Patent Application
20050278130
The invention and its wide variety of potential embodiments will be readily understood via the following detailed description of certain exemplary embodiments, with reference to the accompanying drawings in which:
At least one exemplary embodiment of the present invention includes a method comprising obtaining a first position, a second position, and a maximum motion parameter for a movement between the first position and the second position. The method also comprises creating a table of values comprising a plurality of target frequencies for the movement, the table of values also comprising a pulse width, a pulse count, and a differential pulse width corresponding to each of the target frequencies from the plurality of target frequencies. The method also comprises outputting at least a portion of the values from a pulse generator.
At least one exemplary embodiment of the present invention includes a method comprising obtaining a first frequency and a second frequency. The method also comprises creating a table of values comprising a plurality of target frequencies intermediate to the first and second frequencies, the table of values also comprising a pulse width, a pulse count, and a differential pulse width corresponding to each of the target frequencies from the plurality of target frequencies. The method also comprises outputting at least a portion of the values to a motion device. The method can also comprise transmitting the at least a portion of the values over the Internet.
At least one exemplary embodiment of the present invention includes a device comprising a sub-incremental clock-count-derived pulse generator. At least one exemplary embodiment of the present invention includes a device comprising a pulse generator adapted to produce a change from a first pulse output frequency to a second pulse output frequency by interpolating between pre-computed intermediate pulse output frequencies, a width of each pulse derived from real-time sub-incremental addition of clock counts during periods of acceleration and deceleration.
At least one exemplary embodiment of the present invention includes a system comprising an arithmetic logic unit. The system also comprises a memory comprising a pre-computed table of target pulse widths, changes in pulse width, and pulse counts distributed according to a constrained semi-logarithmic distribution, said memory connected to said arithmetic logic unit via a pipeline mechanism. The system further comprises a state machine adapted to load each of said target pulse widths and changes in pulse width from said memory into said arithmetic logic unit at pre-determined intervals of pulse count while maintaining control of a pulse width generated by said arithmetic logic unit.
At activity 1100, a motion control profile can be created, potentially by a user of the pulse generator. The user can provide a desired angular and/or linear distance to be traveled and a speed at which to cover the distance.
n certain embodiments, the user can provide a maximum velocity, a maximum acceleration, a maximum change in acceleration with respect to time (sometimes referred to as “jerk” or specified as a “jerk time”), and/or a maximum change in jerk with respect to time. The user can determine these values deterministically and/or empirically to achieve desired machine movements and/or avoid undesired effects, such as motor stalling, motor slippage, or other motion device problems (e.g., excess stresses, loss of friction between conveyed items and a conveyor belt, flinging of liquids, etc.).
In response, a motion control profile can be rendered. In certain embodiments, the motion control profile can indicate desired positions and times for a motion device on a position versus time plot. In other embodiments, the motion control profile can indicate a desired frequency ramp for a motion device on a frequency versus time plot. Such a motion control profile can include a beginning and an ending frequency, and can be linear, curvilinear, or a combination of linear and curvilinear portions between the frequencies. Potentially based on the user's input, in some cases the motion control profile can reflect an S-curve, in which the “take-off” from the beginning frequency gradually ramps up in frequency change, and the “landing” to the ending frequency gradually ramps down in frequency change, to avoid abruptness in frequency change near the beginning and ending frequencies. The motion control profile can be stored as, for example, a table of time-frequency points. Other potential plots can include position versus pulse count, frequency versus pulse count, speed versus time, speed versus position, acceleration versus time, acceleration versus pulse count, acceleration versus position, acceleration versus frequency, and/or acceleration versus speed. Note that a motion control profile can provide expressly certain motion parameters, and can imply other motion parameters. For example, a motion control profile can expressly describe a beginning position, an ending position, and a time over which a motion between the beginning and ending positions is to occur, which by implication describes an average velocity for the motion.
At activity 1200, the motion control profile can be stored, such as in a memory, for later retrieval. At activity 1300, the motion control profile can be obtained by, for example, a motion control processor of a pulse generator module of a programmable logic controller (“PLC”).
At activity 1400, the motion control processor can load a first and second frequency from the motion control profile, and can calculate a plurality of intermediate frequencies. The motion control processor can seek to adjust the pulse width of each pulse so as to achieve the frequency (corresponding to speed), acceleration, and/or change in acceleration specifications of the specific application while also meeting intermediate and/or total pulse count (corresponding to distance) specifications. Because each pulse has a finite width, and because pulse width can vary within a pulse train, new values of pulse width can be needed from a few times per second to several hundred thousand or millions of times per second in current typical applications.
For a motion control processor intended to supply a broad variety of end use needs, the range of motion parameters (e.g., frequencies, accelerations, times, and/or distances, etc.) to be accommodated can be represented and calculated as floating point values, and/or in a versatile fixed point format with a large number of significant digits. In certain embodiments of the present invention, the values can be computed before needed and stored in a memory for retrieval and use in real time.
At least one embodiment of the present invention can compute values (e.g., pulse width, change in pulse width, and/or number of pulses, etc.) associated with a plurality of frequencies intermediate in value to the first and second frequencies of the motion control profile prior to the start of motion and can store the values in memory. During execution of the motion, these values can be fetched from memory at pre-determined intervals of pulse count. The pulse widths of individual pulses issued during the intervals between fetched values can be computed as a simple linear addition of a change in pulse width for each pulse, accomplishing in real time an interpolation between the pre-calculated values of pulse width associated with the plurality of intermediate frequencies.
Computation of values prior to the start of motion, when calculation time is relatively unconstrained, can allow calculation to a desired level of precision with a processor that is limited in size, cost, and/or power. Computation of values associated with a sufficiently sparse plurality of intermediate frequencies can allow storage of the pre-computed data in a memory that is limited in size, cost, and/or power. In certain cases, changing each pulse width by means of a simple additive interpolation between the pre-computed points can allow for a closer approximation to the desired motion control profile than would be achieved by using only the plurality of intermediate frequencies, while limiting the cost, size, and/or power of the computing device that is required to have a fresh value available for each pulse as it occurs.
Where a pulse is defined as an event comprising a period of a logical “1” and a logical “0”, simple addition of a constant value to the pulse width with each pulse produces a non-linear change in frequency, since frequency is the inverse of pulse width. Sustained addition of a constant value to the pulse width can produce a markedly curved change in frequency with respect to time, with low acceleration at low frequency and high acceleration at high frequency. In order to achieve a given desired adherence to the specified frequency profile, the intervals between the points of the plurality of intermediate frequencies can be controlled, thereby controlling the curvature produced by the interpolative addition of a constant to the pulse width. For a constant acceleration region of a profile, said curvature is more marked at the lower values of frequency. By distributing the intervals between the points of the plurality of frequencies according to a semi-logarithmic progression, (or equivalently, by advancing each interval of increasing frequency by an equal predetermined percentage of the preceding frequency) a distribution of points can be achieved whereby more points are placed at the lower values of frequency where more attention to curvature control is desired, thus achieving a desired level of adherence to the profile with fewer points than would be achieved by points distributed equally in time.
Complete adherence to this semi-logarithmic distribution of points can result in very few points at the higher frequency portion of a frequency change region. For frequency profiles that include controlled changes in acceleration (S curves, or controlled “jerk”) sufficient points can be included at the higher frequencies to describe, to the desired level of accuracy, the desired limited change in acceleration between a region of high acceleration and a region of low acceleration or no acceleration (constant speed). Thus, the increasing interval between pre-calculated frequency points implied by a semi-logarithmic progression can be constrained, or limited, to provide points at some maximum interval consistent with representing the anticipated range of curvature desired for the S-curve, or limited acceleration case.
Further, there can be practical limits on the minimum interval between points, as required by, for example, a processor interrupt response time needed to load a new set of values, and/or by the requirement that the realizable intervals be greater than a current pulse width. Thus, the intervals calculated by the semi-logarithmic distribution can be examined and suitably constrained to no less than some minimum quantity determined by the implementing system constraints.
Each pulse width can be produced by counting out an integral number of clock events of some suitably higher frequency master clock, which number of clock events can be the stored representation of pulse width value, and each change in pulse width can be produced by adding a positive or negative number to the current value. In order to achieve the desired range of frequencies, accelerations, and/or pulse counts, with a desired level of precision, at least one embodiment of the current invention can represent the change in pulse width with an integral number of clock events and a sub-integral or fractional part number of clock events. A fractional part of a clock event is not necessarily expressed in a non-integral number of clock events per realized pulse width, but can be accumulated on each of the pulse width change additions for the specified interpolation interval. At such time that the accumulation of fractional or sub-integral parts exceeds a whole value, this whole value can become part of the expressed number of integral clock events per pulse width.
At activity 1500, any or all of the calculated values can be adjusted to conform to predetermined motion constraints and/or parameters. For the purposes of this application, motion constraints can include constraints on frequency, the first derivative of frequency with respect to time, and/or the second derivative of frequency with respect to time. Motion constraints can also include constraints on angular and/or linear position, velocity, acceleration, jerk (the first derivative of acceleration with respect to time), and/or the second derivative of acceleration with respect to time. Note that those constraints that are expressed as derivatives can be approximated using predetermined differential values (increments) of time. For example, jerk can be approximated as a maximum allowable acceleration value divided by a chosen increment of time. As another example, a specified minimum motion time can be followed.
Motion constraints can apply at any region of the motion control profile. For example, when first beginning a change in position of a stepper motor shaft, to avoid potential challenges such as slippage, jerk can be limited to a predetermined amount, such as 0.01 radian per second3 or 0.01 mm per second3.
At activity 1600, a table can be populated with the calculated and/or adjusted values. Each row of the table can contain a different intermediate frequency, each intermediate frequency having a corresponding pulse width, change in pulse width, and/or pulse count. Two adjacent rows can be considered to contain an adjacent pair of intermediate frequencies.
At activity 1700, the table can be read by a pulse generator of the PLC to output a pulse train. Table values can be provided to a motion control processor of the pulse generator, which can comprise a pipeline mechanism, an arithmetic logic unit (“ALU”), and/or a controlling state machine. The ALU can provide interpolated intermediate values of pulse width by adding a supplied positive or negative change in pulse width to each successive pulse width value, for the indicated number of pulses. The ALU may add a value representing an integral number and a non-integral or fractional number of master clock events, accumulating the fractional number over multiple pulse widths until such accumulation exceeds a whole number and is incorporated into the expressed pulse width. Control of the reading of the table and output of the pulse train can be governed by the state machine at predefined intervals, which intervals can be indefinitely deferred, interrupted, and/or re-initiated by command and/or external event while maintaining control of pulse width.
At activity 1800, the pulse train can be provided to a motion device. The pulse train can be provided via a direct connection to the pulse generator, and/or via a network connection, such as an Internet connection. The pulse train can be provided as a digital or an analog signal.
Pulse generator 2140 can be integral to PLC 2100. That is, once installed, pulse generator 2140 can be a component of PLC 1100, rather than free standing. Pulse generator 2140 can include a motion processor 2150 having a memory 2160, such as a dual port RAM. Motion processor 2150 can be a commercially available general-purpose microprocessor. In another embodiment, motion processor 2150 can be an Application Specific Integrated Circuit (ASIC) that has been designed to implement in its hardware and/or firmware at least a part of a method in accordance with an embodiment of the present invention. In yet another embodiment, motion processor 2150 can be a Field Programmable Gate Array (FPGA).
Memory 2160 can contain instructions that can be embodied in software, which can take any of numerous forms that are well known in the art. Pulse generator 2140 also can include a communications interface 2170, such as a bus, a connector, a telephone line interface, a wireless network interface, a cellular network interface, a local area network interface, a broadband cable interface, etc.
Pulse generator 2140 can be connected to a motion controller 2300 that is separate from PLC 2100. Motion controller 2300 can be connected to a motion drive and/or an actuator 2400. Pulse generator 2140 also can be connected via a network 2500 to a motion controller 2600 that is separate from PLC 2100. Network 2500 can be a public switched telephone network (PSTN), a wireless network, a cellular network, a local area network, the Internet, etc. Motion controller 2600 can be connected to a motion drive and/or an actuator 2700. Further, pulse generator 2140 can be connected to a motion controller 2180 that is integral to PLC 2100. Motion controller 2180 can be connected to a motion drive and/or an actuator 2200.
Connected to network 2500 also can be an information device 2900, such as a traditional telephone, telephonic device, cellular telephone, mobile terminal, Bluetooth device, communicator, pager, facsimile, computer terminal, personal computer, etc. Information device 2900 can be used to program, interact with, and/or monitor pulse generator 2140.
In one embodiment, network interface 3100 can be a telephone, a cellular phone, a cellular modem, a telephone data modem, a fax modem, a wireless transceiver, an Ethernet card, a cable modem, a digital subscriber line interface, a bridge, a hub, a router, or other similar device.
Each processor 3200 can be a commercially available general-purpose microprocessor. In another embodiment, the processor can be an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA) that has been designed to implement in its hardware and/or firmware at least a part of a method in accordance with an embodiment of the present invention.
Memory 3300 can be coupled to a processor 3200 and can store instructions 3400 adapted to be executed by processor 3200 according to one or more activities of a method of the present invention. Memory 3300 can be any device capable of storing analog or digital information, such as a hard disk, Random Access Memory (RAM), Read Only Memory (ROM), flash memory, a compact disk, a digital versatile disk (DVD), a magnetic tape, a floppy disk, and any combination thereof.
Instructions 3400 can be embodied in software, which can take any of numerous forms that are well known in the art.
Any input/output (I/O) device 3500 can be an audio and/or visual device, including, for example, a monitor, display, keyboard, keypad, touchpad, pointing device, microphone, speaker, video camera, camera, scanner, and/or printer, including a port to which an I/O device can be attached or connected.
Certain exemplary embodiments of the present invention include a position or motion module that connects a PLC to stepper motor via a stepper drive controller. Certain exemplary embodiments of this module are sometimes referred to herein as the EM 253 Motion Module. Certain exemplary embodiments of the PLC are sometimes referred to herein as the S7-200. Certain exemplary embodiments of the stepper motor are sometimes referred to herein as the Simostep P50 motor. Certain exemplary embodiments of the stepper drive controller are sometimes referred to herein as the Simodrive FM Step Drive.
Features of the EM 253 Motion Module
The EM 253 Motion Module can identify itself as an S7-200 intelligent module and can provide local inputs and outputs for interfacing with, for example, a single axis stepper motor as specified herein.
Communication between the module and the S7-200 PLC can be transacted over the Expansion I/O bus. The appropriate hardware can be provided in order to support communication initiated by either the PLC or the module.
The EM 253 Motion Module can provide the pulse outputs for motion control from 12 pulses per second (pps) to 200 K pulses per second (pps). In the event that this span of pulse rates cannot be provided as a single range, the module firmware can automatically select the operating range based upon the maximum speed specified in the module configuration. Based upon the maximum speed (MAX_SPEED) the module can compute the minimum speed (MIN_SPEED) for that range. The following ranges can be supported:
Programming Interface to the EM253 Motion Module
The S7-200 programming software can provide three functions to aid module configuration, profile creation, and module operation (control and status monitoring).
The configuration function can prompt the user to enter the required parameters. Then the profile creation function can prompt the user for the required information for each move profile. The information for each move profile can then be converted into a sequence of steps with a move identification number. Once the user has entered the information for the configuration and all the move profiles, the sequence of steps for each move can be combined into a single table as specified herein. The pointer to the V memory table can be stored in the section of the SDB provided for the module. Then both the data block for V memory and the SDB can be downloaded to the PLC.
A library instruction using the PCALL instruction and its corresponding subroutine can be provided as a standard function for controlling the module's operation. The user can be able to monitor the module's operation by enabling execution status of the library instruction.
Expansion I/O Bus Interface
The module can provide a ten-pin ribbon cable for connection to the expansion I/O bus interface on the PLC or previous I/O expansion module. It can also provide a ten-pin ribbon cable connector (male) into which another I/O expansion module can be connected. The module can return the ID code 0x21 which identifies the module as:
The module can provide a dual port RAM through which communication with the PLC can be accomplished. Configuration information for the module can be accessed using either the MPI or the more efficient Block Data Transfer. The module can utilize a maximum of one MPI request and/or multiple Block Data Transfer (BDT) request per transaction with the S7-200 CPU.
As part of power up initialization, the module can clear all data in the dual port RAM Banks 0 through 7. The fifty bytes of SM data area allocated for the intelligent module are defined in Table 1 (the definition is given as if this were the first intelligent module in the I/O system). In order for the CUR_POS and CUR_SPEED values to be consistent with one another the module H/W design can provide the facilities to capture both values as an atomic operation.
When an error condition or a change in status of the data is detected, the module can indicate this by updating the SM locations corresponding to the module's position. If it is the first module, it will update SMB200 through SMB249 as required to report the error condition. If it is the second module, it will update SMB250 through SMB299; and so on.
The module can implement Banks 15 through 127 for module initiated communication to the PLC. There is no requirement for the module to implement Banks 128 through 255
Module Configuration and Profile
Both the configuration and the profile information can be stored in a table in V memory in the PLC. The EM 253 Motion Module can access its configuration and profile information using the pointer value supplied in the SM locations in the PLC.
The Configuration/Profile Table can be divided into three sections. The first section is the Configuration Block, which can contain information used to set-up the module in preparation for executing motion commands. The second section is the Interactive Block, which can support direct setup of motion parameters by the user program. The third section can contain from 0 to 64 Profile Blocks, each of which can describe a predefined move operation that can be performed by the module.
Before the module can execute a profiled motion, it can perform the calculations to convert the general speed and position data provided in the profile block to the specific data and actions required to actually perform the move. These calculations can be performed whenever the module first sees the profile, but to improve responsiveness on subsequent executions of the profile, the module can provide a cache memory to store the complete execution data for up to four profiles. When the user commands the execution of a given profile, the module can check the cache to see if the profile is resident. If the profile is resident in the cache, the profile can be executed immediately. If the profile is not resident in the cache, the module can transfer the profile from the PLC's V memory to cache before it is executed.
The profile cache can be implemented as a FIFO queue, ordered according to the time a profile was last executed. When the user commands execution of a profile, that profile can become the newest entry, whether or not it previously existed in the cache. If the profile to be executed is not currently resident in the cache, then the resident profile with the longest period of time since it was last executed can be removed from the cache to make room for the newest profile. An example of cache behavior is shown in
The module can manage the cache memory automatically without any intervention required by the user. If the user changes profile information for profiles that have already been executed, the user can be responsible for commanding a module reconfiguration. A module configuration command can cause the module to read the configuration information and empty the cache. If the user does not change any of the configuration information (only changes profile information), then the module can empty the cache.
The cache is not necessarily used for motion controlled from the Interactive Block. When the command to execute a motion is issued, the module can read the data contained within the Interactive Block to obtain the specifications of the move.
The following table defines the structure of the Configuration/Profile Table which can be located in V memory of the S7-200 PLC. This information can be accessible by the module, but typically can not be changed by the module. The Byte Offset column of the table can be the byte offset from the location pointed to by the configuration/profile area pointer.
The Type field for each entry can specify the numeric format of all the double word values. If the MEAS_SYS configuration value is set to Pulses, a double integer value (int) can be used. If the MEAS_SYS configuration value is set to Engineering Units, a floating point value (fp) can be used.
The ranges given in Table 2 for speed and position values are expressed in units of pulses per second and pulses, respectively. When using engineering units, conversion to either pulses per second or pulses can be required to verify that the value is within the allowed range. In absolute mode the position range is −231 to 231 −1. However, each position change in an interactive move or each step of a profile move can have a range of 1 to 230-1 pulses, with the exception that a position change of 0 can be allowed for the initial step. For relative mode each position change in an interactive move or each step of a profile move can have a range of 1 to 230-1 pulses.
As shown in Table 3, the Profile Block Section of the Configuration/Profile Table can contain from 0 to 64 move profiles. If more than 64 move profiles are needed, the user can bear the burden of exchanging Configuration/Profile Tables, by changing the value stored in the configuration/profile table pointer.
Command Byte
The module can provide one byte of discrete outputs, which can be used as the command byte. The command byte can have the following definition, in which R: 0=idle, 1=excute command specified in command_code, as shown Table 4, below.
An interrupt can be generated on each rising edge of the R bit as an indication that a new command for profile execution has been received. Likewise, an interrupt can be generated on each falling edge of the R bit indicating a transition to an idle condition. Module firmware can have the ability to disable this interrupt.
If the module detects a transition to idle (R bit changes state to 0) while a command is active, then the operation in progress can be aborted and, if a motion is in progress, then a decelerated stop can be performed. Once an operation has completed, the module can require a transition to idle before a new command will be accepted. If an operation is aborted, then the module can complete any deceleration before a new command will be accepted. Any change in the command_code value while a command can be active can be ignored.
The motion module's response to a PLC mode change or fault condition can be governed by the effect that the PLC exerts over the discrete outputs according to the existing definition of the PLC function. Potential module reactions are described below:
(a) The PLC changes from STOP to RUN: The operation of the module is controlled by the user program.
(b) The PLC changes from RUN to STOP: The user can select the state that the discrete outputs are to go to on a transition to STOP or that the outputs are to retain their last state. Therefore, the following possibilities exist:
(c) The PLC detects a fatal error and turns OFF all discrete outputs—any motion in progress can be decelerated to a stop; further movement can be prevented as long as the XA_OD signal remains active.
(d) The motion module can implement a watchdog timer that will turn the outputs OFF in the event that communication with the PLC is lost. In the event that the output watchdog timer expires any motion in progress can be decelerated to a stop.
(e) In the event that the motion module detects a fatal error in the module's H/W or firmware, the P0, P1, DIS and CLR outputs can be set to the inactive state.
Command 0-63, Execute Motion Specified in Profile Block 0-63
Execution of this command can cause the module to perform the motion operation specified in the MODE field of the Profile Block indicated by the command_code portion of the command. The specifications for Interactive Block motion operations typically are not cached, so they can be read each time that the module receives this command.
In mode 0 (absolute position) the motion profile block can define from one to four steps with each step containing both the position (POS) and speed (SPEED) that describes the move segment. The POS specification can represent an absolute location, which is based on the location designated as reference point. The direction of movement can be determined by the relationship between the current position and the position of the first step in the profile. In a multi-step move a reversal of direction of travel can be prohibited and can result in an error condition being reported.
In mode 1 (relative position) the motion profile block can define from one to four steps with each step containing both the position (POS) and the speed (SPEED) that describes the move segment. The sign of the position value (POS) can determine the direction of the movement. In a multi-step move, a reversal of direction of travel can be prohibited and can result in the reporting of an error condition.
In the single-speed, continuous speed modes (2 and 3), the position (POS) specification can be ignored and the module can accelerate to the speed specified in the SPEED field of the first step. Mode 2 can be used for positive rotation and mode 3 can be used for negative rotation.
In the single-speed, continuous speed modes with triggered stop (6 and 7) and RPS inactive, the module can accelerate to the speed specified in the SPEED field of the first step. If and when the RPS input becomes active, movement can stop after completing the distance specified in the POS field of the first step. If the POS=0, then the movement can decelerate to a stop without regard to the distance traveled. Mode 6 can be used for positive rotation and mode 7 can be used for negative rotation.
In modes 8 and 9, the binary value of the RPS input can select one of two continuous speed values as specified by the first two steps in the profile block. Mode 8 can be used for positive rotation and mode 9 can be used for negative rotation. The SPEED can control the speed of movement. The POS values can be ignored in this mode. The following table defines the relationship between the inputs and the step within the profile block.
Command 118, Activate the DIS Output
Execution of this command can result in the activation of the DIS output.
Command 119, De-activate the DIS Output
Execution of this command can result in the de-activation of the DIS output.
Command 120, Pulse the CLR Output
Execution of this command can result in the issuance of a 50 ms pulse on the CLR output.
Command 121, Reload Current Position
Execution of this command can cause the module to read the value found in the TARGET_POS field of the Interactive Block and set the current position to that value.
Command 122, Execute Motion Specified in the Interactive Block
Execution of this command can cause the module to perform the motion operation specified in the MOVE_CMD field of the Interactive Block. The specifications for Interactive Block motion operations typically are not cached, so they can be read each time that the module receives this command.
In the absolute and relative motion modes (0 and 1), a single step motion can be performed based upon the target speed and position information provided in the TARGET_SPEED and TARGET_POS fields of the Interactive Block.
In the single-speed, continuous speed modes (2 and 3), the position specification can be ignored and the module can accelerate to the speed specified in the TARGET_SPEED field of the Interactive Block.
In the manual speed control modes (4 and 5), the position specification can be ignored and the user program can load the value of speed changes into the TARGET_SPEED field of the Interactive Block. The motion module can continuously monitor this location and respond appropriately when the speed value changes.
In the single-speed, continuous speed modes with triggered stop (6 and 7) and RPS inactive, the module can accelerate to the speed specified in the SPEED field of the first step. If and when the RPS input becomes active, movement can stop after completing the distance specified in the POS field of the first step. If the POS=0, then the movement can decelerate to a stop without regard to the distance traveled. Mode 6 can be used for positive rotation and mode 7 can be used for negative rotation.
Command 123, Capture Reference Point Offset
Execution of this command can allow for the establishment of the zero position that is at a different location from the reference point position.
Before this command is issued, the reference point position can be determined and the user can jog the machine to the work starting position. Upon receiving this command, the module can compute the offset between the work starting position (the current position) and the reference point position and write that computed offset to the RP_OFFSET field of the Interactive Block. Then, the current position can be set to 0. This can establish the work starting position as the zero position.
In the event that the stepper motor loses track of its position (power is lost, the stepper motor is repositioned manually, etc.) the Seek to Reference Point Position command can be issued to re-establish the zero position automatically.
Command 124, Jog Positive Rotation
This command can allow the user to manually issue pulses for moving the stepper motor in the positive direction.
If the command remains active for less than 0.5 seconds, the motion module can issue the number of pulses specified in JOG_INCREMENT while accelerating to the JOG_SPEED. If the command remains active for 0.5 seconds or longer, the motion module can begin to accelerate to the specified JOG_SPEED. Once a transition to idle is detected, the module can decelerate to a stop.
Command 125, Jog Negative Rotation
This command can allow the user to manually issue pulses for moving the stepper motor in the negative direction.
If the command remains active for less than 0.5 seconds, the motion module can issue the number of pulses specified in JOG_INCREMENT while accelerating to the JOG_SPEED. If the command remains active for 0.5 seconds or longer, the motion module can begin to accelerate to the specified JOG_SPEED. Once a transition to idle is detected, the module can decelerate to a stop.
Command 126, Seek to Reference Point Position
Execution of this command can initiate a reference point seek operation using the specified search method. When the reference point has been located and motion has stopped, then the module can load the value read from the RP_OFFSET field of the Interactive Block into the current position.
Command 127, Reload Configuration
Execution of this command can cause the module to read the configuration/profile table pointer from the appropriate location in SM memory. The module then can read the Configuration Block from the location specified by the configuration/profile table pointer. The module can compare the configuration data just obtained against the existing module configuration and perform any required setup changes or recalculations. Any cached profiles can be discarded.
Reference Point Definition
The location known as the reference point can have one of several pre-defined sequences of module input conditions. The user can select the definition of the reference point that most closely matches the needs of the application. Once the user has selected a definition for the reference point and configured the module accordingly, the user can issue the seek reference point command. In response to this command, the module can automatically seek the reference point position, stop at the reference point and activate the CLR output for a period of 50 msec.
Potential definitions of the reference point that the user can choose from are listed below (the number of options are multiplied by four when all combinations of the RP_SEEK_DIR and RP_APPR_DIR specifications are included):
a) Mode 1: The reference point can be where the RPS input goes active on the approach from the work zone side.
b) Mode 2: The reference point can be centered within the active region of the RPS input.
c) Mode 3: The reference point can be located outside the active region of the RPS input. RP_Z_CNT can specify how many zero pulse counts on the ZP input to move after the RPS input goes inactive.
d) Mode 4: The reference point can be most likely located within the active region of the RPS input. RP_Z_CNT can specify how many zero pulse counts on the ZP input to move after the RPS input goes active.
For
User Interface
Table 5 shows the inputs, outputs and status LED's for the module.
The module's specification for the inputs and outputs are shown in Table 6. The operation of open drain outputs above 5VDC might increase radio frequency emissions above permissible limits. Radio frequency containment measures might be required for certain systems or wiring. Depending on the pulse receiver and cable, an additional external pull up resistor might improve pulse signal quality and noise immunity.
State Diagram
A state diagram for the motion module can include the following variables:
run_q: qualified run command, command register content changes are recognized at chosen safe times—generally in sub-states below the top level diagram.
modes from pc(31:30)
xfr_pipeline: command to load refreshed pipe values to working registers, commanded by CPU or RPS edge.
run_cid: run instruction to pulse generator state machine.
pccapture_time: a timing strobe, name comes from original use to identify when pulse counter contents are stable for read/capture. Now used only for state transition timing.
States:
idle: do nothing, exit on xfr_pipeline command to init_load.
init_load: move initial pipeline values to working registers, wait for run command to transition.
run_a_step: enable pulse machine, set run_ack, periodically examine pulse count & run command, exit on:
pipe_load: load refreshed pipe values to working registers, examine new pc[31:30] to determine next state:
run_continuous: enable pulse machine, acknowledge run or abort command, periodically examine run command and xfr_pipeline.
abort: instruct pulse machine to freeze pulse width, go to run_continuous
estop: disable pulse machine ability to start new pulses, clear run_ack, exit to idle on pulse complete.
finish: disable pulse machine ability to start new pulses, clear run_ack, exit to idle on pulse complete and run command=stop.
Certain potential operating sequences for the pulse generator module are shown in Table 8, below.
For certain motion modules, potential CPU ASIC pin assignments and address mapping are provided in Table 9, below:
Pulse Calculations
A number of motion scenarios can be visualized. One scenario of particular difficulty emerges when the distance an actuator is to travel is short and insufficient to achieve maximum speed. For this scenario, at least three cases can be recognized.
Generally, a linear slope can be defined by F_start, F_end (F0), and t_a (the acceleration time), such that
Also, n_end=number of pulses from F_start to F_end.
In the case of an S-curve:
The user can specify an n_target (nt) that corresponds to the desired motion distance.
Case 1: Linear Acceleration, Acceleration=Deceleration
Solution:
n1=1/2a1t12+F0t1;n2=1/2a2t22+F0t2; t2=(a1/a2)t1
For very short moves, F0*t might be an appreciable part of the total move.
nt=F0(t1+t2)+1/2a1t12+1/2a2t22, or
a1/2(1+a1/a2)t12+(1+a1/a2)F0t1−nt=0
a1/2t12+F0t1−nt/(1+a1/a2)=0
t1=−F0/a1+sqrt{F02/a12+2nt/(a1(1+a1/a2))}
For realizable motions there is always a positive number under the radical and the realizable root is always found by adding a positive square root to the initial term.
Use the above expression to compute t1. Now, finding n1=1/2 a1t12+F0t1, choose way point just less than n1 on the acceleration slope, with associated speed F1. Interpolate as required on the deceleration slope to match speed between acceleration and deceleration slopes, finding an n2 on the decel slope. As before, find the length of a constant speed step that links the two slopes, n_cs=nt−n1−n2. The total move is accelerate for n1 steps, a (short) constant speed step n_cs, decelerate for n2 steps.
Case 3: Short S-Curve—Distance n not Sufficient to Complete S-Curve to Max Speed
For the case of a short S-curve: tj=jerk time, amax=(F_end−F_start)/t_a,
In the initial curve region:
a=kt
F=F0+1/2kt2
F change with jerk time tj: Fj=1/2 k tj2
For the S-curve, recall there is an Fj associated with jerk time, tj, and that the change in F during the curve region is symmetrical during the increasing and decreasing acceleration portions of the curve.
One procedure is to solve for the simple trapezoid as before, identifying a maximum speed F1 reached at count n1 and time t1. Then, reduce the maximum frequency to a new Fmax which can be achieved in the same accel/decel time t1. The n_cs (constant speed region) now gets enlarged a bit from the simple trapezoid case, such that: n_cs=nt−2*n(Fmax).
If t1>=2 tj: There is a linear region+2 full jerk times. Obtain Fmax=F0+2*Fj+(t1−2tj)*amax. Apply standard S curve calculation to Fmax to get n vs. v up to Fmax. Then the constant speed distance is n_cs=nt−2*n(fmax)
If t1<2 tj: There will be a pure S-curve with an inflection point at t1/2. The ΔF around the inflection point: Finflection=1/2 k (t1/2)2, and Fmax=F0+2*Finflection=F0+k(t1/2)2. Apply the standard S curve to Fmax and find n_cs as before.
What follows is an exemplary pseudo-code subroutine listing for determining a plurality of target frequencies, pulse widths, differential pulse widths, and pulse counts, for a portion of a motion control profile, that portion being a single change in frequency according to a specified beginning frequency, ending frequency, maximum acceleration, and maximum change in acceleration(that is, an S-curve case) such as that described above. The target frequencies are distributed semi-logarithmically, with constraints to achieve realizable minimum time intervals, and maximum time intervals consistent with a desired level of accuracy in representation of the upper frequency portion of the S curve. All data input and output has been removed to clarify the algorithm.
Although the invention has been described with reference to specific embodiments thereof, it will be understood that numerous variations, modifications and additional embodiments are possible, and accordingly, all such variations, modifications, and embodiments are to be regarded as being within the spirit and scope of the invention. For example, programming of the motion module can occur over the Internet. Likewise, transmission of the pulse train can occur over the Internet. References specifically identified and discussed herein are incorporated by reference as if fully set forth herein. Accordingly, the drawings and descriptions are to be regarded as illustrative in nature, and not as restrictive.
The present application is a divisional patent application of patent application Ser. No. 10/309,539 filed on Dec. 4, 2002, of McNutt for “Pulse Output Function for a Programmable Logic Controller”. This application claims priority to, and incorporates by reference herein in its entirety, the following provisional applications: Ser. No. 60/346,488 (Attorney Docket No. 2002P00150), filed 7 Jan. 2002; and Ser. No. 60/384,979 (Attorney Docket No. 2002P08887US), filed 3 Jun. 2002.
| Number | Date | Country | |
|---|---|---|---|
| 60346488 | Jan 2002 | US | |
| 60384979 | Jun 2002 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10309539 | Dec 2002 | US |
| Child | 11177275 | Jul 2005 | US |
