Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Bob Gunther]

Here is the latest finger top. I call it ... #2

The stem is soft maple and the rotor is nanciton (Hyeronima alchorneoides) It may also be known as pilón, zapatero or bully tree wood. I used nanciton for the rotor because it has a rather high density similar to teak. The tip is a carbon steel 1/4" (6.35 mm) bearing ball pressure fit into the soft maple tip. The red marks on the stem are from my first attempt to balance a spinning top (as per Iacopo's video)  ... managed to gain valuable insight to a skill that I have yet to develop. Mind you, I did manage to get some extra weight (sticky tack) onto the side of the rotor and it did spin better and a bit longer as well!

The soft maple stem and tip are one piece and fits through a hole in the nanciton. In essence, a dowel through the rotor. To mount the bearing ball, I drilled a 1/4" (6.35 mm) hole into the soft maple, dropped in the bearing ball and thoroughly wet the wood, then used high heat to dry it. The wood gets swollen from the water and the heat turns that moisture to steam thereby breaking the cell walls in the wood. Doing so prevents the wood shrinking as it dries. Finally, I sanded the top to 400, then finished the entire top using CA glue and sanding at 600 between coats. This will provide a very smooth finish and hopefully help in the aerodynamics.

Details
Rotor diameter: 1 9/16" (39.7 mm)
Rotor thickness: 9/16" (14.3 mm)
Rotor center height: 1/2" (12.7 mm)
Launch speed: ?
Best spin time: 2 minutes 8 seconds
Total mass: 18 g (including sticky tack)
Total height: 2 1/4" (57.2 mm)

 

[ta0]

Thanks for the very detailed specs and the explanation on fixing the ball bearing.

It sounds that soon you will be making very high performance finger tops.

[Iacopo]

I used nanciton for the rotor because it has a rather high density similar to teak.

Teak specific weight is 0.65.  Also there are some quite dense species that are also beautiful, which could be interesting for you, (but only if you want longer spins, which is not imperative, of course);
Cocobolo (s.w. 1.09), snakewood (s.w. 1.21), african blackwood (s.w. 1.27)...

my first attempt to balance a spinning top (as per Iacopo's video)  ... managed to gain valuable insight to a skill that I have yet to develop. Mind you, I did manage to get some extra weight (sticky tack) onto the side of the rotor and it did spin better and a bit longer as well!

As I explained in my video, low barycenter tops behave differently from high barycenter tops, because the first ones, when unbalanced, lean towards their heavy side, while the second ones instead lean towards their light side.  If you found difficult to balance this top, it could be that you were unlacky and your top has just an intermediate height of the barycenter, so that the direction of leaning is not towards its heavy side, nor its light side, but somewhere in between, and this makes quite difficult to balance the top with the brush technique.  For knowing what kind of top it is, you can purposely unbalance the top adding much tacky stick (maybe one gram in a top like this your ?) at one side of the flywheel, at the center of its height.
Then you spin the top and mark the stem with the brush.  If the marks are in the same side where there is the added weight, you have a low barycenter top.  If the marks are in the opposite side, you have a high barycenter top.
You will balance the top accordingly.
If the marks are in some other position instead, you probably have an intermediate top.
You can shift the added weight into another sector of the top, and try to mark the stem again, as a countercheck.
Even if more difficult, you can still balance an intermediate top with the brush technique;  observe what is the angular distance of the marks from the added weight.  Remove the added weight and use this distance for knowing what is the heavy side of the top starting from the marks left on the stem by the brush. 
In intermediate tops the position of the marks on the stem depend also on the speed of the top, so you need to spin the top always at about the same speed, when leaving the marks.
   

thoroughly wet the wood, then used high heat to dry it. The wood gets swollen from the water and the heat turns that moisture to steam thereby breaking the cell walls in the wood. Doing so prevents the wood shrinking as it dries.

This is something I never heard before. Thanks for sharing this info.

Finally, I sanded the top to 400, then finished the entire top using CA glue and sanding at 600 between coats. This will provide a very smooth finish and hopefully help in the aerodynamics.

As for what I experienced, there is no advantage in the aerodynamics with a very smooth finish.
I like polished surfaces but I do so in my tops for aesthetics and no other reasons.
The speed of tops is relatively slow, and consequently the close-fitting and steady layer of air "sticked" to the top is thick enough to hide the tiny irregularities of the surface of the top, (like those left by a medium grit sandpaper), to the air moving around it.
At higher speeds, (airplanes, cars..), things change, but we are not in that range of speeds.   

[Jeremy McCreary]

Here is the latest finger top. I call it ... #2
Details
Rotor diameter: 1 9/16" (39.7 mm)
Rotor thickness: 9/16" (14.3 mm)
Rotor center height: 1/2" (12.7 mm)
Launch speed: ?
Best spin time: 2 minutes 8 seconds
Total mass: 18 g (including sticky tack)
Total height: 2 1/4" (57.2 mm)

Nice look and excellent performance from these specs. Once you accurately locate a heavy side or identify a stem or tip misalignment, how would you fix a top like this on a permanent basis? I know very little about woodworking.

[Bob Gunther]

Thank you Iacopo ... I have once again taken notes! I have some Cocobolo and African Blackwood so I will try those next while keeping the same (similar) dimensions. I wasn't happy with the soft maple so I may use hard maple instead.

One of the biggest novice challenges I am having with the paint brush method is that, more often than not when I touch the stem with the brush, I end up knocking the top over.

Nice look and excellent performance from these specs. Once you accurately locate a heavy side or identify a stem or tip misalignment, how would you fix a top like this on a permanent basis? I know very little about woodworking.

That's a tricky question. I imagine that if there is misalignment, it would have been transferred from a misalignment on the part of the lathe itself. In theory, a properly set up lathe, shouldn't have any misalignment and since the top is made in it's entirety while still secured to the lathe, things should remain aligned throughout construction ... in a perfect world.

Correcting misalignment would be difficult I think. I'm not aware of a way of doing so with any accuracy once the top is off the lathe ... going to have to give that some thought. Maybe others with more experience can add to this.

As for correcting weight distribution problems, so far I am just using sticky tack but that has created a new problem. Once balanced, I could drill a small hole in the appropriate area(s) and lets say, glue the sticky tack in. However, since I can not accurately weigh the sticky tack or calculate the amount of weight removed when drilling the hole or the weight added by the glue, there arises an issue of accuracy. I've seen a video on Youtube (attached below) where the maker tapes on metal weights and when balanced, he then drills a deeper hole and puts those metal pieces into them. But the video is not detailed enough to show the final balancing and how those weights are positioned in the deeper hole, or how they are glued in or how the top is further tested for accuracy.

[Iacopo]

One of the biggest novice challenges I am having with the paint brush method is that, more often than not when I touch the stem with the brush, I end up knocking the top over.

This is because your tops are light.  It may help using a brush with a little diameter, (1 mm or so), with long and soft bristles. I would use the tip of the brush (for a softer touch) and not the sides of the bristles as in the video, (the top in the video is relatively heavy so it's a different situation). A correct amount of color in the bristles and a correct dilution of the color also may help; to me it works better when there is a tiny drop of color on the tip of the bristles, so the lightest touch leaves a clear mark. Another trick could be not to hold the brush by hand, but to use a support for it, (I use a glass), placed on the table, so it is easier to control the tiny movements of the brush.
You can try not to touch directly the stem with the brush, go with the brush very very close to it, (I use magnifier glasses here), then wait for the top to touch the brush. As soon as you see the first mark, suddenly remove the brush, to avoid the top to receive other marks in different positions, in case the first touch of the brush was hard enough to make the top to move in a disorderly way.
 

Correcting misalignment would be difficult I think. I'm not aware of a way of doing so with any accuracy once the top is off the lathe ... going to have to give that some thought. Maybe others with more experience can add to this.

If the tip is not centered with very high precision, there will be, less or more, some wobbling in the top.
Unless expensive lathes are used, and proper techniques, it is not unusual that a top comes unbalanced out off the lathe.  That's nearly always the case for my tops.  I check (and correct, if necessary) for misalignment, ( if more than about 1/100 mm) using the lathe.  Then I eliminate residual unbalance, by distribution of weight corrections, or by microscopical shifts of the contact point of the tip, in both cases using the paint brush technique for detecting the heavy side of the top.       

I could drill a small hole in the appropriate area(s) and lets say, glue the sticky tack in.

One hole in one area is enough, if it is in the correct position.

However, since I can not accurately weigh the sticky tack or calculate the amount of weight removed when drilling the hole or the weight added by the glue, there arises an issue of accuracy.

You could calculate the specific weight of the sticky tack.  Then you measure the volume of the sticky tack you used for to balance the top; if you shape it like a ball, or a cube, it will be not difficult to measure its diameter/lenght and calculate its volume.  Knowing the volume, and the specific weight, you can calculate the weight, with some precision.
You can do the same for the weight of the removed wood because of the hole.
You could use a piece of metal (which is denser) instead of the sticky tack, so the hole in the top would be littler.
The added weight should be placed about at the height of the center of mass of the top, which is approximately at the center of the height of the flywheel.
For aesthetics, you may want to make the hole under the top, instead of in the side of the flywheel, so you don't see it; just make the hole deep enough so that you can position the added weight at the center of the thickness of the flywheel.
The amount of weight inserted in the top will be different from the weight of the sticky tack, also because you have to consider the distance of the added weight from the axis of the top;  e.g., 0.65 grams of sticky tack at 26 mm from the rotation axis of the top, are equivalent and have the same effect on the top as 1.30 grams added at 13 mm from the axis.
0.65 (26:13) = 1.30
 

[Jeremy McCreary]

One of the biggest novice challenges I am having with the paint brush method is that, more often than not when I touch the stem with the brush, I end up knocking the top over.

I've had good luck with a slack chalk line -- even with LEGO tops lighter than yours.

Never tire of watching this Japanese master and his wonderful tops. Don't recall this particular video, but it's a good one. Others will surely remember his name.

[Bob Gunther]

... I use magnifier glasses here ...

I actually have one of those with lighting! Great idea!

You could calculate the specific weight of the sticky tack

The math is a challenge for me but I have been trying to learn how to do all that you suggest. I will post the calculations once I understand better how to do them.

I've had good luck with a slack chalk line

An interesting alternative to the brush technique.

[Bob Gunther]

As promised .... Bob's math! 

From google, it says that the relative density of sticky tack is 1.8 and I assume the unit of measure is g/cm³

I rolled the sticky tack into a ball and measured it:
Diameter: 4 mm or 0.4 cm
Radius: 2 mm or 0.2 cm

Volume:
V = (4/3)Pi*r³
   = (4/3) x 3.14 x .2 cm³
   = 0.0335 cm³

Weight:
w = density (rho) x volume (V)
   = 1.8 g/cm³ x 0.0335 cm³
   = 0.0603 g

The biggest struggle so far has been keeping the units of measure correct. Now, assuming I have done this correctly, I need to find a weight that equals 0.0603 grams plus whatever weight I remove when drilling the hole. I see more math in my future, but one step at a time!

[ta0]

Now, assuming I have done this correctly, I need to find a weight that equals 0.0603 grams plus whatever weight I remove when drilling the hole. I see more math in my future, but one step at a time!

The calculations are correct, but the mass is quite small. I recommend you just drill a small hole on the opposite side and not fill it up. You can start with a very shallow one, test it, and continue making it deeper until it's balanced. The important thing is that you found the right location using the sticky tack. The effect of the hole on air drag would be negligible. The cosmetic "blemish" I think would also be unimportant.

[Iacopo]

I need to find a weight that equals 0.0603 grams plus whatever weight I remove when drilling the hole. I see more math in my future, but one step at a time!

I recommend you just drill a small hole on the opposite side and not fill it up.

At times I tend to make things a bit complicated.. 
Bob, Ta0's suggestion is wise and it will allow you the advantage to fine tune the balance of your top, drilling a bit more if necessary, or inserting a bit of mass in the hole if you drilled too much. Simple and effective.

In any case, here are some possible following calculations:

let's suppose you make a hole in the wood like Ta0 suggests, how much large would it be ?



Sticky tack weight:  0.06 grams.
Nanciton wood density: 0.58

0.06 grams of nanciton have volume
0.06 : 0.58 = 0.103 cubic centimeters.

So you should remove 0.103 c.c. of wood, but we have to correct the data because the average distance of the sticky tack from the axis of the top is different from the average distance of the removed wood from the axis of the top;
in fact the effect of added/removed weight on the top is proportional to its distance from the rotation axis.

Average distance of sticky tack from top axis:  mm 20.3, (for sticky tack thickness 1 mm).
Average distance of removed wood from top axis:  mm 14.8, (if the hole will be 10 mm deep).
The wood to be removed will be
0.103 x 20.3 : 14.8 = 0.141 cubic centimeters.

The real amount of wood to remove is  0.141 cubic centimeters, for a hole 10 mm deep.

A cylindrical hole 10 mm deep with volume 141 mm³ has radius:

square root of [(141 : 10) : 3.14] = 2.12 mm
radius 2.12 = diameter 4.24 mm

This is a hole with 4.24 mm diameter and 10 mm depth: practically you can use a 4 mm drill bit and make a hole a bit more than 10 mm deep.

[Jeremy McCreary]

You're in good hands with ta0 and Iacopo here. Just wanted to point out that it's often better to measure masses directly than to calculate them from volumes and densities when you can. Amazon has a digital scale with 100 g capacity and 0.001 g accuracy for $17.

[Iacopo]

You're in good hands with ta0 and Iacopo here.

...and Jeremy, of course !  I learned many things from you.

Just wanted to point out that it's often better to measure masses directly than to calculate them from volumes and densities when you can. Amazon has a digital scale with 100 g capacity and 0.001 g accuracy for $17.

I think it's easier, with a scale, probably more reliable, but not necessarily more accurate.
I have a scale 100 g capacity and 0.01 g accuracy, (so the scale is divided in 10,000 parts), which is decently accurate.
I have another one, 200 g capacity, 0.005 g accuracy, (scale divided in 40,000 parts), and the last number is practically random, I don't even look at it.
The scale you suggest has the scale divided in 100,000 parts, so I believe that it has the last number totally random, like my second one. At that price, too much precision is promised (0.001 g) with too large capacity, (100 g).

[Jeremy McCreary]

...and Jeremy, of course !  I learned many things from you.

That goes both ways, my friend!

I think it's easier, with a scale, probably more reliable, but not necessarily more accurate... The scale you suggest has the scale divided in 100,000 parts, so I believe that it has the last number totally random, like my second one. At that price, too much precision is promised (0.001 g) with too large capacity, (100 g).

Yes, you have to be skeptical about high accuracy and high capacity in the same scale -- especially at low cost. If you plan to do a lot of balancing, there are plenty of more credible but still affordable alternatives claiming 0.001 g accuracy at 20 g capacity or less. If all you're doing is weighing sticky tack and drill shavings, that's plenty of capacity, but to weigh entire tops, you'll of course need a 2nd scale with more.

You also have to be very careful about significant digits in calculations like these. The maximum number of significant digits warranted in any output equals the smallest number among the inputs. So if one of your inputs is a wood specific gravity with only 2 significant digits (e.g., 0.58), and all other inputs have at least 3, your result is still good to only 2 significant digits at best.

NB: To minimize rounding errors, it's generally best to round only the final result to the proper number of significant digits.

Moreover, if you got that 2-digit density off the internet instead of measuring the actual wood in hand, you may not even have 2 solid digits -- especially in a material as variable as wood. At the very least, I think it would be smart to check such calculations against a suitable scale.

[Iacopo]

To minimize rounding errors, it's generally best to round only the final result to the proper number of significant digits.

...if you got that 2-digit density off the internet instead of measuring the actual wood in hand ...

Good suggestions, Jeremy.
Calculations are more accurate, in that way.

I add just for curiosity that a very sensitive scale can be made out of a less sensitive one, by the use of levers.
The increased sensitivity comes at the expense of a reduced capacity.

The following picture is about an instrument I made some years ago;
there is a scale (sensitivity 0.01 grams, capacity 100 grams). There is a lever, the fulcrum is at the center of the lever.  A vertical element at the left, very near to the fulcrum, transmits the movement from the lever to the scale.
The wood plate at the right end of the lever is for weighing objects. 
The arm of the lever at the left, (with a movable weight, a copper cylinder), is for counterbalancing the weight of the lever at the right. The height of the fulcrum can be fine tuned, as the position of the copper weight, until the lever of the balance is horizontal and the reading on the scale is a high number, (for example, 94.17 grams).
The leverage is 1:50.  If I put an object on the plate, (exactly on the center of the plate), and the reading on the scale becomes, e.g. 32.53 grams, its real weight is
(94.17 - 32.53): 50 = 1.2328 grams
Sensitivity is one 5000^th of gram, and capacity is only two grams.



The complete instrument is a magnetometer;
I made it for to to check some gold and silver coins, to see whether they are real or fake: silver and gold are repelled by magnets, (by an extremely weak force, so a very sensible scale is needed here), while metals used to counterfeit them (especially gold) are generally weakly attracted by magnets. This test gives a clue and not a proof of fake silver/gold.