Grinder maintenance and calibration schedules for busy sites
Short answer: the grinder controls the most sensitive variable in espresso and gets the least attention of any machine on the bar. A workable schedule is short — brush down and wipe out daily, clean deeper and check retention weekly, degrease or strip and check alignment monthly, and inspect burrs against throughput rather than the calendar. The part most often missing is not a task, it is a written record of the setting and a named person allowed to change it.
Espresso machines get serviced because they fail visibly and expensively. Grinders rarely fail. They degrade, and degradation gets absorbed by the person on the bar making a small adjustment each week, until the coffee is noticeably worse than it was and nobody can say when that happened.
Why the grinder is the device that matters most
Grind is the variable with the steepest response curve on the bar. A small change in setting moves shot time more than an equivalent change in dose, temperature or pressure, and it also changes the shape of the particle size distribution, which affects how evenly the bed extracts rather than just how fast water passes through it.
It is also the device that changes on its own. A boiler holds its temperature until something breaks. A grinder's output shifts continuously — burrs dull, retention accumulates, oil films build, the burr carrier moves, temperature swings with duty cycle. Every one of those changes the cup without changing the number on the dial.
The asymmetry is complete: the most influential device is the one that drifts, and the drift is invisible because staff compensate for it in small increments. A maintenance schedule exists to make it visible.
The daily routine
Three things, five minutes, at open and at close.
Brush down the grinding chamber and the exit. Fines pack into corners, the chute and the fork or portafilter rest. They are yesterday's coffee, they stale where they sit, and they leave with today's dose. A stiff brush and the willingness to look inside cover this.
Empty and wipe the hopper. Not a deep clean — tip the beans out or run them down, wipe the interior and look at the throat for oil build-up. The reasoning, and when emptying is and is not worth it, is in keeping coffee fresh in the hopper.
Check the first-shot behaviour. Grind, dose and pull one shot before service. Note the time. This is the cheapest diagnostic available: a shot that runs materially differently from yesterday's at the same setting means something has changed, and finding out what before service beats discovering it during. Expect the first shot after a night to differ slightly from the middle of a rush — that difference is thermal and normal, and the problem with overheating grinders explains why.
The weekly routine
A deeper clean. Remove the hopper, remove what the design allows without tools, brush and vacuum the chamber, wipe the throat and the gate. Degrease the hopper properly rather than dry-wiping it.
Check retention. Retention is the coffee the grinder holds internally between doses, and it matters because retained grounds mix the previous grind into the current one — a grinder with high retention delivers a blend of settings rather than the setting you chose. The test: grind a known dose by weight, weigh what comes out, compare. A consistent shortfall is retained coffee. Track it over time; a rising figure means build-up rather than a design characteristic, and build-up is cleanable.
Inspect for oil residue. Look at the chamber walls, the chute and the underside of the top burr carrier. Oily deposits mean the cleaning interval is too long for the roast in use — a darker roast may need weekly what a lighter one needs monthly.
The monthly routine
Cleaning tablets or a strip-down, depending on the design. Grinder cleaning tablets are a food-safe abrasive that runs through the burrs and carries out oil and fines. They are convenient and they suit designs that are awkward to open, but they clean the path rather than the corners. Check the manufacturer's guidance on whether tablets are approved for your burr set, and purge coffee through afterwards before serving. Where burrs come out with hand tools, an actual strip — burrs out, carrier and threads cleaned, burrs brushed — is the better job.
Check alignment. See the next section; this is the check most often skipped and most often the answer.
Check fasteners and the adjustment mechanism. Burr screws loosen under vibration. An adjustment collar that has become stiff, gritty or loose in its detents is telling you that coffee has got into the threads, and a mechanism that moves under vibration will not hold a setting no matter how carefully it is set.
By throughput, not by date: burrs
Burrs wear as a function of how much coffee has passed through them, not how long they have been fitted. A grinder doing heavy volume five days a week and one doing light volume seven days are on completely different schedules despite identical install dates.
That makes throughput the honest metric, and it means tracking kilograms ground rather than months elapsed — either from a counter, if the grinder has one, or from delivery records, which are usually good enough. The replacement figure is specific to the burr set: geometry, material, coating and manufacturing all change it, so ask the manufacturer of your burrs rather than adopting a number heard elsewhere.
Between inspections, the symptom to watch for is setting creep: needing a progressively finer setting to hold the same shot time on the same coffee. That pattern is the clearest field indicator of dulling, and it is why the written log below matters — creep is only visible if the old settings were recorded. How burr sharpness changes your grind covers the mechanism and what dulling tastes like.
| Interval | Tasks |
|---|---|
| Daily | Brush chamber and exit · empty and wipe hopper · pull and time a first shot |
| Weekly | Deeper clean · degrease hopper and throat · measure retention · inspect for oil residue |
| Monthly | Cleaning tablets or strip-down · alignment check · check fasteners and adjustment mechanism |
| By throughput | Burr inspection, then replacement at the interval specified for that burr set |
| On any change | Re-check the setting after a clean, a burr change, a coffee change or a move |
Calibration discipline
The schedule above is the easy half. Calibration is where sites lose consistency, and the fixes are organisational rather than technical.
Name who adjusts. A grinder that anyone may adjust will be adjusted by everyone, in opposite directions, without record. By the end of a week the setting reflects the accumulated opinions of four people and the coffee is somewhere none of them intended. One named person per site owns the setting, with a defined deputy. Everyone else reports a problem rather than solving it.
Write the setting down, every time it changes. A remembered setting is worth almost nothing: it cannot be compared with last month, it cannot be handed over, and it leaves when the person does. Written, it becomes a record of drift — and drift is the signal you are trying to detect. The entry needs four things: the date, the coffee, the setting, and the shot time and yield it produced.
Re-check after anything that could have moved it. Any clean that opens the burr chamber, any burr change, any move of the machine, any new delivery, any change of roast. In each case the correct assumption is that the setting is no longer valid until a shot has confirmed it. A burr change in particular invalidates the setting completely: new burrs have a different zero point and a different distribution, and they need dialling in from scratch — and then again after a few kilograms, once they have settled.
Change one thing at a time. Cleaning and re-dialling and switching coffee in the same session leaves you unable to attribute the result to any of them.
Why alignment is checked before burrs are blamed
When a grinder produces poor or inconsistent shots, the reflex is to suspect the burrs. That is often wrong, and the wrong version is expensive.
Alignment is the parallelism of the two burrs across their whole facing surface. If they are not parallel, the gap varies around the circumference: wider on one side, narrower on the other. Coffee then takes the path of least resistance through the wide side and is over-processed on the narrow side, so the output is a wider particle distribution than the same burrs would give if properly aligned — and, on a badly misaligned set, uneven wear that eventually damages the burrs themselves.
The practical consequence is that a sharp but misaligned set can grind worse than a worn but well-aligned one. Fitting new burrs into a misaligned grinder produces a small improvement, an unexplained disappointment, and a new set of burrs that will wear unevenly. Check alignment first, and check it again after fitting new burrs, because the fitting is itself an opportunity to introduce a problem.
Alignment checks and corrections vary by design, and some grinders are built so that user correction is not intended. Where that is the case, treat it as a service item rather than a project: an unnecessary attempt at shimming can leave a grinder worse than it was found.
Building a log that survives staff turnover
Every argument above depends on a record, and records fail for predictable reasons: they live on someone's phone, or in a book that goes missing, or in a format so demanding that it is abandoned in week three.
What works is short and physical, kept at the machine.
- One line per entry, not a form. Date, coffee, setting, shot time, yield, initials. If a change was made, why in five words.
- Kept where the work happens. A sheet on the grinder or beside it. A log in a back office is a log nobody writes in.
- Tasks with tick boxes, and dates rather than "done". A weekly task with no date does not tell you whether it happened this week.
- A burr line. Install date, throughput at install, expected interval for that set, throughput now. Four numbers that answer the burr question instantly and would otherwise take an afternoon of guesswork.
- Reviewed by someone, briefly, weekly. A log that is never read teaches staff that filling it in does not matter.
The value shows up on the day the person who knew the grinder leaves. With a log, the replacement inherits a working setting, a maintenance history and a burr age. Without one, they inherit a grinder at an unexplained setting and no way to tell whether the coffee is worse than it used to be. This is exactly the difference between a repeatable process and a skilled individual, discussed more generally in automation and consistency: what repeatable means.
The multi-site version
Across several locations, grinder drift is a leading cause of cups that do not match, and it is under-diagnosed because the search usually starts with the espresso machines.
Each site's grinder drifts independently — different volumes, different burr ages, different cleaning discipline, different rooms — and each site's staff compensate independently. Two sites nominally running the same recipe on the same equipment diverge steadily, and neither has done anything wrong. Three practices contain it.
Specify outputs, never settings. A central recipe should state dose, yield and time. It must not state a grind setting, because the setting is not a shared unit: it is a position on that specific grinder's mechanism, dependent on its zero point, its burr wear and its retention. Circulating a number invites sites to match the number instead of the cup.
Standardise the routine, not the dial. Same tasks, same intervals, same log format at every site. That makes comparison possible — a site whose retention figure is twice everyone else's has a cleaning problem, and you can see it from the log.
Track burr throughput centrally. Burr age is the variable most likely to be wrong at exactly one site, and it is the one nobody notices locally, because local staff have adapted to it gradually.
Where the effort stops paying. A single-site bar with one grinder, one coffee and a stable team does not need a multi-site quality system. It needs the daily brush-down, the weekly degrease, a written setting and a burr count; beyond that the returns fall away quickly. The same applies to buying — a maintenance problem does not become a purchasing problem until the routine has been run properly for a few months and the coffee is still inconsistent. Many grinders described as worn out have never been aligned, and a good number have never been cleaned past the hopper.
Frequently asked questions
How often should burrs be replaced? By throughput, at the interval specified for that burr set, which the manufacturer can give you. Calendar intervals are unreliable because two grinders of the same model can be years apart in wear at the same age. Setting creep is the field symptom to watch between inspections.
Are grinder cleaning tablets enough on their own? They clean the path coffee takes and they are genuinely useful on designs that are hard to open. They do not reach the corners of the chamber or the underside of the carrier. If your grinder opens with hand tools, tablets supplement a physical clean rather than replacing it — and check they are approved for your burrs.
Should the grind setting be part of the recipe we send to sites? No. Send dose, yield and time. Grind settings are not transferable between grinders, and a circulated number encourages sites to match a dial position rather than a cup.
Do I need to re-dial after cleaning? Assume yes, then verify. Cleaning removes oil and fines that were affecting flow through the burrs, and any clean that opens the burr chamber may have moved the setting slightly. One shot and a timer settle it.
Who should be allowed to adjust the grinder? One named person per site, with a named deputy, and a written record of every change. Everyone else reports symptoms. This single rule recovers more consistency than most bars expect, because it converts four people's uncoordinated corrections into one person's tracked decision.
This article is part of the Vea Group knowledge base. Vea Group S.p.A. is an Italian designer and manufacturer of professional and premium coffee machines, with heritage dating to 1919 and production in Chignolo d'Isola, Italy, and Suzhou, China.