A cryogenic pump at the ironworks broke down. Facing a wait of at least five weeks for an external replacement, the workshop decided to manufacture it in-house.
Material: 316L | Length: 3237mm | Diameter: 260mm (at the narrowest point) | Outer Diameter Tolerance: ±0.02mm
The job was assigned to the workshop's most experienced machinist.
The result? After finish-turning, the measured outer diameter was out of tolerance-it was too large.
The required tolerance was ±0.02mm, but the actual deviation was 0.06–0.10mm.
The workpiece was ruined.
It wasn't a lack of skill on the machinist's part. He had over twenty years of experience operating lathes, mostly machining ordinary steel parts, and had accumulated plenty of know-how.
The problem came down to one thing-
He treated the 316L as if it were ordinary steel.
How does "tool deflection" happen?
Was the tool dull? Were the parameters too aggressive? Was the machine tool lacking rigidity?
No.
There was one root cause: work hardening.
During the cutting process, the surface of 316L stainless steel undergoes cold-work hardening in a very short time.
The depth of the hardened layer is typically between 0.1mm and 0.3mm (this is data measured during cutting, not an estimate).
Once hardened, the strength of this surface layer can exceed 1500 MPa-making it harder than many tool materials.
When the finish-turning pass began, the tool couldn't cut into this hard shell; instead, it deflected (pushed away). Consequently, the workpiece ended up oversized.
That is how tool deflection occurs.
It wasn't a problem with the tool itself; the material simply hardened during the cutting process.
So, why is 316L stainless steel harder to work with than ordinary steel?
316L contains molybdenum, which increases its toughness and leads to more intense plastic deformation during cutting-causing the hardened layer to form faster, thicker, and harder.
Using the same parameters and the same tool, ordinary steel is easy to machine, whereas stainless steel causes problems.
What did the machinist do wrong?
He made an error based on past experience:
He assumed that using lower parameters would be safer.
He directly applied cutting parameters meant for ordinary steel to the 316L.
Where exactly did he go wrong? The finishing allowance was set at 0.2 mm-too small-resulting in an insufficient depth of cut. It failed to penetrate the 0.1–0.3 mm hardened surface layer.
The tool rubbed against the hardened surface, generating heat that exacerbated the work-hardening effect.
As hardening intensified, tool deflection increased; the more it deflected, the harder it became to cut into the material.
There was a second error: the feed rate (F) was set to 0.05 mm/rev-far too low.
A lower feed rate prolongs the contact time between the tool and the workpiece, accelerating heat buildup and thickening the hardened layer, which in turn worsens tool deflection.
Both parameters were incorrect.
Settings suitable for ordinary steel simply don't work for 316L stainless steel.
Ultimately, only one change was made:
Swapping the tool-not just the brand.
I switched to an M-class insert specifically designed for stainless steel, featuring a 10°–15° rake angle and a 0.5–0.8 mm chamfered edge (T-land).
Most M-class inserts on the market-such as those from Sandvik or Iscar-come with this chamfered edge; it was a straightforward choice.
That single adjustment did the trick.
Why did it work?
A larger rake angle aligns the cutting force more favorably, making it easier to evacuate chips from the cutting zone. Since chips carry heat away, less heat remains, resulting in a thinner hardened layer. A thinner hardened layer allows the next pass to cut effectively, eliminating the tool deflection issue.
The chamfered edge serves to reinforce the cutting edge. A large rake angle thins the edge, making it prone to chipping; the chamfer prevents this.
One feature addresses work-hardening, while the other protects the cutting edge. Consequently, machining efficiency actually tripled.
In conjunction with the tool change, three additional measures were taken:
① For the finishing pass, the depth of cut was set to exceed 0.3 mm to cut through the hardened layer in a single pass, rather than splitting it into two or three passes. This also improved efficiency.
② The feed rate was increased to 0.1–0.2 mm/rev; the feed rate for 316L needs to be higher than you might expect. If the feed is too low, the tool dwells on the material surface, causing rapid heat buildup and severe work-hardening. It is better to accept slightly lower surface roughness than to let the tool rub against the hardened layer. ③ Increase the concentration and flow rate of the cutting fluid, directing the jet precisely into the cutting zone. The thermal conductivity of 316L is only half to a third that of carbon steel, causing heat to accumulate entirely at the tool tip and the workpiece surface. Cutting fluid isn't just for show-it shouldn't merely be poured over the workpiece; it must be forced into the cutting zone.
After addressing these four points, proceed with finish turning. Keep the outer diameter deviation within 0.03mm. The shaft is successfully produced.
To be honest:
Even after ten or twenty years of working with stainless steel, the issue of tool deflection can be tricky-the parameters and experience gained from standard steel often need to be completely reversed when switching to 316L.
Different materials have different work-hardening rates, requiring different machining parameters.
Experience is valuable, but it must be adapted to the specific material and machining conditions-everyone should keep this in mind!
Four key points to resolve 316L tool deflection:
① Switch to M-class inserts designed for stainless steel (featuring a large rake angle and chamfered edge) to ensure chips carry away the heat. ② Set the finish turning depth of cut to over 0.3mm to cut through the work-hardened layer in a single pass. ③ Increase the feed rate to above 0.1mm/rev to avoid rubbing against the work-hardened layer. ④ Increase cutting fluid concentration and flow rate, ensuring it is directed into the cutting zone.
Get the parameters right, and the scrap rate will naturally drop.
It took me less than half an hour to diagnose the problem with this shaft. Meanwhile, the shop floor machinist had already wasted a piece of 316L stock.
That's the reality of machining problems: when experience doesn't provide the answer, you have to figure it out yourself-and the cost is paid in scrap parts and time.





