iTopSpin

Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: A Figure of Merit for Twirler Spin Time  (Read 11669 times)

Iacopo

  • Superhero member
  • ******
  • Posts: 636
    • Spin tops by Iacopo Simonelli, YouTube channel
Re: A Figure of Merit for Twirler Spin Time
« Reply #165 on: January 17, 2018, 02:25:57 PM »

I can do this with my strobe light, and with slow-motion video.  I'll post some stuff in several days.

Alan

A slow-motion video would be fantastic. It would be important to have a mark at one side of the top, so it would be possible to evaluate spin speed in the video, together with the wobbling.
Logged

Aerobie

  • Hero Member
  • *****
  • Posts: 267
Re: A Figure of Merit for Twirler Spin Time
« Reply #166 on: January 18, 2018, 02:51:35 AM »

More on "no-fall" tops.

The simplest approach is a top with a thin "platter" flywheel which has a Cg lower than the center of the (large) ball tip.  I've made many of these.  I even have a spreadshseet which calculates the height of the Cg.   Example, a top with 0.5" diameter ball and Cg lower than 0.25".

In addition to this, I've found that a very well balanced top won't fall if the Cg is only slightly above the center of the ball.

Finally, getting the Cg low, even if not low enough to prevent fall, might provide extended spin time because of low topple speed.  I made a 1.75" diameter, 54g top tonight which topples at 85 RPM, less than half the topple RPM of most of my small tops. 

But there is "no free lunch".  These tops have larger balls and lower scrape angles.  The greater decay rate of the larger ball consumes the benefit of the lower topple speed.  And, do we enjoy 25 minutes spins if the top scrapes half the time we twirl due to the low scrape angle?

Developments full of trade-offs like this consume a lot of time.  I'm constantly trading parameters in pursuit of best compromises.  And best for me, won't necessarily be best for other twirlers.

Alan

PS  I haven't forgotten the original topic of this thread "A Figure of Merit for Twirler Spin Time".  I compute merit on every top I make and twirl.  I find it helpful when comparing tops of different size and weight.  A high figure of merit on a 50 gram top says, "This is a good performer, considering it's 50 gram weight".  Conversely a low figure of merit tells me, "This design needs improving", or "This design is inferior to that (other) design".   etc

« Last Edit: January 18, 2018, 03:50:07 AM by Aerobie »
Logged

Iacopo

  • Superhero member
  • ******
  • Posts: 636
    • Spin tops by Iacopo Simonelli, YouTube channel
Re: A Figure of Merit for Twirler Spin Time
« Reply #167 on: January 18, 2018, 03:46:48 PM »

And, do we enjoy 25 minutes spins if the top scrapes half the time we twirl due to the low scrape angle?

I would say not, unless this is a top for trying to make a record spin.
With my tops I generally scrape once every maybe ten spins.
I tend to make low scrape angles in my tops but the long stems and the narrow knurls make them easy to spin without scraping.
If this didn't work, I would be making larger scrape angles in them.
Logged

Jeremy McCreary

  • Hyperhero member
  • ********
  • Posts: 1228
    • MOCpages
Re: A Figure of Merit for Twirler Spin Time
« Reply #168 on: January 18, 2018, 04:21:26 PM »

But there is "no free lunch".  These tops have larger balls and lower scrape angles.  The greater decay rate of the larger ball consumes the benefit of the lower topple speed.  And, do we enjoy 25 minutes spins if the top scrapes half the time we twirl due to the low scrape angle?

Depends on the audience. When I let visitors at LEGO shows spin my tops, they quickly get frustrated with the chronic scrapers and move on to tops they can twirl with reasonable success. As you'd expect, these more forgiving tops have smaller AMIs, larger scrape angles, or both.

These relatively unskilled show visitors consistently value ease of use over long spin times. Having had a lot more practice, I still find great play value in the tops they abandon, but I also have tops that frustrate even me. Play value is usually a high priority in my design process, but it occasionally takes a back seat to resting appearance, optical effects at speed, or interesting behavior.

Developments full of trade-offs like this consume a lot of time.  I'm constantly trading parameters in pursuit of best compromises.  And best for me, won't necessarily be best for other twirlers.

Amen to that! The parameter space is just too vast to explore fully -- even with a modular construction system like LEGO.
Logged
Playing with the physical world through LEGO

Aerobie

  • Hero Member
  • *****
  • Posts: 267
Re: A Figure of Merit for Twirler Spin Time
« Reply #169 on: January 20, 2018, 11:11:01 PM »

Lube Tests

I had settled on skin oil as being about as good as any lube when developing Euler discs.  I discovered it accidentally when encountering 2x variability in run time which turned out to be due to clean mirrors vs those with some skin oil.  Here's my record 5:54 run.

https://www.youtube.com/watch?v=N_bMbuEQ6wM

Tops are different, so I tested decay rates at about 500 RPM with different lubes.  Once again, skin (forehead) oil proved about as good as any.  I had learned with the Euler discs that an extremely thin layer works best.  I touch my forehead and then touch the mirror, leaving a thumb-print.  Then I spread it, wiping hard with the side of my thumb or finger.  I hard repeat wipes, each time at right angles about 5 times.

I noticed that very thin slippery lubes had much greater decay, even when I tilted the mirror often to move to fresh lube.  I was using a top with .375" diam ball and it skated freely with thin lubes, giving me the impression that drag would be very low - but it wasn't.  Conversely, I tried some thick grease which damped motion and eliminated skating.  With forehead oil, there's moderate skating which settles down as the top self "erects" in one or two minutes.

One other good lube is thin, "Dow Corning 33 Silicone Lubricant".  It's much thinner than the silicone grease that I use to improve seals.  It's even a bit better lube than forehead oil.

Regards,
Alan
« Last Edit: January 20, 2018, 11:14:08 PM by Aerobie »
Logged

Jeremy McCreary

  • Hyperhero member
  • ********
  • Posts: 1228
    • MOCpages
Re: A Figure of Merit for Twirler Spin Time
« Reply #170 on: January 21, 2018, 02:27:22 PM »

I had settled on skin oil as being about as good as any lube.... One other good lube is thin, "Dow Corning 33 Silicone Lubricant"....  It's even a bit better lube than forehead oil.

Of all the tricks I've learned on this forum, your skin oil and tilting mirror tricks are on my short list of favorites. Skin oil is the best tip lube for LEGO tops yet when both performance and convenience are factored in, and I'm currently working on a 3-point tilting LEGO mirror frame. I find good skin oil on the forehead and temples, behind the ear lobes, and around the base of the nose.

For example, I used skin oil lubrication and the weighted burger top below to get my longest spin times to date -- 3:32 by hand and 4:32 with a 1:16 manual starter. (Record runs not shown.) Not coincidentally, this is about as aerodynamic as a LEGO top can get with currently available parts.

https://www.youtube.com/watch?v=52Sw7xaJqQA

I don't use silicone lube on tips because it's almost impossible to remove from fingers, but it's great in less mess-prone settings -- like the rotor bearings in the 2-DOF LEGO gyro below...

https://www.youtube.com/watch?v=l1_ZmYTRmHU
 
« Last Edit: January 21, 2018, 03:15:19 PM by Jeremy McCreary »
Logged

Aerobie

  • Hero Member
  • *****
  • Posts: 267
Re: A Figure of Merit for Twirler Spin Time
« Reply #171 on: January 21, 2018, 05:37:44 PM »

Hi Jeremy,

Here are two other tricks you might use. 

When the top is spinning slowly and you want to move it a millimeter to fresh lube, you have to be very gentle to avoid inducing wobble and fall.  My thumbscrews allow that, but there's another way.  I set the mirror on something that has a bit of give -- a cloth placemat.  Then, when I want to tilt the mirror very gently, I press down slightly on the edge of the mirror - sinking it slightly into the cloth placemat.  When you release pressure, it may return to the prior location, but it has picked up a bit of lube.

The other trick is to blow gently on the perimeter of the top.  I like to do that with a straw.  When I do, the top is attracted to the straw!  That's Bernoulli effect.

Alan
Logged

Jeremy McCreary

  • Hyperhero member
  • ********
  • Posts: 1228
    • MOCpages
Re: A Figure of Merit for Twirler Spin Time
« Reply #172 on: January 22, 2018, 06:21:23 PM »

When the top is spinning slowly and you want to move it a millimeter to fresh lube, you have to be very gentle to avoid inducing wobble and fall.  My thumbscrews allow that, but there's another way.  I set the mirror on something that has a bit of give -- a cloth placemat.  Then, when I want to tilt the mirror very gently, I press down slightly on the edge of the mirror - sinking it slightly into the cloth placemat.  When you release pressure, it may return to the prior location, but it has picked up a bit of lube.

The other trick is to blow gently on the perimeter of the top.  I like to do that with a straw.  When I do, the top is attracted to the straw!  That's Bernoulli effect.

I'll try both of these, Alan! The soft placemat is a simple and elegant solution easily carried to LEGO shows. Beside, I have other uses for the valuable parts a tilting frame would tie up.
Logged

Aerobie

  • Hero Member
  • *****
  • Posts: 267
Re: A Figure of Merit for Twirler Spin Time
« Reply #173 on: January 23, 2018, 03:46:21 AM »

Watching the top move towards you when you blow on the perimeter with a straw is quite unexpected.  Try it.

Alan
Logged

Iacopo

  • Superhero member
  • ******
  • Posts: 636
    • Spin tops by Iacopo Simonelli, YouTube channel
Re: A Figure of Merit for Twirler Spin Time
« Reply #174 on: January 23, 2018, 12:15:25 PM »

Watching the top move towards you when you blow on the perimeter with a straw is quite unexpected.  Try it.

Alan

I tried it and....   it works !
I blowed without straw, the top goes towards me, or towards the right side if I blow at the right side of the top, or towards the left side if I blow at its left side.
Something interesting to think about.
 
Logged

ta0

  • Administrator
  • Olympus member
  • *****
  • Posts: 9905
    • www.ta0.com
Re: A Figure of Merit for Twirler Spin Time
« Reply #175 on: January 23, 2018, 12:25:15 PM »

Watching the top move towards you when you blow on the perimeter with a straw is quite unexpected.  Try it.

Alan

I tried it and....   it works !
I blowed without straw, the top goes towards me, or towards the right side if I blow at the right side of the top, or towards the left side if I blow at its left side.
Something interesting to think about.

Bernoulli has always been a mystery for me  ::) But here I would expect a complicated effect. Remember that when you tilt the surface where the top spins, it travels at 90 degrees to the slope (if the tilt is not large enough to create slipping). Would the walking of the tip plus precession be part of what you are seeing?
As Iacopo said, something to think (and experiment) about. Perhaps a whole new thread . . .
Logged

Iacopo

  • Superhero member
  • ******
  • Posts: 636
    • Spin tops by Iacopo Simonelli, YouTube channel
Re: A Figure of Merit for Twirler Spin Time
« Reply #176 on: January 23, 2018, 04:08:01 PM »

..when you tilt the surface where the top spins, it travels at 90 degrees to the slope (if the tilt is not large enough to create slipping). Would the walking of the tip plus precession be part of what you are seeing?

It seems to work especially with large and low tops with a ball tip.
Blowing on part of the flywheel pushes that part down.  But, because of the gyroscopic effect, the sinking movement happens 90 degrees later, in the direction of spinning.  So, if I blow at the front of the top, the top tilts sideways.  At that point the top walks on the ball tip because of the tilted position, and it walks towards me, which is logical, after some thinking.  When I stop blowing, the top starts precessing, because now it is in tilted position.
Logged

Aerobie

  • Hero Member
  • *****
  • Posts: 267
Re: A Figure of Merit for Twirler Spin Time
« Reply #177 on: January 23, 2018, 04:27:05 PM »

Bernoulli effect is that pressure in a free stream is inverse of velocity.  That's why a cambered airfoil generates lift at zero angle of attack.

Alan
Logged

ta0

  • Administrator
  • Olympus member
  • *****
  • Posts: 9905
    • www.ta0.com
Re: A Figure of Merit for Twirler Spin Time
« Reply #178 on: January 23, 2018, 11:13:18 PM »

..when you tilt the surface where the top spins, it travels at 90 degrees to the slope (if the tilt is not large enough to create slipping). Would the walking of the tip plus precession be part of what you are seeing?

It seems to work especially with large and low tops with a ball tip.
Blowing on part of the flywheel pushes that part down.  But, because of the gyroscopic effect, the sinking movement happens 90 degrees later, in the direction of spinning.  So, if I blow at the front of the top, the top tilts sideways.  At that point the top walks on the ball tip because of the tilted position, and it walks towards me, which is logical, after some thinking.  When I stop blowing, the top starts precessing, because now it is in tilted position.

I haven't had much luck reproducing this effect, but I don't have many large flat tops with ball tips.

Your theory, Iacopo, sounds plausible. If true, the direction of travel should reverse if you blow from the bottom.

I should have mentioned that not only I found the Bernoulli effect counterintuitive, but also the gyrosocopic effect  :-[  :P :P :P
Logged

Aerobie

  • Hero Member
  • *****
  • Posts: 267
Re: A Figure of Merit for Twirler Spin Time
« Reply #179 on: January 24, 2018, 12:42:40 AM »

I've thought of the top walking laterally as due to the portion of the ball which is in contact as being off center.  Think of the off-center portion of the ball pushing the top like an oar. 

Precession isn't applicable here - it causes the axis of rotation to tilt at right angles to applied tilting torque.

Alan
« Last Edit: January 24, 2018, 07:58:18 AM by Aerobie »
Logged
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 »   Go Up