William F. Skinner - 1939 Gravity Power

[armandino]

Armandino, Congratulations on your build, you've done a fantastic job.
I think this is the closest Skinner replication I've seen to date.

There is one area that I think you may have missed, in replicating Skinner.

Watching your video @ :09, you can see that your wobble plate is attached to the input shaft by a ball joint that allows it to stay horizontal to the ground at all times regardless of the angle of the input shaft.

If you watch the original Skinner video, @ :28, you can see that the wobble plate changes angle as the input shaft tilts and remains perpendicular to the input shaft at all times. @ :29 you can see the setscrew that holds the wobble plate to the input shaft. You can also see the input shaft spinning as the machine starts up @ :36.

I think this is a critical part of the design as the angle of the wobble plate is what creates the condition that allows the top weight to always be falling down hill.


Looking forward to seeing more videos and watching your progress.


John

thanks, I will treasure your suggestions and I am sure that will improve the performance of my mechanism

[Aaron]

elliptical drive

To Aaron

I understood your comments and thank you for valuable tips and I have concluded that I will apply an elliptical motion precisely in primary drive motor-pinion replacing the pinion with a pair of elliptical crowns wheel used on bike with its chain drive. It will be seen later on the result.

Looking forward to it.

[MagnaMoRo]

To Aaron

I understood your comments and thank you for valuable tips and I have concluded that I will apply an elliptical motion precisely in primary drive motor-pinion replacing the pinion with a pair of elliptical crowns wheel used on bike with its chain drive. It will be seen later on the result.

Just a simple reminder that this historic image clearly shows that the upper part of the upper tilt arms were connected to brackets. Nothing is visible much more complex then that.

[Aaron]

elliptical orbit

Just a simple reminder that this historic image clearly shows that the upper part of the upper tilt arms were connected to brackets. Nothing is visible much more complex then that.

BroMikey's animated gif shows it clearly: http://www.energeticforum.com/283944-post683.html

Upper input lever rotates on a swinging arm a the end of the brackets and anyone that says otherwise is spreading misinformation. Whether it is intentional or not is another story, but misinformation nonetheless.

[SpecJet]

Alternate design for elliptical drive

The attached pic shows an option for creating elliptical drive of the top shaft.

The red plate is driven by the electric motor.
The blue plate is driven by and synchronized with the red plate via chain and sprocket so that they are always locked, rotationally.

The drive bar is locked to the pivot on the red plate, but slides thru the pivot on the blue plate.

By changing the offset on the red plate you can adjust the length of the ellipse.

By changing the offset on the blue plate you can adjust the width of the ellipse.

A Heim joint on the end of the ellipse shaft would accommodate the top shafts change in angle and height as the top shaft moved through it's elliptical path.

Not knowing exactly how the top of the machine was put together, this is the design I had for my build as it would allow for just about any path configuration of the top shaft from extreme ellipse to circular, without having to rebuild anything to test different settings.

John

[armandino]

Aaron - Hello I ask you a favor. a comment on my movie calculates the power in watts with only a rotation module, is 150 W. I want this forum can compare this calculation because if this result were true, I think with two modules could already reach interesting values. I have no illusions but would be a great encouragement to continue the work.
96 Kg weight, arm length 50 cm, rotation speed 96 RPM.

[Aaron]

skinner machine

Aaron - Hello I ask you a favor. a comment on my movie calculates the power in watts with only a rotation module, is 150 W. I want this forum can compare this calculation because if this result were true, I think with two modules could already reach interesting values. I have no illusions but would be a great encouragement to continue the work.
96 Kg weight, arm length 50 cm, rotation speed 96 RPM.

Hi, I only just now saw this. I apologize for not responding before.

Do you have a link to the video?

[shylo]

Here's the link https://www.youtube.com/watch?v=ib3ZATuJZis
Hope that worked.
Hi armandino, Very nice work , could you take some pictures of the various stages top middle and bottom and post?
I think with just one set of weights it won't show any gain but with all four ....?
A lot of work though.
Thanks for sharing.
artv

[armandino]

hello, thank you, I'm proceeding in this way in an attempt to force consistent with amplification of four modules. for now all I have shared is the movie on you tube.
Armandino

[Danny B]

Rpm

Armandino, you've done great work. I'm getting ready to jump in to a Skinner replication very soon. After I finish my wood-gas truck.
I believe that Skinner ran at 60 rpm. You are 50% higher with what appears to be a larger radius. Here is an interesting vid; https://www.youtube.com/watch?v=pw3lpFvuzps
I believe that the acceleration of gravity is 33 feet per second squared. If you run at a high enough rpm, the acceleration of gravity is too slow to make the weight fall before the angle of the shaft changes to an unfavorable position.
You have a larger radius so, a larger diameter. The surface speed of a larger circumference is going to require a slower rotation so that the weight has time to accelerate to the lowest position.
Skinner built with 4 smaller units rather than 1 larger unit.
You can picture a very large unit running at high rpm. Both weights would be thrown to the outer limit of travel and stay there.
The upper weight is heavy and fat with a large travel circle. It is thrown outwards with centripetal? force and doesn't have to FALL anywhere.
The lower weight is tall and skinny with a relatively small travel circle. It is driven strictly by gravity. The acceleration of gravity is the only thing that contracts the tendency of centripetal force to pin it to it's outer limit of rotation.
That is why the lower weight is tucked in close. It is tall to keep a heaver weight as close to the center as possible. The center axis has reinforcement because any flexing will cause funny oscillations.
Skinner could have gone to a bigger center axle BUT, the upper weight would have to be bigger to flip it around by reaction. A tubular titanium axle would be very nice to keep the weight ratio as high as possible between the axle and the upper weight. Yes, the top of the lower weight can be solidly linked to the shaft without any ill effects.
An IC piston engine is limited by the feet per minute that a piston can tolerate. It appears that the Skinner device has the same type of limit. The bigger the device (circle of travel), the slower it must turn.
Since bearing drag fights the acceleration of gravity, the lower weight must rotate as freely as possible.
Disclaimer, all of the above may be total bull pucky.

[Danny B]

part deux

I'm looking over what I wrote last night. It mostly looks OK. The upper weight is falling BUT, it is a different animal than the lower weight. I only started looking at the Skinner device a couple of days ago and it takes me a while to get up to speed.
The upper weight is trying to continually reposition the axle and it isn't particularly harmed by momentum. The lower weight is trying to stay in one place and it doesn't want momentum. The upper weight can describe a larger circle without causing problems.
IF gravity is a constant,,, IF we allow that Skinner worked out the variables, then we can use the velocity of the "travel circle" of the c/g of the lower weight as a CONSTANT.
Any 60 rpm Skinner device will have the same distance from the c/l of the axle to the c/l of the lower weight. If you double the circumference of the travel circle, you must reduce the rpm by 50%
You could theoretically build a large machine that ran at low speed and connect it to a Faraday disc. The Skinner device would be a good match for an overshot water wheel, in that it requires constant rpm.

Skinner built with the lower weight circle just below the upper weight circle. This too is probably optimized. He needed the lower weight very heavy BUT, compact and tucked in close to the axle. That dictated that it be tall. The weights are as close as possible to get whatever dynamic balance he could find. Aaron pointed out that there is some "give" at the top pivot.
this reminds me of a trebuchet. A trebuchet on wheels works better than one that is solid on the ground. A trebuchet is weights and levers. If I recall correctly, a trebuchet does a transfer of momentum and needs some movement of the main axle to work correctly.
The lower weight could easily be made of lead. I've made weights like this before. First stop is the tire store to buy old tire weights. I used a tin can with a piece of EMT in the center. Just pour the weight and slip it over the center shaft of the weight bar. This allows the weight to be tucked in just a bit closer at it's max O.D.
The upper weight is a fixed quantity. The lower weight is "trimmed" with small weights to minimize the shaking from balance problems.

The starting point is to use the stills from Pathe to obtain all the dimensions for an accurate copy.
1. c/l to c/l of the lower weight & long axle
2. O.D. and height of the weights,,, to get the proportions and weights.
3. c/l to c/l of the upper weight to it's pivot point(s)

The only way to scale up this device is to build it taller OR slower. Skinner built it as a quad to get around this limitation. We can probably assume that he built a LOT of variations to arrive at the height of his final design. The height can be increased if the rigidity of the axle is maintained without raising the weight of the long axle.
A large-diameter, tubular titanium axle could have a tungsten weight bolted right to it.
I'm going to go see what is happening with the economy.

[artoj]

Past posts

Hi Danny,

Here are a few of my past posts that might help you with accurate sizes and dimensions.

http://www.energeticforum.com/257187-post94.html
http://www.energeticforum.com/257297-post100.html
http://www.energeticforum.com/257321-post104.html
http://www.energeticforum.com/257489-post122.html
http://www.energeticforum.com/257597-post135.html
http://www.energeticforum.com/257923-post180.html

I have completed most of the drawings and was ready to start a Kickstarter for replication, when I noticed the general interest had waned and the factory space and engineering support was not as forthcoming as I hoped.
I have a 120 page PDF file that I might release, when I have time to complete the complete engineering and mathematical analysis. It looks like my time has been occupied by getting my bills paid and home repairs, not much time to complete this work soon. I have no time limit to when I will finish the work, so another few more years does not bother me.
I am still working on projects I started 30 years ago, writing, music and art also are long term items, I hope to get all my work finished just in time for my own funeral. Regards Arto

[Danny B]

will work for dimensions

Arto, do you have a preliminary design for your casket?
Thanks very much for the info and links. I'll look later in the day. I don't need any kickstarter to build this. I have everything clear in my head about the layout. i just don't have solid parameters for weights and dimensions. The Skinner device is like the school girl motor. There is a lot more going on than is apparent at first glance.
The weight is falling down a ramp. A body falls vertically at 33 ft. sq. The ramp isn't vertical so, the speed of the fall is proportional to the angle of the ramp. The max speed of rotation must be maintained slightly below this speed.

I'm not perfectly clear on the top pivot yet. I need to know/figure the precession of the top weight relative to the lower weight. Quick logic says that it should be 90 deg.
The more help that I get on dimensions of the original device, the faster I'll have a replication. The finance, materials, engineering and building don't present any impediments to me. I just need parameters.

[mksboysal]

Skinner Replication

Hi guys,
The machine that are being discussed here I built it, it took me around four months to come to this completion, not finished yet. Enjoy the pictures. Frame size is 5 ft by 5 ft and the height is around 7 + ft. The lower weight is 340 lbs and the upper weight is around 75 lbs. As you can see the upper weight it's totally adjustable, you can slide back and forth if you want to. The driver motor is a 3 ph 1/8HP with a 56:1 gear ratio planetary gear it is controlled with 1/4 HP VFD drive. When I turn this thing on it's scary to stay close by! So I stand back 5,6 ft away when that 340 lbs rotates at 33 rpm or little higher you have to respect the "force in motion" under no load prime mover motor consumes 80 to 90 watts, the output power is still being tested!!! During the operation the frame sways a little but I have plans to make it firmer by installing cross braces all around the frame. Come along tool is temporarily hold the lower weight and does a good job with it. 90 decree weight lowers the input power close to 25 to 30 % anyway if you have a ? or comment drop me a line. If some of you need help? I maybe be available to help you design yours.

[Danny B]

The best mechanical OU devices seem to shake like a demented banshee

mksboysal, that is quite a construct. I believe you are in Hawaii. I'm sure that it is more difficult to find parts there.
Artoj has extensive plans for a replication. Until I have a good working model, I don't want to multiply problems by 4. I plan to build a single device. I'm going to build a copy, not a replication. I’ll do as you did and just use an off the shelf reduction motor. The advantage of building a quad device is that you should be able to cancel out some of the huge momentum swings.
Armandino bolted his device to the wall.
At 1:02, this device has a very cute 4-gear system; https://www.youtube.com/watch?v=Bp4t...ature=youtu.be
IF I find the time to work on this and IF I am successful, then, I will look at making the much larger quad machine. Good luck on your project.