MTM5

[Tarsier79]

I would like to Fletcher, but there are a few things I have to wrap my head around first. My first problem is time. I have lots of small chunks of it at the moment and I think I need a large chunk to wrap my head around this build......It is making me feel a bit stupid actually.

My concern over the energy calculations (or maybe it could be an advantage) was the change in reference frame. IE, is it giving a false positive. (I guess sort of like when a wheel rolls around another equal wheel, but turns an extra time through one rotation)...

[preoccupied]

Is it true that when driving the motion it has less torque than when it is in resistance of motion? If it's in continuous motion would it slow down considerably when in resistance? The distance from the spinning weights axle to their main axle is much longer than the weights axle. Usually this will add more resistance when there is resistance (I think). If it weren't free flying and were running on a track using gears it would definitely lock up mechanically. Unless possibly the positive motion doesn't have any resistance only, less positive.

[MrVibrating]

I do mean it when i say any working wheels have to be built as synced pairs though.. Watch the float test result from a few pages back if you don't believe me..

Hey MrV .. I was thinking about that the other day - with the un-mirrored setup i.e. your original 'float'-test .. the big wheel is anchored by the central motor (regardless of FOR) i.e. motors act as pin joints to the background as well as motors IINM .. could you attach it say to a free-in-space (floating) backboard or some such (gravity off) ..

I was thinking that if there were stray torques the whole shebang would jostle about at the very least I would imagine and in some FOR we should detect that ?

That's what this float test is mate:


I've never seen this effect before. Ever. And i have checked other systems this way.

This remains the single most challenging result in this thread.

It doesn't just 'wobble the planet (in the same way any moving mass would), but forcibly accelerates it. Both CoM and CoAM are defeated.

The rig is set to halt while the sim's still running, in order to demonstrate that this change in motion state does not end with operation of the system, but persists permanently.

Hence all implementations for terrestrial use have to mutually cancel their own stray forces. Or else, energy has to be generated in orbit then beamed back down via MW lasers or whatevs.

It also means however that we can take active control over climate - basically remedial geoengineering - rather than just being passive passengers at the mercy of Kepler and Newton. By accurately plotting our state of motion relative to the fixed stars, we can apply carefully controlled changes to all six degrees of freedom of a planet. Suffice to say, if this thing goes wild as single (dirty) systems, we may have no choice but to try and monitor and correct any such momentum pollution.

Obviously better to try and forestall the interplanetary billiards as long as possible though. Everyone considering a build needs to watch that sim above and fully absorb what it means..

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

I think T79 is beginning to think of practical ways to implement a build to test out the hypothesis in the real-world .. which would sooner or later become a requirement to move from theory to fact over and above a bound to be argued about sim result ..

A month ago i had a half-Joule anomaly few folks would even notice. I switched kiiking techniques and the trickle became a flood. Immediately after confirming OU i checked for stray forces (page 2) - and a week later we're onto self-cancelling dual systems. A wheel became a pair of spindly robot arms. I've no idea where this is going next, but i'm not hanging about - i think physical modelling's still premature at this stage, unless someone was particularly bored, keen and sufficiently well-equipped already that they don't need to go spend a packet on a BoM..

..really, given that it's mainly an exercise in controller theory in its current form, i suspect it would be a challenge to physically replicate - i'd be looking at stepper motors for spinning the weight, and probably just a rotary spring for harnessing, or else a torque meter perhaps.. trying to replicate an empirical gain, rather than trying to loop output to input - in principle the latest design could be built as a pair of self-powered robot arms, but that'd be a long-term engineering project for someone like me at any rate.

What we want is a simple mechanical experiment with clear inputs and outputs. In its current guise you can see it's basically a kind of flailing motion - the flailing part has to be a free axis, but whereas a simple flail relies on the whiplash effect (conservation of the momentum wave propagating through a medium with a diminishing cross-section trades wavelength for amplitude, accelerating it), we need to actively control the flailing action by spinning and de-spinning the weight.

This spinning and de-spinning is where the angular momentum is gained - it causes the F*t asymmetry which causes the CoM break and resulting CoE break - but in the current design relies on being able to control these spin phases, forwards of this free axis required for inertial isolation of the spinning weight. So to me, it currently looks like a rigid armature with a floppy flail on the end, and a motor on the end of that flailing part. Feck trying to build and control that - remember, touch a divergent inertial frame, and you inertially interact with it, killing it. So you can't have a heavy bundle of wires snaking across that joint, it'll compromise its effectiveness.

I think the most anyone could do to get involved is try and take on board this seemingly-crazy theory-talk i've been spouting all these years, because this really is the physical manifestation of it. It's doing what it says on the tin. The real breakthrough here is that we're cultivating and harnessing without breaking inertial isolation - this the only reason we're even getting to glimpse such results. Understand the fundamentally circumstantial reasons energy usually squares with velocity, and you already have a handle on the implicit conditions of an exploit. There's more than one way to skin this cat..

For now i just wanna play with it, familiarise myself with its characteristics and how it reacts and scales to different variables. And i'm honestly struggling to believe it myself - you can see how surprised i was when the big gain popped out, i honestly thought i was chasing a piddling effect - but even in its current form; you mean to tell me those robot arms are producing free energy, and that if one's switched off the other will propel the planet? Please. This whole thing seems insane, fantastical nonsense.. but then there's the data, and the theory, and Bessler's prior success..

There's being out of your depth.. and then there's this.

[MrVibrating]

Just float-tested the last sim, running one arm for 10 secs then halting:



..it actually doesn't look so bad in this version, not sure why. The 'planet' here is just 5e9 kg, light enough to show an effect, but the previous sim showed both angular and linear remnants.

Then again, notice how in the first example there was no y component? That seems a little mysterious given the asynchronous mechanism - stray counter-torque would be one thing, but why would the stray linear component be precisely oriented to the sim's x-plane? That can't be a realistic stray momentum?

And then why in this one is linear momentum conserved? Is that residual angular component even realistic? Am i doubling up for no reason?

The thing is, the first one continually accelerates while running - the longer it runs, the greater the remnant momenta. Except only angular and x-plane accelerations.

More worryingly, if we can't trust this result, how reliable are any of the others?

[MrVibrating]

Logically, the only OU actions involved are on the green and blue axes, the spinning weight and its free-wheeling rotor. So if there's a momentum leak, that's where it'll be coming from.

Thus any net momentum delta applied to the planet should be proportionate to the energy gain.

It should only be angular, not linear as far as i can see.

You take the energy gain, invert it by the MoI of the OU body to solve for a velocity, then multiply that by the MoI to get a momentum delta. Any stray momentum applied to the planet should thus be that quantity, solving to a tiny velocity delta when divided into the MoI of the planet.

There is a small angular delta in that last float test above, so in principle it should be possible to show correlation if there is causation..

I'd guess this would need doing in a specially-designed test, with a focus on momentum data quality rather than adding it in as an afterthought like this..

[MrVibrating]

Doubling up does seem to mitigate as you'd expect:

..i'm not sure it isn't really redundant tho.

Am i just being melodramatic about wandering earth scenario?

And yet CoE depends on CoM.

But then again we don't wanna be building everything twice unnecessarily.

Might be a lot of fuss about nothing, which would be for the best.. Should probably just go back to single rotors, maybe float test any design proposals to be sure..

Looks like the reactionless thrust's off the menu anyway..

[JUBAT]

It looks like a flail with a rotating eccentric on the end of it. Mechanically, if these were ratchet assemblies so that everything could only ever go in one direction, it seems to me eventually it would just come into its own rhythm and always flail at the appropriate time.

[MrVibrating]

It looks like a flail with a rotating eccentric on the end of it. Mechanically, if these were ratchet assemblies so that everything could only ever go in one direction, it seems to me eventually it would just come into its own rhythm and always flail at the appropriate time.

Perhaps; the tricky part would be that the weight spin cannot be driven by applying torque against the ground reference frame, since this will ground it - there needs to be a free axis upon which to accumulate our cut-price momentum, and nothing can interact by direct contact with that axis.. The previous sim when i tried using radially-sliding weights instead shows what happens when there's any kind of grounding between the part we want to propel to OU and the outside world. There's obviously an energy gradient here however and i can't believe there'll be any others so if this is the one Bessler was accessing, there is a simple way to produce and harness the effect. Things can only get clearer with time - for now, we appear to have a tiger by the tail..

[MrVibrating]

For anyone interested in replication / confirmation, here's another sample interaction spit into two halves:

1st 180°:
initial KE = 3.59286114
final KE = 200.66404547
dKE = +197.07118433
kP*t = 14.98046635
mP*t = 171.8808745
net in = 186.86134085
anomaly = +10.20984348
CoP 1.054

2nd 180°:
initial KE = 200.66404547
final KE = 13.75113189
dKE = -186.91291358
kP*t = 6.751573475
mP*t = -203.8325154
net in = 193.664487055
anomaly = +10.168028345
CoP 1.053

..this should make clear the options for what's available, in what form, and when. The gist is that we appear to already have the gain in the form of KE halfway through the interaction.

That said, there must be a CF-PE component involved, which i've not shown there - over a complete interaction the net CF-PE is obviously zero so i haven't considered it a priority metric for now. A straightforward application of mass*radius*angular velocity² however implies the above weight undergoes an I/O of 40 J of CF-PE over the cycle. To further complicate this matter however, we're spinning the weight while descending for the precise purpose of braking its fall against CF force, preventing its passive acceleration - that CF-PE component is thus not converting freely to linear / radial KE, but must instead be conserved and tied up in the rotKE of the spinning weight, presumably, and converted back during the lift phase..

And throughout, the wheel motor is resisting the angular accelerations and decelerations from these MoI variations on the wheel as the weight changes radius, so this too is a potential source and sink for CF-PE.

So the above accounting's still not exhaustive, or basically 'complete'.. but it's all we have for now so make of it what you will..


ETA: and for further reference, here's the identical interaction from above, but measured as one full unbroken cycle:

initial KE = 3.59286114
final KE = 13.75113189
dKE = +10.15827075
kP*t = 21.73203938
mP*t = -31.94919639
net in = 21.73203938
anomaly = +42.10746714
CoP 1.94

..calculating the CoP midway is obviously misleading, and the final gain is actually this 42 J figure.. Which is close to the currently-omitted 40 J CF-PE component (even though that's obviously a zero-sum over the full cycle)..

What can i say, when the laws of physics themselves seem to be going awry, things are bound to get confusing! The full cycle looks to be 4x more worthwhile than the half-cycle though..

If anyone can make better sense of any of it, chirp up.. We're basically all in the same boat here..

[MrVibrating]

Despite the superficial similarity in relative motions, the flailing analogy only gets you so far - the wheel turns at constant speed, and the free axis is locked until the system's set rotating, precisely to avoid jerking / flailing - in practice, the motion is more akin to simply full-cycle kiiking under gravity, only using CF force instead.

There is however transfer of torque and work between the weight's CF-PE cycles (akin to GPE cycles) and the central motor.. esp. via the ice-skater effect..

[MrVibrating]

Latest version i'm using:

[Tarsier79]

Real world build:

I recently destroyed what would have been a perfect frame to test this.... that always seems to be the way. Had I kept it, it would have just gathered dust and got in the way.

Getting power to the moving arm, although annoying, isn't the biggest problem. A stepper motor is not efficient. Unless there is a mechanical equivalent action to replace the motor, we will need to install some sensors and control system, again Arduino is easy. Each time we get to a problem, it costs in energy. So it would need to be driven by 2 efficient motors (95% is a pipe dream here), then we might lose some efficiency on the bearing/power transfer, then we need some extra power for the sensors and drive circuitry. So what we need is a configuration in sim world that already maximises COP.

The other problem is efficient motors cost $$$... so there has to be some confidence on my part before I would spend thousands of dollars on a build. (do it once, do it right).

I think as far as motors go, it would be better to have two equal and electronically synced "flails" to balance forces on the main axle. And also to balance the mechanism with gravity (if we are not going to try to use gravity in the interaction.) Gravity could offer an opportunity to harvest energy. I am thinking the main disk as a flywheel, initially spin it up to operating speed, then use the energy gain to spin the wheel/ Gravity will try to accelerate and decelerate the main wheel.

[johannesbender]

If you want to mostly exclude gravity just go horizontal .

BTW look very trebuchet and planetary like.

[Fletcher]

Mornin MrV .. just a bit of light reading since I was last here lol ..

To much to comment on individually ..

I'll just short-cut to this ..

Regarding the "float test" analogy and as you mention the x axis movements being a little suspicious etc ..

And I did note you saying again that once the FOR is anchored the effect disappears or is absorbed, or evaporates etc etc ..

What I was thinking yesterday to further stress test the sim is to have the background "float" that the big wheel is motor joined to .. and for that background float (maybe a square) to have 4 'corner' springs radiating out to the 4 corners of the sim grid - then measure the tension and the length changes against the visual "wobble'" which maybe too small to see .. but the mere fact of anchoring it via imaginary springs connections to space may take the effect away .. dunno .. but to be fair there are lots of things I'm still processing which some days seem clearer and others not so ..

ETA .. the springs should highlight the x and y plane movement, or just x movement as the case may be - after all, the springs if turned OFF will still record any changes verses when they are turned ON (if it is still there) ..

I'm sure the penny will drop one way or another but in the mean time it is an interesting thread to try and get my head around and contribute positively too .. I think T and others feel the same way ..

[MrVibrating]

I jumped to conclusions with the first float test, seeing what i expected to see. In retrospect, there's no bearing friction in WM and the OU axis (the part i'm colouring green) is just free-wheeling, so can't be leaking momentum. Might still be a concern in a real-world rig if things ever get to that stage, but perfect inertial isolation is obviously unrealistic.

Admittedly disappointed (as much as relieved) it isn't also an inertial thruster, but OU will do for now. I'm obviously not imagining the intrinsic link between mech OU and CoM though.. it's a risk to stay mindful of, going forwards..