Does This Method Have Value?

[eeman]

Forgive me if this is elementary to the seasoned users of WM2d. I'm sure there are those skilled in the art of computing all the vectors associated with any configuration of weights and levers on a wheel that can far outshine what I am presenting here.

As a beginner to WM2d with only a few years of dabble time, I was looking for an easy way to observe exactly what forces were acting on a wheel during one or more complete cycles

As an old electronics engineer, I always liked to use an oscilloscope to measure and visually see what was happening in a circuit.

In WM2d, I find that by using a motor attached to a wheel that I am testing, I can get a very nice trace of the forces acting on a wheel during one or more complete cycles.

The motor acts as a sensitive transducer, relaying to the graph the sum of all torques acting on the motor at any instant of time. As the motor is attempting to maintain a constant speed it must supply counter torque to any forces acting on the wheel.

This gives us a nice graph of forces acting upon the wheel at any instant in time. Once we understand how the graph relates to the torque forces acting on the wheel from devices placed on the wheel, we can then tweak the design hopefully in the right direction that adds torque where we want it.

When we believe we have a worthy design, a runner, we can then have the motor disconnect at a certain point after giving the wheel a starting push.

It is important to understand that once the motor gives the wheel an initial push, the motor is no longer imparting torque to the wheel, for a frictionless wheel will spin on it's own forever once set in motion. That is why on the wheel with no weight we see a short pulse of starting torque and then a flat line in the graph. Any other forces imparting torque to the wheel will register in the graph of motor torque.

So as we pin our various mechanism's onto the wheel we, can observe their effects graphically as torques that either aid or detract from the forward motion of the wheel.

Thus this is a simple graphical test method similar to using an oscilloscpe in an electronic circuit.

In this post and the following few posts I will describe some experiments using WM2d and this method of testing.

I will use the motor built into the program spinning at one revolution per 18 seconds and use the torque measuring function to determine if there can be a net gain in the plot of torque vs time.

As a preliminary, first I show a a very light weight wheel with no attached weights that could cause an imbalance.

We have set the run time of the motor for 18 seconds, which corresponds to 20 degrees of rotation per second for one complete cycle.

As expected the torque curve shows a flat line for one complete revolution as the motor does not need to act once the wheel set in motion, therefore we see no +/- torque plotted in the graph.

In the next post we will attach a one pound weight to the rim of the wheel at the 6:00 position and see the resultant sinusoidal wave plotted as the motor expends torque to lift the weight up the hill (6:00 to 12:00) and then has again to expend energy providing torque in the opposite direction in order to maintain the constant speed. this needed to restrain the kinetic energy of the weight as it exerts a torque on the wheel as it falls from 12:00 to 6:00 .

In this manner we can use the torque measurement property of the graphing function to notice when we have kinetic energy flowing into or out of the system. For clockwise rotation, negative plot values (area below the zero line) occur when the motor is expending energy to lift a weight. Positive plot lines (area above the zero line) occur when the motor is attempting to restrain the stored kinetic energy when the weight is falling.

Important to note:For counter clockwise rotation the values are reversed.

In later posts I will show some arrangements of swinging weights and their graphs.

I will post a short .avi file as well as the .wm2d file, and also a small .jpg
of each device being tested. This is so we can become very familiar with the method and use it to improve our designs.

So first we calibrate our oscilloscope by using a perfectly balanced wheel and observe the graph. (files attached to this post)

As expected, our graph shows a flat line. Outside of the initial tiny startup torque, once the wheel is set in motion, there are no further forces acting on the wheel during the remainder of the cycle.

The video download:
http://www.besslerwheel.com/forum/download.php?id=17865

The drawing:
http://www.besslerwheel.com/forum/download.php?id=17864

[eeman]

Next we will attach a one pound weight to the rim of the wheel at the 6:00 position and see the resultant sinusoidal wave plotted as the motor expends torque to lift the weight up the hill (6:00 to 12:00) and then has again expend energy providing torque in the opposite direction in order to maintain the constant speed. This needed to restrain the kinetic energy of the weight as it exerts a torque on the wheel as it falls from 12:00 to 6:00 .

As you can see from the graph, we now have a nice sine wave with equal magnitude of values above and below the zero line.

If we export the data into a spreadsheet, the sum of all values will be zero as expected.

With time you get to be good at eyeballing and summing the torque values above and below the zero line.

The video download:
http://www.besslerwheel.com/forum/download.php?id=17868

The drawing:
http://www.besslerwheel.com/forum/download.php?id=17867

[eeman]

Next we have a wheel that has two equal weights attached, but one of the weights is on a hinge that allows it to swing out at a certain point in the cycle.

The power pulses are seen in the trace as the very large but brief negative going trace, always followed by a return above the zero line.

The power pulses show up as a large instant torque on the wheel, as the bottom of the swinging "L" with the weight attached is abruptly stopped against the wheel, therefore the vertical part of the "L" with the weight acts as an extended torque arm acting on the wheel.

After careful analysis of the graph, one becomes aware of the nuances of a flying weight and it's action on the wheel.

I won't go into detail about it just yet, but there is a lot to be learned from running the video or wm2d file and observing what is happening and correlating events on the graph vs what is happening on the wheel.

No implications are made that this device has merit, rather it is just used as an example tool to show how we can correlate torque forces acting on the wheel with the graph.

I hope it is helpful.

Attached is the annotated graph and attendant files

The video download:
http://www.besslerwheel.com/forum/download.php?id=17876

The annotated graph:
http://www.besslerwheel.com/forum/download.php?id=17874

The drawing:
http://www.besslerwheel.com/forum/download.php?id=17877

[eeman]

I'm at a loss as to why the videos .avi files do not play properly on this forum, as I have posted them on another forum and they play perfectly. Also the .avi and .jpg files do not have a download tab so can't be downloaded and played independent of the forum video player. Can anyone clue me in?

I'll pause any further uploads until I can get this fixed. Meanwhile is there a repository where I can upload the files so viewers can access them?

Thanks in advance

Regards, eeman

[Furcurequs]

Hello eeman,

Welcome to the forum.

Your videos play fine for me in my browser. If they don't play in yours, it may be a browser issue and/or something to do with the default settings for displaying certain embedded file types.

If you want people to have easy access to direct download links for your attachments, when you first upload them you can copy the links given (right click and copy the blue hyperlinks) and paste them into the text pane before actually submitting the post.

Here are the direct download links for your previous videos that I harvested from the page source:

http://www.besslerwheel.com/forum/files ... weight.avi
http://www.besslerwheel.com/forum/files ... ighted.avi
http://www.besslerwheel.com/forum/files ... on_rim.avi

I hope this is of help.

Dwayne

[Art]

Welcome eeman ,

I like your approach !

Is it possible to keep doing a simple line drawing highlighting the action of the particular component of the setup that is actually being graphed (- for those of us who still struggle with anything connected with computers ).

Something similar to the small drawing in your "ex 2 weighted .jpeg " diagram is ideal IMO

I think I'm a computer dyslexic and easily get lost if I can't visualise the set up really clearly

- I have been a bit challenged by computer simulations in other threads and have been too intimidated to say anything about it ! :)

[eeman]

Hi Dwayne

Thanks for the helpful hints. I tried copy link and pasting link as well as just dragging the link into the test box. It seems to work but doesn't give a hyperlink that you can just click on. Users will need to copy and paste it into their browsers.......I must be doing something wrong, because I've pasted links before and it worked. Maybe need to paste it into "a new private window". It would be interesting to know if anyone else has problems with the videos.

Hi Art

I'm sorry I forgot to upload the little sketch on the last post with the hinged weight. If you click on the image it expands to a much bigger clearer image, or you can now download the image by pasting the supplied link into your browser.

I'm glad that you liked this approach.

Kind regards, Ernie

[Furcurequs]

Hey again.

People can click on your thumbnail images to get to the full size pictures already. ...or at least I can in my browser.

To get clickable links to your videos, however, it looks like you may need to paste the links into the same pane as the text of your posts like I did in my previous post.

Having the url in the description box for the file apparently doesn't work. I wonder if adding url tags to it there will? I may do a quick experiment...

ETA: No, tags in the description box don't make it clickable. You'll need to put it here.

http://www.besslerwheel.com/forum/files/logo.gif

[eeman]

Since the videos seem to be playing for some folks, next upload is a pendulum
that alternately hammers on two stop pins as the wheel rotates, just to observe the large but brief torque pulses into the wheel.

The pendulum is 0.1kg counterbalanced with an equal 0.1 kg weight at it's opposite end.

Of course there is no gain here, it's just for observation and recording of the power pulses which are equal and opposite so they null out.

I included a graph of a much longer run and you can see that the positive and negative power pulses ina more compressed view.

Exporting the data and summing it in an Excel spreadsheet is interesting.

The obvious thought is how do we eliminate the large counterclockwise power pulses that counteract our nice clockwise power pulses.

We'll explore that next and show why the first and most obvious ideas don't work.

enjoy

Video download:
http://www.besslerwheel.com/forum/download.php?id=17879

The drawing:
http://www.besslerwheel.com/forum/download.php?id=17880

The graph:
http://www.besslerwheel.com/forum/download.php?id=17882

[eeman]

Hi Dwayne

Thanks again for your help on getting the download links in properly. It would be nice if a download button on the right would appear and do the job as it does for wm2d uploads, but I guess that would have to be associated with all file uploads, and right now it is only associated and appears with .wm2d files.

Does anyone know if Interactive Physics id much better than WM2D.

I'm so new to WM2D, I'm still trying to figure out how to put color into the drawings for differentiation of otherwise grey on grey objects.

Regards
Ernie

[Gregory]

I'm so new to WM2D, I'm still trying to figure out how to put color into the drawings for differentiation of otherwise grey on grey objects.

Hi eeman,

Highlight the object you want to color then hit Ctrl+J, or click the menu Windows/Appearance.

The same works for:
Ctrl+I == Windows/Properties
Ctrl+K == Windows/Geometry

[eeman]

Hi Gregory

Thanks for the heads up on color. I had not got that far to the right of the pull down tabs, and thought Windows / Appearance had something to to with the Windows operating system, so never tried the tabs.

I would have thought the appearance tab would be under "Object" since that is what you want to change the color of.....duh.

I will have to find a manual at some point because the pull down tab help box did not even recognize the word color or have anything in it's library that I could find referring to color.

Thank you
Ernie

[AB Hammer]

eeman

Welcome to the forum. I have been very busy or I would off posted earlier.

I am very familiar with what you are posting. It is what happens when one tries the wheel of death approach. CF is going to be a problem in a real life build causing a timing problem. But I will say you are on a good path in your research.

[eeman]

Hi AB

We have exchanged ideas before on other forums, OU.com and OUResearch.com where I post under the name of Vortex1 and ION respectively.

I continue to admire your works in metal and your insight into "the problem".

I will look closely at the CF, as I believe it does appear as a signature in the graphs.

My intent will be to develop a wheel that is perturbed from an outside source of energy perhaps 12 times per revolution and then examine the signature.

Once the signature is developed it can be used as a benchmark goal to compare future builds to.

Maybe a lot of this doesn't make sense right now as I am only laying a foundation for future experiment, and sharpening the tools in a manner of speaking.

From the poll I noticed about 50% do not understand the testing method so I will go back and edit each post and add more detail in the hopes it can be made more understandable.

Kind Regards and thanks for your input
Ernie

P.S. Does anyone know how to set the grid snap size in WM2D. nothing in the help reference about it. On most programs it is user selectable.

[WaltzCee]

Hello Ernie,
Welcome to the forum. Any analysis prior to building is an efficient way to do it, but that's not for everyone. Some just like to build.

grid snap size
You can position objects in working model by changing their center of mass. You can also zoom in/out and that changes resolution. If describe what you're trying to do maybe someone could be of more help.

My intent will be to develop a wheel that is perturbed from an outside source of energy perhaps 12 times per revolution and then examine the signature.
Where is this energy coming from?