Metal cutting tools for milling, drilling, turning and threading EN/IT/CZ/PL/UA
Get a tool recommendation

MILLING

Solid carbide end millsIndexable milling cutters

DRILLING AND HOLES

Solid carbide drillsIndexable drillsCarbide reamers

TURNING

Turning toolholdersTurning insertsGrooving and parting

THREADING AND SWISS-TYPE

Threading toolholders and insertsTooling for Swiss-type lathes

Technical information

Cutting edge damageTurning: troubleshootingMilling: troubleshootingDrilling: troubleshootingThreading: infeed methods and passes

About the brand

About the brandQuality and inspection

More

Downloads Contacts Get a tool recommendation
Home/Technical information/Turning: troubleshooting

Technical information

Turning: diagnosing and fixing tool damage

This guide is built around what you see at the machine: first the marks on the part, then the state of the cutting edge, and finally the behaviour of the chip.

17cases worked through
3groups of signs: part, insert, chip
19lines of action in the table
5steps of diagnosis

Course of action

The order within each card is not accidental. It goes from the cheapest fix to the most expensive: cutting data first, then chipbreaker geometry, then grade, and only at the end the rigidity of the setup and the machine. Doing it the other way round — changing the grade without checking the cutting speed — is the most common mistake: the new insert fails in exactly the same way and the cause stays unknown.

  1. Look at the insert under magnification. An even wear band, a crater, welded metal and a fracture are four different diagnoses and four different actions.
  2. Work out what actually happened. The same defect on the part can have several different causes, and the cures are opposite.
  3. Change one thing. Two changes at once make the result impossible to read.
  4. Work from cheap to expensive: cutting data → coolant → geometry and nose radius → grade → overhang and clamping → machine.
  5. Write down what you did. The batch will come round again in a month; your memory will not.

Notation: Vc — cutting speed, f — feed, ap — depth of cut, rε — nose radius, κr — lead angle.

What you see on the part 6 cases

These are the cases an inspector or operator notices without necessarily looking at the tool. That is why the first step in each of them is to go back to the edge: the part shows the consequence, the cause is almost always on the insert or in the setup.

The size wanders from part to part

What it looks like

The size does not drift evenly in one direction but jumps both ways. Adjusting does not help — after a correction the scatter is exactly the same.

Why it happens
  • Inserts differ in actual size: tolerance class M has a wider tolerance on nose position than G.
  • The tool and the part deflecting under load.
  • A built-up edge that periodically tears away.
  • The insert does not seat consistently in the pocket: chips under the shim, a worn pocket, an undertightened screw.
What to do
  1. Blow out and inspect the pocket before every index of the insert.
  2. Check for a built-up edge — the cure for that is raising Vc.
  3. Shorten the toolholder overhang, check how the part is clamped.
  4. Reduce ap and f — less force, less deflection.
  5. Move to a sharper positive geometry, reduce rε, increase κr.
  6. Move to tolerance class G inserts instead of M where the size is critical.
What not to do. Do not correct the tool offset in the control for every part. You are compensating for the consequence; the scatter stays.

The size drifts one way through the batch

What it looks like

The first parts are in tolerance, then the size creeps slowly and evenly in one direction. After N parts a correction is needed, and after another N, again.

Why it happens

This is not a fault but flank wear on the edge: the nose recedes, and the size goes with it. The only question is whether it is happening too fast.

What to do
  1. Measure how many parts you get before it goes out of tolerance and set the change by that number.
  2. Reduce Vc — speed has the strongest effect on the rate of wear.
  3. Check that the coolant is actually reaching the cutting zone.
  4. Increase rε and κr.
  5. Move to a harder grade.
What not to do. Do not “stretch” an insert to the end of the shift: beyond a certain point wear grows in a jump, and the size jumps with it.

Poor surface finish

What it looks like

The surface is dull, torn or scratched instead of showing an even feed pattern.

Why it happens
  • A worn edge — the tool is no longer cutting but pushing.
  • Built-up edge: welded metal tears away and leaves torn patches.
  • The chip is being dragged between the edge and the machined surface.
  • The feed is too high for the nose radius.
  • Vibration — in that case the pattern is regular, in waves.
What to do
  1. Change the insert and look at the result.
  2. Reduce f or increase rε.
  3. Raise Vc — a low speed on steel and stainless produces a built-up edge.
  4. Move to a finishing chipbreaker with a polished rake face.
  5. Deal with the chip: a different chipbreaker, a directed coolant jet.
  6. Check the overhang and the rigidity.
What not to do. Do not reduce feed and speed at the same time: as far as surface finish is concerned they work in opposite directions.

Vibration

What it looks like

A whistle or a rattle, a regular wavy pattern on the surface, fine chipping along the whole contact length of the edge.

Why it happens
  • The overhang of the toolholder or boring bar is too long.
  • The part is weakly clamped, or it is thin-walled or long with no support.
  • The feed is too low for a large radius: the edge is rubbing instead of cutting.
  • rε too large or κr too small — the radial component of the force grows.
What to do
  1. Shorten the overhang. This works better than any change of cutting data.
  2. Check how the part is clamped, add a steady rest or support it with the tailstock.
  3. Increase the feed. A thicker chip stabilises the cut.
  4. Reduce ap and Vc.
  5. Reduce rε, increase κr, move to a sharper geometry.
  6. For deep boring — a bar with a carbide body or a damped bar.
What not to do. Do not try to kill vibration by reducing the feed alone: if the cause is rubbing, it will get worse.

Burrs on the edges of the part

What it looks like

Where the tool leaves the cut, a bent-over ridge of metal is left behind. Consistently so on ductile materials.

Why it happens

At the end of the cut a layer is left in front of the edge with nothing behind it: the metal is not cut off but bent aside.

What to do
  1. Change the worn insert: a sharp edge leaves a smaller burr.
  2. Increase the feed.
  3. Reduce ap on the last pass.
  4. Break the corner beforehand, or lead the tool out at an angle.
  5. Move to a sharper geometry with less honing, increase κr.

Chipping of the part edges

What it looks like

As the tool leaves the cut, part of the edge of the workpiece breaks away. Typical of cast iron and hardened steel.

Why it happens

The same cause as with a burr, with the opposite sign: the material does not bend, it fractures.

What to do
  1. Reduce f and ap as the tool leaves the cut.
  2. Break the corner before the main pass.
  3. Change the toolpath: exit at an angle.
  4. Increase κr, move to a sharper geometry.
  5. Check the rigidity of the workpiece clamping.
What not to do. Do not confuse this with insert chipping: here it is the part that is breaking, the tool edge is intact, and a stronger grade will change nothing.

What you see on the insert 9 cases

The first two are normal wear, and the only question is how fast. The other seven should not be happening at all if the conditions are chosen correctly.

Flank wear

What it looks like

An even bright band along the cutting edge on the flank side.

Why it happens

The working form of wear — the abrasive action of the material. It is accelerated by too high a Vc, too low a grade hardness and running without coolant.

What to do
  1. Reduce Vc.
  2. Check the coolant supply: jet direction, pressure, concentration.
  3. Increase rε and κr.
  4. Strengthen the edge with honing or a chamfer.
  5. Move to a harder grade.

Crater on the rake face

What it looks like

A hollow on the rake face a little back from the edge. The edge is intact at first; when the crater reaches it, the edge breaks without warning.

Why it happens

The temperature where the chip contacts the rake face. The main factor is cutting speed.

What to do
  1. Reduce Vc.
  2. Reduce f and ap.
  3. Copious coolant, preferably under pressure into the contact zone.
  4. A chipbreaker with a larger rake angle.
  5. A grade with a thicker thermal barrier coating (CVD).
What not to do. Do not judge the edge from the side alone: a crater is visible only from above, at an angle.

Notch at the depth-of-cut line

What it looks like

A local notch exactly where the edge leaves the material. The rest of the edge is in good condition.

Why it happens

The work-hardened layer left by the previous pass, scale on the workpiece and a concentration of stress at one point. Typical of stainless and heat-resistant alloys.

What to do
  1. Vary ap between passes so that the depth-of-cut line does not stay at one point on the edge.
  2. Reduce κr.
  3. Reduce Vc.
  4. Strengthen the edge, move to a stronger grade.
  5. For workpieces with scale — make the first pass deeper than the thickness of the crust.
What not to do. Do not raise the speed to “get through the scale faster”: the notch grows as a result.

Chipping of the cutting edge

What it looks like

Small breakouts along the edge. Appears suddenly, often on the first part after an insert change.

Why it happens
  • Interrupted cutting, impact on entry.
  • The geometry is too sharp for these conditions.
  • Vibration or a non-rigid setup.
  • A built-up edge tearing away.
  • Running with a very thin chip.
What to do
  1. Reduce the feed.
  2. Check overhang and clamping, get rid of the vibration.
  3. A geometry with a reinforced edge, a larger rε.
  4. A stronger grade: for interrupted cutting toughness matters more than hardness.
  5. A smooth entry into the cut instead of an impact against the face.
  6. If there is a built-up edge — treat the built-up edge, not the wear.
What not to do. Do not put a finishing positive geometry into an interrupted cut just for the surface: it will break up before it ever holds a size.

Cracks in the insert

What it looks like

Isolated cracks along or across the edge, without a thermal comb. They often precede the insert breaking altogether.

Why it happens

Mechanical overload: too much feed and depth, an impact on entry, play in the clamping, a worn pocket or shim.

What to do
  1. Reduce f, then ap.
  2. Check the pocket, the shim and the clamping screw.
  3. Shorten the overhang, clamp the part more securely.
  4. A thicker or larger insert.
  5. A tougher grade, a reinforced edge.
  6. Change the entry path.
What not to do. Do not index the insert to the next edge if there is a crack in the body.

Thermal cracks

What it looks like

A row of parallel cracks across the edge, like a comb, at even spacing. Fragments then break out between them.

Why it happens

Not the temperature itself but the swings in it: interrupted cutting, leaving the cut, and most often an intermittent coolant supply.

What to do
  1. Decide on the coolant: either copious and uninterrupted, or dry cutting. There is no option in between.
  2. Reduce Vc, then f.
  3. A grade with better resistance to thermal shock.
  4. Strengthen the edge, increase rε.
  5. Reduce the number of entries and exits in the program.
What not to do. Do not add “a bit” of coolant in an interrupted cut: a weak jet gives a continuous thermal cycle — the worst possible arrangement.

Plastic deformation of the nose

What it looks like

The nose is neither fractured nor worn but has “flowed” — sagged and spread.

Why it happens

A temperature higher than the grade can hold: Vc, feed and depth too high in combination.

What to do
  1. Reduce Vc, then the feed, then the depth — in exactly that order.
  2. A steady coolant supply.
  3. Increase rε and κr, strengthen the edge.
  4. A grade with better heat resistance and a thicker coating.
What not to do. Do not confuse this with chipping: here the edge is intact, and the cure is in the cutting data, not in a stronger grade.

Built-up edge on the cutting edge

What it looks like

Workpiece material welds itself to the edge, the surface becomes torn and the size jumps. The built-up edge periodically tears away, taking part of the edge with it.

Why it happens

Too low a cutting speed plus a material prone to adhesion: low-carbon steels, stainless, aluminium, copper.

What to do
  1. Raise Vc — the main action.
  2. Increase the feed.
  3. A positive geometry with a sharper rake angle and a polished face.
  4. Check the coolant.
  5. A grade with better resistance to adhesion.
What not to do. Do not reduce the speed “to make it gentler” — it will get worse.

Overheating of the cutting zone

What it looks like

The chip is blue or purple, the part is hot, and the size after cooling does not match what was measured at the machine.

Why it happens

More heat is generated than the chip and the coolant can carry away. It then turns into cratering, deformation or thermal cracks.

What to do
  1. Reduce Vc, then f and ap.
  2. Sort out the coolant: pressure, direction, concentration.
  3. A sharper positive geometry — less heat for the same metal removal.
  4. Reduce the edge honing.
  5. A grade with better heat resistance.
What not to do. Do not measure a hot part and do not correct the program from that measurement.

How the chip behaves 2 cases

The chip is the only process parameter you can see without measuring. What you want is a short spiral or fragments that leave the cutting zone on their own.

Long tangled chips

What it looks like

A continuous ribbon that wraps around the part, the boring bar or the turret.

Why it happens

The chipbreaker is not working: feed and depth have fallen outside its range. Most often the feed is too low.

What to do
  1. Increase the feed.
  2. Change ap, redistribute the stock between passes.
  3. Choose a chipbreaker to match the cutting data you actually run — this is the main solution.
  4. On soft and low-carbon steels with X-type geometries, raise Vc.
  5. Reduce the rake angle, change κr.
  6. Aim the coolant jet to carry the chip away.
What not to do. Do not try to break the chip by feed alone if the geometry was chosen for a different range.

Chip throw

What it looks like

Small hard fragments fly about, clog the slideways and leave scratches on the machined surface.

Why it happens

The chipbreaker is breaking the chip too aggressively for these conditions.

What to do
  1. Reduce feed and cutting speed.
  2. A chipbreaker with gentler chip breaking.
  3. Increase rε.
  4. Aim the coolant so that the jet leads the chips into the pan.
What not to do. Do not accept fine chip throw as normal on finishing passes.

Full table of actions

● — act in this direction · ↑ — increase · ↓ — decrease. An empty cell means this parameter has practically no effect on this case. The order of the actions is given in the card for the case concerned.

What happenedChoice of gradeCutting dataTool geometrySetupMachine
hardnessstrengthheat resistanceresistance to adhesionVcfaptoolpathcoolantchipbreakerrake anglerεκredge strengthclass M→Gtoolholder rigidityworkpiece clampingoverhang
The size wanders from part to part●●●●●●●●●●●
The size drifts one way through the batch●●●●●●
Poor surface finish●●●●●●●●●●●●●●●
Vibration●● *1●●●●●●●●●●●
Burrs on the edges of the part●● *2●●●●●●
Chipping of the part edges●●●●●●●●●●●
Flank wear●●●●●●●
Crater on the rake face●●●●●●●●
Notch at the depth-of-cut line●●●●●●●
Chipping of the cutting edge●●●●●●●●●●●●●●
Cracks in the insert●●●●●●●●●●●●
Thermal cracks●●●●●●●●●
Plastic deformation of the nose●●●●●●●●●
Built-up edge on the cutting edge●●●●●●●
Overheating of the cutting zone●●●●●●●
Long tangled chips● *3●●●●●●
Chip throw●●●●●●

Three rules that cover most cases

Look first, change afterwards

The state of the edge under magnification answers the question faster than working through grades. Fracture and deformation look alike but are cured in opposite ways.

One change at a time

Two changes at once make the result impossible to read: you will not know what worked, and you will not be able to repeat it.

From cheap to expensive

Cutting data, coolant, geometry, grade, rigidity, machine. Changing the grade as the first step is the most expensive way of not finding the cause.

Recommended cutting data

There are deliberately no numerical values on this page: the Vc, f and ap ranges by workpiece material group are collected in a separate block, which goes below.

Related sections