Improvisation of the Day...

[small.planes]

Occasionally (well more often than that actually...) I have to knock up bits for work. This week its an asymmetric, offset ball bar for some testing work.
280mm centers, 4" and 2" dia balls.
The 4" ball was easy, Sir John already wore out some carbide drills on it, and made a hole, which after a little work attaches it nicely to its ball seat.:



The 2" sphere is carbide though. It came with a small (5mm) dowel in it. Not enough to get hold of in a clamp. It was previously fitted to a different fixture using loctite 638. I planned on turning down that seat to make the new part, but its made from $%$***%%$ awful stainless.
So a new seat was required. And it had to be a pretty good fit so that the 638 stands a chance of holding it. The sphere is (IIRC) 1.5Kg, carbide is heavy.
I dont have a ball turner, and the CNC lathe is still in build, so what follows is todays improvised tooling. I did this, but it may not be safe or a good idea, presented for interest and inspiration (If you try it and break anything you get to keep both bits, nowt to do with me)

I knocked up a blank, a 1" head on a 12mm stem and chucked it seat to be out. I drilled the hole for the pin to give the tool somewhere to start the cut.



The topslide on my CVA (long bed 10EE for the yanks) is quite substantial, and has a pivot which with a little fiddling can be set to 1" from the cutting edge of a small boring bar (can you guess where Im going yet?)



Now set the topslide square, and move the tool to the center of the workpiece:



More to follow....

Dave

[small.planes]

continued...

Now take *VERY* light cuts swinging the topslide, and advancing the cut with the carriage. A radiused (1" in this case) seat begins to appear. Note that I checked all the clearances on things like the topslide *before* I turned anything on. In this case the smallish 3 jaw just clears the topslide, and so I can get close enough to make the cut. It would be a bad thing to crash into the chuck whilst doing this.



Done:



And the ball fits nicely. Sorry for the crappy pic, cellphone camera and in a rush.


The finished bar mounted on its support. When viewed end on you can measure both spheres optically, and hence the bar length. This is for some testing on the optical metrology equipment we manufacture (shameless plug: www.phasevision.com)



The bar will get verified on a mechanical CMM, and then we'll see how accurate my mills DRO and my working are...

Dave
.

[rohart]

Nice work, Dave. What was the source for the carbide if the sphere is so dull compared with the larger one ?

BTW, if you hold that fixture about 100,000 miles away, you can probably measure the diameter of the moon with it.

[Black_Moons]

Why was carbide used? Seems kinda a waste of a really hard expensive material just to make a ball to look at...

[winchman]

Is that carbon fiber tubing?

What a strange assortment of stuff in your shop. I see a wooden pallet, a traffic cone, and the trunk lid from a Honda.

[Timleech]

Is that carbon fiber tubing?

What a strange assortment of stuff in your shop. I see a wooden pallet, a traffic cone, and the trunk lid from a Honda.

Something makes me think that second pic isn't Dave's shop. Waay too much floor

Tim

[small.planes]

The 4" sphere is chrome steel (normal ball bearing stuff). Carbide in that size is prohibitively expensive.
Carbide is used for reference spheres because its hard, stiff and stays the correct shape. For mechanical probing thats important. Here it could easily be made of steel, except we had a 2uM calibrated carbide one in the draw
Yes its CF tube, ~100 mm diameter.
It has to be stiff as when it bends the distance between the balls changes, which is not so good for proving how accurate something is.
Ive not done the sums (they are quite complex...), but we are going to check different poses when we CMM calibrate the artifact.

Tims correct, the last pic is not in my shop. (I wish I had that much space!)
Its the assembly unit at work. We have all sorts of things floating about.
The windscreen (at the back, leaning on the cupboards) is especially hard to measure optically

Dave