There's a difference between a maintenance issue you can handle in-house and a repair that requires a trained technician with the right tools and parts. Most shops know this in principle. In practice, the line gets blurred — particularly when production pressure pushes you to keep running machines that are sending clear signals they need attention.
These are the five signs that cross the line. If your machine is showing any of these, it needs professional evaluation — not another reset, and not another week on the schedule.
Alarms That Return After Reset — Especially the Same One, Repeatedly
An alarm that clears when you cycle power and doesn't come back for weeks is different from an alarm that fires, clears, and fires again — whether that's three times in a day or once a week for a month. Recurring alarms are the machine telling you something is wrong at a level that a reset doesn't reach.
Intermittent faults are the most damaging pattern in CNC maintenance, not because any single alarm event causes catastrophic failure, but because they are almost always a symptom of a component in the process of failing. A servo drive that faults intermittently is on its way to a hard failure. An encoder that throws a feedback alarm once a week is degrading. These failures have a reliable endpoint — and that endpoint is an unplanned, hard-stop breakdown, usually at the worst possible time.
The specific alarm codes that should never be treated as "nuisances" to reset and ignore:
- Fanuc SV4xx (servo alarms), SV009 (error excessive), any recurring OH alarm
- Siemens 380xxx (drive alarms), NCK alarms that appear without a program error cause
- Any alarm referencing encoder feedback, pulse coder, or position error on any control
If the alarm code means something to you and you're not sure whether it warrants a call, call anyway. A five-minute conversation with a technician is free. A full breakdown with parts on order is not.
Parts Are Consistently Coming Off Out of Tolerance — Without an Obvious Cause
Dimensional drift is not always operator error or tooling wear. When parts that were in spec last week are consistently running high or low — and you've verified your offsets, checked your tooling, and confirmed there's no programming issue — the machine itself is the variable.
Causes that require professional diagnosis:
- Ballscrew wear or damaged nut: Creates backlash or positional error that offset adjustment can mask temporarily but not correct permanently
- Axis drive issues: Servo following error that's within tolerance but trending upward often indicates a motor or drive issue before it becomes an alarm
- Thermal growth outside compensation range: Particularly relevant in New England shops with variable temperature environments — but if compensation isn't correcting for the drift, the machine needs evaluation
- Spindle runout: Runout that has increased from your baseline affects surface finish and dimensional accuracy in ways that can mimic tooling or setup problems
The danger of attributing consistent dimensional problems to operator error or normal wear-and-tear is that it delays the diagnosis while the underlying issue progresses. A machine that needs a ballscrew preload adjustment now may need a full ballscrew replacement in three months if the wear continues unchecked.
New Sounds That Weren't There Before
Experienced operators know what their machines sound like. When that changes — a new grinding, a rumble at certain RPMs, a knock during rapids, a whine from the spindle that wasn't there last month — that change is information. Machines don't develop new sounds without a reason.
Sounds that require immediate professional evaluation:
- Spindle grinding or rumbling: Bearing wear or contamination — progressive and irreversible without intervention. Every hour of operation accelerates the damage.
- Axis knocking or clunking during direction changes: Can indicate ballscrew coupling issues, way surface damage, or loose mechanical components in the drive train
- High-pitched whining from servo or spindle motors: Motor bearing wear or drive issues
- Coolant pump noise changes: Pump bearing wear or cavitation from low fluid levels or contamination
Sounds are often the earliest detectable sign of a developing failure — before alarms fire and before parts go out of spec. Acting on a new sound is the difference between a bearing replacement and a spindle rebuild.
Visible Fluid Leaks — Coolant, Hydraulic, or Lubrication Oil
Fluid leaks are rarely cosmetic. A coolant leak that's dripping into the machine base may be contaminating way lubrication. A hydraulic leak means reduced clamping force, which affects both workholding security and tool change reliability. A lube system leak means ways and ballscrews are running dry — accelerating wear in ways you won't see until significant damage has occurred.
Leaks that require professional repair:
- Hydraulic fluid leaking from seals, valves, or lines — particularly if the machine has hydraulic clamping or a hydraulic tool changer
- Coolant that is entering the machine base or electrical cabinet through damaged way covers or seals
- Centralized lube system leaks — a lube system with a broken fitting or cracked line may appear to operate (the reservoir empties) while delivering zero oil to the ways
- Spindle coolant-through leaks that suggest seal degradation inside the spindle assembly
The repair itself may be straightforward. The damage caused by ignoring it is not.
The Machine Has Been Through a Crash — Even a "Minor" One
Shop culture around crashes is often to get the machine running again as fast as possible and not mention it. This is understandable. It is also one of the most common causes of delayed major failures.
A crash that trips the servo drive overload protection — even a slow crash that felt minor — transmits force through the spindle, toolholder, and into the spindle bearing and taper. It can shift axis geometry. It can damage encoder couplings. It can crack a toolholder taper bore in ways that are invisible until the next tool pulls loose under cutting load.
After any crash — regardless of how minor it felt — a professional should:
- Measure spindle runout against your pre-crash baseline
- Check axis geometry for shifts in squareness
- Inspect the spindle taper for damage
- Verify servo drive fault history for overload events
- Check the toolholder that was in the spindle at the time
Most post-crash evaluations come back clean. When they don't — when there's runout that's increased by 0.0003", or a taper that shows a hairline — catching it immediately is what keeps a crash from becoming a delayed spindle failure three months later.
The Common Thread
Every one of these signs shares the same characteristic: they are easier and less expensive to address at the point they appear than at the point where they become unavoidable. CNC machines communicate consistently and honestly — and the shops that listen to that communication are the ones that control their repair costs instead of being controlled by them.
If your machine is showing any of these signs, call for a professional evaluation. If you're not sure whether what you're seeing qualifies, review our broader list of CNC warning signs — and when in doubt, call. A diagnostic conversation costs nothing.
Seeing Any of These Signs on Your Machine?
Maz CNC provides professional on-site CNC evaluation and repair across New Hampshire, Massachusetts, Rhode Island, and Connecticut. Call us 24/7 — we'd rather catch it early too.
(603) 562-4759
Schedule a DiagnosticRelated reading: Emergency CNC Repair — What To Do When Your Machine Goes Down | How Long Does CNC Repair Take? | 7 Warning Signs Your CNC Machine Needs Repair
