Ball-Nose Chatter and a Parting-Line Step That Reverses
An H13 mould insert, 48 HRC, is being finished on a 3-axis VMC. The cavity is 80 x 80 x 25 mm deep with 3 mm corner radii. The finishing tool is a Ø6 mm two-flute ball-nose carbide, 45 mm overhang in an ER32 collet, running 18,000 rpm, 1,200 mm/min, 0.25 mm stepover, leaving 0.30 mm radial stock. Runout measured at the ball is 0.008 mm; the spindle is rated to 20,000 rpm. After finishing, the cavity floor shows evenly spaced chatter marks parallel to the toolpath. When the insert is clamped to its opposite half, a 0.06 mm step appears at the parting line on one side only; the step reverses when the halves are rotated 180°. Both halves were programmed from the same CAD model but machined on different days.
1. Diagnose the chatter and specify the changes you would make to the tooling and cutting parameters.
Read the worked answer
Chatter with a Ø6 ball nose at 45 mm overhang is a stiffness problem first: length-to-diameter is 7.5:1, and an ER32 collet at that overhang is compliant. 18,000 rpm on a small ball nose also places the tooth-passing frequency where it can excite the tool. The 0.008 mm runout is acceptable but not ideal for a ball nose.
Changes:
- Reduce overhang to 25-30 mm or move to a shrink-fit or hydraulic holder; this alone often removes the marks.
- Drop spindle speed to 12,000-14,000 rpm and set feed to hold a chip load of about 0.01-0.02 mm/tooth, or increase feed per tooth to get below the stability lobe.
- Reduce finish stepover to 0.15-0.20 mm while keeping the tool engaged; do not let it rub.
- Use a variable-helix ball nose if one is available.
- Confirm air blast or coolant clears chips from the pocket and check the ball for wear.
Verify by cutting a test patch and inspecting under light before committing the whole cavity.
2. Explain what the reversing 0.06 mm step tells you about the cause, and list the checks you would run.
Read the worked answer
A step that reverses when the halves are rotated 180° is not simple tool deflection; deflection or tool wear would produce a step that stays on the same physical side. A reversing step points to a datum or setup error common to the two halves, or to a rotational mismatch between their origins.
Checks:
- Indicate the same datum features on both halves and compare the recorded part origins.
- Verify the probe with a ring gauge and check tool-length offset history for both days.
- Confirm both halves were set from the same corner and the same probe cycle, not edge-finder on one and probe on the other.
- Check parting-line flatness and the clamp sequence; clamping can move an insert if the faces are not flat.
- Measure the step at several points to see whether it is a pure offset or a rotation.
Rotation or origin error explains reversal; clamping error usually does not reverse cleanly.
3. Describe the correction and how you would verify the parting line before tryout.
Read the worked answer
Correction:
- Re-establish a single setup datum for both halves using the same probe cycle, the same calibration and a machine warm-up cycle. Ideally machine both halves in one program run.
- Check parting-line flatness and specify a clamp torque and clamp order.
- Take a light finishing pass on both parting faces with the same tool and the same parameters, or re-machine only the stepped side if the step is confirmed as an offset.
- Rework the finishing strategy from the chatter findings at the same time so the faces are cut without vibration.
Verification before tryout:
- CMM or blue check of the parting faces; target step ≤0.01-0.02 mm.
- Clamp at low pressure and re-check that the step does not appear or reverse.
- Confirm the cavity depth and corner radii against the model in the same setup, so a datum fix does not silently move other features.