View Full Version : Best combination of carbide insert/steel grade for mirror finish on a manual lathe

08-20-2017, 10:59 PM
I'm trying to figure out a combination of carbide insert/steel grade to get a predictable finish on my lathe. I have a manual Harrison M250 so not easy, but the winning combination does exist.

Here is where I am at the moment. I have used finishing inserts from Kennametal (DCMT KC730) and Sandvik (UF 525 and UM 5015), on some 12L14 leaded steel bar and some other offcuts.

Turning at about 500rpm on a 8-16mm bar, no coolant or cutting oil:

EN1A unleaded - crappy finish with KC730 and UM5015, a bit better with UF525
EN1A leaded (12L14) - crappy finish with KC730 and UM5015, a bit better with UF525
AISI 1010 - not mirror, but flawless finish with Sandvik UF525
some broken Snap-on screwdriver shaft steel - mirror finish with Sandvik UF525 and Kennametal KC730

It would be nice to know what steel Snap on is using on those screwdriver shafts. I would expect the tip to be some proprietary material, but maybe not the shaft.

08-21-2017, 12:42 AM
You should try something harder like 4130 prehard steel. Very easy to get so shiny..
And use way more speed, 8mm bar might be a little difficult even on 4130 prehard because correct finishing speed is something like 5000rpm..

With 1010 I have got best results with sharpened ceramic inserts.

08-21-2017, 02:01 AM
500 rpm? on 8mm you need many times that for optimal cutting. 41sfm is not optimal for carbide at all. You MIGHT sneak by with some aluminium cutting inserts in 12L14, but you should be using HSS at those speeds. It's hard not to get a great finish on 12L14 at any reasonable speed. 500rpm on 16mm isn't much better.. just 82sfm. And at those speed don't worry about the coating (KC730 is a coating grade)- that won't make much difference - focus on geometry.

OK, so few have lathes that can cut at 300-800sfm on those sizes... but sometime you can't use typical carbide well. I get by sometimes with hightly positive rake fine tip finishing carbide, but I'm not shy about slapping in HSS tooling, and resorting to emery paper if finish really matters.

Ian B
08-21-2017, 02:03 AM

With an M250, you should be able to get perfectly acceptable results - it's a good lathe.

Have you tried using high speed steel? Maybe worth a try - buy a couple of sticks, maybe 8mm or 10mm square and grind them freehand on a bench grinder. Plenty of advice on Youtube etc on how to do this. Maybe go for a small radius round nose, or a right hand knife, set to rub very slightly. For the cherry on the icing, finish the tool with a fine oilstone.

Then see what finish you get. For the diameters you mention, try doubling the speed, and add some cutting oil - a brush will do, doesn't need to be flood coolant. Shallow cut, fine feed.

This is how it was done for decades in home shops, long before carbide came along.


08-21-2017, 03:49 AM
I have never got even half as shiny result with HSS on soft steels compared to harder steel grade and carbide inserts. Can get very smooth surface on soft steels but not so shiny and thats what op was after.

Maybe the real solution would be polishing wax...

08-21-2017, 03:58 AM
Try one of these vertical shear bits. (http://www.gadgetbuilder.com/VerticalShearBit.html)

Sun God
08-21-2017, 04:39 AM
For a smooth mirror finish you want to 'burnish' the steel as you cut it, and the best way to do that is with lots of speed, a goodly chip, and a nice sharp carbide insert.

I have very good luck with the Tungaloy NS530 grade cermets in 4140HT (about HB300) at about 250SFM cut dry. My go-to ferrous turning insert is the NS530 CCMT060202. Stick within the DOC and feed guidelines on the packet, you want a nicely formed chip that comes off straw-blue and breaks by itself into nice tight little curls no longer than a half inch or so, no stringing. This very reliably gets me a smooth to the touch, iridescent finish.

08-21-2017, 09:51 AM
Thanks all. I can use HSS and I do get a smooth finish, but the point of this exercise it to get to the bottom of using carbide at low rpm. I intentionally left the sfm at about 50, as that's about as high as I can get on small diameter stock. Conventional wisdom says it's not possible, but it certainly is.

As for freecutting steel, I also did tests at 2500rpm and it does get smooth, but still, there is better. I come to think it's the nature of soft, leaded steel. So the answer must be a harder grade, as some of you suggested. I'll try the 4130.

1010 is the best finish so far amongst the steels that I can identify, but I agree - it needs a very sharp insert and a coated insert like KC730 is about ok, I think. My tests were limited as I ran out of bar.

08-21-2017, 10:50 AM
Try something even harder than 4130 prehard if the parts are small
I have turned some small pins from 1055 carbon steel that was hardened to HRC60 and surface finish is fancy even on 4mm pins turned at 1500rpm.

Bob La Londe
08-21-2017, 10:53 AM
Some metals just can't be mirror finished, but sometimes a vertical shear can get a better finish than anything else. Its really only good for .001 to .002 depth of cut, so it can can take a couple passes to remove the grooves from roughing passes. I hand ground one sometime back for mystery steel (usually some form of hot rolled) and it was better. I still had to hand dress with varying emery grades to get as good as it could get.

I just saw one of your metals is 12L14. That is a very machinable steel. Not sure why you would not be getting good results with that.

08-21-2017, 11:23 AM
Lots of very good advice already, I've used them all in the past and they all do give good results.

Yesterday while making up a couple of seal drivers I had the opportunity to try for a nice finish on some very gummy steel, black gas/water pipe.
I wasn't in a hurry and thought this would be a fun little experiment while cutting some black pipe down to the proper diameter just for the heck of it. Remember these are just a couple of quick seal drivers so I didn't spend a lot of time with them, better result can be had with more effort I'm sure.

I didn't take down the exact rpm but I remember it being in the appropriate range of sfm for the diameter and material and the HSS tooling I was using at the time. Depth of cut was .010" and the feed rate was .0025"/rev. I honed the 1/4" HSS to a very small radius and a very positive cutting geometry, similar to what you would find on carbide insert meant for aluminum.

Below is my first attempt on one seal driver with an as cut finish.
The second seal driver was cut using the same setup but finished using a tool I made up to burnish the cut. It consists of a small tang mounted roller bearing that I can insert into my tool post. I present this to the workpiece at a slight angle so as to concentrate the forces to a relatively small area. The steel in question being relatively soft will be burnished quite easily to a very smooth finish with minimal effort.
I've done other projects using this same technique previously and find it also enhances the metal's resistance to rusting because of the smoother surface finish.

This may or may not be a suitable option for you but it is another tool to use in achieving the finish you are after.



08-21-2017, 11:51 AM
I can't see your pictures Willy. All I see is this photobucket placeholder:

08-21-2017, 12:01 PM
Damn PB!
I can see them but then again I don't need to.:)

Have you tried the clicking on it twice trick?

I'll see what I can do on my end.

08-21-2017, 12:03 PM

08-21-2017, 12:05 PM

08-21-2017, 12:06 PM
Hope that worked.
If nothing else it'll increase my post count. LOL

08-21-2017, 12:17 PM
I've noticed that with my own carbide insert experiments that the feed rate plays a large part in the final finish. Next was that the best finish occurred when the chip load was lower where the chip coming off was quite fine. The cut could still be quite deep. Just that the chip was almost like foil instead of thick and chunky. And if running at a rather low SFM like you're doing it would require a very slow traverse. Instead of turning the hand wheel to move the carriage apply a given pressure on the wheel and let the carriage move as and when the chip at the cut allows by making room for the cutter to move. That was how I got the best finish on some test steels.

You also noted that you got your best finish on that screw driver steel which is still a fairly hard and springy grade of steel. But your other testing is all softer and leaded steel. That's pretty much an apples to oranges comparison. All that tells you is that the cutter that gave a nice finish on the tool steel is a good setup for harder steels. You still need to find the "magic insert" for use with softer steels.

08-21-2017, 01:06 PM
Hope that worked.
If nothing else it'll increase my post count. LOL
Thanks Willy, that worked.

Damn PB!
I can see them but then again I don't need to.
I'm guessing you can see the pictures either because you are logged in to photobucket or because there's a photobucket cookie in your browsers
cache that identifies your photobucket account.

Have you tried the clicking on it twice trick?
None of the tricks that are supposed to make the pictures visible have worked for me. Maybe it's because I don't have a photobucket account?

Anyway, that's a damn nice finish you got in the second picture. Well done.

08-21-2017, 01:10 PM
Thanks for the feedback Rich, was wondering for a while there if I was in my own little world.:o

Yes the burnishing did make a difference, probably should have spent more time with it but at the time I didn't know it was going to be for "show and tell".

08-21-2017, 01:13 PM
Carbide likes it like a bad girl. :) Don't dilly-dally--Hard and fast! I wouldn't even use carbide on 12L or most any hot rolled. But I only have a few carbide inserts/tools to begin with. They're over-rated for most home shops.

Paul Alciatore
08-21-2017, 02:53 PM
When I was doing some manual milling in my SB lathe I discovered a very helpful technique to improve the surface finish. Here, on the right in this photo you can see the original results that I got while milling the edge of this spacer plate.


I reasoned that the irregularities in the finish were due to a less than uniform motion as I cranked the cross feed for the cut. On the left you can see the far better finish that I obtained when I made an add-on crank handle that had a much larger radius. This allowed me to move the carriage in a much smoother manner and the finish was greatly improved. I was astonished.

Here you can see the crude, wood handle that I made to fit over the three ball handle on the SB.


I found that the greater radius that it provides for the crank handle makes it a lot easier to move the feed uniformly. Here it is in use.


I think this experience points to a somewhat unappreciated advantage of CNC; uniform motion is necessary for a good finish. I find that I get a better finish when I am turning a diameter in the lathe if I use power feed. Again, uniform motion. What works for milling also works for turning as well.

Makes me wonder if the handles on all the manual machinery are really too small.

08-21-2017, 03:04 PM
That's a VERY good point Paul. Moving the handles smoothly is very much a skill that needs developing and constant practice. And there's no doubt that it's a lot harder on smaller wheels. And even worse on the 3 ball style or other "cross" style handles. Nice job on the temp add-on and a great proof of the concept!

08-21-2017, 03:34 PM
Bit along same lines as Paul I have been using cordless drill on lathe compound slide.
Works great if you have cordless drill with enough slow crawl speed.

08-21-2017, 04:29 PM
Paul, that's cool.. I've calculated it all out for proper feed and then 1 Mississippi, 2 Mississippi'ed my way around the dial...

As for the Carbide vs HSS... The material has absolutely NO CLUE what the cutter is made out of, nor does it care.
The cutter could be made out of a hardened crusty 40 year old marshmallow, the material doesn't care..

Carbide can take more heat than HSS.. A LOT more heat, so, YES, you can run it much faster without destroying anything..
But carbide does NOT have to be run fast.. IF the geometry is proper.

Where a lot of people run into problems (myself included way back when), is assuming a CNMG is a CNMG. Its not... There is a dizzying array of
substrates, edge preps and chip breakers that make picking the proper insert for what you want to do a really crappy and darn near impossible

I don't pick my inserts anymore. I pick up the phone or fire off an e-mail.. "send me something that does XXXX". As much as I would like
to know which insert to use, I don't have the time to devote to it, and at this point, I really don't care as long as it works...

What I suggest for a lot of manual machining where spindle speed and surface speed is generally low, as is HP.. Ground and polished upsharp inserts
for aluminum. I run them all the time in small stuff when I can't get the surface speed, and they are also great on turning long things to keep the
tool pressure down, and this is on a 12k lb CNC with 25hp(rated at 30 minutes, so about 400 HAASpower). Its like a razor sharp piece of HSS (that you don't have to grind) with
a much wider window of speeds that it can be used. And they are sharp!!!! I have the scars to prove that.

Its all about the GEOMETRY.

08-22-2017, 12:03 AM
Hello, Mr. Power Crossfeed, it's nice to meet you!