William F. Skinner - 1939 Gravity Power

[Aaron]

Skinner videos

Hi,

I was interested to see the 1939 Pathe movie clip of William Skinner's "Gravity Power" machine.
Watching the clip, I could see the operating principles clearly so made the attached drawing. I then built two Meccano models to test the drive principles employed in this device.
The leverage assisted, reciprocating drive rotates the upper mass through the gravitational torque produced each time the mass is lifted on its tilted axis. i.e. every 180 deg. of rotation. (The upper mass has to be swung by hand to start this cycle).
The rotating upper mass, again with the assistance of leverage, tilts the shaft
supporting the lower mass, again invoking the power of gravity and rotates
the lower mass with its output drive.
I made videos of the separate and combined stages operating.
My two stage model represents one quarter of Mr Skinner's machine, and operates very smoothly.
Ideally, there should be a mirrored mechanism, which would enable the
reciprocating input moments to balance perfectly and cancel each other out.
As it is, my model is driven by a small 24v D.C. motor running on barely 2v @0.25a.
I hope that this contributes useful information to those interested.
(Sorry, I've found there is no provision to upload video on this
site. I'll be happy to send videos directly to an email address.)

Regards,
Dave Q.

Welcome Dave, can you post your vids on Youtube?

[Aaron]

ellipse

Aaron,

Sorry, perhaps I wasn't being clear enough, I do that sometimes trying to get my thoughts correctly in words. I wasn't asking about circular, I know it doesn't work, I tried it. I was asking what you thought about a linear vs elliptical. The youtube video I linked to shows a straight piece of wood being moved back and forth using a circular input. (easy to implement) The back and forth movement is not a complete linear movement as it slows down at the point where it changes direction then speeds up in the other direction much as very very narrow elliptical would do.

I will try the (non linear) linear first, then contemplate on how to do an elliptical.

Thanks

With that video you showed with the wooden example - can you make an ellipse shaped track for the end of the long piece of wood to follow. As it is moving away from the wheel, it can follow a slight curve then make an abrupt curve to bias it in that direction and it will be up and on the way back it can follow a track back coming with a slight curve down then abrupt at the end to bias it down to the lower part of the ellipse.

I saw this video before, this is pretty cool - it gets the wheels turning! https://www.youtube.com/watch?v=mkQ2pXkYjRM

I need to stop visiting this thread until after the conference but its too addicting! lol

[Dave Q]

William Skinner - 1939 Gravity Power.

Hi,

I was interested to see the 1939 Pathe movie clip of William Skinner's "Gravity Power" machine.
Watching the clip, I could see the operating principles clearly so made the attached drawing. I then built two Meccano models to test the drive principles employed in this device.
The leverage assisted, reciprocating drive rotates the upper mass through the gravitational torque produced each time the mass is lifted on its tilted axis. i.e. every 180 deg. of rotation. (The upper mass has to be swung by hand to start this cycle).
The rotating upper mass, again with the assistance of leverage, tilts the shaft
supporting the lower mass, again invoking the power of gravity and rotates
the lower mass with its output drive.
I made videos of the separate and combined stages operating.
My two stage model represents one quarter of Mr Skinner's machine, and operates very smoothly.
Ideally, there should be a mirrored mechanism, which would enable the
reciprocating input moments to balance perfectly and cancel each other out.
As it is, my model is driven by a small 24v D.C. motor running on barely 2v @0.25a.
I hope that this contributes useful information to those interested.
(Sorry, I've found there is no provision to upload video on this
site. I'll be happy to send videos directly to an email address.)

Regards,
Dave Q.

[goldpro]

Just wanted to mention 2 books about mechanical mechanisms
available at Lee valley. Lots of interesting mechanisms in them that could be useful for builds. I have them. Some of the explanations are not so
good but you can usually figure out most of them.
No affiliation etc.

Mechanics & Metalwork - Lee Valley Tools - Woodworking Tools, Gardening Tools, Hardware Supplies

Tom

[drak]

Aaron,

Sorry, perhaps I wasn't being clear enough, I do that sometimes trying to get my thoughts correctly in words. I wasn't asking about circular, I know it doesn't work, I tried it. I was asking what you thought about a linear vs elliptical. The youtube video I linked to shows a straight piece of wood being moved back and forth using a circular input. (easy to implement) The back and forth movement is not a complete linear movement as it slows down at the point where it changes direction then speeds up in the other direction much as very very narrow elliptical would do.

I will try the (non linear) linear first, then contemplate on how to do an elliptical.

Thanks

[Aaron]

get started

Ok, Aaron, sorry for my "analysis paralisis"!

Hi Hal,

That definitely wasn't directed at you - I don't even recall you saying anything that it would apply to, I apologize if I offended you.

Just a general statement because its so common for some people to get so caught up in details that they never get anything started.

It just comes from the time when I was enthusiastically getting out of my personal unproductive habits by taking on "Think & Grow Rich" type principles. LOL That really isn't just about making money even though that is what it is commonly connected to, but just general success principles and it applies to the mindset in building machines too or really anything.

All the studies show that people who have the greatest success in their endeavors make decisions very quickly and change them very slowly while those who are not very successful in accomplishing what they want contemplate to consider to evaluate and that just drags on making decisions very slowly then changing them very fast. I was guilty of that for the first half of my life.

Almost every successful builder I know follows this exact principle even if they're unaware that they have that personality trait (unconscious competence) - they don't know what they know or what they're doing right in regards to that specific principle of making decisions quick and changing them slowly. They all just get started on at least something even if it is just building a platform to mount it all on without having the rest figured out necessarily because just making the commitment to start making something ASAP triggers something that just makes the rest just unwind really quick.

Not preaching, just my observations. LOL

Anyway, building the Skinner machine obviously has to be physically built ground up to support everything, but understanding the machine actually has to be done from ground up as well since it starts with the lower weight.

Even if someone is unable to really build the whole thing but wants to really get it, I'd recommend everyone to at least build the lower shaft with weight and just play with it by hand. Rotating it in a circle or ellipse, that hands on kinesthetic feel for what it is doing really internalizes mentally and at the cellular/physical level what is going on quickly even without words to describe it.

I can't recall right off but someone recently posted their experiment where they just stuck it in a hole in the concrete and spun it around by hand.

That is all I did for hours after getting the lower weight/shaft built, I just kept spinning the upper end around in circles and ellipses and it just fascinated me that the weight kept falling to the inside incline of the shaft and all I was doing was changing the center of gravity without having to lift it, but that made it "constantly fall".

I hope everyone is having as much fun learning about this machine as I am.

[Michael John Nunnerley]

Mike, i don't see how that can be. I agree with Aaron that the speed of the drive is related to the output speed, and we have to be carefull only to increase/decrease the speed very slowly, otherwise the machine may loose synchronism...

It's a little difficult to understand and to explain, but I will try

Have you ever been to the fair? on one of those centrafugal gravity wheels where once up to speed they take the floor from under you and you don't fall?

Notice the high lighted part above, at a particular speed you will not fall (depending on your weight), those weights on Skinners machine are free running (not powered by source), at a certain speed they will become at equilibrium between the change of the center of gravity and the centrafugal force created upon them, a fine balance which will not allow an increase in speed. If the speed goes to high the falling effect will be lost and so it slows, in the end you get equilibrium (constant speed).

When running without load this restriction is in play and reduces power available, it is at balance, free wheeling, when a load is applied it will slow the weights and automaticly the power increases because that equilibrium is not there, the weights appear heavier due to lack of centrafugal force, power increases to counter load.

This man was brilliant, he had it all worked out.

regards

Mike

[hal.freeman]

The centrifugal force, RPM in relation to gravity, if you think about it, is controlled automatic. The RPM of this machine is controlled by the mass of those bottom weights, think about it, it has auto RPM control, Skinner nor the motor are controlling the RPM.

This machine is self controlled just like what you are asking me. Well thats how I see it at the moment, but not having a working model, I can't prove that, but that's where the logic leads me.

Mike, i don't see how that can be. I agree with Aaron that the speed of the drive is related to the output speed, and we have to be carefull only to increase/decrease the speed very slowly, otherwise the machine may loose synchronism...

[hal.freeman]

I know half a dozen ways to build the entire Skinner machine that will do exactly what it does in the video as there are many ways to skin a cat. If you're in the stone ages and have no bearings, oil soaked leather will work just fine as a bearing and will last just as long if not longer. I'm only posting this to discourage the "analysis paralysis" that often happens in replication attempts and to encourage focusing on what is more important, which is understanding the principles involved.

Ok, Aaron, sorry for my "analysis paralisis"!

[turbogt16v]

i have tried all 3 ways ,liner not working,circle good but not efficient as
elipse,have to make motor to run elipse.

but replying to aron,the best shape wold be egg

http://csirohelixblog.files.wordpress.com/2013/08/drawegg5.jpg

[Aaron]

circular vs elliptical

Aaron,
Are you saying elliptical is not required that it is just to maximize gains? Do you feel that just a back and forth motion such as this The project (circular to linear) - YouTube would also work? It does seem to be just a very narrow ellipse, slowing down to a stop and heading back the other way picking up speed until the center where it starts to slow down, etc... It seems the oberth effect would still work as long as it sling shots?

I'm only asking because elliptical is a pain to try and replicate.

My simple straight answer is: Can circular give gains? It looks like it should based on the primary principle of shifting the center of gravity since an ellipse isn't needed to do that. With elliptical, it is a whole other ballgame that absolutely will give the best results that circular cannot compete with, but again, can circular give any gain? If the primary work is a mass that only needs to be controlled by its weight that is over the center of gravity - it doesn't seem that elliptical should be necessary to at least show some benefit even if small.

My longer indirect answer/opinion is: My main point is that changing the center of gravity of a weight is primary because we only have to work to move the weight of the mass that is over the center of gravity, while that actually puts the weight of the entire mass in motion. Just by that principle, that tells me that there "may be" gains from circular because an ellipse isn't required to change the center of gravity for this particular principle, unless I'm missing something.

With ellipse, its a completely different dynamic though. The changing of the center of gravity still happens, but we have an oscillation that takes advantage of the kinetic energy peaks at the narrow end of each oscillation. The lever's counter action at each end of the ellipse imparts a push that actually punches it at the peak speed and is additive to the system instead of the counter movement bucking it even though it is in the opposite direction.

In circular, there is no advantage moment in the whole rotation to punch it because it is very symmetrical and in equilibrium - movement is what is in equilibrium, not the weight being off balanced.

So if we synergize an imbalanced weight PLUS an imbalanced movement, is the result more than the sum of the parts (off balanced weight and elliptical movement advantage)? It certainly appears so.

The weight is off balance and the elliptical movement is out of equilibrium. Although it can be seen as being symmetrical since each half cycle rotation added together is symmetrical, it actually is not since it is a sequential event so is out of equilibrium and asymmetrical in any given moment while a circular path has no variable sequence - it is constantly symmetrical.

My opinion is that it is a waste of time to go circular. Elliptical mechanism really isn't that difficult. My "elliptical drive" I still feel is one of the simplest ways to drive the input lever in an elliptical path directly from circular movement as it is a direct intrinsic movement from the circular movement. it isn't a secondary path, but a primary one, which is apparent by simply changing the frame of reference.

[drak]

Even the elliptical motion is not primary - that is just a way to maximize the gains through the oscillation of speeding up and slowing down on the ellipse where the input lever pops the rotation at each narrow end of the ellipse when the rotation is at the peak speed.

Aaron,
Are you saying elliptical is not required that it is just to maximize gains? Do you feel that just a back and forth motion such as this The project (circular to linear) - YouTube would also work? It does seem to be just a very narrow ellipse, slowing down to a stop and heading back the other way picking up speed until the center where it starts to slow down, etc... It seems the oberth effect would still work as long as it sling shots?

I'm only asking because elliptical is a pain to try and replicate.

[Turion]

When you begin to understand and apply principle #1, it begins to make sense how someone like Bessler was able to create the Bessler wheel, and why NO ONE has been able to replicate it since. We have been stuck inside our boxes for far too long.

Dave

[Aaron]

1939 Gravity Power

The rpm is directly proportional to the action of the input lever. For every one full back and forth oscillation of the lever, there is one full elliptical rotation. If the input motor that turns the input lever in an ellipse was speeded up a bit more than what he is showing, then the speed would increase by a proportionate amount - but of course any speeding up has to be done slow enough so the lower weight can keep up with the coupler plate rotation speed. And the other way around, if the input motor speed was cut in half, the lever oscillation would cut in half and the rpm of elliptical orbit would be cut in half. This is how I see it anyway as it is an intrinsic part of the design.

----------------------------------

Just a general comment to everyone on replicating devices.

It is very common for many people to want an exact schematic for a circuit and there becomes a big dependency on that. Whereas if the principles are understood, specific schematics or blueprints are irrelevant because it can be done from scratch by anyone else as long as the principles are done.

For me, it is irrelevant the exact method Skinner used to connect this to that or that to this. My lower shaft is cut in half and the copper sleeve keeps them both in alignment. The bottom half of my lower half is welded to the swivel socket at the very bottom while the top half of the lower shaft is in the collar bearing on the translation coupler so that they are disconnected and the lower weight can free spin around independent of the translation coupler. It isn't what Skinner used but I am applying the principles that are used in his machine and my mechanism does 100% exactly what his is accomplishing. I'm making no energy claims but my early tests showed me everything I needed to know. Everyone has to do their own tests and figured out what they believe is going on.

Looking at various comments across multiple sites, so many people are stuck in not moving forward because they're still going in circles over how Skinner built the machine - while a handful of people are simply applying the principles and they already have builds (without energy claims) that are demonstrating the mechanism of what Skinner is doing.

I know half a dozen ways to build the entire Skinner machine that will do exactly what it does in the video as there are many ways to skin a cat. If you're in the stone ages and have no bearings, oil soaked leather will work just fine as a bearing and will last just as long if not longer. I'm only posting this to discourage the "analysis paralysis" that often happens in replication attempts and to encourage focusing on what is more important, which is understanding the principles involved.

I've designed a couple machines that are not the Skinner machine in the last year and a half after figuring out exactly what Skinner was doing - that has less gain that what Skinner is claiming but only takes 1/10th the effort to build. That was only made possible because I applied just a couple of the principles of the Skinner machine. A few people have noticed but most have not - the Skinner machine opened a real can of worms.

Focusing on the principle that I explained from the beginning in my videos, it takes a small amount of energy to simply change the center of gravity of a mass compared to what it takes to lift it. That is in all my papers and videos from a year and a half ago.

Like I mentioned that a child can balance a full 55 gallon drum on edge with no problem. If you tip it 1 degree in one direction from being balanced, the only weight the child has to deal with is the weight that is over the center of gravity and that is it - not the whole weight of the drum. So effectively, the child only has to work to move that much weight around in a rotation, yet the entire weight of the barrel is rotating creating torque. Just this concept alone can be applied to an infinite amount of different machines that are not even remotely close to what the Skinner machine looks like. This is the #1 primary principle that gives the output work and Skinner's entire machine is designed to simply accomplish that in a very clever manner.

The lower weight is always falling like I showed that when the lower shaft is moved at the top, the center of gravity changes and since the lower weight can move independent of the translation coupler, it "falls" to the inside incline of the lower shaft. That is the "always falling" that Skinner is talking about. That is a very incredible principle but is still secondary to the principle #1 up above... this is just a means to an end... the always falling principle is simply a way to help accomplish primary principle #1 in a most elegant way.

Even the elliptical motion is not primary - that is just a way to maximize the gains through the oscillation of speeding up and slowing down on the ellipse where the input lever pops the rotation at each narrow end of the ellipse when the rotation is at the peak speed. It is an important principle, but is still subordinate to the primary principle #1 - of simply shifting the center of gravity of an imbalanced mass.

The lower weight is 100% of the time always imbalanced so nothing imbalances it. It is already like that - moving the top of the lower shaft only shifts where the center of gravity is of the already imbalanced mass.

Anyway, just my 2 cents - having fun watching this all unfold.

[Michael John Nunnerley]

Thanks Mike,
Unfortunately I'm not able to follow the first part of your explanation too well.
I'll cogitate on that triangulation idea.
Maybe I'm making it more difficult than it is.

Would what you've explained be true, if the support bar were half way up the lower shaft? what about if it was up near the top of the lower weight shaft?

Do you suppose at some rpm the centrifugal force of the lower weight overtakes the gravity force?

Tom

Tom, "what if's" don't really help, "what is" is what we have to concentrate on, then when it is working we can say "what if" to make improvements.

The centrifugal force, RPM in relation to gravity, if you think about it, is controlled automatic. The RPM of this machine is controlled by the mass of those bottom weights, think about it, it has auto RPM control, Skinner nor the motor are controlling the RPM.

This machine is self controlled just like what you are asking me. Well thats how I see it at the moment, but not having a working model, I can't prove that, but that's where the logic leads me.

It becomes more interesting the more you delve into it

Oh for a nice fabrication shop

regards

Mike