WO2018159165A1

WO2018159165A1 - Coffee bean cracking occurrence schedule prediction method, coffee bean cracking occurrence schedule prediction device, coffee bean cracking detection method, coffee bean cracking detection device, and roasted coffee bean production method

Info

Publication number: WO2018159165A1
Application number: PCT/JP2018/002159
Authority: WO
Inventors: 山本　裕之
Original assignee: 山本　裕之
Priority date: 2017-02-28
Filing date: 2018-01-24
Publication date: 2018-09-07
Also published as: WO2018159164A1; JP6211223B1; JP2018139577A; JP6261788B1

Abstract

Provided is a method for easily and accurately detecting or predicting cracking occurrence when roasting coffee beans. A coffee bean cracking detection method according to the present invention comprises: an average value calculation step for calculating an average value of sound pressure during a cracking non-occurrence period, when temporally measuring sound having a frequency in a range included within the range of 5-80 kHz generated from coffee beans that are being roasted and a roasting device; and a cracking detection step for, by using the average value of sound pressure as a threshold, determining that cracking of the coffee beans has been detected if the measured sound pressure exceeds the threshold value and, within a time interval of 0.1-99 milliseconds selected from a time period between the time point of exceeding the threshold value and 0.1-100 milliseconds prior to said time point, if a sound pressure of n-fold (n is more than 1/6 but less than 1/3) of the threshold value has not been detected.

Description

Method for predicting coffee bean occurrence time, apparatus for predicting coffee bean occurrence, coffee bean detection method, coffee bean detection device, and method for producing roasted coffee beans

[Technical Field] The present invention relates to a method for predicting coffee bean occurrence, a device for predicting coffee bean occurrence, a coffee bean detection method, a coffee bean detection device, and a method for producing roasted coffee beans.

Coffee beans can bring out their flavor by roasting. However, the flavor of coffee beans varies with roasting conditions. Therefore, it is necessary to control roasting conditions in order to bring out the desired coffee bean flavor.

By the way, in the roasting stage, the coffee beans generate a goby sound (also referred to as “crack sound”) twice with heating. In order to obtain a desired flavor, an operation of changing the amount of heat given to the roasted beans is performed using the generation time of at least one of the two goby sounds as a guide. In addition, an operation of changing the amount of heat given to roasted coffee beans one to two minutes before the occurrence of a goby sound may be performed by predicting the occurrence time of the goby sound based on the operator's experience and intuition. As described above, it is important to detect or predict the generation time of the goby sound in the roasting stage.

Here, for example, Non-Patent Document 1 discloses a method for detecting a goby sound generated from coffee beans by sound pressure level and frequency spectrum analysis. The goby sound produced by one coffee bean lasts for several milliseconds to 10 and several milliseconds. Of these gospel sounds, frequency spectrum analysis is frequently performed on sounds in the band of several hundred Hz to several tens of kilohertz. In order to remove noise generated from the image and other background noise and detect only the goby sound generated from the coffee beans with high sensitivity, a signal processing dedicated processor (DSP) is required, which is costly. Further, according to this method, since it is necessary to distinguish between the first and second goby sounds, it is difficult to distinguish them in real time during roasting.

Patent Document 1 uses the fact that the first goby sound and the second goby sound have a higher sound pressure than the sound (device sound and environmental noise) generated in the period between them. It is disclosed that roasting conditions are adjusted based on the occurrence of goby sound. However, in such a method, there is no great difference between the sound pressure of the sound generated in the period between the first goby sound and the second goby sound, and the sound pressure of the second goby sound. It is often difficult to accurately distinguish and detect the occurrence of

Canadian Patent Application Publication No. 2267608

The present invention has been made in view of the above circumstances, and an object thereof is to provide a method for easily and accurately detecting or predicting the occurrence of goby when roasting coffee beans.

The present inventors have found that there is a correlation between the amount of carbon monoxide generated from coffee beans being roasted and the generation of goby sound, and the present invention has been completed. More specifically, the present invention provides the following.

(1) According to the first aspect of the present invention, no goby is generated when the sound of the frequency within the range of 5 kHz to 80 kHz generated from the coffee beans being roasted and the roasting device is measured over time. An average value calculating step for calculating an average value of sound pressures during the period, and using the average value of the sound pressures as a threshold value, the measured sound pressure exceeds the threshold value and 0.1 times from the time when the threshold value is exceeded. More than n times the threshold (n is more than 1/6 and less than 1/3) in a time width of 0.1 to 99 milliseconds selected from a period from milliseconds to 100 milliseconds ago A coffee bean detection method, comprising: a goby detection step of determining that coffee goby has been detected when sound pressure is not detected.

(2) The second invention of the present invention is the coffee bean gourd detection method according to the first invention, wherein the sound is started to be measured continuously after 2 minutes from the start of roasting.

(3) In the first or second invention, the third invention of the present invention is an initial carbon monoxide for measuring the amount of carbon monoxide generated from coffee beans being roasted until at least 2 minutes after the start of roasting. It is a goat detection device of coffee beans which further has a quantity measurement process.

(4) According to a fourth aspect of the present invention, in any one of the first to third aspects, the sound generated from the coffee beans being roasted when the sound is measured over time is sampled at a sampling period of 0. It is a goby detection device for coffee beans that is measured in 2 milliseconds or less.

(5) A fifth invention of the present invention is the coffee machine according to any one of the first to fourth inventions, based on the sound pressure of the frequency having a half-value width of 0.1 kHz or more and 50 kHz or less in the goby detection step. This is a coffee bean goby detection device for detecting bean goby.

(6) A sixth aspect of the present invention is a sound measuring unit that measures, over time, a sound having a frequency within a range of 5 kHz to 80 kHz generated from roasted coffee beans and a roasting device; From the time when the measured sound pressure exceeds the threshold and exceeds the threshold, using an average value calculation unit that calculates the average value of the sound pressure during the absence of goby and the average value of the sound pressure as a threshold In a time width of 0.1 ms to 99 ms selected from a period from 0.1 ms to 100 ms before, the threshold is n times or more (n is more than 1/6 and 1/3) And a goby detection unit that determines that a goat of coffee beans has been detected when no sound pressure is detected.

(7) According to the seventh aspect of the present invention, no goby is generated when the sound of the frequency within the range of 5 kHz to 80 kHz generated from the roasted coffee beans and roasting device is measured over time. An average value calculating step for calculating an average value of sound pressures during the period, and using the average value of the sound pressures as a threshold value, the measured sound pressure exceeds the threshold value and 0.1 times from the time when the threshold value is exceeded. More than n times the threshold (n is more than 1/6 and less than 1/3) in a time width of 0.1 to 99 milliseconds selected from a period from milliseconds to 100 milliseconds ago Manufacture of roasted coffee beans having a goby detection step for determining that a goat of coffee beans has been detected when sound pressure is not detected, and a control step for changing roasting conditions based on the detection of the goby Is the method.

According to the present invention, it is possible to provide a method for roasting coffee beans based on the result of detecting or predicting the occurrence of goby simply and accurately when roasting coffee beans.

It is the schematic of the goby detection apparatus of coffee beans. It is the schematic of an example of a goby detection part. 1 is a schematic view of a coffee bean occurrence predicted time prediction device. It is the schematic of a roasted coffee bean manufacturing apparatus. It is the output waveform (a) of the apparatus sound of the roasting apparatus and the environmental noise (a), the waveform (b) after passing through the 10 kHz bandpass filter, and the waveform (c) of the peak detector output. The output waveform (a) of the goby sound microphone (a), the waveform (b) after passing through the 10 kHz bandpass filter, and the waveform (c) of the peak detector output. FIG. 4 is a profile of the concentration of carbon monoxide in exhaust gas at each time from the start of roasting in Example 2-1, the number of goses generated every 10 seconds, the set temperature, and the exhaust temperature. FIG. 5 is a profile of the concentration of carbon monoxide in exhaust gas at each time from the start of roasting in Example 2-2, the number of goose generations every 10 seconds, the set temperature, and the exhaust temperature. FIG. 4 is a profile of the concentration of carbon monoxide in exhaust gas at each time from the start of roasting in Example 2-3, the number of goses generated every 10 seconds, the set temperature, and the exhaust temperature. FIG. 5 is a profile of the concentration of carbon monoxide in exhaust gas at each time from the start of roasting in Example 2-4, the number of goose generations every 10 seconds, the set temperature, and the exhaust temperature. FIG. 5 is a profile of the concentration of carbon monoxide in exhaust gas at each time from the start of roasting in Example 2-5, the number of goses generated every 10 seconds, the set temperature, and the exhaust temperature. FIG. 10 is a profile of the concentration of carbon monoxide in the exhaust gas at each time from the start of roasting in Example 2-6, the number of goses generated every 10 seconds, the set temperature, and the exhaust temperature. FIG. 10 is a profile of the concentration of carbon monoxide in the exhaust gas at each time from the start of roasting in Example 2-7, the number of goses generated every 10 seconds, the set temperature, and the exhaust temperature. FIG. 8 is a profile of the concentration of carbon monoxide in exhaust gas at each time from the start of roasting in Example 2-8, the number of goses generated every 10 seconds, the set temperature, and the exhaust temperature.

Hereinafter, specific embodiments of the present invention will be described in detail. In addition, this invention is not limited to the following embodiment, In the range which does not change the summary of this invention, it can change suitably.

<Coffee beans goby detection method>
The coffee beans goby detection method according to the present embodiment, when measuring over time the sound in the range of 5 kHz to 80 kHz generated from roasted coffee beans and roasting device, An average value calculating step for calculating an average value of the sound pressure in the medium, and using the average value of the sound pressure as a threshold value, the measured sound pressure exceeds the threshold value and 0.1 mm from the time when the threshold value is exceeded. Sounds not less than n times (n is more than 1/6 and less than 1/3) of the threshold in a time range of 0.1 to 99 milliseconds selected from a period from a second to 100 milliseconds. And a goby detecting step for determining that the goat of coffee beans has been detected when the pressure is not detected. By having such a configuration, the coffee bean goby detection method according to the present embodiment can accurately detect the occurrence of goby. And based on the result, roasting conditions of coffee beans can be controlled.

Thus, the range of the frequency for measuring the sound pressure is included in the range of 5 kHz to 80 kHz. In such a range, the goby sound generated from the coffee beans tends to rise sharply, and is not easily affected by, for example, the sound generated from the apparatus or the sound of coffee flowing. Therefore, by measuring the sound pressure in such a frequency range and measuring the sound pressure pulse that rises at a predetermined speed, it is possible to detect coffee beans without being influenced by other noises. .

(Roasting process)
A roasting process is a process of supplying heat with respect to coffee beans and roasting coffee beans, for example.

The heat supply means is not particularly limited, and any known means used in conventional roasting of coffee beans can be used. Specifically, for example, a compressor (blower) or a blower that generates heat from a heat source such as an electric heater for coffee beans stored in a roasting container (for example, a rotating drum for roasting coffee beans) Heat can be supplied by blowing air or the like and applying hot air to the coffee beans. As the temperature of the supplied heat, a heating temperature used for normal roasting of coffee beans can be used, and for example, a temperature in the range of 100 to 300 ° C. can be used. Further, as the amount of hot air blown, the amount of air in the range of 0.1 to 1 L per minute per 1 g of coffee beans can be used. Note that the coffee beans to which heat is supplied in the roasting process refer to coffee beans that are raw materials for the roasted coffee beans.

The coffee beans are not particularly limited, and are conventionally known brands of coffee beans (for example, Guatemala, Brazil (Santos, Datella, etc.), Ethiopia (Mochail Gachev, etc.), Costa Rica, Kilimanjaro, Vietnam, Colombia, Tanzania, Mocha, Blue Mountain, Crystal Mountain, Kenya, Mandelin, Mexico, etc.) can be used alone or in a blend of two or more. In addition to the brand, any coffee beans can be used for the lot, the production area, the purification method, and the production year. Further, the coffee bean goby according to the present embodiment can be used to predict the occurrence of goby even if it is a blend of two or more different brands, lots, production areas, refining methods, and production years. The forecasting method of occurrence time is useful.

As the roasting method, for example, the roasting container (rotary drum) is rotated, and the coffee beans can be roasted while mixing the coffee beans. As the rotation condition, a conventionally known condition can be used. For example, a condition of 5 to 60 rpm can be used.

In the roasting process, the temperature in the roasting container may or may not be measured using temperature measuring means such as a temperature sensor. By measuring the temperature in the roasting container, the atmospheric temperature in the roasting container at the time of roasting can be confirmed.

In the roasting process of the present invention, the chaff generated by roasting can be collected in order to remove the coffee beans chaff (so-called silver skin) while roasting.

In the roasting process in the present invention, the roasting time is naturally determined according to the set value of carbon monoxide as an index, but can be performed within a range of 300 to 1500 seconds, for example.

(Sound measurement process)
The sound measuring step is a step of measuring the sound generated from the coffee beans being roasted and the measuring device over time.

Specifically, in the sound measurement process, sound is measured using a device such as a microphone that converts the signal into an electric signal or the like.

The measurement location (measurement target) is not particularly limited as long as it can measure the sound pressure in the frequency range of the specific range described above among the coffee beans being roasted and the sound generated from the measurement device. However, a generally known rotary drum type roasting apparatus may be provided with a cylindrical heat-resistant glass. When the roasting apparatus includes such a cylindrical heat-resistant glass, the goby sound is not transmitted and the goat may not be sufficiently detected. Therefore, for example, the rotating drum type roasting apparatus is preferably made of metal (specifically, brass or the like).

The start timing of sound measurement is not particularly limited, and for example, it is preferable to start 2 minutes after the start of roasting (start of heating). That is, at the initial stage of roasting (for example, 2 minutes after the start of roasting, preferably after 3 minutes, and more preferably after 4 minutes), the sound pressure measurement process may not be performed. At the initial stage of roasting (for example, after 2 minutes from the start of roasting, preferably after 3 minutes, more preferably after 4 minutes), the coffee beans to be roasted are cold and the surface is hard. ing. In such a case, a sound pressure pulse may be generated in the range of 5 kHz to 80 kHz due to a collision between coffee beans or a collision between a coffee bean and a roasting device, which may cause false detection. On the other hand, when the coffee beans are hard like this, it cannot be said that the coffee beans are sufficiently heated, and no goby sound can be generated. Therefore, measurement of sound pressure is not essential at the initial stage of roasting.

Although not an essential aspect, an initial carbon monoxide amount measuring step for measuring the amount of carbon monoxide generated from the coffee beans being roasted can be provided at least for a period of 2 minutes after the start of roasting. As described above, at the initial stage of roasting, sound pressure pulses are generated in the range of 5 kHz to 80 kHz due to the collision between coffee beans or between the coffee beans and the roasting device, and the accurate roasting state is achieved. Difficult to grasp. Therefore, it is also possible to confirm the roasting state more reliably by measuring the amount of carbon monoxide. In addition, as a specific method of the initial carbon monoxide amount measuring step and an apparatus used therefor, for example, a carbon monoxide amount measuring step and a carbon monoxide amount measuring unit described later can be used.

(Average value calculation process)
The average value calculating step is a step of calculating the average value of the sound pressure during the period when the goby has not occurred.

Here, the “seasonal non-occurrence period” refers to an arbitrary period after the start of roasting (starting heating) after a period when no goby has occurred. The specific time varies depending on the heating conditions and the type of coffee beans, and is not particularly limited, and can be set as appropriate. In addition, for example, a period before a predetermined time from when the coffee beans to be roasted are heated and the goby is detected by hearing (for example, a period of 2 minutes to 1 minute before the first goby is detected by hearing) Etc.) can also be defined as a period when no goby has occurred.

Although the goby non-occurrence period is not particularly limited, for example, a threshold may be set for an arbitrary period of less than 540 seconds from the start of roasting (for example, 60 seconds between 480 to 540 seconds). If it is less than 540 seconds from the start of roasting, normal goby does not occur.

As the average value, for example, the peak value of the sound pressure is divided into a period where the goby has not occurred (for example, 60 seconds), and within the equally divided period (for example, 60 seconds is divided into 60 parts every second). What averaged in the generation | occurrence | production period can be used.

(Haze detection process)
The goby detection process uses the average value of the sound pressure calculated in the average value calculation process as a threshold, and the measured sound pressure exceeds the threshold, and 0.1 milliseconds to 100 milliseconds before the threshold is exceeded. When the sound pressure of n times or more of the threshold value (n is more than 1/6 and less than 1/3) is not detected in the time width selected from the period up to 0.1 milliseconds to 99 milliseconds This is a step of determining that the coffee beans have been detected.

Specifically, in the goby detection step, the number of pulse generations is the number of times that a sound pressure pulse exceeding the above-described threshold is generated within a predetermined period (for example, 1 second, 10 seconds, etc.). When the value counted within the predetermined period exceeds the threshold value, it is determined that goby has occurred.

Here, as the threshold value, the average value of the sound pressure during the goby non-occurrence period calculated in the average value calculating step is used. Then, during roasting of coffee beans, the measured sound exceeds this threshold, and 0.1 milliseconds or more within a period from 0.1 milliseconds to 100 milliseconds before the time when the sound is generated 99 or more 99 If a sound pressure not less than n times the threshold (n is greater than 1/6 and less than 1/3) is not detected in a time width of milliseconds or less, it is determined that goby has occurred.

When the measured sound exceeds the average value of the sound pressure during the absence of goby, it means that a loud sound was generated compared to the background noise. However, it is not possible to determine whether the generated sound is due to the occurrence of goby or mechanical sound only by such a criterion. Therefore, it is confirmed that a sound pressure not less than n times the threshold (n is more than 1/6 and less than 1/3) is not detected in a time width before a specific period from the time when the sound is generated. As described above, the goby sound generated from the coffee beans rises sharply, and therefore, in the case of the goby sound, no loud sound is emitted in a time width before a specific period from the time when the goat sound occurs. On the other hand, for example, device sounds, environmental noises, and the like often emit loud sounds even in a time width before a specific period from the time they are generated. The difference between the sound generation behavior in the period before the generation of the goby sound and the device sound can be distinguished from each other.

As a period during which it is possible to determine whether or not a sound pressure of n times or more (n is more than 1/6 and less than 1/3) of the threshold is detected, the period from which the measured sound is generated is 100 from 0.1 milliseconds before. Although it is not particularly limited as long as it is within a period up to milliseconds, for example, it is preferably within a period from 0.2 milliseconds to 50 milliseconds before the measured sound is generated, More preferably, it is within a period from 30 seconds before 30 seconds, and more preferably within a period from 20 milliseconds before 0.4 milliseconds. When such a period is a required value, a goby sound can be detected more accurately.

In addition, the time width for the actual determination within the period is not particularly limited as long as it is a time width of 0.1 milliseconds to 99 milliseconds, but for example 0.5 milliseconds to 90 milliseconds. It is preferably 2 seconds or less, more preferably 1 millisecond or more and 80 milliseconds or less, and further preferably 2 milliseconds or more and 70 milliseconds or less. When the time width is a required value, the goby sound can be detected more accurately.

The value of n is not particularly limited as long as it is more than 1/6 and less than 1/3, but is preferably 5/24 or more and 7/24 or less, and more preferably 1/4. When n is a required value, a goby sound can be detected more accurately.

The frequency range for measuring the sound pressure is not particularly limited as long as it is within the above-mentioned range. For example, it is preferably 7 kHz or more, more preferably 8 kHz or more, and further preferably 9 kHz or more. 9.5 kHz or more is particularly preferable. Further, the frequency range is preferably 70 kHz or less, more preferably 50 kHz or less, further preferably 40 kHz or less, and further preferably 30 kHz or less. When the frequency range is within the required range, it is possible to increase the accuracy of the distinction between the goby sound and the device sound. The “frequency range in which sound pressure is measured” refers to a frequency range including a sound having a predetermined frequency range to be measured. Specifically, for example, it may be a range in which all sounds in a range of a half width described later are included. At the time of measurement, measurement may be performed so as to include the entire range described above, and it is not excluded to measure other frequency ranges together. When other frequency ranges are also measured, only a required frequency range can be extracted by, for example, a band pass filter.

The half width at the frequency at which the sound pressure is measured is not particularly limited, and is preferably 50 kHz or less, more preferably 30 kHz or less, and still more preferably 20 kHz or less, for example. When the full width at half maximum is less than or equal to the required value, it is possible to increase the accuracy of discrimination between the goby sound and the device sound. On the other hand, as a half value width, it is preferable that it is 0.1 kHz or more, for example, it is more preferable that it is 0.2 kHz or more, and it is further more preferable that it is 0.25 kHz or more. When the full width at half maximum is greater than or equal to the required value, the accuracy of distinguishing the goby sound from the device sound can be increased.

The sampling period for measuring the peak value of the sound pressure (for example, the output of the peak detector) is not particularly limited, but is preferably 0.2 milliseconds or less, for example, 0.15 milliseconds or less. More preferably, it is more preferably 0.1 milliseconds or less, and particularly preferably 0.05 milliseconds or less. When the sampling period is 0.2 milliseconds or less, it is possible to improve the accuracy of the distinction between the goby sound and the device sound. The goby sound has a feature that the peak rises steeply, whereas the device sound has a feature that the peak does not rise steeply.

Specifically, as a method for detecting a sound having a predetermined frequency range and a half-value width as described above, for example, a bandpass filter having a predetermined frequency range and a half-value width can be used.

In addition, when it is determined that goby has been detected in this way, for example, the roasting temperature (for example, thermal power) is increased or decreased for a predetermined period, or the amount of air blown in the roasting device is changed. Perform the operation. For example, the air flow rate can be adjusted by opening and closing an exhaust damper provided in the roasting apparatus. Thereby, the coffee bean which has a desired flavor can be manufactured with sufficient accuracy.

<Coffee bean goby detection device>
The coffee bean goby detection apparatus according to the present embodiment is an apparatus that can realize, for example, the above-described coffee bean goby detection method. A coffee bean goby detection device 1 according to an embodiment of the present invention will be described below with reference to FIG.

FIG. 1 is a schematic view of a coffee bean goby detection device. As shown in FIG. 1, the coffee bean goby detection device 1 includes a sound pressure measurement unit 10, a goby detection unit 11, and a control unit 12. The sound measuring unit 10 is disposed in the vicinity of the roasting container 30 of the roasting apparatus 3. The control unit 12 is connected to an adjustment unit 36 that adjusts roasting conditions (temperature, time, etc.) of the roasting device 3.

[Roasting device 3]
Although not an indispensable aspect, first, the roasting apparatus 3 used for roasting the coffee beans to be measured will be described. The roasting device 3 has a heat source capable of supplying heat to the coffee beans, and is a means capable of roasting the coffee beans. The coffee bean roasting device is not particularly limited, and any conventional coffee bean roasting means can be used. In this embodiment, the roasting device 3 is a roasting device. A container 30, a heat source unit 31, a blower unit 32, a temperature measurement unit 33, an exhaust unit 34, a chaff collector 35, and an adjustment unit 36 are provided.

The roasting container 31 is a container for storing coffee beans. The roasting container 31 is not particularly limited, and is preferably configured to be rotatable in order to mix coffee beans. In addition, as the roasting container 31, a container that can be set to the same rotation condition as the above-described method for predicting the occurrence of goby of coffee beans can be used.

The heat source unit 31 is a means capable of generating heat, and for example, an electric heater can be used. Moreover, as the heat source part 31, what can be used as the heating temperature similar to the goat generation | occurrence | production time prediction method of the coffee beans mentioned above can be used, for example.

The air blower 32 is a means capable of blowing heat generated from the heat source 31 to the coffee beans accommodated in the roasting container 30. The blower 32 can use what can be made into the ventilation conditions similar to the goat generation | occurrence | production time prediction method of the coffee bean mentioned above. As the air blower 32, for example, a compressor (blower) or the like can be used.

The temperature measuring unit 33 is a means for measuring the atmospheric temperature in the roasting container 30.

The exhaust part 34 is a part for exhausting the gas discharged from the coffee beans being roasted.

The chaff collector 35 is a means for removing coffee beans chaff (so-called silver skin).

The adjustment unit 36 is a means that can directly change the roasting conditions of coffee beans. For example, the roasting conditions can be changed and adjusted in accordance with an instruction from the control unit 12. For example, in accordance with an instruction from the control unit 12, roasting conditions such as roasting temperature and roasting time are changed, or roasting is terminated (that is, supply of heat from the heat source is stopped), and roasting is performed. Control can be performed so as to start cooling in the container 30 or to stop cooling by using, as an index, that the temperature of the discharge port has become constant after starting cooling.

[Haze detection device 2]
The goby detection device 2 includes a sound pressure measurement unit 20, a goby detection unit 21, and a control unit 22. The goby detection device 2 measures, over time, the sound pressure of the frequency within the range of 5 kHz to 80 kHz among the sounds generated from the coffee beans stored in the roasting container 30 of the roasting device 3. Therefore, the occurrence of goby is detected. Each part will be described below. The control unit 22 is not an essential aspect.

(Sound measurement unit 20)
The sound measurement unit 20 according to this embodiment will be described. The sound measuring unit 20 is configured to be able to measure a goby sound generated from coffee beans stored in the roasting container 31 of the roasting apparatus 3. For this reason, the sound measuring unit 20 is arranged and used in the vicinity of the roasting container 30.

The sound measuring unit 20 is a means capable of detecting a goby sound generated from roasted coffee beans and converting it into a signal such as an electric signal. Specifically, a general microphone or the like can be used as the sound measurement unit 20. More specifically, Knowles SPU0410LR5H-QB 100 Hz-80 kHz or the like can be used. In the present embodiment, the sound measuring unit 20 is disposed in the vicinity of the roasting container 30, but is not limited to this example, and the sound pressure of a frequency within a range of at least 5 kHz to 80 kHz is measured over time. If it can measure automatically, it will not specifically limit.

(Goby detection unit 21)
The goby detection unit 21 extracts and counts goby sounds from the sound signal measured by the sound measurement unit 20.

Hereinafter, more specific embodiments of the goby detection unit 21 will be described with reference to the drawings, but the present invention is not limited to the following specific examples.

FIG. 2 is a schematic diagram of an example of the goby detection unit. The goby detection unit 21 includes, for example, a bandpass filter 211, a peak detector 212, a waveform analyzer 213, and a counter 214.

The band pass filter 211 extracts only a signal having a predetermined frequency range from the electrical signal of the sound detected by the sound pressure measuring unit 20 (for example, a microphone). In the coffee bean goby detection method according to the present embodiment, for example, by using such a bandpass filter 211, only the sound having a frequency in the range of 5 kHz to 80 kHz is extracted, and the frequency range thereof is extracted. Only the sound pressure can be measured.

The peak detector 212 detects the positive-side peak value of the AC component of the electrical signal having a predetermined frequency that has passed through the band-pass filter 211 and holds it for a predetermined period.

The waveform analyzer 213 performs comparative analysis before and after the rising of the waveform of the output waveform from the peak detector 212, and determines that the sharply rising waveform is a goby sound. The waveform analyzer 213 transmits a signal to the counter 214 every time a goby sound is determined.

The counter 214 counts the signal received from the waveform analyzer 213 every predetermined period, and transmits the number of goby sounds generated every predetermined period (for example, 1 second, 10 seconds, etc.) to the control unit 22. .

(Control unit 22)
Although not an indispensable aspect, the control unit 22 is a means capable of adjusting the roasting conditions in the roasting apparatus 3 based on the goby detection by the gourd detection unit 21.

The control unit 22 receives the output of the goby generation from the goby detection unit 21 and instructs the control unit of the roasting apparatus 3 to change the roasting conditions based on the data. Thereby, the roasting conditions in the roasting device 3 can be adjusted. And roasted coffee beans in which roasting conditions are appropriately controlled can be manufactured by adjusting roasting conditions in this way.

The specific method for adjusting the roasting conditions in the control unit 22 is appropriately selected according to the purpose and is not particularly limited. For example, the coffee beans are linked with the roasting degree of the coffee beans. The heat supply conditions (for example, roasting temperature, roasting time, air blowing conditions, rotating drum rotating conditions, etc.) can be changed to adjust the roasting speed and the like. In addition, the supply of heat to the coffee beans can be stopped. More specifically, the adjustment unit 36 can be instructed to stop the supply of heat to the coffee beans.

Although not provided in the present embodiment, the control unit 22 can instruct on / off of the switch of the gas pump in the amount of carbon monoxide. In addition, the control unit 22 records data (for example, data on another computer (personal computer)) in order to record data (for example, the number of occurrences of the goby sound, the elapsed time since the start of roasting, the temperature within the roasting temperature, the air temperature, etc.). Can also be configured to send.

In addition, the goby detection part 11 and the control part 12 should just be connected to the sound pressure measurement part 10 and the adjustment part 36 in the state which can communicate, respectively. For example, the goby detection unit 11 and the control unit 12 can be installed in a server on the network.

<Prediction method for coffee beans goby>
The coffee bean occurrence time prediction method according to the present embodiment includes a carbon monoxide amount measuring step for measuring the amount of carbon monoxide generated from roasted coffee beans, and the coffee beans based on the carbon monoxide amount. A predicting step for predicting the expected time of occurrence of goby. The coffee bean occurrence time prediction method according to the present embodiment can appropriately predict the expected occurrence of goby by having such a configuration. And based on the result, roasting conditions of coffee beans can be controlled. In addition, according to the coffee bean occurrence time prediction method according to the present embodiment, both the first goby and the second goby are predicted from the two goby sounds generated from the coffee beans during the roasting process. can do.

[Roasting process]
A roasting process is a process of supplying heat with respect to coffee beans and roasting coffee beans, for example.

The heat supply means is not particularly limited, and any known means used in conventional roasting of coffee beans can be used. For example, for coffee beans stored in a roasting container (for example, a rotating drum for roasting coffee beans), heat generated from a heat source such as an electric heater is blown by a compressor (blower), a blower, etc. Heat can be supplied by applying hot air to the coffee beans. As the temperature of the supplied heat, a heating temperature used for normal roasting of coffee beans can be used, and for example, a temperature in the range of 100 to 300 ° C. can be used. Further, as the amount of hot air blown, the amount of air in the range of 0.1 to 1 L per minute per 1 g of coffee beans can be used. Note that the coffee beans to which heat is supplied in the roasting process refer to coffee beans that are raw materials for the roasted coffee beans.

[Carbon monoxide measurement process]
The carbon monoxide amount measuring step is a step of measuring the amount of carbon monoxide generated from the coffee beans being roasted.

The amount of carbon monoxide to be measured is not particularly limited, and either the absolute amount or the relative amount (concentration) of generated carbon monoxide can be used.

The carbon monoxide measurement means (sensor) is not particularly limited, and for example, NAP-505 manufactured by Nemoto Special Chemical Co., Ltd. can be used.

The measurement location (measurement target) is not particularly limited, and for example, a location where the roasting is performed (space) or a location exhausted from the location where the roasting is performed can be measured. When measuring the exhaust, remove the foreign matter with a filter (for example, a chaff filter) in order to measure the carbon monoxide more accurately when feeding it to the measuring means for measuring carbon monoxide. It is preferable to carry out. In addition, it is preferable that the air supply is performed at a constant speed in order to accurately measure the amount of carbon monoxide. Carbon monoxide can be supplied with a gas pump or without a gas pump in order to adjust the speed of the supply. When the pressure at the measurement location changes due to air supply, the pressure can be released to the outside of the measurement location in order to keep the pressure at the measurement location constant.

Measured frequency is not particularly limited, and can be performed continuously or intermittently. Whether continuous or intermittent, it is preferable to measure over time. By measuring over time, the amount of carbon monoxide generated in the future can be predicted more accurately, and as a result, the time of occurrence of goby can be predicted more accurately.

[Prediction process]
The predicting step is a step of predicting the occurrence of goby of coffee beans based on the amount of carbon monoxide.

Specifically, in the prediction step, it is predicted that goby will occur after a predetermined period when the amount of carbon monoxide measured in the carbon monoxide amount measurement step exceeds a threshold value. As a method for determining the threshold, for example, the correlation between the amount of carbon monoxide and the time of occurrence of goby is recorded in advance for a plurality of types (brand, year of production, etc.) of the beans, and a predetermined period before the occurrence of goby (for example, 1-2 minutes before) The amount of carbon monoxide is taken as a threshold value. In the prediction step, when the measured amount of carbon monoxide exceeds the threshold value, the time point is determined to be a predetermined period before the occurrence of goby (for example, 1 to 2 minutes before), and after the predetermined period from the time point (for example, A goby is expected to occur after 1 to 2 minutes).

In addition, it is possible to predict the occurrence of goby using the amount of carbon monoxide during or after the occurrence of goby as a threshold. Further, the control device or the like can be learned in actual operation.

In addition, after measuring the expected time of occurrence of coffee beans, when changing the roasting conditions of coffee beans to produce roasted coffee beans, actually, a predetermined period before the expected time of occurrence of coffee beans (for example, Estimate the time for changing conditions 1 to 2 minutes ago. Thus, coffee beans having a desired flavor can be obtained by estimating the condition change timing and changing the roasting conditions.

And when the occurrence of goby is predicted in this way, for example, an operation such as increasing or decreasing the roasting temperature (for example, thermal power) for a predetermined period, or changing the air blowing amount in the roasting device I do. For example, the air flow rate can be adjusted by opening and closing an exhaust damper provided in the roasting apparatus. Thereby, the coffee bean which has a desired flavor can be manufactured with sufficient accuracy.

<Coffee bean occurrence scheduled time prediction device>
The coffee bean occurrence predicted time prediction apparatus according to the present embodiment is an apparatus that can realize, for example, the above-described coffee bean occurrence occurrence prediction method. A coffee bean gossip scheduled production time prediction apparatus according to an embodiment of the present invention will be described below with reference to FIG.

FIG. 3 is a schematic view of a coffee bean occurrence predicted time prediction device. As illustrated in FIG. 3, the coffee bean gossip scheduled production time prediction device 2 includes a carbon monoxide amount measurement unit 20, a prediction unit 21, and a control unit 22. The carbon monoxide amount measuring unit 20 is connected to the exhaust unit 34 of the roasting apparatus 3. The control unit 22 is connected to an adjustment unit 36 that adjusts roasting conditions (temperature, time, etc.) of the roasting device 3.

[Roasting device 3]
Although not an indispensable aspect, as a roasting apparatus for roasting coffee beans to be measured, the exhaust unit 34 is configured to be able to supply carbon monoxide to the carbon monoxide amount measuring unit 20. And the thing similar to the roasting apparatus 3 mentioned above can be used except the adjustment part 36 being connected so that communication with the estimation part 11 is possible.

[Haze occurrence scheduled time prediction device 2]
The goby occurrence prediction time measuring device 2 includes a carbon monoxide amount measuring unit 20, a predicting unit 21, and a control unit 22. The goby generation prediction time measuring device 2 predicts the goji generation time based on the amount of carbon monoxide generated from the coffee beans roasted in the roasting device 3 described above. Each part will be described below. The control unit 12 is not an essential aspect.

(Carbon monoxide measurement unit 20)
Next, the carbon monoxide amount measuring unit 20 according to the present embodiment will be described. The carbon monoxide amount measuring unit 20 is configured to be able to measure the amount of carbon monoxide sent from the roasting apparatus 3. Although not shown, the coffee bean gossip scheduled generation time measuring apparatus 2 according to the present embodiment further includes a filter between the carbon monoxide amount measuring unit 20 and the exhaust unit 34.

The carbon monoxide amount measuring unit 20 is a means capable of measuring the amount of carbon monoxide generated from the coffee beans being roasted. As the carbon monoxide amount measuring unit 20, for example, NAP-505 manufactured by Nemoto Special Chemical Co., Ltd. can be used. In the present embodiment, the carbon monoxide amount measuring unit 20 is configured separately from the roasting device 3 (outside the roasting device 3) so as to measure the carbon monoxide exhausted from the roasting device 4. However, the carbon monoxide amount measuring unit 20 may be provided inside the roasting apparatus 3.

The filter is disposed in the carbon monoxide passage until the carbon monoxide sent from the roasting apparatus 3 reaches the carbon monoxide amount measuring unit 20. With such a filter, when carbon monoxide is supplied to the carbon monoxide amount measuring unit 20, foreign matters (for example, chaff) can be removed, so that the carbon monoxide amount can be measured more accurately. Can do. Although it does not specifically limit as a filter, For example, a chaff filter etc. can be used.

Further, the coffee bean gossip scheduled occurrence time prediction device 1 can appropriately include other components as long as the effects of the present invention are not impaired. For example, a gas pump for supplying carbon monoxide to the measurement unit 1 can be further provided.

(Prediction unit 21)
The prediction unit 21 is a means for predicting the coffee beans gossip generation scheduled time based on the measurement result of the carbon monoxide amount by the carbon monoxide amount measurement unit 20.

(Control unit 22)
Although not an indispensable aspect, the control unit 22 is a means that can adjust the roasting conditions in the roasting apparatus 3 based on the prediction of the goby occurrence scheduled time by the prediction unit 21.

The control unit 22 receives the output of the expected occurrence time of the goby from the prediction unit 21, and instructs the control unit of the roasting apparatus 3 to change the roasting conditions based on the data. Thereby, the roasting conditions in the roasting device 3 can be adjusted. And roasted coffee beans in which roasting conditions are appropriately controlled can be manufactured by adjusting roasting conditions in this way.

The specific method for adjusting the roasting conditions in the control unit 22 is appropriately selected according to the purpose and is not particularly limited. For example, the coffee beans are linked with the roasting degree of the coffee beans. The heat supply conditions (for example, roasting temperature, roasting time, air blowing conditions, rotating drum rotating conditions, etc.) can be changed to adjust the roasting speed and the like. In addition, the supply of heat to the coffee beans can be stopped.

Although not provided in the present embodiment, the control unit 22 can instruct on / off of the switch of the gas pump in the amount of carbon monoxide. In addition, the control unit 22 records data (for example, data on another computer (personal computer)) in order to record data (for example, carbon monoxide concentration, elapsed time from the start of roasting, temperature within roasting temperature, air temperature, etc.). Can also be configured to send.

In addition, the prediction part 21 and the control part 22 should just be connected to the carbon monoxide amount measurement part 20 and the adjustment part 36 in the state which can communicate, respectively. For example, the prediction unit 21 and the control unit 22 can be installed in a server on the network.

<Combination of the goby detection method and the goby occurrence time prediction method>
Note that the above-described goby detection method and goby occurrence time prediction method can be used in combination. Thereby, the occurrence of goby can be detected with higher accuracy.

<Preparation of roasted coffee bean production equipment>
As the roasted coffee bean manufacturing device 4, a device comprising all of the goby detection device 1 shown in FIG. 1, the goose occurrence scheduled time prediction device 2 shown in FIG. 3, and the roasting device 3 shown in FIGS. Was prepared (FIG. 4). Each will be described in detail below.

[Roasting device 3]
As the roasting means 3, GeneCafe CBR-101 (manufactured by GeneSys) was used. The roasting means 3 includes a roasting container 30, an electric heater as a heat source unit 31, a blower as a blower unit 32, a temperature sensor as a temperature measurement unit 33, an exhaust unit 34, a chaff collector 35, and an adjustment And a unit 36.

In addition, GeneCafe CBR-101 usually has a cylindrical heat-resistant glass as the roasting container 30. In the present example, a 0.1-t brass tube was manufactured and used in order to transmit goby sound.

The adjusting unit 36 can directly change the roasting conditions of the coffee beans according to the temperature and roasting time timer.

The exhaust unit 34 was designed so as to supply exhaust gas to the carbon monoxide amount measuring unit 10 of the goby occurrence scheduled time prediction device 2.

In this example, specific roasting conditions were a drum rotation speed of 10 to 11 rpm, an air volume of about 150 L / min, and a temperature adjustment range of 60 to 250 ° C.

[Haze detection device 1]
Details of each part in the goby detection device 1 are as follows.

(Sound pressure measurement unit 10)
As the sound pressure measurement unit 10, a microphone (Knowles SPU0410LR5H-QB) was used.

(Goby detection unit 11)
As shown in FIG. 3, the goby detecting unit 11 includes a band pass filter 111, a peak detector 112, a waveform analyzer 113, and a counter 114.

As the bandpass filter 111, a two-stage amplification type bandpass filter using an analog operational amplifier was used. The peak detector 112 is composed of an analog operational amplifier and an analog comparator. The waveform analyzer 113 is composed of an analog comparator and a microcomputer. A microcomputer was used as the counter 114.
(Control units 12, 22)
In addition, in order to record data, the

control units

12 and 22 are set so as to transmit the number of occurrences of the goby sound per 10 seconds, the CO gas concentration, the elapsed seconds, the set temperature, and the exhaust temperature to the computer (personal computer). .

[Early goby expected time prediction device 2]
The sensor unit as the goby occurrence scheduled time prediction device 2 was further designed to include a filter and a gas pump. The details of each part in the goby occurrence scheduled time prediction device 2 are as follows.

(Carbon monoxide measurement unit 10)
As the carbon monoxide amount measuring unit 10, CO (carbon monoxide) was measured using a sensor device of NAP-505 (CO measurement range: 0 to 1000 ppm) manufactured by Nemoto Special Chemical.

(Chaff filter)
The chaff filter is for removing foreign substances such as chaff mixed with the exhaust gas supplied from the exhaust unit 34 of the roasting device 3. As the chaff collector, an accessory of the above-described GeneCafe CBR-101 (manufactured by GeneSys) was used.

(Gas pump)
The gas pump sends exhaust gas from the exhaust section 34 in the roasting device 3 to the goby occurrence scheduled time prediction device 1. As the gas pump, NF-11 manufactured by KNF was used. The gas pump capacity was 100 ml / min.

<Example 1>
Using the roasted coffee bean production apparatus 4 described above, 250 g of coffee beans (Mocashidamo G4) were roasted. The heating temperature was constant at 230 ° C. At this time, the main sound is the sound of the roasting device when the goby has not occurred (about 440 seconds from the start of roasting) and the environmental noise, and the goby sound when the goby has occurred (about 620 seconds from the start of roasting). And measured and recorded three channels of bandpass filter input / output and peak detector output simultaneously with a digital oscilloscope.

FIGS. 5 (a) to 5 (c) show the output waveform of the roaster device sound and environmental noise (FIG. 5 (a)), the waveform after passing through the 10 kHz bandpass filter (FIG. 5 (b)), and the peak. It is a waveform (FIG.5 (c)) of a detector output.

6A to 6C show the output waveform of the goby microphone (FIG. 6A), the waveform after passing through the 10 kHz bandpass filter (FIG. 6B), and the waveform of the peak detector output (FIG. 6). (C)).

<Example 2>
(Examples 2-1 to 2-4)
Using the roasted coffee bean production apparatus 4 described above, 250 g of coffee beans were roasted. The heating temperature was constant at 230 ° C. Further, when the concentration of carbon monoxide in the exhaust gas measured in the carbon monoxide amount measurement unit reached 600 ppm, heating was stopped in the adjustment unit 36.

As the period when no goby sound was generated, 60 seconds from 480 to 540 seconds from the start of roasting were selected, and the apparatus sound and environmental noise during that period were measured. The peak value of each second for 60 seconds was recorded, the average value for 60 seconds was recorded, and the threshold value was 541 seconds after the start of roasting. The measured sound pressure exceeds this threshold after 541 seconds from the start of roasting, and the voltage is ¼ or more of the threshold in the period of 3 milliseconds before 0.5 milliseconds from when the threshold was exceeded. Is not recorded, it is determined as a goby sound and a signal is transmitted to the counter. On the other hand, if a voltage of 1/4 or more of the threshold value is recorded during this period, it is determined as noise and no signal is transmitted to the counter.

As a brand of coffee beans, Santos No. 2 (Example 2-1), Mocashidamo G4 (Example 2-2), Guatemala SHB (Example 2-3), Brazilian tailor reserve (Example 2-4) were used. FIGS. 7 to 10 are profiles of the concentration of carbon monoxide in the exhaust gas in the time from the start of roasting, the number of goses generated every 10 seconds, the set temperature and the exhaust temperature.

Table 1 shows the concentration of carbon monoxide in the exhaust gas when the first goby sound is generated and the concentration of carbon monoxide in the exhaust gas when the second goby sound is generated in Examples 2-1 to 2-4. The time from the start of roasting to 4 ppm, 5 ppm, 30 ppm, 40 ppm and 50 ppm is shown. In Table 1, the first goby sound detected when the concentration of carbon monoxide in the exhaust gas is 3 ppm or more and less than 30 ppm is “start of the first goby sound”, and the first goby detected at 50 ppm or more. The sound is "start of second goby sound".

(Examples 2-5 to 2-8)
In a temperature profile of 300 seconds at 180 ° C., 120 seconds at 200 ° C., 120 seconds at 224 ° C., and then maintained at 230 ° C., when the concentration of carbon monoxide in the exhaust gas becomes 9 ppm, the set temperature at that time for 20 seconds Example 2-1 to 2-4, except that the temperature was set to be lowered by 30 ° C. for 20 seconds when the concentration of carbon monoxide in the exhaust gas reached 80 ppm. Similarly, coffee beans were roasted. FIGS. 11 to 14 are profiles of the concentration of carbon monoxide in the exhaust gas during the time from the start of roasting, the number of goses generated every 10 seconds, the set temperature and the exhaust temperature.

Table 2 shows the concentration of carbon monoxide in the exhaust gas when the first goby sound is generated and the concentration of carbon monoxide in the exhaust gas when the second goby sound is generated in Examples 2-5 to 2-8. The time from the start of roasting to 4 ppm, 5 ppm, 30 ppm, 40 ppm and 50 ppm is shown. In Table 2, the first goby sound detected when the concentration of carbon monoxide in the exhaust gas is 3 ppm or more and less than 30 ppm is “the start of the first goby sound”, and the first goby sound detected when the concentration is 50 ppm or more. The sound is "start of second goby sound".

When Table 1 and Table 2 are compared, in Examples 2-1 to 2-8, in the exhaust gas when the first goby sound is detected, although the roasting temperature profile is different, The concentration of carbon monoxide and the concentration of carbon monoxide in the exhaust gas when the second goby sound is detected are in the range of 5 ppm to 9 ppm and 55 ppm to 81 ppm, respectively. Therefore, it is suggested that there is a corresponding relationship between the occurrence of goby noise and the concentration of carbon monoxide in the exhaust gas, regardless of individual differences in coffee beans and roasting temperature.

Using the roasted coffee bean production apparatus 4 described above, 250 g of coffee beans (Mocadamo G4) were roasted. The heating temperature was constant at 230 ° C. Change the center frequency (kHz) and half-value width (kHz) of the sound to be analyzed as shown in Table 3. When the first goby (549-789 seconds after the start of roasting) and when the second goby occurs In each (799 seconds to 999 seconds after the start of roasting), the number of goby occurrences (total number of occurrences in each period), the average number of gouge occurrences per 10 seconds is measured, and when The skewness and kurtosis were calculated for the distribution of the number of goby occurrences at the time of goby occurrence. The results are also shown in Table 3.

<Example 3>
Using the roasted coffee bean production apparatus 4 described above, 250 g of coffee beans (Mocashidamo G4) were roasted. The heating temperature was constant at 230 ° C.

As the period when no goby sound was generated, 60 seconds from 480 to 540 seconds from the start of roasting were selected, and the apparatus sound and environmental noise during that period were measured. The peak value of each second for 60 seconds was recorded, the average value for 60 seconds was recorded, and it was set as the threshold value during the period when the goby sound was generated.

Select a period of 30 to 590-620 seconds from the start of roasting as the period when the goby sound is generated, measure the goby sound during that period, and simultaneously input / output the bandpass filter and the output of the peak detector to the digital oscilloscope. Measured and recorded with. In addition, a goby sound, a device sound, and a flowing sound in the bean drum were measured by hearing. Under these conditions and time, the density of the goby sound is relatively small, and the goby sound can be accurately determined by hearing.

And when the measured sound pressure exceeds this threshold, and a voltage of 1/12 or more of the threshold is not recorded in a period of 3 milliseconds before 0.5 milliseconds from when the threshold is exceeded, A goby sound is determined and a signal is transmitted to the counter. On the other hand, when a voltage of 1/12 or more of the threshold value is recorded during this period, it is determined as noise and no signal is transmitted to the counter. Thereby, goby was detected.

Next, the reference of the voltage value when transmitting a signal to the counter is 2/12 or more (1/6 or more) of the threshold, 3/12 or more (1/4 or more) of the threshold, 4/12 or more of the threshold (1/3 or higher), threshold 5/12 or higher, threshold 6/12 or higher (1/2 or higher), threshold 7/12 or higher, threshold 8/12 or higher (2/3 or higher), threshold 9 / 12 or more (3/4 or more), threshold 10/12 or more (5/6 or more), threshold 11/12 or more, threshold 12/12 or more (threshold or more), and goby detection in the same way Went.

Table 4 shows the number of auditory goby detections per second from 590 seconds to 619 seconds and the voltage value reference when transmitting a signal to the counter as described above and measured by the roasted coffee bean manufacturing apparatus 4 Indicates the number of goby detections made.

* In the table, parentheses (()) indicate false detection, and square brackets ([]) indicate those that cannot be determined by hearing.

It was found that when the reference voltage value when transmitting a signal to the counter is 3/12 or more (1/4 or more) of the threshold value, the number of false detections is the smallest and many goby sounds can be detected.

DESCRIPTION OF SYMBOLS 1 Goby detection apparatus 10 Sound pressure measurement part 11 Goby detection part 111 Band pass filter 112 Peak detector 113 Waveform analyzer 114 Counter 12 Control part 2 Gossop scheduled generation time prediction apparatus 20 Carbon monoxide amount measurement part 21 Prediction part 22 Control part DESCRIPTION OF SYMBOLS 3 Roasting apparatus 30 Roasting container 31 Heat source part 32 Blower part 33 Temperature measuring part 34 Exhaust part 35 Chaff collector 36 Adjustment part 4 Roasted coffee bean manufacturing apparatus

Claims

When measuring over time the sound of the frequency within the range of 5 kHz to 80 kHz generated from the roasted coffee beans and roasting device,
An average value calculating step of calculating an average value of sound pressure during the absence of goby;
The average value of the sound pressures is used as a threshold value, and the measured sound pressure exceeds the threshold value and is selected from a period from 0.1 milliseconds to 100 milliseconds before the time point when the threshold value is exceeded. In a time span of 1 ms to 99 ms, coffee beans are detected when sound pressure of n times or more (n is more than 1/6 and less than 1/3) of the threshold is not detected. And a goby detection step for judging that the coffee beans have a goby detection method.
The coffee bean goby detection method according to claim 1, wherein the sound is measured continuously over 2 minutes after the start of roasting.
The goat detection method for coffee beans according to claim 1 or 2, further comprising an initial carbon monoxide amount measuring step of measuring an amount of carbon monoxide generated from the coffee beans being roasted at least 2 minutes after the start of roasting.
4. When measuring the sound over time, the sound generated from the coffee beans during roasting is measured at a sampling period of 0.2 milliseconds or less. The coffee beans according to any one of claims 1 to 3 Goby detection method.
5. The coffee bean detection according to claim 1, wherein the coffee bean detection is detected based on a sound pressure of a frequency having a half width of 0.1 kHz or more and 50 kHz or less. Method.
A sound measuring unit that measures sound over time in a frequency range of 5 kHz to 80 kHz generated from roasted coffee beans and a roasting device;
An average value calculation unit for calculating an average value of sound pressure during the absence of goby;
The average value of the sound pressures is used as a threshold value, and the measured sound pressure exceeds the threshold value and is selected from a period from 0.1 milliseconds to 100 milliseconds before the time point when the threshold value is exceeded. When the sound pressure of n times or more (n is more than 1/6 and less than 1/3) of the threshold is not detected in the time span of 1 to 99 milliseconds, coffee beans goby is detected. A coffee bean detection device having a goby detection unit for judging.
When measuring over time the sound of the frequency within the range of 5 kHz to 80 kHz generated from the roasted coffee beans and roasting device,
An average value calculating step of calculating an average value of sound pressure during the absence of goby;
The average value of the sound pressures is used as a threshold value, and the measured sound pressure exceeds the threshold value and is selected from a period from 0.1 milliseconds to 100 milliseconds before the time point when the threshold value is exceeded. When the sound pressure of n times or more (n is more than 1/6 and less than 1/3) of the threshold is not detected in the time span of 1 to 99 milliseconds, coffee beans goby is detected. Goby detection process to judge,
A control step of changing roasting conditions based on detection of the goby, and a method for producing roasted coffee beans.