Internal threads on a hemisphere

[NCarolinasailor]

I am hoping to get some advice here on a problem I have been attempting to solve for a while. I need to thread the inside edge of SS hemisphere to screw a brass ring onto. My issue is finding a way to hold the bugger and turn it. In the four jaw it either slips or I risk collapsing the part. In this jig I have a brass washer shaped to fit the inside but two issues being the hole was not made bang on center (I won't even go through my troubles with drilling that thing on center). Still slipped in the jig. I know that I need to turn the inside so that the threading surface is cylindrical which works well enough at speed in the lathe but when it is time to thread it, the tool always grabs hard enough to spin the part in the jig no matter how tight I make that little 32-10 bolt. The only thing I can think of at this point is making a small keyway in the jig. My other thought is to create a large ~60mm tap and try to thread the part like an orange in a juicer. I am hoping y'all might have some better ideas!

 

[jims]

What size is part? What machines do you have? Can you use a Step collet? Or pie chuck jaws bored to part size?

 

[NCarolinasailor]

largest running lathe I have is a SB9 (with a 3 Jaw and 4 Jaw Independent) and I have a Hardinge 5C collet mill. I like the idea of jaws bored to fit if I can make it work with what I have. Maybe something would work in a 5C collet block in the lathe? Hemisphere is 60mm diameter.

 

[TGTool]

I see a fundamental problem of geometry, in that any point past the midline of the sphere has only a partial force gripping the part and a partial force trying to push it out of the chuck. And you need a strong grip to resist the torque of threading.

If you can turn at least a partial inside hemisphere in a piece of scrap that your SS part can fit in, you can glue it in with superglue, then heat it when the threading is done to release the glue.

 

[gbent]

You need a pitch chuck. Pitch is basically a meltable glue, used in jewelry making and lens making.

 

[implmex]

Hi Ncarolinasailor:
You need to restrain it at the periphery right near where the threading will take place.
TGTool has it right...a center screw will never have sufficient gripping power to resist the force operating on the periphery.

So you need to glue or solder the hemisphere into a jig close to the place where you intend to thread it.
Start with a piece of pipe with an ID a bit smaller than the OD of the hemisphere.
Turn a tiny bit of the sphere into the ID of the pipe and then attach the hemisphere into the pipe with Superglue or soft solder.

Thread away happily.

Cheers

Marcus
www.implant-mechanix.com
www.vancouverwireedm.com

 

[NCarolinasailor]

I see a fundamental problem of geometry, in that any point past the midline of the sphere has only a partial force gripping the part and a partial force trying to push it out of the chuck. And you need a strong grip to resist the torque of threading.

If you can turn at least a partial inside hemisphere in a piece of scrap that your SS part can fit in, you can glue it in with superglue, then heat it when the threading is done to release the glue.

glue was a thought I had too!

 

[NCarolinasailor]

Even in Stainless, the cost of the metal is trivial compared to your f**kwith TIME plus special workholding.

Simply start over!

Bore the hole and tap it while it is still a straight cylinder.

THEN turn the hemisphere.

There was some sort of problem in there?

I didn't make the hemispheres but you are right; it would be much easier before it was made round and polished.

 

[jims]

Hold part in 3 inch slug bored out to 60 mm pocket and a couple thousands short of part length. Make a 3 inch washer with 4 holes 1/4 dia around washer face . Pinch part to 3 inch slug with 4 shcs.
thread thru washer . Hold in 4 jaw chuck.

edit . That is how I would do part with posters machines.

In my shop I would bore a 5c step collet to od of part about 1/8 deep and hold part with max dia held inside collet. I would only use washer to pinch hold it if part wanted to slip in collet. It would take about 15 minutes to bore collet. longer to make pinch washer.

 

[deltap]

I worked in a factory that ground external threads in transmission parts. I guess they did this because the parts were already heat treated. You could research this method to see if it would work for you. Maybe a milled thread? Neither method would generate the torque of single pointing.

 

[NCarolinasailor]

Hold part in 3 inch slug bored out to 60 mm pocket and a couple thousands short of part length. Make a 3 inch washer with 4 holes 1/4 dia around washer face . Pinch part to 3 inch slug with 4 shcs.
thread thru washer . Hold in 4 jaw chuck.

I thought of this very thing as well essentially trapping the part in a box, it may still want to spin inside the box but the couple thou may produce enough pressure. 3in slug is a bit hard to work with in the 9" lathe but it can be done.

 

[CalG]

Hold part in 3 inch slug bored out to 60 mm pocket and a couple thousands short of part length. Make a 3 inch washer with 4 holes 1/4 dia around washer face . Pinch part to 3 inch slug with 4 shcs.
thread thru washer . Hold in 4 jaw chuck.

This, but consider using "mounting rosin" or solder to hold the part. Since I see in the pictures there appears to be a hole in the dome, just pull the part into the "pot chuck" with that.

 

[NCarolinasailor]

This, but consider using "mounting rosin" or solder to hold the part. Since I see in the pictures there appears to be a hole in the dome, just pull the part into the "pot chuck" with that.

Maybe I will do it all and make the jig, glue it, and bolt it in place. If that won’t hold it I will be really lost haha,

 

[implmex]

Hi NCarolinasailor:
You wrote:
"Maybe I will do it all and make the jig, glue it, and bolt it in place. If that won’t hold it I will be really lost haha,"
It will work...you just need to have the restraint close to where the forces will be when you turn the thread.
It's the cantilever that's making you fail; imagine you trying to hold a bolt against rotation when somebody's got a box end wrench on the other end and all you have is your fingers.
This is what you have been trying to do.
Restraining it at the outer edge is like putting another box end wrench on the other end of the bolt...now it's easy to prevent the bolt from rotating.
You don't need a lot of restraint...you just need to put it in the proper place to be effective.

Cheers

Marcus
www.implant-mechanix.com
www.vancouverwireedm.com

 

[stephen thomas]

I thought of this very thing as well essentially trapping the part in a box, it may still want to spin inside the box but the couple thou may produce enough pressure. 3in slug is a bit hard to work with in the 9" lathe but it can be done.

I've done this, but if you have it start with aluminum tube or pipe, ans Implmex suggested in one post. For one off, you can even use a block of maple or mahogany, so long as you finish all opps the same day, before the wood changes dimenstion. Also, this type stuff is easier on a face plate, than in a chuck.

smt

 

[eKretz]

I was about to say much the same as jims and stephen, but they have already beaten me to the punch, as often happens. The only thing I might add is instead of boring the whole "pot" ID to 60mm I would leave a bit inside at a smaller size and perhaps matched to the contour of your workpiece's hemispherical OD to allow the clamping of the part to occur there, near the periphery. Also adding a tapped hole in the bottom of the "pot" to give it a little extra help with resistance to slipping and help centering. The collet idea should work too as long as you have something big enough. That comes down to what you have on hand and whether you want to spend the money on something you don't.