Tamper, WDT and scale: what do they actually do?

Channeling is what happens when water finds the weak spot in your coffee puck and rushes through it, rather than flowing evenly through the whole bed (Clive Coffee). Water is lazy. It takes the path that resists least, the way a river keeps cutting the same bank.

What lands in the cup is easy to recognise. Part of the grounds gets over-extracted, because water ran through that channel the whole time. The rest stays under-extracted, because water barely touched it. Bitter and sour arrive in the same mouthful, and no grinder adjustment rescues that.

One number deserves correcting here, because marketing keeps muddying it. Consumer machines advertise pumps at 15 or 19 bar, but an espresso extracts at around 9 bar at the portafilter (the handled basket holder you lock into the group head). That pump figure is a peak, a sales number, not the working pressure. So a 19 bar machine compensates for nothing: at 9 bar, water is already hunting for a way through, and the puck alone decides whether it finds one.

Three separate faults open that route. A puck tamped off-level or at uneven density, which is the tamper's territory. Clumped grounds coming straight off the grinder, which is the WDT tool's territory. A dose that drifts shot to shot without you noticing, which is the scale's territory. That's why all three exist side by side. None of them covers for the others.

Start with the myth, because it's everywhere and it doesn't survive contact with a measurement.

The instruction to "press with about fifteen kilograms" runs through most coffee content in several languages, including MaxiCoffee's own Coffee Spirit blog, which prints it with no study and no manufacturer spec attached. Nobody seems able to trace where the figure came from. That isn't an accusation. It's just what the trail shows.

And when someone does measure, the measurement disagrees. A test of 40 shots on a La Marzocco GS/3 across four tamping forces (5, 10, 15 and 20 kg) found no statistically significant difference in extraction from tamping pressure at all (Socratic Coffee, 2015), with p=0.40 for the pressure factor. The only significant predictor of refractometer-measured TDS was the final beverage weight, at p=0.00.

Two widely quoted numbers, neither traceable, and one measurement that finds nothing. Barista tool maker Hauck Barista Tools puts it bluntly: "there is no ideal stamping pressure as a number. There is only the moment when the puck is stable." Which means hunting for the right weight in kilos is the wrong question in the first place. Don't swap 15 kg for some other figure. Drop the figure.

Level is what actually matters. A puck tamped two degrees off is thicker on one side than the other, so it resists more on one side than the other, and water works out which side to take almost instantly. A self-levelling tamper handles that mechanically. A spring stop releases at a set force, and the body of the tool rests on the rim of the basket so it can't tilt. Your wrist stops being the variable.

NEOUZA Tampeur Expresso Auto-Nivelant 58 mm is a good example of the type: its Amazon.co.uk listing states a 58 mm diameter and a self-levelling mechanism, checked on 08/08/2026.

Here's the detail almost nobody mentions. On a precision basket (IMS, VST, La Marzocco), where the internal diameter often reaches 58.4 to 58.6 mm, a 58 mm tamper leaves a 0.2 to 0.5 mm gap all the way round the edge (1988coffee.com). That uncompressed ring of grounds sitting against the wall is exactly where channeling starts. It's also why 58.5 mm tampers exist, and they wouldn't if the problem were imaginary. Measure your basket before you pick a diameter, because your machine's nominal 58 mm says very little about the basket you dropped into it.

Grounds leaving a grinder don't fall as an even dust. They come out carrying static and stuck together in small clumps, which gets worse with lighter roasts and home-grade burrs. Those clumps are denser than the bed around them. When water arrives, it goes around the hard part and straight through the loose part.

WDT stands for Weiss Distribution Technique, and it means stirring the dry grounds in the basket with fine needles before tamping. It didn't come out of a lab or a barista competition. John Weiss, who invented it, says he worked it out after concluding that "good grinds distribution in the basket was a critical factor in pretty extractions," and describes starting with a dissecting needle and a cut-down yogurt cup as a funnel (Daily Coffee News, 2022). A hobbyist, at home, around 2004. Some baristas simply call it wdt coffee prep, same motion, different name.

How deep you go matters more than how long you stir. A study comparing surface WDT against full-depth WDT measured roughly 40% shot-to-shot variation in peak puck resistance from surface stirring alone, with that scatter clearly reduced when the needles reached the bottom of the basket (Coffee ad Astra, 2021). Reduced, not erased. The same work, run over 24 shots with pressure and flow sensing, records that the variation persists. WDT makes your next shot resemble your last one. It doesn't make the machine infallible.

So raking only the top does close to nothing useful. The zone that decides where water travels runs through the full thickness of the bed, and a clump sitting three millimetres off the bottom will steer water down there just as reliably as one on the surface.

Needle gauge is a secondary question, honestly. Thick needles push grounds into heaps instead of separating them. Around 0.3 mm they pass through without displacing much. None of the sources gathered here measures the effect of gauge, so treat it as a comfort preference rather than a decisive spec.

Normcore WDT Outil de Distribution V2 (9 aiguilles) fits that description: nine needles at 0.3 mm according to its Amazon.co.uk listing, checked on 08/08/2026.

A shot of espresso isn't a uniform liquid. It's liquid topped with crema, a foam whose volume shifts with bean freshness, roast level and pressure. Two shots filling the cup to the same line can hold noticeably different masses of liquid. That's why the marks on a shot glass are close to useless, and why specialty coffee counts in grams.

The brew ratio is just the relationship between the mass of dry coffee going in and the mass of drink coming out. The espresso ratio convention sits around 1:2, so 18 g of grounds for 36 g in the cup (Wikipedia, Coffee extraction), the same reference that reports the Specialty Coffee Association's target window: 18 to 22% extraction yield at 1.15 to 1.35% strength. Under 18% the cup reads sour and hollow. Over 22%, bitter and drying.

The point of a scale isn't copying a recipe. It's making your problem diagnosable. If 18 g goes in and 36 g comes out in fourteen seconds, you know bed resistance collapsed somewhere, and you look at puck prep. If those same 36 g take fifty seconds, you look at the grind. Without both numbers you only have an impression, and an impression can't be corrected.

The physics backs that logic into slightly counter-intuitive territory. A study published in the journal Matter (Cell Press) showed that grinding too fine can actually clog the coffee bed and hurt extraction consistency rather than improving it (ScienceDaily, 2020). The team even recommended dropping the espresso dose and grinding coarser, using 15 g instead of around 20 g in their worked example, to gain consistency. Note carefully what that paper covers: grind and dose, not tamping and not WDT. It shouldn't be dragged into a tamper argument.

Two specs matter on the hardware and the rest is comfort. Resolution to 0.1 g, because a whole-gram kitchen scale hides up to a 3% dose error. And a built-in timer, so weight and time get read off the same device at the same instant.

BAGAIL Balance Numérique à Café avec Minuterie, Précision 0.1 g covers both: its Amazon.co.uk listing states 0.1 g resolution and an integrated timer, checked on 08/08/2026.

The order below follows the causes, not a shopping list.

Measuring comes first. Until your dose in and yield out are numbers written down somewhere, you can't evaluate any other change, because you'll never know whether the improvement came from your technique or from half a gram of extra coffee.

Distribution comes next. WDT works on the bed before it gets compacted, so it sits upstream of the tamper. A flawless tamp on clumpy grounds gives you a perfectly flat puck that's perfectly irregular inside.

Tamping lands last, and it asks for two things: the same force every time, and a tool that won't tilt. Everything else, grind fineness, temperature, pre-infusion time, belongs to dialing in, the phase where you tune the machine to one specific coffee. There's a dedicated guide covering the seven settings that make up that process, and the "which one should I buy, in what order" question has its own barista accessory comparison. This piece answered something narrower: what each tool fixes, and why.

None of these three tools makes coffee better by magic. They correct three different ways tamping espresso can go wrong, and they don't substitute for one another. A scale won't straighten a crooked puck, and a self-levelling tamper won't untangle clumped grounds.

If you're still weighing a portafilter machine against an automatic one, the tamping question changes shape entirely, since the machine handles it internally. The bean-to-cup machine comparison covers that decision properly, with the grinder and the grind as the lead criteria rather than the pressure figure printed on the box.