Conical vs flat, and burr geometry

The short version

Lance Hedrick started out repeating the common rule of thumb (cones are chocolatey, flats are clear) and spent 2022 to 2026 taking it apart. These days he sees burr shape as a weak predictor. Spoke count, feed rate, finishing-tooth design, burr size, RPM, drivetrain and alignment all matter more, and each type can reach into the other's flavour territory 0:30 6:13.

How Lance Hedrick's view changed

The Eureka Silencio is a flat that behaves like a cone, sitting at the chocolatey, coating end of his spectrum 3:32.

How burrs break a bean

• Three phases. Pre-breakers crack the bean, then a crushing zone, then a cutting/finishing zone. Changing the setting moves the burr or collar and changes the gap between the finishing teeth 6:36. Burrs cut in the early teeth and crush in the flat final zone 9:14.
• Spoke count sets feed rate. More spokes leave smaller gaps and feed beans more slowly. Fewer spokes pull beans in faster, demand more torque on light roasts and are better matched to darker, softer beans 7:07 4:07 6:44. On cheap hand grinders, pentagonal burrs feed fast and makes more fines for dark roasts, while heptagonal ones feed slower with fewer fines for light-roast filter 5:06. Fewer pre-breakers also make dense light roasts much harder to crank 5:38. Even so, a seventh spoke does not by itself guarantee a better spread 24:36.
• His ranking of 38 mm-class hand-grinder burr styles puts heptagonal multi-purpose first, then hexagonal, then slice-to-cut heptagonal, then pentagonal 10:42.
• Same geometry, bigger size means bigger side gaps, faster feed and more fines (the Comandante C60 against the C40, for example) 20:20 18:50.
• C40 lineage. Many hand grinders sold on clarity use a version of the Comandante C40 geometry, which feeds slowly and suits medium-light roasts 25:00 4:35. Kingrinder P2/K, Comandante, Lume Mini and 1Zpresso Q2 heptagonal burrs are nearly identical 11:13.
• Etzman burr trio. Dark-roast burrs have fewer cutting surfaces and feed faster. Light-roast burrs feed slower so dense beans don't jam. Filter burrs have long straight cuts to feed slowly and cut evenly 3:01.

Flat-burr design levers

• Flattening the finishing triangle shifts a burr towards espresso. Steepening it shifts towards filter 3:35.
• More pre-breakers and a shallower angle feed faster and make more fines, with harshness in the cup 12:58. Burrs with a few large pre-breakers skip a stage and pull beans in fast 28:07. Many small pre-breaker steps, or big pre-break steps followed by a short final stage, both make fines. Going straight into a second crushing phase tends to make fewer 11:12.
• Two-phase burrs with no pre-breakers (such as brew burrs) feed slowly and suit light roasts 9:41. SSP Brew burrs have an aggressive outfall that gives very low fines and high clarity, but espresso needs an uncomfortably tight setting 17:10.
• Burrs with large, non-flush finishing teeth (Fellow Ode Gen 1 stock) make few fines but can't grind fine, so single cups under-extract 11:48.
• Shallow, uneven grooves on cheap burrs produce lots of fines at filter sizes 13:23. Dull, curved burrs mash beans 7:40. Big gaps squash rather than crack 10:13. Ceramic dulls and crushes, steel stays sharp 9:14.
• Ghost burrs slice beans into chunks that drop through the gaps instead of squeezing them through progressively finer teeth. They rely on alignment and can't reach espresso 0:30. Squared tips get closer and finer than pointed tips 5:36. They give a large fines hump and unimodal boulders, which taste hollow to him with light roasts, though some like them with dark 12:15 22:33.
• Hybrid burrs put a cone-like cracking stage above a flat finishing stage 12:15.
• Cast vs machined. Cast burrs differ from batch to batch, so you may or may not get a good set. Machined versions are more uniform 15:41 0:33.
• Coatings (DLC, Red Speed, Silver Knight). Lance Hedrick sees them as mostly about burr life and friction, treats flavour claims as guesswork, and favours uncoated versions 16:47 14:31. He did note Red Speed lengthening drawdown and muting some acidity at the same grind 30:08.
• Specialisation. A burr set designed for one job beats an all-rounder at that job 6:07. All-round burrs do everything adequately and excel at nothing 46:38.
• Humility. He repeatedly says nobody really knows burr geometry. Designs iterate on existing patterns, small changes can have big effects, and tweaks don't always transfer between diameters 6:07 4:36 2:32 6:36.

Size, blind burrs and augers

• Size. Bigger burrs give more cutting area, better feed control, less congestion and less heat, so a higher ceiling 10:13 1:36 22:53. But bigger isn't automatically better, and a well-made small burr can beat a poor large one 12:15 3:05 6:10. Nominal sizes aren't comparable between brands 5:06 50:28.
• Blind (screwless) burrs add cutting surface, reduce retention and grind more evenly 13:15 16:21. At 64 mm they have roughly the surface of a 75 mm burr and need about 20% more power 23:26. Screw holes hold coffee and disrupt cutting 14:13.
• Pre-breaking augers. An auger only pre-breaks if it sits close to the inlet walls so beans crack against them 18:50. Done well, it makes a 64 mm burr behave more like a larger one 9:13, improves the cup over dropping beans straight in 10:45, feeds evenly and reduces heat 11:08 3:07. If the auger pre-breaks, the burr should drop its own pre-breaker teeth and use that area for crushing 4:38 14:46. Big burrs need pre-breaking. A cone-shaped pre-breaker splits beans into big pieces with few fines, which can then swamp the finishing teeth 15:44.
• Wide feed aperture (SSP multi-purpose against a Mazzer I200D, for example) brings beans in faster, loosening the distribution and adding heat 18:22.
• Magnetic burr mounting is criticised by some professionals because grounds can creep behind and misalign the burr 18:38.

Burr sets Lance Hedrick talks about

James Hoffmann's view

James Hoffmann holds the conventional generalisation more firmly than Lance Hedrick does, while also stressing the exceptions.

• Broadly, flat burrs give a more uniform grind than conicals, with exceptions 2:05. Cones generally produce more fines, which brings body and texture to espresso at some cost to clarity, which is why most people prefer flats for filter 1:32 17:57.
• Shape alone doesn't decide the profile, though. The Niche Duo's flat espresso burr produced a near-conical distribution and tasted indistinguishable from the conical Niche Zero 2:03 9:45. In his 2023 test the Lagom Mini measured more like a flat than any other cone, and the Mahlkönig X54 more like a cone than any other flat 24:38. The Fellow Opus and Encore ESP also broke the "conical = heavy" expectation 11:15.
• He felt no loss of filter transparency from a Niche Zero, though side by side a flat-burr Wilfa Uniform was cleaner and sweeter and the Niche more textured 6:36 9:41.
• Cheap grinders use cones because conical burrs offer lots of cutting surface at low speed, so weaker, cheaper motors cope. Flats need stronger motors 2:34 9:43.
• Interlocking (ghost-teeth) burrs touch before the gap is very small, so their finest setting is relatively coarse. This limited the original Fellow Ode and a cheap XeoLeo 16:56 2:01.
• Screwless mounting leaves the cutting surface uniform all the way round 7:39. Small chips or screw holes in the coarse cutting zone probably matter little. Damage to the finishing teeth is the bigger concern 5:06.
• Burrs matter more than the body. The same burr set in a Zerno and in a Fellow Ode gave no tastable difference 27:14. Swapping burr sets changes taste enormously 9:06. Similar burrs at the same size and speed (DF64 and Eureka Oro) gave near-identical distributions 24:08.
• Cheap hand grinders tend to combine ceramic burrs, which cut less cleanly, with a poorly supported lower burr that wobbles. Around £100 buys metal burrs and steadier construction 8:45.
• Roller mills (industrial, multi-stage) can reshape the distribution, moving the target peak while cutting fines 9:44.

Daddy Got Coffee's view

• Daddy Got Coffee uses the broad generalisation, with conicals more chocolatey, blended and forgiving and flats clearer, brighter and narrower to dial, while stressing the overlap 11:43 6:12 19:28 18:56. Many newcomers find entry flat-burr grinders fussier than expected 18:56 10:38.
• Conical sizes are misleading. They're quoted by outer housing diameter. A "48 mm" Sniper burr is functionally about a 38 mm Timemore burr, and a VS3's 48 mm burr has an inner of about 30 mm 8:13 1:50 1:32.
• Why flats are rare in hand grinders. Flats are built to spin fast and fling grounds out. Cones suit low speed and let gravity do the work 2:31 0:30.
• Unimodal and bimodal burr sets. A bimodal burr deliberately makes fines plus boulders for espresso and filter. A unimodal burr aims to remove fines 4:33 5:05. A burr without big pre-breakers feeds slower and grinds more uniformly 6:10.
• SSP 64 mm burrs. Brew is very unimodal and his favourite filter burr. MP is the Brew geometry altered for more fines, with edgier filter and clear espresso. HU is blended and easy for espresso, and Cast V3 is his favourite espresso taste, though it needs very fine settings and suits levers 3:04 4:35 3:35 6:05 5:05.
• Platforms. 64 mm is a standard size, so burr swaps change the profile without a new grinder. 64 mm has more options than 80 mm 10:43 23:00. Blind burrs free almost 9% more cutting area 11:49.

Kyle Rowsell's view

• Kyle Rowsell uses the familiar split, where flats pull flavours apart for clarity and transparency and conicals fold them together into layered, chocolatey, textured cups 7:20 15:51 16:49. In a blind tasting the flat-burr Fellow Ode was easy to spot for that reason 4:42, and he finds some flats one-dimensional 7:20.
• Like Lance Hedrick, though, he finds the particular burr set matters more than its shape. Two flat sets can taste nothing alike 6:11.
• Burr purpose. Espresso-tuned burrs produce surplus fines, costing clarity in light-roast filter, and a more unimodal aftermarket set gave clearer cups 6:06. The 1Zpresso J-Max burr sacrifices some clarity to gain sweetness and complexity 4:34. SSP high-uniformity burrs suit traditional espresso but tasted astringent and blended for light filter 17:27, while the 98 mm high-uniformity set matched 64 mm unimodal clarity with more sweetness 10:11.
• Size isn't everything. In his 2023 big-burr comparison an auger-fed 64 mm Zerno Z1 gave better cups than 83 mm sets in a DF83 and Niche Duo, which turned astringent from about 22 to 23% extraction 23:35 3:38. Bigger conicals grind faster and heat the coffee less, which can shift the profile 3:41.
• Coated burrs grabbed beans a little differently and made slightly more fines than polished ones, without changing the overall distribution style 9:41. Flats for espresso need a strong motor to grind fine, echoing Hoffmann's point about motors 1:31.

The Coffee Chronicler's view

• The Coffee Chronicler credits the burr set with roughly 80 to 90% of what a grinder tastes like, the rest of the machine mostly shaping workflow 0:30 1:32, though by 2026 he adds that speed, burr orientation and any auger shape taste together with the burr 2:01.
• Geometry beats diameter, a theme he returns to often. 50 mm Eureka Filtro burrs out-sieved bigger sets 24:08, 54 mm sets can match 64 mm 14:03 and even beat some 98 mm sets 12:05, a Wilfa Uniform fitted with 58 mm SSP Brew burrs frequently out-tasted the 64 mm MP 4:38, and 83 mm adds only a little over 64 mm 10:28. Sprometheus reached the same conclusion with measurements (below).
• Flats have the higher ceiling for flavour purity while conicals give texture and body 10:02. Flats generally make better filter and suit modern light-roast espresso, and premium conicals chase espresso body and sweetness 6:54 9:12. A grinder that separates filter flavours sharply can taste hollow as espresso 8:46.
• Hand conicals are easier to set up and accept a wider band of filter grind sizes than flats 21:00. Good heptagonal-style hand burrs cluster closely in taste 7:42. Heptagonal burrs produce a tighter spread than pentagonal ones and are slow for espresso and better for filter, in line with Lance Hedrick's spoke-count point 2:31.
• The Millab M01's long "pineapple" burr with two pre-breakers grinds in stages, and the effect seemed genuine to him 3:56. On the Mx. Cool Aries the outer burr spins, so there is no chute or impeller and less retention and clumping 1:33. Burrs that look alike can still behave very differently 8:43, while Timemore's distinctive Spike-to-Cut design made no obvious real-world difference 3:37.

Sprometheus's view

• How each type breaks a bean. A conical drags beans down with its teeth and gravity and crushes them in stages. A flat throws them outwards to cut and chip, which changes particle shape and leaves more grounds inside 5:04 3:04. Conicals spin slower on smaller motors, so they run quieter, cooler and cheaper, while flats need about 700 RPM and bigger motors 2:03, which agrees with Hoffmann. That low-speed crushing is why conicals suit hand grinders 3:15, as Daddy Got Coffee also says.
• In the cup, flats are brighter and cleaner. Conicals come out ahead on mouthfeel, complexity and a broader sweet spot 6:18 6:43.
• Same geometry, different size. A DF64P and a DF83 with the same steel burr design gave near-identical espresso when time and yield were matched, and only grinding speed changed 7:06. A DF64V and a Lagom P64 with identical burrs and RPM also tasted the same, the differences being build and motor 7:06. Larger burrs add cutting area, so they finish sooner and offer a wider usable dial, without changing the cup on their own 3:10 2:45. He preferred the Lagom 01's cup for its geometry, not its size 8:00. This lines up with the Coffee Chronicler and with Hoffmann's "burrs matter more than the body".
• Blind burrs. Without bolt holes a 64 mm set gains about 8% surface area 2:08 (Daddy Got Coffee says almost 9%). Heat and speed barely changed in his test, but cups were slightly clearer with fewer defects and extraction rose a little. They're a nice refinement, but you don't need them 2:38 5:24 5:59.
• Eureka's classic bimodal flat geometry gives a broad spread, with body and balance at the expense of clarity 4:52, a reminder, as Lance Hedrick says, that flats aren't all clarity burrs.

European Coffee Trip's view

• Comandante's three Lab Series burrs (Nitro Blade, Tiger Shark, Hammer Head) differed mostly in feed speed and feel during short side-by-side grinding rather than in visible uniformity 10:06.
• The C60 Baracuda puts larger burrs into a C40-sized body and grinds much faster 4:28, consistent with Lance Hedrick's note that a bigger version of the same geometry feeds faster.

Brian Quan's view

• Talking with Denis Basaric of Kafatek, Brian Quan gives the usual split (flats clearer with sharper flavour peaks, conicals smoother and more chocolatey) but says modern sharp-cutting conicals handle light roasts and flats can do dark ones. The gap has shrunk, and the real differences are now workflow, cost and the extremes (maximum extraction on flats, very dark roasts on conicals) 2:04 6:41. That sits close to Hedrick's later view and is looser than Kyle Rowsell's.
• On one coffee and one profile, a flat exaggerated the fruit more than a conical, which he calls a difference of style rather than quality 1:30. Pushing very fast flow through conicals tends to leave a harsher finish than a good flat does 2:34 17:21.
• Big conicals (63 mm Niche, 68 mm Pharos, 71 mm Helor Flux) taste much alike at normal ratios. Only very long ratios show the larger set's extra body 9:47 2:31. Kafatek's 71 mm Shurikone lifts clarity and finish above the stock Mazzer cones while keeping conical texture 4:36. Denis curved its grinding path so it is as long as a 120 mm burr's 1:00.
• SSP 64 mm naming. High uniformity suits 1:2 traditional shots and the unimodal sets suit long 1:3 shots 4:34. MP is very clear and quick to extract with thin espresso body, and HU sits between MP and a Niche Zero, nearer the Niche 1:02 4:07. Coatings such as red speed say nothing about geometry 0:00. The 98 mm high-uniformity set behaves like a louder, cleaner 64 mm MP 6:21 7:34. This runs in the same direction as Daddy Got Coffee's SSP breakdown.
• Forgiveness. Ultra-clear flats such as 98 mm SSP Brew magnify roast and brewing mistakes, while Kafatek's Shuriken burrs have a soft, silky finish that hides them 3:35 2:34. Burr sets designed in China often lean towards sweetness and blended acidity 4:34. Uncoated Moonshine burrs on a Lagom Mini produce fewer fines than the coated Obsidian set, so you grind finer and get a clearer cup 4:04, which agrees with Kyle Rowsell's coated-burr observation.
• Flats cost more to make because they need extra machining, and carrier tolerances limit how well a flat set performs in a different housing 9:46.

For Notchbook

Because two grinders with the same nominal burr type can need very different settings, Notchbook's per-grinder history is the right model. Burr type alone isn't enough to transfer a setting.

Summarised in our own words from the creators' videos, with every point linked to the moment it comes from. Notchbook isn't affiliated with or endorsed by any of these creators, so please watch the originals for full context. Sources

Keep reading

• Burr alignment
• Astringency
• Grind size and extraction
• Particle size distribution and fines
• Retention, static and RDT
• Seasoning new burrs
• Slow feeding and RPM
• Zero point and grinder scales
• Grinder workflow: single dosing, WDT, RDT, purging and storing beans
• Troubleshooting: what to change when a cup tastes wrong

Want new Notchbook builds first? Join the beta on iPhone or Android.

Notchbook is a free app that remembers the grind setting that worked for each coffee, grinder, brew method and filter paper, and keeps a journal of every brew. Get it free on Google Play. The iPhone app is coming soon. More about Notchbook.