milling a rounded rectangle

Hi all,

I would like to machine a shape which is a rounded rectangle, out of a

3/8 " thick Aluminum sheet. The shape and the basic dimensions are here:

formatting link
The tolerance I need is ~0.01". (Sorry for the drawing... I am a total newbie).

Can this be machined on a regular milling machine, in one's garage, with some kind of an adaptor to make the circular curves? A friend of mine who consideres himself an experienced machinist just laughed at me and said that I will need to give this job to a professional shop which has a million-dollar electronic mill, the rounded curves to be iterated by pixelated steps in X-Y. At the same time, to me it seems like there must be a simpler way.

I would not mind spending up to $1000 in equipment, if it can be done in my garage. The shape is for a prototype, and I am sure I will have to go through many of them, with variable dimensions, before it starts working, and paying a shop for every trial not sound very efficient, in terms of time and $$.

Thank you all in advance for suggestions!

RunnerXCskier

Reply to
revyakin
Loading thread data ...

Do you have DRO's on your mill? That makes it easier, but I have done similar job's without them. It is a darned good reason to part with that $1000, My Mitutoyo DRO was the best $900 I ever spent on mill improvements.

The straight parts are no problem. Treat the curves as multiple points on a bolt circle. Lot's of free software that will generate the X-Y points for the circle. For instance, if you do the end curves as

24 points of a 48 point circle, make the bolt circle say 1.25" and use a 3/4 mill, you will generate a 2" semi-circle that will be well within your .01" tolerance. This technique can be used to generate any constant radius curve, the larger the radius or smoother the desired finish, the more points you use.

Bill

Reply to
Bill Marrs

Buy a rotary table

Reply to
Chief McGee

A friend of

Did you ask how it was done before computer control?

There are lots of ways to do this, starting with a hacksaw and a file. One method would be the use of a rotary table on a mill. You could make a swivel jig and round the ends on a belt sander. You could make a template and rout the shape with a wood router. Since you are not in production, grinding out N units per minute, you could make a setup to use whatever tools you had. Don't get hung up on having to have a particular machine to do something. As long as you get the end product, the method doesn't matter.

Kevin Gallimore

Reply to
axolotl

easy easy easy.. just like the previous posters said... start with the hand methods first. Called template routing.

Variable Spee Router/w carbide spiral bit pattern bit will cut AL just fine. Make a master template by hand.. create curves with a sharp compass and sand curves to shape on a disk sander. Phenolic is a good template material, but anything will work.. masonite, plywood, MDF, plexi.... whatever.

Pattern makers routinely can "sand to a line" and get accurate to an easy .005" by watching the for the scribe mark to just crumble away. Easy easy easy.

For 10 or so mulitples -- Fasten the AL blank (blanks can easily be cut on a regular old table saw with a triple chip grind carbide blade) to the template with double sided tape. Use a pattern bit and and each blank will be an almost perfect replica of the template (within the bearing tolerance, which is usually here:

Reply to
Steve Koschmann

you can hit that tolerance just fine working carefully with a bandsaw and a disk sander.

Reply to
bridger

This would be trivial with one of the better benchtop CNC mills. Even with a manual milling machine capable of handling the workpiece size, it could be done fairly easily. The slot can be done with a 1/2" end mill, and the radius at the ends could be done with a boring tool and boring head set for the right radius. A DRO on a manual machine would help a lot.

Jon

Reply to
Jon Elson

All you need is a mill and a small rotary table.And there are alot on e-bay

formatting link

Reply to
TLKALLAM8

snipped-for-privacy@yahoo.com (revyakin) wrote in news:fa06e058.0406021706.6ab80c0 @posting.google.com:

Runner, this is a perfect job for a shop with an abrasivejet machine. You will be well under $100 for the part, including material, probably under $50, with most of that cost being material.

Reply to
Anthony

Drill or bore the end radii and use an end mill for the long sides. Or, a 2" end mill will do this in one pass. On a smaller machine, or to get different radii if a boring head isn't available, use a fly cutter. Just grind the tool correctly for plunge cutting. You could even make one, if need be. If you need specifics, email me.

Ron Thompson On the Beautiful Florida Space Coast, right beside the Kennedy Space Center, USA

formatting link

The ultimate result of shielding men from the effects of folly is to fill the world with fools.

--Herbert Spencer, English Philosopher (1820-1903)

Reply to
Ron Thompson

Hey revyakin,

Just going by the "sketch" (very nicely done by the way), it appears that you will just want to make a plain full radius on the ends. If standard "widths" of the aluminum is the norm, then you can buy aluminum bar stock 3/8" thick material in various widths including 2". Just hack/cut it off at the length you need, and round the ends by any of many simple methods, belt sanding being easiest.

One thing those suggesting a rotary table have not mentioned is the "problem" of how to hold the stock while machining. I think I would not use a rotary table, but be looking at making an adjustable pivoted clamping jig that allowed the stock to be pivoted against a belt sander, in increments.

Take care.

Brian Laws>Hi all,

Reply to
Brian Lawson

Well, I see I mis-interpreted the drawing. Looked like a slot to me.

Ron Thompson On the Beautiful Florida Space Coast, right beside the Kennedy Space Center, USA

formatting link

The ultimate result of shielding men from the effects of folly is to fill the world with fools.

--Herbert Spencer, English Philosopher (1820-1903)

Reply to
Ron Thompson

A combination of rotary table and sander is the method of choice. To reiterate and combine some of the previous suggestions:

  1. Buy stock of the proper thickness and width.
  2. Scribe the length with a square, locate the center, and scribe the radius you want to cut.
  3. Cut off length (+ a little for finishing -- e.g., .+.05) on a band saw.
  4. Chop the corners off on the bandsaw to reduce milling time and effort.
  5. Use a vertical belt sander to rough cut up to the scribe line.
  6. Use the rotary table for the finishing cut.
  7. Back to the sander (if necessary) with a fine belt for the final finish.

Making a clamping jig for a rotary table is no more work than making one for a belt sander. A belt sander with a rotary jig as mentioned above is workable, if you're careful. I would personally prefer a rotary table with a jig. The advantage of such jigs, in either case, is that first you do one end and the the other. You would need different jigs for different sized pieces. However, when you're in the prototype making stage, clamping directly to the rotary table (with a spacer underneath so that you don't cut the table) is the easier way to go. If you are careful, with careful scribing and layout, a vertical belt sander or a disc sander will do it for the prototypes. In production, though, the table will be a lot faster and not require as much care. But even then I would rough out with a coarse belt or disc sander and do the finishing on the mill.

Boris

Boris

Reply to
Boris Beizer

Hey Ron,

Whoops!! You're thinking pocket milling, and I'm thinking plate with rounded ends. I wonder which the OP means?

Take care.

Brian Laws>>Hi all,

Reply to
Brian Lawson

[ ... ]

I think that he wants the negative of what you are suggesting. He wants the *remaining* metal to have that shape, not a hole in the metal. (Assuming that I -- and most of the others -- are reading him correctly.)

Enjoy, DoN.

Reply to
DoN. Nichols

Get a spraycan of layout blue and a 4" x 36" bench belt sander, some coarse (40 or 50 grit) belts for roughing and a finer belt (80 to 120 grit) for finishing. The layout blue (like a lacquer) makes it much easier to see a scribed line. You can get a pretty decent beltsander for well under $200.

You should have no problem holding .01" tolerance and you could have the part made in less time than it would take to set up a rotary table on a mill and locate the stock twice (once for each end).

You'll also want some leather gloves. The metal will get pretty hot.

Reply to
Don Foreman

Yeah, that's what I thought :)

Reply to
revyakin

Thank you all for the responses - what great people there are in this group!

Yes, a rotary table with a vise (I saw one for $100) is what I was thinking about. What I don't understand is why one can't use the rotary table from the very beginning. Why do I need to use the jigsaw and the belt sander first? Can't I just use a fine face mill to cut the curves out - pass by pass, deeper and deeper, - instead of end milling? Or does it sound like herecy?

runnerXCskier

Reply to
revyakin

Brian, I don't like the belt sanding idea b/c I need a nice 90-deg angle between the sidewalls and the face. Wait... I can just hold the thing against the table surface, I guess... Nevermind.

Reply to
revyakin

No reason you couldn't use the rotary table without any trimming first, with end mill, face mill, side cutting mill. Even a fly cutter. Depends on what you have.

Lacking a rotary table, you could use a half-round mill and cut an end in one straight feed. You could stack a bunch of parts and cut them all in one pass.

Lots of ways to skin a cat.

John Martin

Reply to
JMartin957

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.