Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Need a tree taken down or milled?

spheres and a question

Posts

spheres and a question

#1

spheres and a question

John Lucas

My good friend Pat Matranga broke her arm and needed some spheres turnes ASAP so I turned them this morning (see below). The cherry one is 4 1/2" and the Hackberry ones are 3 1/2". No finish on them yet.

My question is what kind of centers do you all make to hold the spheres for turning. I turned these spheres with the sphere turning jig and they are almost perfectly round except for the 2 tenons on each end. I put them between these homemade cups and they never seem to center perfectly. I'm wondering if I should make the cups V shaped instead of somewhat concave sphere shaped. It seems like the spheres of this size are held by the outer rim which seems to run true but still they are always off just a hair. Any suggestions or comments.


Re: spheres and a question

#2

Re: spheres and a question

Lan Brady

John,

I use similar chuck cups. Don't seem to have the same problem. But then again, I do not measure to nearest 0.001" either.

Looks good and Good Luck, Lan

Re: spheres and a question

#3

Ric Taylor

Re: spheres and a question

Ric Taylor

John,

My cups are kinda similar, and it does take a bit of fiddling to get the spheres correctly centered. I like to have the cups as small as possible, while still being large enough to hold the sphere securely. Sometimes there is some tradeoff needed to make it work.

Ric

Re: spheres and a question

#4

Alan Sweet

Re: spheres and a question

Alan Sweet

My first naive thought has to deal with diameter of the sphere. While I may be able to create holders for a sphere of a particular diameter, it would appear to me that those holders would be off some for a sphere of a different diameter. Whether that amount would be significant enough to cause them not to center properly, I would have to experiment (or sit down and analyze). But I would think it has some effect.

In answer to your question, I would think a V shape with a center that does not touch the sphere and edges that are very narrow. The narrower the width of the contact would seem to minimize any variation. The cup shape, the bottom should not touch.

Of course the wider gives you better tension adhesion.

Re: spheres and a question

#5

Steven Antonucci

my thought too


Re: spheres and a question

#6

Re: spheres and a question

John Lucas

My thought it a V shaped cup of about 45 degree or 90 degree included angle would work with more sizes of spheres. You would probably need 2 or 3 to cover a wider range because you don't want the outer portion of the V to get in the way ay more than necessary.

Re: spheres and a question

#7

have u tried...

Darrin Hill

putting a small, bright lightsource straight above the piece? Put a piece of white paper under the sphere with a circle drawn on it that is the size of the end product.

The light will cast a crisp shadow under the sphere and if u compare the shadow to the line drawing you can see the bump you need to turn away. when the shadow matches the line, its perfect..

saw it somewhere in a book...

Re: spheres and a question

#8

Alan Sweet

Re: spheres and a question

Alan Sweet

Given my lack of experience at turning spheres, your logic sounds fine. I was wondering about experimenting with You got my curiosity up and I am wondering about experiment with a chuck-type device where the size of the holder can be changed. It would not have 100% contact around the edge, but not sure that is necessary.

Now you got me wondering... Ok off to the shop I go

Re: spheres and a question

#9

Steven Antonucci

Mine is more donut profiled

steven antonucci

I liken it to the rim of a mug, only thicker. If you can somehow get a leather "donut" to stick, it won't mar the ball. The inside needs to be deep enough that the ball doesn't bottom out. If you make it larger, you may have to flip the ball a couple of times, but it grips better.

Steve

Re: spheres and a question

#10

Unfortunately it would have to be perfectly ...

Geoff Whaling

Unfortunately the light would have to be a very strong point source and perfectly centered above the zenith - nadair axis of the ball, otherwise you will be turning an oblate spheroid - one axis longer than the other. This is caused by scale error - the light rays from one side trace a longer path than the other side if the light source is offset. Its a very well studied and understood principle in aerial photography - very similar camera axis orientation errors from tilt, pitch, roll & yaw. The shaddow cast will also be quite blurry if the light source is not a strong point source.

ps all of my spheres become oblate spheroids no matter what I do ;-) the grain forces that. Even Dave Springett syas to not get too carried away about attaining "perfect" spheres for his chinese balls.

John very nice work on these. Vermec Engineering (Australia) also have a very robust sphere turning jig that uses a tilted cup cutter & I think its carbide?

Re: spheres and a question

#11

Dennis in Southern Oregon

Re: spheres and a question

Dennis in Southern Oregon

John, there are two main sources of positioning error in the setup you are using. One is inaccuracy and instability of the support cups. The other is the residual error of roundness in the partially turned sphere at the points where it contacts the support cups.

Regarding the first source, unless you true up the cups every time after mounting them, they likely will not be perfectly round and centered on the turning axis. Changes in moisture content, stress relaxation, and the degree of tightness of the chuck on the tenon will cause deviation. You might want to consider using a more stable medium such as Corian or Nordic plywood instead of plain wood for the cups.

Regarding the second error, the partially turned sphere will not be perfectly round (even though close) where it contacts the cups. To minimize the effect of these errors on the overall centering error of the sphere, the cups should contact the sphere at a diameter that is a reasonably large fraction of the maximum diameter of the sphere. With a very small contact circle size as in your photo, small errors in the sphere (or the cups) at the points of contact can lead to much larger errors the position of the sphere. AI expect that a 45 degree cone as you suggested, would be a reasonable shape for the cups and would not lead to very much amplification of error.

Dennis

Re: spheres and a question

#12

Re: Unfortunately it would have to be perfectly ..

Darrin Hill

thats why i said.. 'small" light..

the smaller the light source, the harsher your shadows...

and i'm sure he has the stuff somewhere in a closet to do so...

darrin hill

master photographer, photographic craftsmen, CPP-PPA

;)

Re: spheres and a question

#13

Re: spheres and a question

Barry Irby

I turned a bunch of spheres years ago. I turned cylinders between centers and then rounded the ends leaving a stub. Then I reoriented them between cups like you show above and turned away the "blur". This worked fairly well but the size was mostly a random event. Sometimes I had to reorient them numerous times. I got to the point that if they looked like spheres and felt like spheres, they were good enough.

Then somebody asked me to turn some dogwood croquet balls. As I recall they had to be 3 5/8" in diameter with a very small tolerance. I turned cylinders again with the rounded ends and stubs, but this time I used a parting tool and cut an 1/8" groove around the "equator" and with calipers got it to 3 5/8". I made a piece 1/8" Masonite with a semicircle cut cut in it. I mounted that in a chuck with it between the jaws. Place the groove onto the Masonite and press the ball into place with the tail stock, keeping the "equator" level and "neat". Then you turn away everything "above" the equator. This gives you a "bottom". You still have to reorient to turn away the part capturing the Masonite, but you can get to a very reliable dimension.

With your sphere turning jig, none of this is important to you.

Re: spheres and a question

#14

Another variable that would......

JamieDonaldson

...impact "perfect" roundness would be misalignment on the axis from headstock to tailstock, not very likely on a Mustard Monster, but possible with chuck cups. :b

Re: spheres and a question

#15

Re: spheres and a question

Dale Miner

John,

As Dennis mentioned, the cups should be recut before each use, or at least after each remounting. This is pretty easy for the one in the chuck. The one on the tailstock, not so easy, but here is a way with an assumption;

The assumption, using a Oneway or Powermatic tailstock center.

Chuck up a piece of dry diffuse porous hardwood spindle grain orientation. Dogwood, hard maple, apple, etc. Turn the outside round, face the end square, and drill and tap 3/4" - 10 tpi to a blind hole depth of about 1.5". Part off the piece at about 2" long. Thread the piece onto the live center. Cut a 16 penny nail or other piece of stock to a length of about 1.25". Put the nail or other into the hole in the center that is used to hold the rotating part while threading accesories on and off. Wrap tape around the live center to prevent the nail from falling out. The live center will now not be able to spin. Put the center into the morse taper in the headstock. Being mindfull that the center can pop loose, take light cuts and hollow the end of the wood threaded onto the live center. Remove the center from headstock and remove the tape and nail.

This makes a cup that runs as true as possible, and can be taken off and placed back on the center with virtually no error.

Once I started making my cups this way, the error in remounting partially turned spheres was much better.

As to the shape of the cup, I don't think it makes any difference. the contact between the cup and the partially turned sphere will be essentially a circle regardless of the cup shape. the larger the diameter as previously mentioned, the better the centering will be, at the expense of access to the sphere for turning.

With no experiance using a jig, the following applies to hand turning and may apply to using a jig. The closer the shpere is to perfect before placing in the cups, the better it will remount. If the sphere is not close to perfect, the contact area will theoretically be at best two points in each cup instead of a full circle. Having learned that perfect hand eye coordination seldom visits my shop, I do spin by hand and adjust to reduce runout when putting in the cups.

Re: spheres and a question

#16

Dennis in Southern Oregon

Re: spheres and a question

Dennis in Southern Oregon

Dale, it occurs to me that if the cup diameter of contact on the headstock end is large there should be little loss of accuracy if the tailstock contact is essentially a point instead of a cup. In order not to damage the sphere I would propose to take a small scrap of thin plywood or other non-splitting wood, hold it against the outboard side of the sphere and then bring up the tailstock center and drive it into the plywood. This should hold the sphere firmly in place against the headstock cup without biasing its position.

Dennis

Re: spheres and a question

#17

Re: spheres and a question

John Lucas

Dale that's exactly how I turn the tailstock cup but I use a cotter pin so I can bend the tabs back to hold it in place and don't need the tape.

Re: spheres and a question

#18

Flat surface on tailstock side

Doc Green

I have used a flat surface on the tailstock side of a sphere to press it against a cup center. The flat surface was made using the method Dale suggested above - thread it onto the OneWay type tail center, etc., but instead of turning a cup, turn it flat.

The flat surface then makes contact with the sphere only at a point, as Dennis suggested in the above post. Having the flat screwed onto the threads of the live center eliminates having to deal with another piece of wood (the plywood) when you're mounting the sphere.

For best results, turn the best sphere you can between centers before you remove the blank and mount it between the cups, or between a cup and a flat surface (as someone suggested).

Re: spheres and a question

#19

Re: Flat surface on tailstock side

Dale Miner

My concern with a flat, or single point contact would be denting of the sphere. Even when using cup centers with a radius egde and a fairly good sized contact area, denting is a concern. Perhaps if the face of the block had an elastomeric or other resilient surface.

Re: spheres and a question

#20

Re: Flat surface on tailstock side

John Lucas

My thoughts exactly Dale. the cups I have now leave a tiny dent in soft woods. That's why I thought about the 45 degree cups. It would be a flat surface but would have a relatively large area of support touching the wood. I'm looking right now to see if I can find some plastic of some sort in about 2 o4 3" rods to make the cups. I did find some UHMW but it's too expensive for this project. I'll keep looking through my other catalogs.

Re: spheres and a question

#21

Re: Flat surface on tailstock side

Dale Miner

John,

I have some small shorts of uhmw. Got an snail mail address?

Re: spheres and a question

#22

Re: Flat surface on tailstock side

tomB

I use a small rubber O ring on a wooden flat on the tail stock side. I made a wooden cup that threads on the live center, since it snugs up the same every time, there is no re-centering necessary

Re: spheres and a question

#23

Re: Flat surface on tailstock side

Dale Miner

The O Ring ideas sounds like it is worth pursuing.

Thanks

Re: spheres and a question

#24

Dennis in Southern Oregon

Re: Flat surface on tailstock side

Dennis in Southern Oregon

Doc, not to make a big issue of it, but there is slight difference between a rigid flat contact mandrel and the plywood interface that I mentioned. In the first case, if there is a small ridge say 1/4" one side of center, the flat contact will tend to deflect the sphere away from center until a balancing contact is made on the opposite side of center. In the second case, the plywood will rock with almost no pressure until it contacts the opposite side of the sphere, after which the force on the sphere will be reasonably well balanced. However, if a layer of conforming material is interposed between the mandrel and the sphere, the difference may not matter.

Dennis

Re: spheres and a question

#25

Re: Unfortunately it would have to be perfectly ..

Geoff Whaling

Darrin, as a professional photographer you understand the physics and the vagaries of light paths. I come from the other side as a practical user - a surveyor. I haven't seen to many pin point light sources that would be better that using the turn away the shaddow method. I can't see how you could set up the light in a practical position to efficiently use the pin point light and shaddow board method. It will work but not very practical in use. Horses for courses.