PDA

View Full Version : Milling on a periphery



firbikrhd1
12-20-2010, 10:09 AM
Once again I need the advice of the experts, this time with milling a periphery. Milling is a new “art” for me and I am nearly devoid of actual experience. I have read considerable information regarding the process but as usual, I seem to always want to begin with an oddball job that the books don’t cover.

I need to mill the periphery of a piece, but only an arc about 300 degrees of the circumference. I have my work mounted in a 5C collet in a spin indexer on a horizontal mill with the surface of the circumference parallel to the endmill I am using, making this a peripheral milling operation (using the side of the endmill on the circumference of the work). I have been making a pass around by turning the handle of the spin indexer through the 300 degree arc, returning to the beginning, feeding inward toward the spindle and making the next pass. I am using about 1/3 of the diameter of the endmill.

So far I’ve had two difficulties. First, the spin indexer wanted to move on the table. Apparently the ground surface on the bottom and the smooth table surface didn’t have enough friction to hold it when clamped with strap type clamps. Perhaps I didn’t have the nuts holding the clamps tight enough. To resolve that issue I cleaned both surfaces completely free of oil and placed a piece of newspaper between in hope of improving the coefficient of friction between them and tightened the nuts as tightly as I dare. I’m not sure if this is accepted practice for mounting fixtures but so far it seems to have solved the problem. Perhaps someone can advise whether placing paper or other substance between the milling table and fixtures is proper technique.

The second issue is that of the endmill wanting to climb at the beginning of the cut. I am not using a climb cutting technique, however when I return to the starting point for the next pass and increase the depth of the cut the mill seems to flex and climb out of the cut, up and around the periphery of the work It doesn’t matter whether I turn the machine off to return to the beginning or not. My technique has been to shut off the machine, return to the beginning point, lock the spin indexer at that point, turn on the machine, feed toward the end of the endmill .015” upon which the endmill will climb. I am fortunate that this is a belt driven machine and has stalled the spindle rather than cause damage to the machine. The damage to the work so far has not ruined it, yet.

The material is mild steel. The endmill I am using is a 4 flute HSS 5’/16” diameter mill. My depth of cut has been about .015”. I have made every attempt to insure rigidity of the setup. Where am I going wrong? Should I be using the endmill perpendicular to the circumference rather than parallel with it? Would a 2 flute endmill be more resistant to flex, or perhaps a larger diameter 4 flute endmill?

Your thoughts, critique are appreciated.

lynnl
12-20-2010, 11:59 AM
I'm not sure I'm understanding the situation correctly. I have next to "no" experience on a Horizontal mill, so I tend to think in terms of a vertical. And the way that's (normally) done on a vertical is to orient the endmill perpendicular to the "plane" of the workpiece.

You say you're using about 1/3 of the dia of the endmill, but then later you say your depth of cut is only about .015. ...what's the "1/3 ...." all about?

You don't say how thick the mild steel piece is, but I would use the beefiest endmill I had handy, or maybe up to 1/2" at any rate, and probably a DOC of at least .025. If .025 seemed an easy cut I'd bump it up from there.

I've never personally had a need to use it, but I've read others talk about clamping a layer of paper, or thin cardboard between work and table for better "hold". I think that's OK.

Re: "....the mill seems to flex and climb out of the cut, up and around the periphery of the work ..."
Are you sure your cutter hasn't gotten dull? What kind of RPM are you using here? Is this carbide? or HSS?

derekm
12-20-2010, 01:52 PM
Once again I need the advice of the experts, this time with milling a periphery. Milling is a new ďartĒ for me and I am nearly devoid of actual experience. I have read considerable information regarding the process but as usual, I seem to always want to begin with an oddball job that the books donít cover.

I need to mill the periphery of a piece, but only an arc about 300 degrees of the circumference. I have my work mounted in a 5C collet in a spin indexer on a horizontal mill with the surface of the circumference parallel to the endmill I am using, making this a peripheral milling operation (using the side of the endmill on the circumference of the work). I have been making a pass around by turning the handle of the spin indexer through the 300 degree arc, returning to the beginning, feeding inward toward the spindle and making the next pass. I am using about 1/3 of the diameter of the endmill.

So far Iíve had two difficulties. First, the spin indexer wanted to move on the table. Apparently the ground surface on the bottom and the smooth table surface didnít have enough friction to hold it when clamped with strap type clamps. Perhaps I didnít have the nuts holding the clamps tight enough. To resolve that issue I cleaned both surfaces completely free of oil and placed a piece of newspaper between in hope of improving the coefficient of friction between them and tightened the nuts as tightly as I dare. Iím not sure if this is accepted practice for mounting fixtures but so far it seems to have solved the problem. Perhaps someone can advise whether placing paper or other substance between the milling table and fixtures is proper technique.

The second issue is that of the endmill wanting to climb at the beginning of the cut. I am not using a climb cutting technique, however when I return to the starting point for the next pass and increase the depth of the cut the mill seems to flex and climb out of the cut, up and around the periphery of the work It doesnít matter whether I turn the machine off to return to the beginning or not. My technique has been to shut off the machine, return to the beginning point, lock the spin indexer at that point, turn on the machine, feed toward the end of the endmill .015Ē upon which the endmill will climb. I am fortunate that this is a belt driven machine and has stalled the spindle rather than cause damage to the machine. The damage to the work so far has not ruined it, yet.

The material is mild steel. The endmill I am using is a 4 flute HSS 5í/16Ē diameter mill. My depth of cut has been about .015Ē. I have made every attempt to insure rigidity of the setup. Where am I going wrong? Should I be using the endmill perpendicular to the circumference rather than parallel with it? Would a 2 flute endmill be more resistant to flex, or perhaps a larger diameter 4 flute endmill?

Your thoughts, critique are appreciated.
You need to tighten the clamps correctly - use a torque wrench.
The studs and the t-nut size need to match the bed for strength, so that when torqued up the studs are stretched elastically, but are not going to rip up the bed. a couple of low carbon steel M8 studs correctedly torqued up will give considerable grip.

Carld
12-20-2010, 03:41 PM
First, the paper under the base is ok and will work.

Second, are you sure your not climb cutting. If your undercutting the rotation of the end mill is counterclockwise looking at the end of it. Then you would be rotating the spin index in a counterclockwise direction to make the cut. If that is what your doing then your undercutting. The problem is, at the start it would be a climb cut and that can spin the work and/or break the end mill or work piece.

Third, doing it with a spin indexer is a dangerous way to do it but even with a rotary table or dividing head would be tricky. I would probably do what your doing on a vertical milling machine with a rotary table and the work in a chuck. It would still have the issue of the end mill grabbing the work and breaking things.

The problem with using a spin indexer is the only thing that is keeping the work from spinning is your hand and I can tell you if it grabs, your hand will not keep it from spinning.

For safety reasons I would not use a spin indexer to do what your doing. If you do, wear a safety shield for your face and hope your fingers don't get broken.

darryl
12-20-2010, 04:38 PM
The paper shim will increase the grip for the same clamping power, so that's ok. The various plays in the axis need to be taken up in the direction that the endmill would move the workpiece when you initially plunge it in to the depth of cut you're using. Not much you can do about that except try to begin rotating the workpiece at the same time you crank in the depth of cut. This works if you're milling the full circumference, but if you need to go less, then you're also forced to either have rigidly defined starting and ending points, or allow for some climb cutting action. It's easier to deal with the smaller forces during each plunge to depth of cut than it is to climb mill to 'backtrack' to clean up the cut to the starting point. I hope this makes sense.

In essence, you have to take up all play before the cutter touches the workpiece. That's play in the table and play in the indexer. Then when the cutter 'digs in', it will only grab as much material as flex in the setup will allow. If you have to climb cut to some extent to complete the machining operation, it would help if you can restrict the rotation of the indexer to the point where the cut must stop. You might have to arrange some kind of a hard stop for it.

I do what you're calling circumferential milling by using a pivot point attached to the mill table. The workpiece has a center hole for that pivot, and I attach a handle to it. Then I bolt a stop block to the table and the handle can come up against it. To start, I rotate the workpiece so the handle comes to that stop, then crank the work towards the cutter, then move the handle around by hand. The initial contact of the cutter against the work wants to crank the workpiece around, but it can't spin. The hand pressure I use on the handle rotates the work towards the cutter, so if I can't apply enough force, it simply doesn't proceed. If I let go the handle, the workpiece rotates back to the starting point, but can't go beyond because of the stop.

The longer the handle, the more inertia it presents to counter the pulsing from the cutting edges entering the workpiece. At the same time, the longer handle makes my job of controlling the rotation of the workpiece easier.

rohart
12-20-2010, 06:33 PM
I'd say that given the limits of the tools you have you're doing fine.

You seem to understand well how to try to take up the play in the spin indexer.

All I would suggest is that you feed in 5 degrees round from the start, and then let the mill cut the 5 degrees you missed using climb, and then back, doing the rest conventially. If you know there's going to be slack, and the cutter will always try to climb if there's slack, you might as well let it do it in as controlled a fashion as you can.

Even better, let it climb, and take up the slack, as you feed in for the next pass. That way the climb will be at minimal depth of cut. Let it climb to zero degrees, finish feeding in, and off you go.

Of course a rotary table is much better, and the longer and stiffer the spin indexer lever you have, the better too.

All this is conditional on my understanding your set up correctly, of course.

Carld
12-20-2010, 08:58 PM
There is something I forgot to add. If your turning the handle counterclockwise as I described to do an undercut it is also the direction to loosen the collet and if the cutter grabs it can also open the collet and let the work piece move around.

As I said before, using a spin indexer to do that job is not a good idea.

firbikrhd1
12-20-2010, 11:20 PM
Thank you to all who have responded. Since I wrote the original post my luck with this operation has changed some as I have changed technique. To assist with clarity I am including some photos of the setup.

http://i283.photobucket.com/albums/kk287/firbikrhd1/PC200110Small.jpg

http://i283.photobucket.com/albums/kk287/firbikrhd1/PC200109Small.jpg

The changes I have made are as follows: I replaced the original collar on the Spin Indexer which was held in place with 3 set screws with a clamp type collar. This has taken any end play out of the spin indexer and allows it to turn very smoothly without binding due to a collar that is slightly crooked. Under the aluminum lock screw I placed a piece of plastic to act as a brake providing pressure to prevent easy unwanted movement of the spindle in the wrong direction. Lastly, I locked the indexer at the ends of desired travel and made plunging cuts into the work at those points thus avoiding the need to work up to that edge when cutting the remainder of the circumference. This eliminated the possibility of the mill flute grabbing the edge and pulling it up and over as has already happened. Evidence of the previous mistake can be easily seen in the first photo.

While this may not be the ideal way to perform this operation my resources are currently limited so I must make do with what I have. It is now working pretty well given that limitation.

I also realize the indexer not being fully supported by the table is not the best situation and I have found a remedy for that as well. Today I ordered collets to fit the mill spindle which will take up much less real estate than the endmill holder you see here. This will allow me to center the indexer for future operations.

Derekem mentioned torquing the nuts holding the clamps. That is probably not a bad idea. I am concerned about how much pressure I can apply as these are only 7/16" T slots. The table already has a chunk out of the T Slot where a previous owner either bottomed out a bolt or pulled too hard while tightening. The threaded stock I'm using is all thread which is 5/16-18 thread. I used a short combination wrench (about 5") to tighten what you see here but any torque specifications anyone might suggest are welcome.

darryl
12-21-2010, 12:14 AM
Another suggestion- it looks like you're cutting on the bottom side of the workpiece. This puts upward pressure on the indexer and the table. I'd be lowering the table and cutting on one side of the workpiece- putting the cutting force downwards. In other words, the workpiece should be to the left of the cutter, and the axis of the indexer and that of the cutter at about the same height. If anything, have the cutter axis a bit higher. That will help keep play and flexing to a minimum.

Paul Alciatore
12-21-2010, 12:16 AM
I like your idea of doing the two ends first with plunge cuts. You could expand on this and take about 90% of the metal off with spaced and overlaping plunge cuts before finishing it by rotating the work.

As I understand it, you are rotating the spin indexer by hand. This is never the first choice and the handle used for tightening the collet is far to short to serve in this capacity. You would be far better off if you constructed a longer handle, 18" or more long, and attached it directly to the OD of the indexer's spindle. Bore a matching hole in the handle and slit it's end off half way through the bore to create a removable cap to allow attachment and tightening. This will not mar the spindle. I made a long handle for my lathe spindle for threading operations and it is just great. It provides tremendous control. In order to use this, you may need to keep the indexer over the table edge.

Fine sandpaper also works good to keep things from slipping. Also aluminum foil: the foil is soft and can distort under pressure to fill the small grooves in both the table and the indexer. This helps to increase friction between the two.

Another thing you may consider is doing the job in sections to lessen the forces on the work/tool. Instead of what looks like a 3/4" to 1" (measured horizontally in your pictures) width of the cut, do it in several 1/4" increments, leaving a few thousanths for the final cut to the the finished OD which would be done the full width. You will have better control.

firbikrhd1
12-21-2010, 08:16 AM
Darryl, I hadn't thought of how forces acted upon the indexer & table in that way. Originally I had my work set up much as you suggest but thought, perhaps over thought, that by having the cutting take place at the lowest point the forces would be minimized as pressure applied by the cutter during plunging cuts would have less "leverage" pushing against the work from a low point.

Paul, thank you for your thoughts. Others have mentioned a longer handle as well and I agree that manual turning of the indexer isn't the best. In the future I plan to use a worm gear system to convert the spin indexer to a dividing head. Of course that would eliminate the handle issue entirely as I could simply crank the spindle around with the worm. Believe it or not I haven't had much difficulty had much difficulty with controlling rotation with the current setup. The "brake" I added under the locking screw has worked very well as an aid to control.
I had thought of using fine sandpaper but decided against it. I will take it into consideration in the future now that I've learned it is an acceptable method. I like the aluminum foil idea. It is soft enough that it won't mar anything but conforms to both surfaces to make a solid contact area.
The picture you see is after several cuts about .1" deep, axially. The picture is misleading as it looks as if I have made the entire depth (width) of cut in one pass, which is not the case.

Carld
12-21-2010, 09:48 AM
firbikrhd1, using a longer handle is not a good idea. The handle on the spin idexer is used to tighten the collet holding the work. Because of that when turning the handle in a counterclockwise direction your loosening the collet so keep that in mind when your thinking about using a spin indexer as your doing.

Benta
12-21-2010, 10:51 AM
That end mill holder is not a good idea, it's hanging way too far out of the spindle and will flex.
But you mentioned that you are getting collets, so that should be fixed soon.

Benta.