Spinning Top Science Wiki?

[spincakes]

Hey all! Quickly introducing myself, as this is my first post here - I'm David, an engineer and science nerd. I found an old spinning top from my childhood in an old box of stuff, and have gotten incredibly curious about the science behind spinning top 'efficiency', and how all the parameters (center of gravity, AMI, scrape angle, tip diameter, mass, stem diameter, aerodynamics etc.) influence spin duration. I'm most drawn to small tops (~50grams) that take a single spin, that don't need a base to spin on.

I'm able to find a lot of fantastic information on this on the forum, but am wondering if there's a wiki somewhere, where all these aspects are discussed, perhaps with formulas, or some of the practical experiments I've seen on this forum and in Simonelli's videos.

Is there something of that sort, and if not, could we maybe start something like that?

[ortwin]

Welcome to the forum David!
Today it is exactly three years that I joined this forum.  I had a lot of nice experiences, good discussions and by now I also met some great forum members also in person.


In you introduction you  say: "[size=0px]I'm most drawn to small tops (~50grams) that take a single spin, that don't need a base to spin on."[/size]
What exactly do you mean by "... don't need a base to spin on." Most likely not a top that hovers in the air like a Levitron? The surface you spin a top on,  almost always makes a huge difference. I can only think of bearing tops that are an exception to this general rule.


I do not know of a wiki dealing with spinning tops specifically. I would help to contribute if one gets started.


Hope to hear more about your spinning top activities soon!

[Jeremy McCreary]

Hey all! Quickly introducing myself, as this is my first post here - I'm David, an engineer and science nerd. I found an old spinning top from my childhood in an old box of stuff, and have gotten incredibly curious about the science behind spinning top 'efficiency', and how all the parameters (center of gravity, AMI, scrape angle, tip diameter, mass, stem diameter, aerodynamics etc.) influence spin duration. I'm most drawn to small tops (~50grams) that take a single spin, that don't need a base to spin on.

I'm able to find a lot of fantastic information on this on the forum, but am wondering if there's a wiki somewhere, where all these aspects are discussed, perhaps with formulas, or some of the practical experiments I've seen on this forum and in Simonelli's videos.

Is there something of that sort, and if not, could we maybe start something like that?

Welcome aboard!

I share exactly the same interests — including not just the physics but also the engineering behind real tops. I also like non-throwing tops that work on flat surfaces, including finger tops and similar tops that need a starter to reach full play value (mainly by having too much AMI, too small a scrape angle, or both). Starters are fun to design and use in their own right.

The tops and starters I make are built with LEGO. This modular medium extends the design space in some interesting directions, but the aerodynamics are always a challenge. Spin times long enough to be entertaining are usually achievable nonetheless.

A top science wiki would be wonderful, but (a) I don't know of one, and (b) hard as it might be to believe for such a "simple" system in this day and age, any wiki would still have lots of unresolved controversies to dance around. Not the least of these would be a quantitative understanding of spin decay in real tops and a general theory of wobble in its many forms.

The forum search function isn't the best, but it's probably your best bet. I can tell from some of the terms mentioned in your post that you've already been using it.

A list of the members most interested in top STEM (science, technology, engineering, math) might help in your searches. Besides myself, look for posts from ta0, Iacopo, ortwin, Bill Wells, and johnm. Sorry if I left anyone out.

I like formulas, too. They help me think. So I'll share 2 posts below with formulas relating mass properties to spin decay rate and critical speed (the minimum angular speed for stability against small perturbations).

https://www.ta0.com/forum/index.php/topic,1496.msg46533.html#msg46533

https://www.ta0.com/forum/index.php/topic,5686.msg60562.html#msg60562

Critical speed is an important concept with practical implications. Since real spin decay curves are roughly exponential in shape, you'll gain a lot more spin time by shaving, say, 20 RPM off a top's critical speed than by adding 20 RPM to its launch speed. For a topmaker, knowing and understanding the formulas guides the shaving strategy.

[ortwin]

....

A list of the members most interested in top STEM (science, technology, engineering, math) might help in your searches. Besides myself, look for posts from ta0, Iacopo, ortwin, and johnm. Sorry if I left anyone out.

...

The projects and posts of "Bill Wells" might also interest you.
Can you give us an example of the ~50 g tops that you are interested in? Are some commercially available, and we might have mentioned them in this forum ("Spartan", "Quark top","Kemner top" are a few examples) ?   

[Jeremy McCreary]

....

A list of the members most interested in top STEM (science, technology, engineering, math) might help in your searches. Besides myself, look for posts from ta0, Iacopo, ortwin, and johnm. Sorry if I left anyone out.

...

The projects and posts of "Bill Wells" might also interest you.

Dang, knew I was forgetting some names. Wiil add Bill now.

[spincakes]

Thanks for the welcome, ortwin and Jeremy!

With no base I indeed mean on a smooth flat or large concave surface - thus having the tip protrude from the bottom of the spinning top. The ones I feel most drawn to are beautifully optimized for a low center of gravity, a high axial moment of inertia, and low weight for the other parts, like the following Kemner top: https://www.kemnerdesign.com/products/dynamo-tungsten-and-aluminum-spin-top-w-super-grip-spindle-ruby-bearing

I've seen similar ones from "Plier Spinning Tops", and various other sellers.

Diving more into the threads (really happy with the names and links to the formulas!) it's also becoming clear to me that it's an incredibly complex problem with the sheer amount of variables that affect each other. For example making a top wider would increase mass (and thus tip friction losses), AMI (and thus storable energy / critical speed), surface area (and thus air resistance) and scrape angle in different proportions, with there being a human limit to the amount of energy that could be put into the top, influenced by the stem diameter, etcetera, etcetera, where I'm probably still neglecting (or missing) some of the more complex aspects of the spinning behavior on the surface.

That said, I'm enjoying slowly getting a grasp on how everything influences each other, and the precise math being too complicated means it's probably best to empirically determine the ratios in which various parameters influences the spindown curves and -performance, which sounds like an enjoyable endeavor.

If there would be a way / place to gather some of these experiments and conclusions, together with the knowledge already here, I'd love to help out. I would imagine some sort of wiki with all the different scientific terms influencing top performance, with explanations and the way things influence each other, and examples / experiments showing those. Added to that perhaps similar topics on the practical parts of tops. Different tip / ball materials and considerations, the influences of stem length and width, etcetera. For me, having that knowledge available would make it more fun and accessible to start playing around with various designs, and I'd love to share that somehow.

Health reasons are limiting my time in the workshop at this moment, but I'm able to design and 3D print plastic prototypes (with ceramic balls for tips). Not too representative of a metal top in a lot of aspects, but a fun start to play around with aerodynamics due to the lower tip friction losses, and fast deceleration rate due to the low mass / AMI.

Hope to present some tests soon!

[ortwin]

....
With no base I indeed mean on a smooth flat or large concave surface - thus having the tip protrude from the bottom of the spinning top. ...

Around here we seem to have agreed to call this setup: - external tip, no pedestal -
But those terms are  not carved in stone, I mean appear in no Wiki.


You being a new member, you might have some different ideas and experiences than us who are here together for some time now. So, what are the longest spin times for different categories of tops you have heard of? (I guess I ask questions about spin times a lot, so you might as well get used to this quirk of mine right from the start   

[Jeremy McCreary]

We'll clearly benefit from your engineering perspective, David.

Like a boat, everything in or on a top affects everything else, as you noted. As maddening as such systems can be, I'm masochistically drawn to them. (My self-educated deep dive before spinning tops was naval architecture.)

Those deeply intertwined couplings make design trade-offs among endurance, interesting spin-down behaviors, and appearance at rest and at speed unavoidable and difficult. For example, the deeply knurled, fenstrated finger grip on the Kenner top is an aerodynamic liability. It's also probably too thick for max release speed. But I like the way it looks.

Of course, it's all too convenient for an aerodynamically challenged LEGO topmaker to say that spin time isn't everything. But it really isn't — as any big collection of Japanese or vintage European non-throwing tops will show.

Looking forward to your tests.

[spincakes]

Thanks Ortwin, great to learn the terminology.

You being a new member, you might have some different ideas and experiences than us who are here together for some time now. So, what are the longest spin times for different categories of tops you have heard of? (I guess I ask questions about spin times a lot, so you might as well get used to this quirk of mine right from the start   

I've seen the recent 'battle' from Simonelli and Jamer / HiPer, beating the 1 hour mark without shrouds, and have looked into the '50 gram external tip category'.

Here I've found one of Kemner tops spinning close to 50 minutes with a standing stop (where I also recognized some of your names in the comments) https://www.youtube.com/watch?v=rwk9QKSOPxk, though the owner says it went down to 20 minutes after tip wear.

And also a remarkably simple / rectangular design claiming to pass the 30 minute mark on Etsy: https://www.etsy.com/listing/1387791779/3557-by-an-etsy-customer-20-30-minutes?show_sold_out_detail=1&ref=nla_listing_details

I'm also intrigued by the wildly varying spin times in tops when spun by different people / within different tops of the same design. Assuming a simplified 90% 'efficient' top (in the RPM decline per minute way I've seen used around here), 10% more rotational speed (and thus 21% more stored kinetic energy) should only give one minute extra spin time. A top sometimes spinning 20 minutes, and sometimes spinning 30, can't be explained by a slightly varying starting RPM alone. What's happening here? Seems like quite an important variable.

Those deeply intertwined couplings make design trade-offs among endurance, interesting spin-down behaviors, and appearance at rest and at speed unavoidable and difficult. For example, the deeply knurled, fenstrated finger grip on the Kenner top is an aerodynamic liability. It's also probably too thick for max release speed. But I like the way it looks.

But how big is this aerodynamic liability? The surface speed and surface area of the top there isn't too high. And how much would the diameter influence release speed with a fixed AMI? So many questions!

[ortwin]

Here I've found one of Kemner tops spinning close to 50 minutes with a standing stop (where I also recognized some of your names in the comments) https://www.youtube.com/watch?v=rwk9QKSOPxk, though the owner says it went down to 20 minutes after tip wear.
...

For most of us around here the "standing stop" disqualifies it for being a top. The definition many adopt is, that something to be a "top", it must fall when at rest.


Those 50 minutes also seem to be a singular time for this top. You will also find that that exact video was mentioned in this forum a few times in the past.

... A top sometimes spinning 20 minutes, and sometimes spinning 30, can't be explained by a slightly varying starting RPM alone. What's happening here? Seems like quite an important variable.
...

What I found for my tops is, that those huge differences are due to the balancing of the top. Perfect balance is hard to gain, I found, and easily lost. The contact point may shift a bit due to wear of the tip or for other reasons. Iacopo says that even some microns matter. The weight distribution can probaly also change due to thermal effects among many others.
The over simplified picture in my mind is, that if the principal axis of inertia does not run perfectly through the intended contact point, the tip describes tiny little circles and thereby losses due to surface friction are increased. At the same time topple speed increases since more force coming from spinning motion is needed to keep the top upright when the CM is not exactly above the contact point.

And also a remarkably simple / rectangular design claiming to pass the 30 minute mark on Etsy: https://www.etsy.com/listing/1387791779/3557-by-an-etsy-customer-20-30-minutes?show_sold_out_detail=1&ref=nla_listing_details
...

I did not know that one, thank you for pointing it out.

[spincakes]

For most of us around here the "standing stop" disqualifies it for being a top. The definition many adopt is, that something to be a "top", it must fall when at rest.

It does feel like cheating to evade the critical speed problem like this.

The over simplified picture in my mind is, that if the principal axis of inertia does not run perfectly through the intended contact point, the tip describes tiny little circles and thereby losses due to surface friction are increased. At the same time topple speed increases since more force coming from spinning motion is needed to keep the top upright when the CM is not exactly above the contact point.

This makes a lot of sense, and sounds like an incredibly tricky thing to balance. I suppose a top that looks balanced might not have a perfectly center tip still, and a top that seems to wobble could possibly have a tip that is rotating true? Or do you think these always go hand in hand?

Only having a 3D printer at hand now is probably going to pose a problem in that department as well. I planned to make a setup today to measure a top's air resistance, similar to Iacopo's youtube video, with a tiny high RPM motor, but ran into the issue that a slight deviation of the center position causes offset loading of the bearings, which resistance seems more significant than the actual air resistance I'm trying to measure. My printer doesn't seem accurate enough for this.

I also initially planned to measure the current increase at a fixed RPM, to try and compute the required extra torque (which does still depend on motor efficiency) and thus the actual power lost due to the air resistance, but doing some calculations I found the average power a 50 gram top loses is less than a milliwatt... The noise from the motor current itself would be higher than the tiny amount of current change I'd be measuring. Grrrr.

To be continued.

[ortwin]

...I suppose a top that looks balanced might not have a perfectly center tip still, and a top that seems to wobble could possibly have a tip that is rotating true? Or do you think these always go hand in hand?...

If it looks balanced to the naked eye it can be still unbalanced to a degree that makes difference of a factor two or more in spin time I would say. But there are more sensitive methods to detect wobble than the naked eye. There is the paint method , the laser method and some more. Please ask again if you do not find with these keywords the description of the methods in the forum. What seemed originally very promising to me, was the "cell phone vibration sensor" Method that Bill Wells brought to the forum.


"seems to wobble" is not well defined I think because we did not reach any final agreement on what wobble means.
Do the tops in this thread seem to wobble badly to you?


I think it is not necessary to have a perfect balanced top to begin with, as long as you planned measures how to increase the balance of your top. There are different ways to do that:  a ball of putty, grub screws, ...
I have grub screws in my tops, but I still do not have a good method to detect the amount of unbalance in my tops.

[Jeremy McCreary]

For most of us around here the "standing stop" disqualifies it for being a top. The definition many adopt is, that something to be a "top", it must fall when at rest.

I also subscribe to that definition of a true top, but I bet that Kenner falls most of the time, and that's good enough for me. When it doesn't, the spin time is undefined.

On close examination, spin time's a pretty fuzzy concept anyway. Time to first scrape is the best-defined spin time, but in a spiraling fall, that's subject to scrape angle and hence the height of any pedestal in use.

My favorite LEGO ball tip (5.5 mm radius of curvature) has a submillimeter flat at bottom dead center. (Probably a mold scar.) An exceptionally well-balanced sleeper will come to rest on that flat without falling maybe 1-5% of the time. Like a worn flat, these built-in flats are overall liabilities, as they surely increase tip resistance and can excite energy-waisting wobbles in susceptible tops.

But 95-99% of the time, tops with this ball tip behave like true tops, and I still consider them as such.

[ortwin]

@Spincakes: where abouts do you live? Continent and time zone should be good enough. Or maybe you want to put that in your forum profile?

[Jeremy McCreary]

Only having a 3D printer at hand now is probably going to pose a problem in that department as well. I planned to make a setup today to measure a top's air resistance, similar to Iacopo's youtube video, with a tiny high RPM motor, but ran into the issue that a slight deviation of the center position causes offset loading of the bearings, which resistance seems more significant than the actual air resistance I'm trying to measure. My printer doesn't seem accurate enough for this.

Those bearing loads are valid concerns but may not be deal-killers.

Built an alignable LEGO test rig with a 9V LEGO high-speed, low-torque brushed DC motor. The top is kept vertical and coaxial enough for my purposes by the slightly loose splined coupling at the stem end and a shallow convex lens at the tip end.

Can't measure total resistance directly with this setup, but the speed drop between the rig's no-load speed and the top's terminal speed (both at 9V) is a rough but reproducible way to compare total resistance in 2 tops — or more commonly, between 2 variants of the same top.

The method doesn't require high total resistance to be of value. Works even with low-drag tops with low tip resistance — probably on par with the printed tops you're contemplating.

You can see the test rig put to another use in this video at 6:00...

https://www.youtube.com/watch?v=Un0okmmpmgk