Method for roasting coffee beans, and roasting device
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- ROAST PRECISION SYSTEMS GMBH
- Filing Date
- 2026-01-19
- Publication Date
- 2026-07-30
Smart Images
Figure EP2026051137_30072026_PF_FP_ABST
Abstract
Description
title
[0001] Method for roasting coffee beans and roasting device Technical field of invention
[0002] The invention relates to a method for roasting coffee beans and a roasting device with which, in particular, the method can be carried out. The taste of coffee is influenced by a multitude of factors. Besides the choice of coffee beans – hereinafter also simply referred to as “beans” – The processing of the coffee beans, which can be wet or dry, the grind size, and the brewing method, are all factors, but roasting the coffee beans has a particularly significant impact on the coffee's flavor. Roasting coffee requires consideration of a multitude of parameters that influence the process. In non-industrial and non-automated roasting environments, this often results in roasting outcomes that depend on chance, unless the person operating the roasting equipment has extensive experience. A typical coffee bean roasting process includes the following steps: First, a roasting chamber is preheated to a predetermined temperature. This is typically measured by a bean temperature sensor inside the chamber, which primarily measures the bean temperature, i.e., the temperature of the beans at their surface. This means that the roasting chamber—especially before the first roast—continues to be heated for a longer period after the bean temperature sensor has signaled that the preheating temperature has been reached, before the coffee beans are added, to ensure that the temperature is essentially the same throughout the chamber. Once the predetermined preheating temperature is reached, meaning that the roasting chamber is at a substantially uniform temperature, a predetermined quantity of coffee beans is added.The preheating temperature is typically set between 140 °C and 190 °C, preferably around 180 °C, but can also be higher or lower. When the coffee beans, which are usually stored at room temperature, are added, the temperature at the bean temperature sensor initially drops sharply before rising again after reaching a tipping point. The temperature at this tipping point is generally between 75 °C and 120 °C and depends on the preheating temperature and other bean-specific factors such as moisture content, quantity, and temperature. The actual roasting process therefore only begins at this tipping point, which is also referred to as the starting temperature.
[0003] In the first phase, also known as the drying phase, the coffee beans are dried. In the roasting chamber, once the turning point is reached, the temperature is continuously increased from the starting temperature to approximately 150 °C, or it increases automatically to this temperature if the preheating temperature is sufficiently high. The water contained in the beans evaporates, and the beans turn yellow until the final temperature of the first phase, approximately 150 °C, is reached.
[0004] In a second phase, which directly follows the first, the coffee beans are roasted by initiating a Maillard reaction. This process converts reducing sugars such as glucose and galactose, along with various amino acids, into melanoidins, resulting in the browning of the beans. Simultaneously, the sugars caramelize. During this second phase, the temperature of the roasting chamber is continuously increased to approximately 200 °C. In both the second phase and the third phase, described below, it is preferable that the temperature rises steadily rather than remaining at a constant level, in order to avoid bready or baked aromas. Ideally, the first derivative of the bean temperature curve, the so-called "rate of rise," should exhibit a linearly decreasing slope.The second phase lasts approximately 50% of the total roasting time, but the duration can vary depending on the roast and the type of bean.
[0005] While the second phase also influences the quality and taste of the coffee beans, the decisive phase is the third phase, also known as the development phase. In this phase—where the temperature is further increased—chemical transformations continue, complex aromas develop, and the beans darken. The color of the beans corresponds to so-called roast levels. The roast level of the coffee beans is an important factor in the taste and aroma of the coffee. A lighter roast results in a fruity and acidic taste, while a darker roast leads to a stronger, bitter, and spicy aroma.
[0006] The first and lightest roast level among the established roast levels is the so-called Cinnamon Roast. The event that marks the achievement of the Cinnamon Roast is known as the First Crack. Simultaneously, the First Crack also marks the beginning of the development phase, the transition from the second to the third phase. During the development phase, the roast levels become darker, with the darkest practically achievable roast level being reached at or shortly after the Second Crack. These roast levels are referred to as Vienna Roast or Italy Roast. Most roasts end with medium roast levels—examples include American Roast, City Roast, and Full City Roast—in the temperature range between the First Crack and Second Crack. Worldwide, approximately seven roast levels are distinguished, depending on the region. These lie between the First Crack and the Second Crack.This means that roast levels can be associated with bean core temperatures ranging from 196 °C to 225 °C, in approximately 5 °C increments per roast level. Depending on the bean's thermal conductivity, a different gradient develops between the bean's surface temperature and the bean core temperature, i.e., the temperature in the center of the coffee bean. This means that a roast level cannot be precisely assigned a single, readily measurable temperature at the bean's surface (also referred to as bean temperature), but rather depends on the heat transfer to the bean's core.
[0007] The bean temperature, i.e., the temperature measured at the surface of the bean, varies depending on the bean variety when the first crack occurs. This is due to the formation of a temperature gradient between the bean's surface and its inner core. The first crack occurs when the bean's core reaches a temperature of 196 °C. The gradient varies depending on the bean's thermal conductivity. Beans with higher density conduct heat better, so the first crack occurs at lower surface temperatures. Conversely, beans with lower density experience a higher surface temperature for the first crack. Therefore, the surface temperature of the coffee beans at the time of the first crack provides information about the heat transfer within the bean.
[0008] The timing of the first crack, and also the second crack, are therefore crucial reference points for determining the roast level of a roast. The third phase is the most sensitive of the roasting process, as it has the greatest influence on aroma development. At the same time, with durations between approximately 50 seconds and six minutes for specialty roasts, it is the shortest of the three phases, and even slight differences in roasting time can drastically alter the roast level and aroma. The first and second cracks are characterized by audible cracking sounds, which arise from the overlapping cracking sounds of individual coffee beans—hereafter referred to simply as beans.Differences in the beans and / or temperature fluctuations in the roasting device lead to a statistical distribution of the number of cracking beans over time: The beans crack audibly over a period of 60 to 80 seconds, with the cracking initially being only faintly audible, becoming louder as the number of cracking beans increases, and then decreasing again.
[0009] To gain information about the roasting process, various machine parameters can be measured during roasting. The moisture content of the beans, the bean variety, and their material properties, such as density and cell structure, significantly influence the outcome of a particular roasting process. Therefore, it is not possible to derive a roast level solely from machine parameters and the roasting progress and define it as a criterion for stopping a roast. Without knowing the point in time known as the "first crack" during roasting, it is virtually impossible to determine the roast level of a roast. Knowing the first crack makes it possible to estimate the roast level. Depending on which parameters are included in the estimate, such as...The more accurate the estimation becomes, the more precise the estimation becomes, considering factors such as bean variety, temperature rise from the first crack, temperature gradient from the first crack, time from the first crack until roasting is stopped, bean temperature at the first crack, and bean moisture content. Knowing the first crack, therefore, helps to estimate a termination criterion, allowing for the closest possible approximation of the desired roast level. Thus, determining the exact time of the first crack is desirable to obtain reproducible results. State of the art
[0010] Manually operated roasting machines for non-industrial use, as well as control software for such machines, often offer the option of marking the occurrence of the first crack by entering a value or pressing a button. Upon hearing the crackling sound, the operator must subjectively decide at what point they consider the start of the development phase. This then starts a timer that displays the duration of the third phase, the development phase, to the operator. While the total roasting time ranges from eight to 25 minutes, the third phase, the development phase, lasts a maximum of only a few minutes, primarily to prevent the establishment of a constant temperature level and the associated loss of flavor.As explained in the previous paragraph, the phase in which cracking sounds are audible can last up to 80 seconds, making the subjective determination of the start of this development phase subject to considerable uncertainty. This can lead to significantly different roasting results, even with the same bean variety, precisely because of the relatively short development phase. Furthermore, various parameters, such as the amount of heat absorbed by the roasting chamber, the ambient temperature, air pressure, moisture content, and bean density, mean that the first crack does not always occur at the same time.
[0011] Reliable determination of the first crack time is therefore a prerequisite for reproducible, high-quality roasts with consistent quality. Established methods exist for acoustically determining the first crack time. For example, ROEST® offers such an acoustic detector for its roasting equipment at https: / / www.roestcoffee.com / first-crack. This detector records acoustic signals from which so-called countable cracks are derived. The customer then decides after how many cracks should the timing for the development phase begin, which then ends after a predetermined time. However, this method of determining the first crack is subject to uncertainties, as the cracking sounds and their measurable duration vary considerably depending on the density of the raw coffee beans – meaning the time at which the predetermined number of cracks is reached can vary.
[0012] Another approach is pursued in EP 3311 675 A1. It is known that during the drying phase, the chemical transformation begins in the cells of the coffee beans, characterized by an endothermic, phase-exothermic reaction. This process generates a so-called roasting gas, which consists of various individual components, in particular different oxidizable gases such as carbon monoxide, carbon dioxide, and a multitude of volatile organic compounds in varying concentrations. It is known that with increasing temperature during the roasting process, the quantity of these gases increases significantly and exerts enormous pressure on the cell walls.The exact causes leading to first and second cracking are still unclear, but it is assumed that in the phase known as first cracking, the pressure of the gases causes the cell walls to rupture, resulting in the audible cracking sounds; simultaneously, heat is released, and the reaction proceeds exothermically in this phase. Second cracking, also characterized by audible cracking sounds, is partly attributed to the rupture of the cell matrix, combined with oils migrating to the surface. EP 3311 675 A1 describes a method in which the degree of roasting is derived from the concentration of carbon monoxide produced during roasting. There is a direct correlation between the degree of roasting and the level of carbon monoxide concentration. The carbon monoxide is measured in situ during roasting, and the concentration is determined from this measurement.Once the carbon monoxide concentration corresponding to the desired roast level is reached, the roasting process is stopped. A disadvantage of this method is that a gas sensor, necessary for determining the carbon monoxide concentration, must be calibrated. However, stable calibration is practically impossible due to the constant and increasing contamination of the sensors during roasting caused by the oily components in the roasting gas.
[0013] The object of the invention is therefore to further develop a method for roasting coffee beans of the type described above, in which, at least during the second and third phases of roasting the coffee beans, the volume concentration of at least one component of a roasting gas released during roasting is determined continuously or at discrete intervals, such that at least the first crack, but preferably also the second crack, can be determined with higher accuracy, so that the results of the roasting process can be readily reproduced. Furthermore, a roasting device is to be developed which is particularly suitable for implementing the method.
[0014] In a process of the type described above, this problem is solved by continuously determining the rate of increase in the volume concentration of at least one component of the roasting gas during the second and third phases of coffee bean roasting. From this rate of increase, a first point in time is then determined at which the second phase—in which the Maillard reaction takes place—transitions into the third phase—the development phase. This first point in time corresponds to a first maximum in the rate of increase. This first point in time is the point of first cracking.
[0015] The first crack is thus characterized by a maximum rate of increase in volume concentration, i.e., the release rate of one or more components of the roasting gas. In contrast to the prior art, the rate at which this concentration changes is determined, not the level of the gas concentration over time. Studies have shown that in the first crack phase, the amount of gas released increases sharply until a maximum release rate is reached, after which the rate of increase decreases again. This behavior is related to the statistical distribution of the number of coffee beans that crack over time, with temperature as a parameter.
[0016] Similarly, a second point in time can be determined from the rate of increase in the volume concentration of at least one component of the roasting gas in the third phase, which corresponds to a second maximum in the rate of increase of the volume concentration of at least one component of the roasting gas. This second point in time corresponds to the point of the second crack. While the cracking sounds at the first crack are attributed to the rupture of the cell walls of the coffee beans due to excessive internal pressure, which also leads to the release of heat in an exothermic reaction, the cracking sounds of the second crack are attributed, at least in part, to the rupture of the cell matrix, which is associated with oils migrating to the surface of the bean.
[0017] The timing of the first crack, and optionally the timing of the second crack, can be determined with high precision using the method according to the invention. This timing can be indicated or displayed by the roasting device, for example acoustically or visually, so that the person operating the roasting device has a precise indication of when to stop the roasting process, provided that the timing of the first or second crack is used as a reference point. The accuracy of the method according to the invention is significantly increased compared to acoustic measurements, since the maximum is determined precisely, whereas with an acoustic measurement the maximum can easily be missed, especially if the statistical distribution of the number of cracking sounds over time shows a flat profile.
[0018] The method works particularly well when at least the volume concentration of carbon monoxide, a component of the roasting gas, is determined. The proportion of carbon monoxide in the roasting gas is very high, which simplifies the measurement and increases accuracy. However, the method can also be performed with other roasting gas components, such as volatile organic compounds, provided these are present in sufficient quantities and the sensors for such compounds have sufficiently high sensitivity to determine the volume concentration of the compound in question and its changes with sufficient accuracy to establish the first and, if necessary, the second time point. With appropriate sensors, several components of the roasting gas can also be determined simultaneously, enabling redundant determination of the first and second time points, if necessary.
[0019] To determine the first and second time points as accurately as possible, a high frequency of measurements of the volume concentration of at least one component of the roasting gas is desirable. From a control engineering perspective, the simplest approach is to measure the proportion of at least one component of the roasting gas and subsequently determine a value proportional to the volume concentration in equidistant time steps. In the case of a metal oxide gas sensor, for example, the resistance RSENS of the gas-sensitive layer is determined by a voltage measurement and processed by a control circuit. The time interval between two measurements can advantageously be set depending on other parameters that influence the control of the roasting device used, such as the duration of a roasting process. For example, the interval between two measurements can be between 0.1 and 10 seconds.Setting the time interval between two measurements to 1 second has proven advantageous, as this ensures close monitoring sufficient for many bean varieties, heating curves, and other control parameters, while also minimizing processing power requirements. However, it is within the scope of expert skill to use other values, for example, in the range of 0.5 to 2 seconds. Furthermore, a variable time interval between measurements is also possible: If, for instance, the roasting system's control unit registers an increased volume concentration of the roasting gas, the interval between measurements can be reduced to determine the first crack as precisely as possible, and then increased again once the first crack has occurred. A similar approach can be used for the second crack.
[0020] Once the first crack is reached, this is indicated to the person operating the roasting machine. Depending on the bean variety, desired roast level, and roast profile, the operator can then manually stop the roasting process, as the duration of the roasting process from the first crack is usually known to the operator from daily routine and repeated roasting cycles. To further increase the accuracy and reproducibility of a roast, it is advantageous to stop the roasting process during the third phase after reaching a predefined termination condition for a specific roast level, whereby the termination condition takes into account a minimum time interval from the first crack. In the simplest case, once the termination condition is reached, this is indicated to the person operating the roasting machine, who can then stop the roasting manually.Preferably, the termination condition is stored in the roasting device's control system, for example, in conjunction with a control curve defined for the period between the initial point in time and the termination point—which typically also depends on the bean variety and the desired roast level—so that the roasting process is automatically terminated when the termination condition is reached. Other, non-predefined parameters, such as humidity, ambient temperature, and the temperature and temperature gradient of the beans at the initial point in time, can influence the roasting process and be taken into account accordingly. For example, the time interval can be set for standardized parameter ranges, and the control system can vary this interval if some of the parameters—determined by measurements during roasting—lie outside these standardized ranges.The adjustment can be made by accessing local or non-local databases, such as those stored in the cloud. Additionally or alternatively, the controller can also access lookup tables containing the relevant values. Calculating the timing of adjustments is also possible, as is accessing control curves stored within the controller. The only essential parameters to be entered by the operator of the roasting machine are the bean variety and the desired roast level. Other parameters, which may also determine the predefined termination condition, such as the bean temperature at first crack, can be determined automatically.
[0021] As mentioned previously, the time elapsed since the first crack is only one, albeit crucial, parameter influencing the termination condition. In addition to this time elapsed, it is also advisable to consider the specified roast level, the surface temperature of the coffee beans, the temperature profile of the surface of the coffee beans after the first crack, the moisture content of the coffee beans before and / or during roasting, and / or the coffee bean variety. All these parameters ultimately influence the time elapsed since the first crack and can be used in the control system to adjust this interval, after which the roasting process is terminated. Furthermore, these parameters can also be used to create a more general roasting profile, which can be stored directly in the control system, in a connected database, or in a lookup table.Such a roasting profile is then preferably transferable and can be implemented accordingly on other roasting devices.
[0022] By determining the first point in time not only by measuring the instantaneous volume concentration, but also by differentiating the release rate from its temporal profile as the first derivative of the volume concentration as a function of time, errors that arise from limiting the determination to just the volume concentration are eliminated; only in this way can the first point in time be determined precisely. As described above, knowing the first point in time and other parameters, such as the bean variety and the desired roast level, a termination condition can be defined that ends the roasting process after a predetermined time, dependent on these parameters, has elapsed.To further increase the accuracy in determining the roasting process termination point and to reduce other sources of error, the termination condition can also include one or more dynamic components that are determined only during the roasting process and after the initial time point has been established. For this purpose, after the initial time point has been determined, the subsequent temporal profile of the volume concentration of at least one component of the roasting gas is also measured. The termination condition takes this profile into account in one or more of the ways described below.
[0023] One approach is to determine an initial quotient by dividing the maximum rate of change of volume concentration, determined at the first time point (the first crack), by the rate of change of volume concentration over time (i.e., the ongoing release rate of the roasting gas), starting at the first time point. This calculation is performed continuously at various measurement points, so that the instantaneous rate of change relative to the first time point is used when calculating the quotient. The measurement points can be chosen continuously or quasi-continuously, but intervals of several seconds or even several tens of seconds are also possible. The roasting process is terminated when this quotient reaches a predetermined initial threshold.The values of this quotient increase from 1 upwards, as the counter value represents the maximum over time, and are suitable for the sensitive control of particularly light roast levels. A second possibility is to determine a second quotient by dividing the volume concentration value over time – again, the value at a given measurement point, i.e., the instantaneous volume concentration – by the volume concentration determined at the first measurement point. When this second quotient reaches a predefined second threshold, the roasting process is terminated. The values of this second quotient also increase from 1 upwards. Both methods can be combined. Additionally or alternatively, the first quotient and / or the second quotient can also be combined with the instantaneous bean temperature, i.e., the temperature of the coffee beans at the respective measurement point, as an additional parameter.Weighting factor; the roasting process is stopped when the weighted quotients reach their respective first and second thresholds as specified above.
[0024] The invention relates, in addition to the method for roasting coffee beans as described above, to a roasting device for roasting coffee beans, with which, in particular, the method described above can be carried out. Such a roasting device comprises (a) a heated roasting chamber, (b) a first temperature sensor arranged in the roasting chamber, with which the temperature at the surface of the coffee beans is determined, (c) a first gas sensor for the continuous detection of first measured values, which correspond to the volume concentration of at least one component of a roasting gas released from the coffee beans during roasting, (d) a first temperature control unit, with which the temperature in the roasting chamber and / or the temperature of the coffee beans at their surface is regulated and / or controlled, (e) an evaluation unit,(f) a control unit connected to the first temperature sensor, the first gas sensor, the first temperature control unit, and the evaluation unit, which terminates the roasting process after reaching a termination condition that takes into account at least the time interval to a first maximum of the rate of increase. If both the surface temperature of the coffee beans and the temperature in the roasting chamber are to be measured and controlled, the roasting device has an additional temperature sensor at a different location. The temperature sensor for determining the surface temperatures is located in an area of the roasting chamber where the beans move during roasting.so that it is at least predominantly surrounded by the coffee beans during roasting. The temperature sensor for determining the temperature in the roasting chamber is located outside this area.
[0025] The first gas sensor is designed to detect measurements corresponding to the volume concentration of at least one component of the roasting gas. For a more precise determination of the first maximum of the rate of increase, redundant measurements of other components of the roasting gas can advantageously be detected. This can be done with the first gas sensor, but the roasting device typically includes additional gas sensors connected to the control unit for detecting these further components. In addition to the first maximum of the rate of increase, which corresponds to a first time point (the first crack), a second maximum of the rate of increase of the volume concentration of the corresponding component of the roasting gas, corresponding to a second time point (the second crack), can also be determined.
[0026] The first gas sensor, and possibly subsequent gas sensors as well, are installed in locations within the roasting device where they come into sufficient contact with the roasting gas. This could be, for example, directly in the roasting chamber. Preferably, however, the gas sensors are installed in the exhaust duct or its inlet area, since all components of the roasting gas are discharged there and the gas sensors inevitably pass through; the specific gravity of the roasting gases therefore plays no role here.
[0027] The first gas sensor can, in principle, be designed to detect any component of the roasting gas. However, it is advantageous to use the first gas sensor for detecting carbon monoxide, a component of the roasting gas. The proportion of carbon monoxide in the roasting gas is very high, which simplifies the measurement and increases accuracy.
[0028] For the detection of at least one component of the roasting gas, any sensor that delivers accurate results under the given operating conditions during roasting—where oils are released alongside the roasting gas and can foul sensors, thus distorting absolute measurements—is fundamentally suitable. Tests have shown that metal oxide gas sensors are particularly well-suited for the purpose of the invention, namely the continuous determination of the volume concentration of one or more components of the roasting gas over time. These types of gas sensors are cost-effective and robust under the given operating conditions; moreover, they have proven their worth in numerous practical applications in other contexts. Various types of sensors exist, differing in their construction and the underlying principle of signal processing.There are sensors that are temperature-modulated and whose signal waveforms are evaluated according to specific characteristics, or that are subjected to alternating voltage signals to determine specific capacitances in addition to resistance. Electrochemical sensors or infrared sensors are also suitable for detecting carbon monoxide.
[0029] Metal oxide gas sensors (MOx gas sensors) comprise a heated layer contacted with a metal oxide. Depending on the type of metal oxide used, the gas sensors react to different gases and change their conductance as a function of the volume concentration of the gas to be detected. However, the generally considered disadvantages of MOx gas sensors, such as nonlinearity, low selectivity, cross-sensitivity, drift, and aging, do not affect the results of the measurements according to the invention. For example, the GGS series MOx gas sensors from UST Umweltsensortechnik GmbH are well-suited; for carbon monoxide (CO) as a roasting gas component, the GGS 2530 T gas sensor, for instance, is suitable due to its high sensitivity to this component, even with changes in concentration. The 2600 series MOx gas sensors from Figaro Engineering Inc. are also suitable.are well suited for determining the volume concentration of roasting gas components, for example the TGS 2620 gas sensor.
[0030] Metal oxide gas sensors are offered by many different manufacturers and are constantly being further developed. In addition to the classic evaluation of electrical conductance, gas sensors are being developed, among others, in which the imaginary part of the AC impedance is used to determine the volume concentration of the roasting gas component by means of dielectric excitation. Thus, there is a large number of gas sensors that are suitable for the purpose of the invention with regard to their design and signal evaluation. Regardless of the specific sensor, it is sufficient if the gas sensor used converts the volume concentration of the relevant component of the roasting gas into an evaluable signal.
[0031] If an MOX gas sensor is used, the roasting device preferably includes a second temperature control unit, which is connected to the control unit and regulates the temperature of the first gas sensor. Depending on the type of gas sensor and its operating temperature, the gas-sensitive surface of the MOX gas sensor is heated to temperatures between 100 °C and 600 °C. Depending on its installation location on the roaster, the gas sensor is exposed to hot gases with temperatures of up to 300 °C. This changes the operating temperature of the gas sensor; excessively high ambient temperatures can destroy the sensor. The second temperature control unit maintains the heated surface at a constant operating temperature to prevent damage.
[0032] It is understood that the features mentioned above and those to be explained below can be used not only in the combinations given, but also in other combinations or on their own, without leaving the scope of the present invention. Brief description of the drawings
[0033] The invention will now be explained in more detail with reference to exemplary embodiments and the accompanying drawings, which also disclose essential features of the invention. These exemplary embodiments serve only for illustration and are not to be interpreted as limiting. For example, a description of an embodiment with a multitude of elements or components should not be interpreted as meaning that all of these elements or components are necessary for its implementation. Rather, other embodiments may also contain alternative elements and components, fewer elements or components, or additional elements or components. Elements or components from different embodiments may be combined unless otherwise specified. Modifications and variations described for one embodiment may also be applicable to other embodiments. To avoid repetition, identical or corresponding elements in different figures are designated with the same reference numerals and are not explained multiple times. The figures show:
[0034] Fig. 1 shows a flowchart for the process of roasting coffee beans,
[0035] Fig. 2 shows an exemplary construction of a roasting device,
[0036] Fig. 3 shows an example of a gas concentration and a calculated release rate from it.
[0037] Fig. 4 shows a section of the curves shown in Fig. 3 with the first crack, and
[0038] Fig. 5 shows the relationship between the temperature of the beans at their surface and the occurrence of the first crack. Detailed description of the drawings
[0039] Figure 1 shows a flowchart illustrating a process for roasting coffee beans. This process comprises the following steps, which are carried out sequentially: In a first step 100, the roasting chamber is preheated to a predetermined preheating temperature. During heating, in an intermediate step 110, it is continuously checked whether the preheating temperature has been reached. As soon as the preheating temperature has reached the predetermined value, which is usually between 140 °C and 190 °C, but can also be outside this range – at which point the temperature should be the same throughout the roasting chamber – in a step 120, a predetermined quantity of coffee beans is introduced into the roasting chamber and dried in a first phase.Since the temperature of the coffee beans – which are usually stored at room temperature or refrigerated – is lower than the temperature in the roasting chamber when they are introduced, the temperature in the roasting chamber initially drops and then rises again after reaching a tipping point, as the roasting chamber is heated further. During this first phase, the water contained in the beans evaporates; the final temperature of this first phase is approximately 150 °C. Even during this first phase, the system continuously monitors whether the final temperature has been reached; this step is not shown in Fig. 1.
[0040] Subsequently, in a further step (130), the coffee beans are roasted by initiating a Maillard reaction in a second phase, as described earlier. The temperature in the roasting chamber is continuously increased to approximately 200 °C, ensuring that the temperature does not remain at a constant plateau. During roasting, the volume concentration of at least one component of a roasting gas released from the coffee beans is continuously or at discrete time intervals determined. Carbon monoxide is particularly well-suited as a component of the roasting gas for these measurements. During the second roasting phase, in step 140, the rate of increase of the volume concentration of this component is continuously determined from its temporal profile.Also in step 140, a first point in time at which the second phase transitions into a third phase is determined from the time course of the rate of increase in the volume concentration of at least one component of the roasting gas. This first point in time, the so-called first crack, corresponds to a first maximum of the rate of increase; at this point, the coffee beans have simultaneously reached a core temperature of 196 °C. In total, step 140 continuously checks whether the first maximum of the rate of increase has been reached. If this is not the case, the second phase continues and the temperature in the roasting chamber is further increased. The measurements of the volume concentration of at least one component of the roasting gas are preferably taken at equidistant intervals of one second. The point in time at which the first maximum is reached, i.e.,The time of the first crack can optionally be stored in a database (150) to be used, along with other data, for creating a roasting profile. This other data includes, for example, the specified roast level, the surface temperature of the coffee beans, the temperature profile of the coffee beans after the first crack, the moisture content of the coffee beans before and / or during roasting, and / or the coffee bean variety.
[0041] After the first maximum temperature rise rate is reached, which is noticeable over time as a decrease in the rate of rise after exceeding this maximum, the third phase of roasting the coffee beans begins in step 160. This is the so-called development phase, in which the temperature is further increased and which immediately follows the first point in time, the first crack. In this third phase, the rate of rise of the volume concentration of at least one component of the roasting gas is continuously determined, at least initially, from the time course of its volume concentration. This is necessary to accurately determine the first maximum, which occurs within seconds or fractions of a second, depending on the measurement frequency, after reaching the first crack.Optionally, the profile of this volume concentration can continue to be determined even after the first maximum has been determined and exceeded. This can be particularly useful in the third phase for determining a second time point that corresponds to a second maximum in the rate of increase of the volume concentration of at least one component of the roasting gas. This second time point is also referred to as the second crack.
[0042] In any case, the end of the roasting process is determined based on the first crack. This can be done manually, for example, by notifying the person performing the process that the first crack has occurred. This person can then decide when to stop the roasting, often based on experience. Preferably, however, a termination condition is specified that takes into account at least the time elapsed since the first crack, i.e., it specifies how much time may elapse between the first crack and the end of the roasting process. This can also be displayed to the person in question. However, the reproducibility of the roasting process is increased if the termination condition is stored in database 150.Using a control unit for the roasting process, the roasting process can then be automatically terminated, for example, after the time period specified in the termination condition has elapsed—if only the time of the first crack is considered. Accuracy is further increased if the additional data, which can optionally be stored in database 150 as described above, are also taken into account when defining the termination condition. When defining the termination condition, the subsequent temporal profile of the volume concentration of at least one component of the roasting gas after the first time point can also be considered, from which various quotients can be calculated for comparison with the volume concentration at the first time point. Accordingly, in a process as shown in Fig.As shown in Figure 1, in the optional step 170, it is monitored whether the termination condition has been reached. To determine the termination condition, the data stored in database 150 is used, specifically the time of the first crack and the time elapsed between the first crack and the end of the roasting process. If a roasting profile has already been stored, the roasting process can also run essentially automatically until it is terminated in step 180. The roasted beans are then removed from the roasting chamber and cooled.
[0043] Fig. 2 shows a schematic diagram of the structure of a roasting device for roasting coffee beans, which can also be used to carry out the previously described method for roasting coffee beans. Such a roasting device comprises a heated roasting chamber 10. The coffee beans are roasted there. A first temperature sensor 20 is arranged in the roasting chamber 10, with which the temperature at the surface of the coffee beans is determined. The first temperature sensor 20 is located at the bottom of the roasting chamber 10. The roasting chamber 10 can be designed so that the beans are moved during roasting, for example by a mechanical mixing device (stirrer, curved paddles) or, as in hot air or fluidized bed roasters, by the airflow. This ensures a more uniform temperature distribution in the beans when the heat for roasting is introduced into the roasting chamber 10, usually from a central point.The roasting device also includes a first gas sensor 30. The first gas sensor 30 must be located in the roasting device where it can come into sufficient contact with the roasting gas released from the coffee beans during roasting, for example, in the exhaust duct. In the roasting device shown here, the first gas sensor 30 is located at the top of the roasting chamber 10, where an exhaust duct (not shown) connects. The first gas sensor 30, which is preferably a metal oxide gas sensor, serves to continuously detect initial measured values that correspond to the volume concentration of at least one component—for example, carbon monoxide—of a roasting gas released from the coffee beans during roasting. Optionally, the roasting device can also include further gas sensors for detecting other components of the roasting gas.
[0044] Also connected to the roasting chamber 10 are a first temperature control unit 40, an evaluation unit 50 and a control unit 60. The control unit 60 is connected to the first temperature sensor 20, the first gas sensor 30, the first temperature control unit 40, and the evaluation unit 50. During operation, the first temperature sensor 20 transmits values to the control unit 60, which can optionally be integrated with the evaluation unit 50 in a common component or module. The control unit, in turn, is connected to the first temperature control unit 40 and instructs it to change the temperature in the roasting chamber 10 and how quickly this should occur.
[0045] The temperature control unit 40 controls or regulates the surface temperature of the coffee beans, as this temperature is measured by the first temperature sensor 20. Typically, the temperature of the coffee beans is a key reference point for the operators and indicates the different roasting phases. Alternatively, the temperature of the roasting chamber 10 can be directly regulated and / or controlled without measuring the surface temperature of the coffee beans, for example, in a highly automated process where the coffee beans already have a predetermined temperature and humidity level before being placed in the roasting chamber. For this purpose, a second temperature sensor (not shown) can be installed in the roasting chamber 10, which measures the temperature of the roasting chamber 10 outside the area where the coffee beans are moving.
[0046] During operation, the evaluation unit 50 continuously determines the rate of increase in the volume concentration of at least one component of the roasting gas, as well as at least one initial maximum of this rate of increase, from the initial measured values supplied by the first gas sensor 30. This initial maximum of the rate of increase corresponds to the first point in time described above, the first crack. This point in time is stored in the control unit 60, which then terminates the roasting process—in the case of a largely automated process—after a termination condition is reached. This termination condition takes into account at least the time interval since the first maximum of the rate of increase, the first crack.In simple terms, when the first time is determined in the control unit, a timer is started, and after a period of time has elapsed, which takes into account the time interval defined in the termination condition, the roasting process is automatically terminated. This time interval, which in the simplest case solely defines the termination condition, can be read by the control unit 60 from a database 150, which itself can also be integrated into the control unit 60. Alternatively, it can be entered manually by the person operating the roasting device before the start of the roasting process, at the latest during the drying phase.
[0047] Depending on the type of sensor used for the first gas sensor 30, the roasting device can include a second temperature control unit 70, which is also connected to the control unit 60 and regulates the temperature of the first gas sensor 30. For metal oxide gas sensors, for example, such a second temperature control unit 70 is advantageous for maintaining a constant operating temperature of the detection surface and thus extending its service life. Furthermore, the roasting device can include additional gas sensors connected to the control unit 60 for detecting other components of the roasting gas. These additional sensors can also be connected to their own temperature control units.
[0048] The essential aspects of the process flow are explained in more detail below in connection with Figures 3, 4, and 5. Carbon monoxide was used as the component of the roasting gas in each case, and its volume concentration and its change were determined. Figure 3 shows, as an example, the course of the gas concentration with the dashed curve and the resulting calculated release rate, which corresponds directly to the rate of increase of the volume concentration of carbon monoxide and is simply the first derivative of the time course of the gas concentration. The first maximum occurs approximately eleven minutes after the start of the residual process and corresponds to the first time point, the first crack. The second maximum occurs at approximately 14 minutes and corresponds to the second time point, the second crack.In the region of the first and second crack, a sharp increase in the release rate is initially observed. After exceeding the maxima, the release rate drops again almost as sharply, and the increase in the amount of gas released, i.e., the gas concentration, slows down. This pattern results from the statistical distribution of the bursting beans. Both times can therefore be determined directly from the measurement data with high accuracy, unlike what would be possible with a purely acoustic measurement of the cracking of the beans, as is known in the prior art.
[0049] Fig. 4 shows a section of Fig. 3 in the area between the eighth and thirteenth minutes after the start of the roasting process, where the release rate, again represented by the solid line, has been smoothed. The maximum rate of increase / release rate is clearly visible at 11.5 minutes.
[0050] Finally, Fig. 5 shows the relationship between the bean temperature and the carbon monoxide release rate. The dotted curve shows the surface temperature of the coffee beans, and the solid line represents the release rate, or the rate of increase in the volume concentration of carbon monoxide. For the sake of clarity, the gas concentration curve is not shown here. Heated intake air introduces the heat required to raise the bean temperature into roasting chamber 10. After the beans enter the roasting chamber, the bean temperature rises continuously. Above a certain bean temperature, the carbon monoxide release rate increases. This results in the release rate curve typical of the first crack, characterized by a pronounced maximum (see also Fig. 3 and Fig. 4).
[0051] The roasting process and roasting device described above enable a precise, objective, and reproducible determination of the first crack and second crack during coffee bean roasting. This makes roasting processes easier to reproduce, even for laypersons, and ensures consistent quality of the roasted beans, especially when used with semi- or fully automatic roasting equipment, but also during manual operation. 10 roasting chamber 20 first temperature sensor 30 first gas sensor 40 First temperature control unit 50 Evaluation unit 60 Control unit 70 second temperature control unit 150 database
Claims
Patent claims 1. A method for roasting coffee beans, comprising the following steps: Preheating a roasting chamber (10) to a predetermined preheating temperature, introducing a predetermined quantity of coffee beans into the roasting chamber (10) after reaching the preheating temperature, Drying the coffee beans in a first phase, Roasting the coffee beans in a second phase by initiating a Maillard reaction, Roasting the coffee beans in a third phase, wherein, at least during the roasting of the coffee beans in the second and third phases, the volume concentration of at least one component of a roasting gas released from the coffee beans during roasting is determined continuously or at discrete time intervals, characterized in that, during the roasting of the coffee beans in the second and third phases, the rate of increase of the volume concentration of the at least one component of the roasting gas is continuously determined from the time course of the volume concentration, From the course of the rate of increase of the volume concentration of at least one component of the roasting gas, a first time point at which the second phase transitions into the third phase is determined, whereby the first time point corresponds to a first maximum of the rate of increase.
2. Method according to claim 1, characterized in that a second time point is determined from the course of the rate of increase of the volume concentration of the at least one component of the roasting gas in the third phase, which corresponds to a second maximum of the rate of increase of the volume concentration of the at least one component of the roasting gas.
243. Method according to claim 1 or 2, characterized in that at least the volume concentration of carbon monoxide as a component of the roasting gas is determined.
4. Method according to one of claims 1 to 3, characterized in that the volume concentration of the at least one component of the roasting gas is measured at equidistant time intervals of one second.
5. Method according to one of claims 1 to 4, characterized in that the roasting process is terminated during the third phase after reaching a predetermined termination condition for a predetermined degree of roasting, wherein the termination condition takes into account at least a time interval from the first time point.
6. Method according to claim 5, characterized in that the predetermined termination condition additionally takes into account the predetermined degree of roasting, the temperature at the surface of the coffee beans, the temperature profile at the surface of the coffee beans after the first time point, the moisture content of the coffee beans before and / or during roasting, and / or a variety of coffee beans and is preferably used to create a roasting profile.
7. A method according to claim 5 or 6, characterized in that the predetermined termination condition additionally takes into account the further temporal course of the volume concentration of the at least one component of the roasting gas after the first time point and the roasting process is terminated when - a first quotient of the value of the rate of change of the volume concentration determined at the first time point divided by the value of the rate of change of the volume concentration relative to the first time point reaches a predetermined first threshold value, and / or - a second quotient of the value of the volume concentration over the time of the volume concentration divided by the value of the volume concentration determined at the first time point reaches a predetermined second threshold, and / or the first quotient and / or the second quotient, each weighted by the current bean temperature, reaches the respective predetermined first or second threshold.
8. Roasting device for roasting coffee beans, comprising - a heated roasting chamber (10), - a first temperature sensor (20) arranged in the roasting chamber (10), with which the temperature at the surface of the coffee beans is determined, - a first gas sensor (30) for the continuous detection of first measured values, which correspond to the volume concentration of at least one component of a roasting gas released from the coffee beans during roasting, - a first temperature control unit (40), with which the temperature in the roasting chamber (10) and / or the temperature of the coffee beans at their surface is regulated and / or controlled, - an evaluation unit (50) with which the rate of increase of the volume concentration of at least one component of the roasting gas and at least one first maximum of the rate of increase are continuously determined from the first measured values, - a control unit (60) which is connected to the first temperature sensor (20), the first gas sensor (30), the first temperature control unit (40) and the evaluation unit (50), and which terminates the roasting process after reaching a termination condition which takes into account at least the time interval to a first maximum of the rate of increase.
9. Roasting device according to claim 8, characterized in that the first gas sensor (30) is arranged in the roasting chamber (10) or in an exhaust air duct of the roasting chamber (10).
10. Roasting device according to claim 8 or 9, characterized in that the first gas sensor (30) is a metal oxide gas sensor.
11. Roasting device according to any one of claims 8 to 10, characterized in that the first gas sensor (30) is designed to detect carbon monoxide as a component of the roasting gas.
12. Roasting device according to any one of claims 8 to 11, comprising a second temperature control unit (70) which is connected to the control unit (60) and with which the temperature of the first gas sensor (30) is controlled.
13. Roasting device according to one of claims 8 to 12, comprising further gas sensors connected to the control unit (60) for detecting further components of the roasting gas. 27