Decoupling Per-Cycle Momemtum Yields From RPM
Moderator: scott
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
I realised the solution last night, while trying to get to sleep.
Mulled it over all day at work. I'm sure i've at least tried to measure the interaction before, dunno what went wrong that time, probably hadn't quite conceived the right experiment, but can't find it in me notes either so i'll just fill yuz in from the top:
• an overbalancing weight adds momentum to a wheel
• it will add the most when starting from stationary
• the faster the speed the drop begins at however, the less momentum it's worth, because of the inverse relationship between G-time and RPM
• consider a pendulum: upswing and downswing periods - ie. negative and positive G-times - are equal, hence net momentum is constant
• yet we can vary G-times by applying inertial torques to either or both swing phases
• one such type of inertial torque would be that caused by changing MoI whilst rotating - the ice skater effect
• another is simple counter-torque, and the corresponding counter-momentum it induces
• this latter principle is all we're going to use
• make the pendulum a rigid rod
• attach the bob to the rod's lower end with a motor
• by applying torques to the bob, spinning it up about its own axis, we can direct the counter-torque against its own gravitation, cancelling gravity's acceleration
Now you have the concept, here's the rig:
The beam is 2 meters in length / radius, and the rotor is nominally 1 meter radius and 1 kg.
The system begins in rotation at 1 rad/s.
As it departs TDC, the motor begins to spin up the bob, using however much torque as required to cancel gravity's acceleration.
Thus the downswing proceeds at constant velocity (near-as for a quick sim anyway).
Note that this required us to spin up 41.1 kg-m²-rad/s of momentum..
..this is the momentum yield for that GPE and RPM; if the bob weight and starting speed don't change, neither will the momentum yield..
..however to raise enough counter-momentum to cancel that GPE's momentum yield, we also had to raise 846.1 J of rotKE on the bob!
Obviously, if we need to collide or brake the bob back against the beam in order to repeat the process (and to avoid having to keep spinning it ever-faster, we do), then most of that input energy's gonna get dissipated in an almighty thwack! and it just wouldn't be very efficient, to put it mildly..
But the MoI, for a given bob weight and thus GPE, is a variable function of its radial distribution, per mr², right?
So, we could keep the bob weight at 1 kg, keep the pendulum at 2 m, keep the starting speed at 1 rad/s and just consider a small range of bob sizes of arbitrary radii and thus angular inertias..
For instance, let's repeat that last run, making this one, single difference:
• in the bob's properties widget, i've changed the MoI from '1' to '10' - observe the result:
..as you can see, near-identical results, but for the fact that the input workload has been cut by a factor of ten...
Now just run with me on this for a sec - we all appreciate the difference between 'practical' and 'physical', and it's the latter possibilities that concern us; all that matters is: is it physical? - as in, at a fundamental level, does the universe allow it?
You know what we've got to do now, right? Yep, let's add another hypothetical zero to that bob's MoI:
..once again, same-same, 'cept now we've just cancelled out 41.2 J of output GPE using just 8.5 J of input work!
So...
...U thinking what i'm thinking..?
..yeah, me too...
Go on then:
..so just to clarify what you're seeing there:
• beginning with 1 rad/s of upswing velocity as it departs BDC, we perform 8.5 J of input work to apply 41.2 L of CM worth 41.2 J of GPE
Don't even need to take the integral, it's all plain as day..
;P
Mulled it over all day at work. I'm sure i've at least tried to measure the interaction before, dunno what went wrong that time, probably hadn't quite conceived the right experiment, but can't find it in me notes either so i'll just fill yuz in from the top:
• an overbalancing weight adds momentum to a wheel
• it will add the most when starting from stationary
• the faster the speed the drop begins at however, the less momentum it's worth, because of the inverse relationship between G-time and RPM
• consider a pendulum: upswing and downswing periods - ie. negative and positive G-times - are equal, hence net momentum is constant
• yet we can vary G-times by applying inertial torques to either or both swing phases
• one such type of inertial torque would be that caused by changing MoI whilst rotating - the ice skater effect
• another is simple counter-torque, and the corresponding counter-momentum it induces
• this latter principle is all we're going to use
• make the pendulum a rigid rod
• attach the bob to the rod's lower end with a motor
• by applying torques to the bob, spinning it up about its own axis, we can direct the counter-torque against its own gravitation, cancelling gravity's acceleration
Now you have the concept, here's the rig:
The beam is 2 meters in length / radius, and the rotor is nominally 1 meter radius and 1 kg.
The system begins in rotation at 1 rad/s.
As it departs TDC, the motor begins to spin up the bob, using however much torque as required to cancel gravity's acceleration.
Thus the downswing proceeds at constant velocity (near-as for a quick sim anyway).
Note that this required us to spin up 41.1 kg-m²-rad/s of momentum..
..this is the momentum yield for that GPE and RPM; if the bob weight and starting speed don't change, neither will the momentum yield..
..however to raise enough counter-momentum to cancel that GPE's momentum yield, we also had to raise 846.1 J of rotKE on the bob!
Obviously, if we need to collide or brake the bob back against the beam in order to repeat the process (and to avoid having to keep spinning it ever-faster, we do), then most of that input energy's gonna get dissipated in an almighty thwack! and it just wouldn't be very efficient, to put it mildly..
But the MoI, for a given bob weight and thus GPE, is a variable function of its radial distribution, per mr², right?
So, we could keep the bob weight at 1 kg, keep the pendulum at 2 m, keep the starting speed at 1 rad/s and just consider a small range of bob sizes of arbitrary radii and thus angular inertias..
For instance, let's repeat that last run, making this one, single difference:
• in the bob's properties widget, i've changed the MoI from '1' to '10' - observe the result:
..as you can see, near-identical results, but for the fact that the input workload has been cut by a factor of ten...
Now just run with me on this for a sec - we all appreciate the difference between 'practical' and 'physical', and it's the latter possibilities that concern us; all that matters is: is it physical? - as in, at a fundamental level, does the universe allow it?
You know what we've got to do now, right? Yep, let's add another hypothetical zero to that bob's MoI:
..once again, same-same, 'cept now we've just cancelled out 41.2 J of output GPE using just 8.5 J of input work!
So...
...U thinking what i'm thinking..?
..yeah, me too...
Go on then:
..so just to clarify what you're seeing there:
• beginning with 1 rad/s of upswing velocity as it departs BDC, we perform 8.5 J of input work to apply 41.2 L of CM worth 41.2 J of GPE
Don't even need to take the integral, it's all plain as day..
;P
- Attachments
-
- GPE_vs_MoI_2.wm2d
- (12.45 KiB) Downloaded 56 times
-
- GPE_vs_MoI_1.wm2d
- (12.45 KiB) Downloaded 50 times
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
re: Decoupling Per-Cycle Momemtum Yields From RPM
- JC's APBut I would just like to add this friendly little note of caution:- A
great craftsman would be that man who can "lightly" cause a
heavy weight to fly upwards! Who can make a pound-weight rise
as 4 ounces fall, or 4 pounds rise as 16 ounces fall. If he can sort
that out, the motion will perpetuate itself. But if he can't, then his
hard work shall be all in vain. He can rack his brains and work his
fingers to the bones with all sorts of ingenious ideas about adding
extra weights here and there. The only result will be that his
wheel will get heavier and heavier - it would run longer if it were
empty! Have you ever seen a crowd of starlings squabbling
angrily over the crumbs on a stationary mill-wheel? That's what it
would be like for such a fellow and his invention, as I know only
too well from my own recent experience!
He refers back to this riddle later ("even Wagner, wherever he is, will have heard that one pound can cause the raising of more than one pound").
The preceding paragraph speaks of weights alternating radii - but also denying a false attribution of this principle by Wagner to himself, since this is the first time he's ever mentioned it; obvs Wagner was referring to a 'classic OB' mech of radial lifts w/ angular drops, which also involves weights changing radii, yet this is evidently not their purpose wrt whatever it is Bessler has in mind.. and if not for OB, their only other effect is in modulating the MoI.
Indeed, the paragraph after gives an explicit rebuttal of any possibility of an effective GPE asymmetry..
..so the riddle definitely isn't hinting at a GPE asymmetry.
It's probably thus not even referring to an energy gain.
So what purpose, then, to raise an output one quarter of an input?
It's obvious we must actually be dealing in the currency of OU, right?
Momenta, then.
Momentum exchanges / G*t yields.
Trading MoI for velocity / reactionless angular accelerations / decelerations (the ice-skater effect). Which also thus causes I/O G-time asymmetries and closed-loop momentum gains from gravity and time (per kiiking).
These 'quarters' are the stuff - one of them tips the scissorjack on the Toys page (note that it's mechanically non-functional), and each vertical link down the chain of A & B represents a 25% efficient (ie. 'one quarter') net momentum yield, four of which sum to unity but five of which make 125%.
Thus '4 oz in for 1 lb out' is alluding to one of these momentum exchanges / G*t yields.
IOW, this cryptic 'four-factor', whatever it is, dishes out RPM-invariant momentum yields - that is, it's a process that fixes the unit energy cost of momentum in the rotating FoR, decoupling it from its accumulating KE value in the ground FoR; doubling the speed thus giving 4x the energy at only 2x the cost.
Thus, 'rising' doesn't necessarily mean 'gaining height'..
'Accelerating', perhaps? It has to be one or other component of momentum, right?
Note again how the riddle is structured to shut down misintepretations - another example of this careful word play is the repetition of the quadratic function - "4 oz in / 1 lb out, or 16 oz in for 4 lb out" - clearly emphasising the same mathematical relationship.
Further, since "pound-weight" refers to a single body / object, might we also assume that "ounces" are individual quanta, rather than one homogeneous lump?
Again, the Toys page shows an interaction that's 75% inefficient each cycle, but the quarter of each cycle's input energy that remains buys the same 'rise' in momentum, culminating in a 125% gain at the fifth cycle.
These are the 'quarters' we wanna be buying.
Whatever they are, is the solution to his 'pound-weights' riddle.
Is MT 137 relevant? ("Pounds in equilibrium.") Dunno.
Nonetheless, he keeps referring back to this riddle (scan the search term "..in part one.." thru AP), basically telling us to "go back and study it, cuz it already covers everything you need know, if only yuz can get ye heads 'round it.."
'Raising' = momentum.
Not 'GPE'.
The solution to the 'pound weights' riddle is the solution to this thread's riddle.
The scissorjack, or whatever it represents, is obviously key to solving this.
Mechanically, it converts opposing torques into linear displacement, or vice-versa.
No real clue how all this fits together yet, but a penny's dropped re. the significance of this riddle - of what it definitely isn't talking about, and thus by elimination, whatever's left..
It's clever because of course if you had an effective GPE asymmetry, Bob's yer uncle.. and if me nan had wheels she'd be a tea trolley..
But if you can raise equal momenta at rising RPM yet at constant 25% efficiency - absorbing the initial (75% !) losses - the fourth helping hits unity, and the fifth, 125%.
Not much, but a toe-hold.. see if it leads anywhere..
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
re: Decoupling Per-Cycle Momemtum Yields From RPM
FWIW, here's those AP quotes:
Wagner seems almost to have run out of fancies. He says nothing
can be achieved with "mechanical implements", the gist being that
my Mobile must be impossible because I designed it to be driven by
some "mechanical power". But did I not, in Part One, devote more
than one line to a discussion of the type of "excess impetus" that
people should look for in my devices? Once more I will humbly extol
the virtues of this passage to my next worthy reader. Even Wagner,
wherever he is now, will have heard that one pound can cause the
raising of more than one pound. He writes that, to date, no one has
ever found a mechanical arrangement sufficient for the required
task. He's right! So am I, and does anyone see why? What if I
were to teach the proper method of mechanical application? Then
people would say: "Now I understand!�
Listen - my weights are not likethose in turnspits and clocks. They don't
need to be raised up - it's a different arrangement altogether from what
you see in mill-wheels, turnspits and clocks. This is all mentioned in
Part One; read it at your leisure. Have I got to slap you on the snout with
it, you ignorant half-wit, before you properly understand it?
Since Wagner would be only too eager to know the secret of my
device, I'd like to tell him that he'll find all the ins and outs of the
matter in Part One.
re: Decoupling Per-Cycle Momemtum Yields From RPM
Hi MrVibrating
I like this section of the text 'He's right! So am I, and does anyone see why?'
Regards
I like this section of the text 'He's right! So am I, and does anyone see why?'
Regards
[MP] Mobiles that perpetuate - external energy allowed
re: Decoupling Per-Cycle Momemtum Yields From RPM
Hi Mr V .. I think basically you are on to it. You got more than a toe hold.
We know a wheel is a closed system where weights shift around, be that radially or otherwise. The ol' "Height for Width" conundrum. And that is what Wagner refers to imo. And for Agor95 why Wagner is also right !
We also know that because gravity force is conservative the path of weights makes not a jot of difference to the NET Energy and Momentum sums. The ol' What Goes Up Must Come Down axiom. Or put in a way I prefer to think about it "What Goes Down Must Come Up An Equal Vertical Displacement". The ol' WGDMCUAEVD which is far too long to catch on like H4W or W4H lol.
IOW's, the GPE is recycled or restored in each complete revolution, or even each individual sector depending on your non-asymmetric (or temporary torque) OB system. Even if weights were lifted at tdc etc by an unknown force it would still follow a defined path that circulates GPE and repeats. This assumes for expediency no frictional energy losses like a sim can achieve.
================
B. tells us his wheels generate innate excess impetus. Excess force, excess "weight", preponderance etc.
================
Karl tells us it turns from innate momentum IIRC.
DT Page 192-195 John Collins
By The Grace Of God, We, Karl, Landgrave Of Hessen, Prince Of Hersfeld, Count Of Katzenelnbogen, Dietz, Ziegenhaven, Nidda And Schaumburg, hereby make the following testimony and proclamation: -
"is a revolving wheel, which is able to run, by means of its own innate momentum"
Kassel, 27th May, 1718 KARL
And this is where B. is also right !
================
So excess Momentum imo is the answer we seek, as you deduce. Produced by some internal ingenious mechanical arrangement where it is both easy to understand the mechanical actions when seen (but maybe not to understand the Physics reasons for the PM Principle at first glance), and simple to build, according to Karl's quote to his ministers in the Bernoulli letter.
Somewhere in the Laws of Newtonian Physics and Newtonian Mechanics is a "workaround" mechanical condition that leads to a natural gain in system Angular Momentum, IMO. An example might be MT13 where excess torque is produced in all of wheel by lifting the lws early full height at tdc, or MT21 side lift to higher position. These require a readily usable force generated internally and redirected for purpose.
There are limited possibilities and, as you do, you work thru them one by one looking for the chink in the armour to be exploited for a real physical advantage. A cautionary note. The mechanical actions (but not necessarily the mechanical PM principle) have to be easy to understand once seen and simple to build. Find it first and refine it to its simplest buildable form, which is also your sensible approach.
"Findeth the PM Principle, cometh the wheel."
.................................
MT137 :
I have on occasion used a similar line drawing method when building a physical wheel to mark out equi-distant circumference points for sectors. It's a bit cumbersome imo. I usually just use a protractor and ruler etc.
It also makes a good template for building gears imo, tho again I use free online gear templates off the web.
Thirdly it symbolizes imo a balanced torque neutral arrangement which of course it must be because it has complete symmetry.
We know a wheel is a closed system where weights shift around, be that radially or otherwise. The ol' "Height for Width" conundrum. And that is what Wagner refers to imo. And for Agor95 why Wagner is also right !
We also know that because gravity force is conservative the path of weights makes not a jot of difference to the NET Energy and Momentum sums. The ol' What Goes Up Must Come Down axiom. Or put in a way I prefer to think about it "What Goes Down Must Come Up An Equal Vertical Displacement". The ol' WGDMCUAEVD which is far too long to catch on like H4W or W4H lol.
IOW's, the GPE is recycled or restored in each complete revolution, or even each individual sector depending on your non-asymmetric (or temporary torque) OB system. Even if weights were lifted at tdc etc by an unknown force it would still follow a defined path that circulates GPE and repeats. This assumes for expediency no frictional energy losses like a sim can achieve.
================
B. tells us his wheels generate innate excess impetus. Excess force, excess "weight", preponderance etc.
================
Karl tells us it turns from innate momentum IIRC.
DT Page 192-195 John Collins
By The Grace Of God, We, Karl, Landgrave Of Hessen, Prince Of Hersfeld, Count Of Katzenelnbogen, Dietz, Ziegenhaven, Nidda And Schaumburg, hereby make the following testimony and proclamation: -
"is a revolving wheel, which is able to run, by means of its own innate momentum"
Kassel, 27th May, 1718 KARL
And this is where B. is also right !
================
So excess Momentum imo is the answer we seek, as you deduce. Produced by some internal ingenious mechanical arrangement where it is both easy to understand the mechanical actions when seen (but maybe not to understand the Physics reasons for the PM Principle at first glance), and simple to build, according to Karl's quote to his ministers in the Bernoulli letter.
Somewhere in the Laws of Newtonian Physics and Newtonian Mechanics is a "workaround" mechanical condition that leads to a natural gain in system Angular Momentum, IMO. An example might be MT13 where excess torque is produced in all of wheel by lifting the lws early full height at tdc, or MT21 side lift to higher position. These require a readily usable force generated internally and redirected for purpose.
There are limited possibilities and, as you do, you work thru them one by one looking for the chink in the armour to be exploited for a real physical advantage. A cautionary note. The mechanical actions (but not necessarily the mechanical PM principle) have to be easy to understand once seen and simple to build. Find it first and refine it to its simplest buildable form, which is also your sensible approach.
"Findeth the PM Principle, cometh the wheel."
.................................
MT137 :
I have on occasion used a similar line drawing method when building a physical wheel to mark out equi-distant circumference points for sectors. It's a bit cumbersome imo. I usually just use a protractor and ruler etc.
It also makes a good template for building gears imo, tho again I use free online gear templates off the web.
Thirdly it symbolizes imo a balanced torque neutral arrangement which of course it must be because it has complete symmetry.
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
Re: re: Decoupling Per-Cycle Momemtum Yields From RPM
Because Wagner's talking about an effective GPE asymmetry, while Bessler's thinking about the equivalence principle (not even codified until Einstein) - ie. in terms of 'a pound of momentum'.. not GPE.agor95 wrote:Hi MrVibrating
I like this section of the text 'He's right! So am I, and does anyone see why?'
Regards
Remember, this is before official resolution of the vis-viva dispute, let alone development of SI units, so the terms 'L' or 'kg-m²-rad/s' simply hadn't been devised yet. Newton's Principia described conservation of the mv product, but if Bessler talked in such explicit terms he'd be giving the game away.. it probably seemed perfectly natural to thus consider momenta in terms of the 'weights' of the masses / inertias.
The same point applies to angular inertia - a property he must've been keenly adroit of - which might naturally be considered in terms of the 'weight' equivalence of the orbiting / rotating masses.
Bottom line, the remark re-affirms the futility of seeking an effective GPE asymmetry, while insisting there's another, plausible interpretation of some process that does fulfill the criteria.. and moreover, it's the lynchpin of his energy gain principle, if not an energy-gain mechanism itself.
Another potential intimation is that this process is, in some sense, 'non-mechanical'.. tho this seems less consistent; i suspect the intended point is simply that it's not a prospective GPE asymmetry.
The form of excess impetus we should be looking for is the one that achieves the same net rise in system momentum from the same internal work done each cycle, invariant of some range of RPM.
Somehow, the solution to the riddle has to furnish us with that..
- eccentrically1
- Addict
- Posts: 3166
- Joined: Sat Jun 11, 2011 10:25 pm
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
Re: re: Decoupling Per-Cycle Momemtum Yields From RPM
MT 13 features an internal stator, so would be exchanging momentum with Earth. MT 21 doesn't have this problem, satisfying the EMGAT principle, but obvs needs further input energy..Fletcher wrote:Hi Mr V .. I think basically you are on to it. You got more than a toe hold.
We know a wheel is a closed system where weights shift around, be that radially or otherwise. The ol' "Height for Width" conundrum. And that is what Wagner refers to imo. And for Agor95 why Wagner is also right !
We also know that because gravity force is conservative the path of weights makes not a jot of difference to the NET Energy and Momentum sums. The ol' What Goes Up Must Come Down axiom. Or put in a way I prefer to think about it "What Goes Down Must Come Up An Equal Vertical Displacement". The ol' WGDMCUAEVD which is far too long to catch on like H4W or W4H lol.
IOW's, the GPE is recycled or restored in each complete revolution, or even each individual sector depending on your non-asymmetric (or temporary torque) OB system. Even if weights were lifted at tdc etc by an unknown force it would still follow a defined path that circulates GPE and repeats. This assumes for expediency no frictional energy losses like a sim can achieve.
================
B. tells us his wheels generate innate excess impetus. Excess force, excess "weight", preponderance etc.
================
Karl tells us it turns from innate momentum IIRC.
DT Page 192-195 John Collins
By The Grace Of God, We, Karl, Landgrave Of Hessen, Prince Of Hersfeld, Count Of Katzenelnbogen, Dietz, Ziegenhaven, Nidda And Schaumburg, hereby make the following testimony and proclamation: -
"is a revolving wheel, which is able to run, by means of its own innate momentum"
Kassel, 27th May, 1718 KARL
And this is where B. is also right !
================
So excess Momentum imo is the answer we seek, as you deduce. Produced by some internal ingenious mechanical arrangement where it is both easy to understand the mechanical actions when seen (but maybe not to understand the Physics reasons for the PM Principle at first glance), and simple to build, according to Karl's quote to his ministers in the Bernoulli letter.
Somewhere in the Laws of Newtonian Physics and Newtonian Mechanics is a "workaround" mechanical condition that leads to a natural gain in system Angular Momentum, IMO. An example might be MT13 where excess torque is produced in all of wheel by lifting the lws early full height at tdc, or MT21 side lift to higher position. These require a readily usable force generated internally and redirected for purpose.
There are limited possibilities and, as you do, you work thru them one by one looking for the chink in the armour to be exploited for a real physical advantage. A cautionary note. The mechanical actions (but not necessarily the mechanical PM principle) have to be easy to understand once seen and simple to build. Find it first and refine it to its simplest buildable form, which is also your sensible approach.
"Findeth the PM Principle, cometh the wheel."
.................................
MT137 :
I have on occasion used a similar line drawing method when building a physical wheel to mark out equi-distant circumference points for sectors. It's a bit cumbersome imo. I usually just use a protractor and ruler etc.
It also makes a good template for building gears imo, tho again I use free online gear templates off the web.
Thirdly it symbolizes imo a balanced torque neutral arrangement which of course it must be because it has complete symmetry.
And oops i was actually thinking of MT 143 "pounds in equilibrium" tho i've also been thinking more about 137 too (without much progress tho) - the Roberval thing, since MT 134 also seems to feature a similar (but weirder) planar linkage, and MT's 133 & 134 seem to relate to the Toys page, as well as these mysterious 'quarters'..
The key necessity of inertial isolation (EMGAT / no stators) further elucidates the following paragraph from AP (emphases mine):
When torquing against a stator, the rising relative speed between it and the rotor is distributing the same given torque over a squaring range of angle; IOW maintaining constant acceleration means increasing the torque per unit angle to compensate, thus enforcing unity.The wheel's own inner force must come into being without
external momentum being applied by such devices. It must,
simply put, just revolve, without being wound-up, through the
principle of "excess weight", as I describe in Part I.
The whole point of going statorless is that momentum gained sans-stator is immediately freed from this practical constraint - essential for fixing the unit energy cost of momentum to a minimum value invariant of RPM.
However gaining momentum from G*t per kiiking - where the gain in rotKE is equal to the net work done against CF force - comes up against the same fundamental problem; CF force squares with RPM, so once again input work / PE tracking output KE.
So the '4 ounces fall / 1 pound rises' riddle must refer to some other way of gaining momentum from G*t.
It can't come from anywhere else - cuz N1. G*t is the only possible momentum source / sink in a statorless vertical wheel.
Thus the only practical objective between us and OU is fixing the energy cost of that momentum, so that it doesn't square up with rising RPM. Even tho KE always does. Breaking PE:KE symmetry by buying momentum at low velocity-cost, but for accumulated-velocity KE value; rotKE = ½Iw² so the min value of a 1 rad/s acceleration of a 1 kg-m² system is ½ J: 10 of those costs a net input PE of 5 J, leaving us with a 1 kg-m² system at 10 rad/s and thus having 50 J of rotKE.
That's actually a crudely over-optimistic simplification, but really does convey the plain mathematical mechanics; in reality of course we're looking for an interaction that's 75% inefficient, so maybe 2 J / 1 kg-m² / 1 rad/s, or something in that area - again, it only has to be constant across some finite range of RPM, not necessarily 'cheap' on a per-cycle basis - the key dynamic simply that input energy is scaling linearly with velocity while output KE is squaring, hence for any given per-cycle efficiency there's a break-even point where the exponential curve intersects the straight-line diagonal plot of input work, and then shoots straight up thru it - everything under the curve in front being loss, and everything behind, gain..
So far i've considered two different potential solutions that would result in this 25% per-cycle efficiency:
• a 3:1 ratio between inertias in an asymmetric inertial interaction
This could for instance resolve to a reactionless acceleration applied to one of four masses (or conversely, the other three), followed by a collision that redistributes that momentum gain back amongst all four; this would duly dissipate 75% of input energy each cycle..
• alternatively, consider an interaction in which half the input work each cycle generated some momentum, while the other half had to make equal opposite counter-momentum of equal value - so, a 1:1 inertia ratio - and this counter-momentum portion has to be sunk to gravity in order to thus seize its orphaned positive component; so half the input energy's sunk to gravity each cycle, and then we lose another 50% of whatever's left in the consolidating collision, for a net 75% loss
Suffice to say i've not yet found a mechanical arrangement that can embody either solution..
But this mysterious 'quarters' riddle from AP hints at some other solution i've hitherto overlooked..
I don't know what it is, yet - mind's racing looking for anything that might match, only so many jigsaw pieces left in the box (and i think we have most of the edges done)..
How about that central radial displacement thru the axle? A solid post like a stamper might reasonably cause an MoI variation of a factor of two, double the MoI thus halving the speed and quartering the energy... dunno, but these kinds of dynamics seem to fit - logically, it's got to refer either to a momentum exchange - so, trading I and w components - or else, the means of actually nabbing that momentum from G*t in the first place.. either way, tho, in a manner that provides some degree of RPM-invariance on the input workload / momentum yield.
-
- Addict
- Posts: 2879
- Joined: Sat Jul 31, 2010 12:19 am
- Location: W3
Again, cool meditation:
• drop ball bearings by hand onto a simple 'mill wheel', letting them roll off in a big, spreading mess..
Give the wheel some decent mass so each ball only imparts so much momentum. Make the bearings frictionless.
The point is that you're not picking any of them back up - no momentum is being shed back to gravity for the lifts; all we're doing is compounding our momentum gains by successively dropping more and more output GPE's. Or else, fuggit - pick 'em up by hand, to close the loop; so long as their lifting applies no load to the wheel, so that it's only ever under positive gravitational torque..
• even tho no momentum at all is being given back to gravity, the system never breaks unity!
Instead what happens is that due to the ever-rising speed, each ball simply spends less time under gravity's time-constant acceleration than its predecessors, thus as RPM's rise, per-cycle momentum yields fall..
This drop-off in p/c momentum yields perfectly tracks ½Iw² - IOW PE is locked to KE via the diminishing momentum returns on the constant p/c input GPE, diminishing by the inverse square of the RPM rise.
Think about what this means - it seems quite a powerful insight, to me: no passively over-balancing scheme can ever achieve OU levels of momentum gain...
That is, the momentum we get from passive OB can only be had at unity efficiency; it's always going to cost the same input work to raise it, but the amount of momentum it can induce in falling is always going to reduce with rising RPM, thus enforcing unity efficiency..
OU is impossible from passive OB.
Yet this is only to reiterate that effective GPE asymmetries are not on the cards..
As much as it tells us what isn't gonna work, it likewise points us towards the only possible way forwards:
• we can only aspire to fix the work / energy cost of additional momentum we introduce to the descent, over and above that which is provided by a drop's natural G-time alone
The momentum we get from over-balancing is only ever going to be had at unity efficiency; we have to try and make bread on a sideline operation - orchestrating additional momentum - from G*t (nowhere else) - at constant unit-energy cost and yadda yadda collide and consolidate.. it's that, that we might be able to control terms on.
This would of course be consistent with Wolff's impression of the OB weights landing with more momentum than could've been imparted by their gravitation alone:
• it's that momentum the weights land with that we need to fix the cost of
It's tricky here to go with the Merseburg or Weissenstein two-way wheels, since these "began to turn as soon as an internal weight was heard to begin falling".
What is that sound, exactly? Well, we know the wheels emitted 'scraping' sounds, "as if parts or poles were being pulled over one another", but surely, the 'sound of something falling' is pretty much any sound that terminates in a collision - quite possibly, with an accelerating attack slope..
IOW, the 'sounding weight' that begins falling is the same weight heard landing on the descending side.
This is where the probing power of logic begins to fog somewhat:
• if the wheel begins accelerating as soon as the weight begins its descent, then is it descending faster or slower than the wheel is accelerating?
Because if faster, maybe it's torquing the wheel against some other inertia, or perhaps even sinking the counter-torque to gravity..
..whereas if slower, the torque is being applied between it and the wheel - basically using it as a gravitating 'stator', and hence using its weight to sink counter-momenta, perhaps.
But then this also may determine whether the weight 'lands' on the wheel / its rimstop, rather than the wheel itself catching up with the slower-moving descending weight..
Further clouding the issue is the two-way nature of the wheels, which must rely on some kind of process to initiate continual OB in whichever direction, from its perfectly-balanced standing start - IOW are any of these sound effects those of the energy-gain mechanics, rather than the priming apparatus or other actions incidental to the energy-gain principle? We can't be sure.
The sounds made by the one-way wheels are thus that much more useful to go on, tho besides 'scraping sounds' i'm not sure what more details there are - did the Gera wheel make banging sounds? I need to check up on this..
It would seem most consistent however if the over-balancing weights landed on their rimstops and imparted fixed-price momentum gains in the process, having been somehow endowed with reactionless momentum from G*t. The rest of the OB game being a zero-sum.. just making on that extra lil' bump of fixed-cost momentum each collision..
The thing is, where's the under-balancing weight in the tied-off & stationary Gera wheel? Already lifted, or part-way up? Because when the system's turning, that UB weight has angular momentum, and when it undergoes a radial lift that momentum must be conserved..
..could this 'additional' momentum being imparted by the weights landing on the descending side be comprised in some part or whole by that given up by the under-balancing weights radial retraction?
I don't see how 'the riddle' would apply here, tho.. that whole '¼:1' thing.
Still, something semi-chewable to grind on for a while anyhoos..
• drop ball bearings by hand onto a simple 'mill wheel', letting them roll off in a big, spreading mess..
Give the wheel some decent mass so each ball only imparts so much momentum. Make the bearings frictionless.
The point is that you're not picking any of them back up - no momentum is being shed back to gravity for the lifts; all we're doing is compounding our momentum gains by successively dropping more and more output GPE's. Or else, fuggit - pick 'em up by hand, to close the loop; so long as their lifting applies no load to the wheel, so that it's only ever under positive gravitational torque..
• even tho no momentum at all is being given back to gravity, the system never breaks unity!
Instead what happens is that due to the ever-rising speed, each ball simply spends less time under gravity's time-constant acceleration than its predecessors, thus as RPM's rise, per-cycle momentum yields fall..
This drop-off in p/c momentum yields perfectly tracks ½Iw² - IOW PE is locked to KE via the diminishing momentum returns on the constant p/c input GPE, diminishing by the inverse square of the RPM rise.
Think about what this means - it seems quite a powerful insight, to me: no passively over-balancing scheme can ever achieve OU levels of momentum gain...
That is, the momentum we get from passive OB can only be had at unity efficiency; it's always going to cost the same input work to raise it, but the amount of momentum it can induce in falling is always going to reduce with rising RPM, thus enforcing unity efficiency..
OU is impossible from passive OB.
Yet this is only to reiterate that effective GPE asymmetries are not on the cards..
As much as it tells us what isn't gonna work, it likewise points us towards the only possible way forwards:
• we can only aspire to fix the work / energy cost of additional momentum we introduce to the descent, over and above that which is provided by a drop's natural G-time alone
The momentum we get from over-balancing is only ever going to be had at unity efficiency; we have to try and make bread on a sideline operation - orchestrating additional momentum - from G*t (nowhere else) - at constant unit-energy cost and yadda yadda collide and consolidate.. it's that, that we might be able to control terms on.
This would of course be consistent with Wolff's impression of the OB weights landing with more momentum than could've been imparted by their gravitation alone:
• it's that momentum the weights land with that we need to fix the cost of
It's tricky here to go with the Merseburg or Weissenstein two-way wheels, since these "began to turn as soon as an internal weight was heard to begin falling".
What is that sound, exactly? Well, we know the wheels emitted 'scraping' sounds, "as if parts or poles were being pulled over one another", but surely, the 'sound of something falling' is pretty much any sound that terminates in a collision - quite possibly, with an accelerating attack slope..
IOW, the 'sounding weight' that begins falling is the same weight heard landing on the descending side.
This is where the probing power of logic begins to fog somewhat:
• if the wheel begins accelerating as soon as the weight begins its descent, then is it descending faster or slower than the wheel is accelerating?
Because if faster, maybe it's torquing the wheel against some other inertia, or perhaps even sinking the counter-torque to gravity..
..whereas if slower, the torque is being applied between it and the wheel - basically using it as a gravitating 'stator', and hence using its weight to sink counter-momenta, perhaps.
But then this also may determine whether the weight 'lands' on the wheel / its rimstop, rather than the wheel itself catching up with the slower-moving descending weight..
Further clouding the issue is the two-way nature of the wheels, which must rely on some kind of process to initiate continual OB in whichever direction, from its perfectly-balanced standing start - IOW are any of these sound effects those of the energy-gain mechanics, rather than the priming apparatus or other actions incidental to the energy-gain principle? We can't be sure.
The sounds made by the one-way wheels are thus that much more useful to go on, tho besides 'scraping sounds' i'm not sure what more details there are - did the Gera wheel make banging sounds? I need to check up on this..
It would seem most consistent however if the over-balancing weights landed on their rimstops and imparted fixed-price momentum gains in the process, having been somehow endowed with reactionless momentum from G*t. The rest of the OB game being a zero-sum.. just making on that extra lil' bump of fixed-cost momentum each collision..
The thing is, where's the under-balancing weight in the tied-off & stationary Gera wheel? Already lifted, or part-way up? Because when the system's turning, that UB weight has angular momentum, and when it undergoes a radial lift that momentum must be conserved..
..could this 'additional' momentum being imparted by the weights landing on the descending side be comprised in some part or whole by that given up by the under-balancing weights radial retraction?
I don't see how 'the riddle' would apply here, tho.. that whole '¼:1' thing.
Still, something semi-chewable to grind on for a while anyhoos..
re: Decoupling Per-Cycle Momemtum Yields From RPM
Bessler said he used different "principles" in his different wheels. So don't get hung up on which OB system he used, imo.
Take a look at MT20 and MT's 44 and 48. Completely different operational OB systems.
The common factor is the Prime Mover.
Where we agree is that the Prime Mover costs less than it produces and that which is deployed to the various OB systems to create surplus torque, imo.
Regardless, if we each pull it all apart (in agreement or not on what those significant parts are) eventually someone will put it back together again like Humpty Dumpty on the wall. So discussion is always worth having.
Take a look at MT20 and MT's 44 and 48. Completely different operational OB systems.
The common factor is the Prime Mover.
Where we agree is that the Prime Mover costs less than it produces and that which is deployed to the various OB systems to create surplus torque, imo.
Regardless, if we each pull it all apart (in agreement or not on what those significant parts are) eventually someone will put it back together again like Humpty Dumpty on the wall. So discussion is always worth having.
-
- Devotee
- Posts: 1373
- Joined: Thu Mar 09, 2006 2:34 am
- Location: Wisconsin, U.S.A.
The weather has been so beautiful the last few days. Tomorrow I'm going to finish cleaning up the leaves in front. It's so weird how the maple in front loses all of it's leaves before the maple in back even starts. It's just, that they're the same thing, but their behavior is different from one side to the other. It's been a consistent pattern that I've observed. It's helpful, because this way I only have to care about one section at a time. In the end, everything falls. It always falls. There's nothing I can do about that. It's just its nature. It's a beautiful sight, every time, all of these years.
re: Decoupling Per-Cycle Momemtum Yields From RPM
That is perpetual momentum we can all appreciate.
Regards
Regards
[MP] Mobiles that perpetuate - external energy allowed