JP2513451B2 - Door drive device for elevator car by linear induction motor - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an elevator car door drive system using a linear induction motor, and more particularly to a door drive system for an elevator car that reduces an audible perturbation of a force induced in a door of an elevator car by a linear induction motor. Regarding the device.

[0002]

PRIOR ART US Pat. No. 4,30 owned by the applicant.
The elevator door operation mechanism as shown in No. 5,481 uses a rotary motor that operates a complicated lever mechanism. In an attempt to simplify and thereby reduce cost, equipment, adjustment and maintenance costs, Applicant's US Pat. No. 3,891,907 and US Pat.
No. 29,203 (filed Mar. 10, 1993) proposes the use of a linear induction motor. The use of a linear induction motor can reduce mechanical parts more than the electric induction secondary winding. Here, on the secondary side, a door can be attached and the door is driven in the opening direction and the closing direction according to the current applied to the primary winding.

[0003]

The use of a digital computer to generate a variable phase motor current waveform to drive a variable voltage, variable frequency linear induction motor has the advantage of producing a pure sine wave waveform that produces the proper linear induction motor characteristics. Is only an approximation of. Distortions in the digitally generated waveform caused unwanted noise and increased motor heating. Techniques for improving current drivers by closed loop current control implemented through digital signal processing are known in the art. However, closed loop current control becomes extremely complicated with the increase in cost of the device.

Using open current loop control and pulse width modulation approximation of the desired sinusoidal drive signal, both contribute significant perturbations at frequencies in the human audible range. The perturbation is
A. C. Drive frequency harmonics (between 4 and 50 hertz) and pulse width modulation switching frequency double harmonics (1
0 kilohertz).

The present invention has been made in view of the above problems, and an object thereof is to provide an elevator car door drive device that reduces noise in an elevator car door drive system using a linear induction motor. .

[0006]

In order to achieve the above object, the method of the present invention comprises a primary side comprising three windings connected in Y and an elevator door for providing linear motion. A three-phase variable voltage / variable frequency linear induction drive motor having a secondary side, a three-phase drive signal source having polarities alternating at a frequency of 50 hertz or less, and three capacitors connected in a delta shape, and Each inductor is connected to one free end of the winding, one end of each inductor is connected to the corresponding connection point of the capacitor pair, and The other end of the inductor is 3
Connected to the corresponding phase of the phase drive signal source, the capacitance of the capacitor is selected with respect to the inductance of the motor winding to form a low pass filter having a break point between 100 and 200 Hertz. It is characterized by Also, the capacitance of the capacitor is selected to form a low pass filter having breakpoints on the order of 140 hertz.
A three-phase drive signal source supplies current to the winding in an open loop.

[0007]

According to the present invention, a drive system for an elevator car door includes a three-phase variable voltage, variable frequency induction motor, the induction motor being coupled to the car door to provide a linear motion to the car door. And a primary side that includes three windings that are Y-connected.
A three-phase drive signal whose polarity changes at a frequency of 50 Hertz or less is applied to the primary winding through a low pass filter.
The low pass filter has a break point between 100 and 200 Hertz and is Y-connected to the winding on the primary side of the motor.
It is composed of three capacitors and three inductors connected to the respective windings, through which a drive signal is applied. In an embodiment of the present invention, the break point of the low pass filter is on the order of 140 hertz.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described below with reference to FIGS.

Referring to FIG. 1, the elevator door 17
Is shown as a solid line in the closed position and as a dotted line in the open position. The door 17 is attached to the secondary side 18 of the linear induction motor, the primary side of which is fixed to the elevator car. In the embodiment of the present invention, the primary side 19 of the linear induction motor has six windings 22-25 (FIG. 2), and as shown by winding U
Pairs 20,21 and 22,23,24,25 are connected so as to form windings U, V, W, each forming a north pole and a south pole separated by 180 electrical degrees. ing. The windings are always driven in a three phase relationship, with the windings having phases that are 120 ° apart from each other. For example, as shown in FIG. 3, at some point winding U has a relatively small negative current, while winding V is at maximum current and winding W
Is an intermediate negative current. As is known, multiple sets of 6 windings per set are possible.

Secondary side 18 typically consists of a conductive split where the secondary current is formed by a magnetic field, the conductive split having magnetic feedback. This magnetic feedback is achieved in any implementation of the invention by an optical encoder (not shown) on which the primary side of the linear motor (e.g., the magnetic flux between the poles of the windings 20 and 21) is mounted is the secondary side 18. ), Whose encoder split has a pulse line and supplies it to a computer 30. It is provided on line 29 for a fixed time (which in this example is 1 millisecond). The number of pulses shown indicates the distance traveled at that time and hence the speed of the door, and the assembly indicates the current value of the door, as described below with respect to Figures 6-11. In turn, the computer 30 uses this information to derive a force command which achieves the desired velocity for the position motion profile of the door, and which is shown in FIG. The positive and negative of the driver 31 is fully described in terms of voltage selective on and off.The voltage supplied by the driver 31 produces the desired current as shown in FIG. As shown in Fig. 1, the linear induction motor on one side requires linear force of 2 to 5 times the normal force without linear force. Move to)
Cannot be supplied. Normal Force activates the solenoid on the primary side and attracts the magnetic material on the secondary side towards the primary side. In the art, linear forces have been selected to provide a given acceleration by selecting the frequency and amplitude of the alternating current in the primary winding that produces such a result. The relationship between normal force and linear force is shown roughly in FIG. 4 in order to show it as a function of the frequency and amplitude of the current in the primary winding. The selection of the frequency and amplitude of the linear force supply in various combinations provides maximum electrical efficiency in using the linear induction motor.

FIG. 5 shows a typical sine wave of AC current that can be used to drive a linear induction motor. As shown in Sine wave, it is known to generate a sine wave comprehensively with steps, and its magnitude is
It is the average magnitude of the sine wave during the period of the step. A constant magnitude (not shown in FIG. 5, but equal to the maximum magnitude of the sine wave) is equal to the value corresponding to the step shown on the sine wave (here 1 ms). A) conventional pulse width modulation as it is introduced into the winding for a selected time period to produce the average magnitude. In the example, instead of a pulse width that varies between 0 and 1 millisecond, a pulse width that varies between 0 and 64 microseconds is used, the same pulse width obtained over one millisecond.

Therefore, in order to obtain an approximate sine wave at 11 °, during a second one millisecond period of 50 milliseconds, a drive current of 20 Hertz and each of 16 pulses of about 12 microseconds is applied. It is shown in the center of FIG. Similarly,
Sixteen negative pulses of voltage, each with a duration of 16 microseconds at 195 °, are applied to the windings. Thus, the current in the winding is a fixed amplitude voltage for any period in one millisecond period, which is a function of the current sine angle along the sinusoidal drive function of the desired current (ie force) times (that is, according to the invention). This is accomplished by this example by applying a constant) throughout the life cycle of the device. The current in that winding is separated by 120 ° between the three windings (as in Figure 3). In the present embodiment, normal force is used to offset the clockwise torque, which is the force in the right direction on the upper right corner of the door 17, and when the door begins to open, the normal force is applied. As a result of the above, by applying a vertical force in the direction of the primary side 19 of the linear induction motor, there is a counterclockwise torque about the gravity at the center of the door. All of this is described in the above-mentioned prior US application.

Referring to FIGS. 6 and 7, each millisecond real-time interrupt is provided to reach one millisecond through entry point 34. First test 35
Determines whether the ramp down flag is set. This is a flag that is set when the door is fully open or fully closed, and, as will be fully explained below with respect to FIG. 10, the residual voltage is slowly dropped in order to avoid drooping. Specifies the ramp down period. Whenever the door is opened and closed, the flag 35 is set and the result of the test 35 is positive. However, most of the time the flag is not set and the negative result of test 35 reaches test 36 to determine if the door is fully closed. As considered in this example, test 36 is based on a physical switch on a normal elevator door indicating complete door closure. However, the door position or other indicators of a fully closed door can be used if desired.

In the normal case, the door is fully closed, as in other cases than landing. In such a case, the positive result of test 36 reaches test 37 to see if a door open command signal was sent to the door controller by the elevator controller. If the door is not fully closed (test 36) and there is no command to open the door (test 37), the one millisecond interrupt routine of FIGS. 6 and 7 ends and the computer returns to other programming through return point 38a. Eventually, the elevator will be close to landing and when it is within the landing area, a door open will be provided to the door controller and the result of test 37 will be positive. This leads to a pair of steps 38, 39 that set a door open flag which is used as described below, and the 1.5 millisecond timing is used to determine when the door operation should be complete. In this example, the door moves approximately 55 cm and is fully opened after about 1.4 seconds.
Of course, this timing can be adjusted in any case suitable for the particular door operating system to which the invention applies.

For 1.5 seconds, the time is 2.5 while the door is closed.
A timeout of seconds has begun and hence, here
They are called 2.5 second clock and 1.5 / 2.5 second clock. The actual process for formulating the command and the applied voltage to the primary side 19 begins at step 42 in the pulse counter of the encoder counter, which is advanced by the optical position sensor pulse on line 29 (FIG. 1). It is remembered by the word called "count". The counter is then restarted with a millisecond counter to begin updating immediately in step 43. Step 44 counts the velocity constant times (kv) over one millisecond period to calculate the average door velocity. The old position (where it was the start of one millisecond period) is taken as the adjusted indication of the deviation in the position, which is a count of the position constant (kp) times the position obtained during the current one millisecond. By adding, step 45 determines the current position of the door.

Then, in step 46, the old position is updated to equal the new position. Subroutine 49 issues a speed finger using the door open constant (ko) in a manner that produces a speed profile as a function of door position, similar to that in the aforementioned application. This profile is shown overwritten over the leading edge of the door in FIG. In the usual way, the speed error is taken as the difference between the speed command calculated in step 50 and the current speed (determined in step 44).

Then, the force command is generated in a subroutine as a proportional and integral function of the velocity error using the proportional constant (kp) and the integral constant (ki). The force command is processed through a low pass washout filter subroutine in a well known manner and the filtered result is step 5
3 multiplied by the amplitude constant (ka), which gives an amplitude factor indicative of the current required to obtain the desired linear force due to the amplitude-frequency relationship chosen to drive the linear motor at the desired speed. appear.

As is well known, a linear induction motor requires a magnetizing current to establish a magnetic field that produces a secondary current in the conductive secondary 18 and thus to establish a magnetic field that produces a magnetizing current. . This is the primary side 19
Is a constant alternating current for each winding U, V, W of the. This current is a square wave with linear force and the two are combined in subroutine 55 as the square root of the sum of the rectangles. Since the magnitude of the magnetizing current is constant and all operations in one millisecond interrupt routine are the same, a number equal to the number of pulse width magnitudes generated in step 54 is easily stored and the subroutine 55
There is a pulse width associated with the number as an output flow from.

According to the present invention, the magnitude of the magnetizing current can be approximately one-quarter that of the conventional one used to achieve maximum current efficiency. In this example, this is about 32.
Expressed as a pulse width count of. Subroutine 5
After 5 is completed, the program proceeds through the transformation point 56 to the phase portion, as shown in FIG.

In FIG. 8, the first step 59 is the count generated in step 54 (FIG. 7):
Determine if the force-induced amplitude is equal to or greater than 8. It is the first break point of a linear approximation of the desired slip frequency as a function as shown in FIG. 3, which is shown in FIG. 1 by the dotted curve. However, in order to operate the door, FIG.
It has been determined that the straight line approximation shown by the solid line in 4 is sufficient. This simply makes the process easier.

However, table or square root formula calculations may be used if desired in the practice of the invention. P,
If the W, M values are not equal to or greater than 8, there is a linear slope of 4 Hz for the desired slip frequency of 8 and is generated at step 60. But the count is 8
If greater, the positive result of test 59 reaches test 61 to determine if the count is greater than or equal to 128. If so, step 62 is reached and 18
Generate a slip frequency of Hz. But the count is 8
, Between 128 and 128, the negative result of test 61 leads to test 161, where there is a 4 Hz plus multiple of P, W, M.
It is set equal to -8 counts and follows the solid line in FIG.

The slip frequency thus generated is
It is the slip frequency between the secondary and the primary of a linear induction motor, which is known to be necessary in order to generate a suction force to move the secondary side by the primary side in the desired manner. Such frequencies generated on the primary side become ineffective due to door movement once the door is opened. In other words, the phase change necessary to generate the force must be added to that resulting from the movement of the secondary with respect to the primary.

In order to adjust for this situation, the phase factor related to the speed between the second and first order is generated in step 67 as the speed of the speed constant (Kv) times, and the spatial unit between the windings is To 60 electrical angles, which in the example is on the order of 16 millimeters. The slip frequency phase occurs in step 68 in a 360 ° relationship per cycle. The total phase is the sum of the phases generated in steps 67 and 68 and is calculated in step 69.
Is induced in. The actual phase achieved during the current cycle is the sum of the phases previously made along the sine wave of FIG. The current phase of step 70 is step 7
Saved for the next one millisecond interrupt cycle at 1.

As described with respect to FIG. 3, the phases for the three windings are the same at each instant. The phase generated in step 70 is the phase of the winding U in step 73, and other phases can be obtained if necessary. A test 74 determines if the door force is positive or negative (open or closed) as a function of the amplitude of step 53. If the door is open, the amplitude is positive, and the windings V and W are in steps 75 and 76 at 120 ° and 240 in this example.
° Established as a lead. On the other hand, if the door is closed and the force is negative, the result of the test is that the windings V and W are -120 out of phase U.
There is a delay (retreat) of ° and -240 °. The 1 millisecond interrupt routine associated with winding U in FIG.

In FIG. 9, the phase of winding U is tested 84.
And 85 to normalize between 0 and 359 °. The pulse width of winding U is established in step 86 as the AC amplitude of step 55. Step 55 is represented as a pulse width count.

In FIG. 15, the drive current of the winding U is shown by a solid line. For whatever reason (probably related to adjacent pole fields and magnetic hysteresis), the positive-to-negative crossover of the drive voltage sine wave is compensated by closed current loop control as well as negative-to-positive. If not done, it is known to cause a current delay. In this embodiment, there is no current feedback indicating the actual current of the motor, and the current delay caused by zero crossing cannot be corrected. Therefore, according to the present invention, the current delay is basically reduced by performing a voltage boost in the first few cycles (following each crossover through each door opening / closing operation). To achieve this, a zero cross is detected, a flag is set, and in the next few cycles a fraction of the maximum pole width is applied by adding a count equal to the pulse width during those few cycles.

In FIG. 9, test 87 determines if the zero-cross flag is set. A flag is set to determine whether the triangular sine wave of the phase for the U phase is a positive or negative number except for a few cycles following the crossover, and if not, a test 88 is reached. If it is a positive number, the result of test 88 is positive and it is determined if the phase shift of the U winding in the next millisecond cycle was a positive sign. If it is,
The number of steps is bypassed instead of zero crossover. However, if the previous cycle was negative, as in FIG. 4, the negative result of test 89 leads to step S91. Step 91 sets the zero-cross flag equal to zero (to count a few cycles where the boost signal is applied).

Then, in step 93, the sine wave of the phase of the U winding is stored for use in the next cycle (sine wave phase U). The boost number for winding U then occurs in step 98 as a fraction of exponential, whose Kb is an important fraction, for example a count of 60 and 90, or 75 in this example.
The fraction is the number of cycles boost should be applied, which is 4 to 8 or U counter (0 in step 92).
However, it may be between 1/2 and 2, which is a count of 75 in this embodiment. The index (Kb) may be one, and as shown in the example of this embodiment, what is required is
The parameter of step 78 is devised so that the current in the winding is actually close to a sine wave.

Once the boost factor occurs in step 98, the U counter value is incremented in step 99. Then, test 100 determines if the U counter has advanced to setting 6. After the zero-crossing, the value is not set to 6 in the first cycle, and if the result of step 100 is negative, step 101 is reached. In step 101, the pulse width value for the U winding (set in step 86) is calculated in step 9
It is added to the boost value calculated in 8.

As can be seen in FIG. 15, if the index (K
If b) is not less than or equal to the predetermined value, then during the first millisecond after the zero cross, the boost value is a sufficient value of (Kb), for example 7
5 counts. In the second millisecond cycle following the crossover, the fraction of step 88 is 4/5 and the third
For a cycle of 3/5 and thus 0 for the fifth cycle. The U counter is then incremented to 6 in step 99 and test 10
A positive result of 0 leads to step 102 where the zero cross flag is set. This allows the program to look for zero crossings in tests 88-90, as described above.

Subroutine 103 for winding V and subroutine 104 for winding W are the same as and execute the steps and tests 84-102 described for winding U. Once this is complete, the ramp-down portion of the millisecond interrupt routine passes through conversion point 105 to FIG.

In FIG. 9, a first test 107 determines if the door has been closed. This flag only sets when the door is closed. If the flag is not set, the result of test 107 is negative, leading to test 108 to see if the 1.5 / 2.5 second clock has timed out. In the general case, as the door opens, there will be no timeout and the end of the 1 millisecond interrupt routine, and other programming will return through return point 110.

When the door is opened or closed, 1.1 / 2 seconds after opening / closing (FIG. 6).
5 / 2.5 seconds clock timed out, test 109
If the result is negative, step 111 is reached. Step 1
In 11, the flag is set and a series of steps 112-1
At 14, the pulse width values of windings U, V, W are decremented by one count. The ramp down flag is used to determine whether the bundle of millisecond interrupt routines should be bypassed, as described above. After decrementing, each pulse width is set to steps 115-117.
Is tested to see if it has decreased to zero. If either value is not zero, the result is no and return point 11
Reaches 0.

On the next millisecond interrupt, the routine reaches entry point 34 in FIG. As a result of the ramp down flag having a 1.5 second timeout, conversion point 105
Since the result of test 35 is positive, the ramp down portion of the millisecond interrupt routine in FIG. 10 is directly reached. The 1.5 / 2.5 second clock is always in a time-out condition, except when it lasts for the duration of the door motion, and the result of test 109 is positive, leading to steps 111-114. In steps 111-114, the ramp down flag is set extra (without harming anything) and the pulse width is determined again.

Tests 115-117 determine if all of the pulse width has decreased to zero. Initially, steps 115-117 do not reach return point 110. In the following millisecond period, the millisecond interrupt routine again passes through entry point 34 to reach FIGS.
Results in the ramp down switching point 105 and again the test 109.

Since the maximum counts used in this example are always less than 256, even if the values do not have them at the end of 1.5 seconds, they will drop to zero within 256 milliseconds. . At some point (typically 1.5
Within a few cycles after the second timeout, all of the pulse width values have decreased to zero and a series of positive results from tests 114-116 leads to test 120, which
At 20, it is determined if the door is open by testing the door open flag set in step 38.

If the door is open, the flag is set, and if the test 120 is positive, step 121.
In step 122, the door open flag is reset, and in step 122, the door open flag is set. The door open flag indicates that the door is open and remains until a door close command is received from the elevator controller, as described more fully below. Step 112-11
Since it is not necessary to ramp down the count in 4, the ramp down flag is reset in step 123 and the door close flag is reset in step 124.

The purpose of the door closing flag will be described below, but there is no harm. The computer then returns to other programming through return point 110.

When the downramp flag is reset at step 123, the next time a millisecond interrupt occurs, tests 35, 36 and 109 reach test 127 to determine if the door open flag has been set. Since it was set in step 122, the routine looks for a door close command in step 128. While the door is open and passengers are moving in and out (determined by the elevator controller and the door open button in the elevator car for a certain period of time), there is no door close command and the result of test 128 is NO and return point 129 indicates another Revert to programming.

Eventually, the elevator controller commands the door to be closed and the millisecond interrupt tests 128.
Detection of the positive result of step 130 and step 131. Step 130 resets the door open flag and step 131 starts the 2.5 second clock. This is the beginning of the door closing operation. When the door is closed, it is necessary to limit the total inertia of the door for safety reasons.
It takes pulse seconds. As long as the door is closed, in each cycle, step 132 determines whether the passenger flipped the door by shutting off the light beam between the two doors and safely operated the shoe. When the door is turned over, the door closing operation fails and the door opening operation starts. Whenever an inversion occurs, the signal processing described above with respect to Figures 6-9 for the door opening operation is performed. Since the routine is a position loop, this pose is not a problem at all, it picks up the close operation at the position where the door is in. There can be some large phase increase, speed error, etc. However, this is simply the result of returning to the proper door opening routine.

If the door has not been inverted, the result of test 132 is negative and steps 135-139 are reached. Steps 135-139 are the same as steps 42-46, except that the velocity and the increased position are negative (relative to the closing direction). Subroutine 140 is a closed constant (KC)
And a negative velocity command is generated as a function of the difference between the maximum position and the current position, and is otherwise the same as the subroutine 49. Then step and subroutine 50-55
Is AC (expressed as a count indicating pulse width)
Executed before matching amplitude.

In FIG. 8, the same steps and 59-
Since these are not the direction detections, 73 is executed as the door opening operation as described above. Test 74 is step 53
If the result of the test 74 is negative, the process goes to steps 77 and 78, and the phases of the windings V and W which are -120 ° and -240 ° from the phase of the winding U are determined. appear.

In FIG. 9, steps and tests 84.
The -102 and the subroutines 103 and 104 have the same closing direction as described above regarding the opening direction. In the first pulse through FIG. 10, the door closed flag is not set because the door has just started to close and has not yet been closed. The 1.5 / 2.5 second timeout does not occur, the result of test 108 is no, and other programming is reversed through return point 110.

In the next millisecond period, the millisecond interrupt program reaches through FIGS. 6 and 7 through entry point 34. In this case, tests 35, 36, 109 and 12
If the result of 7 is positive, test 132 determines if there was a reversal. If there is no inversion, the millisecond routine is executed. This continues to occur until eventually a 1.5 / 2.5 second clock timeout, as shown by the positive results in FIG. Then, the ramp down flag is set in step 111, and the determination and testing of the pulse width values for the three windings are performed in steps 112-114 and test 1.
15-117.

After a cycle after the 1.5 / 2.5 second clock times out, the pulse width has decreased to zero and test 120 is reached if the results of tests 115-117 are positive. In this case, since the door is closed, the door open flag is not set and it is denied, and the process goes to step 143 to set the door close flag. This flag is used to simply synchronize the routine of FIG. 10 with a physical switch that is common to all elevators to indicate that the door is fully closed. The switch settings are determined in test 144. On the next pass through the millisecond interrupt routine, the ramp down flag (FIGS. 6, 7) and the door close flag (test 107, FIG. 10) reach test 144.

If the physical switch is not closed, the result of the test 143 is negative and the return point 110 is reached. After all,
Door closed completely Switch closed, test 1
If the result of 44 is positive, steps 123 and 124 are reached. In test 123, the lamp down flag is reset, and in step 124, the door close flag (step 143
(Set in) is reset of course. It should be noted that the door flag is naturally reset at the end of the period when the door opening operation is irrelevant.

The millisecond interrupt routine consists of winding U,
It has two numbers each for V and W. One of these numbers is the pulse width generated in step 86, considered in step 101 of FIG. 9, the other number is the phase, which is normalized by step 84 and tests 84, 85 of FIG. To be done. The monitor winding is used when the phase produces the pulse width and all phases are
A triangular sine wave to determine if the motor windings should be driven in the positive or negative half cycle of the driving sine wave.

In FIG. 11, the 64 microsecond interrupt routine reaches through entry point 145 and a series of steps 146-148 generate U, V and W winding counters in steps 86 and 101 for each winding. Set to pulse width. The U-wound phase triangular sine wave signal is then tested in test 149. If the signal is positive, this means that a positive voltage should be applied to the U winding shown at (20, 21, FIG. 2) on the primary side 19 (FIG. 1) of the linear induction motor. This is accomplished by step 150, which connects to the input terminal 154 (with respect to the U winding) of filter 155 through a constant positive voltage (FIG. 13) switch 153 on line 152. On the other hand, if test 149 determines that the phase sign of the U winding is not positive, then step 156 reaches terminal 154 and applies a constant negative voltage from power supply 157 through switch 158. Similarly, the test and steps 161-166 in FIG. 11 cause a positive or negative voltage to be applied to the other terminals 169, 170 of the filter 155 from the power supply on lines 152, 157 through switches 171-174. In FIG. 11, when all counters are set and the three switches are turned on, other programming is returned through return point 175.

The filter 155 is arranged between the switches 153, 158, 171-174 and the primary side 19 of the linear induction motor in this embodiment.
The filter comprises three capacitors 177-179 connected in delta via a pair of windings (U, V; V, W; W, U).
Consists of Between each switch and the corresponding winding, coil 18
0-182 is the capacitor 177-179 is the switch 15
Prevents short circuits via 3,158,177-174. The capacitor and coil of the filter 155 are approximately 140
This breakpoint frequency is chosen because it is chosen to be a low pass filter that attenuates frequencies above hertz and frequencies below 140 hertz are inaudible to humans.
Filter 155 filters out the fundamental door current frequency harmonics. This door current frequency is shown in Figure 5.
Although 20 Hz is shown in the above, a range between 0 and 24 Hz may be used in the present invention.

As shown in FIG. 5, each winding is provided with a series of about 16 pulses within a millisecond period, each having a pulse width exhibiting a sinusoidal amplitude capable of high frequency winding in a particular winding.
The timing of these pulse widths is taken by decrementing the U, V and W counters and the correct pulse widths are set in steps 146-148 of FIG. When they decrease to zero, they cause an associated interrupt, such as the U counter interrupt for the U winding shown in FIG. Once the pulse width is complete for the U winding, the U counter will have an entry point of 1
Through 85, the interrupt is brought to FIG. 12, and a series of steps 186, 187 turn off the positive U switch 153 and the negative U switch 158 (even if one is on) and then reverse through the return point 188 to another plumbing. . Similar counter interrupt routines (not shown) are provided for the V and W windings.

The foregoing is a brief description of the physical devices and logical and arithmetic functions that the present invention may be used to implement. The logic flow is
It should not be the literal equivalent of the computer software that should be used to implement the invention,
Rather, it is merely an example of an implementation that can be implemented with well-known programming techniques. The present invention merely describes specific numbers and values for purposes of illustration. Obviously, other numbers and values can be used, as described to increase the benefit of using the invention in any embodiment. Instead of a few point calculations in the routine described here, a table can be used. The invention describes using a linear induction motor having a vertical pole on the primary side and a horizontal secondary side. This invention is of course also useful for other types of linear induction motors, for example with horizontal poles and secondary arranged horizontally from the primary side.

Thus, while the present invention has been shown and described with respect to representative embodiments, those skilled in the art can make various modifications, omissions and additions to the foregoing and various other modifications without departing from the spirit and scope of the invention. Should be understood by.

[0053]

According to the invention, mechanical perturbations due to alternating harmonics of the drive signal and double harmonics of the switching frequency used to supply the drive signal are reduced. A very quiet door operation can be achieved by reducing all input signals in the audible range.

Other objects, features and advantages of the present invention will be apparent from the detailed description of the embodiments as shown in the accompanying drawings.

[Brief description of drawings]

FIG. 1 is a simplified partially cutaway schematic view of an elevator door operating system using a linear induction motor according to an embodiment of the present invention.

2 is a schematic view of windings in the linear induction motor of FIG.

FIG. 3 is a waveform diagram of a current relationship of three windings in FIG.

4 is a waveform diagram showing a normal force and a linear force as a relationship between frequency and amplitude of current in the linear induction motor of FIG.

FIG. 5 is a series of time-based waveform diagrams showing the pulse width within a control cycle.

6 is a logic flow diagram of one millisecond interrupt routine used within the computer of FIG.

7 is a logic flow diagram of one millisecond interrupt routine used in the computer of FIG.

8 is a logic flow diagram of one millisecond interrupt routine used in the computer of FIG.

9 is a logic flow diagram of one millisecond interrupt routine used within the computer of FIG.

10 is a logic flow diagram of one millisecond interrupt routine used within the computer of FIG.

FIG. 11 is a diagram used in the computer of FIG. 1;
Logic flow diagram of millisecond interrupt.

12 is a logic flow diagram of a pulse width counter interrupt for winding U used in the computer of FIG.

13 is a block diagram of the driver of FIG. 1, including the connection of the filter of the linear induction motor of FIG. 1 and its windings.

FIG. 14 is a slip frequency diagram required to supply a predetermined force.

FIG. 15 is a simplified waveform diagram illustrating boosting according to the present invention.

[Explanation of symbols]

 17 ... Elevator door 18 ... Secondary side of linear induction motor 19 ... Primary side of linear induction motor 20-25 ... Winding 28 ... Photodetector 29 ... Line 30 ... Computer 31 ... Driver 155 ... Filter

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Edward E. Ahijan United States, Connecticut, West Hartford, Maplewood Avenue 65 (72) Inventor Jerome F. Jaminet United States, Connecticut, South Windsor, Birch Road 59 (72) Inventor Thomas Hey United States, Connecticut, Unionville, Woodside Drive 54 (72) Inventor Thomas Em. Kowartsik United States, Connecticut, Farmington, Sycamore Lane 15 (72) Inventor Richard E. Cullack Aldbone Drive, Connecticut, Bristol, United States of America 80 (72) Inventor: David W .. Burlet United States, Connecticut, East Heartland, Pell Road 29 (56) References JP-A-3-207255 (JP, A) JP-A-4-345486 (JP, A) Jitsukai-hei 3-207255 (JP, U) )

Claims (3)

(57) [Claims] 1. A three-phase variable voltage, variable frequency linear induction drive motor having a primary side including three windings that are Y-connected and a secondary side connected to an elevator door to provide linear motion. , A three-phase drive signal source having polarities alternating at a frequency of 50 hertz or less, and three capacitors connected in a delta shape and a filter composed of three inductors, each connection point of the pair of capacitors being the above-mentioned. One end of each inductor is connected to a connection point corresponding to the capacitor pair, and the other end of each inductor is connected to a corresponding phase of the three-phase drive signal source. And the capacitance of the capacitor is equal to the capacitance of the motor winding to form a low pass filter with a break point between 100 and 200 Hertz. Inductance is selected for the elevator car door driving device using a linear induction motor, characterized in that. 2. A door drive device according to claim 2, wherein
An elevator car door drive using a linear induction motor, wherein the capacitance of the capacitor is selected to form a low pass filter having breakpoints on the order of 140 hertz. 3. The door drive apparatus according to claim 1, wherein a three-phase drive signal source supplies a current to the winding in an open loop, and a door drive of an elevator car using a linear induction motor. apparatus.