The tool is very worn: how can I repair it quickly and easily?

• 1
  Key takeaways
• 2
  Decide whether the tool is worth repairing at all
• 3
  Read the wear pattern before you remove metal
• 4
  Regrind the worn edge without losing the geometry
• 5
  Coating, runout and the offset reset
• 6
  Why the same tool keeps wearing out early
• 7
  Repair a worn tool in 4 steps
• 8
  Repair or replace: quick judgment table

Before you touch a grinding wheel, look at what failed. A cutter with uniform flank wear along every flute is a repair candidate. A cutter with one chipped corner, a cracked shank or a bent neck is usually scrap. The money you spend regrinding it is money you will spend again when it breaks mid-cycle.

The material matters more than the price tag. Solid carbide end mills in the Ø6–20 mm range are the classic repair case. So are HSS and cobalt drills, taps, and reamers. Brazed tools with a carbide tip can be re-tipped once or twice. Indexed inserts are the opposite: you rotate the corner, then throw it away.

Size sets the limit. Below Ø3 mm there is not enough body left to remove a worn layer and keep any rigidity. Above Ø25 mm the grinding cost starts to close in on a new tool, so repair only makes sense if the geometry is custom.

One more check: was the tool ever correct? If first-article dimensions drifted from the start, regrinding brings back the same wrong geometry. Fix the program or the holder first.

A quick test. Put the tool under a loupe. If you see a shiny wear land of even width with no cracks, repair it. If the edge looks ragged, glazed or discolored, the heat damage goes deeper than the flank and grinding will not save it.

• 1
  Repair: solid carbide end mills, HSS drills, taps, reamers, brazed tipsWear is even and the body is sound.
• 2
  Replace: inserted tips, micro-tools under Ø3 mm, cracked or bent shanksRepair cost or risk is too high.

Wear tells you what the cutting edge was doing. Flank wear on the clearance face is normal and grows at a predictable rate. Crater wear on the rake face means the chip was sliding too hot. Built-up edge means the material was sticking, often in aluminium or soft stainless.

Chipping is different. Small chips at the corner come from interrupted cuts, hard spots in the casting, or a feed rate pushed too far. A single large chip usually means the tool hit something it should not have: a clamp, a fixture, or a hard inclusion in the stock.

Measure flank wear with a toolmaker's microscope or a 10× loupe and a scale. For carbide in steel, a wear land of 0.2 mm is a working tool. At 0.3 mm the cutting forces rise fast and the surface finish drops. At 0.5 mm the edge begins to break down and sparks appear.

Check the shank and the neck too. A shank that has spun in the holder leaves a bright ring. That means the holder lost grip and the tool was running off-center. Repairing the flutes will not fix that.

Write the numbers down. Diameter, wear land width, corner radius, flute count. Six months of these records tell you whether your cutting data is reasonable or whether you are burning tools on purpose.

Regrinding removes the worn layer and produces a fresh edge. The amount to remove is roughly twice the wear land width. A 0.3 mm land means 0.6 mm off the diameter, so a Ø10 mm end mill comes back at Ø9.4 mm. This is normal, and you must record the new size.

The primary clearance angle on a carbide end mill usually sits between 6° and 10°. Keep the original angle unless you are changing the application. Increasing it makes the edge weaker; decreasing it makes the tool rub. Both change the cutting forces and the finish.

Heat is the enemy. Carbide loses hardness above roughly 800 °C, and a dry grind can reach that at the edge in seconds. Use a diamond wheel, flood coolant, and light passes. If the edge turns blue, you have softened it and the tool will fail early.

For coated tools, grinding exposes bare carbide. That is why a reground tool either gets recoated or gets used in a softer application. A TiAlN-coated tool reground and used dry in stainless will not last like the original.

Radius the corners when you can. A small corner radius, 0.2–0.4 mm, removes the most fragile point and adds life on profiling passes. It also changes the part geometry, so update the program if the corner is critical.

• 1
  Remove twice the wear land0.3 mm land means 0.6 mm off the diameter.
• 2
  Hold the original clearance angle6°–10° for most carbide end mills.
• 3
  Cool the grindBlue edges mean lost hardness.

A recoated tool recovers most of its life. PVD coatings such as TiAlN, AlTiN and AlCrN are applied at temperatures around 450–500 °C, low enough that the carbide substrate keeps its strength. Send the tool out for coating together with its new diameter and geometry data.

If recoating is not practical, use the reground tool for roughing, where surface finish and edge sharpness matter less. Keep the coated tools for finishing passes. This split gets useful life out of both groups.

Runout is the step most people skip. After repair, indicate the tool in the holder. Total indicated runout at the cutting edge should stay under 0.01 mm. Above 0.02 mm one flute does most of the cutting, and it wears out fast.

Then reset the length offset. A reground end mill is shorter and smaller. If the machine still thinks it is Ø10 mm, the first part comes out undersized and the finish suffers. Touch off, update the offset, and cut one test feature before releasing the job.

Check the holder as well. Dirty taper seats, worn collets and over-torqued nuts all push runout up. A clean holder is part of the repair, not a separate task.

If you repair a tool and it fails again in the same spot, repair is not the answer. Look at the cutting data first. Surface speed above the supplier range overheats the edge. Feed per tooth below 0.05 mm for a carbide end mill rubs instead of cutting, which wears the flank faster than a proper chip load.

Coolant delivery is the second suspect. Flood coolant that never reaches the cutting zone does nothing. Through-spindle or through-tool coolant puts fluid where the heat is. In deep pockets, air blast plus minimal lubrication often beats a weak flood.

Runout is the third. A holder with 0.03 mm TIR loads one flute and lets the others idle. That flute chips first, and the operator blames the tool. Check the holder, the collet and the nut torque before blaming the cutter.

Then check the material. Castings with hard skin, flame-cut plate, and work-hardened stainless surfaces all punish the first pass. Take a lighter first pass below the skin, or use an insert tool for that cut and keep your regrindable tools for the clean metal underneath.

Keep a simple log. Tool number, material, speed, feed, hours in cut, and the wear land at removal. After a few months the pattern is obvious. Most early failures trace back to one number that drifted.

• 1
  Feed too lowUnder 0.05 mm per tooth rubs the edge instead of cutting.
• 2
  Weak coolantFluid that misses the cutting zone does not cool anything.
• 3
  High runoutOne flute does the work and chips first.