JP2791008B2 - Control method of tape transfer device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an apparatus for directly transferring a tape (for example, a magnetic tape) from reel to reel, and is particularly suitable for conveying a tape at high speed and with high precision. And a tape transfer device. [Prior art] A conventional apparatus is described in Japanese Patent Application Laid-Open No. 60-140559,
In the process of positioning the magnetic tape at the initial position, the magnetic tape is started / stopped, the characteristic difference between the reel drive systems is detected, the correction value is determined and held using the difference, and after positioning at the initial position, In this case, the motor operation signal is corrected using the correction value. However, the causal relationship between the characteristic difference between the respective reel drive systems and the correction value has not been sufficiently considered. By calculating and controlling the drive current of each motor that drives the supply reel and the take-up reel based on the diameter of the tape wound on each reel, the speed and tension of the tape in the magnetic head section are kept constant. It is known to do so. See, for example, JP-A-53-144307 and JP-A-52-62004. [Problems to be Solved by the Invention] The prior art described above does not sufficiently consider the causal relationship between the detected characteristic difference between the reel drive systems and the correction value, and cannot determine an appropriate correction value. However, there has been a problem that the adaptive correction cannot be performed well within a short time.
For example, the signal of the characteristic difference is a binary signal of ON / OFF,
Depending on how the correction value is set, there is a problem that the convergence does not occur, and a number of trials are required until the correction is completed. SUMMARY OF THE INVENTION An object of the present invention is to provide a control method of a tape transfer device capable of performing appropriate correction of a drive command for compensating for variations in characteristics of a reel drive system and performing stable tape transfer within a short time. . [Means for Solving the Problems] The object of the present invention is to, when starting the tape, determine the time required for one of the reels to stop from a stopped state until the tape reaches a predetermined reference speed, and the average tension error of the tape within the time. Calculating the acceleration errors of the two reel drive systems using the two parameters, the time and the average tension error, and correcting the drive commands to the two reel drive systems so as to eliminate the respective acceleration errors. Is achieved in. Further, the object is to calculate a tension error by calculating a difference between a tape tension and a target tension at the time of acceleration or deceleration of the tape transfer device, correct a drive command to a reel drive motor based on the tension error, and then correct the error. This is achieved by controlling the motor in accordance with the given drive command. [Operation] A characteristic error between the two reel drive systems, particularly a characteristic variation in a transient state during acceleration / deceleration, is a bottleneck in realizing stable tape running. Stable tape running is necessary for a transient state during acceleration and deceleration, but in the present application, description will be given below of acceleration. At the time of acceleration, when the tape is started, the acceleration error of the two reel drive systems is directly obtained from the time from the stop state to the predetermined reference state and the average tension error of the tape within this time. Since this acceleration error is caused by the already given drive command, stable tape running can be realized by correcting the drive command so as to eliminate this acceleration error. In the present invention, a correction coefficient for a motor drive command is obtained from a state amount such as a tape tension at the time of acceleration or deceleration during tape running, a tape speed, a tape feed amount, a time from a stopped state to a predetermined reference speed, and a motor current. And the motor driving command is corrected by the correction coefficient at the time of the next tape running operation, that is, at the time of deceleration or acceleration. By repeating this correction, that is, by performing learning, appropriate correction can be performed. EXAMPLES Hereinafter, the present invention will be described in detail based on specific examples. FIG. 1 is a block diagram of a tape transfer device according to an embodiment of the present invention. In this apparatus, the running direction of the tape is arbitrary, but the following description will be made on the case where the running direction of the tape is from the reel 7 to the reel 6 side. Therefore, the reel 7 is referred to as a supply reel, and the reel 6 is referred to as a take-up reel. This apparatus does not have a tape buffer such as a vacuum column, and drives the reel motors 8 and 9 to rotate the reels 6 and 7 according to the motor drive signal determined by the digital controller 54 as described later. Is sent out from one of the reels, passes through the magnetic head 2 at a predetermined speed, and is wound on the other reel. The reel motors 8 and 9 are driven by drive currents 65 and 66 output from power amplifiers 59 and 60 formed of constant current amplifiers, respectively, and inputs 56 and 58 of the power amplifiers 59 and 60 are DA converters 55 and 5, respectively.
7 is the output. The DA converters 55 and 57 receive the motor drive signal data as digital signals 67 and 68 from the digital controller 54 via the output ports 52 and 53, respectively, convert the data into analog signals, and output them. The digital controller 54 controls the motors 8 and 9 and controls reading and writing of data from and to the magnetic tape 1 via the magnetic head 2. An appropriate tachometer 17 is attached to the motor 8, which generates one pulse C (18) for each rotation of the take-up reel 6 connected to the motor 8, and a pulse counter 19.
Send to A fine tachometer 21 is attached to the motor 9. The fine tachometer 21 similarly detects the rotation amount of the supply reel 7 and outputs a number of fine pulses A proportional to the rotation amount of the supply reel 7.
(23) The direction counter 25 with direction discrimination and the pulse counter 19
Send to The fine tachometer 21 simultaneously outputs a fine pulse B (22) having a phase shift of 90 degrees and sends it to the period counter 25 with direction discrimination. Therefore, the counter 25 can determine the rotation direction of the supply reel 7 by examining the phase relationship between the fine pulses A and B. The direction counter with direction discrimination 25 detects the moving direction of the magnetic tape 1 based on the phase relationship between the two, and outputs the rotation direction signal 27 to the input port
Send to 32. The direction counter with direction discrimination 25 counts the pulse period of the fine pulse A (23) using the clock 24 output from the digital controller 54 through the output port 63, and counts the count value n ₂ (26) to the digital controller. Send to 54 input port 33. The pulse counter 19 outputs the number of fine pulses A (23) per rotation of the take-up reel 6, that is, a counter value n ₁ (20) indicating the rotation amount of the supply reel 7, and outputs the same to the input port 30 of the digital controller 54. send. The tension sensor 10 detects the tension of the magnetic tape 1 from the pressure value in the tape guide 5 and sends a signal to the detector 11. Detector 11 outputs a measured value f _a tape tension (12), and sends the compensation filter 14 and AD converter 28. The compensation filter 14 receives the target tension value 13 and feeds back the compensation signal ic ₁ (15) to the power amplifier 59 on the take-up reel side so that the tape is moved while maintaining the predetermined tension. Signal ic ₂
(16) is fed back to the power amplifier 60 on the supply reel side. The AD converter 28 samples the tape tension measurement value f _a (12) using a clock (24) output from the digital controller 54 via the output port 63, converts the sample into a digital signal, and converts the tape tension measurement value f _a (29) is output and sent to the input port 31 of the digital controller 54. Input ports 30 to 33 are provided in the digital controller 54, and signals n ₂ (20) and f _a (2
9) In response to F / B (27) and n ₁ (26), a calculator with built-in memory
Send to 34. The arithmetic unit 34 has a function of controlling the entire digital controller 54, and the reel motor 8,
9 is driven to rotate the reels 6 and 7, and a motor drive command for controlling the tape speed and tension of the magnetic tape 1 when passing through the magnetic head 2 to predetermined target values is determined. The computing unit 34 computes control parameters necessary for determining the motor drive command using input data sent from the input ports 30 to 33 and mechanism characteristic data stored in advance in a memory. . Then, the control parameters, which are the outputs, are stored in the registers 61, 3 at appropriate time timing.
The feed control parameters are updated to 5, 36, 37, 38, 40, 42 and 43. The data signal (26) sent to the input port 33 is sent to the arithmetic unit 34 and the adder 44. On the other hand, the adder 44 has a register (−
receives the output from the n _ref) 43 output from the input port 33, performs a comparison of the two inputs the result of the (n ₁ -n _ref) to the limiter 45. The limiter 45 outputs the output signal (46)
To the multipliers 48 and 49. The multiplier 48 receives each output of the limiter 45, the register 35, and the register 36, multiplies each signal, and sends the output to the adder 50. The adder 50 receives the outputs of the multiplier 48, the register 61, and the register 38, adds the signals, and sends a motor drive command for driving the reel motor 8 as the output to the output port 52. The multiplier 49 receives each output of the limiter 45, the register 37, and the register 42,
The signals are multiplied and the output is sent to the adder 51. The adder 51 receives the outputs of the multiplier 49 and the register 40, adds the signals, and sends a motor drive command for driving the reel motor 9 as the output to the output port 53. A work register 64 is provided in the arithmetic unit 34,
Here, a limit value for determining whether the tension error is within an allowable value is set. In FIG. 1, reference numerals 3 to 5 denote tape guides. Next, the operation of FIG. 1 will be described with reference to FIG. Second
The figure is an operation flowchart showing the operation in the digital controller 54. First, the process of step F10 in FIG. 2 is performed. This is because the arithmetic unit 34 sets the correction coefficient Ca = 1.0 for correcting the drive command for the motor 8 in the register 36, the correction coefficient Cb = 1.0 for correcting the drive command for the motor 9 in the register 37, and the register 61 , Tension current correction value itc = 0
Is a step of setting Subsequently, the process proceeds to step F20, where the calculator 34 calculates and stores the radii R ₁ and R ₂ of the reels 6 and 7. The calculation of this radius may be obtained by any algorithm, but here, it is obtained as follows. Since the length of the tape wound by the take-up reel 6 during one rotation is equal to the length of the tape drawn out by the supply reel 7, the following expression (1) is established. Since the length of the tape does not change, the sum of the lengths of the tape wound on each reel is constant, and the following expression (2) is established. Where R ₁ : radius of reel 6 R ₂ : radius of reel 7 n ₂ : count value of pulse counter 19 (20) N: number of pulses per rotation of fine tachometer 21 R ₀ : tape on each reel is not wound Time radius L: Tape length T: Tape thickness When the two equations (1) and (2) are solved, the radius of each reel becomes the count value n ₂ (20) as a variable as in the following equation (3). Can be determined. The calculator 34 calculates the radius of each reel by the calculation of the above equation (3). This calculation is performed each time a new count value n ₂ is input through the input port 30. R ₀ , L, T, N constants (part of the mechanism section data) required for this operation are
It is stored in an internal memory in advance. Then, the process proceeds to step F30, with a radius R _1, R ₂ obtained by the calculation, and the speed control gain G _1, G ₂ for driving the reel motor in the reel driving system of the respective
Drive command (tension current command) If, by calculating a set count value n _ref corresponding to the angular velocity required for the tape speed and set speed V _0, and sets the result in register. That, G ₁ in the register 35, G ₂ in the register 42, Is in register 38, Is set in the register 40, and -n _ref is set in the register 43. Each parameter in this embodiment is obtained as follows. First, the control gain set in the registers 35 and 42 is
It is obtained from the following equations (4) and (5). Here, G ₁ : Speed control compensation gain of motor 8 G ₂ : Speed control compensation gain of motor 9 K _T : Motor torque constant J ₁ : Inertia when winding tape is not wound J ₂ : Tape of supply reel Inertia when not rolled α Constant of magnetic tape The calculator 34 with built-in memory reads K _T , J ₁ , J ₂ , α, R ₀ from the mechanical part data stored in advance in the memory and stores it in the memory. Using the data of the stored reel radii R ₁ and R ₂ , G ₁ and G ₂ are calculated by the equations (4) and (5), and the results are sent to the registers 35 and 42. This value is the radius R of the reel in the calculator.
_The value is updated each time ₁ and R ₂ are rewritten. Next, the control constants registered in the registers 38 and 40 are obtained from the following equations (6) and (7). The arithmetic unit 34 with a built-in memory uses the data of K _T , f _ref , F and the reel radii R ₁ , R ₂ stored in the memory from among the mechanism unit data stored in the memory in advance, According to equations (6) and (7) And sends the result to each of the registers 38 and 40. This value is calculated every time the reel radii R ₁ and R ₂ stored in the calculator are rewritten. The value -n _ref set in the register 43 is obtained from the following equation. Here, n _{ref is the} number of clock pulses within the pulse period of the tachometer 21 when the radius is R ₂ (corresponding to the angular velocity required to set the tape speed to the target speed V ₀ ) V ₀ : target value of the tape speed t _d : clock The pulse period calculator 34 calculates _{nref according} to equation (8), adds a minus to the result, and sets the result in the register 43. This calculation is performed each time the radius R ₂ is updated. Then, the process proceeds to step F40, where each motor drive command i
₁ and i ₂ are calculated and output via the respective output ports 52 and 53. This operation is It is performed as follows. However, at this stage, C _a = C _b
= 1.0, (see step F10) i _tc = 0, so, C _a, correction by C _b or tension correction value i _tc no or work al. By outputting the drive commands i ₁ and i _{2 for} the respective reel motors, the process proceeds to step F50, and the reels are driven. At this stage, the digital controller
Numeral 54 is a time t _a from the stop state, which is two parameters necessary for calculating the acceleration error of each reel drive system during acceleration, until the predetermined reference speed Vs (≧ V ₀ ) is reached, and The average tension error f _av of the tape at is obtained. Here, the theoretical basis for _{obtaining the} acceleration error from t _a and f _av is as follows. First, the reel motor
The actual accelerations ₁ and ₂ generated at 8 and 9 can be expressed as follows. Here, K ₀ : set acceleration β ₁ : acceleration error coefficient of the motor 8 β ₂ : acceleration error coefficient of the motor 9 The tension error f _s generated when the motors 8, 9 rotate at the accelerations ₁ , ₂ is represented by the following equation. become that way. Here, t: time after the start of acceleration K _a : spring constant of tape The coefficients β ₁ and β ₂ in the equation (12) are values obtained by linearly approximating the specific errors of the respective reel drive systems and accelerating the errors. is there. From equations (12) and (13), equation (14) is obtained. Next, the time t _s to reach the reference speed V _s are accelerated by setting the acceleration _{K 0 (= V 0 / K} 0), between the time t _a which actually reaches V _s, the following relationship is there. K ₀ β ₂ t _a = V _s = K ₀ t _a (15) From equation (15), β ₂ is From Equations (16) and (14), β ₁ is From these results, β ₁ and β ₂ can be calculated from t _a and f _s . However, the tension error f _s is because it contains a disturbance such as an error or mechanism vibration on measurement, low reliability of the data. Therefore, the average value f _av of the tension error is used. The specific detection method of t _a and f _av is as follows. First,
It will be described how to obtain the time t _a from the start-up reaches the reference speed V _s. The count value n ₁ indicating the tape speed is
Since the input to the arithmetic unit 34 compares the count value n _s corresponding to the reference velocity V _s which is previously stored with the n _1, n ₁ ≦
the time until the n _s be detected using an internal clock. This time is t _a. Further, the average tension error f _av can be obtained using the detected tension value f _d input to the calculator 34. FIG. 3 shows the state of the detection. Since the target tension value f _ref is known in advance (stored in the memory), f _ref is subtracted from the detected tension value f _di input every sampling time, and this is added. Then, by dividing this by the number of samples k, the average tension error f _av
Is obtained. By the way, when the processing in step F60 is completed, step 7F0
Proceed to. In step F70, performs determination of whether the detected tension error resulting from the tension value f _s or an average value f _av that exceeds limit value (allowable value). If the result of this determination is that the value falls within the allowable value, no further correction operation is performed. If the tolerance is exceeded, step F8
Go to 0. In step F80, f _av are already determined, using a t _a, acceleration error coefficient of each reel driving system beta _1, it calculates the beta _2. This calculation is based on equations (16) and (17). Next, in step F90, using this acceleration error coefficient, a correction coefficient for each reel drive system is obtained and stored. In this embodiment, β ₁ , β
Reciprocal of ₂ Is set, and this is obtained from the radius and the motor drive command set in the register Is corrected by multiplication. Next, in step F95, a correction value _itc for correcting the response in the steady state among the characteristic errors of the reel drive system is calculated and stored. In this embodiment, the operation result is stored in the register 61.
The correction is performed by setting _itc and adding the drive command with the adder 50. Here, the determination of the correction value _itc will be described. The average tension error f _av described above and the reel diameter R ₁
Is used, _itc is calculated by the following equation. The correction coefficient and the correction value determined as described above are set in the registers 61, 36, and 37. Then, the process proceeds to step F40, where the corrected drive commands i ₁ and i ₂ of each reel motor are calculated using these set parameters, and the tape operation is actually performed based on the result. The drive command obtained as described above is for compensating the drive command based on the variation in the characteristics of the actual drive reel system, and is close to optimal compensation. Therefore, at this stage, the acceleration characteristics and the tension are controlled so as to be close to the initial set values.
However, the actual operation result after the correction
If the correction is still insufficient at this stage, the processing of steps F80 to F95 and the processing of steps F40 and F50 based on the result are further executed. The effect of the present embodiment is that, in the process of extracting the characteristic error of the reel drive system, the characteristic error can be correctly extracted, so that the characteristic error can be appropriately corrected. The point is that correction can be performed by one tape operation. FIG. 4 is a view showing a main part of another embodiment of the present invention. FIG. 4 differs from the configuration of FIG. 1 only in the digital controller 54. Digital controller 54
In FIG. 1, the output of the registor 37 is a multiplier.
Although sent to 49, it is different from FIG. 4 in that it is sent to both the multiplier 49 and the multiplier 48, and the other configuration is the same. The parameters for extracting the characteristic error of the reel drive system use t _a and f _av as in the example of FIG. The difference is
An algorithm for determining a correction coefficient and a correction value from the two parameters and a correction method. FIG. 5 is a flowchart showing a process of determining a correction coefficient and a correction value in this embodiment. In the figure, the processes up to the tape operation (steps F10 to F50) are the same as those in the previous embodiment, and the description is omitted. In the first tape operation, the digital controller 54 detects the parameter f _av in steps F86 and F87, and determines _a correction coefficient C _a for correcting the response in the transient state among the characteristic errors of the reel drive system. I do. The specific method of detecting f _av is the same as in the previous embodiment. The correction coefficient _Ca is calculated and determined by the algorithm of Expression (16) obtained from Expressions (11) and (15). In this process, the unknown coefficient β ₂ is calculated as 1. So, from equation (17) At this time, the previously set tape transfer direction is referred to. This is because the sign of f _av is inverted _{depending on} the tape transfer direction. The determined correction coefficient _Ca is set in the register 36. Next, the digital controller 54 performs a second tape operation with the new correction coefficient. The second tape operation by detecting the parameter t _a, determines the correction coefficient C _b proceeds to step F88. The specific method of detecting t _a is the same as in the previous embodiment, and a description thereof will be omitted. The correction coefficient C _b value is β
₂ is calculated and determined by the equation (16). The determined correction coefficient _Cb is set in the register 37. Then, step
Proceeding to F95, a correction value _itc for correcting the response in the steady state among the characteristic errors of the reel drive system is determined. The method of obtaining this is the same as in the previous embodiment, and the description is omitted. Computing unit with built-in memory
34 operates the tape again with the new correction value to check whether the adaptive correction has been performed correctly. Subsequent processing steps are the same as in the previous embodiment, and a description thereof will be omitted. This embodiment is characterized in that two parameters t _a and f _a are detected twice in the process of extracting the characteristic error of the reel drive system. Since the tape speed and the tension are mutually related characteristics, the present embodiment solves the problem that the characteristic error is particularly large and the point that the detection accuracy of t _a and f _a deteriorates is eliminated. Aim. The effect of this embodiment is that the characteristic error of the reel drive system is extracted twice, so that the detection accuracy of the error is improved and more appropriate correction for the characteristic error can be realized. In the above embodiment, the two parameters
Although the order of detecting t _a and f _a is determined by performing f _a first and then performing t _a , the order may be reversed. In this case, it is the same as dividing the detection process of the previous embodiment into two. FIG. 6 is a diagram showing another embodiment of the present invention. FIG. 6 shows the configuration of FIG.
Has been changed so that it can be detected in addition to the pulse c18, sending the fine pulse of the tachometer 17 to the period counter 70,
The cycle counter 70 receives the clock (24), counts the pulse cycle of the fine tacho pulse 69, and sends a count value n ₃ (71) to the input port 72 of the digital controller 54. There is a difference in inputting the count value n ₃ sent from. Other configurations are the same as those in FIG. The parameter for extracting the characteristic error of the reel drive system is Is different in using. FIG. 7 is a flowchart showing a process of determining a correction coefficient in this embodiment. In the figure, the process up to the tape operation is the same as in the previous embodiment (FIG. 2), and a description thereof will be omitted. After the tape operation, the digital controller 54 sets two parameters in step F91. Is detected, and the correction coefficients C _a and C _b for correcting the response of the reel drive system in the transient state in the characteristic error are determined. Parameter t _a used in this embodiment not described are the same as parameters t _a used in the embodiment of Figure 1. Now,
New parameters The tape speed of the take-up reel side, corresponds to the time to reach the tape speed V _s of the time t _a which is in the transient state, there is relation of the following equation (20), the error coefficient of acceleration of the motor 8 beta
₁ can be detected. Here, β ₁ : acceleration error coefficient of motor 8 K ₀ : set acceleration t _s : measurement time Further, acceleration error coefficient β ₂ of motor 9 can be detected from equation (16). Therefore, Can be selected as a parameter for detecting the characteristic error of the reel drive system. Next, two parameters It will be described in detail specific detection methods, but for t _a is the same as in the previous examples not described. The count value n ₃ indicating the tape speed on the take-up reel side is sent to the arithmetic unit 34 with a built-in memory via the input port 72. The arithmetic unit 34 compares the count value n _s1 preset in the memory with n _3, and measures the time when n ₃ ≦ n _s1 as the number of clocks using the built-in clock (24). Note that n _s1 is calculated using the following equation and set in the memory. The arithmetic unit 34 with a built-in memory stores the detected parameter t
_a1, using t _a2, correcting coefficient C _a, calculates the C _b decision to correct the response of the inner transient state characteristic error of the reel drive system. The determined correction values C _a and C _b are set in registers 36 and 37, respectively. Next, among the characteristic errors of the reel drive system,
The determination of the correction value _itc for correcting the response in the steady state is the same as in the previous embodiment, and the description is omitted. As described above, two types of correction values for correcting the characteristic error of the reel drive system are determined. The arithmetic unit with a built-in memory operates the tape again with the new correction coefficient and correction value in order to check whether the adaptive correction has been correctly performed. Subsequent processing steps are the same as in the previous embodiment (FIG. 1), and will not be described. In the present embodiment, in the process of extracting the characteristic error of the reel drive system, since it is detected from the tachometer of the motor shaft, it is hardly affected by the vibration of the mechanical part and the detection accuracy is high, so that an appropriate correction coefficient and correction value are determined. it can. Further, correction can be performed by one tape operation. [Effects of the Invention] As described above, according to the present invention, an appropriate drive command for compensating for the characteristic error of the reel drive system is obtained, and the tape is driven by this. Can be realized.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is an operation flowchart of the embodiment of FIG. 1, and FIG.
diagram for explaining the calculation of f _av, Figure 4 is a block diagram of another embodiment antelope portion of the present invention, FIG. 5 is an operation flow diagram of an embodiment of FIG. 4, FIG. 6 is the present invention FIG. 7 is a block diagram showing another embodiment, and FIG. 7 is an operation flowchart of the embodiment shown in FIG. 1 ... magnetic tape, 2 ... magnetic head, 6,7 ... reel, 8,9 ... reel motor, 10 ... tension sensor, 17 ...
Tachometer, 19: Pulse counter, 21: Fine tachometer, 25: Period counter with direction discrimination, 28: AD converter, 34: Operation unit, 36: Register (C _a ), 37: Register ( C _b ), 45 …… Limiter, 48,49 …… Multiplier, 50,
51 ... Adder, 54 ... Digital controller, 59,60
... power amplifier, 61 ... register ( _itc ).

Continuation of front page    (72) Inventor Munetake Kanna               2880 Kozu, Odawara City Hitachi, Ltd.               Odawara Factory                (56) References JP-A-60-136937 (JP, A)                 JP-A-59-148175 (JP, A)                 JP-A-55-150152 (JP, A)

Claims (1)

(57) [Claims] 1. A method of controlling a tape transfer device for giving a drive command to two reel drive motors so that a tension of a tape transferred between a pair of reels becomes a predetermined tension. A correction coefficient is calculated from the state amount at the time of acceleration or deceleration during the tape running period, and a drive command to each motor used after the next deceleration time or acceleration during the tape running period is calculated. A method for controlling a tape transfer device, wherein the correction is performed using a coefficient. 2. 2. The tape transport according to claim 1, wherein the state quantity is any one of a tape tension, a tape feed amount, a time from a stop state to a predetermined reference speed, a tape speed, or a motor current. How to control the device. 3. The state quantity uses a tape tension and a time from a stop state to a predetermined reference speed, or a time from each stop state to a predetermined reference speed on each of the pair of reels. A method for controlling a tape transfer device according to claim 1. 4. In a control method of a tape transfer device for giving a drive command to two reel driving motors so that a tension of a tape transferred between a pair of reels becomes a predetermined tension, a method of controlling a tape when the tape transfer device is accelerated. A correction coefficient for correcting a motor drive command for reducing a tension error with respect to a predetermined tension is calculated and held, and a drive command to the two reel drive motors used for tape running after deceleration is calculated. A method for controlling a tape transfer device, wherein the method is corrected by a correction coefficient. 5. In a control method of a tape transfer device for giving a drive command to two reel drive motors so that a tension of a tape transferred between a pair of reels becomes a predetermined tension, a method of controlling a tape when the tape transfer device is decelerated is used. A correction coefficient for correcting a motor drive command for reducing a tension error with respect to a predetermined tension generated is calculated and held, and a drive command to the two reel drive motors used for tape running after acceleration is calculated. A method for controlling a tape transfer device, wherein the method is corrected by a correction coefficient.