Hulling and dry milling
The dry-mill steps that turn dried parchment or dried cherry into export-ready green coffee – hulling, polishing, grading, sorting and bagging.
Across the chain: Equipment
Contents
In brief
- Hulling removes the dried parchment, or the whole dried husk of a natural, to free the green bean.1,3
- After hulling, the coffee is graded, sorted, polished, cleaned and bagged.1
- The parchment is about 20% of the weight of dry parchment coffee; a dried cherry gives about half its weight as green coffee.10,11
- Coffee is hulled at about 11–12% moisture; over-dry beans are brittle and break, and too-wet beans can be crushed.2,3
- Hulled coffee keeps less well than coffee stored in parchment or cherry, partly because of hulling damage.13,14
What the dry mill does#
Drying ends with coffee that still wears its outer layers. Washed coffee leaves the patio as parchment coffee; a natural leaves it as dried whole cherry, called husk coffee.1 The dry mill turns both into green coffee.
Hulling is the removal of the dried endocarp of parchment coffee – the parchment – to produce green coffee.1 For parchment coffee, the huller takes off the parchment and silverskin.2 For dried cherry, the dry pulp and the parchment come off in a single operation.3 That dried husk is made up of skin, pulp and parchment.12
Hulling is done by machine, or on small farms with a wooden mortar and pestle.4 In Kenya, central private hulling companies hull farmers’ coffee for a fee; hulling plants and equipment are expensive and technically demanding.3
The sequence#
The Codex code of practice lists the steps after drying: the coffee can be stored, then hulled, and then passed through sizing (grading), sorting, polishing, cleaning and bagging.1 In practice the order varies by mill.
- Reception and storage – dry coffee is received and often held in parchment or cherry until it is sold.1,14
- Hulling – the parchment or dried husk is broken and separated from the bean.1,3 In friction hullers an air current carries the loose shell away.7
- Polishing – residual silverskin is removed by purely mechanical means.1 Some silverskin already comes off in the huller; if it does not, a polisher is used.4
- Grading and sorting – beans are separated by size, density and colour, by machine and by hand; hand sorting is still widely used.4 Sorting removes foreign matter, fragments and defective beans.1
- Cleaning and bagging – the finished lot is cleaned and bagged for export.1
Screen sizes, defect counts and sorting machines are covered in grading and sorting. This entry deals with the milling itself.
Hullers#
The sources opened for this entry describe two main designs.
Ribbed-cylinder (friction) hullers. An 1890 US patent by Evaristo Conrado Engelberg describes a ribbed cylinder turning inside a sectional casing, with an adjustable hulling plate and a slotted screen.6 The ribs carry the grain continuously past the plate and towards the discharge end.6 The gap between plate and cylinder is set wider at the feed end than at the delivery end.6 The machine was built for rice, but the inventor reports using it for coffee by widening the plate’s range of adjustment.6 A 1912 coffee huller and polisher uses the same idea with ribs that are partly helical: the ribs hull the coffee and feed it forward, and an air current removes the loose shell.7
Rubber-roll hullers. A Philippine national standard for coffee hullers states that rubber-roll hullers shall be used for wet-processed coffee.5 The same standard also lists steel hullers as the other type, and defines a coffee huller as a machine that removes the hull from dried parchment coffee or dried cherry.5
Yield: how much green coffee comes out#
Mills and buyers work with conversion factors.
- Parchment to green. The husk is about 20% of the weight of dry parchment coffee, a ratio of 1.25:1.10 Cenicafé found the same 1.25 ratio in Colombia (1.26 for unselected coffee), a conversion factor of 0.80 (0.79 for unselected coffee), and reports hulling losses of 17.75–18.40%.9 The trade factor multiplies parchment weight by 0.8 to get green weight.11
- Dried cherry to green. The standard ratio is 2:1.10 The trade factor multiplies dried-cherry weight by 0.5.11 Dry processing gives about 1 kg of husk for every kilogram of coffee bean produced.12
These are rounded standards. Specialty chains that sort hard lose more: all five Ethiopian cooperatives in one study had post-farm-gate conversion rates, from fresh cherry to clean coffee, of 8:1 or higher (up to 11:1) when producing specialty coffee.10 The authors note that the cooperatives did not keep records that would allow these rates to be measured precisely.10
Reference values#
The table collects the values reported in the sources. They describe practice and standards, not a single correct setting.
| Parameter | Value reported | Source |
|---|---|---|
| Moisture at hulling | About 12%; “fully dry” when the green bean is translucent jade green | FAO Lao PDR2 |
| Moisture after drying, before storage and hulling | 11% (Kenya, dry process; São Paulo manual) | FAO Kenya3, Codeagro14 |
| Moisture, storage safety | < 12.5% (wet basis), aw 0.67–0.70 for parchment | Codex1 |
| Moisture, export | Below 12% (wet basis), dried uniformly | FAO15 |
| Parchment share of dry parchment | About 20% (1.25:1) | Schuit et al.10, Cenicafé9 |
| Hulling loss, parchment | 17.75–18.40% (older reference: 18.00%); basis not stated | Cenicafé9 |
| Green per kg parchment | 0.8 kg (Cenicafé: 0.80, unselected 0.79) | ITC11, Cenicafé9 |
| Green per kg dried cherry | 0.5 kg (2:1) | ITC11, Schuit et al.10 |
| Husk from dry process | About 1 kg per kg of coffee bean produced | Eckhardt et al.12 |
| Huller standard: hulling recovery | Minimum 80% (dried parchment), 40% (dried cherry) | PNS/BAFS/PAES 2125 |
| Huller standard: hulling efficiency | Minimum 95% | PNS/BAFS/PAES 2125 |
| Huller standard: mechanically damaged beans | Maximum 10% | PNS/BAFS/PAES 2125 |
| Huller standard: purity | Minimum 97% | PNS/BAFS/PAES 2125 |
| Bean temperature in the huller | No limit found in the sources opened; heating should be avoided | Haile & Kang4 |
The Philippine standard defines hulling recovery as the weight of green coffee at the green-coffee outlet divided by the weight of dried parchment or dried cherry fed in, and counts broken or scratched beans from hulling as mechanically damaged.5
Data table
| % | Source | |
|---|---|---|
| Parchment, standard ratio 1.25:1 | 80% (Schuit et al.) | 10 |
| Parchment, trade factor 0.8 | 80% (ITC) | 11 |
| Parchment, Cenicafé measured conversion factor | 80%; 79% unselected (Cenicafé) | 9 |
| Parchment, huller standard minimum recovery | ≥ 80% (PNS/BAFS/PAES 212) | 5 |
| Dried cherry, standard ratio 2:1 | 50% (Schuit et al.) | 10 |
| Dried cherry, trade factor 0.5 | 50% (ITC) | 11 |
| Dried cherry, huller standard minimum recovery | ≥ 40% (PNS/BAFS/PAES 212) | 5 |
For parchment, 1.25:1 means 80% green; Cenicafé’s measured conversion factors of 0.80 and 0.79 mean 80% and 79% green.9,10 For dried cherry, 2:1 means 50% green.10,11
Mechanics of friction hulling#
A ribbed-cylinder huller rubs the parchment off by friction rather than cutting it.4 The ribs co-operate with a stationary hulling plate to remove the hulls, and the plate can be adjusted, one side more than the other.6 Engelberg set the gap wider at the feed end and narrower towards the delivery end.6 Broken husk leaves through a perforated screen whose slots taper or flare outwards.6 The same machine carried a polishing section with flexible leather fingers acting as whips on the grain.6
The 1912 design adds two things that matter for coffee. Helical ribs move the coffee forward while it is hulled, and an air current draws the loose shell and dust away before the beans pass to a second casing, where air again separates the remaining shell.7 Cool air is also let in between the rotating sleeve and the hollow cylinder to cool the working parts.7
Heat. A review notes that hullers which rub the parchment off by friction might create heat.4 It warns that beans should not be heated during hulling because heat affects the colour and taste of the coffee.4 No source opened for this entry measured bean temperature in a huller or set a limit, so no number is given here.
Polishing and silverskin#
Polishing removes the residual silverskin from green coffee by purely mechanical means.1 A 1991 patent for a polishing machine gives two reasons for doing it: silverskin left on the bean can burn during roasting and give a bad smell to the roasted coffee, and patches of silverskin can make colour sorters reject sound beans as defective.8 These are the patent’s stated reasons; no independent study opened for this entry tested them.
No study opened for this entry measured the heat produced by polishing or compared the cup of polished and unpolished lots from the same coffee.
Risks and defects#
- Broken and crushed beans. A badly set huller or over-dry, brittle coffee damages beans; coffee that is too wet can be crushed.2 Over-dried coffee has a faded colour and damages easily in hulling.2 A São Paulo good-practice manual likewise warns against over-drying to avoid weight loss and broken beans at hulling.14 A huller standard allows at most 10% mechanically damaged beans.5
- Heating. Heat during hulling can affect colour and taste.4
- Loss of shelf life. Hulled coffee loses quality faster in storage. In a Brazilian trial with Arabica (‘Catuaí Amarelo’) dried to 11% and stored in permeable bags, hulled natural and washed coffee kept its initial quality for up to 12 months at 10 °C and 50% relative humidity, but only up to 3 months at 25 °C.13 The authors attribute part of the loss to mechanical damage from hulling and to the missing fruit layers.13 A São Paulo good-practice manual therefore advises storing coffee in cherry or parchment rather than hulled.14
- Residual silverskin. A polishing-machine patent states that silverskin left on the bean can burn in the roaster and cause off-odours; this has not been tested in the sources opened.8
- Defective beans passed through. Black and sour defects carried the highest ochratoxin A levels in lower-grade coffee, so defects should be removed or screened before they enter the food chain.1
- Hulling losses. In Kenya, competition between private hulling plants reduced hulling losses and charges.3
Measurement#
The Codex code expresses moisture on a wet basis and asks for moisture meters to be calibrated against the ISO 6673 method.1 Huller performance is measured as hulling recovery, hulling efficiency, purity and the share of mechanically damaged beans.5 In storage trials, electrical conductivity and tetrazolium tests have been used alongside cupping to track damage to the bean.13
Open questions#
- How hot beans get in friction hullers and polishers, and at what bean temperature cup quality suffers. No measured values were found.
- Whether polishing changes cup quality once roasting differences are controlled.
- How moisture at hulling, within 11–12%, changes the share of broken beans.
- How much the moisture content of green coffee shifts between hulling and bagging.
Key studies#
Abreu et al., 2023 – Hulling and storage. Natural and washed Arabica dried to 11% moisture were stored for 12 months either hulled or unhulled, refrigerated or at 25 °C. Refrigeration preserved quality; at room temperature hulled coffee held its quality for only about 3 months, and the authors link part of the loss to hulling damage.
Montilla-Pérez et al., 2008 – Conversion factors in Colombian processing. Cenicafé measured physical properties and conversion factors at each stage of wet processing. The parchment-to-green ratio came out at 1.25, as in the older reference, with hulling losses of about 18%.
Schuit et al., 2021 – Specialty coffee economics in Ethiopia. A study of five Ethiopian cooperatives that also records the conversion ratios used along the chain. It gives the standard 2:1 for dried cherry and 1.25:1 for parchment, and shows that specialty sorting pushes overall cherry-to-export ratios to 8:1 or more.
- Parchment-to-green conversion: the standard is 1.25:1, which means the parchment is 20% of the weight9,10, and Cenicafé measured the same ratio (conversion factor 0.80; 0.79 for unselected coffee). The same Cenicafé table reports a hulling loss of only 17.75–18.40%; the report does not state the basis of that loss figure, so the two values cannot be reconciled from the source.9
- Moisture at hulling: FAO guidance for Lao PDR treats coffee as fully dry at 12%2, while Kenyan dry-process practice and a São Paulo good-practice manual dry coffee to 11% before storage and hulling.3,14 No source opened for this entry tests which moisture minimises hulling damage.
Coffearium (2026). Hulling and dry milling. Coffearium – The open coffee compendium. Retrieved 22 October 2026, from https://coffearium.org/entries/hulling-and-dry-milling/
@misc{coffearium2026-hulling-and-dry-milling,
author = {{Coffearium}},
title = {{Hulling and dry milling}},
year = {2026},
howpublished = {Coffearium -- The open coffee compendium},
url = {https://coffearium.org/entries/hulling-and-dry-milling/},
urldate = {2026-10-22},
note = {Retrieved 22 October 2026}
}Entries are updated when the evidence changes; the retrieval date tells your readers which version you used.
Sources
- ACodex Alimentarius Commission (FAO/WHO) (2009). Code of Practice for the Prevention and Reduction of Ochratoxin A Contamination in Coffee (CAC/RCP 69-2009). Codex Alimentarius.
- AWinston, E.; Op de Laak, J.; Marsh, T.; Lempke, H.; Chapman, K. (2005). Arabica coffee manual for Lao-PDR. FAO Regional Office for Asia and the Pacific, Bangkok.
- AMutua, J. (2000). Post Harvest Handling and Processing of Coffee in African Countries, Annex 5: Coffee Post Harvest Handling and processing in Kenya. FAO.
- BHaile, M.; Kang, W. H. (2019). The Harvest and Post-Harvest Management Practices' Impact on Coffee Quality. Coffee – Production and Research, IntechOpen. doi:10.5772/intechopen.89224 · other copies
- ABureau of Agriculture and Fisheries Standards (BAFS), Department of Agriculture, Philippines (2017). PNS/BAFS/PAES 212:2017 Agricultural Machinery – Coffee Huller – Specifications. Philippine National Standard.
- AEngelberg, E. C. (1890). Machine for hulling, cleaning, and polishing rice (US Patent 424,602). United States Patent Office.
- AOkrassa, R. (1912). Coffee huller and polisher (US Patent 1,035,631). United States Patent Office.
- CSatake, T. (Satake Engineering Co., Ltd.) (1991). Coffee bean polishing apparatus (US Patent 5,076,157). United States Patent and Trademark Office.
- AMontilla-Pérez, J.; Arcila-Pulgarín, J.; Aristizábal-Loaiza, M.; Montoya-Restrepo, E. C.; Puerta-Quintero, G. I.; Oliveros-Tascón, C. E.; Cadena-Gómez, G. (2008). Propiedades físicas y factores de conversión del café en el proceso de beneficio (Avances Técnicos 370). Cenicafé (Colombian Coffee Research Centre).
- ASchuit, P.; Moat, J.; Gole, T. W.; Challa, Z. K.; Torz, J.; Macatonia, S.; Cruz, G.; Davis, A. P. (2021). The potential for income improvement and biodiversity conservation via specialty coffee in Ethiopia. PeerJ 9, e10621. doi:10.7717/peerj.10621 · other copies
- BInternational Trade Centre (ITC) (2011). The Coffee Exporter's Guide, Third Edition. International Trade Centre (UNCTAD/WTO), Geneva.
- AEckhardt, S.; Franke, H.; Schwarz, S.; Lachenmeier, D. W. (2022). Risk Assessment of Coffee Cherry (Cascara) Fruit Products for Flour Replacement and Other Alternative Food Uses. Molecules 27(23), 8435. doi:10.3390/molecules27238435 · other copies
- AAbreu, G. F.; Rosa, S. D. V. F.; Coelho, S. V. B.; Pereira, C. C.; Malta, M. R.; Fantazzini, T. B.; Vilela, A. L. (2023). Influence of hulling and storage conditions on maintaining coffee quality. Anais da Academia Brasileira de Ciências 95(4). doi:10.1590/0001-3765202320190612 · other copies
- ASistema de Qualidade de Produtos Agrícolas, Pecuários e Agroindustriais do Estado de São Paulo Manual de Qualidade: Boas Práticas de Produção de Café. Secretaria de Agricultura e Abastecimento do Estado de São Paulo (Codeagro).
- AFAO Guidelines for the Prevention of Mould Formation in Coffee. Food and Agriculture Organization of the United Nations.
Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.
Discussion
A space for readers to share experience from the farm, the roastery or the bar, ask each other and offer other views. The Coffearium team does not reply here. Please keep to the community rules; to change the entry itself, use Suggest a change.