Preventive Maintenance Scheduling for Stone Fabrication Equipment

A stone shop runs on a short list of machines, and that is precisely the problem. A cabinet shop with a dead sander moves the work to another station; a fabrication shop with a dead bridge saw has nothing cut today and no install the day after. The equipment is capital intensive and mostly single threaded, so the cost of an unplanned failure is counted in lost days across the whole schedule rather than in the price of the bearing that seized.

Nearly every shop already performs maintenance of some kind. What most lack is a schedule that survives a heavy week, an honest record of what was done, and a way to decide which of forty possible tasks deserves the two hours a technician has free on Friday. This guide builds that system from nothing: an asset register, a criticality ranking, interval logic matched to how each machine wears, a split between operator and technician work, staged parts kits, and a compliance figure that shows whether the plan is real.

What a Maintenance Program Has to Deliver in a Wet, Abrasive Shop

Water is what separates stone equipment from most other production machinery. Cutting, profiling, and polishing all run wet, so each machine lives inside a mist of slurry carrying silica and stone fines. That slurry finds bearing seals, way covers, limit switches, and enclosure gaskets. Hard water scales nozzles; recycled water carries abrasive solids back into pump impellers and rotary unions. A plan that ignores water management will not prevent the failures that stop production.

The second driver is what equipment emits when it is not working correctly. OSHA sets a permissible exposure limit for respirable crystalline silica of 50 micrograms per cubic meter as an eight-hour time-weighted average, with an action level of 25 micrograms per cubic meter. A clogged filter or a leaking water line can turn a compliant station into a non-compliant one with no visible change on the floor. Under the construction silica rule, employers using the tabulated control methods must operate and maintain water delivery systems and dust collectors according to the manufacturer instructions.

Noise belongs in the same category. The OSHA noise standard requires a hearing conservation program once employees reach an eight-hour time-weighted average of 85 dBA and requires engineering or administrative controls above 90 dBA, with impulse noise not exceeding 140 dB peak sound pressure level. Worn spindle bearings, a dry gearbox, and loose guards all push levels upward. When a machine gets louder, that change is data, and a note on a daily sheet often finds a bearing weeks before it seizes.

Structurally the work falls into three buckets: reactive repair after a breakdown, preventive tasks on a fixed interval, and condition-based tasks triggered by a measurement. Cheap, low-consequence items can be run to failure on purpose. Wear items with predictable life belong on fixed intervals. High-value assemblies such as CNC spindles and saw main bearings deserve a measurement, because calendar replacement throws away service life and post-failure replacement throws away a week of production.

Start With an Asset Register, Then Rank It

The register lists everything that can stop work, not only the obvious machines. Include the bridge saw and secondary saws, the CNC workcenter or router, edge and line polishers, manual grinders, the water recycling system with its pumps and clarifier, the slurry pit, dust extraction, the air compressor and dryer, the overhead crane or gantry, A-frames and carts, vacuum lifters and slab clamps, and the forklift. Anything that can idle a crew for half a day earns a line.

Fields worth capturing on day one

Each line needs an asset number, make, model, serial number, install date, and location. Then add what a technician needs with the machine down: control voltage and full load amps, lubricant grades and capacities, filter and belt part numbers, bearing designations, the tooling interface, and where the manual and schematic live. Photograph the nameplate and store the image with the record, because half the delay in an emergency repair is somebody hunting a part number.

Ranking criticality so the schedule has priorities

Score each asset from one to five on four questions: what happens to safety if it fails, how much production stops, whether a backup exists, and how long replacement parts take to arrive. Sum and sort. In most shops the bridge saw, the CNC, and the water system land at the top; dust extraction and lifting equipment come next because their failures are safety events rather than throughput events. That ranking decides where technician hours go when the week gets away from you.

Choosing the Interval Basis for Each Asset

Three bases exist, and picking the wrong one is the most common reason a schedule stops matching reality. Runtime intervals suit anything whose wear tracks operating hours: spindle lubrication, gearbox oil, pump seals, compressor service. Calendar intervals suit anything that degrades whether or not the machine runs, including corrosion, hose ageing, and statutory inspections. Condition-based intervals suit expensive components with detectable early warning, where the trigger is a temperature, a current draw, or a pressure differential rather than a date. Mixing all three on one asset is normal.

Capturing machine hours when the equipment has no counter

Plenty of shop equipment ships without an hour meter, and counters that exist often log control-on time rather than spindle time, overstating wear on a machine left powered all day. Check the drive first: many variable frequency drives already record run hours in a diagnostics screen nobody has opened. Second, fit an inexpensive elapsed-time indicator across the spindle contactor or coolant pump circuit so it accumulates only while the machine cuts. Third, use a proxy the shop already records.

Useful proxies include linear feet cut, square feet processed, slab count, and blade changes. Calibrate once by timing a representative week and dividing measured run hours by measured output, then convert the manufacturer hour intervals into output intervals and post them at the machine. A proxy that is roughly right and always recorded beats a precise counter nobody reads.

Condition checks a shop can genuinely perform

Condition monitoring does not require an analyst on staff. An infrared thermometer and a paint pen give a repeatable temperature check: mark the exact spot on each bearing housing and motor, read at the same point in the cycle, and log it. A clamp meter shows rising current as a drive train binds, a differential pressure gauge across dust collector filters shows loading long before suction visibly drops, and gauges on water headers show nozzle scale and pump wear.

Splitting the Work: Operator Checks and Scheduled Technician Tasks

Operators find failures first because they stand next to the machine all day, but only if the check is short, physical, and unambiguous. A daily sheet should fit one page, take under ten minutes for the whole cell, and ask yes-or-no questions with space to write a number. Anything requiring guards to come off, lockout of energy sources, or interpretation of a specification belongs to a trained technician on a scheduled work order.

What belongs on the morning walk

Water pressure and flow at each head, coolant level and clarity, visible leaks, condition and mounting security of the blade or tooling, air pressure and moisture at the filter regulator, function of the emergency stop, condition of way covers and bellows, slurry on rails and limit switches, and a look at the slings, clamps, and vacuum pads that will lift slabs that day. End the shift by washing the machine down before slurry cures on it.

Scheduled technician work

The table below is a defensible starting point for a shop running one production shift; adjust it against manufacturer documentation and your own failure history. Several entries are not optional. Powered industrial trucks must be examined before being placed in service and at least daily, and after each shift where they run around the clock. Slings must be inspected each day before use by a competent person, with a thorough periodic inspection at intervals not exceeding twelve months.

Build a Parts Kit for Every Recurring Task

A scheduled task that cannot be completed because a filter is missing is worse than no task at all, because it teaches people to skip the schedule. Every recurring job should have a written bill of materials: filter numbers, gasket sizes, belt designations, seal kits, lubricant grades and quantities, and any consumable replaced whether or not it looks worn. Bag those items together, label the container with the asset number and task name, and stage it before the work order is released. The technician should never have to go shopping.

Stock levels follow criticality and lead time. Cheap fast-moving items sit at a comfortable minimum on a two-bin reorder trigger. Expensive, slow-moving, long-lead items get a deliberate decision: hold one on the shelf, or accept the downtime and document that choice. Spindle bearings, main saw drive belts, water pump seals, contactors, proximity switches, and rotary union rebuild kits are the usual candidates for on-site stock.

Measuring Whether the Program Is Real

Preventive maintenance compliance is the percentage of scheduled tasks completed on time, and it separates a real program from a document. The common convention is the ten percent rule: a task counts as compliant only if performed within plus or minus ten percent of its interval, so a thirty-day task must land between day twenty-seven and day thirty-three. Counting a task done six weeks late as complete produces a comfortable number and no reliability benefit.

Commonly cited targets put strong programs at ninety percent compliance or better, with sustained results below roughly seventy percent indicating that reactive work has taken over. Track two supporting figures. The ratio of planned to unplanned hours shows whether the program is buying you anything, and the share of corrective work orders generated by a scheduled inspection rather than a breakdown shows whether the checks are catching things.

Phasing In a CMMS Without Stalling Production

A disciplined spreadsheet beats abandoned software. Start with one workbook: an asset tab holding the register, a task tab listing every job with its asset, interval, basis, duration, parts kit, and acceptance criterion, and a log tab where completed work is dated and signed with meter readings. Sort by next-due date, print the week, and post it. That much can be running by Friday, because the benefit comes from discipline rather than tooling.

Phase the rollout by criticality instead of covering everything at once. Spend the first month on the top three assets, prove the checks are done and the parts are on the shelf, then extend to the next tier. When you move to software, choose on how technicians will use it: mobile entry, scannable tags at the machine, offline capability, and painless export of your history.

Long-Term Considerations: Water, Geometry, and Machine Life

The water system deserves its own long-horizon plan because it silently damages everything downstream. Solids that are not settled or filtered out become an abrasive slurry pumped through seals, unions, and nozzles all day. Set clean-out intervals on volume processed rather than the calendar, and treat pump seal replacement as a scheduled task instead of a repair. Where water is recycled, watch filtration as closely as pump output.

Lubrication discipline is dull and decisive. Use the grades the manufacturer specifies, respect quantities rather than pumping grease until it emerges, and avoid mixing incompatible greases in one bearing, which can break down the thickener and leave the bearing effectively dry. Label every lubrication point with product and quantity, keep a dedicated colour-coded gun per product family, and store lubricants sealed away from the slurry mist.

Geometry and calibration checks are the tasks most often skipped, and they quietly increase your remake rate. Verify blade squareness to the table, rail level and straightness, head tram on polishers, and axis positioning on the CNC at least annually, and immediately after any collision or major service. Record measured values rather than a pass mark, because the drift between checks tells you whether a component is wearing faster than expected.

If you are rebuilding a maintenance program, review consumables and wear parts at the same time, since tooling condition and machine condition tend to fail together. The catalog at Dynamic Stone Tools covers blades, profiling tooling, polishing consumables, and handling equipment, and the reference material at dynamicstonetools.com covers selection and setup for the machines on your register.

Free Tool

Free Guides & Tools — practical shop resources, including checklists and reference charts you can adapt into the daily operator sheets and scheduled task lists described above.