energy producing experiments

[Wubbly]

Nick's experiment is from this thread HERE.

... if all of the "motion", and by that pequaide means "momentum", is given to the smaller mass, you come out with energy gain.

All we have to do is run some simple numbers. Let's say you have a large mass, which is actually two masses M1 and M2. They are travelling at a velocity V1. They would have a momentum of p1 = (m1 + m2) * V1

If all of the "motion", and by that pequaide means "momentum", is given to one of the masses (let's call that mass m2), It would have a new velocity (let's call that v2), with a new momentum of p2 = m2 * v2. According of pequaide's hypothesis, there should be energy gain.



You don't even need a huge mass difference to get energy gain. All you need to do is set m1 equal to m2 and run some simple numbers with the momentum transferrance equation (m1 + m2) * v1 = m2 * v2 which is actually just the reverse of the ballistic pendulum equation.


The Initial momentum of the "heavy" mass is (before transferrance)
p1 = (m1 + m2) * v1 but since m1 = m2 we get
p1 = (m1 + m1) * v1
p1 = 2 * m1 * v1

The final momentum of the "light" mass is (after transferrance)
p2 = m2 * v2 (but since m1 = m2, then)
p2 = m1 * v2

Since you are transferring all of the momentum p1 to the mass m2, then p1 would have to equal p2.
Setting p1 equal to p2 and solving you get

p1 = p2
2 * m1 * v1 = m1 * v2
or
v2 = 2 * v1 Your velocity doubled. Mass m2 is now going twice as fast as the heavy mass.

Now let's calculate Kinetic Energy
Initial kinetic energy is
KE = 0.5 (m1 + m2) * v1 * v1 but since m1 = m2 we get
KE = 0.5 (m1 + m1) * v1 * v1
KE = 0.5 (2 * m1) * v1 * v1
KE = m1 * v1 * v1

kinetic enrgy after the momentum transferrance is:
KE = 0.5 (m2) v2 * v2 but since m2 = m1, and since v2 = 2 * v1 we get
KE = 0.5 m1 (2*v1)(2*v1)
KE = 0.5 m1 4 * v1 * v1
KE = 2 * m1 * v1 * v1

The kinetic energy doubled. Did anyone notice that the kinetic energy just doubled?

All you need is the larger mass to be only twice the smaller (flung) mass, and if pequaide's hypothesis were correct, you would double your energy. You would get a 100% energy increase.

In Nick's experiment, where m1 = m2, what was found? After all was said and done, ZERO energy gain was found. Zero, nada, zilch. It has measurable inputs, measurable outputs, a hypothetical energy increase of perhaps 100%, (using pequaide's momentum transferrance hypothesis), and yet zero energy increase was found. If pequaide's momentum transferrance hypothesis were correct, Nick's experiment should have showed some energy increase. It showed none.

And to add insult to injury, the "fall time" is longer than the "rise time". It even satisfies the "slow fall", "fast rise" hypothesis espoused by some, but with zero energy gain.

[jim_mich]

In Nick's experiment, where m1 = m2, what was found? After all was said and done, ZERO energy gain was found.

I disagree with the interpretation of the results. Two weights fell from the top of the wheel until even with the wheel axle, where they disengaged from the wheel and were no longer causing wheel rotation. From that point the tethered weight fell lower and then rose back up to the center of the wheel axle, thus the rise cancels the fall from the axle. From the wheel's axle the weight rose to a height of four times its original fall distance. This shows a clear energy gain, contrary to the ZERO energy gain that is claimed. The energy gain is consistent with conservation of momentum and not conservation of KE.

[ovyyus]

Jim, my replication of Nick's experiment could not achieve his seemingly extraordinary results. IIRC, it was discovered that Nick was inaccurately measuring his weight and wheel balance. When corrected there was no extraordinary result. The above pictures are of the flawed test setup and are therefore irrelevant.

I've attached two diagrams. Diagram 1 shows Nick's setup and extraordinary result using his inaccurate measures (diagram derived from his posted photos). Diagram 2 show the actual result (my replication) using accurate measures. Seems clear enough.

[Wubbly]

Well then we'll just have to disagree, Jim. That's what makes a market. There is the mass still attached to the rim that continued to fall from the 3 o'clock position down to the 5 or 6 o'clock, which you can't ignore. Also, the flung mass in Nick's original experiment got up as high as it did because the mass attached to the rim was heavier than the flung mass. In subsequent experiments, by others on the forum, where the masses had a similar weight, the flung mass only got up to below your fourth line drawn (counting from the bottom).

If I call the 6 o'clock position zero, and measure in units of R, then you start out with two masses at a height of R = 2, and you end up with one mass at a height of around R = 0, and one mass at around a height of around R = 4. Clearly zero energy gain. Experiments done by others had a heavier background mass (plywood) which should have contributed to the ball flying even higher (if the pequaide hypothesis of momentum transferrance were correct). Also, some of those other experiments had the flung mass release lower than the axel with the flung mass getting no higher than the other experiments. I would still have to disagree with your opinion, Jim.

[Wubbly]

If the above math is correct, you only need the heavy mass to be twice the flung mass to get an energy increase of 100%. If the plywood mass (in the experiments presented in Nick's close shave thread) is included in the heavy mass, the heavy mass is easily more than twice the flung mass. If you start out with M1 and M2 at a height of R = 2, for the energy in the experiment to double, you would need M1 at a height of R = 0, and M2 at a height of R = 8. Clearly the energy did not even come close to doubling (as predicted by the pequaide momentum transferrance hypothesis), let alone increase.

[pequaide]

Jim is correct: the ball drops off the wheel at 3 o’clock; it accelerates the wheel for only one R. So the ball drops 1 R and the attached mass drops 2 R. this gives us a total drop of 3 R and a rise of 4 R. That is a gain of one R. Four R / 3 R = 133%; it drives for 3 R and is rewarded with a rise of 4 R.

In the highest potential energy position the ball is 3 R above its original position and the other equal mass is at 2 R below. Again: this is a gain of 1 R.

Plus: we know that the ball could have risen higher if it were released from the wheel.

And: is the wheel stopped when the attached mass is at 6 o’clock?

Is the wheel moving when the ball is at its highest position?

I have a few thoughts about this experiment.

If the ball has achieved its maximum possible height then there will be no lateral motion; because the lateral motion could have been used to help the ball rise. Remember small increases in velocity cause larger increases in height. An increase in velocity from 3.7 m/sec to 4.0 m/sec (8.1%) will cause an increase in height from .6977 m to .8155 (16.9%). If all the energy is used for rise the ball would fall straight down from its maximum achieved height, and this condition is almost met.

After looking more closely I believe that the ball is beginning to come off the wheel at 2 o’clock not 3 o’clock. If the ball is already sliding at 2 o’clock its drive contribution is negligible between 2 and 3; and after 3 it has no drive contribution. So the ball has less than one R drive.

Also the attached mass is not at 6 o’clock, when the ball is at the high point, it is at about 5 o’clock. I believe this is mentioned by others but then not used in the calculations. The attached mass does not fall 2 R it falls less than 2 R.

Also: I see the wheel moving backwards. This could greatly reduce the height of the throw.

With the fact that the ball does not drive even 1 R and the attached mass does not drive even 2 R and you have backward movement of the wheel, I think the prediction that this is a 33% increase in energy is very reasonable.

Great experiment Nick.

[nicbordeaux]

The weights were indeed wrong when I worked up the guts to dissemble the thing, at fixed driver weight 1.5 for 1 of flung. The guts, because when you hit on an "anomaly" , invariably if you take the device apart, you can never replicate the results, a small change in position, anything... and it don't work.

Wubbly is entirely right, 1 to 1 is no go in the setup used. Two to one is borderline 100 or unity. Around 2.5 to one, there's more experimentation needed. This because the velocity of the wheel seems to command fling height in a non-linear manner. Meaning that as you get faster circ speed, your tether mass appears to be gaining on a "logarithmic" (sounds clever, dunnit ?) manner. And the faster wheel speed arises from more fixed drive weight to push it from zero to maximum: quicker acceleration and bigger end speed.

I offer no proof for this, no vids, no claims. Call it an impression, that will hopefully avoid trouble.

I do however have one question, actually been pondering over it (vaguely) a long while because it fits in with another experiment : a soccer ball given a real heavy kick is travelling at speed "x" and on some trajectory, when it hits a wet slippery pitch and bounces forwards it picks up speed. This must be at the expense of height and trajectory. Is that correct ?

ps : actually, there is another question : if anyone knows the latin for "It's too bloody cold to be out in the shed or garden mucking around with bouncy balls, this will have to wait a bit" , I'd like to use that as a signature.

[jim_mich]

It is currently 20ºF (-7ºC) in Michigan right now, which is too cold to be outside!

[Fletcher]

Jim is correct: the ball drops off the wheel at 3 o’clock; it accelerates the wheel for only one R. So the ball drops 1 R and the attached mass drops 2 R. this gives us a total drop of 3 R and a rise of 4 R. That is a gain of one R. Four R / 3 R = 133%; it drives for 3 R and is rewarded with a rise of 4 R.

In the highest potential energy position the ball is 3 R above its original position and the other equal mass is at 2 R below. Again: this is a gain of 1 R.

Plus: we know that the ball could have risen higher if it were released from the wheel.

And: is the wheel stopped when the attached mass is at 6 o’clock?

Is the wheel moving when the ball is at its highest position?

I have a few thoughts about this experiment.

If the ball has achieved its maximum possible height then there will be no lateral motion; because the lateral motion could have been used to help the ball rise. Remember small increases in velocity cause larger increases in height. An increase in velocity from 3.7 m/sec to 4.0 m/sec (8.1%) will cause an increase in height from .6977 m to .8155 (16.9%). If all the energy is used for rise the ball would fall straight down from its maximum achieved height, and this condition is almost met.

After looking more closely I believe that the ball is beginning to come off the wheel at 2 o’clock not 3 o’clock. If the ball is already sliding at 2 o’clock its drive contribution is negligible between 2 and 3; and after 3 it has no drive contribution. So the ball has less than one R drive.

Also the attached mass is not at 6 o’clock, when the ball is at the high point, it is at about 5 o’clock. I believe this is mentioned by others but then not used in the calculations. The attached mass does not fall 2 R it falls less than 2 R.

Also: I see the wheel moving backwards. This could greatly reduce the height of the throw.

With the fact that the ball does not drive even 1 R and the attached mass does not drive even 2 R and you have backward movement of the wheel, I think the prediction that this is a 33% increase in energy is very reasonable.

Great experiment Nick.

pequaide .. now that you've found some enthusiasm for Nick's experiment, this time around, take that open mindedness & build your own version with the tweaks you mention, to your comfort.

For instance a peg & tether release mech [remember to put a peg on the opposite side to counter balance] so that the flung mass can rise unfettered - then you might add a simple ratchet to halt the wheel & drive mass at its lowest position or whatever - you could increase or decrease the moment of inertia for the wheel to alter the acceleration of the wheel & drive mass etc to test out your 'more time' theory.

Use your bb bags & a known drive mass etc.

Then video it & analyse the results by frame or with the aid of photo-gates if you desire.

What was found by Nick, ovvyus & wubbly [who replicated it & analysed it] was there was no Net Energy [Pe] gain at all when all the Pe of all the masses was calculated.

If you find any different then a closer look would be required, as happened with Nick's excellent experiments, but at least the anomalous results were critically & logically explained.

[pequaide]

I saw were some mentioned that the drive mass was heavier but since Nick had not said so himself I had not responded. Thank you for the correction. So if the drive mass is actually 1.5 then we are looking directly at unity. Except that the 1.5 does not drop 2 R and the ball does not drive for 1 R and there is backward motion of the wheel. So you are producing a small amount of energy.

I have found that when the ball slides off and drops limply, allowing the tether to loosen, (as it does in this experiment) the throw is pathetic.

This statement is correct Nick: “This because the velocity of the wheel seems to command fling height in a non-linear manner.�

I think you will find that doubling the mass of the wheel at the same RPM will also have non-linear results on the height.

[pequaide]

I have been doing so thinking. Are we good scientists if we don’t admit what we see?

We are at unity (input energy = output energy) if the 1.5 mass falls to 6 o’clock. But it does not fall to 6 o’clock it falls to only 5 o’clock. This experiment is already producing energy. The 1.5 mass will generate a useful amount of momentum when it rotates from 5 to 6.

Second: we don’t care how far the ball mass drives, we only care where it ends up. The ball mass could be physically attached to the wheel and it then would not begin sliding at 2 o’clock. We could in fact make it drive to 4 o’clock and release it there. This would place the end position of the 1.5 mass at 6 o’clock, where it belongs. The drive angle between 2 and 4 is your premium drive angle and would produce substantial amounts of momentum.

Third: The ball is pulling the wheel backwards; this means that the tether has not reached a 90° angle to a tangent line at the point of connection. Therefore the ball has not achieved maximum height when it begins pulling the wheel backwards. This robs the ball of a substantial quantity of momentum (velocity). And we are already passed unity.

Getting all these angles right is quite a challenge. I guess that is why my early experiments were done in a horizontal plane.

[broli]

I guess that is why my early experiments were done in a horizontal plane.

In my opinion that's the only right way to do these experiments. Some day someone will perform a real quantitative experiment and end the debate once and for all. I'm still as baffled as many months ago that this has yet to be done. But prepare yourself peq.

[broli]

I also forgot to mention something about simulators. Let me tell you why any physics simulator out there will never show an energy gain.

It's called constraint based physics. Newton or Leibniz law's don't really define constraints. A constraint could be anything. For instance if you attach a rod to two masses, that's a constraint of distance. These objects will thus be simulated using newton's law and their position, velocity and acceleration is adjusted continuously in order to never break these physical constraints.

Now any physics engine out there uses additional non physical constraints on top of the physical ones. One notable constraint is the energy constraint. Any time you attach an object with a rod to/or with another object in wm2d for instance, you automatically put an energy constraint on the system on top of your three position/velocity/acceleration (PVA) constraints. In theory nothing can break that constraint, however in practice it depends more on the simulation's time step size. If they do break wm2d usually gives this famous error:

"Inconsistent constraints or physical instability has been detected. Some constraints may be ignored. Do you wish to continue."

And it either explodes or does questionable things afterwards.

In other words all these simulation software will do anything they can in order to abide by these constraints. This is physics engine 101. There are many mathematical solver techniques to continuously adjust bodies based on these constraints (see linear equation matrix solving) but all have the energy constraint at their heart.

So in order to get results only based on inertial motion, that is using only a PSA constraint you either use a physics engine where you can disable the energy constraint, or you write your own. However the same goes for momentum, you can't use a momentum constraint as that would defeat the point.

Here's a good read in constraint based physics:
http://www.cs.cmu.edu/~baraff/pbm/constraints.pdf

[path_finder]

A first element of answer: this PE-KE conversion experiment (see it again):
http://www.youtube.com/watch?v=GuTMYgQDUzs

[Fletcher]

pequaide .. you are in a position to 'tweak' that setup anyway you want, till you are satisfied - not with the result perhaps, but with the elements & the relationships you talk about - alter the tether length & positions of things, change the moments of inertia, add latches & ratchets - do whatever is required to be a fair test that will give you a maximum result.

Then let the chips fall where they may.

Your theory is either correct & can be reliably demonstrated & replicated in some circumstances [the proof], or not.

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

broli .. your thoughts about physics programs make sense - however, I view them as a tool to learn about actions & reactions - from studying them it often becomes apparent exactly what is happening & my illusion of what I thought would happen is dissolved or at least modified.

If I had anything that showed close to OU on a physics program & was not a simple & easily understood device I'd build it - I've done this a few times over the years & retro'd the sim with the physical dimensions & masses etc, then compared results - usually some small sim advantage disappears when made accurate & it mirrors the real build pretty closely.

If I saw something that the sim produced that I disagreed with, seeing more potential in it than the sim produced, I'd build it to gain the knowledge, safely knowing that the sim would object to the findings, if there were some of note, because of those constraints.