Decoupling Per-Cycle Momemtum Yields From RPM

[MrVibrating]

ME's response:

"that would require actually following the work and i can't be bothered"

But regardless, unsupported ad hom and lots of non-sensical gibberish.

Claims to be able to replicate the measurement, but can't be arsed, but certainty regardless that it's wrong, whatever it is.

More vacuous gibberish philosophy.. more ad hom..

Whining about old grudges, more ad hom, a cursory "build it!"

Then a bookmarked grudge (bet he has a lot of them, eh?) from when i caught him copy-pasting web formulas he didn't know how to use.




That's the sheer intellectual prowess of the response you've been waiting for, Mr Justsomone.. it's utterly typical and predictable, and the reason he's on so many ignore lists.

I'd recommend he be on yours too..

..but if you need someone else to interpret what i'm posting for you (least of all that specimen), maybe it's not really for you after all..?

I'm minded to put you on mine just for summoning that troll to this thread (don't speak their names!), however you get a pass, this time, since you're already on mine anyway (for making consistently inane posts)..

I've explained the interaction in plain english, and now i'm measuring it in ever-finer detail using plain metrics. Anyone can follow the sums, and any simmers able to construct basic equations can replicate the control conditions and measurements from scratch.

OU results are provocative, but it's our whole raison d'etre here - we're not doing it simply to be punk. If you have nothing constructive to contribute to the work, you don't need to comment!

If it IS a mistake - as almost every other example in the history of 'ever' - it'll be found and eliminated by the current methodology; zooming in on the exact cause of the gain, until it either disappears, or converges to a consistent anomaly... which is the current stage we're at. Further analysis will reveal precisely why it's arising, by tying it conclusively to its causative conditions, and showing how changes to those conditions cause proportionate changes in the gain characteristics.

Fundamentally, PE to KE gain can only be caused by anomalous acceleration of the system's reference frame - ie. by an effective N3 violation - hence either the gain will be attributable to such an effect, or else it must be in error, but finding out which requires patient methodical analysis..

Appealing to the authority of trolls ain't gonna get us there any quicker..

[silent]

A quick shout-out to once again thank you for your hard work. I agree with you about no need to comment and people dragging the trolls around. Never before have seen so many people in one spot who are legends in their own minds. No humility whatsoever! What I find utterly laughable is they claim expert status, yet have no wheel to show for it! I don't either, therefore it is impossible for me to be an "expert" on Bessler wheels until I figure out how to build a working one...and then refine it. Maybe someday I'll get to that point, but until someone actually knows how to build a wheel - it's like bumfighting. You can get a bunch of bums fighting and the winner is still still a bum. (BUM in the USA sense of the worse although I guess you could use Australia's definition and it would still aptly fit.) Likewise, you get a bunch of supposed experts on the forum each trying to outdo each other and throw their egos around, but in the end they're still a bunch of people with NO Bessler Wheel to show for all their supposed knowledge. Sitting on a hill, expounding knowledge, while gathering followers for yourself does not an expert make. You only think you are.

True - you can be an expert in running simulations or at the very least be really damn good at it - such as your MrVibrating. You're way over my head and that's why I ponder over your words to try and get a grasp of which you speak. Utterly fascinating!

Until someone can apply the knowledge gleaned from who-knows-where and actually build a working Bessler wheel, nobody can rightly claim to be an expert at Bessler wheel building. One must show results in order to be believed - a fact that the self-made legends just absolutely can't stand! And I sit here and laugh at them because I too am not an expert at Bessler Wheel production and never claimed to be.

But I'm trying and a humble spirit can learn oodles more than a haughty one I'm convinced.

By the way I've never enjoyed being on the list more than since I blocked the comments of the supposed experts. It's been a whole new world where I can actually enjoy going through over a decade (almost 2) of posts and not have to read *ANY* of their drivel. I'm with you - get those culprits in your blocked list and ride off into the sunset. Let 'em sit and rot for all I care. They honestly believe that nobody will be able to build the wheel without gleaning knowledge from their posts. LOL!

Thank you sincerely for hours and hours of hard work running those simulations!

silent

[MrVibrating]

Cyc #3:


freq = 32765 / 0.6148

i-s/f = 10,000

initial KE = 97.1182 J

final KE = 97.5905 J

KE rise = 0.4723 J

motor T*a = 0.305976805 J

motor P*t = 0.342624995 J

..since these have diverged somewhat, let's not average 'em - one's likely more accurate than the other, so i'll keep 'em separate..

initial net momentum = 9.8549 kg-m²-rad/s

final net momentum = 9.8788 kg-m²-rad/s

momentum rise = 0.0239 kg-m²-rad/s



efficiency:

per T*a:

0.4723 / 0.305976805 = 1.5435x OU for PE to KE

KE pre-collision = 117.4545 J

KE post-collision = 117.2045 J

KE dissipated = 0.25 J

total output energy = 0.25 + 0.4723 = 0.7223 J

0.7223 / 0.305976805 = 2.36x OU for net of KE + heat

per P*t:

0.4723 / 0.342624995 = 1.3785x OU for PE to KE

0.7223 / 0.342624995 = 2.1081x OU for net of heat plus KE


So the latter, P*t integral seems the most consistent with the previous cycles (surprising, usually T*a is more accurate) - either way tho, efficiency on this third cycle shows a tantalising increase..

..this being the case, i'll proceed with further consecutive cycles - i'd been intending to jump ahead to cycs #10, 20, 100 etc. if the first few were consistent, but if efficiency is changing with RPM - and especially, if it's increasing - then this is of course consistent with a diverging inertial frame, the longstanding objective; a system that gets more efficient the faster it spins.. so, cycle #4 in progress now, and will keep at it until it stops being exciting..

[MrVibrating]

@Silent - you're far too kind sir, i'm a complete dilettante and dabbler, and all i know about Bessler's wheel is deduced from the descriptions of it, in relation to the basic laws of physics - it was statorless, everything went around together, it made banging noises and it was somehow OU. In compliment to those facts, mech. OU actually does arise from an effective workaround to N3.. so, it's really just putting two and two together; round peg, round hole..

I like the 'bum fight' analogy tho - i've thought of it in the past as like trying to knock down a wall by running into it full-whack - everyone's got their preferred techniques and theories, but all of us just keep bouncing off of it, nursing various degrees of insult.. "watch this lads, this'll be a good 'un" - 'chaaarge' - 'splat'.. again and again. We're tenaciously committed to failure, and the pathology of repeating it, over and over.. trying to maintain any kind of dignity just makes one more of a clown; we ARE the poster child for a 'fools errand', this IS the peanut gallery, and trying to take it too seriously only makes for further comedy..

As you say, turning the tables means producing results. We're all armchair experts til then..

[ovyyus]

I sit here and laugh at them because I too am not an expert at Bessler Wheel production and never claimed to be.

Who on this forum claims to be a 'Bessler wheel production' expert'? Who are these 'experts without wheels' that you laugh at and crusade against? AFAIK, ME has never claimed Bessler wheel production expertise. Curious.

[MrVibrating]

..still waiting on cyc #4 (won't be done til A.M.), but chewing over potentially-influential factors already known to be changing with RPM..

..for example the torque applied is multiplied up by the system RPM, to compensate the diminishing 'G-time' per cycle as speed increases; however i didn't fine-tune it to perfectly offset the timing, hence it actually advances over successive cycles - accomplishing the target relative speed earlier in the cycle, the higher the RPM's get..

Could this be causing the efficiency to improve? Is it the case that applying more torque earlier in the cycle - despite attaining the same speed each time - achieves a better energy return?

Alternatively, there's another place it might be creeping in; once the TRS has been reached, it is held constant until the bob reaches 6 o' clock BDC, at which point the motor is swapped out for the brake.. So, might this second phase of the drop be responsible for the gain? Under this condition - the TRS having been reached - the bob and rotor are both being accelerated by gravity, however the rotor is already rotating at 1 rad/s (or whatever TRS) faster than the bob, so is it actually gaining more KE from the bob's fall than it does itself, despite both having identical inertias..? Since KE squares with velocity, if the rotor's turning a constant 1 rad/s faster than the bob, maybe where the bob gains 1 J from output GPE, the rotor gains 2 J, from that same fall, or something?

Obvioushly, accomplishing the TRS earlier in the cycle increases the 'duty cycle' of this second phase wherein the weight's still dropping but whilst the rotor's locked to a higher fixed speed relative to it, type situation.. so this could be consistent with efficiency improving with RPM..

Dunno, for now, but all will become clear once the data's in - i'll take further integrals from a single cycle; one halfway through the spin-up phase, and then another after the TRS is reached, and another just before the brake engages.. somewhere along that process we'll be able to see exactly where the gain develops.. then we can really start optimising..

[MrVibrating]

cyc #4:


freq = 32765 / 0.6136

i-s/f = 10,000

initial KE = 97.5905 J

final KE = 98.0618 J

KE rise = 0.4713 J

motor T*a = 0.3287939 J

motor P*t = 0.299423645 J

initial net momentum = 9.8788 kg-m²-rad/s

final net momentum = 9.9026 kg-m²-rad/s

momentum rise = 0.0238 kg-m²-rad/s


efficiency:

0.4713 / 0.3287939 = 1.4334x unity for T*a to KE

0.4713 / 0.299423645 = 1.574x unity for P*t to KE

initial net momentum = 9.8788 kg-m²-rad/s

final net momentum = 9.9026 kg-m²-rad/s

momentum rise = 0.0238 kg-m²-rad/s

net KE pre-collision = 117.9258 J

net KE post-collision = 117.6758 J

KE to heat thus = 0.25 J

0.25 + 0.4713 = 0.7213 J total output energy

0.7213 / 0.3287939 = 2.1937x unity for T*a to net of heat + KE

0.7213 / 0.299423645 = 2.4089x unity for P*t to net of heat + KE

..still rising..

Cyc #5 in progress..

[MrVibrating]

..here's what's bugging me:

Look at that 'KE dissipated' each cycle - 0.25 J, each and every time..

This is because the target relative speed is 1 rad/s, and both parts have an MoI of 1 kg-m², and accelerating 1 kg-m² by 1 rad/s costs half a Joule, per KE=½Iw²..

..not 0.3 J..!


So, get this; we're seeing the 1 rad/s relative acceleration, no question. That's legit. An inelastic collision between two 1 kg-m² MoI's at 1 rad/s relative dissipates half the energy - and half of 0.5 J is of course 0.25 J...


Are you following this bit? It seems it could be important.. the KE dissipated in the collision is consistent with KE=½Iw²; the 1 rad/s relative acceleration is worth 0.5 J, and half of that figure gets dissipated in the collision, consistently (coming in at exactly 0.25 J every cycle thus far)...

..yet the input work done by the motor is only ~0.3 J per cycle!

This 0.3 J is consistently achieving a target relative velocity of 1 rad/s per cycle, which then consistently dissipates 0.25 J per cycle.. and then we end up with a 0.47 J net KE gain each cycle! What the actual fuck?

..so it appears that the collision is somehow causing 0.3 J of work to turn into 0.47 J of KE..

..plus the 0.25 J dissipated, hence 0.3 J is actually transformed into 0.47 + 0.25 = 0.72 J each cycle..

..and 0.72 / 0.3 = 2.4x input..

Looks like it might help to meter the two parts' KE's independently, at some point, just to see what their respective gains / losses are...

[agor95]

Hi MrVibrating

Apologies for not committing the time your analysis deserve.

We are well onto 17+ pages of information to read.

Your methods of calculating joules from K.E. etc will be helpful later.

However I would like you to look at this effect from another frame of reference.

The disk is balanced and the pendulum is of cause not.

Imaging this device being accelerated literally up the screen with no gravity present.
So the force of Gravity is replaced by mass times acceleration.

That would mean virtual work being done to move, also accelerate, the mass with respect to time.
That work is being done from outside the system, as gravity is external to the device.

The pendulum moving from 12 O'clock to 3 O'clock

The pendulum at first is pushed but less so until it is horizontal.
The pendulums appears in 'free fall'; however the mass is stationary.

This is from your external point of view.

From 3 O'clock to 6 O'clock the pendulum is pulled more so until it arrives at the
end position.

In this alternative interpretation the Kinetic Energy in the pendulum is a result of the work done by the external force on the system.

The virtual work being done on the disk during this process is constant.
The virtual work being done to the pendulum drops and then increases.

However if you slow the pendulums motion by increasing the disk rotation;

Then the virtual work done on the pendulum also increases.

Regards

[ME]

That reply only shows that you defend your overunity claims with a lot of assumptions towards me.

I am in no position to defend myself against the label "troll", that's up to others, but I try no to be one.
Especially for that reason I try to avoid assumptions by giving links to posts. In previous case, I simply searched this forum: "Unicode", member:"mrVibrating".
That link also points to one post up that shows a calculus that should be easy to follow for everyone with a bit algebra knowledge to verify and recalculate. While at the same time avoids assumptions about me talking nonsense where I could just have said "humbug".

Claims to be able to replicate the measurement, but can't be arsed, but certainty regardless that it's wrong, whatever it is.

I'm not replicating your errors. As you show, there are too many ways to make them.
More than once I corrected your miscalculation and just showed how thing balance in contrast of your claimed 'overunity'. That is: all energy components are accounted for, and no surplus.
No grudge list, but I think I can find the links on this forum that substantiate this with an examples. I suspect that at least one should be easy to find, because I remember recreating the whole calculus in a spreadheet so I'd have to search my album for that.

Actually, now I mention, we never heard of those previous claims again. What's up?

I am not replicating your stuff so often, because it only just takes a lot of time and effort to figure out what you tried exactly and then see how WM2D might fluctuate in its measurement.
Hence it would be best to do the recalculations on paper, so you'd have ideal target values to check against and discover how it differs.
And even when I invest these efforts, as I did before, I'll still be called "troll".

We could for example look at https://www.besslerwheel.com/forum/view ... 631#169631
But no one cares: By the quick looks of it you over complicated that motor thingy by ramping the torque up till it reaches 10 rad/s, while this torque is coupled to current speed..why ? I understand you want to reach to 10 rad/s very fast, but the resulting function is now in effect an exponential function --and as the result is also the input, it needs a solver, but is not part of the solver, and adds to inaccuracy-- where a solid torque value, or a constant increase would get you the same effect.
Also, it looks like your green bob is not in free-fall. Likely this motor pulls itself relative to the green-bob layer. It adds yet another constraint that's is in need of a solver because of mutually depended values.
In effect of such coupling and relative to freefall you pull the green-bob upwards while also slowing it down.
Maybe this is where some of your unaccounted energy comes from: The difference between measured values relative to the stationary background, versus values relative to the green-bob-layer. But if that path really needs investigation depends on how you exactly calculated those Motor/Brake (T*a) values.

Also your brake is also doing non-ideal stuff.
When your brake would be locked, then we could easily calculate that the angular velocity before motor engagement (green-bob at 9:00) would be 10.2945 rad/s.
But that isn't the case: With fluctuations, your wheel goes: 10.3041 rad/s, while your the green-bob goes 10.2859 rad/s.
That doesn't seem much and it averages well enough, it also doesn't influences the principle of the impact.
Yet that final difference in speed (0.0182 rad/s) will certainly have its influence when you do an W=T*w*t on an massive break system.
I don't know if you did use you break calculus as such, or applied a numerical integral on that part, or skipped this quarter...
It shows one possibility where you could discover your unaccounted energy if you'd really be interested.

When you would make the brake more sturdy, then you should see an increase in accuracy.
When you would make the motor more predictable, then you should be able to calculate its effect manually.

Oh well, when you really want a calculus attempt, here we go.
The cycle of that situation is observed to go like this:
In the first quarter then Green bob drops a quarter and drags the main wheel with it. The gain is 9.80665 Joules because of the potential drop.
We measure that in the first cycle the motor adds about 31.65 Joules to the whole system, and at the end it's 29.85 Joules (because it's closer to 10 rad/s). This quarter also gains 9.80665 joules because of the potential drop of the bob.
The collision robs it of 25 Joules.
Why this constant, because the you designed the motor to gain a speed difference of 10 rad/s. In inelastic collision this difference gets equalized over two inertia because momentum is what gets reserved.
Then in the last half they are locked (almost), and the green-bob loses potential again: twice 9.80665 Joules for the semicircle.

So each cycle you gain about ±30.6 - 25 = ±5.6 Joules, After 10 cycle this is ±56 joules.
No matter how you calculate the motor gain, as can be deduced from this simple analysis, it is the motor gain you'll measure as gain.

But apparently I'm interfering with some kind of world-record attempt for creating the most overunity claims in one continuous stream.
Sorry.

tl;dr; It show its very easy to yell overunity, when the calculus is difficult (or impossible) and that ignoring it gives a reason for a brand new claim of overunity.
That you cant sufficiently recreate the values with manual work-around calculus on spiky data does not proof it is overunity for "regardless and whatever it is".

Maybe I should take the easy route and write it as simple as mr Silent does:

• No Physical Wheel - No Overunity

[eccentrically1]

From page 1

Ecc1:

What mechanism is spinning, rotating, and braking your flywheel?

Mrv:

A motor, bearing and brake.

I lost interest after this answer to my question.
Momentum gain is never reactionless.
I tried to show that in another thread, I don’t remember which one.

[MrVibrating]

cyc #5:


freq = 32765 / 0.6121

i-s/f = 10,000

initial KE = 98.0618 J

final KE = 98.5320 J

KE rise = 0.4702 J

motor T*a = 0.324552895 J

motor P*t = 0.34217574 J

initial net momentum = 9.9026 kg-m²-rad/s

final net momentum = 9.9263 kg-m²-rad/s

momentum rise = 0.0237 kg-m²-rad/s



efficiency:

per T*a:

0.4702 / 0.324552895 = 1.4487x OU for PE to KE

KE pre-collision = 118.3961 J

KE post-collision = 118.1461 J

KE dissipated = 0.25 J

total output energy = 0.25 + 0.4702 = 0.7202 J

0.7202 / 0.324552895 = 2.21x OU for net of KE + heat

per P*t:

0.4702 / 0.34217574 = 1.3741x OU for PE to KE

0.7202 / 0.34217574 = 2.1047x OU for net of heat plus KE


..getting bored (more like impatient) with this now - was gonna try plotting out the performance stats with simple graphs, but little point with such an arbitrary rig design.. better to spend some time analysing any one of these cycles in closer detail - i keep thinking of more useful tests to run, but the PC's constantly hogtied with these long runs.. How about running a quick test cycle with gravity disabled? Then we could see how the input work integrals compare to the gravitating example..

I think i already got an angle on this last night - it's a 1 rad/s acceleration of 1 kg-m² of inertia, so should've cost ½ J; the collision dissipates ¼ J - the 'right' amount of KE loss for the speed and inertia - so the anomaly is that the motor only performed a little over 0.3 J of work in achieving that 1 rad/s relative acceleration..

..and so this is what i wanna check with a 'no gravity' cycle - is the motor workload still 0.3 J, or does it rise to 0.5 J?

Because if that's the case, then the bottom line is that we're getting a 0.2 J discount - from gravity - on the energy cost of causing a 1 rad/s acceleration..


Additionally, i need to add independent KE meters on both parts, so that we can see the relationship between the work done by the motor, and the corresponding changes in KE of the two inertias..

Like i say, slow going since i only get an hour or two for this each day (if i even have the energy)..

[MrVibrating]

Hi MrVibrating

Apologies for not committing the time your analysis deserve.

We are well onto 17+ pages of information to read.

Your methods of calculating joules from K.E. etc will be helpful later.

However I would like you to look at this effect from another frame of reference.

The disk is balanced and the pendulum is of cause not.

Imaging this device being accelerated literally up the screen with no gravity present.
So the force of Gravity is replaced by mass times acceleration.

That would mean virtual work being done to move, also accelerate, the mass with respect to time.
That work is being done from outside the system, as gravity is external to the device.

The pendulum moving from 12 O'clock to 3 O'clock

The pendulum at first is pushed but less so until it is horizontal.
The pendulums appears in 'free fall'; however the mass is stationary.

This is from your external point of view.

From 3 O'clock to 6 O'clock the pendulum is pulled more so until it arrives at the
end position.

In this alternative interpretation the Kinetic Energy in the pendulum is a result of the work done by the external force on the system.

The virtual work being done on the disk during this process is constant.
The virtual work being done to the pendulum drops and then increases.

However if you slow the pendulums motion by increasing the disk rotation;

Then the virtual work done on the pendulum also increases.

Regards

kind of see what you're saying, i think...

..i'm planning on doing a 'no gravity' baseline, to see if the workload on the motor increases to 0.5 J per 1 kg-m² per 1 rad/s (as you'd expect from ½Iw² - it should be a 0.25 J change of KE in each direction); if this is confirmed then it means we're getting free work from gravity.. in the form of a discount in the amount of torque required..

Dunno if i'll get any more done this eve, been zombified all day as it is, need slepp..

[MrVibrating]

From page 1

Ecc1:

What mechanism is spinning, rotating, and braking your flywheel?

Mrv:

A motor, bearing and brake.

I lost interest after this answer to my question.
Momentum gain is never reactionless.
I tried to show that in another thread, I don’t remember which one.

Short of anything new to say but keen to pipe up anyway, you simply requote yourself from a year ago! Feel better tho, right? Getting that off your chest? Good.


Yes, we're all aware of N3, thanks.

Ever managed to gain height on a park swing, tho? Were you applying counter-torque at the swing's axis, i wonder? If you were to swing around in complete loops, would you be inducing counter-angular momentum to earth, via gravity?

You're presumably aware that the motor and brake only connect between the two co-rotating parts - not between them and the background / world? So you can appreciate, what with your finely-honed sense of N3, that because they both have equal angular inertia, if gravity were disabled, the motor would spin each to equal speeds in opposite directions, agreed? Except here, we do have gravity, and so this skews the distribution of momentum procured from said motor; w/o gravity it's symmetrical, but with gravity, it's asymmetric! The rotor thus speeds up more than the pendulum slows down, even though they both have equal MoI are are subject to equal torque from the motor.. because gravity.


If the motor WERE mounted to the background / world, then both the torque * angle and power * time integrals would show that the target speed of 1 rad/s has a PE cost of ½ J... not 0.3 J..!

We're not definitive yet - still a chance of error - but both metrics are consistently showing that the work done by the motor is less than the resulting KE value of that work.

As i keep pointing out, 'excess KE' is an oxymoron, not a logically-consistent concept - a system or mass can only ever have precisely the right amount of KE for its given inertia and velocity, per the standard KE equations. Therefore if mech. OU is to be 'a thing' at all, then it can only mean discounted momentum; paying less in F*d or T*a than the resulting KE value. Anything else is just magical thinking.

The rig's designed to accommodate a diverging reference frame - its system momentum is isolated from the environment, there is no torque between the system and the environment, it is a closed system of masses interacting about a common axis, but for the additional presence of gravity. Closed loop momentum gains from gravity are already 'a thing' (per swinging / kiiking). The closed-loop momentum gains in this rig however are vanishingly small (and gradually shrinking with rising RPM) - the FoR divergence must thus be happening instantaneously, in real-time, as the torque is being applied.. because there's fuck-all momentum accumulating between cycles!.

This was an "if you build it, they will come" type thing - i hadn't plotted out an expected gain - but regardless, it looks like they're here (woo!) - or maybe it's all just a silly mistake. Who knows, or cares. It's this, or 2 hours of pointless base-building in Subnautica. This 0.2 J discount per cycle has a cause, and square in the sights is gravity. Tomorrow night i'll try the 'no gravity' test to see how it affects the input energy..

[MrVibrating]

@Mar-troll-o

The only calculus you need is the actual test data - you output that into a text file then copy-paste it into Excel and apply a Riemann sum; each animation frame produces one row of data, so more frames - and more integration steps per frame - improves the integral accuracy. It is what it is - i don't 'claim' that 2+2=4 or anything else; the result is OU - the motor's doing less work than its resulting KE value.. that's the anomaly.

The two integrals being used for input energy are torque * angle, and torque * angular velocity * time.

Your ongoing obsession with linking to old grudges is creepy and weird. Snap out of it man. Focus on the here, now - what's the rotKE for 1 kg-m² at 1 rad/s? How much energy should that acceleration cost to perform? Now go back and look at the test data.. see the anomaly? You're not gonna find the answers in old threads, my bunny-boiling compadre..