Wear Condition: Replace Now vs Monitor (No-Dollars) Decision Tree

Why this matters

Field techs face the replace-now-or-monitor call on almost every preventive visit. A part that is worn but still operating sits in a grey zone. Replace too early and you generate waste and customer pushback. Replace too late and you own a callback, an after-hours emergency, or worse, a safety event. This decision tree gives you a condition-based framework that ignores price tags entirely. The trigger is observable wear state, run-time risk, and consequence of failure, not the customer's budget or your stocking cost.

This approach aligns with condition-based maintenance principles in ISO 17359 and the failure-modes guidance in ISO 14224. The point is to act on what the part is telling you, not on a calendar or a guess.

Symptom presentation

You see a component that is no longer pristine but is still doing its job. Common cross-trade examples:

• A contactor with light pitting on the face but clean pull-in
• A drive belt with surface cracking but no missing chunks or glazing
• A bearing with a faint hum that is louder than last visit but not yet rough
• A flue gasket with discoloration but no soot trail past the seal
• A flexible hydraulic hose with surface checking but no weeping
• An anode rod at half mass remaining
• A blade edge with rolled tips but no chips

The customer is not complaining. The unit is running. You have to decide whether to write it up as monitor, scheduled replace, or do-not-leave-it.

Quick checks before any call

1. Confirm the component is actually in the wear zone and not in the failed zone. A pitted contactor that chatters is failed, not worn.
2. Read the duty cycle. A part at 50 percent wear on a unit that runs 200 hours a year behaves differently than the same part on a unit that runs 4,000 hours.
3. Check the failure mode. Wear-out failures (predictable degradation) are monitor candidates. Sudden-death failures (insulation, bearings, springs) bias toward replace.
4. Check the consequence. Loss of a contactor in a residential air handler is an inconvenience. Loss of the same contactor on a walk-in cooler full of inventory is a disaster.
5. Confirm access. A part you can reach in five minutes can be monitored. A part behind a torn-down assembly should be replaced while you are in there.

Decision thresholds

Apply these in order. Stop at the first one that fires.

Replace now if any of the following are true:

• Wear has crossed the manufacturer's stated limit (groove depth, contact erosion, vane clearance, blade gauge, gasket compression set)
• The failure mode is sudden-death and the part is past the midpoint of expected life
• Consequence of in-service failure includes safety risk, refrigerant or fuel release, water damage, or loss of a freezer/cooler load
• The part is buried behind work you have already opened and re-access cost is high
• You see two or more independent wear indicators on the same part (e.g., belt cracking plus glazing plus visible dust trail under the sheave)
• The part is on a life-safety circuit (smoke control, emergency lighting transfer, gas valve seat) and shows any measurable wear

Schedule replace (next planned visit, parts on order) if:

• Wear is in the upper third of the published range but no single threshold is crossed
• The duty cycle is high and the next PM is more than half the remaining life away
• The part has failed prematurely on similar installations in your service area
• The customer has an event window coming (peak season, inspection, lease turnover) where in-service failure would be costly to them

Monitor if all of the following hold:

• Wear is in the lower two-thirds of the published range
• Failure mode is gradual and you have a clear next indicator to watch for
• Duty cycle is light or seasonal
• Re-access is cheap (front panel, single fastener, no refrigerant recovery)
• Consequence of failure is contained (annoyance, not damage)

Confirming the call

Before you write monitor, ask: what specifically am I asking the next tech (or myself) to watch for, and how will they know it crossed the line? If you cannot name the next observable indicator and the threshold it has to cross, you do not have a monitor plan, you have a deferral. In that case bias toward replace or schedule replace.

Document the wear state with a photo and a measurement where one exists. Contact erosion in thousandths, belt deflection at midspan, anode mass in inches remaining, bearing temperature delta over ambient, vibration band reading. Numbers turn a subjective call into a trend.

Next steps

When you replace, capture the removed part's actual wear state in your records so the next service plan calibrates against reality, not the OEM brochure. When you monitor, write the next indicator and the threshold into the customer record, not just into your head. The tech who follows you cannot read your mind.

If the customer pushes back on a replace-now call, the answer is the failure mode and the consequence, not a price comparison. Show them the wear indicator, name what fails next, and name what that failure damages. The decision becomes theirs to make on facts.

References

• ISO 17359 Condition monitoring and diagnostics of machines, general guidelines
• ISO 14224 Petroleum, petrochemical and natural gas industries, collection and exchange of reliability and maintenance data for equipment
• ISO 13379-1 Condition monitoring and diagnostics of machines, data interpretation and diagnostics techniques
• NFPA 70B Recommended Practice for Electrical Equipment Maintenance, contact and conductor inspection criteria
• ACCA Standard 4 Maintenance of Residential HVAC Systems, component-condition inspection guidance