Day eighteen of a twenty-one day hatch. The setter is full, the eggs candled clear and moving, and at two in the morning the display goes dark. By sunrise you know the power is fine and the controller is not. The board that failed is a small thing, not a gearbox and not a motor, and it costs less than a bag of feed.
Nobody within reach has one. The nearest stockist wants the model number, printed on a sticker that three years of humidity have steamed off. Three days later the right part arrives and the machine runs perfectly again. The hatch does not, because embryos keep no calendar of their own.
Almost nothing in that story is about the failure. It is about the gap between failure and repair, and that gap is the part of a breakdown you can manage. This article is about closing it: which parts genuinely fail on Kenyan farm machines, which of them deserve shelf space, how to record what you need before you need it, and how to store spares so they are still good when you reach for them.
The Expensive Part of a Breakdown Is Almost Never the Part
Every stoppage carries two costs. The first is the price of the component: visible, fixed and usually modest. The second is the cost of waiting, invisible on the day you buy the machine and often many times the first.
The asymmetry matters. The price of a part does not change whether you bought it in advance or in a panic, while the cost of waiting changes enormously. A belt bought calmly in August and one bought desperately in November cost close to the same money. What differs is the four stopped days between them.
Waiting has three components: diagnosis, sourcing and fitting. Fitting is usually the shortest and the only one most people plan for. Sourcing is the long one, and it is not a fixed property of a part. It depends on how common the item is, whether your town has a stockist, whether it is a standard industrial size or something moulded for one model, and whether you can state exactly what you want. That last condition costs nothing and is entirely within your control.
Downtime Cost Is a Rate, and the Calendar Sets It
The same machine stopped in two different weeks produces two completely different losses, and treating downtime as one flat number is the mistake that leads farms to under-stock.
A chaff cutter idle in a wet month with standing fodder in the field is an inconvenience. The same cutter idle in the dry season, when every kilogram the herd eats passes through it, is a daily milk loss that does not come back when the machine restarts. Biology keeps its own schedule: cows lactate, chicks pip and irrigation windows close whether or not your equipment is running.
Count a stopped day in five directions. Lost output: the litres, trays or bags you did not produce. Irreversible spoilage: a failed hatch, feed gone sour, a crop that passed its spray window. Substitution: the mill you hire or the dairy meal you buy instead of making. Idle labour, which is still paid. And for a service business such as a posho mill, customer loss, because people who walk to the next mill do not all walk back.
Work out one number and keep it. What does a single stopped day cost you in the worst week of your year, not the average week? That figure decides nearly every stocking question in this article, and without it you are guessing low.
What Actually Fails: A Machine by Machine Map
Failures are not random. Every machine has a small number of sacrificial components designed, or at least resigned, to wear out. Knowing which they are turns a vague worry into a short shopping list.
Egg incubators. Heating elements are the classic. They cycle on and off thousands of times per hatch, and it is that thermal cycling, not the running temperature, that eventually breaks the element or its terminal connection. Next come the controller and the sensors, which drift before they die, so the early symptom is a machine holding temperature less tightly than it used to. Then the turning motor and its gear train, which run continuously for weeks, the circulation fan and the door seal. We stock incubator spare parts because these items are predictable and the machine is worthless without them for the twenty-one days that matter.
Chaff cutters. The blade and the shear bar are a pair of scissors, and scissors fail as a pair. Blades go blunt, then chip, and the bolts holding them stretch and loosen under impact. When the clearance between them opens, fodder is torn rather than cut: power draw rises, output falls and the cut length goes ragged. Behind those sit the V belts, the cutter shaft bearings and the feed roller chain. Anyone running a chaff cutter daily through the dry season should treat blades, bolts and a belt as consumables rather than emergencies.
Hammer mills and posho mills. Hammers and screens are the sacrificial pair and they wear together. Maize dried on a compound floor carries abrasive soil and grit, so hammer tips round off and the pins they swing on wear oval. Screens blind with fines first, then tear at the edge of a hole. The symptom sequence is familiar: output slows, meal comes out coarser, the motor sounds more loaded, then a screen lets a stone through. Behind those sit the beater shaft bearings and the drive belts.
Water pumps. The mechanical seal decides a pump's life. It survives on a microscopic film of water between two polished faces, so a few minutes of dry running can destroy it. Next comes the impeller, which erodes where sand is present and loses head long before it looks broken. Then, on single phase units, the start or run capacitor, a cheap component that fails silently and leaves a motor humming without turning. Then the pressure switch, the foot valve, the strainer and the motor bearings.
Sprayers. Nozzles quietly cost money long before they fail outright. The orifice enlarges with use and the pattern degrades from an even cone toward a stream, so you apply more chemical per hectare for a worse result. Then the O-rings through the lance and tank fitting, the pump diaphragm or piston cup, the trigger valve and the in-line filter.
Everything with a belt or a shaft. Belts and bearings are the universal spares. They sit on cutters, mills, pumps and threshers alike, they carry standard numbers, and they are cheap relative to what their absence costs. A bearing announces itself by running hot and then changing note; a belt announces itself by squealing on start-up and by glazed or cracked flanks.
Pre-Buy or Source on Failure: A Decision You Can Make on Paper
You cannot stock everything, and trying to is simply a way of converting working capital into shelf decoration. The decision is made with four questions, asked per part rather than per machine.
Does it wear, or does it die? Wear parts give notice: blades, hammers, screens, nozzles, belts, seals. You can inspect them and replace them on your own timetable, so you rarely need one urgently. Dying parts fail without warning: a controller board, a capacitor, a bearing that took a knock during fitting. Those create the three-day gap.
How fast can you honestly get it on the worst day? Not on a Tuesday morning with the road dry and the shop open, but on a Sunday, on a public holiday, in a rainy week, when the one person in your town who stocks it has travelled.
What does a stopped day cost you? The number from the previous section, taken at its peak-season value.
Does it keep? A bearing in its sealed wrapper keeps for years. A steel blade under a film of oil keeps indefinitely. A rubber diaphragm in a hot iron-sheet store is aging from the day you buy it. Shelf life is the constraint that stops the list getting silly.
The result surprises most people. The parts that deserve shelf space are usually not the expensive ones. They are the cheap ones with awkward supply: a specific nozzle size, an O-ring kit, a motor capacitor, a particular belt length, a controller sensor.
The cheap part with a long lead time is the one that hurts. A nozzle, an O-ring, a capacitor, a drive belt: each is a small purchase, and each can stop a machine for a week when the only stockist is three towns away. Build that end of your list first.
One category sits outside the matrix: the insurance spare, a complete assembly held only because the consequence of its absence is catastrophic. A standby generator that will not start during an outage is the example, since the whole point of the machine is to work on the day everything else has failed. If you rely on one, the routine in our guide to generator maintenance matters more than any spare, because most standby failures come from a machine that sat still.
Write Down the Numbers Before You Need Them
This is the highest-return, lowest-cost habit in the entire article, and almost nobody does it. Part numbers are stamped, printed or moulded on components while the machine is healthy, and they become unreadable at precisely the moment you need them. A bearing that has run hot has a scored, discoloured face where the number used to be. A snapped V belt usually snaps through the printed section. A controller sticker steams off in an incubator room.
So capture the data while the machine is working. For each machine record the make, model, serial number and the branch you bought it from. Record the motor nameplate in full: voltage, current, power, frame size, speed and the capacitor rating. Then the consumables: the belt marking including section letter and length, the bearing numbers off the bearing face including the seal suffix, the seal dimensions, the screen aperture in millimetres, the nozzle type and size, and the blade count and bolt size.
Photograph all of it, then store the record in two places because one will fail: an album on your phone named for the machine, and a printed page in a plastic sleeve in the store.
The third step is the one people skip and the one that pays. Take your numbers to a counter while nothing is broken and ask whether the part is held and how long it takes if it is not. That turns a guess about lead time into a fact, and facts are what the matrix above needs.
One caution on what you buy. A spare that fails early costs two stoppages instead of one, so the quality question applies to components as much as to machines. The reading skills in our guide to telling genuine equipment from imitation apply just as well to a bearing in a plain box.
Storing Spares So They Are Still Good When You Reach for Them
A spare that has degraded on the shelf is worse than no spare, because you find out only after stripping the machine. Different materials have different enemies, and storage is a matter of keeping each away from what destroys it.
Rubber: belts, seals, O-rings, diaphragms, hoses. The enemies are ultraviolet light, heat, ozone, oil and mechanical strain. Keep rubber in the dark, in a bag or closed box, at the coolest steady temperature you have. Hang belts over a wide peg or lay them flat, never kinked over a nail, because a permanent crease becomes a crack. Keep rubber away from the welding bay and from any motor that arcs, and never store it stretched.
Bearings. Leave them sealed in the original packaging with the greased paper intact until the minute they go on the shaft. The enemies are moisture, dust and vibration: a bearing stored on a bench that a mill shakes all day can be marked in its races before it ever runs, and unwrapping one to admire it invites dust into a precision fit.
Electronics: controller boards, sensors, capacitors. The enemy is the daily condensation cycle inside an iron-sheet store that bakes in the sun and cools at night. Keep boards in a sealed bag with a desiccant sachet inside a rigid box, and keep that box in the house if the store is a hot one.
Steel wear parts: blades, hammers, screens, impellers. The enemy is rust and the remedy is a film of oil or grease and a wrap of paper, kept off the floor because concrete wicks moisture. Pitting on a cutting edge is a defect, not a cosmetic issue, since a pitted blade will not take a clean edge. Store gaskets and filters flat and dry.
An unlabelled spare is not a spare. On the day it arrives, write on the box which machine it belongs to and the number printed on the part. In two years nobody will remember, and a bearing in a plain white box that might be a 6204 or a 6205 will be left there while somebody drives to town for a known one.
Three rules make the shelf a system rather than a pile. Keep everything in one place, so there is a single answer to whether you have one. Keep one written list, updated when something goes on or comes off, because a spare nobody knows about is a spare nobody uses. And rotate first in, first out, replacing what you use in the same week you use it. That last discipline stops a shelf quietly emptying over two seasons until the day you need it and it is bare.
Building the Shelf Without Tying Up Your Working Capital
Nobody needs to build this in one purchase, and doing so would be a poor use of money. Build it in three tiers, in order.
Tier one: universal consumables. These serve several machines at once and are the cheapest insurance you will buy. The belts your machines actually use, the two or three bearing sizes that recur across them, an assorted O-ring kit, the fuses your control boxes take, a motor capacitor of the rating you run, spare sprayer nozzles and a set of blade or hammer bolts. One of each beats three of one.
Tier two: machine specific wear parts. Chaff cutter blades, a mill screen and hammers, a pump mechanical seal, an incubator element, a sprayer diaphragm. Add these for the machines that carry your production and skip them where a machine could sit idle for a fortnight without loss.
Tier three: insurance spares. Complete assemblies held only where the consequence justifies it, such as a controller for an incubator running continuous hatches. This tier should be short, and every item on it defensible with your downtime number.
Two habits sharpen the list over time. Keep a failure log, one line per event: the date, the machine, the part, the hours you were stopped and where you eventually got it. After one season that log is worth more than any general guide, including this one, because it is about your machines and your suppliers. And standardise where you can: a farm running three cutters of one model carries one set of blades, while a farm running three models carries three.
Buy your first set of spares when you buy the machine, while the dealer has the exact match in stock and knows which model you own. We deliberately do not print prices in articles, because they move and a stale figure is worse than none: the figures on our price list are the current ones, and any branch will confirm what is on the shelf today.
Key Takeaways
- The expensive part of a breakdown is the gap between failure and repair, not the component. Buying in advance does not change the price of a part, but it removes the cost of waiting.
- Work out what one stopped day costs you in your worst week, counting lost output, spoilage, substitution, idle labour and lost customers. That number decides what earns shelf space.
- Failures are predictable by machine: elements and controllers on incubators, blades and belts on cutters, hammers and screens on mills, seals and impellers on pumps, nozzles and O-rings on sprayers, bearings everywhere.
- Stock the cheap parts with long lead times first, and record part numbers while the machine is healthy, because those numbers become unreadable exactly when you need them.
- Store by material: rubber dark and unstretched, bearings sealed until fitting, electronics dry with desiccant, steel oiled and off the floor. Label everything on arrival and replace what you use the same week.
Frequently Asked Questions
Which spares should a small farm buy first?
Start with items that are cheap, serve more than one machine and are awkward to source locally: the specific V belts your machines use, the two or three bearing sizes that recur across them, an O-ring kit, fuses, a motor capacitor and spare sprayer nozzles. Only once that layer is in place should you argue about blades, screens, elements or complete assemblies.
How long do rubber parts keep on the shelf?
Much longer if they are stored properly and much shorter if they are not. Rubber is degraded by ultraviolet light, heat, ozone and being held stretched or creased, so a belt in a dark closed box is in a different condition from an identical belt hung on a nail in a window. Before fitting any stored rubber part, flex it and look for cracking, hardness or a chalky feel. If it looks tired, treat it as expired.
Is it worth keeping a spare pump or a spare motor?
Only where the downtime number justifies it. A complete assembly ties up real capital, so it earns its place when the machine cannot be substituted in its season and the loss cannot be recovered later, such as irrigation in a critical growth window. For most farms a spare seal, impeller and capacitor covers the majority of pump failures at a fraction of the cost.
How do I find the part number if the machine is already broken and the label is gone?
Measure rather than guess. For bearings, measure the bore, outside diameter and width in millimetres, which identifies the standard size even when the face is unreadable. For belts, measure the top width and total length and note whether the profile is A, B or a narrow section. For seals, measure the shaft diameter, housing bore and depth. Better still, bring the old part, because a counter matches a sample far faster than a description.
Fuga keeps a spares counter and not only a showroom: incubator elements, controllers, sensors and turning motors, chaff cutter blades and belts, mill screens and hammers, pump seals and impellers, sprayer nozzles and seal kits, and bearings in the common sizes. Send a photograph of the nameplate and the failed component, and any branch will tell you what is on the shelf today rather than what might arrive next week.
