Into the Vanishing Point..

A Bessler, gravity, free-energy free-for-all. Registered users can upload files, conduct polls, and more...

Moderator: scott

Post Reply
User avatar
Fletcher
Addict
Addict
Posts: 8200
Joined: Wed Nov 05, 2003 9:03 am
Location: NZ

re: Into the Vanishing Point..

Post by Fletcher »

Then that rationale might equally apply to the coupled stampers actually seen, and the never observed pendulums.

IOW's there is a Goldilocks zone for optimal output.
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Re: re: Into the Vanishing Point..

Post by MrVibrating »

Fletcher wrote:hmm .. so correct me if I'm off the reservation here.

The system total momentum (thru your very specific range of translation conditions) forms a quadratic parabola shape loosely like an n, waxing and then waning to a datum.

But the currency for doing Work is Energy and not momentum (as currently defined in physics) - so does the energy curve show the same peculiarities albeit a taller thinner curve ?

If there is a specific rise in system energy by specific translation conditions then would not the cost of obtaining that translation against cf (the reset) be the expense side of the ledger ? And if the system energy spike is greater than the energy expenditure would not this be the gross revenue side of the P & L ?

If there is a Net Surplus of energy after costs then (in theory) energy symmetry from momentum induction has been theoretically broken ?

All that would remain would be to engineer efficient mechanics to aid the translation conditions that give a theoretical surplus of energy and momentum ?

Am I visualizing this correctly ? Thanks.

Honestly mate, i've no idea how or where an energy anomaly may arise here - nothing specific, anyway.

What i'm hoping is that thoughts of energy are "greedy" and a shift of focus to momentum may pay dividends by other means..

When i say i've no idea how, that's not entirely true - because of CoE's dependence on N3. The way energy asymmetries evolve from N3 breaks are usually non-intuitive (incidentally there's been some evidence supporting the view that an implicit presumption of N3 is hardwired into us and in evidence at a few months of age, so maybe that's part of our handicap here), but the general principle is that once you've busted N3, pretty much anything else you do is likely to create energy.

An effective momentum asymmetry is a gravy train for pocketing free energy. It's a free launching platform for further accelerations, giving us a free leg-up to higher-value momentum than we've paid for.

And so this is why i was manically plotting up the unit energy cost of momentum for the 10:1 config on the previous page - the energy value of momentum is highly non-linear in this interaction, varying from cheap to expensive.

This non-linearity in the energy value of momentum is specifically interesting in relation to the 'gold standard' metric of 1/2mV^2. The unit energy value of momentum usually accumulates as the half-square of rising velocity, and so any other metric - which may be perfectly valid within its own context - but which varies from the usual rigorously-enforced half-square exponent, could represent a potential energy gradient.

For now though, my immediate goal is to see if it's possible to simply cause a closed-loop change in angular momentum, from an otherwise elastic system, at any energy cost, since CoM is not energy-dependent; if we can buy in more of the stuff without recourse to external reaction mass then we have a breach in CoAM, and a major hurdle down.

Other points to keep in mind are the potential for variable-MoI flywheels, which get fatter instead of faster, converging up to a finite size determined by the juxtaposition of Hooke's law for springs and CF as a function of mr^2 (demonstrated in my previous thread), and so offering the potential of storing PE rather than just KE, limiting the rise in forces for any other CF-critical interactions in other layers of mechanism, but also, the potential role/s for gravity and GPE interactions with regards to your question - so far i don't even have gravity enabled, yet vertical operation seemed integral to Bessler's success, so, since we both know G, M and H are constants and that closed-loops thru static fields yield zero energy - ie. no GPE asymmetry is possible - then we must look to juggling GPE for alternative reasons... if it's a necessity, then it must be a critical step in generating or harnessing an energy asymmetry. GPE inputs and outputs are of course another means of modulating momentum and inertia.

With regards to the usual tail-chasing problem of CF rising with putative RKE gains, another potential angle could be MoI variations aboard counter-spun angular momenta - ie. consider two angular inertias rotating at uniform RPM about a common axis; apply an impulse between them as from a rotary spring or motor, further accelerating one, while decelerating the other... then perform an MoI variation (ie. radial translation) upon the decelerated (potentially even stationary) 'counter momentum', before re-colliding it with the unadulterated 'primary' momentum...

Yet another angle might be the fact that a mass could potentially find its way to the center of a rotating system by various means - for instance, a paternoster (per MT 48 etc.) could raise weights into the middle while bypassing the main rotation, there to be dropped out into the wheel. The momentum plots already show that we get these transient boosts in momentum regardless of whether we're entering or leaving the center via radial motion - so pulling masses in against CF might not be the only option - i'm doing it this way for now simply as a consistent set of starting conditions for a test battery.

So this is all totally speculative and i'm merely prospecting, at best. As mentioned previously, an energy gain doesn't necessarily come gift-wrapped in momentum gains, too - yet Bessler's wheels were constantly offloading momentum to their applied loads along with its excess RKE - ie. tap off RKE without generating more momentum and you'd grind to a halt. So the hope here is that maybe an energy exploit will become clear if we can figure out how to harness closed-loop momentum gains... even if they're expensive. Because whatever the cost, the usual metric is 1/2mV^2 so if we can trade by any other exchange rate, there's likely to be a breakeven threshold of velocity beyond which cost < benefit..

For now tho i'm just going with my gut... and it's saying enough with the energy already, maybe try following the momentum..
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Re: re: Into the Vanishing Point..

Post by MrVibrating »

Fletcher wrote:Then that rationale might equally apply to the coupled stampers actually seen, and the never observed pendulums.

IOW's there is a Goldilocks zone for optimal output.
For momentum variation, yes.

Too much non-radially translating inertia, and there's no variation in system momentum. Remove the NRT inertia entirely and there's no momentum variation.

So there's definitely an upper limit to the ratio, beyond which the translating mass is actually undergoing angular deceleration all the way in, shedding momentum at constant rate, while the non-translating mass is gaining at equal rate, and so keeping the net momentum constant. Below that threshold ratio, the inbound mass first accelerates, likewise accelerating the fixed mass, and so raising the net momentum, before shedding it again as it spirals into the center.

But whatever the degree of variation, the starting momentum, when both masses are out at the rim, and the ending momentum when one's at the rim and the other's been drawn into the center, are equal. If we begin with 10 kg-m/s, then no matter how high it goes while performing the translation, it always dives back down to its initial value as the sliding mass enters the center, albeit at a now-greater velocity.

So again, these starting and ending states, possessing identical system momentum but significantly different energies, represent a variation in the energy value of momentum. Granted it ain't free, yet, but these are the kinds of landmarks by which we can at least map out the territory, in the search for ways to exploit it..

Bessler's weights alternated inner/outer positions, after all, so would almost certainly be subject to these dynamics to some extent, and as noted, the question of how his wheels generated momentum may be considered a separate issue to the energy gains, since for example, if we could somehow pull an orbiting mass inwards for free, this would raise its energy without raising its momentum - and tapping that kind of gain over successive cycles just drains the finite system momentum. So yeah, don't really know what i'm doing, it's just something i need to do, hopefully something more concrete will fall out of it..
User avatar
Fletcher
Addict
Addict
Posts: 8200
Joined: Wed Nov 05, 2003 9:03 am
Location: NZ

re: Into the Vanishing Point..

Post by Fletcher »

Well you've got the tenacity and smarts to reach down and pull the rabbit from the hole, if anyone does.

So keep having at it.
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

Itching to try yours and Greg's solutions for metering the net momentum. Might leave it til the w/e now, but once set up i should be able to rip thru a bunch of tests in short order, establishing the precise limits for the effect, gain margins within those limits, and ultimately the KE costs of those momentum gains.

Presumably after all this, the KE / momentum value will still prove to be 1/2mV^2 dependent. If not, then we may already have a possible PE gradient. But i don't expect that, at this stage. I think it's more likely we'll need a further trick or two before we can start consolidating closed-loop momentum gains, and it's only then that we'll start to qualify for concessions, by riding the coat tails of the diverging system momentum, in relation to the static frame of gravity and GPE.

The only exciting discovery, for now, is this transient momentum rise in the absence of a corresponding counter-momentum - all else being equal, ie. following the same path back out, we will encounter a counter-momentum, but usually counter-momentums are induced at the same time, hand in hand with a primary momentum, whereas here, they're separated in time, on alternate strokes of the interaction, opening the possibility of reneging on or changing the terms of a momentum transaction mid-way thru the deal..

The mere presence of a reactionless momentum variation seems an intriguing curiosity in its own right, quite aside from the potential exploits it may or may not present. When pulling anything at all out of the hat would be something, this looks a lot like a pair of bunny ears... so even if there's no bunny on the end of 'em, it at least seems a step in the right direction..
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

Another couple of points that seem worth airing again at this stage, are that this transient momentum rise is caused by conservation of angular momentum; the fixed-radius mass is being accelerated, and thus its momentum raised, by the very act of CoAM attempting to hold the momentum of the inbound mass constant, by raising its speed to compensate its reducing MoI.

So we're relying on CoAM to do exactly what it's supposed to do, yet in that process, causing the exact opposite result - essentially we've commandeered a key conservation law to work for us, betraying its own cause.

That's at once kinda cool in its own right, but it's also tentative encouragement that a successful OU result is going to have to depend upon the conservation laws holding exactly as they're supposed to. Every single step in an OU interaction has to be fully conservative under its own terms and conditions - we can expect no excess force or energy to materialise ex nihilo..

But here, we're just playing the game, going along with the rules. And it's a trivial thing - a torque's being applied to a mass, so it's accelerating, increasing its momentum. Nothing doing. But it's also reactionless, in the first instance - the instantaneous counter-torque is zero. The instantaneous reaction is a radial force, CF / CP, instead of an angular one of opposing sign. Everything's as it should be, there's no anomaly... yet we've coaxed CoAM to stick its neck out, pushing the boat out with this temporary advance of momentum - all bought and paid for - but sans counter-momentum. A mere gentleman's agreement, compared to the usually-binding terms of instantaneous counter-momentum.. It's a closed-system rise in momentum, caused by conservation of momentum itself, attempting to prevent just such a phenomenon. So whether or not it's really over-extended itself depends upon what we can next wrangle out of the small print.. fully consistently with all laws and regulations, naturally...


The other point that ties into this, is the ever-arcane esoterics of Noether's theorem.. we already know that in the system we're looking for, energy is not the conserved quantity, and that its Lagrangian is asymmetric with respect to the gravity vector, so we're likely blinkering our prospects by trying to analyse in terms of force / distance integrals (and i'm the worst one for this) - what we're obviously looking for is a differential - a field that has a time rate of change factor, with properties of inductance and capacitance and hysteresis.. basic logic says the name of our game has to be momentum. That's our 'conserved' field...

...and here we have an example interaction where conservation of momentum is the very cause of further momentum induction... so if we can close-loop momentum gains over repeated cycles, we'll end up with no more energy than we should have.. given the runaway inertial frame.

Energy schmenergy. It's putting the cart before the horse..
User avatar
Gregory
Aficionado
Aficionado
Posts: 566
Joined: Sat Sep 23, 2006 10:33 pm
Location: Europe

Post by Gregory »

So we're relying on CoAM to do exactly what it's supposed to do, yet in that process, causing the exact opposite result - essentially we've commandeered a key conservation law to work for us, betraying its own cause.

That's at once kinda cool in its own right, but it's also tentative encouragement that a successful OU result is going to have to depend upon the conservation laws holding exactly as they're supposed to. Every single step in an OU interaction has to be fully conservative under its own terms and conditions - we can expect no excess force or energy to materialise ex nihilo...
Exactly, I had similar conclusions. If you somehow can change symmetry for asymmetry, then possibly you can also escape some of the consequences of that original symmetry...
The other point that ties into this, is the ever-arcane esoterics of Noether's theorem.. we already know that in the system we're looking for, energy is not the conserved quantity, and that its Lagrangian is asymmetric with respect to the gravity vector, so we're likely blinkering our prospects by trying to analyse in terms of force / distance integrals (and i'm the worst one for this) - what we're obviously looking for is a differential - a field that has a time rate of change factor, with properties of inductance and capacitance and hysteresis.. basic logic says the name of our game has to be momentum. That's our 'conserved' field...
If a wheel is genuine and woks by creating asymmetris, then Noether's theorem would not apply to it.
Are you sure that momentum can stay constant while the wheel is rotating faster and faster? Both momentum and kinetic energy depends on m and v. My personal feeling is that if you end up changing one in a repetitive manner, then most likely you'll end up changing the other as well.
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

Been procrastinating... doubtless trying to delay the inevitable and keep some kind of suspense going.. Pitiful eh, but still tinkering with things slowly..

I've been looking out for any prominent instances of the number five here, as i've long been of the opinion that this is precisely the kind of place to look for one... and over the last week i've noticed a possible example:

- when we do this using two equal masses - one in, one out - the net momentum rises by a fifth.

For instance if we begin with both point masses at 1 kg-m/s, then the total system momentum is 2 kg-m/s.

If we then draw one mass into the center, en route, the net system momentum will converge towards a gain ratio of 2 to 0.4, and .4 goes into 2 five times, IE. momentum rises by precisely one fifth when using two equal masses this way. We begin and end the interaction with 2 kg-m/s, but in between we hit this plateau of 2.4 kg-m/s.


As such, if these transient momentum rises could be rectified somehow (via collisions perhaps), then we'd need five of them per cycle to double our momentum per cycle...

Rank numerology. But hey Bessler started it..
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

...Just realised an obvious reason why Bessler's system may have needed to remain statorless & with everything going around together, if it was based on these kinds of principles - it would basicaly be rotating mass and reaction mass. If the reaction mass was acting in the opposite direction of rotation, then the two momentums would be of opposite sign and so subtract from one another. Whereas, if they're both, say, clockwise, then they sum..

Take the 10:1 config for example - 10 kg pulled in 10 meters from 1 RPM, while 1 kg remains at 10 m fixed; initial system momentum is 11 kg-m/s, rising to a peak of 25 kg-m/s at the 2.3 meter mark, then plunging back down to 11 kg-m/s over the final 2.3 meters.

So applying a casual assumption of N3, if we collide that 25 kg-m/s peak with some other angular inertia, it'll divide itself out equally between the two MoI's, with 12.5 kg-m/s shared between the orbiting masses, and another 12.5 on whatever they've collided with.

Since momentum rose 2.27x at the 2.3 meter mark (from 11 up to 25 kg-m/s), if we apply that same multiplier to the remaining 12.5 kg-m/s, it should reduce by 2.27x as we reset the sliding mass back out to the rim, or equally, pull it all the way in to the center, leaving 5.5 kg-m/s final on the orbiting masses, plus the 12.5 we have on the 'reaction' mass we collided with, for 18 kg-m/s total remaining momentum... 1.6x more than the 11 kg-m/s we began with..!

So that seems like a nice closed loop gain of momentum... but if however the colliding masses were of equal opposite sign, we'd be subtracting that 5.5 kg-m/s from the 12.5 kg-m/s reaction momentum, leaving us with just 7 kg-m/s net remaining, a loss of 4 kg-m/s over our starting amount of 11 kg-m/s.

So in short, keeping both momentums to the same sign causes them to add, while opposing signs subtract - if the above casual application of N3 is valid, of course. Will be interesting to see what actually falls out...
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

I'd been trying to harvest this apparent momentum gain via a collision - spin up an orbiting mass by pulling it inward (ice skater effect), causing a gain in momentum as measured by its linear terms (rest mass times velocity), itself caused by CoAM attempting to keep the system momentum constant as a function of MoI times RPM..

And that's the key distinction here - angular or linear, speed is just speed, rate of change of location WRT time. But MoI and linear inertia / rest mass aren't quite so interchangeable; the former varies as a function of radius, whereas rest mass is constant...

And so, perhaps unsurprisingly, my attempts to consolidate this apparent momentum gain got off on slightly the wrong foot..

With the 10:1 config (10 kg translating radially, 1 kg at fixed radius), the summed X+Y momentum begins and ends at 11 kg-m/s. In-between though, it rises to a peak of 24.829 kg-m/s, at the 2.3 meter radius mark.

So my hope was that this momentum gain could be creamed off by colliding it with something else. To that end, i paused the sim at the peak momentum point, and placed a stationary target mass in the path of the fixed-radius mass. Gravity and losses remain disabled (ie. perfectly elastic collisions).

I observed the following results:

- if the target mass was below 1.530 kg, the armature 'drives through' when striking it, applying momentum to it, but without giving up all of its own momentum, resulting in both remaining momenta being of equal sign, and thus a rise in system momentum, at no net cost (ie. net system KE does not change) - however it must be noted that the system isn't closed, since this static 'target' mass was added, so this principle can only add momentum in a closed system if something else causes a component within it to decelerate relative to the rotating system. Obviously, gravity would seem to be one potential means of attempting this, and MT134/135 could be relevant here (the hammer-levers droop under gravity)..

- if the target mass is greater than 1.530 kg, the armature bounces off of it in the opposite direction; momenta of opposite sign subtract, so net momentum has decreased, with energy again conserved.

- if the target mass is exactly 1.530 however, we get a 'Newton's cradle' type effect; the armature is brought to a full stop by the collision, and thus all of the system momentum is transferred over to the target mass.

In this final scenario, the final net system momentum - all of it represented by the target mass, being the only moving mass in the system - was just 11 kg-m/s... precisely what it began with.

I'd been hoping that instead it would be the 24.829 kg-m/s figure, since that's how much X+Y momentum it actually had prior to the collision.

So the gain appeared to be illusory! Or was it? Maths can't lie, they can only be misinterpreted...

What i was yielding from the collision was momentum as a function of MoI times RPM. But what i was measuring was momentum as a function of rest mass times velocity.

So to actually get at the linear quantity being indicated, we need to break from the angular term of MoI - not just in the maths, but physically, in the system itself..

And so it turns out that the way to create momentum from within a closed system is simply to pull your orbiting mass inwards under centripetal force, which accelerates it precisely by conserving its angular momentum... and then to simply let go of it - sever the CP constraint and let the mass fly off in whatever direction it likes. We don't even need any non-radially translating mass at fixed radius - just pull all of the system mass inwards, if you like, it matters not, angular momentum is conserved, but in so doing, linear X+Y momentum increases.. and can be harvested by simply letting go of it, flinging it away, somewhere. Obviously though, what's flung up must fling down - gravity reverses the sign of momentum so offering a potential trick for keeping it all of the same sign (ie. reversing the sign of reaction momentum).

In a nutshell though, the principle reduces to a freeze-frame difference: in the moment prior to being released, the orbiting mass's momentum remains a function of its MoI times angular velocity. In the next moment upon release however, its momentum becomes a function of its rest mass times its velocity! It's that simple - pull it inwards then fling it upwards; let gravity pull it back down and you've created a load of momentum from within an otherwise closed system.

Dunno how useful this is yet - turning it into a closed-loop gain of angular momentum would seem cool, if possible. But also need to keep an eye on unit energy cost of momentum bought this way, to check for any variation from 1/2mV^2..

So still tinkering along, incremental steps, no brick walls yet. The overall aim of closed-loop momentum gains remains on-mission...
User avatar
Grimer
Addict
Addict
Posts: 5280
Joined: Tue Apr 14, 2009 9:46 am
Location: Harrow, England
Contact:

Re: re: Into the Vanishing Point..

Post by Grimer »

Fletcher wrote:Well you've got the tenacity and smarts to reach down and pull the rabbit from the hole, if anyone does.

So keep having at it.
He's certainly got the tenacity. :-)
Who is she that cometh forth as the morning rising, fair as the moon, bright as the sun, terribilis ut castrorum acies ordinata?
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

Really wish i did mate, could so use a month off work... but short of a broken leg there's little chance of it..

Pulling all these conclusions together though, now that i'm clearer on what's happening, i realise there may yet be use for the 'Robernoster' concept..

Previously i was pulling mass inwards on the curved sections to try accelerate and thus raise the linear momentum of masses traveling along the straight sections.

However the whole concept of masses acting in pairs, accelerating / decelerating one another - the whole 'inertial drag' thing i was reaching for - now appears to have been misguided and superfluous to requirements anyway, since it would seem that only a single mass is necessary; with a curved-into-linear trajectory, we can harvest a rise in linear momentum, from an acceleration applied by conservation of angular momentum.. all acting upon the same individual mass.

In the above experiments, it was found that peak linear (X + Y) momentum, induced in the form of purely angular motion via radial translation / MoI reduction, corresponded to a ~77% reduction in radius - when the inbound mass was at ~23% of its original distance from the axis.

And when released to fly off in a straight line at this point, it carries with it that rise in linear momentum.

However in my misconceived attempts at this result from the Robernoster concept, i was pulling the inbound mass all of the way in to the center... whereas the above experiments show that the transient rise in linear momentum disappears again as the mass is drawn further in to the center, ending up with the same amount it began with.

In other words not only should i have been trying to harvest the momentum rise from the inbound mass itself, not from some other mass accelerated by it, but i also i pulled the mass in too far...

...so the obvious thing to try now is a Robernoster track with a single mass and armature; pull the mass inwards to 23% radius on the curved section, timed so that it reaches peak X + Y momentum just as it clears the curve coming back onto the straight.

If the system behaves the same as the 'slingshot' effect described previously, then this transient rise in linear momentum should be preserved on the mass as it travels onto and along the straight... Then, as per the previous plan, its 'radius' can be raised again without incurring angular inertial forces. As it comes back onto the next curve, it's now at higher velocity / momentum and radius, and a repeat cycle should be able to double those gains..
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

Post by MrVibrating »

I tested this conjecture this evening, with success.

The setup is a Robernoster with single arm and sliding mass. The mass is pulled in halfway while rotating around the curve, arriving onto the straight with a 1.8x rise in momentum.

The mass can then be re-extended while traveling along the straight, and so not causing a deceleration, to be pulled in again on a subsequent curve.. ie. it can be cycled, gaining more momentum per cycle.

It's such a simple revision - pull the mass all the way in while it's rotating and the transient momentum rise is over by the time the mass reaches the straight. There was a rise in linear momentum (rest mass times velocity), caused by conservation of angular momentum (MoI times RPM), but it was all over and done with while still rotating, so arriving onto the straight with no more momentum than it began with.

But pull the mass only halfway in, and the mass enters the straight with 1.8x more momentum than it began with. Reliably, repeatably. It's incredibly simple. CoAM is causing a rise in momentum.

Because of this result, i've gone back to re-examine the mirrored Robernoster inertial motor idea - two Robernoster tracks side-by-side, one having a clockwise armature, the other counter-clockwise, in sync, masses pulled in at one end, accelerating the whole system linearly via centripetal force, re-extending on the straights - basically exactly as the single version, but in mirror symmetry... with both tracks connected to the same base, and that base is free floating...

I suspect that previously it didn't work for the same reasons, of pulling the mass in too far, but with the revised technique, the original hypothesis still stands - if we can gain momentum around a closed loop, then two such loops in mirror symmetry should produce an uncancelled, unilateral, force..

So that's my next goal - to try and gain linear momentum in a closed system. Cuz if that works, then OU is almost a triviality..
User avatar
Fletcher
Addict
Addict
Posts: 8200
Joined: Wed Nov 05, 2003 9:03 am
Location: NZ

re: Into the Vanishing Point..

Post by Fletcher »

I find myself holding my breath with you MrV.

Looking forward to future installments.
MrVibrating
Addict
Addict
Posts: 2875
Joined: Sat Jul 31, 2010 12:19 am
Location: W3

re: Into the Vanishing Point..

Post by MrVibrating »

Image

Ok so here's a working Robernoster concept.

The bob weighs 1 kg. Everything else (armature, slot & track etc.) is notionally massless. Thus the bob represents all of the system mass.

As you can see, net momentum rises, and can keep doing so indefinitely.

The bob begins with a small amount of motion (0.1 m/s). From thereon, it is pulled inwards while rounding the bends, and pushed back outwards on the straights.

Pulling the bob in while it's rotating causes an inertial torque, accelerating it due to conservation of angular momentum (CoAM).

During this acceleration, the bob's momentum is a function of its angular inertia times its angular velocity, and remains constant.

Indeed, it's accelerating because its momentum is not rising.

However when it arrives at the straights, these angular terms no longer apply and its momentum is simply its rest mass times its velocity. Which has increased.

And so while still carrying this rise in momentum, the bob is extended again but while traveling along the straight, and so not incurring a reciprocal angular deceleration.

This cycle is open-ended and can be repeated any number of times, limited only by how fast you wanna go.

And so in conclusion, it is possible to rectify inertial torques of one sign only (could be positive or negative), by pumping an orbiting mass in and out, varying its radius, while traveling in a closed loop trajectory.


Just to re-iterate, at no time is any other torque or artificial acceleration being applied. The only forces here are inertial and radial (no gravity or friction). The only source of torque is CoAM, and input energy via these radial excursions.

In other words, some input energy + conservation of angular momentum, directly causes a rise in both angular and linear momenta.

Notably, when we're inputting our energy, momentum is not increasing. The increase only arises spontaneously after we've input some energy, when the momentum terms suddenly switch from angular to linear, at the straight flanks; all of the input energy is applied within a 90° arc between 45° and 135° of the rotations, so the rise in momentum really is temporally isolated from the rise in energy - technically, we're not inputting more momentum, only energy. Only after energy input is complete do the momentum terms switch, causing the momentum rise to manifest.


Dead-simple system, and not particularly surprising to watch in action... yet could arguably win you a bet with any level-headed physics bods..

Still, took me two months to get it working as intended..!

For anyone interested, the technique used here is from this YT vid "Driving an Actuator Using Table Data". The sim and table are enclosed below.


____

ETA:

..in retrospect, to emphasise the above points it would've helped to plot both angular as well as linear momenta, to see how they interpolate - for now tho, remember that the rise in momentum being plotted during the input strokes is linear (rest mass times velocity), not angular (MoI times RPM), which is not rising during the input strokes - only materialising after the bob re-extends on the straights and re-enters the next curve; at which point the angular momentum does, spontaneously, increase.

In summary, both the angular and linear momentum rises are spontaneous, corresponding to the moments the momentum terms switch, and not the intervals during which energy is being input. So we're only generating these momentum rises after the fact - they're made incarnate by an instantaneous change in their respective terms, almost as if the momentum is being plucked from the aether.. at no point is any momentum raised directly by the application of torque.

And remember, these torques are reactionless - the only applied forces are radial, not angular, so there is no counter-torque being applied back to the axis.

At this stage, energy is still evolving via the half-square of velocity, so no energy gains yet.

I also spent the last week trying to take this the next step to an inertial motor, without success, unless anyone needs a mechanism that wobbles quite robustly. I'll post up those failures in due course, but for now i'm counting this as a minor win..
Attachments
RN_single_1.txt
(128 Bytes) Downloaded 53 times
RNS1.wm2d
(29.9 KiB) Downloaded 53 times
Last edited by MrVibrating on Tue Apr 11, 2017 1:40 am, edited 3 times in total.
Post Reply