CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Roasting · Roaster

Fluid-bed roasters

Hot-air roasters with no moving parts in the roasting chamber – a stream of hot air lifts, mixes and heats the beans, mainly by convection. Common in research, in sample roasting and in some industrial designs.

Across the chain: Science · Equipment

Source-checkedby Coffearium verification pass· Last source-checked 14 Oct 2026· Prepared with AI assistance· 12 sources (5 tier A)· 2 contested· Updated 14 Oct 2026
Reading depth
Contents
  • A fluid-bed roaster has no moving parts in the roasting chamber; hot air blown up through a perforated plate keeps the beans in motion.1
  • According to a roasting chapter, convection is the main heat transfer, whereas a drum roaster may transfer a substantial share by conduction from its hot wall.1
  • Lifting the beans needs a relatively high air velocity.1
  • In a spouted bed, beans rise fast in a central spout and sink slowly back through the bed; their flow pattern changes as they lose density during roasting.3
  • Thermocouples in the bean bed read a combination of bean and air temperature3; laboratory studies have measured the bean core with fine thermocouples.2,5

How a fluid-bed roaster works#

A fluid-bed roaster has no moving parts inside its roasting chamber. The beans are kept in motion only by the current of hot air.1 The air enters at the bottom of the chamber through a perforated plate and leaves at the top.1 It must flow fast enough to lift the beans.1 A patent for such a roaster describes the air velocity as sufficient to levitate the beans in a bed of air.9

The same air mixes the beans and heats them; convection is the main heat transfer.1 A fan draws the air in, a heater warms it, and a probe just before the chamber measures its temperature to control the heater.2,5 At the end of the roast, the beans fall by gravity into a separate cooling unit.1

1165471098321
  1. Fresh air and fan. A fan or blower draws in air and pushes it through the system.2,8
  2. Heater. Electric heating elements in laboratory and patented designs; a patent from 1976 also allows a direct gas burner.2,8,9
  3. Air-temperature probe. Measures the hot air just before the chamber and controls the heater.2,5
  4. Air inlet. Hot air enters from below through a perforated plate.1
  5. Bean fountain. In a spouted bed, beans rise fast in a central spout and sink slowly back down along the wall.3
  6. Bean-mass probe. Reads a combination of bean and air temperature.3
  7. Expansion and exhaust. The hot air leaves at the top of the chamber.1
  8. Cyclone. Separates chaff from the exhaust air.8,9
  9. Recirculation. In larger systems, a major part of the exhaust returns to the heater to save energy.1,8
  10. Vented share. Part of the exhaust must always leave, sometimes through off-gas cleaning.1
  11. Cooler. A separate section where ambient air cools the beans.1,2,5
A batch fluid-bed (spouted-bed) roaster with partial recirculation of its exhaust air; the layout combines elements of the laboratory and patented designs cited. Schematic, not to scale.

Some designs shape the chamber so that the air stream rotates, a “rotating fluidised bed”.1 Others add inclined walls on which beans slide down. While sliding, they take up a larger share of heat by conduction before re-entering the fast air.1 A 1976 patent used an inclined back wall to guide the beans back into the airstream.8

Spouted beds and the bean fountain#

In a spouted bed, the air drives a jet of beans rapidly upwards at the air inlet.3 A 2021 tracking study followed single beans in such a roaster.3 It found two zones: a dense bed in which beans move slowly, below 0.5 m/s, and a spout in which they rise at 0.5–1.5 m/s.3 From the bed, beans fall downward to the spout, parallel to the wall.3

The motion changes during the roast. Beans become less dense, so they are more easily lifted and move faster in the spout.3 At the highest air flow with a small batch of green coffee, the study saw a circulating flow with no true bean bed.3 Batch size and air flow therefore shape how the beans move.3

00.511.522.533.5m/sBeans in the bed(spouted bed, PEPT)below 0.53 Beans rising in the spout(PEPT)0.5–1.53 Suction air at roasterinlet, slow roast(fluid bed)2.0372 Suction air at roasterinlet, fast roast(fluid bed)3.0072
Data table
Pointm/sSource
Beans in the bed (spouted bed, PEPT)below 0.50–0.53
Beans rising in the spout (PEPT)0.5–1.50.5–1.53
Suction air at roaster inlet, slow roast (fluid bed)2.0372.0372
Suction air at roaster inlet, fast roast (fluid bed)3.0073.0072
Bean speeds measured in a pilot spouted bed3, and suction-air velocities (ambient air at the roaster inlet, before the heater) set for a fast and a slow roast in a 100 g laboratory fluid bed.2 Two different machines; the values show orders of magnitude, not one roaster.

Air temperature and air flow as controls#

A fluid-bed roaster is steered through the hot air: its temperature and its velocity.1,2 The laboratory fluid bed of a 2005 study allowed accurate control of both.2 The research roasts cited here held the air at one temperature for the whole roast.2,4 In one study, a fast roast used 260 °C air for 170 s and a slow roast 228 °C air for 720 s, both to the same colour.2

Air flow matters too. In a 2025 study on a 0.5 kg spouted-bed roaster, batch size and air flow changed the time–temperature profiles more than the inlet air temperature did.4 More air also speeds up drying: a high air-to-bean ratio can raise the air velocity at the bean surface, and the bean then dries faster.1 The air-to-bean ratio is the mass of air per mass of green coffee.1

200220240260280300°CLab fluid bed, 100 g(720 s; 170 s)2282 2602 Experimental spouted bed(best quality at 5 min)2507 Pilot spouted bed, 0.5 kg(constant inlet air)220–2804 1976 patent: air not above530 °F≈2778
Data table
Point°CSource
Lab fluid bed, 100 g (720 s; 170 s)2282282
Lab fluid bed, 100 g (720 s; 170 s)2602602
Experimental spouted bed (best quality at 5 min)2502507
Pilot spouted bed, 0.5 kg (constant inlet air)220–280220–2804
1976 patent: air not above 530 °F≈2772778
Hot-air temperatures used in a laboratory fluid bed2, an experimental spouted bed7 and a pilot spouted bed4, and the upper limit in a 1976 patent (530 °F, converted).8

Fluid bed and drum compared#

According to a 2017 roasting chapter, both are hot-air roasters, and both transfer heat by convection, conduction and radiation at the same time.1 The shares differ. In a fluid bed, convection dominates; in a drum, a substantial amount of heat may pass by conduction, especially with direct heating under the drum.1 A drum is limited in air flow by the speed at which beans would be carried away with the exit air.1 A fluid bed needs high air speed to keep the beans moving at all.1

When researchers made a laboratory fluid bed and a commercial drum follow the same bean-temperature profile, the kinetics of aroma formation were similar in both.6

Reference values#

Values reported in the sources for particular machines and coffees. They are not targets.

Parameter Value reported Source
Batch size, laboratory fluid bed 100 g green coffee Geiger et al.2
Air temperature and time, fast roast (HTST) 260 °C, 170 s Geiger et al.2
Air temperature and time, slow roast (LTLT) 228 °C, 720 s Geiger et al.2
Inlet air velocity during roasting 3.007 m/s (fast); 2.037 m/s (slow), measured at the suction inlet Geiger et al.2
Roast degree reached by both L* 22–23; roast loss 15.38% (fast), 15.86% (slow) Geiger et al.2
Bean-core thermocouple Type K, 0.25 mm, in a 0.3 mm drilled hole; bean tied to a wire inside the fluidised batch Geiger et al.2
Batch size, pilot spouted bed (0.5 kg) 200, 350 and 500 g Al-Shemmeri et al.3,4
Inlet air temperatures, same roaster 220, 235, 250, 265 and 280 °C, each held constant Al-Shemmeri et al.4
Bean speed in the bed Below 0.5 m/s Al-Shemmeri et al.3
Bean speed in the spout 0.5–1.5 m/s, upwards at the air inlet Al-Shemmeri et al.3
Best-quality condition, experimental spouted bed (peaberry) 250 °C air, 5 min Nagaraju et al.7
Overall heat-transfer coefficient, same study 12.023 W/(m²·K), from the time–temperature relationship Nagaraju et al.7
Upper air temperature in a 1976 patent Not above 530 °F (about 277 °C) Sivetz8

Heat transfer in a stream of air#

Convection first. In a fluid bed, convection accounts for the main heat transfer.1 In a spouted bed, the heat-transfer coefficient in the spout is high, because beans there meet the full air-to-bean convection.3 As the beans lighten, more of them are carried by the air, so a greater fraction is heated by convection late in the roast.3 A 1997 spouted-bed study described the heating as an unsteady-state heat-transfer process and derived one overall heat-transfer coefficient for it.7

Conduction as an option. Inclined walls add a phase of higher conduction while beans slide back.1 A 2000 doctoral thesis on hot-air roasting recommended a fairly high proportion of conductive heat transfer and low air-to-bean ratios.5

Radiation is disputed. A roasting chapter treats radiation as usually negligible in hot-air roasters.1 In a laboratory fluid bed, however, the bean core never reached the set air temperature; the authors attributed this to radiation and called this typical of small equipment with a large air-to-bean ratio (see Contested).2

Air-to-bean ratio and the bean. High ratios dry the bean faster and change the combinations of bean temperature and water activity during the roast.1 According to the same chapter, they slightly decrease final bean volume, and changing the ratio alone makes a limited but statistically significant difference in flavour.1 A flavour chapter links higher ratios to aroma stripping and a flatter flavour.10

Measuring temperature in a fluid bed#

Air temperature is the control signal. In the laboratory fluid bed of the 2005 study, a probe placed just before the chamber controlled the heaters.2,5 A patented fluid-bed roaster places its thermocouple at the air nozzle.9 This is the traditional way of automating a roaster: set and hold a hot-air temperature. The bean temperature then varies with green-coffee moisture, weather and a cold roaster.1

The bed probe is a mix. Thermocouples in a spouted bed measure a combination of bean and air temperature.3 A 2025 model of a spouted bed, fitted to logged inlet-air, outlet-air and product thermocouple data, shows the expected difference between thermocouple and bean temperature; it estimates system parameters such as a thermal response coefficient.4 According to a roasting chapter, most roasting systems use a probe that touches both beans and hot air, so their reading is a mix temperature.1

Bean core in the laboratory. Research fluid beds made direct measurement possible. Fine thermocouples went into holes drilled in single beans, which were tied to a wire and moved partly freely within the fluidised batch.2 The thesis behind that roaster averaged at least 10 curves from individual beans to allow for differences between them.5

Exhaust measurement. Air above a fluidised bed is turbulent. In the 2005 study, exhaust air flow could not be measured directly during roasting; it was calibrated in empty runs instead.2 Carbon dioxide and water vapour were then measured in the exhaust in real time.2

Fast and slow roasts in research#

Several laboratory studies of high-temperature short-time (HTST) and low-temperature long-time (LTLT) roasting used fluid beds.2,5,6 These roasters allowed defined process conditions, with accurate control of hot-air temperature and air velocity.2

  • Gas and water. A 2005 study roasted 100 g batches fast and slow to the same colour and followed carbon dioxide and water in the exhaust (see Roasting and Degassing).2
  • Aroma. A 2008 study compared fast and slow roasting in a commercial drum and laboratory fluid beds. When both roasters followed the same bean-temperature profile, aroma kinetics were similar.6
  • Structure. According to a roasting chapter, at the same colour, fast-roasted beans expand more and have greater volume and porosity than slow-roasted beans.1
  • Air flow. In the 2025 spouted-bed study, batch size and air flow shaped the time–temperature profile more than inlet temperature.4

In these studies the air temperature was held constant through the roast, while the bean temperature rose.2,4

Recirculation, chaff and off-gas#

According to the same roasting chapter, smaller roasters usually draw in fresh air and release the hot exhaust (an open system), which wastes energy.1 Most large-scale operations recirculate a major part of the off-gas to the heater and back into the chamber.1 A 1976 fluid-bed patent recycled vented air through a recirculation duct to conserve heat.8

Part of the off-gas must always leave the system. This share may pass through an off-gas cleaning step to meet air-pollution rules.1 Chaff comes off the beans in the air stream. Patented designs collect it in a cyclone in the exhaust8,9; the laboratory roaster of the 2000 thesis removed silverskin by suction.5

Continuous and industrial designs#

Hot-air roasters can run in batches or continuously.1 The 1976 patent covers both batch and continuous versions of its fluid-bed roaster.8 A 1968 CATIE paper, also presented at a 1969 chemistry congress, reported that coffee could be roasted continuously in a fluid bed within limits of temperature, air velocity and residence time.12 According to a 2017 roasting chapter, continuous systems were popular decades earlier but had become nearly extinct; batch roasters dominate because they are more flexible and easier to control.1

A patent granted in the United States describes a fluid-bed system for batches from small profile sizes to large production sizes.9 No source opened for this entry gives market shares of fluid-bed roasters.

Sample and small roasters#

The fluid-bed principle works at very small scale. Research roasters have roasted 100 g batches2, and a pilot spouted bed 200–500 g.3 Small patented systems cover “profile” (sample) batch sizes as well as production.9 No study opened for this entry examined small home air roasters.

Risks and defects#

  • Bean breakage. Any bean movement causes mechanical stress, and a good design minimises breakage.1 Decaffeinated beans can be less stable mechanically and need more attention in fluid-bed roasters.11
  • Batch size out of range. With a small batch and high air flow, the bean bed can disappear into a circulating flow.3 As beans lighten they fluidise more easily; the tracking study suggests that commercial roasting may be best started at moderate air flow, reduced as density changes.3
  • Aroma stripping. High air flows have been linked to flatter, less complex flavour.10
  • Reading the wrong temperature. Air-temperature control does not guarantee the same bean temperature from batch to batch.1 The bed probe reads bean and air together.1,3
  • Off-gas. Recirculating systems must vent part of their off-gas.1

Open questions#

  • Air-to-bean ratio and flavour. Sources disagree on how much the ratio itself matters (see Contested).1,5,6,10
  • Radiation. Its share in small air roasters has not been measured separately in the sources found.1,2
  • Sensory studies. No peer-reviewed sensory comparison of specialty coffees roasted in fluid beds and drums at production scale was found.
  • Home air roasters. No study opened for this entry measured their air flow, temperatures or roast uniformity.

Key studies#

Geiger et al., 2005 – Gas and water in a laboratory fluid bed. Roasts 100 g batches in a fluid bed with controlled air temperature and velocity, fast (260 °C, 170 s) and slow (228 °C, 720 s) to the same colour. Measures bean-core temperature with drilled-in thermocouples and carbon dioxide and water in the exhaust; the core stays below the air temperature.

Journal of Food Science 70(2), E124–E130 · doi:10.1111/j.1365-2621.2005.tb07084.x

Al-Shemmeri et al., 2021 – Following single beans in a spouted bed. Tracks labelled beans by positron emission particle tracking in a pilot spouted-bed roaster at three batch sizes and several air flows. Beans move slowly in the bed and fast in the spout, and the flow pattern changes as they lose density.

Journal of Food Engineering 311, 110709 · doi:10.1016/j.jfoodeng.2021.110709

Al-Shemmeri et al., 2025 – Modelling a spouted-bed roast. Builds a batch-scale heat and mass transfer model from logged inlet, outlet and product temperatures of a 0.5 kg spouted bed. Batch size and air flow shaped the profiles more than inlet temperature, and the model estimates system parameters such as a thermal response coefficient.

Beverages 11(6), 162 · doi:10.3390/beverages11060162

Baggenstoss et al., 2008 – Fluid bed versus drum. Follows 16 aroma compounds in fast and slow roasts in a commercial drum and laboratory fluidising beds, all stopped at the same lightness. When both roasters followed the same bean-temperature profile, the kinetics of aroma formation were similar.

Journal of Agricultural and Food Chemistry 56(14), 5836–5846 · doi:10.1021/jf800327j

Contested
  • Air-to-bean ratio and cup quality: a doctoral thesis on hot-air roasting found that low air-to-bean ratios gave superior cup quality5, and a flavour chapter links higher ratios to flatter, less complex flavour through aroma stripping10; a roasting chapter describes the effect of the ratio alone as limited but statistically significant1, and when a laboratory fluid bed and a drum followed the same bean-temperature profile, the kinetics of aroma formation were similar.6
  • Radiation in small air roasters: a roasting chapter calls the contribution of radiation in hot-air roasters usually very limited and negligible1; a laboratory fluid-bed study attributes the bean core staying below the set air temperature to excessive heat transfer by radiation, which it calls typical of small-scale equipment with a large air-to-bean ratio.2
Share or cite this entry

Sources

  1. BSchenker, S.; Rothgeb, T. (2017). The Roast – Creating the Beans' Signature. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 11, pp. 245–271. doi:10.1016/B978-0-12-803520-7.00011-6 · other copies
  2. AGeiger, R.; Perren, R.; Kuenzli, R.; Escher, F. (2005). Carbon Dioxide Evolution and Moisture Evaporation During Roasting of Coffee Beans. Journal of Food Science 70(2), E124–E130. doi:10.1111/j.1365-2621.2005.tb07084.x · other copies
  3. AAl-Shemmeri, M.; Windows-Yule, C.; Lopez-Quiroga, E.; Fryer, P. J. (2021). Coffee bean particle motion in a spouted bed measured using Positron Emission Particle Tracking (PEPT). Journal of Food Engineering 311, 110709. doi:10.1016/j.jfoodeng.2021.110709 · other copies
  4. AAl-Shemmeri, M.; Fryer, P. J.; Farr, R.; Lopez-Quiroga, E. (2025). Batch-Scale Simulation of Heat and Mass Transfer of Coffee Roasting in Spouted Bed Roasters. Beverages 11(6), 162. doi:10.3390/beverages11060162 · other copies
  5. BSchenker, S. (2000). Investigations on the Hot Air Roasting of Coffee Beans (Diss. ETH No. 13620). ETH Zürich (doctoral thesis).
  6. ABaggenstoss, J.; Poisson, L.; Kaegi, R.; Perren, R.; Escher, F. (2008). Coffee Roasting and Aroma Formation: Application of Different Time−Temperature Conditions. Journal of Agricultural and Food Chemistry 56(14), 5836–5846. doi:10.1021/jf800327j · other copies
  7. ANagaraju, V. D.; Murthy, C. T.; Ramalakshmi, K.; Rao, P. N. S. (1997). Studies on roasting of coffee beans in a spouted bed. Journal of Food Engineering 31(2), 263–270.
  8. A/BSivetz, M. (1976). Coffee roasting system (US Patent 3,964,175). United States Patent and Trademark Office.
  9. A/BFluid bed coffee roasting system (US Patent 10,765,137). United States Patent and Trademark Office.
  10. BPoisson, L.; Blank, I.; Dunkel, A.; Hofmann, T. (2017). The Chemistry of Roasting – Decoding Flavor Formation. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 12, pp. 273–309. doi:10.1016/B978-0-12-803520-7.00012-8 · other copies
  11. BPietsch, A. (2017). Decaffeination – Process and Quality. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 10, pp. 225–243. doi:10.1016/B978-0-12-803520-7.00010-4 · other copies
  12. BRolz, C.; Menchú, J. F.; Sandoval, J.; Simón, P. (1968). La tostación continua del café en cama fluida. CATIE, Turrialba, Costa Rica; also in the 10th Latin American Congress of Chemistry, San José, Costa Rica, 1969, p. 185.

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”.

Suggest a change

On Fluid-bed roasters. Tell us what you would change and why. A source that supports it helps most.