PDA

View Full Version : milling a circle, manually.



wagnerite
01-08-2010, 04:27 PM
This is not related to any project, mainly curiosity. Before the days of CNC, how would one mill a circle manually? I tried this and its like an impossible etcher sketch nightmare!

i know with the right size boring heads and indexible holders one can do this... is this just not done with only regular endmills?

bob308
01-08-2010, 04:30 PM
rotory table have done it many times.

beanbag
01-08-2010, 04:48 PM
This is not related to any project, mainly curiosity. Before the days of CNC, how would one mill an ellipse manually? I tried this and its like an impossible etcher sketch nightmare!

i know with the right CAD and CAM software one can do this... is this just not done with only manual tools?

wagnerite
01-08-2010, 04:53 PM
rotory table have done it many times.

gotcha!! duh, why didn't i think of that!!!!!!!!!!:mad:

Your Old Dog
01-08-2010, 05:00 PM
If you don't have a RT you need one :D Pretty handy thing to have. Like, lets say you wanted to mill a circle. You'd be home free with an RT :D

wagnerite
01-08-2010, 05:06 PM
Old Dog.. indeed. :D

Doozer
01-08-2010, 05:09 PM
That is how die makers do it. Lay out the shape with a scribed line,
etch-a-sketch close to the line in the mill, then use the die grinder or die filer to get it to the line.

--Doozer

JCHannum
01-08-2010, 05:13 PM
There are compound cross slide milling attachments such as the Volstro milling head that can do circles, compound angles and other interesting tricks. There are several other attachments for K&T and some of the other machines that were used for die sinking and other functions that are capable of some pretty intricate shapes. The operator played a large part in the success of the undertaking of course.

Bridgeport made hydraulic tracing mills that would duplicate a shape in 3D that worked on the same principle as a hydraulic tracer for a lathe. Then, there were the pantograph machines. The list goes on.

arkiehilbily
01-08-2010, 05:26 PM
I have cut several circles/radii and one ellipse on my manual mill using just an end mill. If you want a 1.085 diameter hole and the biggest end mill you have is only. 750, you have to make multiple cuts on a .1675 radius around the center (Manual Numerical Control). You can do it with a scientific calculator, pencil,and paper. I use my Visual Cadd software. The smoother the hole has to be, the more steps you have to take.
Needless to say, this is in hobby mode only. No way could you do it for pay.
Back in the 60's, we made Apollo parts with a rotary mill. Sorry, I don't remember the brand, but you could make about any shape an engineer could dream up with one of those. The spindle rotated and a leadscrew adjusted the distance from center.
Jim

Bguns
01-08-2010, 05:27 PM
http://i95.photobucket.com/albums/l142/m37b1/snow2783.jpg

Roughing out a circle (with tangs) 3/4 thick steel and a rougher end mill

Fasttrack
01-08-2010, 05:31 PM
... or one can use the Kearney Trecker 2D milling machine :D


Beanbag - there was a thread here awhile ago about how to cut elipses. Clever folks have come up with linkage for many different strange shapes.

kf2qd
01-09-2010, 10:19 AM
No pictures - did this at a job 2 years ago - OD Radius only -

Take the part you want to radius and drill & ream the center hole for a dowel pin and make a plate for to be placed under it and bolted to the milling table. Clamp the part you wan to radius to the spacer plate and position the milling cutter for the correct radius and make a plunge cut. Loosen the workpiece clamps and index the part slightly, tighten and take a plunge cut. Repeat until the part is radiused. Us a larger diameter cutter for a better finish. If done with care it will require a minimum of finish polishing .

drof34
01-09-2010, 11:02 AM
Lathe
Vertical Turret Lathe

loose nut
01-09-2010, 11:54 AM
Most DRO's have a built in function for machining arcs but it is slow and if a half decent finish is wanted it takes a lot of small movements in the x-y axis's, not recomended. Sort of a poor mans CNC, it calculated the movement and you are the servo motors that do the work. It will work though.

Old model engineer, live steam type of magazines have lots of home made jigs for machining arcs and other curved shapes for locos and small engines parts. Worth looking at. These old timers were quite ingenious.

Black_Moons
01-09-2010, 12:08 PM
Iv seen a weird device that allows you to 'scribe' ovals
http://lumberjocks.com/wdkits1/blog/8097
Maybe it could be adapated for milling directly too :)

wagnerite
01-09-2010, 12:13 PM
That is how die makers do it. Lay out the shape with a scribed line,
etch-a-sketch close to the line in the mill, then use the die grinder or die filer to get it to the line.

--Doozer

ALL DONE MANUALLY?!?!?!?

Dawai
01-09-2010, 02:09 PM
Tracer mill..??

I just sold my big Rotary table to a local HSM lurker.. Wish him the best and thanks.

Pantograph tracer milling machine... okay.. when you move the lever say 1", the end mill can move .1".... when you trace something it comes out a bunch of straight lines that resemble a circle, just like a circle drawed on a monitor, made up of dots and lines..

Just you wait, I been playing with a pantograph, who needs a cnc when you got one?

sansbury
01-09-2010, 03:21 PM
ALL DONE MANUALLY?!?!?!?

CNC didn't start to leave the lab even for aerospace applications until ~1960.

If you want to be blown away, look into the early days of marine steam turbines--the Olympic-class liners were in the range of 50,000 tons and could do close to 25 knots, or slightly reduced speeds for days on end. Fuel efficiency aside, these numbers are still respectable a hundred years later.

On a much smaller scale, I still find things like screw machines to be quite amazing.

whitis
01-09-2010, 08:38 PM
This is not related to any project, mainly curiosity. Before the days of CNC, how would one mill an ellipse manually? I tried this and its like an impossible etcher sketch nightmare!

i know with the right CAD and CAM software one can do this... is this just not done with only manual tools?

There are a variety of procedures and mechanisms that exploit several mathematical properties of an ellipse. These are just the ones I can quickly think of, though I did find a few links to illustrate some of them, except where noted.

Use two pins, a loop of string or wire rope, a scribe, and layout fluid to layout the elipse then cut slightly outside the line with a mill or bandsaw, then finish with a file.

Use a router with a piece of aircraft cable wrapped around the guide bushing and two pins in a manner similar to the layout method above.

Install two pins in the workpiece at the focii. Put a groved ball bearing assembly around the end mill and one around each pin. Wrap a loop of wire rope around the three. Manipulate X and Y hand wheels by feel (see below). If you prefer, float the lead screws and move table by hand but you need to keep machining forces within manual limits.

You can use a framing square and a stick to draw an ellipse:
http://zo-d.com/stuff/squares-and-measuring/how-to-use-a-framing-square-to-draw-an-ellipse.html
Feel free to substitute linear guides. In the tradition of patenting that which is sufficiently obvious to those familiar with the art (and has prior art), here is an example:
http://www.freepatentsonline.com/4306598.pdf
It turns out this is called a Trammel of Archimedes or ellipsograph
http://en.wikipedia.org/wiki/Trammel_of_Archimedes
Note that every point on the beam connecting the two slides describes an ellipse so you can put the attachment in the center and cut around it or you can have the slide rails off to the edges and cut in the middle.

Remove X and Y axis leadscrews on mill. Install two rotary tables on table and gear them together, 1:1. Mount an eccentric on each table, each a different distance from the center. Install a groove or slide which constrains X axis motion over one rotary table and one which constrains Y axis motion over the other. If you can mount an eccentric to both top and bottom of rotary table (spaced above mill table), you only need one rotary, otherwise you are likely to have mechanical interference issues.

A circle with radiused edge titled at an angle projects an elipse and can be used as a template with appropiate mechanism.

A cone cut at an angle forms an ellipse, providing one way to make a template. You can also mount flat work on spindle at an angle and turn which will produce an eliptical profile, however edge will not be square.

You can layout an arbitrary cam, not just an ellipse, by layout out a number of points (possibly in polar coordinates) on its surface and then using a flexible spline to interpolate the curve in between.

Convert the ellipse to polar coordinates and with the work mounted on the rotary table or dividing head make hundreds of radial cuts or adjust the radial position then turn rotary table. Smooth high spots between cuts with file. An old method of making cams.

Install synchronized cam followers on X and Y axes and install an appropriate cam or use an eccentric.

It was not uncommon to approximate an ellipse using 4 circular arcs.

Many of the other methods depend on already having an elliptical template. Template can be made out of softer material and only needs to be made once. Then you can crank the parts out.

Attach an undersized elliptical template to the top of the work, put an instrument ball bearing over the end mill to make a guide, and move the X and Y handwheels on a mill by feel. Apply pressure to one while cranking the other. Which is which depends on which portion of the curve you are on.
Note that you can use a cutter shape that allows the bearing to be the same diameter as the cutter eliminating requirement for undersize ellipse.

Use an eccentric to drive the X and Y axes in place of the lead screw (or added to the displacement of the lead screw similar to a lathe taper attachment) such that it cuts a circle. By changing the gain on X and/ or Y, you can make an ellipse.

Slap the part on the rotary table on the mill, install template and bearing as above, remove the X axis leadscrew, preload X with a pulley and a bucket of concrete, and crank the rotary table.

Tracer mill.

Use a rose engine lathe with an eliptical cam.

Pantograph.

Templates can also be inverse. I.E. an eliptical hole to cut an ellipse.

Install template on one end of a dividing head and work on the other. Install a end mill in a horizontal mill and a guide bushing in the opposite end of the overarm support. Float and preload X axis or adjust crank by feel.
(This one comes from Machinery's Reference Series No. 9: Designing and Cutting Cams).

A method of drawing an elipse that utilizes the fact that the coordinates of an ellipse are the X and Y coordinates of two different diameter circles:
http://books.google.com/books?id=OqphSjpcHJwC&pg=PA764

Today, of course, one would typically just use CNC as CNC is cheaper than large numbers of exotic attachments.

I came across this 1921 citation for cutting an ellipse in Industrial Arts Index: "Fixture for milling an ellipse. A. A. Fredericks. diags Am Mach 53:752-4 O 21 '20". The Engineering index has a slightly less terse citation:
" Fixture for Milling an Ellipse, P. A. Fredericks. Am Mach., Vol 53, No 17, Oct 21, 1920, pp 752-754. 4 figs. Fixture devised for milling small eliptical shapes on production basis."

Colvin and Stanley, Milling Machine Kinks (1921) used a boring head set at an angle in a right angle milling attachment on a horizontal mill traveling through the work to generate an elliptical hole. Produces square edges. This is a variation of the titled circle method mentioned above. They also make reference to the Trammel of Archimedes ellipsograph.
http://books.google.com/books?id=JmsEAAAAMAA