mv's and agree that Total Momentum is conserved ?

Here's the simple diagnostic formula for a calculator when you don't know the after contact velocities ..

......................



......................

[Gregory]

... but all in all they all must use the formulas we know to adhere to what we know in the real world of motion and physics .

I am counting on it jb !!!

Background .. for decades I've wondered about and asked the question whether kinematic sim programs like wm2d were bottom up or top down driven .. we know they adhere to the Laws of Physics and Motion - I devised various sim experiments to test the boundaries to try and determine exactly how the coding was structured .. as an aside any online momentum calculator will give you Conservation of Momentum if you know the velocities before and after contact ( no brainer ) ..

These latest very simple sim experiments show that the Conservation of Momentum is the paramount consideration, from bottom up .. because ropes in wm2d can't deform and change length etc etc, and because the sim program can't identify and simulate each and every dissipative energy loss, so it handles this 'constraint' by keeping to momentum conservation - it finds the velocities, and since velocities are common to both mv and KEt then it gives remaining KEt and assigns all other energy losses ( summed ) to the ether as part of COE as a secondary accounting consideration ..

I think the way these simulators work is something like this:
1. Identify what the user constructed (like bodies, connections, constraints, etc) and what initial condition are set (like forces, velocities, torque, etc).

2. Come up with the conserved quantity and the equations of motion for the system. Usually, it's KE or KE + potentials = Total. Probably can come up with a system kinetic energy equation too. Then doing some algebra, moving things around, it can come up with formulas for the position, velocity, and whatever parameter for all the different components.

3. Then put everything inside an iterative loop based on position for example, always move things a tiny bit in each step and calculate the new positions, velocities, etc. for each component. This usually assumes Conservation of Energy and/or Momentum. But depends on considerations and what's the purpose, how the creators implemented it... This is mostly unseen for the user.

4. Checking for collisions, preset conditions, etc. When collision detected you just redistribute momentum/KE, and call it a day. Possibly check for conservation here too, so collision can't go unrealistic.

Something like that, but of course it can be more sophisticated and use much more advanced maths and methods.
A few times I have written my own simulators in python and jupyter notebook to double check something, or when I didn't trust the sim, or wm2d was not really suitable for the design, or wanted to see if my math on paper made sense...

We know wm2d sometimes screws up collision, friction, etc. But with a critical eye we figure that out and change to omit those silly things.
I also noticed that in some cases wm2d tries to force Conservation of Energy, which can produce a ridiculous output.
Errors can accumulate in any numerical process, and in special cases errors can also produce a gain in energy. So, you also need to look for that kind of thing somehow. But here can come the question: How can you distinguish between errors and possible gains? Or do you need to?

So, yeah. That's a hard question. Depends on how you approach the situation, and perhaps depend also on the situation itself.
Wm2d might do this mostly by allowing you to use artificial forces, torque, formulas, etc... It evaluates and validate those, let those pass even if they increase energy, but might treat other gains as errors to be handled or checked... hard to say.

Question .. so can we all agree that for these simple experiments that if we know the mass/inertia of 2 objects set to collide, and we know their start velocities, and we know the COR, we can easily calculate what the after collision velocities for both objects will be ? ( we could use the SUVAT equations but my method is very very simple, and accurate ) - and hence also their individual mv's and agree that Total Momentum is conserved ?

I would say, yes. I guess we only need the elastic/inelastic collision formulas for that?
But I didn't messed much with collisions, so I am not the best with this.

[Fletcher]

Hey Gregory .. thanks for your thoughts and detailed reply - I completely agree with yours and jb's logic of what we think happens inside the brain ( coding ) of a kinematic sim program like wm2d - fortunately we don't need to know how to code to use it, and it is user friendly and reliable by design ( with mountains of beta testing ) - for its minor grumblings it is a valuable tool that we can choose to use - as I've always said it doesn't absolve us from our personal responsibilities and we can't check our brains in at the door and collect on the way out - to have some high level of faith in a predicted result from an experiment we have designed we have to know or be able to spot or anticipate some of the 'bugs' we might encounter, and how to work around them if they could be 'show stoppers' - usually we can spot the behaviour or outputs that seem nonsensical or erratic, and redesign the experiment as a cross-check etc, simplify etc .. the bottom line is we still need to have a good working understanding of Classical Physics to run as a ruler over any sim result just like any real-world experiment we might design - I long ago gave up doing spreadsheets to predict motion and movements and cross checking against Conservation Laws - the sim does all this in an instant ..

Question .. so can we all agree that for these simple experiments that if we know the mass/inertia of 2 objects set to collide, and we know their start velocities, and we know the COR, we can easily calculate what the after collision velocities for both objects will be ? ( we could use the SUVAT equations but my method is very very simple, and accurate ) - and hence also their individual mv's and agree that Total Momentum is conserved ?

I would say, yes.

I guess we only need the elastic/inelastic collision formulas for that?

But I didn't messed much with collisions, so I am not the best with this.

Thanks Greg - and for the record it was NOT a trick question ..

It's just a simple collision test .. the question everyone else must be wrestling with is can we trust the predicted result, or does it make sense ?

Imo it is obvious that Momentum is conserved ( regardless of COR/Elasticity value ) because the Output meters show it is unchanging from start to pause, and it is what I would expect - and since Momentum is both inertia and velocity then it is also obvious that KE can NOT be conserved ( n.b. tho Total Energy is thru CoE ) after the collision except in just 1 circumstance ..

That circumstance is when the COR/Elasticity is set to 1.00 ( from elasticity range 0.0 lowest to 1.0 highest ) n.b even the best materials in the world can't and don't have a COR value of 1.00 or even close to it - steel and rubber 0.95, a cricket ball or tennis ball in the 0.7's to 0.8's ..

WHEN the COR/Elasticity for ANY collision test is a hypothetical 1.00 , then for like any other collision , BOTH Momentum AND Energy is Conserved ( aka CoM and CoE conserved ) - but in this special case TOTAL ENERGY = KEt ( with NO dissipative energy losses ) ..

Breakdown ..

when COR = 1.00 Momentum is conserved , and Total Energy is conserved ( i.e. where Total Energy = KEt + zero dissipative energy losses ) ..

when COR is less than ( < ) 1.00 Momentum is conserved , and Total Energy is conserved ( i.e. KEt is les than above, and assumed apportioned dissipative energy losses increase to balance Total Energy and fulfill the CoE caveat ) ..

Conclusion ..

If COR is 0.0 then mv is conserved ( for all COR ranges ) and KEt is half of when the COR is 1.00 ( i.e. KEt is COR dependent ) ..

If COR is 1.0 then mv is conserved and Total Energy = KEt ..

..............

[Fletcher]

Continuing the case for free lift conditions - approaching the problem and solution from slightly different perspectives to build a foundation that can stand up and be added to ..

Here is the same Leg 1 "Impulse" sim test with a slightly different arrangement - this time I have kept total masses the same etc etc but have added a fixed swinger and a swinger able to rotate CCW and be caught at its highest GPE gain ..

The COR / Elasticity of all the Body's is set to 1.0 ( effectively a perfectly elastic collision which results in no dissipative energy losses like my original Impulse / Momentum given to swinger carts a few pages back ) ..

Note from the previous sim results that in Leg 1 it is exactly as I would expect - the control and the fixed swinger comparison have the same momentum from start to pause - their Total Energy's are the sum of their KEt, which is also unchanging before and after collision ..

* The bottom Red test with the swinger able to "swing" after collision has the same total momentum from start to pause as per the tests above it - we see the Total KEt is reduced significantly after collision ( see graphed outputs ) - see how C1 continues moving at a reduced velocity because of the swingers inertia as it swings upwards - and that the swinger has gained significant GPE ( see black output meter ) - as anticipated when the Red test GPE energy is added to Total KEt the Total Energy is the same for all tests ..

** This is consistent with WD = Energy ( as per previous sims of Leg 1 ) - we expect this, WEEP is secure, the logic of Work Energy Equivalence is not threatened, .. for now ..

................



................

[Gregory]

Continuing the case for free lift conditions - approaching the problem and solution from slightly different perspectives to build a foundation that can stand up and be added to ..

Here is the same Leg 1 "Impulse" sim test with a slightly different arrangement - this time I have kept total masses the same etc etc but have added a fixed swinger and a swinger able to rotate CCW and be caught at its highest GPE gain ..

The COR / Elasticity of all the Body's is set to 1.0 ( effectively a perfectly elastic collision which results in no dissipative energy losses like my original Impulse / Momentum given to swinger carts a few pages back ) ..

Note from the previous sim results that in Leg 1 it is exactly as I would expect - the control and the fixed swinger comparison have the same momentum from start to pause - their Total Energy's are the sum of their KEt, which is also unchanging before and after collision ..

* The bottom Red test with the swinger able to "swing" after collision has the same total momentum from start to pause as per the tests above it - we see the Total KEt is reduced significantly after collision ( see graphed outputs ) - see how C1 continues moving at a reduced velocity because of the swingers inertia as it swings upwards - and that the swinger has gained significant GPE ( see black output meter ) - as anticipated when the Red test GPE energy is added to Total KEt the Total Energy is the same for all tests ..

** This is consistent with WD = Energy ( as per previous sims of Leg 1 ) - we expect this, WEEP is secure, the logic of Work Energy Equivalence is not threatened, .. for now ..

Hey Fletcher,

I like the idea of this passive GPE lift with C3. Made me thinker a bit. It's interesting...
So, momentum is conserved here as it should be, but we have gained some GPE instead of stuck with the usual KE. But the GPE + KE is the same as the initial energy, as usual.

Now, what's next?
GPE might be used to increase momentum. That's actually sounds like a possibility...
I get the impression of some special arrangement of things bent into a circular form... playing a kind of chain reaction where energy and momentum is conserved inside the sub-process, while the GPE gained is tasked to increase the total momentum of the system... which leads to an increase of both energy & momentum in essence. However, using up GPE will decrease the total energy of the system.

So, does that make sense? Is it possible somehow to produce an overall gain?
I wonder where this is going...

[Gregory]

destroying duplicate (post)

[Roxaway59]

As I have said before Fletcher I’m not sure about this or where you are taking it but I would like to add a notion that occurred to me.

When I relate this idea to my main wheel idea that you have helped me with many times there are certain things that fit.

Imagine these sliders working in the Z axis as with my wheel. Then imagine that they are not sliders but hammer men.

You and most people here know how my wheel is suppose to work and you will be aware that the real wheel I demonstrated actually works quite well and it has impacts on it.

My idea was based on the idea of raising GPE on the left of a clockwise rotating wheel which then sets a pendulum around 12 o’clock.

There is no reason though why a hammer men impact couldn’t impact a spring loaded mechanism so that a weight swings downward or upward (depending on the mechanism) as with your approach on the right, thereby having the same effect.

The impact would then be on the right and it could be argued that it was a better way of doing it because gravity might be helping to move the weight.

Its just a thought and if you do actually have a mechanism in mind that can work on a wheel I’m just wondering what form its going to take.

Graham

[Fletcher]

Continuing the case for free lift conditions ...

Hey Fletcher,

I like the idea of this passive GPE lift with C3. Made me thinker a bit. It's interesting...

So, momentum is conserved here as it should be, but we have gained some GPE instead of stuck with the usual KE. But the GPE + KE is the same as the initial energy, as usual.

Yes Greg, all is as Classical Physics predicts for Leg 1 ( the Pump Leg ) - there is no magic there, nor extra energy - that happens in Leg 2 ( the Dump Leg ) - but we can't get to Leg 2 without first passing thru Leg 1 i.e. we have to look at the cycle in totality, the complete Pump and Dump phases ( Legs 1 and 2 as one operation ) - then we can run some math and maybe get to give WEEP a black-eye .. see Page 9 of this thread for animations etc and the summary spreadsheet analysis of Momentum changes, and Work Done ( WD ) verses Energy ( KEt + GPE ) - note that WD is calculated on the distance the swinger cart itself has moved from and to where the force is applied ( not average swinger component distance moved ) .. n.b. see the reproduced animation below of the original Leg 1 Pump phase of the 2 part acceleration and deceleration cycle of giving momentum and taking it away again - it has the same result as the last slack rope and collision body sims with a COR of 1.0, because the impulse / momentum input is not a collision ..

Now, what's next?

Looking again at the complete Pump and Dump cycle animations and spreadsheet results to see if my WEEP workaround proposal is in theory plausible, and there is a mathematical advantage of free GPE gain that I propose falls out of the spreadsheets and sims ..

GPE might be used to increase momentum. That's actually sounds like a possibility...

Exactly Greg ! - if the complete cycle results in more Energy ( KEt + GPE ) than WD then that extra energy is in the form of GPE, and is technically "free" .. this is hypothetically usable to create ongoing wheel torque ( gravity force x horizontal distance from axle ) - this would increase wheel baseline wheel angular momentum cycle on cycle, and the excess-torque cost next to nothing ..

I get the impression of some special arrangement of things bent into a circular form... Yes .. playing a kind of chain reaction where energy and momentum is conserved inside the sub-process, while the GPE gained is tasked to increase the total momentum of the system... .. correct .. which leads to an increase of both energy & momentum in essence. .. right ..

However, using up GPE will decrease the total energy of the system.

Only if WEEP is always true - my sims suggest ( to me ) that there is a mechanical workaround to dismantle WEEP doctrine ( so it's not always true ) so that it is not conservative in my special arrangement of mechanical inputs ) ..

So, does that make sense? Is it possible somehow to produce an overall gain?

Follow the Energy ( Math ), and let's see if it makes sense and can we get an overall gain ( to me the Math is logical in the final analysis .. )

I wonder where this is going... forward ;7)

...................



...................

[Fletcher]

As I have said before Fletcher I’m not sure about this or where you are taking it but I would like to add a notion that occurred to me.

[ ... There is no reason though why a hammer men impact couldn’t impact a spring loaded mechanism so that a weight swings downward or upward (depending on the mechanism) as with your approach on the right, thereby having the same effect.

The impact would then be on the right and it could be argued that it was a better way of doing it because gravity might be helping to move the weight.] Short answer - collisions are VERY energy wasteful ..

Its just a thought and if you do actually have a mechanism in mind that can work on a wheel I’m just wondering what form its going to take.

Graham

Hey Graham .. here is an animation from page 9 of this thread - it is of a generic wheel with generic OOB generating one-way swingers ( except I can't reliably sim one-way bearings at this scale so they just flop about and you have to use your imagination ) - the Prime Mover in this instance is an external Pendulum connected by short crank to the wheel - it acts as both a Prime Mover to impart a first motion to the wheel carcass and also as an escapement analogue to both give and takeaway again an equal impulse ( technically a net zero momentum change ) - this will be replaced later in the thread with a MOI changer apparatus that does the same basic job as the external pendulum but is fully enclosed within the wheel and goes around with it, and many can occupy the internal area available n.b. I deduced this from the Toy's Page Hammermen toys in particular, and one other important evolution of that device imo found in MT .. hope this helps give you a clearer picture of the driver and driven ..

N.B. both alternative Prime Mover structures cause the wheel carcass to speed up, and slow down as can be seen in the somewhat volatile RED sine-wave graph plot of wheel rpm in the animation of the sim ..

.....................



.....................

[Roxaway59]

Hi Fletcher it reminds me of something I did a few weeks ago.

Turn the motor off and as so often happens in Algodoo it goes over unity.

You might have to set the speed to normal I think its moving at 30 rpm if memory serves.

Graham

[Gregory]

Only if WEEP is always true - my sims suggest ( to me ) that there is a mechanical workaround to dismantle WEEP doctrine ( so it's not always true ) so that it is not conservative in my special arrangement of mechanical inputs ) ..

Right. If work energy equivalence can be tricked, that's a great starting point for sure!
I'm not 100% convinced yet, guess I need to experiment with it myself. I might thinker around with those carts, and see what can I do with them.

Follow the Energy ( Math ), and let's see if it makes sense and can we get an overall gain ( to me the Math is logical in the final analysis .. )

I wonder where this is going... forward ;7)

Haha, right Fletch... That's a good direction! :-)

Hey Graham .. here is an animation from page 9 of this thread - it is of a generic wheel with generic OOB generating one-way swingers ( except I can't reliably sim one-way bearings at this scale so they just flop about and you have to use your imagination ) - the Prime Mover in this instance is an external Pendulum connected by short crank to the wheel - it acts as both a Prime Mover to impart a first motion to the wheel carcass and also as an escapement analogue to both give and takeaway again an equal impulse ( technically a net zero momentum change ) - this will be replaced later in the thread with a MOI changer apparatus that does the same basic job as the external pendulum but is fully enclosed within the wheel and goes around with it, and many can occupy the internal area available n.b. I deduced this from the Toy's Page Hammermen toys in particular, and one other important evolution of that device imo found in MT .. hope this helps give you a clearer picture of the driver and driven ..

N.B. both alternative Prime Mover structures cause the wheel carcass to speed up, and slow down as can be seen in the somewhat volatile RED sine-wave graph plot of wheel rpm in the animation of the sim ..

Years ago I had the impression and tried to explain that creating a controlled variation in speed can be an important concept.
The idea is that you need some kind of potential/usable energy, before you can start working with something.
For example hot air vs cold air, or high pressure vs low pressure, that kinda thing...
And in a macro-mechanical wheel environment, the most natural suspects might be velocity and acceleration. It's like there can be "a flow" manifested between a faster moving (hot) and a slower moving(cold) component or state of a system.

Now, that all might sound quite mystical or complicated, but for example to put it in a very simple way:
1. Variate speed of wheel body in a controlled manner
2. While freely exercising its own motion...
Let a passive inertial mass "react" to the speed variations, and for example drag a weight OOB or do something else.
3. Enjoy your free lift, and let the OOB do its job.
4. Beers and cigars!

Of course, it's not that simple. But still a good starting concept.
Certainly better, than trying to suggest some weights to lift themselves or go a little bit further this way or the other...
They never listen... just sitting there hungry and lazily, snacking some more from the bowl of spicy Momentum.

So, first in essence some environment/process needs to be established, before anything else can happen. That was my conclusion.
And after that maybe, just maybe... new or interesting movements/things/ideas might start to emerge.
I experimented a lot around these ideas, but could not make it work as I wanted so far. The devil is in the details as usual. Still it was great and I learned a lot, and then drifted away to different territories and concepts.

By the way, I developed a couple of different speed variator mechanisms. Some work via MOI change, others via spring potential energy storage. I attach some, it might help you in your research. Oh yeah... and a one-way clutch.
Hail the great inertial magic! Let it flow through all our humble wheel carcass! :)

[Kattla]

Can wm2d do hydrostatic pressure and weight simulations?