What's new
What's new

Turning low micro finish, tips?

Nick H

Aluminum
Joined
Jun 13, 2007
Location
Westfield, WI
Well I've got a few jobs that I've ran over the past year that require some decently low micro finishes of 32 max. I took the job from the multi-spindle department in our shop because they could not turn 32 micro and had to have parts sent out for grinding. Using some Sandvik CNMG 432 WF 4015 grade inserts I was able to hold finishes in the teens and 20's easily and with decent tool life. Well Sadvik has cut this grade from production and replaced it with a 4215 grade. Problem is the insert life is more then cut in half now.

I'm turning 12L14 steel on a Puma 10HC lathe, 1.747 o.d. sfm of 950 feed .005", worked excellent on the old inserts. I had to turn the sfm up to 1200 to get a decent finish on the new inserts, but as stated it only last for 30 pcs instead of 150+ pcs per cutting edge! Then it jumps up to 32+ on the micro finish. I'd like to keep the spindle speed under 3k since it's rather large stock and being feed by bar feeder in 4ft. sections. Also might need to note that O.D tolerance is only +/- .00025". At spindle speeds over 3k it wants to use the whole tolerance instead of holding .0001 or so. Coolant is a synthetic water based, Benz Oil Dura Kool H.j

Anyone got any tips or ideas?
 
You might want to try inserts with a wiper type geometry; the ones I have seen are fron WNT and are called 'Masterfinish'. I guess other manufacturers have similar products. They claim such a good finish that grinding is unneccesary.

I must admit that I have never used this type of insert; just looking at the catalogues.

Regards,

Mike.
 
A feedrate just below F.003 should give you almost exactly a 32 finish bsed on a 1/32 noseradius insert.
SFM of 1000 sounds OK.
If you go to a little bigger noseradius, you will get a better finish or you can feed a little faster.
Heinz.
http://home.columbus.rr.com/hputz

Is 32 or better really that low? You should be able to get the finish you want at 2500 RPM's and a feed of .0028-.0035. You may be way low on the insert start point for feed and speed, which is why they aren't lasting as long. Just my opinion. :) Make sure you are leaving at least the nose rad amt in finish pass amt.
 
Last edited:
A lot of the bigger width groovers are also made to be used as a turn toy, and such have a .004-.010 Rad. You can even git'm with chip breaker geometry. (Kennys Metal NG3156LK [RK])

A "Thinbit" groover for .031 prolly aint gunna be a good target. ;)

I was running a Manchester Dogbone on 12L on Sat with really swell SF's! (Looked burnished!) Also leaves a good finish on 316 - but I don't know what a TNMG would'a looked like in comparison on that material. 12L I can compare.

EDIT:

ACTUALLY - you could possibly git your sharp cornered inserts to doo this in a pinch - if you kick the holder on a slight taper so that the face of the toy is cleaning up what the side/corner originally peeled. (Follow that?)

------

I am Ox and I approve this here post!
 
Two things influence surface finish and appearance.

Tearing and rewelding of the material leaving a dull finsh and a kind of fuzz on the part. Up-sharp inserts or cermets eliminate this. Cermets are great for mirror like shiny surfaces.

The other is the grooves or cusps left by the tool geometry. It is the shape of the area just BEHIND the high point of the tool radius that generates this form.
Larger rads curl away from the part slower leaving better finishes.
Wiper inserts have a small flat (or sometimes a 2" radius) just behind the radius high spot to smooth out the part.

As Ox said you can do the same thing with a groover using the flat tip to act as a wiper smoothing out the tool marks.
Also noted by Ox is that if you can reorient the tool so that the trailing edge behind the rad is flat to the part it will cut off the high spots acting like a wiper. You have to get this just right, almost flat but not quite. If too much of the tool is touching on the side you'll get chatter.

If the gometry is right you can routinely generate finishes in the single digits to low teens using feed rates from .008-.015 IPR.
Bob
 
Sandvik CNMG WF is a wiper insert...

Regardless, I'd absolutely be using a cermet for this. As you're using sandvik, the 5000 series grades are cermets. gc5015 is the one I generally use, in a dcmt or vbmt geometry.

Gregor
 
You are still running the 4225 too slow according to the Sandvik catalog. It suggests 1650 sfm for an F.004 feedrate. Other lathe programmer tells me you have to really keep the SFM up on this insert for it to work right.

I agree that a cermet is the way to go. I am partial to the XT3 grade from NTK, tho I must admit to having excellent results with cermets from several other manufacturers.

I ran a 250 piece order of 316L SS off using one corner. It required a 20 finish. Ran at 850 SFM, but don't remember the feedrate. Think F.002. It was a TNMG-332 series insert.

Currently running 416 SS with a Ceratip CCGT21.50.8 FL-U (.008R) insert. 32 max. finish. Would love to use a bigger radius insert, but size of hole dictates size of tool, which dictates size of insert radius to avoid chatter. One corner normally runs for one shift, about 230/240 parts.
 
Sandvik CNMG WF is a wiper insert...

Regardless, I'd absolutely be using a cermet for this. As you're using sandvik, the 5000 series grades are cermets. gc5015 is the one I generally use, in a dcmt or vbmt geometry.

Gregor

I think I have that exact insert. I've got some CNMG 432 inserts with 5015 as the grade, they are a silver color? If this in fact the insert, where should I start for sfm and feed? I stuck one of the inserts in, and started at 1200 sfm, feed of .006" Micro was in the low 20's but quickly climbed into the 30's and it's persistant to stay in the low 30's which is to close for comfort. I tried upping the sfm to as high as 1450, and feed as high as .009" and as low as .003" Could really not tell much of difference in finish. As it would come down back into the upper 20's and bounce around there and go right back into the low 30 Ra finish.

Normally I'm all about testing to see what works and does not work, but I've got very limited stock for this job, waste is not so much an option.
 
Just a dumb question, but how are all of you checking your surface finish? Are you using a digital device to compare the surface finishes, or are you just familiar enough to look and know? Sorry, but after reading all of that it just left me wondering.
-Greg
 
Just a dumb question, but how are all of you checking your surface finish? Are you using a digital device to compare the surface finishes, or are you just familiar enough to look and know? Sorry, but after reading all of that it just left me wondering.
-Greg

No that's a good point. I use a Mitutoyo surface finish tester. It's a device with a needle that runs across the part and reads the Ra. We've got a test standard, and means to calibrate it also.
 
Nick, that is the correct grade, but I'd suggest that the reason you are struggling is the geometry of the insert. Negative rake inserts like CNMG are not the best for fine finishing.
I'm all metric, typical surface speed for me in that kind of material would be 500-550 m/min with a feedrate of about .1 -.15, depending on the tip radius.

I don't know how much you're leaving for a finish pass, but if you're stuck with that particular insert geometry you might have better luck if you leave more material for your finish pass. I don't know how much you can get away with and maintain that tolerance, but leaving around 1mm DOC for your finish pass will help improve the surface finish with that insert.

If you have any positive finishing inserts you can use instead, I'd go with them...
 
Sometimes turning the coolant off for the finish pass helps the surface finish of the part. Chips get hotter and shear off cleaner...or so the theory goes.
 
Sometimes turning the coolant off for the finish pass helps the surface finish of the part. Chips get hotter and shear off cleaner...or so the theory goes.

Great example of "thinking outside of the box", if I may use that cliche. I agree that that can be one of the many solutions. Thing is, you find what works and the tool company changes things. Been so long (maybe) since the outcome needed to be expiermented... Part of the reason I love this site!! :D You get your options from experience. :cheers:
 








 
Back
Top