Gravity assisted Normal Energy

[jim_mich]

Blimey, this can send you nuts!

start tearing your hair out, because you'll soon find, you splendid mechanic, that this is a nut you can't crack!

Once again Bessler was writing about gravity-wheels not working. You can't crack this nut by trying to use gravity.

Cheers,

[ME]

It's possible the KAS-mechanism is as efficient as you can get: In terms of GPE vs (R)KE.

Come on ME! Where's you sense of invitation? This is nowhere near as efficient as you can get

:-)
I must have traded it for "the benefit of the doubt"....

Let's face it, it has to be connected to a secondary device in some way. This design has turned that on its head by ensuring that the contact anchor point is fluid and free to find its own position.

Sounds good.
"normally" a mechanism is only able to change 1 parameter (1D) along some path it takes - even though it not always looks like it, or not designed that way.
When highly connected to the orientation of the wheel, such mechanism actually lacks choice: thus it simply finds balance somehow, somewhere.

Having two parameters seems good:
Still having that first parameter 'ensuring' overbalance for the wheel and thus connected to the wheel orientation;
The second (fluid) parameter 'ensuring' balance for the mechanism itself, where ever that mechanism might be on the wheel.
Their 'fight' should result in rotation.

(B) secondly, it is virtually friction free. If we discount air resistance for a moment, I have never been able to reduce contact friction to this low level before which has to be encouraging.

Well latency of a balancing effect could also be beneficial.

This is because I may have discovered a way to raise (yes raise!) the weight to a higher level and only by utilizing the Ke generated by the same weight. Blimey, this can send you nuts!

Listening :-)

[KAS]

A possible modification to the arc rocker could improve efficiency if it proves feesable. You may have noted from my drawings, descriptions and Mick's video that there is clear evidence of acceleration and increased Ke toward the end of the moment just before the weight falls away. This acceleration could be increased if, as ME pointed out, a slight weight overhang was evident.
Enter the next object from Bessler's toy page; that of the parallelogram puppet toy.
If we add this set up to each end of the rocket, I believe it may add two improvements.
(A) a bias overhang containing the weight should encourage the oob moment to start earlier or at a Lower angle.
(B) the hinged section on the tail of the arc could lift the weight at the end of the moment when the Ke produced is at its greatest.

The Bob hinge strut can be added to the top of the arc but the tail of the arc will need to be shortened by approx 10deg to cater for the lower hinge bar.

Forgive the rough sketch but I am in the process of purchasing a technical drawing software package.

ME, Fletch, Mick, do you think you could try this mod out on your exiting sim to check out its feesability? If you are unable to do this, I will try and manufacture a 360 deg testing jig to test out this new set up and to check if the reset if affected.

[KAS]

Blimey, this can send you nuts!

start tearing your hair out, because you'll soon find, you splendid mechanic, that this is a nut you can't crack!

Once again Bessler was writing about gravity-wheels not working. You can't crack this nut by trying to use gravity.

Cheers,

Not trying to use gravity Jim. That has Issues with COE laws. Trying to harness Nf. (Normal force).

[Fletcher]

Sorry for the delay.

Here is my sim of KAS's device on a 30 degree ramp - very similar to ME's (used the same 0.5 meter radius, Blue (Clear) mass 1 kg, Red Arc Mass 0.045 kg (ratio 22:1).

N.B. the CoM for the Arc and the Blue mass are approximately 3/5ths of radius.

Note the Arc CoM shows as does the System CoM (that's why the Blue mass is clear so you can see it).

It will roll on thru and down only if the Arc mass is increased.

Right now the Blue mass has varying amounts of KE .. BUT .. it does not exceed the System loss of GPE at any time. This is the result for sim world !

If this is correct it means the Nf is not being harnessed in this arrangement to give Blue mass KE > System GPE lost as KAS would like.

Feel free to alter and play with the sim in any way.

[Mark]

re: KAS-ME1-End.png

A 1 kg Blue mass and a 0.045 kg Arc are firmly affixed to each other.
The Blue mass CoM and the Arc CoM are at the same radius, but on radii 90 degrees apart.
The Blue mass CoM and the System CoM are in very close proximity to each other.

But the Blue mass KE peaks later than the System KE?

Is the GPE [negative] peak centered between, and overlaping them both?


Or am I just overanalizing ?

[Fletcher]

OK .. I turned all energy signs into positive for direct comparison side by side.

[KAS]

Thanks Fletch.

Could you tell me if this data is for the full half turn? It looks as though from the graph that it has only rotated a a quarter turn or to the point of slow down that ME pick up on his version.

Thanks

Kas

[Mark]

Cool. Thanks, Fletcher.

Let me ask my question a different way...
If the Blue doesn't move in relation to the Arc, and the Blue and System CoM's are practically at the same spot, why the delayed peak?

I get the feeling that I'm ignorant of something that's obvious to everyone else. :-/

Mind you, I'm not familiar with sim representations, maybe this is just normal?

If I were looking at a physical model doing it's thing, I wouldn't be able to perceive this [admittedly small] aberration.

Kudos to Kas for sharing this device, and to you guys for working through the thought process. It's quite interesting.

[ME]

Good one Fletcher, I see you used a polygon - I was less adventurous :-)

In my sim the Main-weight was located off-axis by 0.46°, and had a radius 0.304 (or 0.608 times the radius) -It seems I just arbitrarily moved it into position in that version)

I spend some time fiddling around with some parameters, and I found (within my sim) a relation (regression line) between the weight and the radius (I call it sim-science) to get things rolling at 30°.
When it's able to go over that hill it just goes - as Mickegg noticed.
---
(m= arc mass, r= arc radius, M= Main-mass, R=operating radius of Main mass)

For your simulation with (m=0.045 kg, r=0.49 m)
The Main-weight radius is determined to be about (R=0.31195 m)
I calculated: for r=0.490 m -->The main Mass (M<=0.5360 kg)
I calculated: for r=0.495 m -->The main Mass (M<=0.6015 kg)

Manual sim adjustments (with ramped-up accuracy and friction) gave (M<=0.601 kg).

With (M=1) I calculated (R<=0.3) - your sim doesn't roll.

---
According to my calculation on Mickegg's device, it should have worked - WM2D agrees.

m=62, r=(110+3.3)/2 = 56.65 (when 110 is the inner diameter)
A weight (M=167) needs to be at radius (R<46.98)
As I measured that radius being 80% of the arc-radius (that's R=45.32), it has a high probability it would roll down a 30° angle.
---

Just noticed, which I could have known by my own off-axis inaccuracy:
It could be the rolling/not-rolling also depends highly on the Main mass being off-axis (less then 90°) - KAS's is about 15°, Mickegg's about 5°.
Now I have to recalculate that parameter in this rolling pendulum, as my calculus was done with exactly 90°.... :-/

[mickegg]

Hi Guys

ME.....it's the outer diameter that is 110mm


I'm progressing on the spiral form template for a ramp angle of 20 degrees
hoping to keep the weight on a horizontal path.

Is this still worth pursuing?

Regards

MIck

[ME]

r=53.35 -->weight (M=167) needs to be at radius (R<44.24), ok it shouldn't work.

I'm progressing on the spiral form template for a ramp angle of 20 degrees hoping to keep the weight on a horizontal path.
Is this still worth pursuing?

Depends on what you want...
Any kind of lever, or rotating pendulum only moves by itself when it can lose GPE.
But you might discover something else, or combine it as KAS suggests or something similar.

[Fletcher]

Thanks Fletch.

Could you tell me if this data is for the full half turn? It looks as though from the graph that it has only rotated a a quarter turn or to the point of slow down that ME pick up on his version.

Thanks

Kas

Yeah .. quarter turn was achieved until it stopped.

If you look at the graph you can see all values climbing steadily to peak values, then steadily decline back to baseline. In the form of distribution curve.

That compares to how you'd expect an ordinary pendulum to behave in many ways.

So the KE's increase (velocity increasing) in proportion to the loss of System GPE (the rate of change) etc.

As I said earlier if we increase the mass of the arc (proportion to the blue mass) it will roll on much further and even tip over, loosing lots of GPE at that stage.

[KAS]

Well, I'm certainly not giving up on this yet purely for the two reasons I mention earlier.
I'm not in denial or anything. If it doesn't work, it doesn't work, but my mistrust of sim systems together with a nagging feeling that there is something new here is spurning me on.

You can never be sure that sim software will ever give you results beyond the boundaries of known physics. There's two reasons why; it could be that no one has ever managed to to archive it. Or it could be that there is underlying program code that will not allow results beyond newtonion physics by reverting to best guess when confronted by non conforming data. A sort of run home to mama function.
It is also worth noting that certain scenarios simply cannot be simulated. I gave an example with the arc friction. Simply entering a value to prevent it from sliding on the ramp doesn't reflect reality and could possibly throw all the data out.

Nevertheless, I am greatful to the sim masters here and others for their help and I will certainly let you all know if something new comes out of my continued physical research.

Not finished with this yet! Watch this space.

[Fletcher]

Cool. Thanks, Fletcher.

If the Blue doesn't move in relation to the Arc, and the Blue and System CoM's are practically at the same spot, why the delayed peak?

That's a good question for which I don't have an answer right now. There appears a slight delay. I'd suggest it has something to do with inertia. The Blue mass accelerates initially (because the Sys CoM is dropping fast) then decelerates because the Sys CoM is climbing. I think the observation to take away is that neither the Sys KE nor the Blue mass KE exceed the System GPE loss - IOW's, in order for something to move theer must first be a loss of GPE: and the KE will be less than that GPE loss; at any vertical height (much like a pendulum swinging or a mass rotating under gravity on the rim of a wheel (and why physicists says that gravity is conservative and path independent).

I get the feeling that I'm ignorant of something that's obvious to everyone else. :-/

Mind you, I'm not familiar with sim representations, maybe this is just normal?

If I were looking at a physical model doing it's thing, I wouldn't be able to perceive this [admittedly small] aberration.

That's the beauty of sim modeling and comparing to actual real world builds as a cross check - we are looking for that aberration, that anomalous behaviour that probably a sim can't predict.

Kudos to Kas for sharing this device, and to you guys for working through the thought process. It's quite interesting.