» » »

 
  post #1  
11-01-2010, 10:16 PM
Senior Member
 
About
: Dec 2008
: 183
: 6 @ 10
  is on a distinguished road
Icon23 Mill Cutting Tools & Surface Finish


Mill Cutting Tools & Surface Finish
Herein are my notes on selecting the right cutting tool and achieving a decent surface finish with it. They are to date a rather disorganized collection of anecdotes found elsewhere. Over time I will organize these and test each one, discarding those that don't work so well and emphasizing those that do. For now, I am using a font convention to differentiate the tips:
If I have written a tip in this font, I have personally verified it.
If it is written like this, I found it on the Internet and am awaiting verification. These are the ones to take with a grain of salt.
Climb Versus Conventional Milling
This has to do with the direction of the cut and how it relates to the rotation of the cutter. Climb milling is when the direction of cut and rotation of the cutter combine to try to "suck" the mill up over (hence it's called "climb" milling) or away from the work. It produces the best surface finish. Here is a diagram showing climb versus conventional milling for a number of orientations:

Keep in mind that for this illustration, it is the workpiece that moves, not the spindle. On some machines, like a gantry router, the spindle moves, so the labels would reverse. I keep it straight by thinking of the spindle as a pinch roller that can either help move the workpiece in the direction it was already going (climb milling), or that might fight that movement (standard or conventional milling).
Try the experiment on your mill of cutting both ways and you'll see that climb milling is a lot smoother and produces a better surface finish. Note that depending on which way you are milling, you will need to make sure your workpiece is supported well in that direction.
Some further thoughts on Conventional Milling:

  • The width of the chip starts from zero and increases as the cutter finishes slicing.
  • The tooth meets the workpiece at the bottom of the cut.
  • Upward forces are created that tend to lift the workpiece during face milling.
  • More power is required to conventional mill than climb mill.
  • Surface finish is worse because chips are carried upward by teeth and dropped in front of cutter. There's a lot of chip recutting. Flood cooling can help!
  • Tools wear faster than with climb milling.
  • Conventional milling is preferred for rough surfaces.
  • Tool deflection during Conventional milling will tend to make the cut deeper.
Some further thoughts on climb milling:

  • The width of the chip starts at maximum and decreases.
  • The tooth meets the workpiece at the top of the cut.
  • Chips are dropped behind the cutter--less recutting.
  • Less wear, with tools lasting up to 50% longer.
  • Improved surface finish because of less recutting.
  • Less power required.
  • Climb milling exerts a down force during face milling, which makes workholding and fixtures simpler.
  • Climb milling reduces work hardening.
  • It can, however, cause chipping when milling hot rolled materials due to the hardened layer on the surface.
  • Tool deflection during Climb milling will tend to make the cut shallower.
There is a problem with climb milling, which is that it can get into trouble with backlash if cutter forces are great enough. The issue is that the table will tend to be pulled into the cutter when climb milling. If there is any backlash, this allows leeway for the pulling, in the amount of the backlash. If there is enough backlash, and the cutter is operating at capacity, this can lead to breakage and potentially injury due to flying shrapnel. For this reason, many shops simply prohibit climb milling at all on any manual machines that have backlash. Some machines even came equipped with a "backlash eliminator" whose primary purpose was to enable climb milling and its attendant advantages.
One way to think of it is to consider the concept of chip load. This is a measure of how much material each tooth of the endmill is trying to cut. Typical values for finish work would be 0.001 to 0.002" per tooth. For roughing work, that might increase to 0.005". Now in the worst case, climb milling may grab the table and slam the work into the cutter by the full amount of backlash during the instant when a single tooth is cutting. You can therefore add the backlash to the chip load to see what your new effective chip load might be in this worst case. Suppose you are roughing 0.005" per tooth and have 0.003" backlash. In the worst case, your chipload will soar to 0.008". That's probably not the end of the world, but it is a strain. Now suppose you have an older machine with 0.020" of backlash and are running an 0.005" chipload. If the worst happens there your chipload will soar to 0.025", which is probably going to break the endmill or something else.
The second thing to consider is whether cutting forces are strong enough to pull the table through the backlash in the first place. A lot will depend on the exact cutting scenario together with your machine. If you've got a fancy low friction linear way machine, it can grab easily. If you've got a lot of iron in the table, and maybe you're running with the gibs tightened a bit, it'll be harder. There are ways to calculate the cutter force, but in general, smaller end mills, less depth of cut, lower feeds, and lower spindle speed will all reduce the cutting force and make it less likely the cutter can drag the backlash out of your table and create a problem.
In general, the home machinist should investigate climb milling carefully and responsibly. Don't run afoul of the backlash problems that can lead to spectacular failure. Know what your backlash is, and therefore what kind of chipload you can run before getting into trouble.
Postscript: Conventional Milling for Accuracy, Climb Milling for Surface Finish
I came across this interesting diagram during a Google search:

The arrows show where the cutting force is attempting to deflect the cutter...
The arrows show where the cutting force is attempting to deflect the cutter. The takeaway is that when the accuracy of the wall's location is critical (as when profiling), conventional cutting yields a better result. It deflects the cutter in a direction that is less directly vectored towards or away from the wall. OTOH, it is well known that surface finish is better when climb milling.
I'll be tempted to try conventional for roughing, followed by a light depth of cut climb milling pass for the final finish. At the very least, one should avoid too much depth of cut when climb milling. The same article suggests that when deflection is to be minimized, use no more than 30% of the diameter of the cutter for conventional milling and 5% for climb milling.


Basic Milling Cutters

Fly Cutters
A fly cutter will often produce the best surface finish because they allow you to finish a very wide area in one pass. They make a distinctive sound as their single cutting edge makes its cut on the workpiece. It's a noise that drives some machinist's crazy and can best be characterized as a loud thumping that sounds like its about to tear itself apart.
I bought one of the cheap R8 fly cutters from Lathemasters (really great people to deal with!):

Lathemaster Fly Cutter with a Brazed Carbide 3/8" Cutter...
There is no harm in buying an inexpensive import fly cutter like this--they're not exactly precision tools. This particular one is designed to work with lathe-style left-hand (e.g. cuts left to right) cutting tools. It's built for a 3/8's inch tool, but you can use shims to run smaller tools if you have a small lathe like me. I like to use brazed carbide tools in mine, although I suppose you could also try a carbide insert tool as well.
I find my fly cutter does not want to take incredibly deep cuts, but it does get the job done by covering a wide area. In general, when face milling, you should try to use a mill that is slightly wider than the area you are cutting. For most home machinists, a fly cutter will be that solution when facing wide areas. If you have the $$$, a wide indexable face mill can produce a finer finish, but they're not cheap.
To figure your "feeds and speeds", think about getting 0.003" of feed per revolution. The math looks like this:
Feed Per Revolution = IPM / RPM (You want to get this equation to yield about 0.003" per revolution for best results)
Let's try some examples. Suppose you are turning your feed handle on the mill about once per second. On my mill, that's 100 * 0.001" = 0.1" / second. Converting to IPM, we get 60 * 0.1" = 6 IPM. Now to figure the best spindle RPM for that feedrate, I get RPM = IPM/0.003 = 2000 rpm. The moral on this story is you can probably turn your handwheel a bit faster unless you want to run your spindle at 2000 rpm!
Like most things, your fly cutter will perform better with positive rake on the cutting tool. I haven't yet seen any positive rake brazed carbide tools, so I suppose you would either need an insert like a CCMT or you'd need to grind your own tool from HSS with some positive rake.
Someday I want to build an R8 fly cutter that has a built in CCMT insert. I'll use a design similar to my dovetail cutter, and I don't think it would be a hard tool to make.
Here is a sample of the surface finish possible with a fly cutter:

This was one of my first cuts with the tool, I was squaring a block for my vise stop project. I didn't run the spindle nearly as fast as I should have for a great finish, but it still came out pretty decently. The moral of the story is that fly cutters are cheap and pretty easy to use.
Fly cutters are very sensitive to tram. If you think about it, the cutter is apt to cut a concave instead of flat cut if your mill's head is not trammed properly. Watch your workpiece carefully as the leading edge moves through the center of the fly cutter. All the cutting should have been done as it was travelling the first half of the way if you head is in tram (think about it carefull, that's one geometry for one direction, so you actually have to see this behavior cutting in both directions to be sure). If it cuts more on the second half, you can be sure the tram is such that the head has that half tilted downward slightly. You can see the tramming effects in this photo:

The mill head is close to being in tram, but you can see that one set of marks is a little heavier than the others. If you're really out of tram, you can only see one set of marks. Also, the larger the fly cutter diameter, the more the tram effect is exaggerated.
I believe in fly cutters for achieving a good surface finish over wide areas such as plates. If the workpiece is narrow enough, I prefer indexable face mills. I did buy an even bigger fly cutter on eBay, and it's something that would be pretty easy to make:

If you are going to grind your tool, here are some useful pointers I have gleaned from my Internet travels:
Fly Tool Cutter Geometry
Fly tool cutter geometry is a craft within itself. For steels such as 1018 I usually use C-8 carbide bits with zero rake on the face and about 5 degrees relief. The point radius should be kept to less than .06". Larger radiuses will deflect the tool/holder and resist, resulting in chatter, squealing and poor finishes. This same tool can be used with aluminum however, aluminum requires extremely sharp tooling. For aluminum such as 6061-T6, I usually use a High speed bit blank. Grind about 60 degrees of rake on the face 1/2 from the end. 5 degrees relief on the bottom. The radius is left large, As much as .25" Strange looking tool but, they leave a very good and more importantly, accurate surface. They can even be used to fly tool rubber without freezing it. Handlap the cutting edge on a whetstone for best results. This type of cutter actually cuts, zero rake cutters on aluminum merely smear the material away. The file surface experienced was probably caused by too small of a radius or no radius. A big no-no with flytools or any cutter for that matter is allowing it to dwell spinning in one spot against the stock, not cutting anything. It takes the edge off of it it immediately. Keep it moving. This same cutter geometry works for lathe tooling as well in soft materials.
I would use carbide, allot of rake, allot speed, and a small radius on the cutting edge, use coolant is all you need most of the time, or brush on a bit of cutting oil or varsol, or WD 40, perhaps on your very last cut, otherwise you have too much smoke.
Using a single tooth fly cutter with a .03R on a 3" flycutter turning 1200rpm at 2.5-5"/min, you should get a very nice finish. The smaller the tip radius, the slower the feed to keep the finish decent. Too large of a radius increases tool cutting pressure and the tendency to chatter. That can also lead to greater warpage in the part. I don't know if that is from friction induced heat or odd stress relief over a larger area. Not much you can do if the radius is required. Lighter finish cuts and a very sharp edge should help. No fancy cutter material is needed unless cutting steel. Interrupted cuts are not as kind to thin brittle cutting edges. Rake can help and hinder depending on the chip desired. Long stringy chips are not fun to deal with. I prefer to have a chip breaker as was presented, but not as thin at the cutting edge with relief on the bottom so only the radius portion is cutting. The basic positive rake left cutting lathe tool works in most instances. No need to complicate it with fancy grinds for the most part.
In response to, "It seems like I was running more like 1000 rpm at 15-20 ipm. I'll try a lower feed tonight." Yikes!, that is .015-.020/rev with a single tooth fly cutter? Not surprising it was groovey! LOL! Getting the feed rate down to .003/rev is going to be a big improvement. At least below 5"per min. 1000/3"per min=.003 If you see a nice cross hatch pattern develop, it means the head is trammed properly. If you see any steps for successive passes in either the x or y directions, this will indicate the head is out some or wear in the ways. Be very aware if you have an automatic oiler on your ways. The oil can lift the bed floating on the oil during a cut when least expected at the worst possible moment.
I tried running at .010 deep 1200 rpm and 5 ipm, results; Beautiful finish! and no step between passes. Head seems to be trammed in pretty good, my only complaint was that the cutter started squealing once it was fully engaged in the part. I reached for some undiluted cutting oil (only thing that was handy) but it didn't seem to help, i'm sure that vibration doesn't help the finish. I have a job starting tonight that I will have to take .090 off a bunch of 3" wide peices. The finish on one side really doesn't matter so I'm thinking I will face the visible side as done last night, turn it over and then do the major metal moving. How much is safe to take off in one pass?
Trying to draw the appropriate cutter geometry, the cutter I am using right now is carbide so I can't do much modifying to it. I wan't to grind one out of HSS today. The attached is 30 degree rake, 5 degree relief and .062 radius. I am not sure how to incorporate the chip breaker, it seems like it would go right through the cutting edge. When grinding, do you grind the radius first then the relief and rake?

Joe, I've added some arrows to your image. This will remove metal on a light cut, but not as nicely as it could and not on heavier cuts. Assuming it's oriented in the image as it would be in the mill, the top rake is ground as if you want to plunge into the material in the direction of the green arrow. Think of this bit mounted in a lathe, a nice knife tool. It will cut well going from left to right. But take this tool this to the end of your work and try a plunging facing cut and you'd have problems. Thats what's happening when you fly cut - its cutting in the direction of the red arrow, not the green.
If you study the image I posted a few posts above, you'll see there is a crescent shape ground out of the end - the is what gives the tool the top rake for cutting in the direction of the red arrow, its not a chip breaker (well it is, but thats just an added bonus). When using a fly cutter, its usually an interrupted cut so long chips aren't an issue anyway imo
This radius is exaggerated in the image I made - grind it on the corner of the wheel, its nothing fancy it doesn't have to be exact - anything that'll give you 10-15 top rake (30 is too much, does hurt the action of cutting but it does weaken the cutting edge). Btw, this is a standard (and the best imo) grind for facing tools for lathe work.
Also, don't forget clearance It looks like you have no front clearance - the front of the tool (as per the red arrow) needs to be angled away from the cutting edge. If used as drawn, the edge after the cutting edge leading up the blue arrow will hit/rub, preventing the cutting edge from engaging properly
After grinding, stone the tool a little, sharpness does count, and put a small radius on corner. no need to grind the radius, doesnt have to be big, just do it with a stone. This will give both a great finish and allow fairly high removal rates. As always, work out your feeds and speeds!
The grind you drew would be perfect for fly cutting if the tool was held vertically. Build a fly cutter of that type to use the brazed carbide tools if you don't to modify them. the type of fly cutter that has the tool more oriented as in your image, is better off with hss ground as per above.

I think the confusion is over what exactly is the cutting edge - the thick red line is the cutting edge - this is a what traverses the work forming a a chip. we are used to, because of lathe work, thinking of the cutting edge as being along the side of the hss bit. in this case its on the end - its the same principal as any other cutting tool but you have to turn your head sideways to see it btw, if you don't grind your lathe facing tools like this, give it a try - it will be your new defacto. check out http://easyweb.easynet.co.uk/~chrish/tl-tools.htm. Theres a diagram half way down showing rake/clearance angles etc. every cutter, from simple lathe bit, to drills to shell mills has them, you've got to orient yourself to what the cutting edge is then where the rake and clearance has to be. Its good to learn this as the angles change substantially for different materials knowing how to grind a bit or drill for different tasks is fundamental to good workmanship. Next youll be making your own specialty cutters from drill rod . my comment in blue - you don't need this angle. purple- shows how the top rake is formed by the crescent shaped grind. front clearance, yours doesn't yet have any, i've shown it on the side view in orange.

The chatter can be a result of excess radius, cutter angles, RPM, or thin material to begin with. Thin material should be fully supported, not just on parallels in a vise. Reading the material, cutter or reaction to changes can help pinpoint what is contributing to the chatter. The cutter you have posted with the chip breaker is pretty close. I would not cut the groove that large or deep and if I did, I would reduce the hook at the cutting edge some. The idea is to get the chip to curl against itself so it will break off or at least be compact. I would also cut it at an angle behind the radius tip and not directly parallel with the front edge. Application dependent again. As a side note.....on thin materials in taking more material off of one side verses the other. That can set up differing stress relief, which will spring the parts toward the side with greater material removal when released from workholding. With a single tooth cutter, you could cut .125, but realistically around .06-.09 is not uncommon with a smaller tip radius. I have taken .375-.5 with a 4 stepped 4" flycutter. Just try to avoid taking a heavy pass that envelopes the full cutting radius of any fly cutter. Wiser to step over one half its diameter or use a larger fly cutter.
Well I fly cut 20 3"x6" parts last night with something similar to the last geometry posted and it is still cutting well. Now my biggest problem is how to keep the chips off my keyboard! I'll post a picture if I get a chance, I don't know how you could get much better of a surface finish.
For an even better finish, instead of using cutting oil for lube use a little kerosene. Careful for fire though! In my experience when cutting alum, a little kerosene goes a long way and the finish is prismatic.
Hogging Mills
Fly cutters do a pretty good job, but they can be a bit slow at times. You can remove material the fastest with a "corncob" style hogging mill. Here's a little Niagara 1 1/2" hogging mill I use very successfully on my mill:

1 1/2" Niagara Hogging Mill...
This is a 6-flute mill. Surface finish is much rougher than with the fly cutter, but you can remove material a lot faster. Here is a comparison of the surface finishes:

Hogging cutter on the left, fly cutter on the right...
Coolant
Rules of thumb for high speed machining:

  • 10 gallons/minute per inch of tool diameter
  • 0.5 gallons/minute per HP on the spindle

Feeds and Speeds
One professional shop uses the following guidelines:
Spindle RPM = 4 * CuttingSpeedFPM / CutterDiameterInInches
CuttingSpeeds (Idealized for perfect conditions including flood cooling, you may want lower!):
500 plastic
300 aluminum
200 brass
100 mild steel
50 stainless steel
These numbers are good for slot machining, with a cut on both sides. If you want to be a bit more aggressive you could add 50% when peripheral machining (i.e. only cutting 1 edge).
Example: What RPM for a 1/2" end mill cutting mild steel? RPM = 4 * 100 / 0.5 = 800 rpm. How about aluminum? 2400 rpm. You can go 3x faster with aluminum according to these numbers.
FeedSpeed = ChipLoad * n * Speed
ChipLoad = amount of material removed by each flute per revolution. 0.005" for roughing, 0.001" to 0.002" for finishing is typical, but exact values are in Machinery's Handbook.
n = number of flutes
Speed = spindle rpm
For example, take my dovetail cutter. If I run my spindle at 900 rpm and I want to take finishing cuts only, I should be feeding at 0.001 * 1 * 900 = 0.9 inches per minute! If roughing, it should be 4.5 inches per minute. Now, to cut aluminum, I should be running at 4*300/1 = 1200 rpm. My handwheel feeds 0.100" per revolution, so if I am turning it 1x per second, that's 60 rpm * 0.1 = 6" per minute.
Let's go back to the 1/2" end mill on mild steel and aluminum. My Chip Load at 6" per minute feed and the spindle speeds calculated above is 6/(4*200) = 6/800 = 0.0075". That's fast even for roughing, so I'll need to slow it down. 3/4's to 1/2 speed would be good for roughing. I need to go really slow for finishing, close to 7 seconds per revolution on the handwheel. Same with the aluminum case. However, I could also slow down the spindle too. Let's run it at half speed. Now the chip load is 6/(4*100) = 6/400 = 0.015" That's even worse! Basically, I am forcing my end mill to cut off too much steel on each revolution. The bottom line is I need to feed more slowly.
You can see why people will pay some money for power feeds. It takes quite a long time to feed at these speeds and a lot of patience to do it by hand!
Depth of Cut should never be more than maybe diameter/4 to diameter/2 for a home shop. I use diameter/8 to diameter/4 for roughing and less for finishing.
Coated cutters allow 25-50% more aggressive use as well.
There are some simpler ways to look at this as well. If you are getting long stringy chips, crank up the speed. If the chips are coming off as blue 6's and 9's, you may have it too fast, or just right. For best tool life, try to get brown chips. Chip color assumes steel, of course.
Drills Versus Endmills
A twist drill can remove cubic inches of material faster than an endmill for nearly any situation. Consider chain drilling (drill a series of holes along the tool path) and then milling out the chain to a smooth surface. Also, when cutting a closed slot, start the slot by drilling a hole and then extend the hole with your endmill. If it's a production job, remember you'll have to factor in toolchange time if you introduce drilling to your process. Lastly, drill bits are often cheaper than endmills.
Tips On Surface Finish

  • Surface finish is a function of feedrate (in/rev), axial depth of cut, radial depth of cut, and cutting speed (in/min) in that order. To achieve a better finish maximize feedrate, minimize axial depth of cut, minimize cutting speed, and minimize radial depth of cut.
  • Keep a fine finish pass. Use a different cutter than you roughed with for best finish.
  • Keep your finish and rouging cutters separated. If you use the same type of cutter for both, start new cutters as finish cutters and move them to roughing after a little while.
  • If the finish is down a hole, relieve the upper cutting surface so it won't be in contact with the hole and just cut at the bottom. This will reduce chatter, and works well when profiling with ball nosed end mills as well.
  • Use a larger diameter cutter, if possible, as it will flex less.
  • Keep the gibs tighter for less chatter.
  • Make sure the z gib is tight by lowering the spindle on a aluminum block on the table to put a bit of upward pressure on the head and tighten the gib on my x4 i was able to turn the screw of the z gib by a turn and a half this will reduce most if not all the chatter you might get from the head assembly.
  • I've seen Widgitmaster use a spacer to lock the quill on his Bridgeport. I try to cut with the quill locked on my Industrial Hobbies machine whenever possible.
  • Try a 4 flute on aluminum for the finish pass. By the time you're ready to finish, there should be ample room for chips to fly and the extra flutes and light depth of cut will make for a nicer finish (the equivalent of more spindle rpm with a 2 flute).
  • Lots of folks swear by 3 flute cutters on aluminum.
  • Reduce the flex in deep holes on a 2 flute cutter by grinding down one flute. Now the cutter acts like a boring bar.
  • Balance the diameter of a ball end cutter versus the rigidity. Remember, the part of the ball near the axis moves slowly. A smaller ball interpolated exposes more of the surface to a faster moving cutter, leading to a better finish. But, the smaller cutter can flex more. Hence the need to balance these two factors.
  • TiAl coatings should not be used to cut aluminum. The aluminum in the coating sticks to the workpiece aluminum. Uncoated won't last as long but give the best finish on aluminum. If you want to use a coating for longer cutter life, use ZrN or TiB. Micrograin C-3 (low cobalt) carbides will have the best life.
  • Try to get the mill head as low on the column as possible to maximize rigidity.

Other Tips and Techniques

  • Don't cut a slot in a single pass if the surface finish or accuracy matter. Cut the middle, then cut each edge on its own.
  • Be careful not to squeeze a slot in your vise. Ideally, the slot should be cut on the Y direction so the slot walls are not being squeezed together in the vise. Failing that, sit the workpiece up high so the slot is above the jaws.
  • Prefer center-cutting endmills. Non-center cutting have a little hole in the center where no cutting action occurs. If you try to plunge the cutter, you get a little nib there that quickly prevents further progress plunging.
  • Blue on the tips of a mill cutter indicates the chips are welding themselves to the cutter. That cutter is probably done for (in need of sharpening) and you need to either slow down or use some flood coolant.
  • Shorter milling cutters and drills will be more rigid.
  • Keep your finishing cutters separate from roughing cutters. When finish degrades, you may choose to demote the cutter to "roughing" status. Use made-for-the-purpose roughing cutters for hogging. These often have serrated flutes.

Sponsored Links
  post #2  
07-05-2010, 09:26 PM
Junior Member
 
About
: Mar 2010
: 24
: 0 @ 10
 is on a distinguished road

  post #3  
26-06-2010, 03:21 AM
4   4
Member
 
About
: Mar 2009
: 40
: 0 @ 10
  4 is on a distinguished road


()



« | »



Surface Coatings wise_55 0 03-11-2010 07:57 PM
1 19-07-2010 02:31 AM
Measuring the Flatness Laser 2 16-01-2010 08:55 AM
Milling 0 08-10-2009 02:07 PM
milling operations 1 24-04-2009 09:48 PM


09:48 PM.

..

RSS RSS 2.0 XML XML2 SiteMap SiteMap2 ARCHIVE HTML HTML2 EXTERNAL JS URLLIST

- -     

Powered by vBulletin® Version 3.8.4, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd. TranZ By Almuhajir