Fanuc spindle alarm 46 means the one-rotation signal of the position coder could not be detected normally during thread cutting — the once-per-revolution Z pulse that synchronizes the Z-axis feed to spindle rotation misbehaved exactly when it matters most: mid-thread. Fanuc's manual sends you to the alarm 41 procedure — the causes are the same (parameters, coder, cable shielding, cable routing, amplifier). What AL-46 adds is context: the signal is good enough to pass the power-on and idle checks, but fails under the specific conditions of threading — and that narrows the suspects considerably.

Why threading exposes marginal signals

During thread cutting the control re-synchronizes to the Z pulse every revolution, at cutting speed, with the axes under load — servo drives switching hard, coolant flowing, the whole machine electrically noisy. A Z signal that is marginal (weak, noisy, intermittent) survives idle rotation but drops or doubles exactly here. So AL-46 usually means: signal-quality problem, load-dependent — which reorders the alarm-41 checklist:

  1. Shielding and grounding first: shield continuity end to end, frame-ground via the cable clamp per the B-65282EN connection diagram. A floating shield is the #1 cause of ‘threads fine sometimes’ machines.
  2. Routing: is the coder cable bundled with servo or spindle power leads? Threading is precisely when those leads carry hard-switched current. Separate them — this fixes a remarkable share of AL-46 cases.
  3. The cable itself: oil-degraded or crushed sections — wiggle-test, and check whether the alarm correlates with specific axis positions (a damaged section in the drag chain flexes at particular Z positions).
  4. The coder: a contaminated disc or slipping drive belt gives a Z pulse that jitters under vibration — observe it per the AL-41 procedure at speed, not just hand-rotation.
  5. Parameters and amplifier last — as in the AL-41 guide.
Tip: Before the alarm ever appears, marginal Z signals scrap parts: threads with pitch errors, multi-pass threads that don't track the previous pass (‘crossed threads’), taps that break on the retract. If operators report any of these on a machine that later shows AL-46 — it was the same fault all along. Treat thread-quality complaints as this alarm's early warning.

Diagnosis checklist (in order)

  1. Reproduce: does it trip on a specific thread cycle, speed, or Z region?
  2. Verify shield continuity and the frame-ground clamp.
  3. Walk the cable route — separate from power leads; inspect drag-chain sections.
  4. Observe the Z signal at threading speed (check board / scope), not just by hand.
  5. Then coder, parameters, amplifier per the alarm 41 guide.
  6. After the fix: cut a multi-pass test thread and verify the passes track before production.

Related guides

AL-46 mid-thread — or threads that mysteriously stopped tracking? JJ Automation diagnoses feedback signal quality on site in NCR with proper instrumentation, stocks αi position coders and shielded cables, and repairs amplifier input stages at component level. WhatsApp us or call +91 88022 12394.
FanucOne-rotation signalPosition coderSpindle alarm 46Thread cuttingTroubleshooting