Slow feeding and RPM

Slow feeding

The mechanism. Dumping a whole dose floods the pre-breaker. Beans crack quickly and then queue at the narrow exit (about 500 to 600 µm), so grounds rub on grounds, heat up, build static and get mashed instead of cleanly cut 1:33. Longer time in the burrs makes particles smaller, so a dumped dose grinds finer and pulls slower than the same beans fed one at a time 18:55.

What changes when you slow-feed

• At the same setting the grind comes out coarser with fewer fines, so you have to go finer to get back to the same size. Clarity and distribution improve 8:39 8:42. Your grinder may even run out of fine range, which means dialling in again 9:13.
• The shift can be big. On a DF83V a fast-fed espresso took 24 to 25 s against about 10 s slow-fed at the same setting 7:38. On a DF64 Gen 2 it was 28 s against about 12 s 8:08.
• Auger-fed grinders (DF83, Timemore, Zerno, EK43) show a much smaller effect. On a Zerno, slow-fed shots were only 1 to 3 s faster. Lance Hedrick still finds the Niche's feed disc too fast 4:07.
• On a muddy filter cup from an espresso-oriented or bimodal grinder, his first suggestion is to slow-feed and then dial in finer 0:31.
• Feeding gradually gives more consistent grinds than pouring everything in 20:42. Time in the burrs shapes the distribution, which is why swappable augers matter 19:20. A tight, slow auger matters as much as variable RPM, while a big aggressive blade auger favours speed 16:21.
• Feed rate is also designed into burrs. Faster-feeding burrs stall more and make more fines on light roasts 13:30, and a faster feed likely muddies filter body 23:23. A flow-control disc above the burrs slows the feed, nearly stops stalling and reduces regrinding 9:41. On grinders without a pre-breaker, a slow acceleration ramp avoids beans bouncing when hot-loading 6:35.

Hand grinders. Hold the grinder close to horizontal so only one to three beans drop in at a time, and keep a steady cadence 5:38 5:08. Tilting slows the feed and makes the grind slightly coarser, so dial in again if you change your tilt 29:11. It's Lance Hedrick's preferred way to hand grind 18:17 41:09.

Stalls. Slow feeding is the fix for small or weak motors on dense, very light roasts at fine settings 14:24 11:54 4:05 12:48 27:27. Combine it with a hot start by running the motor before adding beans 32:16 6:08 10:06. Darker roasts are less dense and break more easily, so they stall less 13:18.

RPM

Cones: lower is generally better (consistent view)

• Lower RPM tends to raise clarity and shot quality on conical burrs. There's a big difference between 300 and 100 rpm, with diminishing returns below about 100 (maybe 80) 14:19.
• Fines production rises with RPM and slower rotation tightens the distribution on cones 2:05. On cones RPM has a big effect, and faster means more fines 9:09 4:31.
• Hand-grinding speeds (roughly 60 to 80 rpm by his estimate in 2021, 100 to 120 rpm in 2024) are linked to the clarity of good hand grinders 12:52 5:09 9:41 2:32. Low-RPM cones are easier to dial and reduce heat and fines 16:00, and cones at about 60 rpm give a pleasing distribution 4:05. He admits part of his preference for slower motorised cones may be in his head 18:39.
• Budget electric cones at 550 rpm and above are a recurring complaint (Turin SD40, Baratza). The slow, stepless budget grinders produced similar, nicer cups 4:10 25:30.

Flats: Lance Hedrick's view moved from "RPM shapes fines" to "RPM mostly moves the peak"

Manufacturer claims differ. Timemore cites laser data showing fewer fines at lower RPM on its burrs 4:36, and Bentwood fixed 1400 rpm as where coffees tasted best 34:30.

Torque, gearing and controllers

• Torque matters more than wattage, and a gearbox step-down gives more torque than a bigger motor 5:36 5:07 2:31. Low variable RPM stalls more unless geared 10:11.
• When a grinder stalls at low RPM, Lance Hedrick usually blames the controller board rather than the brushless motor 13:43.
• RPM works as a fine-tuning knob. Get close with the dial, then adjust RPM 12:41. He usually only needs low, medium and high presets 8:12.

Stated speeds (as reported)

James Hoffmann's view

• Slow feeding (2020, earlier than Lance Hedrick's 2024 video). Regrinding coarse grounds at an espresso setting choked the shot when fed fast but gushed (about 15 s) when fed very slowly, at the same setting 1:31 2:31. Feed rate dramatically changes the effective grind, and once dialled finer the slow-fed grind tasted sweeter 2:01 4:33. His theory is that congestion at the narrow burr exit keeps grounds in the burrs longer and regrinds them. Popcorning may actually be a better grind that's simply too coarse at that setting 3:33 4:03. Anti-popcorn discs may help partly because they slow the feed 8:42 5:04.
• His "grand universal theory" (2025). Where coffee jams up, usually at the finishing teeth, it stays in the burrs longer and gets cut repeatedly, and each extra cut creates fines below about 90 µm 17:30. Feeding single beans removes the bottleneck and gives a much coarser grind with fewer fines. Espresso ground that way at a normal setting gushes 20:01 21:03. Burr gap sets grind speed far more than auger feed rate 7:41.
• Hot start. He recommends starting most grinders before adding beans 3:35 3:34.
• RPM: open question, with one surprising result. He has repeatedly called variable RPM largely unknown territory with no clear guidance yet 6:05 4:37, and likens RPM profiling to early pressure profiling, a tool still looking for a use 3:04. On the Weber Key, higher RPM gave faster shots at the same setting (against expectation) and more texture, while lower RPM gave clearer, more linear cups 13:22 13:54. He grinds filter at lower RPM for clarity 20:04. He suggests RPM moves where the bottleneck sits, which might explain inconsistent RPM effects between grinders 19:01. Burr design drives grind speed far more than RPM 19:15.
• He saw no evidence that slower motors improve coffee, only that very cold beans grind better (2019) 9:13. Cheap electrics spin weak motors fast, which makes them loud and short of torque for espresso 11:48. Low RPM on dense light roasts needs a lot of torque and stalled the Key 11:19.

Daddy Got Coffee's view

• Pre-breakers and feed control reduce regrinding. When beans flood the burrs they spin faster than they can exit 14:52 15:23. The Mazzer Philos auger feeds big doses faster and small doses more gently, while the Zerno feeds consistently 8:43. A flow-control disc gives a more uniform grind 5:39.
• RPM varies by grinder. On the Aries conical, about 90 rpm tasted better for filter and higher rpm added body to espresso 11:14 12:14. On the Weber Key, max rpm ran shots up to 10 s faster with more body 10:17 10:49. On the Lagom P80 and Timemore 078S, higher or mid-range rpm worked best, while top rpm on the 078S was poor 6:14 7:14. Low rpm is stall-prone 7:40 6:46.
• Many flat-burr grinders run at around 1400 rpm 10:08.

Kyle Rowsell's view

• Lower RPM gives a more even distribution and clearer cups, which Kyle Rowsell links to hand grinders' strong results. Higher RPM gives more fines and body 5:16. A slow Varia VS3 was clearer than a faster Turin SD40 15:18. Changing RPM moves the distribution's peak, so expect to dial in again 12:49, which matches Lance Hedrick's later "RPM moves the peak" view.
• Hand speed matters. A Goat Story Arco cranked by hand runs at nothing like its motor base's speed, so numbers don't transfer and he went a little coarser by hand 5:05 5:35. How fast you crank changes the cup 8:36.
• An auger keeps the mass on the burrs steady however full the hopper is, avoiding popcorning 6:37, but on a DF83V dense light roasts overloaded it at low RPM, and running above about 1500 RPM avoided that 9:10.

The Coffee Chronicler's view

• Hand grinders spin at something like 60 to 100 rpm and budget electrics at 400 to 500, so hand grinders run cooler with fewer fines and cleaner cups 4:31. Newer slow electric conicals such as the Lagom Mini, MHW-3Bomber Sniper and Timemore Sculptor bring that to electric grinders 7:42, avoiding the regrinding typical of fast electric cones 5:13 3:36. This agrees with Lance Hedrick on cones.
• On flats, The Coffee Chronicler finds RPM changes the taste as well as the peak. With the same burrs, 800 to 900 rpm tasted cleaner and more distinct than 1400 in repeated blind tests 3:01. On one 2025 grinder, espresso was astringent at 1400 to 1600 rpm and balanced at 1000. At 1200 it was still harsh, at 900 clean yet clinical, and at 800 gentle and sweet but short on intensity, and about 800 suited filter 9:48 11:19 12:21. He sets a DF64V to 600 to 800 for filter 6:40.

Sprometheus's view

• Measured. Lower burr speed gave a tighter distribution with fewer fines 4:00. At matched espresso time and yield the slowest speed came out about 0.5% higher, whereas for filter the fastest speed drained slower and came out about 0.5% higher, as fines plugged the gaps 4:30 4:00. Higher speeds produced more heat in his test 6:33. A 530 RPM Niche Duo gave more clarity than a 1400 RPM DF83, and the same DF64V setting at 600 and 1800 RPM made very different shots 7:21 3:52. On a Lagom P80 analyser runs, higher RPM raised the median, cut the share in range and widened the spread 5:25. Sprometheus suggests 300 RPM for the clearest filter and 900 as an all-rounder 6:30.
• Feeding has no single rule. On a Zerno Z1, the slow auger gave a much tighter grind when beans went in a few at a time, while the medium auger did better with the whole dose dumped in. Match the feed to the auger 5:29 6:30 7:01. On the P80 at 300 RPM, tipping in the whole dose beat slow feeding 5:59. That complicates Lance Hedrick's and James Hoffmann's finding that slow feeding is better.
• The DF83V's pre-breaker auger let fewer whole beans through as RPM rose 4:52. Larger burrs only run cooler on big batches. With a single 20 g espresso dose the Lagom 01 ran a few degrees warmer than a P64 5:25.

Brian Quan's view

• RPM strongly changes fines and so the effective fineness, and high RPM makes many fines 9:17, and the same burr set can taste different at different speeds 13:17. Low RPM on an EG-1 made very good light-roast espresso but risked stalling 12:21. Variable RPM lets you grind slowly without stalling, then speed up to purge, and runs quieter than the same burrs at fixed full speed 11:45.
• Shot design should follow burr speed. Quick-extracting 98 mm flats and 64 mm MP burrs work best around 20 to 35 s, while traditional cones (a Niche Zero, a Lagom Mini) can run 35 to 45 s with longer pre-infusion 7:44.
• Gearing a hand grinder down (the Helor Flux needs three handle turns for each burr turn) lowers the effort enough to grind light roasts for espresso by hand 1:30, as Lance Hedrick says about gearing.

For Notchbook

Feeding style and RPM are hidden variables in a grind setting. If a user changes either (tilting a hand grinder, adding a slow-feed disc, changing an RPM preset), the same dial number can produce a noticeably different grind.

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
• Conical vs flat, and burr geometry
• Grind size and extraction
• Particle size distribution and fines
• Retention, static and RDT
• Seasoning new burrs
• 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.