Planned Maintenance in PCB Manufacturing

Planned maintenance is the work done on equipment before it fails, scheduled at a time that suits production rather than at a time that suits the failure. A board shop depends on a small number of expensive, highly loaded machines, and the loss from an unplanned stop at the drilling area or the plating line is far larger than the cost of the maintenance that would have prevented it. The difficulty is not in believing this but in finding the time, which is why the schedule has to be defended rather than fitted in.

Planned and Preventive Maintenance

Planned maintenance covers work that is organised in advance, including inspections, lubrication, calibration checks and component replacement at defined intervals. Preventive maintenance is the subset driven by time or usage rather than by condition, and it is the part most shops start with because it requires only a schedule and a record. Both differ from reactive maintenance, which is the repair of something that has already stopped.

The balance between the two matters. Too little planned work and the shop spends its time reacting, with the production schedule at the mercy of the next failure. Too much and the shop replaces components that still have life, spending money and consuming production windows for no benefit. The right point moves as the equipment ages, and it is found by tracking failures rather than by choosing a policy in advance. A shop that records what failed and what was found at each inspection will converge on the right interval within a year.

Building the Maintenance Schedule

A maintenance schedule begins as a list of assets with the tasks that keep each one reliable. For a board shop that list typically covers the drilling spindles, the exposure optics, the plating rectifiers and filters, the press platens and heaters, the reflow ovens and the compressor and exhaust systems. Each task needs an interval, an estimated duration and the skill required to perform it, because a task that nobody is trained to perform is a task that will be deferred indefinitely.

The interval should come from the manufacturer’s recommendation adjusted by the shop’s own experience. A machine running three shifts wears faster than the handbook assumes, and a filter in a dusty area clogs sooner. Recording the failures that occur and the condition found at each inspection is what allows the intervals to be corrected over time, and that correction is what turns a generic schedule into a useful one.

Scheduling Around Production

Maintenance competes with production for the same hours, and the competition is usually resolved in favour of production until a failure resolves it in favour of the maintenance team. The way out is to plan the maintenance into the production schedule rather than to look for gaps, which means reserving defined windows in advance and treating them as commitments. A window that is cancelled whenever the shop is busy will always be cancelled.

Grouping tasks helps. Maintenance that requires the same isolation, the same shutdown or the same skill can be combined into one window, reducing the total production time lost. Grouping also makes the window easier to justify, because a four hour shutdown that completes six tasks is an easier decision than six separate interruptions of half an hour each. Publishing the maintenance window alongside the production plan removes the argument about whether the time is really available.

Condition Monitoring

Condition monitoring moves work from the calendar to the actual state of the machine. Vibration measurement on rotating equipment, temperature logging on heaters, pressure drop across filters and current draw on rectifiers all reveal deterioration before it becomes a failure. The advantage is that the maintenance is performed when it is needed, which reduces both unplanned stops and unnecessary replacements.

The practical barrier is cost, and the answer is to apply monitoring where the consequence is largest. Monitoring the spindle of the drilling machine or the rectifiers on the plating line is worth the instrumentation; monitoring a pump that can be swapped in twenty minutes may not be. The decision should be made from the cost of an unplanned stop rather than from the availability of the technology, and it should be revisited once the first results are in.

Records and Equipment Reliability

A maintenance record has two uses: it proves the work was done, and it accumulates the evidence needed to improve the schedule. Recording the date, the asset, the task, the condition found and the time taken turns a stack of forms into a dataset that shows which assets fail, how often and why. Without the condition notes, the record proves the work but teaches nothing.

Equipment reliability improves when the same failure is analysed rather than simply repaired. A repeat failure of the same component usually has a cause that the previous repair did not address, such as an alignment, a supply condition or an operating practice. Treating the second failure as a signal rather than as bad luck is what moves a shop from repairing to preventing, and the change is usually visible in the maintenance records within two or three months.

Measuring Downtime Reduction

Downtime reduction should be measured with the same figure used for equipment performance, which is the time the machine was unavailable when it was planned to run. Splitting that figure into planned maintenance time, unplanned breakdown time and changeover time makes the progress visible and prevents the shop from claiming an improvement by moving work from one category to another.

The most useful trend is the ratio of unplanned to planned downtime over time. A shop moving from mostly reactive to mostly planned work shows this ratio falling, and the improvement usually appears as better delivery performance before it appears as a cost saving. Delivery reliability is the commercial benefit that justifies the maintenance windows, and a shop that can promise a date because its equipment is reliable is selling something its competitors may not be able to sell at all.

Practical Rules

List the assets, define the tasks and intervals, reserve windows in advance and record the condition found at each visit. Use the records to correct the intervals, and apply condition monitoring where an unplanned stop is expensive. The list should be reviewed once a year, because the equipment changes and so does the consequence of its failure.

Analyse repeat failures rather than repairing them again, and review the unplanned to planned ratio each month with the production records. Planned maintenance in a board shop is not an overhead activity; it is the mechanism that keeps the quality of the process stable, because a machine that is drifting produces defects long before it stops.

FAQ

How often should maintenance be scheduled? According to the manufacturer’s recommendation adjusted by your own failure history. Three shift operation usually needs shorter intervals than the handbook suggests.

What is condition monitoring? Measuring the state of the equipment, such as vibration or temperature, so that maintenance happens when it is needed rather than on a fixed calendar.

How is the benefit measured? By the ratio of unplanned to planned downtime over time, and by the delivery performance that improves as failures become rarer.