A micro drill does not wear out quietly. It snaps. The reason is arithmetic: strength scales with the square of diameter, and cutting force does not.
A 0.3 mm drill has a cross-section of 0.07 mm². A modest 50 N of thrust puts 700 MPa into it before any bending is added. Once bending starts, the tool lives on the difference between carbide strength and the stress a small misalignment creates.
Read the fracture before changing anything
The break face tells you which load killed the tool.
| Appearance | Load | Usual cause |
|---|---|---|
| 45° shear lip, flutes twisted | Torsion | Chip packing, feed too high, hole bottom |
| Transverse break, flutes intact | Bending | Runout, wander, holder, overhang |
| Corner chipping, no twist | Impact | Interrupted entry, cross-hole, scale |
| Wear land, then crumbling | Heat and abrasion | Speed too high, coolant starved |
| Crack from the margin, slow growth | Fatigue | Vibration, unstable holder |
A torsional break means torque exceeded the section. A bending break means the tool was pushed sideways. They have different fixes, and guessing wastes tools.
Cause one: runout
Below 1 mm, total indicated runout should stay under 0.005 mm. Above that the drill rotates off-centre, one flute does most of the cutting, and the side force bends a tool with almost no bending stiffness.
Check in this order: collet condition, collet nut torque, spindle taper, shank cleanliness. A 10 µm chip on the shank is 3% of a 0.3 mm diameter, and that is enough.
Cause two: feed that is too low
This one is counter-intuitive. Operators reduce feed when a drill breaks, and a drill that rubs is more likely to break than a drill that cuts.
Below a threshold feed the edge cannot penetrate; it burnishes. Thrust rises, the material work-hardens, and the next revolution is harder still. The drill breaks and the operator reduces the feed again.
For micro diameters, keep feed per revolution between 0.5% and 2% of diameter. At 0.3 mm that is 0.0015-0.006 mm/rev.
Cause three: chip evacuation
A 0.5 mm hole at 10 x D holds about 0.002 cm³ of material. That does not sound like much, but the chips must leave through two flutes with almost no clearance.
- Peck at 1 x D in steel and stainless, 2-3 x D in aluminium.
- Use through-coolant above 5 x D when the series supports it.
- Retract at rapid, return at rapid, and keep cutting feed inside the hole.
- Watch for chips dragged back into the flutes on the return.
Cause four: entry
A drill entering a curved, inclined or interrupted surface is in single-point contact. The side force is large and the tool is at its least supported.
Spot with a 90° or 118° spot drill, or use a flat-bottom drill for the first 1 x D. On cross-holes, reduce feed to about 30% through the intersection.
Cause five: overhang
Stiffness falls with the cube of overhang. Doubling the stick-out makes the tool eight times more flexible. Use the shortest drill that reaches the depth, and prefer an extended-length series over pulling a standard drill out of the collet.
A short diagnostic sequence
- Measure runout at the tool tip. Above 0.005 mm, stop and fix it.
- Match the fracture face to the table above.
- Torsional break: halve the peck depth, verify feed per revolution.
- Bending break: fix runout before touching parameters.
- Discoloration: reduce speed by 20% and check coolant delivery.
FAQ
Can a harder coating stop breakage?
No. Breakage is a mechanical overload. A coating changes friction and hot hardness; it does not change the section modulus.
Why does the drill break at the same depth every time?
Chip packing. The flutes fill at a predictable depth, torque spikes and the tool twists off. Shorten the peck.
Is a left-hand drill stronger?
No. Left-hand geometry exists for machines that run in reverse and for pull-drilling.
Do I need a torque-adaptive control?
For production at 0.2 mm and below it helps, but runout and peck depth fix more breakages for less money.